Disorders A - Z:   A    B   C    D    E    F    G    H    I    J    K    L    M    N    O    P    Q    R    S    T    U    V    W    X    Y    Z

Skip secondary menu

Final Report of the Stroke Progress Review Group - Topic area working group full reports

Additional information about Stroke PRG 

Table of Contents


Co-Chairs: Jeffrey Saver, Helmi Lutsep, Patrick Lyden 

Members: William Barsan, Kyra Becker, Gabrielle DeVeber, Jeff Frank, Kama Guluma, Clarke Haley, Michael Hill, Reza Jahan, Pooja Khatri, Tom Kwiatkowski, Brett Meyer, Raul Nogueira, Larry Wechsler

NINDS Liaison: Scott Janis


Progress has been made on the 2002 and 2006 SPRG goal to develop therapeutic agents that open blood vessels in more patients and do so more frequently, faster, and with greater safety, but further progress is a high priority. We also recommend prioritization of studies that will potentially increase eligibility for IV rtPA.

The 2002 Stroke PRG correctly foresaw that neuroprotective therapy started late with agents targeting single mechanisms was unlikely to be successful in humans. The Stroke PRG helped foster novel strategies, including developing single agents with pleiotropic, robust effects, like hypothermia and albumin, and prehospital techniques for hyperacute initiation of treatment with the first minutes after onset.

Since the 2006 report of the SPRG, there has been a significant increase in the number of certified stroke centers and attention to systems of care with an associated improvement in delivery of stroke therapies and improvement in outcomes. Continued expansion and integration of care across the spectrum of acute treatment facilities is needed to ensure access to acute stroke treatment for all Americans.

The field of pediatric stroke has developed rapidly, and has reached the stage where focused funding prioritization is required to make further advances including understanding mechanisms and optimal treatments. The existence of highly motivated and collaborative researchers in this field provides tremendous potential for leveraging research productivity.

In the past 10 years, there has been a substantial improvement in the integration of Emergency Medicine investigators in NINDS funded stroke research, most prominently in the successful research networks - NETT and SPOTRIAS. There has also been increased involvement of prehospital providers in emergency stroke research and a successfully recruiting phase 3 clinical trial in stroke which enrolls in the prehospital setting.

Since the 2006 Stroke PRG report, much progress has been made in telestroke: more telestroke programs are in place, rt-PA rates have increased, and clinical trials showed excellent decision-making efficacy using telestroke. With continued support from NINDS, clinicians and researchers can expand current telestroke procedures, leverage telemedicine for research recruitment, and innovate on smartphone and other emerging platforms.

Recent changes in hospital reimbursement improved the climate for development of stroke centers, but also created new challenges: a disincentive to transfers of “drip and ship” patients in a stroke system of care and a barrier to recruitment of patients into acute stroke clinical trials. Further research regarding the impact of these issues on the public health problem of delivering acute stroke services is urgently needed.

The SPRG process furthered awareness of potential sex differences in acute stroke trials. However, future clinical trials need to include sufficient numbers of women and under-represented race-ethnic groups to evaluate treatment effects in these populations.

In acute ischemic stroke therapy, the mechanism of stroke is less important than the location of the occluded artery. Thrombus pathology would be useful if it could then be shown to help stratify therapeutic decision-making. Assessment of interventions to prevent early recurrent stroke or to prevent progression of existing deficits is needed during the acute hospital stay. Further assessment of brain, vascular, and cardiac imaging technologies, their diagnostic utility and the timing of their use, is needed. ?



Several novel fibrinolytics were explored, but none entered clinical practice. The NINDS-funded tenecteplase (TNK) trial ended prematurely due to slow recruitment, but additional efforts to study this drug are ongoing. The industry-funded ancrod (viprinex) trial was ended after interim analyses revealed futility and a trend towards increased bleeding, demonstrating the value of new, stronger approaches to early futility analyses with rigorous stopping rules. The industry-funded DIAS 2 trial, testing desmoteplase in imaging selected patients, did not demonstrate clinical efficacy. Successor DIAS 3 and 4 trials with a refined imaging selection strategy are under way. Intra-arterial plasmin advanced in early clinical studies.

Studies combining fibrinolytic agents with additional drugs showed safety and sparked further study. The NINDS-funded CLEAR Trial combining IV rtPA with eptifibatide within 3h of stroke onset suggested safety; a successor trial, CLEAR-ER, is now testing higher doses. The NINDS-funded TARTS trial studies combined fibrinolysis and thrombin inhibition with argatroban and showed safety with a trend towards reduced rates of re-occlusion; a follow-up project (ARTTS) extended the findings to a larger population.

Pilot trials analyzed co-administering ultrasound microsphere contrast agents with rtPA and transcranial Doppler (TCD) ultrasound. TCD has already been demonstrated to enhance recanalization, and gaseous microspheres could further potentiate this effect. As the spheres expand and oscillate at the target thrombus site in response to the ultrasound pressure wave, they can produce microdisruptions of clot integrity and facilitate lytic penetration throughout the thrombus. Two small industry pilot trials found evidence of enhanced recanalization with this approach, but potentially increased rates of hemorrhagic transformation at higher microsphere doses.

Combining endothelioprotection and neuroprotection with lytics, to reduce hemorrhagic conversion and increase salvaged brain volume, continued to be explored. The second industry-funded trial of NXY-059, SAINT 2, failed to confirm prior evidence of protective effect. A NINDS-funded safety and feasibility trial (MINO) found minocycline safe when given alone or in combination with IV rtPA. Hypothermia in combination with lytics was studied in the SPOTRIAS funded ICTuS-L trial; immediately after rt-PA therapy the placement of an endovascular cooling catheter was shown to be safe. There was a trend towards fewer hemorrhages with hypothermia compared to normothermia. Starting neuroprotectants prior to IV rtPA exposure was shown feasible in the NINDS-funded, phase 3 FAST-MAG Trial.

Substantial advances occurred in mechanical recanalization therapies. In addition to the MERCI retrieval device, the Penumbra aspiration device was FDA-cleared in 2007, based on single-arm safety and preliminary efficacy data. These two device families are increasingly used in the United States. More recently, stent retrievers demonstrated safety and high recanalization rates in limited case series leading to regulatory approvals in Canada and Europe. Randomized trials in the US comparing stent retrievers with the predicate Merci retriever device are nearing completion.

The clinical efficacy of mechanical revascularization compared to IV rtPA, or compared to supportive medical management in later time windows, remains to be determined in a randomized trial. The NINDS-funded SPOTRIAS MR RESCUE trial of mechanical embolectomy versus medical management within 8 hours of stroke onset is within one year of completion. The NINDS-funded, international IMS III Trial of IV rtPA versus IV rtPA followed by intra-arterial therapies has completed over half its recruitment, and will require several additional years for completion.

A novel, collateral enhancement approach towards reperfusion, partial aortic occlusion, did not demonstrate clinical efficacy on its primary endpoint. Potential signals of effect observed on secondary endpoints require confirmation.


The SAINT 2 trial failed to confirm the benefit seen in SAINT 1 with NXY-059, a putative free radical scavenger. This result, on top of multiple past failures of single-mechanism, tightly targeted neuroprotective agents started relatively late after stroke onset, led to a marked diminution of industry-funded adjuvant drug clinical trials.

Paradoxically, the exodus of large pharmaceutical sponsors from the field occurred just as the principle of neuroprotection for human brain ischemia was validated by positive trials and wide adoption in clinical practice of hypothermia treatment for global brain ischemia/reperfusion injury in cardiac arrest.

These developments suggested that neuroprotection was achievable in human patients, but would require new clinical-translational research paradigms. NIH-NINDS funded projects helped to establish two new strategic directions: 1) prehospital neuroprotective treatment start in an ultra-early time window, and 2) testing pleiotropic agents that exert multiple mechanisms of action and are highly potent in preclinical models.

NINDS-supported pilot and pivotal trials of the neuroprotectant magnesium sulfate pioneered the strategy of achieving hyperacute treatment start by enrolling patients in the ambulance, prior to hospital arrival. By July 2011, the FAST-MAG Phase 3 Trial had enrolled 1350 of its planned 1700 patients, with median onset to treatment time of 47 minutes. Fully 73% were enrolled within the first (“golden”) hour after onset. Building on FAST-MAG, a British academic group launched a second prehospital neuroprotective study for stroke – the pilot RIGHT trial of the nitric oxide donor glyceryl trinitrate.

While regulatory and scientific barriers prevented realization of the original Stroke PRG’s call for multiple trials of combinations of neuroprotective agents, an allied course was pursued by studies testing single interventions that exert pleiotropic effects on multiple pathways in the “ischemic cascade” – the molecular elaboration of injury within hypoxic environments.

  • Hypothermia is a highly pleiotropic intervention that is the most robust neuroprotective treatment known in preclinical studies. Over the course of the decade, several feasibility trials, including two NINDS SPOTRIAS trials, refined methods of surface and endovascular cooling, shivering suppression, and rewarming. The ICTuS-L trial randomized 58 patients, most also treated with rt-PA, to cooling or no cooling. Favorable results led to NINDS funding of the currently ongoing, multicenter ICTuS 2/3 Trial, a phase 2b to 3 pivotal study.
  • Albumin exerts antioxidant, endothelial relaxation, and microvascular permeability stabilization effects in acute brain ischemia. In the NINDS-supported, phase 3 ALIAS 1 trial, 434 subjects were enrolled. The trial was stopped early to analyze potential systemic adverse effects in older enrolled patients when an increased death rate was noted in patients over age 83. A signal of potential efficacy was nonetheless noted, with favorable final outcome noted in 44.7% of albumin treated patients vs. 36.0% of placebo. The NINDS-supported successor ALIAS 2 phase 3 trial is underway with modified entry criteria and drug administration regimen to enhance safety.
  • Several additional pleiotropic adjuvant protection agents progressed in development. Successful NINDS-funded phase 2 trials of high dose lovastatin, minocycline, and the combination of caffeine and ethanol (caffeinol) were performed.
  • Cytidyldiphosphocholine or citicoline, a naturally occurring nucleoside essential for the formation of phosphatidyl-choline and maintenance of membranes, is approved in several countries for acute stroke and is available in the US as a dietary supplement. An industry-sponsored international trial, ICTUS, made substantial progress toward completing enrollment in a large, pivotal trial.

Transcranial near-infrared laser therapy, a novel combined neuroprotection and early neuroplasticity-enhancing intervention intervention, showed promise in two industry-supported clinical trials. Through absorption of 808 nm wavelength light by cytochrome c oxidase and other chromophore molecules, photobiostimulation may re-energize mitochondria, exert anti-apoptotic effects, and promote neurogenesis and synaptogenesis. Strong signals of potential benefit were observed in a pilot and a first pivotal trial.

Stroke Centers

The number of Joint Commission certified Primary Stroke Centers (PSCs) increased dramatically. By the end of 2010, more than 50% of Americans resided in a state or County in which EMS systems preferentially delivered acute stroke patients directly to designated Primary Stroke Centers.

The benefits of receiving stroke care in a PSC are now demonstrated, with documented decreases in mortality and increased use of thrombolytics.

Metrics for measuring the quality of care at Comprehensive Stroke Centers (CSCs) have been published and plans for launch of Joint Commission certification of CSCs are in advanced stage.

Increased attention to quality performance issues led to more hospitals using programs like the American Heart Association’s Get with the Guidelines (GWTG) – Stroke. National registry studies demonstrated that hospitals that use GWTG-Stroke show better adherence to performance measures over the course of time.

Pediatric Stroke

The field of Pediatric Stroke grew substantially with numerous retrospective observational studies published. Multicenter collaborative networks have formed and stabilized primarily through foundation funds and voluntary efforts of investigators. Two NINDS funded prospective clinical observational studies were launched: PedNIHSS and VIPS.

However, outside of one sickle cell trial funded by NHLBI, no clinical trials have been proposed or funded. [Editorial comment: The NINDS is funding the Thrombolysis in Pediatric Stroke (TIPS) trial to assess the safety of tPA in childhood stroke.]

The International Pediatric Stroke Study (IPSS), a collaborative network of pediatric stroke researchers, formed in 2003 to conduct multi-center research studies in pediatric stroke. The IPSS has expanded to 45 centers, which have enrolled over 2800 neonates or children with either arterial ischemic stroke or cerebral sinovenous thrombosis. Three and 12 months post-stroke outcomes are collected at approximately 50% of centers utilizing validated standardized measures including the Pediatric Stroke Outcome Measure. The IPSS infrastructure, communications and two annual meetings are funded by a Canadian private foundation research grant.

Prehospital and Emergency Department Investigators and Protocols

The Neurological Emergencies Treatment Trials (NETT) Network is an interdisciplinary network with teams of emergency physicians, neurologists, neurointensivists, neurosurgeons and biostatisticians with specific expertise in the conduct of emergency clinical trials. The NETT was initially funded by NINDS in 2006 and expanded successfully in the last 5 years, to 17 hubs with associated spokes. Two acute ischemic stroke trials have begun enrolling patients within the NETT: 1) ALIAS 2 (albumin); and 2) POINT- a study investigating an early loading dose of oral clopidigrel in patients with high risk TIA or minor stroke to prevent secondary stroke. NETT sites have to date contributed over half of the patients enrolled in both trials.

The NIH-funded Specialized Program for Translational Research in Acute Ischemic Stroke (SPOTRIAS) phase 2 trial network requires joint leadership by emergency physicians and vascular neurology specialists. Emergency physician-investigators have played lead roles in several network studies, serving as the principal investigators of trials evaluating external counterpulsation for acute ischemic stroke, the combination eptifibatide with rt-PA for acute ischemic stroke, the use of computed tomography angiography to predict hematoma expansion in intracerebral hemorrhage, and preclinical models of ultrasound thrombolysis.

The NINDS Field Administration of Stroke Therapy – Magnesium (FAST-MAG) Phase 3 Trial, testing the neuroprotective agent magnesium sulfate, pioneered prehospital initiation of research stroke treatment. The FAST-MAG consortium in Los Angeles and Orange Counties brought together over 41 EMS Provider Agencies, 2900 paramedics, and 400 Emergency Physician investigators.

The Increasing Stroke Treatment through Interventional behavior Change Tactics (INSTINCT) trial is funded by NINDS to evaluate a standardized multi-level barrier assessment and interactive educational intervention in increasing appropriate tPA use in stroke and emergency physician knowledge about the use of tPA. The primary investigator in INSTINCT is an emergency physician. Many of the INSTINCT study sites are rural, helping extend appropriate stroke care to patients in rural areas.


Telemedicine for stroke evolved into a standard part of some stroke clinical practices.

Studies demonstrated that, when telestroke was introduced into a region, rates of rt-PA use increased dramatically. The Bavarian TEMPiS study noted an approximate 10 fold increase in rt-PA treatments. In the NINDS-SPOTRIAS funded STRokE-DOC project, the rate of rt-PA use in rural ‘spoke’ centers increased from essentially nil to 25% of all acute ‘code stroke’ patients.

The geographic extent of telestroke networks grew dramatically. Hub and spoke telestroke networks became common throughout the US.

The technological specifications and capabilities for telestroke systems advanced. Generally accepted requirements included high resolution (400 x 300) capabilities, full screen video abilities, full frame video rate (30 frames per second), synced audio with video, echo cancellation, high data compression and bandwidth, HIPAA compliant security measures, DICOM radiology CT image access and viewing software, and the ability to access a system with minimal, or no delay. As industry embraced these specifications, technology became less expensive, and many more telemedicine vendors entered the marketplace. Leading telecommunications companies developed their telemedicine systems to include medical friendly applications, enhancements and decision support tools (instead of simply utilizing teleconferencing software for these purposes). Following the STRokE DOC model, almost all telestroke vendors have transitioned from a “point to point” ISDN connectivity platform to an Internet based “site- independent” platform, supported by advances in bandwidth and variable Internet access tools. Broadband technologies (EVDO/CDMA and cell phone signal combining techniques) and enhanced quality of service (QoS) algorithms have enabled the realistic use of telestroke software on an ever changing array of wireless devices including base station, desktop, site- independent laptop computer, and even smartphone technologies.

A NINDS SPOTRIAS-supported randomized clinical trial comparing telemedicine to telephone for acute stroke medical decision-making showed efficacy for telestroke (STRokE DOC).

A 2009 AHA/ Scientific Statement found substantial evidence supporting the use of telemedicine for remote neurologic examination, remote neuroimaging interpretation, medical decision-making regarding thrombolytic stroke therapy, standardized disability rating scale assessments, consent elicitation for prehospital hyperacute trials, and remote physical and occupational therapy delivery.

Financial Barriers to Acute Stroke Care

Changes in reimbursement for acute stroke care improved delivery of services to stroke patients. The greater payment to hospitals associated with diagnosis-related groups (DRGs) for stroke patients treated with thrombolytic therapy and mechanical devices helped offset the higher costs and encouraged hospitals to create and sustain stroke centers. Providing physician leadership with appropriate resources and administrative stipends became financially viable with this DRG fee structure.

Recent research has suggested that both intravenous tPA up to 4.5 hours after stroke onset and mechanical thrombectomy are cost-effective interventions.

Diversity Issues

Sex has substantial effects on stroke outcomes and trajectories through the acute stroke care system. Investigation of these determinants of acute stroke care and outcome increased.

  • Building upon earlier studies describing a gender-by-thrombolysis interaction, a study investigated effects of sex on mechanical embolectomy outcomes in the MERCI and Multi MERCI databases. This study found that the rates of favorable outcomes were not different in women and in men when the vessel was revascularized.
  • Preclinical studies investigated mechanisms of sex-linked ischemic brain injury.

Reports showed blacks are less likely than whites or Hispanics to be eligible for or treated with IV tPA.

A retrospective multicenter study in Japanese patients treated with low dose alteplase (0.6 mg/kg) showed outcomes similar to studies using 0.9 mg/kg. A prospective, multicenter observational study in Taiwan investigating two doses of alteplase in Chinese patients, a standard dose (0.90 mg/kg and a lower dose (0.72 mg/kg), found the standard dose associated with higher intracranial hemorrhage and mortality rates.

Ischemic Stroke Mechanisms

The mechanism of stroke is very important when considering minor stroke and TIA or major stroke where thrombolysis has showed very early success in recanalization. These patients are at risk of early recurrence/early deterioration. Knowledge of the mechanism may help to guide therapy.

The quality of investigations to determine stroke mechanisms improved as imaging technology for the heart, brain and vessels advanced. CT angiography and cardiac MR were particularly noteworthy innovations.



There has been tremendous progress in the availability of effective mechanical revascularization therapies for strokes with proximal arterial occlusions. In early studies of currently available devices, 80-100% partial recanalization rates were reported, though complete recanalization rates still remain suboptimal. As these devices become increasingly common in clinical practice, understanding the clinical settings that will yield improved patient outcomes becomes pressing. Further, optimizing their use to limit complications and intracranial hemorrhage remains an important goal. Large, well designed studies are needed to determine the relative merits of IV rtPA alone, endovascular therapy alone, and combined treatments, as does the role of IA fibrinolysis as an adjunct to mechanical embolectomy.

Other charges regarding reperfusion therapies from the 2006 SPRG remain relevant. The development of novel lytics with better safety profiles remains a desired goal. Combining lytics with other anticoagulant and antiplatelet agents to open arteries more often and more quickly and to prevent reocclusion needs further study. Hemorrhage protection agents would also be a boon to the field. In addition, ultrasound enhancement of thrombolysis continues to show promise, and development of an operator-independent delivery method to make this strategy more widely applicable is an important priority.

One new question that is a priority for the field of acute stroke reperfusion therapies is whether the eligibility criteria may be beneficially broadened for currently available reperfusion therapies. In particular, population-based studies suggest that 15% of strokes that present to emergency departments in the US are identified upon awakening from sleep. A study of IV rtPA in this group is occurring in the NINDS SPOTRIAS-funded MR Witness Trial, and study of endovascular therapy in this setting is a consideration. Studies also suggest that up to half of patients otherwise eligible for IV rtPA are not treated due to perceived mild deficits, and 1/3 of these patients may experience disability. These frequent causes for exclusion from reperfusion therapies need definitive studies to determine their appropriateness. Additionally, the safety of IV rtPA for children will be studied by the recently NINDS-funded TIPS study.



1) The most fundamental unresolved question is whether neuroprotection can be an effective treatment strategy in acute ischemic stroke, complementing and potentiating well-established reperfusion therapy.

2) The practicability of trials testing combinations of neuroprotective therapies also remains challenging. Factorial design trials would allow disambiguation of agent-specific side effects, but require a degree of collaboration across multiple pharmaceutical sponsors, investigators, and regulators not readily achieved. Government leadership could prove transformative.

New questions that have emerged in the last five years include:

1) Is intra-arterial delivery of neuroprotective agents an effective strategy? IA neuroprotection theoretically allows delivery of agent to the center of the ischemic zone in high concentration, not just to the periphery in low concentration, and would permit direct intervention for reperfusion injury following endovascular recanalization therapy. As endovascular recanalization treatment centers have spread, the infrastructure for IA neuroprotective therapy is growing rapidly.

2) Are agents targeted at reperfusion injury beneficial? For the last 30 years, the focus of clinical neuroprotection trials has been in conferring cytoprotection during the initial period of ischemia. Reperfusion injury was not an urgent concern as reperfusion occurred infrequently. Now that reperfusion is achieved with increasing regularity, the clinical impact of reperfusion injury and its possible amelioration by agents targeting reperfusion-specific mechanisms of cell death becomes relevant.

3) Are intra-ischemic and post-ischemic conditioning potentially beneficial? Inducing intermittent ischemia in peripheral limbs or the brain itself has recently demonstrated potent neuroprotective effects in some animal models and is readily achievable in humans, but dosing and intensity schedules need to be developed and refined.

Stroke Centers

Despite the marked increase in the number of certified PSCs, a large portion of the US population still live considerable distances from PSCs and centers that offer endovascular therapy. Strategies for dealing with patients who live in limited access areas need to be devised. And while telestroke networks have increased access to specialized care for many patients in such areas, there are currently no procedures in place to measure adherence to protocols at spoke sites. Further, the integration of care across PSCs and CSCs needs to be addressed. When is it appropriate to refer patients from a PSC to a CSC, and does retention of patients at PSCs impact research being done at CSCs? Finally, if endovascular therapy is to be considered a standard of care, transfer delays to stroke centers are a major issue for the delivery of such therapy.

Pediatric Ischemic Stroke

1) We urgently need clinical trial network infrastructure funding for RCTs. For example, studies of acute anticoagulation in neonatal cerebral sinovenous thrombosis are needed and have been proposed to the NETT. This could leverage the existing initiative ‘NETT’ by giving ‘teeth’ to the request for pediatric involvement ideally leveraging funding trial centers in the IPSS, but will also require new funding initiatives.

2) We need studies to improve timely recognition and specific diagnosis of stroke in children. Without this we will never be in a position to study hyper-acute treatments. We need to develop tools to enhance education of frontline providers (ED providers, EMTs) and tools that can be used by these providers to identify children who are having strokes, and more easily distinguishing stroke from stroke mimics. These tools should include quick/simple bedside exam methods, serum markers, and bedside cerebral perfusion tests (optical imaging, TCD, fast urgent neuroimaging protocols: MRI first given that CT misses 50% initially in children).

3) A very urgent concern is the need for studies and guidelines to ensure safety data and selection criteria prior to implementation of endovascular acute stroke devices (Merci, Penumbra etc) in young children (e.g. <10 years age) with stroke. These devices are currently used in young children in centers without pediatric stroke expertise and without outcome reporting despite very significant risks and adverse outcomes.

4) There is a need to identify and aggressively treat seizures that are symptomatic of acute stroke in children. We don't know how common or severe such acute seizures are, particularly sub-clinical seizures. We need to evaluate the contribution of electrophysiologic monitoring during acute stroke in neonates and children. Other acute neuroprotective measures need refinement and validation including blood pressure, temperature, blood glucose and oxygenation parameters.

5) Hypothermia for acute neonatal stroke is a potential therapy that would reduce adverse outcomes.

6) Placental banking for 5 days post-natally in large birthing centers would enable over time, collection of valuable tissue for understanding the cause of neonatal stroke.

7) Animal Models for juvenile, not just neonatal, stroke require development. These would enable evaluation of stem cell therapies, acute and sub-acute vasculopathies and endothelial-vessel wall-coagulation interactions.

Emergency Department and Prehospital Investigators and Protocols

The original agenda to build an infrastructure for Emergency Department stroke research investigators and protocols has generally been achieved. Sustaining and capitalizing upon this framework is the next task.

New research opportunities include:

1) Prehospital care: further work is needed to determine optimal treatment of blood pressure and glucose in the prehospital setting. In addition, there is little data regarding the benefit or harm of supplemental oxygen in acute ischemic stroke. Further testing of neuroprotective agents in the prehospital phase beyond FAST-MAG must be explored. There is a need for better diagnostic tools in prehospital care to distinguish stroke from non-stroke, probable ischemic from likely hemorrhagic stroke, and probable large artery from likely small artery ischemic stroke, to optimize ambulance routing.

2) Emergency phase of care: as in prehospital care, there is the need to better define optimal treatment of blood pressure and serum glucose in the emergency department as well as determining whether supplemental oxygen is beneficial or harmful. Determining which patients are best treated with intravenous fibrinolysis alone, endovascular recanalization therapy alone, or both combined will be an increasingly urgent issue.

3) TIAs: there is the need for data to determine the optimal evaluation, treatment and disposition for patients seen in the emergency department with transient ischemic attacks, what testing is indicated acutely, and whether patients can be safely discharged, admitted or observed.

4) There is a need to additionally target NIH funds towards the pre-hospital and hyperacute aspects of stroke care inherent to emergency medicine.


Although telestroke is used widely, and evidence has shown Class I support for its use in medical decision-making, the long term outcomes of patients evaluated by telemedicine has not been fully shown in randomized clinical trials. Published data has shown good outcomes, but the ultimate assessment of telemedicine's efficacy in stroke will include both an increased number of patients receiving thrombolytic therapy and also scientifically assessed good long term outcomes.

Additional areas of future assessment and clarification may include streamlining credentialing of telemedicine providers, facilitating reimbursement, clarifying legal responsibilities, determining telemedicine's role in stroke center designation, assessing the use of telemedicine for clinical trial enrollments and utilizing telemedicine in various systems of care models.

Telemedicine could become a tool for clinical research trial enrollments. NINDS should foster research to assess the reliability of telemedicine for clinical trial recruitment and enrollment. Telestroke is also now more frequently considered for use in the pre-hospital field as well. With cell phone technologies and site-independent telemedicine systems now firmly in place, NINDS should foster the development of stroke care models outside of the traditional ED or inpatient arena. Data access is moving toward ubiquity, with use of mobile phones and wireless signals - data, radiology, and video images are now being evaluated on laptops, tablets, and even smartphones. NINDS should again serve as a barometer to help set technical standards and ensure that scientific rigor is applied to guide fuller dissemination of these technologies.

Financial Barriers

Unfortunately the revised DRGs for acute stroke patients treated with thrombolysis did not account for the relatively common situation of a patient treated with IV tPA at a community hospital and then transferred to a stroke center for post-tPA care. Despite providing the same post-tPA care at higher cost at the stroke center, hospitals receiving these “drip and ship” patients do not qualify for the DRG payment for thrombolysis because tPA was given at the originating site. This inequity discourages stroke centers from creating a referral network or stroke system of care to facilitate rapid treatment at community and rural hospitals. Additional research is necessary to document the growing number of drip and ship patients and the impact of this issue on the development and structure of stroke centers.

The higher reimbursement for acute stroke patients may also have a detrimental effect on recruitment into acute stroke trials. Patients entered into a trial may have otherwise received mechanical thrombectomy or a thrombolytic agent qualifying the hospital for a higher DRG payment than the reimbursement associated with a clinical trial. In view of the slow recruitment experienced by most acute stroke randomized trials, examination of this potential conflict might lead to viable solutions for stroke centers that benefit hospitals, investigators and patients.

Reimbursement of physicians caring for acute stroke patients also requires further study. Many stroke physicians believe that current coding system does not permit adequate reimbursement for the time and expertise delivered at the bedside in evaluating and treating patients with acute stroke. When treating a patient at the bedside, critical care codes can be charged based on total time spent caring for a critically ill patient. Reimbursement for telemedicine evaluation is limited to E&M coding which is reimbursed at a lower rate and is currently only provided by CMS when the remote hospital is in a rural region, but often these hospitals are in underserved urban and suburban areas. In many cases, a stroke physician must leave the office or come in from home on an emergent basis and adequate reimbursement is essential to encourage physicians to participate in acute stroke care. Modeling of acute stroke care delivery models based upon greater physician reimbursement or improved telemedicine payments represents an important research opportunity.


While a potential gender-by-thrombolysis interaction has been found in analyses of randomized clinical thrombolysis trials, the pathophysiology behind such an interaction remains unknown. Despite the presence of studies showing gender differences in ischemic pathways and responses to treatment, sex differences in the clinical efficacy of neuroprotectants have not been systematically sought. To date sufficient numbers of women to explore these questions have not been included in clinical trials. Effects of race and ethnicity on acute stroke outcomes have not been investigated sufficiently. Barriers to acute stroke treatment such as those in the black population deserve study and appropriate action. Two studies raise the question whether a lower dose of alteplase may have a better safety profile and similar efficacy compared to usual dose alteplase in Asian acute stroke patients.

Ischemic Stroke Mechanisms

The pathology of the thrombus is not easily determined with current imaging prior to treatment. However, it is clear that some thrombi are more susceptible to biochemical fibrinolysis and others are not. Their microarchitecture and composition requires further detailed pathological study to determine whether some thrombi are particularly susceptible to intravenous fibrinolysis, some more responsive to other intravenous agents (e.g. platelet disaggregating agents), some especially responsive to combined intravenous regiments, and some most responsive to an endovascular approach. Correlation with stroke mechanism – e.g. mature thrombus from the left atrial appendage vs. fresh thrombus from an acutely ruptured atherosclerotic plaque at the carotid bifurcation – would be useful.

The widespread use of CT angiography and to a lesser extent MR angiography as a tool to decide about the use of thrombolysis requires validation and evaluation of utility. These methods clearly prolong the ‘door-to-needle’ time if imaging occurs prior to intravenous thrombolysis. Time to treatment is a therapeutic imperative. A time tradeoff implies utility of the diagnostic information but the utility is unknown. More study of acute neurovascular imaging would be useful so that this time trade-off can be better understood. The same applies to the utility of perfusion imaging.

Aspirin remains the only antithrombotic of established benefit in preventing early recurrent stroke. Trials of early generation anticoagulants, including unfractionated and low molecular weight heparin in all comers and in atrial fibrillation patients showed no net advantage over aspirin. Several newer anticoagulants and antiplatelet agents have recently become available and offer the possibility of seeking improved protection against early recurrence in trials with larger sample sizes.

Patients with minor stroke or TIA, or those who have responded very early to thrombolysis are at high risk of early recurrent stroke. Treatment depends upon mechanism but it remains unclear who is at the highest risk. Patients with atherosclerotic carotid artery disease at the bifurcation may be at the highest risk. The FASTER trial suggested that double antiplatelet therapy might be a reasonable strategy short term. However, it is absolutely clear now with MATCH and SPS3 interim results that the use of double antiplatelet therapy long-term (aspirin and clopidogrel) results in a higher rate of major hemorrhage, primarily gastrointestinal hemorrhage. The POINT trial will help to unravel this problem and recruitment should be encouraged.

Diagnostic and quality evaluation of new diagnostic tools and the timing of their use in acute stroke and the immediate few days thereafter are needed, including duration of cardiac monitoring, algorithms for echocardiography, and use of new cardiac MRI and CT imaging techniques.


1) Making Reperfusion Therapy Swifter, Safer, and Surer: Early recanalization currently is, and likely always will be, the most effective therapy for acute ischemic stroke. Present recanalization treatments are limited by infrequent achievement of complete reperfusion, long delays to reperfusion, and high rates of hemorrhagic complication. New drugs and devices, alone and in combination, are needed to enable cerebral reperfusion interventions to achieve rapid, complete and sustained vessel patency in all patients harboring salvageable tissue, with no risk of hemorrhagic transformation.

2) Cerebral Cytoprotection Early and After Reperfusion with Potent, Pleiotropic Interventions, and with Optimized Management of Physiologic Parameters: Cytoprotective therapies continue to hold promise as adjunctive treatment to reperfusion, and the most promising avenues for deriving benefit include initiation in the hyperacute 0-2 hour period to stabilize the penumbra until reperfusion, use of agents blocking reperfusion injury in the post-recanalization period, and study of single and combination agents that have demonstrated extreme potency and polymorphic effects in interrupting injury pathways. In addition, developing firm evidence for optimal protocols to manage blood pressure, glucose, oxygen, and other physiologic parameters would provide improved cytoprotective care to all acute stroke patients.

3) Refining and Leveraging Effective Acute Stroke Clinical Trial Networks: In the last decade, NINDS and other agencies successfully launched cooperative study groups to conduct acute stroke trials, including SPOTRIAS (NINDS, phase 2), NETT (NINDS, phase 3), and IPSS (pediatric). This established clinical trial infrastructure is poised to drive rapid, iterative advance in acute stroke care. NINDS should prioritize a review of these networks and then refine and optimize them based on both internal experience and experience in kindred disease states; such network reviews should drive to make translational treatment trials inexpensive, efficient, and successful.



Co-Chairs: Michael Chopp, S. Thomas Carmichael, Jack Parent

Members: Steven Cramer, Theresa Jones, Randolph Nudo, Sean Savitz, Gary Steinberg, Raghu Vemuganti, Zheng Gang Zang

NINDS Liaisons: Francesca Bosetti, Tim LaVaute


Three Priorities:

1) Identify critical stroke repair/restorative mechanisms at the molecular, cellular, tissue and systemic levels

2) Define and optimize restorative therapies, including cell-based, pharmacological, brain-machine interface, and brain stimulation approaches, based on mechanistic advances

3) Translate cell-based, pharmacological, brain stimulation and behavioral manipulations with associated biomarkers and neuroimaging from animal models to clinical trials.

New Stroke research opportunities, emerging topics, and unresolved areas since 2007

miRNA: miRNAs likely play important roles in mediating neurological recovery and inducing restorative processes.

Epigenetics: Investigate the role of epigenetic phenomenon that include DNA methylation, acetylation and histone functions after stroke.

Endogenous neurogenesis: The role of adult-generated neurons in repair after stroke remains unclear.

Cell transplantation: The biological mechanisms through with cell transplants act to promote repair should remain a scientific focus in order to optimize such therapies.

Coordination of vascular and neural circuit remodeling: A better understanding of spontaneous vascular structural remodeling process after stroke, the interdependence of activity-dependent circuitry remodeling, new vessel formation and vessel stabilization, and sources of variability in these responses (e.g., with age) are important for optimizing and timing restorative treatment strategies.

Diabetes and aging: Both the diabetic and aging brain may have a distinctive biology from the young brain and in recovery, but this area has not been not been investigated in detail.

Imaging recovery: There is a need to improve the MRI capabilities and other imaging modalities to non-invasively monitor brain recovery. Translation of results to humans is paramount, underscoring the value of imaging methods to provide direct comparisons between rodent and human brains.

Multiphoton and other live imaging approaches: Linking repair processes to behavioral function requires sophistication in behavioral as well as imaging approaches.

White matter injury and repair: The cellular mechanisms of white matter repair have not been identified.

Stroke repair biomarkers: Biomarkers for neural repair are needed for mechanistic studies and clinical trials.

Developing and diversifying stroke animal models: There is a major need to both better encompass the diversity of human stroke in animal models.

Electrical and magnetic stimulation: TMS and tDCS require a rigorous understanding of underlying neurobiological principles in order to be fully utilized in clinical populations.

Brain machine interfaces: Studies that focus on developing an understanding of the neurobiological effects of such artificial communication bridges are needed to advance this technology to stroke and other neurological conditions.


Glia mediate recovery

A major research advance which can impact the development of therapy for stroke designed to promote recovery is the finding that cell-based therapies for the treatment of stroke not only produce neurotrophic and other factors but that these cells stimulate parenchymal cells to remodel the brain. Attention is focused on the astrocyte, the most numerous brain cell. Glia are highly sensitive to injury and responsive to changes within the brain microenvironment. Cell therapies have been shown to activate astrocytes, as well as other parenchymal cells and subsequently these stimulated parenchymal cells initiate a set of interacting restorative processes, involving the production of factors that amplify endogenous restorative processes such as angiogenesis, neurogenesis, neurite outgrowth and remyelination, which act in concert to enhance neurological function. Astrocytes may also serve as multipotent progenitors when reactive after stroke. Data from multiple publications in both cell and pharmacological restorative therapies also highlight the interactions present in the coupling of neurogenesis, angiogenesis/arteriogenesis and neuronal remodeling.

The dichotomous role of microglia in the reparative phase after stroke has been underscored by work performed since 2007. Microglia amplify brain injury after ischemic insults in the acute phase by secreting various pro-inflammatory cytokines and activating the innate immune response. These effects may also suppress subsequent endogenous reparative mechanisms by inhibiting neural stem/progenitor cell proliferation. Recent studies in the chronic phase after stroke, however, suggest that microglia support potential reparative processes. Such effects may occur by stimulating the neuronal differentiation and survival of adult-generated cells via secretion of insulin-like growth factor-1 or other trophic factors. The above findings are consistent with Priority 1 noted in the 2002 SPRG Report  -understanding the molecular, cellular and network changes in brain that lead to good outcome.

CNS and systemic responses to restorative stroke therapy

Treatment of stroke with neurorestorative therapies, particularly with cell-based therapies, has also shown widespread and systemic indices of plasticity and multi-organ responses. Stroke induces neurite outgrowth in the cortical spinal tract, corticorubral tract, spinal cord and the contralateral cerebral hemispheres. With treatment with a restorative cell-based therapy, this neuronal remodeling is substantially amplified and significantly correlated to functional improvement, indicating that rewiring that occurs throughout the CNS post stroke and that recovery of neurological functions, particularly motor function, can be accomplished by stimulating new connections in remote areas of the CNS. Stroke also produces a systemic immune response. Treatment with a cell-based therapy modifies immune responses and affects other organs, such as the spleen which alters cytokine release, and mesenchymal cells within the bone marrow in which gene expression is greatly altered. The beneficial therapeutic effects of cell therapy appear to be connected with stimulating other organs to evoke immune responses that may be beneficial to stroke recovery. This work has led to several Phase I/IIa clinical trials in stroke for MSC and other bone-marrow derived cell therapies as well as cell therapies from tissues such as umbilical cord and placenta. A major set of advances since 2007 has been in the cellular effects of these cell-based therapies, and in the determination of timing and dose, that have paved the way for these clinical trials. These findings are consistent with Priority 1 noted in the 2002 SPRG Report - understanding the molecular, cellular and network changes in brain that lead to good outcome.

Time window for optimal therapeutic effect

Studies ranging from longitudinal examination of gene expression to behavioral interventions in rodent models have accumulated over the past few years that suggest a relatively broad window for maximal therapeutic effects. While the optimal time window for acute neuroprotection is limited to several hours, recovery mechanisms, including upregulation of growth-promoting genes, axonal and dendritic sprouting, etc., occur most robustly over a period of several weeks. These fundamental neurobiological findings will ultimately drive the prescription for physical, cell-based and pharmacotherapeutic interventions in humans.

Imaging of neuroplasticity and recovery after stroke—MRI

Since 2007, there have been major advances in the ability to monitor recovery of neurological function non-invasively using MRI. Improvement in neurological recovery was shown to be driven by vascular as well as neuronal remodeling. MRI also provides biomarkers for recovery and possibly the potential for recovery, as well as insight into the structural substrates that underlie recovery of function, with an emphasis on axonal and remyelination changes. Imaging of tissue repair as potential translation markers of tissue remodeling that can be (and some already have been) translated into the clinic include imaging vascular remodeling (MRI mean microvessel density, diameter, mean segment length, susceptibility weighted imaging (SWI) , blood brain barrier (BBB) transfer constant angiogenesis, and imaging of neuronal remodeling (diffusion tensor imaging - DTI and QDTI), neurite density, stem-cell migration and distribution, and Q-space fiber tracking. These imaging processes are applied to both preclinical and clinical conditions after stroke. These studies are consistent with Priority 2, to develop neuroimaging and other methods for detecting molecular, cellular, synaptic and circuit mechanisms of recovery.

Imaging of neuroplasticity and recovery after stroke—Multiphoton/PET

Since 2007 advances in in vivo multiphoton imaging have mapped the changes in neuronal connections and dendritic architecture that occur after stroke and are associated with neurological recovery. Stroke triggers dendritic sprouting, changes in dendritic spines and, by extension, alterations in the connections of peri-infarct tissue. There is an initial loss of connections in peri-infarct tissue, and then a rebound increase in dendritic spines that indicates a net synaptogenesis. The map of sensory function in stroke follows these changes in local neuronal architecture, with an initial loss of responsiveness of peri-infarct tissue, and then a remapping of sensory function in new locations. These data, coming from multiple labs and consistent across groups, indicate stroke alters the circuitry and cognitive mapping of function in peri-infarct tissue in a way that promotes new functional networks and recovery. This remapping of cognitive function in peri-infarct tissue in experimental animals parallels the new functional maps seen in human imaging during stroke recovery and establishes structural targets for neural repair therapies. Multimodal imaging using micro PET has also been employed to image in vivo gene expression and to visualize/quantitate recovery processes such as angiogenesis (radiolabeled VEGF receptors), metabolic activity, receptor-ligand interactions, neurotransmitter binding, as well as tracking stem cell transplant fate and function. These studies are consistent with Priority 2, to develop neuroimaging and other methods for detecting molecular, cellular, synaptic and circuit mechanisms of recovery.

Molecular mechanisms of neural repair after stroke

Studies since 2007 have identified key molecular systems that mediate tissue reorganization and repair after stroke. These tissue responses include axonal sprouting from neurons that survive in both peri-infarct cortex and from contralateral cortex into spinal cord. A molecular program that mediates post-stroke axonal sprouting in neuronal circuits within peri-infarct cortex has been identified: a sprouting transcriptome. Similarly, the transcriptome of neurons in contralateral cortex from the stroke that sprout a projection into the cervical spinal cord has been identified, allowing a comparison of molecular “regeneration” programs after stroke. These two gene programs share similarities and distinctions indicating that neuronal circuits respond in different ways on a molecular level to stroke during neural repair, and provide possible targets for drug therapies. Several molecular pathways within the tissue responses for reactive astrocytosis and angiogenesis, both in the normal state and after a cell therapy, have been identified. These include tissue remodeling and signaling systems, such as tPA and ADAMTS, and local growth factor production, such as with a BDNF-angiogenesis link. These studies initiate a molecular identification of neural repair pathways that relate to Priority 1 noted in the 2002 SPRG Report - understanding the molecular, cellular and network changes in brain that lead to good outcome.

Stunned neuronal circuits in peri-infarct cortex

Human and animal imaging studies in stroke recovery indicate that remapping of sensory and motor function within peri-infarct cortex is closely linked to functional recovery. Physiological and molecular studies since 2007 have identified that neuronal circuits adjacent to the stroke have altered excitatory and inhibitory neurotransmission and distinct distributions of growth factors. In experimental animals, the peri-infarct motor cortex is over-inhibited by an increased tonic GABA current. This current can be reversed pharmacologically to promote greater motor recovery. Excitatory signaling within this same region is altered after stroke. Neurons are more sensitive to glutamate-induced BDNF production: activation of the glutamate AMPA receptor induces BDNF within these peri-infarct circuits and not other brain regions after stroke. The potent vascular and neuronal growth factor VEGF redistributes to peri-infarct motor neurons during recovery in these same areas. These findings of altered inhibitory and excitatory signaling and growth factor distribution in tissue adjacent to the stroke provide pharmacological targets for stroke recovery, and relate to Priority #1 of the 2002 SPRG Report. These findings also harmonize with the multiphoton brain imaging data noted above to suggest cellular mechanisms for the initial loss of responsiveness of peri-infarct tissue in brain maps, and then the remapping of sensorimotor function during recovery.

Behavioral mechanisms of brain repair

Behavioral experience potently modulates neural and non-neural cellular responses after stroke. It can impact axonal sprouting, dendritic growth and cell proliferative responses, as well as new cell survival, astrocytic reactions and numerous other cellular events. This is relevant not only for designing rehabilitation strategies and their combination with other therapies, but also for understanding normal brain repair mechanisms, which can be expected to be shaped by ongoing behavioral activity. A natural response to the lateralized motor impairments common after stroke is to learn to rely on the unaffected body side. Research in animal models indicates that this compensatory learning can drive dendritic and synaptic growth in the uninjured hemisphere while simultaneously reducing neural remodeling events in peri-infarct motor cortex that mediate functional improvements in the stroke-affected upper extremity. Thus, some experiences promote plasticity that subverts more functionally beneficial restorative responses. Clinical research findings indicate that interhemispheric activity patterns can become unbalanced after stroke and contribute to dysfunction in the affected upper extremity. Suboptimal experience-driven plasticity and disruptive interhemispheric activity may need to be overcome with behavioral interventions (e.g., physical therapy, robotic therapy) and other strategies (e.g., cortical stimulation) to optimize treatment strategies. Behavioral training focused on stroke-affected extremities is known to remodel functional circuits in peri-infarct motor cortex. Such training can promote large scale neuroanatomical rewiring, as evidenced by axonal sprouting between major subdivisions of motor cortex. These findings are consistent with Priority 1 of the 2002 SPRG Report to understand molecular, cellular and network changes in the brain that lead to good versus poor behavioral outcome.



miRNAs are 22 nucleotide long RNA molecules that act as post-transcriptional regulators. The miRNAs silence mRNAs by binding to specific 8-nt complimentary seed sequences in their 3’-UTRs. The current view is that miRNAs play a more diverse role than just silencing mRNAs as certain miRNAs can also upregulate translation as well as transcription. Binding of certain miRNAs to promoters activate gene expression. This phenomenon known as RNA-induced gene activation may play an important role in post-ischemic gene induction. Furthermore, miRNAs are emerging as co-master controllers as they can control as well as be controlled by transcription factors, DNA methylation and histone deacetylation. Thus, miRNAs and other epigenetic phenomenon can control each other bi-directionally and hence their relationship is complex. In the developing brain, neural miRNAs are involved at various stages of synaptic development, including dendritogenesis, synapse formation and synapse maturation. There are emerging studies suggesting that miRNAs likely play important roles in mediating neurological recovery and inducing restorative processes like neurogenesis, angiogenesis and axonal remyelination, among others. Each miRNA can bind to and regulate many mRNAs thereby acting as a master switche for the regulation of gene translation.

Greater than 98% of the transcriptional outcome is represented by non-coding RNAs that include miRNAs. The significance of other non-coding RNAs including, but not limited to piRNAs, lncRNAs, scaRNAs, rasiRNAs, tasiRNAs in controlling post-ischemic pathophysiology needs to be understood to use in guiding plasticity. For example, piRNAs play a major role in controlling transposons, and modulating them can also significantly influence stem cells.


While other fields like cancer have extensively studied the role of epigenetic phenomenon that include DNA methylation, acetylation and histone functions, not many studies on these important cell regulators in the stroke field are available. Understanding these phenomena in relation to other controllers like non-coding RNAs (in addition to miRNA function noted above) and transcription factors will significantly help advance the understanding of post-stroke plasticity and regeneration.

Endogenous neurogenesis

The role of adult-generated neurons in repair after stroke remains unclear. Subventricular zone neural stem/progenitor cells are activated by ischemic insults, produce increased numbers of immature neurons that migrate to injured striatum and perhaps cortex and, at least in the striatum, these differentiating neurons appear to take on the phenotypes of neurons damaged by the insult. Although these findings have been replicated in multiple models, critical challenges and questions remain. A major issue is that very few of the newly differentiated neurons survive and integrate. In fact, unequivocal integration of adult-generated neurons after stroke remains to be shown in any model. Another question is whether these cells contribute to recovery, and if so, whether the mechanism is neuronal replacement or stimulation of other reparative pathways. Moreover, it will be critical to determine whether stimulating the survival and integration of adult-generated neurons will restore function. The advance of transgenic mouse models to track adult-generated neurons and manipulate their signaling pathways should allow these questions to be answered in the next few years.

Cell transplantation

Many different cell types, primarily stem or progenitor cells, have been used to examine the role of cell transplantation in regenerative therapy for stroke. Despite this work, the optimal cell types, differentiation states, timing after stroke and grafting methods (including cell encapsulation, use of scaffolds and related technologies) for stroke therapy are unknown. The biological mechanisms through which cell transplants act to promote repair should also remain a scientific focus in order to optimize such therapies. In addition, the ability to generate patient-specific neural progenitors and neurons from adult somatic tissues via induced pluripotent stem cell reprogramming or direct transdifferentiation raises new possibilities for autologous grafting that need to be explored. Both efficacy and safety are important issues to address. The mechanisms, not aimed at cell transplantation per se, underlying how some types of cell therapies from non-neural tissues enhance recovery require further investigations. For example, the biological mechanisms that link how cell therapies promote recovery through their effects on peripheral organs and immune tissues need to be elucidated. What are the key paracrine, endocrine, and neural mechanisms that explain how some types of cell therapies when given systemically restore neurological function after stroke without penetrating the CNS in large numbers?

Coordination of vascular and neural circuit remodeling

Prolonged reductions in cerebral blood flow and vascular densities may occur well outside of the ischemic core, which conceivably creates an insufficiency in blood supply needed to support neural activity dependent circuitry remodeling. There is now strong evidence supporting the importance of angiogenesis in functional outcome. The expression of angiogenic factors is increased by ischemia and positively correlated with outcome in stroke patients. Besides angiogenic factors, many other treatments that promote neural plasticity and improve function, such as exercise and cortical stimulation, also promote angiogenesis. However, vascular responses are multiphasic and unfold over long time spans after stroke, and there is also considerable variability in their manifestation. A better understanding of spontaneous vascular structural remodeling process over long periods after stroke, the interdependence of activity-dependent circuitry remodeling, new vessel formation and vessel stabilization, and sources of variability in these responses (e.g., with age) is likely to be important for optimizing and timing restorative treatment strategies.

Diabetes and aging

Diabetes is a major risk factor for stroke and the diabetic brain responds differently to stroke with more severe damage and possibly reduced neurological recovery. Similar observations are present in the aged brain. For example, aging results in both a decline and altered time course of poststroke angiogenesis. The temporal profiles and the events associated with recovery from stroke have not been sufficiently investigated in the diabetic and aged brain. Studies since 2007 indicate that both the diabetic and aging brain may have a distinctive biology from the young brain and in recovery, but this area has not been not been investigated in detail. Preclinical studies in the response to stroke and restorative therapy in type 1 and type 2 diabetic brains and in both young and aged animals are warranted.

Imaging recovery

There is a need to improve the MRI capabilities to non-invasively monitor brain recovery. Although great strides have been made in imaging vascular and neuronal remodeling, there is a compelling need to develop more quantitative imaging of these processes in both animal and human brain. An additional area that warrants investigation that can have a major impact on the treatment of stroke as well as neurodegenerative diseases is functional connectivity mapping using functional MRI (fMRI). Advanced fMRI methods have been developed to investigate the interactions between brain regions and to investigate changes in brain connectivity after stroke. In addition to fMRI, there is a need to further develop DTI-based anatomical connectivity mapping which provides the capability to monitor structural changes in the  brain that are altered after stroke and that are remodeled in response to stroke and recovery of neurological function.

Improved functional recovery after transcranial magnetic stimulation (TMS) treatment corresponds to increased effective connectivity between primary motor cortices, and between primary motor cortices, the basal ganglia, and the motor nuclei of the thalamus, measured by positron emission tomography (PET). Stroke patients, compared with healthy controls, exhibited decreased intrinsic neural coupling between primary motor cortex and cerebellum corresponding to dysfunctional motor cortex and cerebellum connection, increased coupling from supplementary motor area to primary motor cortex and from supplementary motor area to cerebellum, suggesting that changes in supplementary motor area and cerebellum connectivity may occur to compensate for a dysfunctional primary motor cortex. Further work in this area will provide insight into the brain network response to stroke and how this network  interaction is altered during recovery.

Multiphoton and other live imaging approaches

The newer and still emerging high resolution imaging approaches, such as multiphoton imaging, are likely to continue to rely predominantly on mice (because of YFP and GFP transgenics and optogenetics). However, mouse models lag far behind rats in behavioral methodology, stroke model refinement and systems-level neuroscience foundation (e.g., cortical organization and intrinsic connectivity). This is a major issue because sensitively linking repair processes to behavioral function requires sophistication in behavioral as well as imaging approaches. There is a major need to refine mouse models of chronic impairments that resemble those common in stroke, as well as a need to improve outcome measures sensitive to these impairments. There also is a need to bridge multiphoton approaches with those that are more feasible in other animal models and in humans, and to extend imaging capacity to deeper brain regions.

Validation of non-invasive imaging data

There has been a rapid proliferation of non-invasive imaging data demonstrating changes in functional activity and connectivity in human brain. While analogous animal studies are more limited, they offer the advantage of pre-post examination, and perhaps most importantly, post-mortem validation of connectivity patterns. However, such validation of imaging approaches used in humans has rarely been done in animal models. This is a critical area for exploitation in the near future.

White matter injury and repair

White matter stroke in humans constitutes approximately 20% of all strokes. This type of stroke produces focal neurological deficits, such as weakness or numbness, but also accumulates to produce vascular dementia. In addition, white matter injury is a component of large artery stroke. The cellular mechanisms of white matter repair have not been identified. These repair mechanisms are distinct from those studies in traditional small animal stroke models, and relate to oligodendrocyte progenitor responses to stroke, the integrity of injured axons, and the communication of axons and surrounding glia in what has been termed the axoglial unit. Are there distinct mechanisms underlying oligodendrocyte death? Is there a different time course compared with ischemic neuronal death? Is there a surviving peri-infarct region of partially damaged axons? Oligodendrocyte progenitors proliferate after stroke—can they differentiate into oligodendrocytes and repair damaged white matter? Can they repair damaged white matter even in the progenitor state?

Stroke repair biomarkers

Clinical trials in many neurological diseases benefit from a biomarker that indicates the presence of a target process, limiting sample size for clinical trials and allowing more detailed mechanistic studies of the disease process. Studies in multiple sclerosis have benefited from imaging biomarkers, and to a certain extend MRI imaging sequences have allowed definition of the stroke core and penumbra that might serve as biomarkers in acute stroke/neuroprotection trials. A similar biomarker or biomarkers for neural repair are needed for mechanistic studies and clinical trials. These might relate to brain imaging modalities. Repair and recovery in humans after stroke involves plasticity in cognitive maps, for example in the cortical motor hand map. This plasticity in the ipsilesional hemisphere is strongly correlated with recovery. Can fMRI, TMS, MEG or other non-invasive imaging or mapping techniques be utilized to develop a neuroplasticity biomarker after stroke that would allow smaller sample sizes and more rapid Phase II clinical trial screening of candidate stroke recovery compounds? Studies since 2007 have shown that recovery after stroke is associated with angiogenesis, in a way that likely indicates links between angiogenesis, neurogenesis and other aspects of cellular reorganization. Can an imaging modality for tissue reorganization, such as angiogenesis, be included in functional network mapping to produce a biomarker of functional and tissue reorganization after stroke?

Recent studies showed that RNAs in blood are very stable and can serve as biomarkers to differentiate different disease states. In particular they can be used to identify various subtypes of stroke. A combinatorial approach to identify gene expression changes together with the presence of different subtypes of non-coding RNAs, including microRNAs, in blood can be a powerful approach to understanding the dynamics of stroke repair. As obtaining blood samples is a routine practice and RNA analysis with microarrays and real-time PCR is well-developed, this is a strategy worth considering.

Developing and diversifying stroke animal models

Chronic stroke represents a diverse class of brain injuries and behavioral deficits. There is a major need to both better encompass this diversity in animal models and to investigate the mechanisms of its contribution to variability in treatment responses. There is an unresolved need to better refine stroke models for these purposes. To study this diversity in repair and recovery processes requires strong discourse among clinical and basic researchers and may require some disentanglement of model refinement priorities from those of stroke neuroprotection. There is a major disconnect between animal models of stroke and the clinical condition of human stroke. It has been widely recognized, and cited in each of the SPRG reviews that the majority of studies in the pre-clinical and basic science literature are conducted in young, healthy, male rats. However, human stroke occurs in elderly, diabetic individuals of both genders. There has been an increase recently in the number of studies in aged rodents. This is a step in the right direction. However, information on the effects of these individual variables alone will not necessarily lead to better models of human stroke. It is important to understand whether these variables are even relevant in translating findings to humans. Does matching age, co-morbid condition and gender to the human population experiencing stroke provide any more predictability in treatment effects?

A related issue regarding animal models is the influence of scientists who have sought to develop standards for pre-clinical trials in animal models in order to eliminate bias. What influence has this had on the literature, and how should scientists go forward in improving both pre-clinical trials and basic science studies? Are the models for studying acute stroke pathophysiology equally valuable for studying neural repair mechanisms? A particular model or technique for infarct induction that doesn’t replicate the acute conditions of human stroke, may be very valuable for understanding repair strategies.

Electrical and magnetic stimulation

Several modalities of stimulation, including epidural, subdural, TMS and tDCS have risen in popularity as a potential therapeutic adjunct to traditional rehabilitation approaches. To date, relatively little is known regarding the neurobiological effects of electrical and magnetic stimulation in inducing plastic change in brain tissue. This promising form of therapy requires a more rigorous understanding of underlying neurobiological principles in order to be fully utilized in clinical populations.

Brain machine interfaces

Brain machine interfaces (BMIs), or electronic interfaces between intact brain regions and other parts of the nervous system, have been implemented in animal models and a few case studies in humans in the past few years. While still in early stages, they have demonstrated that spike discharges and local field potentials can be used to drive activity in prosthetic limbs, muscles and spinal cord. In the majority of these studies the focus has been on spinal cord injury. However, this technology is advancing rapidly, and applications for stroke recovery are likely to emerge soon. Studies that focus on developing an understanding of the neurobiological effects of such artificial communication bridges are needed to advance this technology to stroke and other neurological conditions.


1. Identify critical stroke repair/restorative mechanisms at the molecular, cellular, tissue and systemic levels

  • axonal and dendritic sprouting, neurogenesis, angiogenesis, epigenetic and small RNA alterations, remyelination and white matter repair, glial activation, altered network activity, contralateral plasticity
  • biomarkers to monitor effects of manipulations on recovery (e.g., imaging neural plasticity and repair, neuroplasticity and repair)

2. Define and optimize restorative therapies, including cell-based, pharmacological, brain-machine interface, and brain stimulation approaches, based on mechanistic advances

For cell therapies, investigate and identify:

  • cell types, physical properties of grafted cells and associated material/scaffolds, optimal non-invasive monitoring of graft or endogenous response
For all restorative therapies, investigate and identify:
  • Therapeutic windows and timing of restorative intervention

3. Translate cell-based, pharmacological, brain stimulation and behavioral manipulations with associated biomarkers and neuroimaging from animal models to clinical trials.

  • identify animal stroke models particularly useful for studying neural repair, specifically investigate animals models associated with elevated stroke risk and morbidity/ mortality, including ageing, diabetes and hypertension
  • optimize behavioral and cognitive outcome indices of recovery
  • identify optimal therapeutic window for intervention
  • support preclinical safety and efficacy data gathering



Co-Chairs: Costantino Iadecola, Marilyn Cipolla, Frank Faraci

Members: Nabil Alkayed, Robert Bryan, Turgay Dalkara, Donna Ferriero, Edith Hamel, Zvonimir Katusic, Raymond Koehler, Charles Leffler, Chris Schaffer, Danica Stanimirovic

NINDS Liaisons: Thomas Jacobs, Erik Runko


The endothelium has emerged as a key player not only in the regulation of vascular tone, but also in maintaining the structural and functional integrity of the neurovascular unit (NVU). As a major site of end-organ damage in cerebrovascular disease, it represents a key target for prevention, as well as therapy and recovery of function. Understanding the molecular bases of the interaction between endothelial cells and other cells within the vessel wall and NVU may harness the broad protective potential of the endothelium.

Progress has been made in understanding the structure and function of vascular smooth muscle (VSM) and pericytes, but fundamental questions concerning their normal and pathophysiological roles remain. For example, the mechanistic bases of autoregulation and the role of newly discovered ion channels remain elusive. Excitation-transcription coupling in VSM and pericytes and their interaction with neurons and astrocytes needs to be better understood. New technologies to investigate large and small vessel function in health and disease will advance this area further.

Advances in neurovascular and gliovascular signaling in health and disease have occurred as a result of more precise structural and functional in vivo imaging of neurons, astrocytes and cerebral vessels. These newer technologies need to be applied to unravel the mechanisms of neurovascular dysfunction in cerebrovascular pathologies. Protecting the NVU as a whole may limit brain damage and enhance repair in stroke, as well as in Alzheimer’s dementia and other neurodegenerative conditions.

Progress has been made in understanding blood-brain barrier (BBB) formation and maintenance, transporters and ion channels, but the mechanisms of BBB disruption by risk factors and cerebrovascular diseases remains understudied. Knowledge gaps include a lack of detailed understanding of molecular and functional changes in BBB in disease, and strategies to protect and repair the BBB. Novel approaches for drug delivery across the BBB, as well as BBB models that better mimic the in vivo situation, should promote new treatments for stroke and other neurological diseases.

A greater level of complexity has emerged in the vascular damage caused by stroke risk factors and cerebrovascular diseases, centered on the interactive role played by immunity, inflammation, and oxidative stress. New efforts need to define cell-specific pathways and molecular mechanisms of disease, as well as the pathogenic interactions between cell types in large and small vessels. Models that take into account risk factors, gender and age, as well as new genetic, molecular and imaging tools are needed.

Challenges remain in understanding sex-specific mechanisms, course, and response to therapy of cerebrovascular disease at all ages. Knowledge is still lacking in understanding the cerebrovascular correlates of the menopausal transition and the impact of hormone replacement therapy. Clinical studies on the predisposition to preeclampsia with a focus on cerebrovascular disease and stroke risk factors should be conducted in parallel with the development of relevant animal models.

Understanding normal cerebrovascular regulation and the pathological alterations induced by asphyxia/ischemia in the developing brain is critically important for prevention, diagnosis, and treatment in fetal, neonatal, and childhood hypoxic-ischemic events. More knowledge is needed on the pathological events triggered by asphyxia-ischemia and stroke in newborns and children, and how the causes, effects, and healing processes may differ from the adult. In pediatric stroke, efforts should be directed at developing pre-clinical models so that mechanisms and risk factors can be studied more appropriately.



Continued progress has been made in defining the paracrine signaling originating from endothelial cells. The endothelium-derived prostanoid prostacyclin (PGI2) is vasoprotective by preventing thrombosis, atherogenesis, hypertension, wall stiffening, and pathological remodeling after vascular injury. Clinical experience with both selective and nonselective inhibitors of cyclooxygenase (COX) isoforms indicated that preservation of PGI2 biosynthesis is critical for vascular health, and the adverse effects associate with COX2 inhibition have been attributed to loss of PGI2 resulting in enhanced effects of thromboxane A2. Progress has been made in endothelial nitric oxide (NO) synthase (NOS) biology, including new insight into regulation of endothelial NOS (eNOS) expression, post-translational modifications, subcellular trafficking, substrate and cofactor availability, and interactions with other regulatory molecules, such as the endogenous eNOS inhibitor ADMA. One of the more significant advances has been the realization that eNOS has many effects beyond regulation of vascular tone. eNOS has key effects on angiogenesis, formation of collateral vessels, adaptations to changes in blood flow, neuronal transmission, neurogenesis, oligodendrocyte function, and processing of amyloid precursor protein. The loss of trophic function of the endothelium may contribute to the impaired neurogenesis that occurs with aging and other stroke risk factors. Thus, endothelial dysfunction, in addition to its deleterious vascular effects, also promotes the progression of neurological injury and impairs recovery following stroke due to loss of eNOS-dependent signaling. The cerebroprotective effects of exercise are linked to eNOS, although the signaling pathways remain unclear. A third endothelial vasodilator mechanism exists which is independent of COX or eNOS and involves primarily intermediate and small conductance endothelial Ca++-activated K+ channels (endothelium-dependent hyperpolarization or EDH) This mechanism is more prominent and constitutive in smaller arteries and arterioles, and seems to be enhanced following stroke. Thus, EDH may be particularly important in local regulation of cerebral blood flow (CBF) and in maintaining CBF during pathological states.


Newly discovered ion channels [Na+ (ENaC), K+ (K2P), and non-selective cation (TRP and ASIC)] have significantly advanced our understanding of VSM function. This includes compartmentalization and functional domains within the vascular smooth muscle (VSM). The gain of Ca++ signaling in VSM can be regulated by protein kinases, and altered in hypertension, vasospasm, and ischemia. While there are similarities in the regulation of VSM in cerebral arteries and smaller cerebral arterioles, there are also striking differences in receptor/channel expression and function, making the myogenic response of particular importance to regulation of vascular resistance and blood flow within the brain parenchyma. TRP channels have been shown to be involved in the myogenic response and appear to be the mechanosensitive channels by which pressure depolarizes smooth muscles to elicit contraction. Along with structural determinants, regulation of the contractile state of VSM is the final step controlling cerebral perfusion. It is increasingly apparent that VSM, and probably pericytes, are not only influenced by neurotransmitters and paracrine/autocrine factors, but also through direct cytoplasmic communication via myo-myocyte and myo-endothelial gap junctions. Pericytes have received increasing attention in recent years. Pericytes communicate with endothelial cells by direct contact, gap junctions or through signaling pathways. In vivo studies suggest pericytes may play a role in the regulation of capillary flow in brain. These recently described functions add to known roles in angiogenesis, blood-brain barrier (BBB) stability, and macrophage-like activity. Continued progress has been made in the control and remodeling of VSM during pathological states. Like VSM, pericytes are affected by pathological states including stroke, where sustained contraction may lead to incomplete reperfusion/reoxygenation after recanalization of occluded arteries (“no reflow” phenomenon). Detachment of pericytes from the vessel wall can lead to loss of BBB integrity. Amyloid deposition is observed in degenerating pericytes, suggesting pericyte dysfunction may play a role in cerebral hypoperfusion and impaired ?-amyloid clearance in Alzheimer’s disease.

Neuro- and glio-vascular signaling

In line with previous SPRG recommendations, significant progress has been made on the cellular interplay between neuronal signals and their processing by the neuronal, astroglial and vascular compartments of the NVU. With the development of better imaging modalities, a shift from studies in brain slices towards in vivo models permitted visualization of regional blood flow, vascular diameter, and neuronal or astroglial activity. These advances have provided better insight into the spatial relationships of neurons and astrocytes with arterioles, capillaries, and venules in different brain regions, the temporal and spatial characteristics of columnar ascending and descending vasodilation, and the close relationship between astrocyte calcium signaling and hemodynamic changes. Imaging techniques have progressed to provide finer resolution of temporal and spatial changes in NADH redox state as well as in blood and tissue oxygenation. The neuronal populations recruited and the vasoactive mediators involved were shown to depend on the neurotransmitter released by the incoming afferents. Different subsets of interneurons in cerebral cortex seem to be activated with thalamocortical, corticocortical, and basal forebrain afferents. Moreover, whereas interneurons in the activated area and, particularly, in the cerebral cortex may instigate the changes in pyramidal cell depolarization and activity, the latter appears to drive the hyperemic response to sensory and other incoming stimuli. Some recruited neurons release vasodilator messengers such as prostaglandins and NO, the formation of which may require tissue plasminogen activator. Other neurons act by modulating neuronal and astroglial activity by the release of glutamate or GABA. For glutamate, it is thought that stimulation of metabotropic receptors on astrocytes produces an increase in calcium that stimulates the synthesis of epoxyeicosatrienoic acids (EETs). EETs promote opening of TRP channels and large conductance Ca++-activated K+ (BK) channels in astrocytes and VSM. Opening of BK channels in astrocytes can induce sufficient K+ release to hyperpolarize and relax VSM via inward-rectifying K+ channels. Of interest, NO may inhibit the cytochrome P450 that produces 20-hydroxyeicosatetraenoic acid (20-HETE) that acts to close BK channels and increase VSM tone. Hence, maintaining NO bioavailability may be particularly important for preserving NVC in conditions where basal 20-HETE is elevated, such as hypertension, SAH, and ischemia. Progress was made in understanding the impact of vascular dysfunction and for interpreting fMRI signals. The increase in blood flow during neuronal activation is independent of blood oxygenation. Mitochondrial permeability transition pore appears to contribute to impaired NVC after cortical spreading depression, which may occur in ischemic stroke and SAH. Functional hyperemia becomes impaired by increased NADPH oxidase (Nox) production of ROS in models of aging, Alzheimer’s disease, and angiotensin-II-induced hypertension. The scavenger receptor CD36 is required for ß-amyloid-induced cerebrovascular impairment, whereas COX1-derived prostaglandin E2 acting on EP1 receptors is required for angiotensin-II-induced impairment. The latter can be countered by estrogen.

Blood-brain barrier

Significant progress has been achieved in understanding the role of non-endothelial elements of the NVU in establishing and maintaining the BBB during development and under physiological and pathological conditions. Most notably, the critical role of pericytes in BBB formation (ontogenesis) has been established using new PDGF transgenic animal models. The emerging role of Wnt/beta-catenin and hedgehog signaling pathways in induction and maintenance of the BBB phenotype has created new opportunities to target these pathways to ‘repair’ barrier function damaged by disease. Advances in understanding the molecular make-up of the BBB have been made, most notably from proteomic studies. Among these, systematic cataloguing of proteins, receptors, ion channels and transporters polarized to luminal and abluminal membranes of brain endothelial cells led to the discovery of new BBB transporters involved in ion, metabolites, and vesicular transport. Signaling pathways involved in the regulation of expression/function of P-glycoprotein and of the ABC transporter breast cancer resistance protein have been established, whereas a role of these transporters in the pathogenesis of neurodegenerative diseases, such as AD, is emerging. The nonselective cation (NC) channel NCCa-ATP can cause BBB disruption through increased intracellular Ca++ and/or by promoting cell swelling (cytotoxic edema). The regulatory subunit of NCCa-ATP channel, SUR1, prevents cytotoxic edema and improve stroke outcome. In line with recommendations from the previous SPRG report, recent studies have also provided important new insights regarding the role of aquaporins in astrocyte end-feet in the control of water distribution and brain edema in pathologic conditions. Substantial advances have been made in understanding the roles that aging and matrix metalloproteases play in the BBB disruption and hemorrhagic transformation after tPA administration. Notable progress has also been made in understanding immune cell interactions with and their transmigration across the BBB. T-cell subset-selective interacting molecules such as ALCAM-1 and ninjurin-1, expressed in membrane sub-domains of brain endothelial cells have been identified. This is particularly relevant for neuroinflammatory diseases and stroke, since targeting these cell-specific interactions may provide a path for clinical translation of immunomodulatory therapeutic strategies. The efforts to discover new strategies for drug delivery across the BBB have resulted in discovery of novel peptide and antibody carriers using new methods such as phage-display screening. For the first time, some of these new carriers, most notably Ang1005, have been tested in clinical trials, while others are being evaluated pre-clinically.

Vascular risk factors

Vascular risk factors including hypertension, aging, diabetes, hypercholesterolemia, hyperhomocysteinemia, among others, have a profound impact on cerebrovascular structure and function. There has been significant progress in defining effects and mechanisms that underlie vascular changes in models of stroke risk factors (usually single risk factors studied in isolation). Oxidative stress has emerged as a common underlying component of cerebrovascular disease in diverse models. Elevated production of ROS and impaired antioxidant defenses contribute to ROS-dependent effects. These changes include reductions in CBF, endothelial dysfunction, impaired NVC, increased BBB permeability, and altered vascular growth. Cerebral vessels can generate relatively high levels of ROS. Several major sources of vascular ROS have been identified with considerable attention given to Nox and defining its role in models of stroke, hypertension, and Alzheimer’s disease. Virtually all studies in brain have focused on Nox2. Mitochondria are a major source of ROS particularly in endothelial cells, and deficiency in mitochondrial SOD creates predisposition to oxidative stress, vascular dysfunction, and atherosclerosis. The role of antioxidant defenses has begun to be characterized. The complexity of oxidant-related effects is due in part to the many molecular species involved. Inactivation of NO by superoxides, produces peroxynitrite which alters vascular tone and permeability, but may also feed-forward promoting further oxidative stress. The role of H2O2 is even more complex since it may contribute to normal signaling, but also elicits diverse pathophysiological effects. The RAS has emerged as a common underlying mechanism promoting oxidative stress in large and small vessels. Through pleiotropic effects, the RAS contributes to end-organ and vascular cell damage in a continually growing list of diseases. New subcellular systems have been identified including a mitochondrial angiotensin system. The RAS interacts with other pathways that promote vascular disease including the endothelin and eicosanoids. Additional studies, particularly in models of hypertension, have emphasized the importance of rho-kinase signaling, inward remodeling of cerebral arterioles, and rarefaction of microvessels, all of which can have a negative impact on local perfusion. Some work suggests a modulatory role for P450 metabolites in these processes. Activation of PPAR? inhibits inward vascular remodeling and promotes vasodilation whereas interference with PPAR? signaling mimics effects of angiotensin II. Inflammatory cells and inflammatory-related pathways in vascular cells play a major role in atherosclerosis, a major risk factor for ischemic stroke. Continued insight has been gained into the cellular and genetic basis of atherosclerosis in carotid artery and aorta, with lipids and inflammatory pathways having a major role. Aging activates inflammatory signaling, an effect linked to nuclear factor-?B (NF-?B). Increased production of angiotensin II, and increased expression of pro-inflammatory mediators, inducible NOS, matrix metalloproteinases and ROS are present in aged arteries. Oxidative injury and inflammation are highly interactive processes, but some details of the cross-talk between these mechanisms are starting to be recognized. While risk factors, e.g., hypertension, cigarette smoking, diabetes, and some molecules, e.g., angiotensin II and endothelin 1, promote inflammation, oxidative stress, and atherosclerosis, much less is known about how these elements interact. Many of these same pathogenic pathways may affect the microcirculation promoting small vessel disease (SVD). Definitions of SVD vary, with clinical SVD characterized largely by parenchymal effects: lacunar infarcts, white matter lesions, and microbleeds. These abnormalities are linked with dementia and predispose to major stroke. Clinical data suggest SVD has a complex etiology, with age, hypertension, genetic predisposition, and cerebral amyloid angiopathy, all acting as important risk factors. In effort to understand the basis for SVD, as well as to develop new therapies, this report takes a broader view including the pathophysiology of pial and parenchymal microvessels (small arteries, arterioles, capillaries and veins) that may contribute to reduced CBF, reduced local perfusion pressure, impaired vasodilator responses, increased permeability, loss of trophic support, and cellular injury. In this context, studies of the pial microcirculation have continued while studies of parenchymal vessels have increased. These efforts have highlighted common but also unique features in each segment. Elements of SVD that occur commonly (particularly with hypertension) such as wall thickening, lumen narrowing, and microaneurysms, has continued to be studied. Mechanistically distinct processes may underlie these different structural changes. Vascular hypertrophy for example is common in models with oxidative stress. Inward remodeling develops much more selectively, but may have the greatest impact on local perfusion pressure and blood flow. New genetic models (eg, mice with mutations in Notch3, collagen type IV, and PPARg) have provided insight into the role of these factors in SVD. SVD may also suppress trophic effects of the vasculature on neuronal tissue and thus may contribute to neuronal dysfunction independently of alterations in CBF.

Gender disparities

New data have highlighted sex differences in neurovascular function, autoregulation and dysfunction after stroke. Vascular effects of angiotensin II are sex-dependent, being generally much larger in males compared to females. However, the deleterious impact of these pathways may grow in women after menopause. Novel mechanisms have been investigated including endothelial mitochondrial effects of estrogens, local cerebrovascular sex steroids, sex-specific endothelial signaling pathways, antioxidant mechanisms, and the role of novel estrogen receptors such as GPR30. A significant sex disparity in stroke death rates has recently been recognized. Specifically, stroke became the fifth leading cause of death in men, but remained the second leading cause of death in women. Sex differences have been observed in the use of and response to tPA and catheter-based therapy (stents and thrombectomy devices), with women seemingly benefitting more than men from tPA, yet less likely to receive thrombolysis treatment compared to men. In pediatric arterial ischemic stroke (AIS), male predominance is observed. Gender disparity also occurs in neonates. Data suggest that neonatal AIS and cerebral sinus thrombosis are more commonly diagnosed in boys. Most deaths from childhood stroke are due to hemorrhagic stroke, and infants and males have the highest risk. Birth cohort mortality rates fell for each successive generation since the 1950s suggesting the influence of prenatal or perinatal factors. After eliminating cases with coexisting sickle cell disease, excess stroke risk persisted in blacks; after elimination of trauma, excess stroke risk persisted in boys. Total testosterone concentration was recently measured in children with AIS, children with cerebral sinovenous thrombosis, and 109 healthy controls. In children with AIS or cerebral sinovenous thrombosis, testosterone levels above the 90th percentile for age and sex increase risk, such that for each 1 nmol/l increase in testosterone the odds of cerebral thromboembolism were increased 1.3-fold. Sex differences have also been investigated in pediatric animal models of cerebrovascular disease and ischemia. A neonatal rodent study investigating the effect of 2-iminobiotin (2-IB) showed that 2-IB treatment reduced long-term brain damage in female, but not male rats. Cerebrovascular dysregulation during hypotension occurs after fluid percussion brain injury (FPI) in the newborn pig owing to impaired K+ channel function. Hypotensive pial artery dilation is impaired after FPI in a gender-dependent manner. A number of studies over the past decade utilizing neonatal rodent models of stroke and hypoxia-ischemia (HI) have shown gender differences in apoptotic pathways, especially caspase signaling which is enhanced in the female newborn brain and thus more susceptible to blockade.

Developmental and pediatric issues

The high incidence of cerebrovascular ischemic and hemorrhagic events in preterm infants and newborns, coupled with devastating outcome and high cost, make pediatric stroke an important public health problem. Successes in basic and clinical research have led to the establishment of hypothermia as standard of care for neonatal brain injury caused by asphyxia-ischemia. Sickle cell stroke is one area where focused intervention has brought effective therapies (transfusions, hydroxyurea) and perhaps could serve as a model for studies of stroke in childhood. Basic research continues to improve understanding of the regulation of fetal and neonatal cerebrovascular circulation, how this regulation differs from that of older individuals and how pathologic conditions impact that regulation often in ways different from in the older child or adult.

Pregnancy, preeclampsia and the risk of stroke

Preeclampsia shares many pathophysiological features of atherosclerosis, including endothelial dysfunction, activation of the coagulation cascade, and lipid abnormalities. Women with prior preeclampsia have increased risk for cerebrovascular disease and stroke. A recent clinical study of formerly eclamptic women found that they had subcortical white matter lesions twice as often as women with normal pregnancy, suggesting a link with cognitive impairment. Animal studies have shown that the vascular endothelium is more sensitive to insults during pregnancy, an effect that was recently shown in the cerebral circulation.



The signaling pathways in endothelium activated by PGI2 are incompletely understood. Existing evidence support the concept that PGI2 may activate not only receptors coupled to the canonical adenylate cyclase pathway, but also peroxisome proliferator activated receptors (PPAR) isoforms, including PPAR?. Since activation of PPAR? regulates many target genes it is likely that protective effects of PGI2 is very complex. Defining signaling mechanisms underlying endothelial protective effects of PGI2 might help in identification of new therapeutic targets in prevention and treatment of cerebrovascular disease. Moreover, the regenerative function of the cerebral circulation, e.g., repair of injured endothelium appears to be dependent on arachidonic acid metabolism and production of PGI2 by endothelium and in circulating endothelial precursors. The impact of endothelium and eNOS-derived NO is very diverse with new functions continuing to be discovered. Thus, continued work in this area is needed to understand subcellular and molecular mechanisms that 1) underlie normal endothelial function, 2) become dysfunctional and contribute to vascular disease, and 3) participate in endogenous mechanisms of vascular protection. Aspects of eNOS signaling are poorly understood. For example, how rho kinase inhibits eNOS function in endothelium as well as NO signaling within vascular muscle is not known. PPAR? may suppress rho kinase function and facilitate NO signaling through effects that are largely unknown at present. The pathogenic contribution of eNOS ‘uncoupling’, a condition in which eNOS produces superoxide instead of NO, remains to be defined. Inward vascular remodeling can reduce vasodilator capacity results in a mismatch between energy requirements and substrate delivery. Although inward remodeling has emerged as a risk factor for cerebrovascular events, the impact of endothelium on these changes is poorly understood. There is increasing evidence that endothelial dysfunction and loss of NO plays a role in the vascular component of cognitive impairment and Alzheimer's disease. However, definitive causal links between endothelial function and cognition are still lacking due to limitation of current models and approaches. The overall impact of EDH-related mechanisms during vascular disease, particularly in vivo, remains poorly defined. While endothelial cells are clearly a key element in coordinating segmental vascular resistance, a process essential for optimal regulation of local perfusion pressure and thus blood flow, the mechanisms that control communication along and between vascular segments remain poorly defined. How these mechanisms are altered by disease or after stroke remains unclear. The role of endothelium in the coupling between neural activity and blood flow may be substantial in some regions, but little is known regarding such mechanisms. Similarly, to what extent astrocytes affect vessels through endothelium-dependent mechanisms is unclear.


How mechanical stimuli (pressure and flow) affect vascular tone and vascular gene expression are prime examples. The identification of new ion channels has provided additional insight; however, key elements of autoregulatory and flow-mediated responses are still not well understood. While it is clear that ion channels, Ca++ signaling, and other pathways are altered by these mechanical stimuli and in disease, many fundamental issues remain to be resolved. For many newly identified mechanisms, it is often unclear how functionally important they are under physiological or pathophysiological conditions in vivo. Our understanding of the properties of VSM in intraparenchymal arterioles remains limited. How VSM and pericytes integrate the various signals derived from other cell types needs to be elucidated. The role of pericytes in controlling CBF, their mechanisms of proliferation and remodeling and gap junction communication, are not well understood. There is still no single molecular marker that unequivocally distinguishes pericytes from related cells under normal and disease states. Conducted vascular responses and network control of VSM/pericyte function is very complex and requires greater knowledge under normal physiological conditions and pathological states. Our understanding of CBF regulation derives mainly from cortical regions, which can be visualized from the brain surface, but CBF regulation in deeper structures and other brain regions is less well understood. Considering that mechanisms of neurovascular coupling (NVC) differ in cerebral cortex, cerebellar cortex and basal forebrain, VSM/pericyte function may differ across brain regions as well. Unveiling the pericytes' role in the pathophysiology may increase the success of reperfusion therapies. Furthermore, the pericytes' role in tissue revascularization and repair after stroke also deserves further investigation. While there has been progress in our understanding of mechanisms of NVC and autoregulation, advances in other major areas of CBF regulation have been more incremental. For example, although hypercapnia and acidosis have powerful effects on vascular tone and are very important clinically, there are major gaps in our understanding of mechanisms that control these responses.

Neuro- and glio-vascular signaling

The signaling cascade linking astrocytes and neurons to blood vessels requires more detailed clarification, and tools to better target cellular function in vivo need to be developed. Since EETs released from astrocytes may be key intermediaries in NVC coupling to various stimuli, the cellular localization in brain tissue of the enzymes involved in this synthetic pathway is needed. Communication between astrocyte foot processes and other NVU cells including the role of gap junctions in this intercellular communication requires further exploration. In cerebral cortex, gliovascular coupling accounts for only part of the steady-state vasodilation during increased neuronal activity and does not appear to contribute to the initial dilation. Moreover, combining inhibitors of the known astrocyte and neuronal signaling pathways does not eliminate the cortical vascular response. Thus, other pathways remain to be identified or the interplay between the various pathways better understood. Furthermore, the functional role of increased astrocytic glycolysis and lactate production in metabolic coupling with neurons needs further elucidation. Signals arising from increased glycolysis, redox state, ROS, and glutamate and GABA uptake need further study. No in vivo studies have investigated Ca++ dynamics in identified neurons producing vasoactive substances or the vascular effects of their stimulation. Such studies will be possible with the development of transgenic mice expressing fluorescent reporters and/or optogenetic tools in discrete subsets of cortical neurons together with the emergence of ultrafast multispectral imaging systems that allow simultaneous monitoring of Ca++ events and hemodynamics. Further, novel optical reporters of other physiological variables, such as blood and tissue oxygenation, will extend the range of information on stimulus-induced changes in cortical physiology and metabolism that can be obtained. Such studies are needed to evaluate the contribution of specific neuronal types in NVC in different brain regions. They should provide decisive conclusions on the temporal, spatial and extent of the neurally driven hemodynamic alterations and how the latter can be interpreted in the context of brain imaging of normal or pathological physiology. Brain ischemia and stroke alters NVC. However, detailed study of this phenomenon has been hindered, in part, by the difficulty of distinguishing neuronal electrophysiological alterations from impaired neuro-glio-vascular signaling. The effect of ischemia on astrocyte Ca++ needs to be delineated. Changes in astrocytic domains and gliovascular coupling during reactive gliosis and inflammation, and changes in angiogenesis and dendritic plasticity in the repair phase after stroke needs to be studied. Determining how hemodynamic signals arising from the endothelium and other vascular elements might influence astrocytic or neuronal function needs to be explored and then examined in the context of neurodegenerative processes. The impact of these neurovascular alterations produced by stroke and stroke risk factors on cognitive function remains to be assessed. Furthermore, whether vascular pathology plays a primary role in the cognitive decline associated with neurodegenerative diseases remains to be defined.

Blood-brain barrier

Key achievements, future directions, as well as barriers to advancing translational research in BBB biology have been summarized in Nature Neurosci Rev (12:169) on behalf of the International BBB Consortium. The following ‘knowledge gaps’ and emerging questions have been identified that pertain more specifically to the mandate of SPRG. Since the BBB is one of several functions of the NVU, advances in understanding this function are contingent upon continuing understanding of how molecular changes involved in cell-cell contacts (including tight junction regulation) and communication (endothelium- astrocyte – pericyte -inflammatory cell) and cell-matrix interactions within NVU impact BBB establishment, maintenance and response to disease. For such ‘systems-scale’ understanding, integration of information from ‘omics’ profiling with function assessment using novel models and imaging methods is critical for further progress. Recent application of advanced, high resolution in vivo imaging techniques provided preliminary evidence that the BBB permeability change in stroke and other diseases is not an all or nothing event, but rather complex, regionally scattered and size-selective for molecules of different hydrodynamic size. However, detailed understanding of dynamics, spatial characteristics and size-selectivity of BBB permeability during disease process (age-dependency; acute and chronic BBB ‘opening’ in stroke, tPA-triggered opening; early BBB dysfunction/failure in AD, etc), as well as of mechanisms governing these changes are lacking. This understanding will help exploit opportunities for delivery of protective therapies, and using BBB function as an early indicator (biomarker) or target to modulate disease process. The luminal surface of endothelial cells contains a particularly thick glycocalyx, which is involved in key BBB functions, including the transport of solutes and macromolecules, permeability, vasoreactivity, and interactions with circulating cells and platelets. Due to its complexity and lack of appropriate techniques, there have been virtually no studies of this ‘sub-compartment’ of the BBB. Yet, changes in molecular composition of the glycocalyx may be particularly important for stroke, given its role in thrombogenesis, blood flow, and cell trafficking, as well as its tendency to ‘shed’ upon ischemic/inflammatory stimuli, providing a source of potentially unique circulating biomarkers of BBB dysfunction. Along with advances in molecular understanding and dynamics of the BBB changes in disease, it is an opportune time to exploit this understanding through increased efforts in developing strategies to protect and repair damaged BBB. This is of particular importance in stroke, where such strategies would help extend the therapeutic window for tPA administration. Although the field of BBB drug delivery technologies has recently experienced a dramatic expansion, emerging drug delivery methods are still vastly inadequate to address the spectrum of neurological diseases. The most important impediment for clinical translation of emerging biopharmaceuticals - growth factors, genes and vaccines - is their delivery to the brain. The delivery across the BBB also hinders the development of ‘targeted’ molecular imaging agents needed for early and precise diagnosis of neurological disease. Therefore, discovery and development of ‘vectors’ (antibodies, carriers, particles) to deliver macromolecules across the BBB in a controlled and non-invasive manner should remain a high priority. These efforts should be coupled with better understanding of the fate of macromolecules in the brain – their diffusion to target, elimination routes from the brain, reverse transport across the BBB, etc. These discoveries should promote further advances in the field.

Vascular risk factors

Development of therapies specifically designed to protect against vascular oxidative stress remains a major challenge due in part to our incomplete understanding of the role of ROS in the vasculature. Further studies are needed to determine the sources and compartmentalization of ROS, and to identify which ROS species contribute to different pathologies. In addition, the effects of oxidative stress on trophic and regenerative functions of the vasculature are poorly understood. New techniques for early identification of pathology are needed if potential therapies to halt progression of disease are to be tested. Additional work on large and SVD are needed to better define mechanisms that promote and protect against vascular disease in general. The mechanisms by which vascular dysfunction leads to parenchymal damage remain poorly understood and animal models that better mimic the human condition are needed to enable studies of potential neuronal dysfunction and other complications that result from microvascular problems. Studies defining cell-specific and molecular mechanisms of disease remain a priority for all segments of the brain circulation. While some diseases or risk factors affect predominantly large vessels, e.g., atherosclerosis, others affect all segments, e.g., hypertension. In relation to BBB permeability, many changes occur at the level of capillaries and veins. Veins are also a key site for interactions between endothelium and circulating immune cells. A reasonable priority would be to focus on understanding systems, pathways, and molecules that have been implicated as contributing to vascular abnormalities in multiple models of disease. Studies of vascular protection may ultimately lead to the exploitation of endogenous protective mechanisms in a way that slows the onset and progression of vascular disease. Molecules like PPARg and other transcription factors may be promising targets due to their potential to suppress oxidative stress and inflammation, slow progression of atherosclerosis, and provide a mechanistic balance against the RAS and endothelin systems. Although effects of many molecules are cell specific, many or our current models do not take these issues into account. While effects of hypercholesterolemia on cerebral vessels have received attention, studies that focus on basic aspects of intracranial atherosclerosis are rare. It remains unclear whether mechanisms of atherosclerosis in intracranial arteries have unique aspects, reflecting the unique immune characteristics and blood flow dynamics of the brain. Despite its clinical impact, the contribution of aging to large and small vessel disease continues to be greatly understudied. How combinations of risk factors, e.g., hypertension and aging, accelerate vascular abnormalities at the molecular level are rarely pursued. Unappreciated mechanisms and new therapeutic targets may arise from the study of vascular disease due to combined risk factors. Lastly, the increasing recognition of the importance of vasculature in the pathogenesis of neurodegenerative diseases calls for efforts to define the cellular and molecular features driving the complex interaction between vascular and non-vascular cells. A major area of focus in the future is expected to be the immunologic impact of stroke risk factors on the components of the vasculature. A need for newer experimental models exists. While some progress has been made in developing models of hypertension-induced intracerebral hemorrhage and aneurysm rupture, the approaches used do not closely replicate clinical conditions. Aortic stiffness, hyperhomocystinemia, blood pressure variability, obesity, and obstructive sleep apnea are recently emerged risk factors for stroke, and experimental models are needed to better understand their impact on vascular structure and function. Molecular mechanisms relevant for understanding of cellular aging include telomere shortening, damage of DNA, alterations in mitochondrial function, calorie intake, regenerative capacity, genetic polymorphisms, autophagy, function of the circadian clock, and post-transcriptional gene regulation (microRNAs). The exact contributions of these mechanisms to vascular aging and their impact on vascular regulation are largely unknown. The regional heterogeneity of aging–induced endothelial dysfunction including NOS isoforms, EDH, and metabolisms of arachidonic acid should be defined. The ability of circulating progenitor cells to repair injured cerebral vessels during progression of aging has not been systematically studied. Phenotypic characterization of cerebral circulation in experimental models of accelerated aging might help to identify novel pathways and molecules critical for healthy aging. Efforts to improve understanding of the role of aging vasculature in pathogenesis of cognitive impairment should be intensified. Genetic and pharmacological approaches that may slow vascular aging should be explored.

Gender disparities

We need to: (a) investigate effect of sex steroids on cerebrovascular function and disease to alleviate the hormones’ adverse effects, while capitalizing on their protective properties; (b) continue to study mechanisms underlying the age-related increase in stroke rates in women after menopause; (c) investigate mechanisms underlying sex-based differences in stroke therapy, including aspirin, tPA and thrombectomy, with special emphasize on the impact of these therapies on cerebrovascular function and the endothelium; (d) continue to investigate sex-specific signaling pathways, especially endothelial mechanisms of injury in cerebral microvessels; and, (e) understand mechanisms of gender effects in stroke in childhood as well as neonate. A major problem remains that a limited number of investigators are interested in childhood stroke basic mechanisms. This is detrimental because extrapolating adult data to the young has proven to be paradoxical at times. Therefore, training programs to encourage young scientists to enter this field of research would be highly desirable.

Pregnancy, preeclampsia and the risk of stroke

It is not clear whether preeclampsia is a causal risk factor for stroke or whether preeclampsia and cerebrovascular disease share common risk factors that are unmasked by preeclampsia. This issue needs to be addressed. The underlying causes of seizure during pregnancy, which can occur independently of hypertension, are still unknown. Better models of preeclampsia and eclampsia are needed in which the cerebral circulation can be studied in greater detail.

Developmental and pediatric issues

Animal studies suggest sufficient promise of therapeutic agents alone or in combination with hypothermia in countering neurovascular injury to warrant further basic and clinical experimentation. Early safety and pharmacokinetic study of large-dose erythropoietin in very premature babies was recently published. More recently, there are pilot clinical studies that suggest erythropoietin may be efficacious in neonatal asphyxia-ischemia induced brain injury with improved neurological outcomes being reported. Many questions remain regarding roles of gestational age, candidates for treatment, treatment regimens, and prolonged follow-up. Knowledge of mechanisms involved in physiological regulation of the cerebral circulation and pathological alterations to this regulation induced by asphyxia/ischemia is needed. Our understanding remains very incomplete in age dependence of contributions of paracrine mediators in control of CBF and our knowledge of the structural differences that exist among fetal, newborn and adult remains very incomplete. Understanding normal function and pathological alterations will be necessary to treat all components of brain injury rather than just neurons, and will be critical for the development of new therapies. Potential for iatrogenic hemorrhagic and ischemic stroke and brain injury caused by mechanical ventilation, ECMO, permissive hypercapnia, and hyperventilation deserve further investigation. Another area ripe for investigative focus is diagnosis of developing, progressing and regressing injury to NVU integrity as well as improvements in monitoring for seizures and imaging of lesions. Advanced imaging techniques have pointed to disruption of brain development in the setting of early stroke or white matter injury. Studies are needed to understand the timing and mechanisms of brain injury in newborns with congenital heart disease and how brain abnormalities, now documented by imaging studies, can inform mechanisms. Poorly understood is the age specificity of vascular, astrocyte, and neuronal effects of and responses to components of tissue damage coupled with unique pathologies and potential recovery in the brain not already developed. Animal studies have found gestational age and sex to be an important contributor to the nature of the brain injury caused by asphyxia-ischemia with different injury patterns and neurological expression. Clinical observations suggest similar differences in humans, but much more study is needed.


1) Elucidate the genomic, proteomic features and the corresponding functional correlates of vascular and perivascular cells, and cells of the NVU. Priorities include: (a) defining cell-specific and molecular mechanisms of disease for all segments of brain circulation, from large cerebral vessels outside the brain to intraparenchymal vessels in intimate contact with neurons and astrocytes; (b) elucidating the changes that occur during development and aging, and investigating their sex specificity and the modifications induced by functional activation in the normal state, after injury and during repair processes.

2) Elucidate the signaling mechanisms governing the functional and trophic interactions among the cellular elements of the vasculature, as well as their relationships to neurons and perivascular cells in the developing, adult and aging brain, and in both sexes. Assess how activation, stroke risk factors and injury alter such interactions, and how damage to one cell type alters the homeostasis of the vessel wall, including function of the NVU and the BBB.

3) Achieving these goals will require novel applications of current technologies, as well as development of new molecular, cellular, and imaging tools to dissect the inner workings of vascular cells and their interactions with neurons, astrocytes and perivascular cells. Translation of newly identified mechanisms and pathways into therapies will require well controlled disease models incorporating multiple stroke risk factors and taking into account age (neonatal, pediatric, adult, and old age) and sex. Efforts to develop a new generation of investigators in cerebrovascular and neurovascular biology will be vital to advancing the field to the next level.



Co-Chairs: Joseph Broderick, Thomas Brott, Karen Johnston

Members: Colin Derdeyn, Pam Duncan, Nicole Gonzales, Yuko Palesch, Rema Raman, Robert Silbergleit, Tanya Turan, Kenneth Cavanaugh (reviewer), Natalie Getzoff (reviewer)

NINDS Liaisons: Scott Janis, Claudia Moy, Salina Waddy


Key concepts gleaned from recent trials that will drive future research.

1) The results of the COSS and SAMPMPRIS Trials suggest that future stroke prevention trials should probably include aggressive treatment of vascular risk factors in the protocol and open the question of whether the results of the asymptomatic endarterectomy trials performed over a decade ago are still valid in the modern era of statins and newer antihypertensive medications.

2) The past five years has also seen the development of several advances in stroke clinical trial methodology and infrastructure that can impact present and future trials.

3) The impact of CMS reimbursement policies on trial recruitment has become very apparent, with the success of SAMMPRIS and CREST and the struggles of IMS 3 and RESPECT.

Future Priorities:

Improved efficiency of clinical trials

  • Continued support/expansion of infrastructure and best recruitment practices to facilitate timely and efficient completion of trials (networks, NIH-sponsored national IRBs, resources for study monitoring)
  • Reimbursement for Phase IIb-III Trials by CMS/payors for promising but clinically unproven therapies prior to reimbursement for clinical delivery of technology of care, requiring coordination of federal agencies
  • Partnership with other countries to improve recruitment and generalizability
  • Minimization of overlapping and concurrent competing stroke trials

Improved trial design, conduct, and outcome assessment.

  • Centralized tools for trials for getting studies started and identification of well designed
  • outcome variables that are consistent across studies
  • Training tools for new coordinators/investigators (building upon past successful trials)
  • Innovative statistical designs

Conduct clinical trials that advance current evidence-based therapies or for those conditions without a proven therapy:

  • Acute ischemic stroke trials that test treatments potentially more efficacious than IV tPA alone or use imaging to test treatments in imaging selected subgroups beyond 4 ½ hours
  • Acute intracerebral hemorrhage (medical and surgical)
  • Trials of patients with ruptured aneurysms
  • Trials of aggressive medical therapy vs. interventional treatment of asymptomatic carotid stenoses
  • Trials of new anti-thrombotic agents in patients without a cardioembolic source who fail aspirin
  • Trials of behavior change/primary stroke prevention (beyond scope of NINDS alone)
  • Phase I, II and III trials of neurorecovery therapies


Key clinical trials since 2007

Key trials for primary and secondary stroke prevention include RELY, Prevention Regimen for Effectively Avoiding Second Strokes Trial (PRoFESS), Carotid Revascularization versus Endarterectomy and Stenting Trial (CREST), Carotid Occlusion Surgery Study (COSS), and Stenting and Aggressive Medical Management for the Prevention of Recurrent Ischemic Stroke (SAMMPRIS). RELY was the first study to establish a safer and easier alternative to warfarin for stroke prevention in patients with atrial fibrillation. PRoFESS showed similar recurrent stroke rates with 2 anti-platelet agents: clopidogrel and the combination of aspirin and extended release dipyridamole. CREST found similar overall outcomes with angioplasty and stenting and surgical endarterectomy for patients with symptomatic and asymptomatic carotid stenosis. However, in older patients, endarterectomy was associated with better outcomes. COSS found no advantage for surgical bypass over medical therapy for patients with complete atherosclerotic carotid occlusion. SAMMPRIS reported better outcomes at one year with medical therapy over angioplasty and stenting for symptomatic intracranial atherosclerotic disease. Two important acute stroke treatment trials that were concluded were the European ECASS III and STICH. ECASS proved the benefit of IV TPA out to 4.5 hours in selected patients with acute ischemic stroke. STICH demonstrated similar outcomes in patients with intracerebral hemorrhage treated with surgical therapy as compared to those treated initially with medical therapy alone. A subset of patients with superficial ICH tended to do better with surgical removal and this observation led to ongoing STICH II Trial of lobar ICH. Several important randomized studies of recovery after stroke demonstrated both the promise and limitations of various approaches to improve recovery of function following stroke (LEAP, EXCITE, and VA cooperative Robotic Trial).

Key concepts gleaned from recent trials that will drive future research.

1. Improved outcomes with medical therapy for patients with atherosclerotic arterial disease (COSS and SAMMPRIS). In both COSS and SAMMPRIS, event rates in the medical group were much lower than in prior trials. These results and other supporting data have opened the question as to whether the results of the asymptomatic endarterectomy trials performed over a decade ago are still valid in the modern era of statins and newer antihypertensive medications. Furthermore, these results suggest that future stroke prevention trials should probably include aggressive treatment of vascular risk factors in the protocol.

2. Hemodynamic impairment is a powerful risk factor for stroke in atherosclerotic occlusive disease (COSS). Patients without perioperative complications had much lower risks of stroke than medically-treated patients. If surgical complication rates been lower, surgery, EC-IC bypass may have been effective in these selected patients. Other revascularization procedures with lower risks may be effective in this population.

3. The past five years has also seen the development of several advances in stroke clinical trial methodology. Web-based technology has been widely adopted and has facilitated CRF completion, randomization processes, and imaging cores with easy remote access. Trial networks have continued to develop and grow as they have demonstrated success in facilitating recruitment and execution of trials of all phases: Phase 1 and 2 (SPOTRIAS), Phase 3 (NETT). The NINDS has advanced the use of cooperative agreements for large clinical trials, and this has helped run these studies more efficiently. Finally, the impact of CMS reimbursement policies on trial recruitment has become very apparent, with the success of SAMMPRIS and CREST and the struggles of IMS 3 and RESPECT.


Needed Clinical Trials

We remain without a scientifically proven treatment for intracerebral hemorrhage (ICH) – the most devastating type of stroke. We now have evidence of treatments that effect biologic markers of hemorrhage (e.g. recombinant factor VIIa to slow bleeding and catheter installation of t-PA to accelerate removal of clotted blood from ventricles and parenchyma). We also have demonstrated the ability to rapidly and effectively lower elevated blood pressure in the clinical setting of acute ICH. However, we have yet to clearly demonstrate improved clinical outcome. Thus, ongoing trials of minimally invasive removal of clot and refined methods to slow bleeding acutely in appropriately selected patients are needed. Other medical approaches such as potentially accelerating removal of blood products by pioglitazone and minimizing the effects of intraparenchymal iron from the bleeding by deferoxamine are also ongoing. These trials will require large number of centers with coordinated neurosurgical and neurological care to provide adequate recruitment.

The emergent treatment of acute ischemic stroke still has no scientifically proven treatment other than intravenous t-PA. Ongoing randomized trials comparing endovascular approaches to IV t-PA, or to standard therapy if beyond the 4 ½ hour window for t-PA, need to be completed. Inclusion of imaging selection in patients beyond the 4 ½ hour window will be important. Evolving technology needs to be compared to standard therapy in randomized trials. Improved medical therapies for reperfusion, including enhancement of lysis by a combination of medications or by extracranial ultrasound, are also potential subjects for a Phase III Trial if Phase II trials are suggestive. Finally, neuroprotection for acute stroke remains a focus despite a frustrating past experience. Hypothermia, which has been demonstrated to be effective in global ischemia with reperfusion, remains the best potential treatment under investigation and is currently in Phase II trials. Other approaches under study include albumin infusion in acute ischemic stroke, closer control of hyperglycemia in acute stroke patients, and pre-hospital treatment with magnesium therapy. Ideally, neuroprotection strategies in the field and pre-hospital setting could provide the most potential benefit as long as therapy can be administered without need for brain imaging.

Medical therapy for stroke prevention has evolved substantially over the past 25 years and earlier trials comparing standard medical therapy to an intervention may need reconsideration in the light of more aggressive current approaches. A prime example of this is the SAMMPRIS Trial. One priority is a re-evaluation of aggressive medical therapy as compared to carotid artery stenting/endarterectomy in patients with an asymptomatic carotid artery stensosis.

Determining the best approach for medical prevention of recurrent stroke in someone previously on anti-platelet medication is another area of clinical need. Comparison of the various anti-platelet agent regimens has not demonstrated a clearly superior approach with plavix and Aggrenox having similar efficacy and safety. Earlier trials of stroke prevention comparing warfarin and aspirin in non-cardioembolic stroke showed no significant difference in stroke recurrence but were accompanied by higher rates of hemorrhage in the warfarin group. The newer generation of antithrombotic agents (anti-thrombin and anti-factor X) has the potential of superior safety as compared to warfarin. Thus, comparison of these new anti-thrombotic agents to aspirin in various populations of non-cardioembolic ischemic stroke patients should be strongly considered. These trials are more likely to be funded by industry.

Neurorecovery trials include devices, physical therapy approaches, stem cell infusions (locally and intravenously), and growth factors. This area of clinical trials has the biggest potential for advances over the next decade and next century. Pilot trials in animals and humans are ongoing and accelerating. Phase I and Phase II stem cell studies need to be rigorously designed regarding outcome measures, procedures, and safety monitoring (short and long term). Phase III trials of cell therapies are premature until more preliminary data in animals and humans are available. Phase III Trials are currently best suited for physical therapy approaches with or without devices or brain stimulation, although Phase II studies of these approaches are also needed.

Finally, technology (e.g., Imaging, genetic testing) has enabled determination of the differential efficacy of treatments. Examples include MRI/CT imaging in patients with acute ischemic stroke, and genetic testing for medication effects such as clopidogrel and warfarin). In some respects, this is an extension of determining the efficacy and safety of acute and preventive therapies in various clinical subgroups.

Design, Conduct and Successful Completion of Clinical Trials

Despite the progress made in recent years, there remain several critical areas of need in stroke clinical trials. As the ultimate goal of stroke clinical trials is to evaluate new treatments and demonstrate improvement in stroke therapy, the efficiency of completing stroke clinical trials has been identified as a critical area of need. Areas of particular focus include trial design, trial recruitment, translation of trial results to clinical practice and careful consideration of appropriate resources to support the research activities required.

Stroke clinical trials must be designed and conducted in an efficient way. New approaches to design include adaptive design (Frequentist or Bayesian) that may maximize information and update estimates used to determine success. Consideration of the Bayesian framework for adaptive designs, when there is high confidence in the prior information, may be appropriate. Internal pilot studies where the sample size can be re-estimated or consideration of adding or dropping a group midcourse are examples of adaptive designs that may be valuable. These designs would have to be pre-specified prior to initiation of the trial to maintain a trial’s validity. Efforts to better understand appropriate design strategies as well as interpretation of results are urgently needed to maximize efficiency of stroke clinical trials. The advantages and limitations of adaptive designs must be well understood by the stroke community.

The identification and standardization of validated outcomes that will maximize outcome information (appropriate sensitivity/specificity), be easy and reliable to capture and be relevant to long term clinically relevant patient outcome are critical. Additionally, capture of cognitive outcome, patient perspective outcomes and effectiveness outcomes are of great interest.

The engagement of the clinical trial community in the use of the newly developed NIH tools including NeuroQOL, PROMIS, the NIH toolbox and the NINDS Common Data Elements (CDEs) is a critical area. Finally, methods to allow rapid and efficient regulatory activities including all activities that will streamline clinical trial infrastructure and regulatory processes are of great importance. Such activities may include but are not limited to the use of central IRBs, streamlined reporting of adverse events, rapid negotiation of master trial agreements between sites and trial sponsors, and efficient interactions with the FDA.

Currently, the large number of simultaneous stroke clinical trials has resulted in slow recruitment for many trials. A critical area of focus must be new approaches to rapidly and efficiently recruit stroke patients to clinical trials research. Partnerships with the new NIH–NINDS recruitment experts to utilize new strategies will be important, including but not limited to the use of new technologies/ social networks, and the use of repositories of successful recruitment strategies. Additionally, recruitment of specific stroke populations is a critical area of focus in stroke trials.

The pediatric population is of particular interest as treatments approved for adults are of unclear risk and benefit in children. A network of investigators for pediatric trials is now in place and preliminary feasibility work suggests these trials are feasible. Minority populations and populations identified in disparities research are also a critical focus as need for improved treatments for those populations are clear. Novel approaches to attracting these populations to clinical trials or retaining them in clinical trials may be necessary. The development of new tools and strategies to engaged underrepresented groups is a priority such as demonstrated by the CRUiSE workshop highlighting recruitment issues (http://www.nhlbi.nih.gov/meetings/workshops/cruise.htm).

More rapid successful completion of clinical trials will depend upon maintenance and expansion of recruitment networks, creative solutions to training of coordinators/investigators, methods to insure call coverage for acute stroke studies, finishing ongoing NINDS trials before starting competing NINDS trials which have major overlap, and harmonization of the decisions of governmental regulatory agencies with needed and ongoing clinical trials. This latter issue is really limited to scientific testing of devices. The SPOTRIAS and NETT networks have been very useful in completion of Phase I-II and Phase III trials, respectively. However, the networks still have a limited volume of patients and involvement of spokes/other centers will be important going forward. Restarting networks or trials from scratch is very expensive and time-consuming. Turnover of coordinators and investigators is frequent in stroke trials. Use of teleconferencing and web-based methods can decrease costs for NINDS trials. Use of the recruitment person at NINDS to help gather best practices for center and trials regarding improving recruitment is needed.

Finally, CMS and FDA should be aware of ongoing clinical stroke trials in relationship to new technology and should consider the effects of approval on ongoing studies addressing the clinical efficacy of devices or approaches. For example, knowledge of existing studies could however be used to enhance the new study design and interpretability of the resulting data. For instance, new studies could have eligibility criteria that mirror those in existing studies to facilitate comparisons of results. In particular, any decisions regarding reimbursement should not be made without consultation of clinical experts and investigators about existing science documenting the clinical efficacy and safety of new technology and ongoing trials. The call for a Federal Partners Working Group by participants of the CRUISE workshop reflects this same priority and includes reimbursement strategies to promote clinical trial participation for clinically unproven therapies/technologies, standard and transparent procedures for assessment of efficacy and safety for all governmental agencies, and harmonization and promotion of best regulatory practices.

The recent focus on the importance of effectiveness in addition to efficacy has also resulted in a critical need for strategies to move clinical trial results efficiently into clinical practice. Overall, the highest priorities for critical needs are the allocation of appropriate resources and development of key partnerships to support continued clinical trials activities. 



Priorities include clinical trials that are most needed in the next 10 years as well as those issues which need to be addressed to accelerate successful completion and dissemination of clinical trial results into clinical practice.

Improved efficiency of clinical trials

  • Continued support/expansion of infrastructure and best recruitment practices to facilitate timely and efficient completion of trials (networks, NIH-sponsored national IRBs, resources for study monitoring)
  • Reimbursement for Phase IIb-III Trials by CMS/payors for promising but clinically unproven therapies prior to reimbursement for clinical delivery of technology of care, requiring coordination of federal agencies.
  • Partnership with other countries to improve recruitment and generalizability
  • Minimization of overlapping and concurrent competing stroke trials

Improved trial design, conduct, and outcome assessment.

  • Centralized tools for trials for getting studies started and identification of well designed outcome variables that are consistent across studies
  • Training tools for new coordinators/investigators (building upon past successful trials)
  • Innovative statistical designs

Clinical trials that advance current evidence-based therapies or for those conditions without a proven therapy.

  • Acute ischemic stroke trials that test treatments potentially more efficacious than IV tPA alone or use imaging to test treatments in imaging selected subgroups beyond 4 ½ hours
  • Acute intracerebral hemorrhage (medical and surgical)
  • Trial of ruptured aneurysms
  • Trials of aggressive medical therapy vs. interventional treatment of asymptomatic carotid stenoses
  • Trials of new anti-thrombotic agents in patients without a cardioembolic source who fail aspirin
  • Trials of behavior change/primary stroke prevention (beyond scope of NINDS alone)
  • Phase I, II and III trials of neurorecovery therapies



Co-Chairs: Stephan Mayer, Neil Martin, Jaroslaw Aronowski

Members: E Sander Connolly, Nestor Gonzalez, Dan Hanley, J. Claude Hemphill, Richard Keep, Loch MacDonald, Adnan Qureshi, Lauren Sansing, Paul Vespa, Jeff Weitz, William Young

NINDS Liaisons: Tom Jacobs, Scott Janis


Seminal Advances in CNS Hemorrhage Research 2007-2011

Reducing ICH growth. The FAST and INTERACT Trials confirmed that acute hemostatic and BP reduction therapy can reduce active bleeding and hematoma volume in ICH. Unfortunately, no clear and consistent benefit on outcome was demonstrated. Recent metanalyses have determined that a reduction in absolute ICH volume growth of 6 to 12 ml is required to improve outcome. These studies have set the stage for future studies combining these interventions within a very early time window.

Treatment of vasospasm. The CONSCIOUS 2 and 3 trials showed that clazosentan, an endothelin receptor antagonist, reduces angiographic vasospasm after SAH when given in combination with nimodipine. CONSCIOUS 3 showed benefit on its primary outcome, which incorporated death, infarction or deterioration from vasospasm, and rescue therapy. However, no clear benefit on outcome was realized. The sponsor of this industry-funded trial is considering a final phase III trial comparing clazosentan head-to-head against nimodipine.

Clot evacuation and tissue neuroprotection. STICH-2 is continuing to explore the role of open surgical evacuation for lobar ICH. Preclinical research and phase II clinical trials have demonstrated the feasibility of early clot evacuation (CLEAR, MISTIE, STICH) and novel tissue protection strategies (deferoxamine, pioglitazone) for ICH and IVH. These studies have set the stage for further phase III trials including possible combined treatment trials.

Important New Concepts and Opportunities in CNS Hemorrhage Research

Electrically-mediated brain injury. There is now recognition that cortical spreading depression (CSD) and related phenomena, such as spreading depolarization and ischemia, are frequent events when humans with SAH and ICH are studied with electrocorticography. These phenomena, in combination with non-convulsive EEG abnormalities, are a potentially important and treatable mediator of secondary brain injury that deserve further study.

Early brain injury in SAH. With improvements in aneurysm repair and treatment for vasospasm, there is emerging recognition that early brain injury resulting from the primary bleeding event is an important cause of SAH morbidity and mortality. There is a pressing need for further experimental and clinical research to elucidate these mechanisms of injury.

Novel outcome measures and assessment techniques. The failure of studies such as FAST and CONSCIOUS-2 to demonstrate clinical efficacy despite evidence of benefit using intermediate endpoints (such as hematoma growth and angiographic vasospasm) highlights the need for novel clinical trial designs with endpoints designed to optimally capture clinically meaningful treatment effects.

CNS Hemorrhage Research Priorities Moving Forward

Blood and the neurovascular unit. Basic science research is needed to better understand the special nature of hemostasis and coagulation within the CNS and how it affects the neurovascular unit, with particular regard to electrical disturbances, cellular signaling, microvascular dysfunction, tissue inflammation, and matrix biology. There is still a need for more relevant models of acute hemorrhage-induced brain injury.

Advanced imaging and physiologic studies in humans. Human data using time-based acute phase neuroimaging, multimodality physiological monitoring, tissue/molecular/biomarker analysis and genetic profiling is needed to better define the time window, mechanisms, and clinical impact of potentially-modifiable primary and secondary injury pathways during the acute phase of bleeding. Target pathways should be identified and validated in preclinical experimental studies.

Surgical hematoma evacuation. Despite the negative results of the STICH trial, the proper role of surgical hematoma evacuation remains the most pressing unresolved clinical question in ICH management. More research is needed to better define the role of very early surgical intervention, and to develop novel image-guided minimally-invasive surgical interventions for deep hematoma aspiration. The phase II MISTIE trial and STICH-2 trials will prove useful in starting these efforts, but more work is needed.


FAST (industry-sponsored) found no clinical benefit in patients who received the hemostatic agent recombinant Factor VIIa early after acute ICH onset, despite a clear reduction in hematoma expansion. Focus has now turned to developing ways to identify ICH patients most likely to undergo hematoma expansion, such as contrast extravasation on CTA (the “spot sign”), as candidates for hemostatic therapy. This approach is currently being studied by STOP-IT (NIH funded) and SPOTLIGHT (Canada).

INTERACT, a phase II blood pressure lowering trial from Australia and China, found that lowering systolic blood pressure to 140 mm Hg within 6 hours of ICH onset decreased mean proportional hematoma growth. ATACH (NIH funded) tested the safety of progressive tiers of acutely lowered blood pressure. Pivotal phase III trials testing the impact of acute BP lowering on clinical outcome after ICH are in progress. These trials specifically address the compelling clinical question of blood pressure lowering, which was emphasized as a top priority in the 2005 NINDS ICH Workshop.

Intraventricular hemorrhage (IVH) has devastating effects on survival and recovery. CLEAR 1 and 2 provided crucial phase II dose-ranging and safety data for this intervention. CLEAR 3 is an NIH-funded phase III clinical trial of this intervention that is actively enrolling patients. MISTIE is an NIH-funded phase I/II trial of thrombolytic aspiration for deep ICH.

The STICH-2 trial is testing the hypothesis that early surgical evacuation improves outcome in patients with large superficially-located lobar hemorrhages.

Basic and translational research advances have focused on the role of inflammation and iron toxicity in creating peri-hematoma secondary injury in ICH. Deferoxamine has now been tested in a NINDS-sponsored phase I clinical trial in ICH patients; this represents the first translation of a potential treatment from preclinical animal models to humans.

The CONSCIOUS 2 and 3 trials (industry sponsored) evaluated clazostentan, an endothelin receptor antagonist, for the prevention of vasospasm after SAH. Although significant reductions in angiographic vasospasm and the need for rescue therapy were found, side effects such as hypotension and pulmonary edema resulted in no overall benefit in outcome. Driven by a compelling body of basic science research, these trials demonstrate the difficulty of successfully translating bench research to effective intervention at the bedside.

CoSBID is a European consortium of centers that has investigated the role of CSD and related electrical abnormalities detectable with electrocorticography as potential mediators of secondary injury after SAH. These studies have clearly established that these phenomena occur abundantly after SAH and ICH (as well as with ischemia and trauma) and are related to both ischemia and hyperemia. These studies have set the stage for further research to define the biological and clinical relevance of these phenomena, and to develop and test targeted therapeutic interventions.

MASH-2 was a European phase III trial that tested high-dose magnesium infusion as an intervention to reduce morbidity and mortality from vasospasm. No clinical effect was noted.

STASH and ALIASAH (NIH funded) are phase II trials of high-dose statin therapy and albumin, respectively, for the prevention of delayed ischemia after SAH. These trials are in progress.

Observational studies have documented an increase in the likelihood of good outcome when SAH and ICH patients are cared for in high-volume centers. This research supports the rationale for the development of comprehensive stroke centers and regionalization of stroke care. Outcomes research has also identified therapeutic nihilism (early DNR status without a trial of aggressive intervention) as an important determinant of death in the real-world setting. Economic forces and a lack of centralized policy directives have hampered to the establishment of stroke systems that ensure transfer to the most qualified institution.

There has been a shift in focus in basic research towards defining early brain injury mechanisms which may lead to therapeutic strategies following SAH. These include mechanical blast injury, acute ischemia, endothelial and microvascular dysfunction, and microthrombosis.

With the negative results of the CONSCIOUS trials indicating that the prevention of angiographic spasm does not necessarily translate into improved clinical outcome, basic science research has increasingly focused on the role of distal small vessel or microvascular spasm in producing delayed ischemia after SAH.

ARUBA is a phase III NIH-funded clinical trial designed to test whether functional outcome and the risk of spontaneous AVM rupture at 5 years with best medical therapy is superior to procedural intervention (embolization, microsurgical resection or radiosurgery). The trial is in progress.


Study of aneurysm and AVM growth and rupture

This should focus on chronic health risk factors, lifestyle, and integrate genetic, hemodynamic and treatment-related factors using advanced imaging, genetics, and biomarker analysis. The presence of inflammatory cells, active thrombosis and fibrinolysis, connective tissue degradation, or changes in flow hemodynamics can conceptually be detected using biomarkers or imaged in order to better understand who bleeds, and why they bleed. MR imaging of ultra-small particles of iron oxide (USPIO), for example, is one such method that holds promise. Also, vessel wall imaging with MR or optical coherence tomography may yield information about aneurysm formation, growth and rupture. In order to devise hemorrhage prediction models very large sample sizes are needed. This will require a sea change in the level of cooperation and collaboration of teams that care for and do research on this disease. Multicenter disease registries have the greatest opportunity to address this unmet need.

Vessel stabilizers to reduce the risk of primary or secondary hemorrhage

The concept of medical or biological “vessel stabilizers” for primary and secondary prevention of brain hemorrhage due to ICH, SAH and AVM is attractive, but little work has been done in this area to date. Attractive biological or cell-based therapies might influence vascular remodeling and the formation or degradation of vascular matrix tissues. Approximately 20% of AVM patients receive no treatment, and radiosurgical treatment accounts for roughly 30%, but there is a long period between irradiation and nidus obliteration, and then only about 70-80% achieves eradication. Many small unruptured aneurysms are diagnosed and followed with serial imaging for years. A major impediment in this area is the lack of animal models that recapitulate the long-term natural history of SAH, ICH, and AVM.

Biology of vascular hemostasis

There is extensive evidence implicating thrombin in the pathogenesis of secondary tissue injury after ICH, but how the process of hemostasis contributes to BBB permeability and extension of bleeding is poorly understood. There is emerging evidence that the contact system of coagulation, which is linked to fibrinolysis, inflammation and vascular permeability, plays a role in this process. Factor XII or factor XI represent novel targets for attenuation of coagulation without disruption of hemostasis. The reduction in intracranial hemorrhage observed with targeted inhibitors of thrombin or factor Xa compared with warfarin is novel and exciting. Plasma kallikrein, especially in the setting of hyperglycemia and hyperosmolality, has emerged as a potent antihemostatic regulator (by preventing platelet-collagen attachment), and is a new potential contributor to a greater hematoma expansion in the context of hyperglycemia. This priority is echoed in the SPRG CHE report.

Clinical trials of treatments to prevent early rebleeding

Further trials of hemostatic therapy for ICH should build on current efforts to identify high-risk patients with CTA and other forms of neuroimaging. In the meantime, preclinical studies are needed to identify more effective and less expensive alternatives to rFVIIa. BP control strategies will need to be intimately incorporated into these trials.

Emergency reversal of anticoagulation

Anticoagulation and antiplatelet therapy contributes to the risk of ICH, and may promote the active bleeding process. There is a paucity of data on what emergency reversal protocols are most effective to limit active bleeding and improve outcome. Further studies exploring the optimal methods for reversal of new anticoagulants, such as dabigatran, rivaroxaban and apixaban, are needed. This priority coincides with that of the SPRG CHE group.

Blood product detoxification

This concept is not new, but is important, and progress has been made. Robust pre-clinical research with deferoxamine (an iron chelating agent) over the past several years has demonstrated in animal models that deferoxamine has a potent therapeutic effect in ICH. A phase II study testing the feasibility and tolerability of deferoxamine in ICH patients found this treatment to be feasible and safe. This line of investigation may provide proof of concept that the neutralization of toxic products of hemolysis (e.g. hemoglobin, heme or iron) is a viable therapeutic target in ICH. Work in this area in ICH models can also be extended to direct clot toxicity from SAH.

Cerebral inflammatory response

It is now well-established that thrombin engenders a cerebral tissue inflammatory response, but this process is incompletely understood. Recent work has identified key roles for innate immune receptors (protease-activated receptors, CD36, TLR4), complement, and blood-derived leukocyte recruitment in the inflammatory response. These pathways can now be exploited for new therapeutic targets. Importantly, future work must distinguish between inflammatory responses that contribute to injury from those which aid in repair and recovery (e.g. microglial activation after thiazolidinedione treatment aids in phagocytosis while diminishing inflammation). Little progress has been made since 2007 regarding the need for analysis of human tissue samples to clarify the translational relevance of basic science advancements.

Hematoma resolution

Though it seems intuitive that surgical evacuation of a hematoma might improve outcome, therapeutic strategies aimed at promoting biological processes that promote hematoma resolution are new. Recent experimental studies have demonstrated that activators of transcription factor PPARy can be used to augment the phagocytotic capacity of microglia and macrophages toward erythrocytes and to speed-up hematoma resolution after experimental ICH3. Based on this study, a phase I clinical study with pioglitazone in patients after ICH has been initiated, and hematoma resolution is considered as one of the outcome variables.

Development of small animal models

Although progress is being made, there is still a paucity of animal models that recapitulate the dynamic bleeding process of acute ICH that reproduce the known clinical effects of SAH, or the distinct anatomy of brain AVM. Breakthroughs are needed. Collagenase injection is the most commonly-used model for acute ICH, but does not replicate the biology of hypertensive ICH. For brain AVM, a mouse model using Alk-1 deletion and VEGF stimulation that displays many—but not all— of the phenotypic characteristics of the human disease has been reported. Further refinements of the model are needed to more precisely phenocopy the human disease as a platform for screening medical therapies (i.e. vascular stabilizers). AVM animal models should also focus on the role of post-natal angiogenic stimulation, and potential overlap with the Hereditary Hemorrhagic Telangectias in terms of causative mechanisms such as genes or pathways related to disordered TGF-ß signaling.

Human Tissue Studies

Modest success has been made in identifying angiogenic, molecular, and inflammatory pathways of AVM and perihematomal brain injury by studying tissue from resected lesions. However, these observations have been largely descriptive and not coordinated with studies of important clinical events including growth and symptom generation. Human tissue studies have been underemphasized to date. Larger, more concerted studies coordinated via a disease registry network could address this need.

Systemic immunological dysfunction

Recently, it became clear that brain trauma, including cerebral hemorrhage, proportionally to its severity and volume, triggers an innate immunological and systemic inflammatory response that is proportional to the severity of the trauma or the volume of the bleed. . These responses may lead to increased susceptibility to infections, and may affect brain chemical homeostasis and increase susceptibility to secondary injury. Modulation of these responses could represent an important target for treating ICH related brain injury and its associated complications.

Early brain injury after SAH

There is emerging consensus that with improved treatments for vasospasm, early brain injury, defined as the initial damage resulting from the acute effect of hemorrhage, has become the most important determinant of patient outcome. This complex phenomenon includes elevation of intracranial pressure and reduction of cerebral perfusion pressure, global cerebral ischemia, brain inflammation and edema, acute vasospasm and microcirculatory dysfunction, and BBB breakdown. There is a pressing need for further experimental and clinical research to elucidate these mechanisms of injury. Resuscitation strategies directed at early brain injury will depend on clinical-pathological correlation of early events to long-term outcome.

Electrically-mediated secondary brain injury

There is small but growing body of human data implicating cortical spreading depression, depolarization, and ischemia as a related set of electrical derangements that may contribute to hemorrhage-related secondary tissue injury, secondary ischemia, or hyperemia and BBB dysfunction. These phenomena result in cellular hypermetabolism, near-complete breakdown of ion gradients, neuronal swelling, and either increases or decreases in regional microvascular tone, which can in turn engender lesion progression due to hyperemia or hypoperfusion. The interplay between these electrical phenomena and non-convulsive seizure activity on EEG is unknown.

Cytoprotection based and the neurovascular unit

The pathogenesis of cerebral hemorrhage induced secondary tissue injury is highly complex and affects all brain cell types. The primary unit of intercellular organization in the brain is the neurovascular-unit (NVU), which consists of microvessels (endothelium, basal lamina, astrocyte end-feet), intervening astrocytes, neurons and their axons, and other supporting cells (e.g. pericytes, microglia, and oligodendroglia). The effect of extravasated blood on intracellular signaling and its role in producing BBB and autoregulatory dysfunction has become a major new topic for investigation. Pleiotropic restorative cytoprortection to preserve the structural integrity and function of the NVU is a promising new vista for further research. This priority coincides with priorities set forth by the SPRG CHE group.

Pediatric hemorrhagic stroke

Germinal matrix hemorrhage is the leading cause of childhood hemorrhagic stroke. In recent years there has been little in the way of organized research efforts for this condition. Efforts to support pediatric stroke and neonatology research should include specific programs directed at neonatal IVH.

Impact of systemic physiological derangements

Brain hemorrhage patients often experience a protracted course in the ICU, during which they are subject to many system physiological derangements, such as fever, hyperglycemia, hypoxia, arterial hypertension, anemia, SIRS, and a hypermetabolic state. Clinical epidemiological studies have increasingly linked these insults to poor clinical outcome. There is a need for studies of specific ICU management strategies that modify or ameliorate these secondary physiological insults. This priority overlaps with the SPRG AST report, and was cited in the last HEM report, but little progress in this area has been made to date.

SAH outcome measures

SAH is unique among other forms of stroke in its tendency to cause disabling neurocognitive rather than motor and physical disability. Apart from the existing ordinal scale measures of global disability and handicap (i.e. the modified Rankin Scale), there is a need for novel and robust measures that capture more subtle cognitive, emotional, and quality of life outcomes after SAH.

Comparative effectiveness research

There is considerable data indicating that patients with severe hemorrhagic stroke have better outcomes when treated in high-volume centers. Outcomes research has also drawn attention to the powerful impact of therapeutic nihilism on survival and recovery from CNS hemorrhage. However, economic incentives and lack of clear guidelines for how to regionalize care for high risk patients (as occurs with trauma) have led to a breakdown of existing stroke referral systems over the past 5 years. There is a need for comparative effectiveness research to identify clinical practices that correspond with improved outcome and cost-effective care for this high-risk population. The role of telemedicine in developing a “golden hour” model for brain resuscitation in acute ICH and SAH should be included in telemedicine research initiatives directed at acute ischemic stroke treatment.

Disease registries and research networks

Most cohort-based research on SAH, ICH, and AVM is based on large single center studies. Ongoing comprehensive disease registries using simple common data elements, internet-based acquisition protocols, and pooled resources for outcome assessment can help achieve more comprehensive understanding of natural history, prognosis, treatment outcomes, and optimal treatment strategies. Biomarkers (i.e. SNPs, inflammatory mediators) are needed that are associated in a predictive capacity to natural history and treatment outcomes. In AVM research there are inadequate means to study the molecular pathophysiology of the disease in a longitudinal context in intact subjects. Such studies will be possible with improved MR or PET tracer development to label molecular level participants in key mechanistic pathways. These registries should be linked to existing infrastructures created for clinical trials or other research networks, including the NeuroNEXT program. This approach also holds the best promise for understanding the affect of race and gender on disparities in care, disease course, and outcome.

Recovery and neuroplasticity after hemorrhage

Little is known about out the mechanisms of recovery that are unique to brain hemorrhage, or how to augment the endogenous healing process. Serial imaging studies relating acute injury patterns, brain reorganization and plasticity, and assessments of functional outcome may provide important insights. The anatomical or functional substrate of injury that underlies the cognitive and emotional disturbances that often occur after SAH needs to be defined. There is a large gap between hemorrhagic and ischemic stroke in research on mechanisms of recovery.

Cellular replacement therapy

Cellular therapies using pluripotential stem, progenitor, or engineered cells derived from bone-marrow or umbilical cord blood have demonstrated cytoprotective and restorative properties in ischemic models. A growing body of evidence has accumulated suggesting that cell-based therapy may also be a promising mode of intervention after cerebral hemorrhage.


Several genetic variants including single nucleotide polymorphisms in the key components of pathways linked to cerebral hemorrhage, aneurysm, and AVMs pathogenesis has been identified and demonstrated to have association with the clinical outcome. Examples of these gene products include: complement factor H, APOE, p53, TGF-beta receptor, haptoglobin, and angiopoietin-like 4, Catechol-O-methyltransferase. Genetic studies—even in the sporadic, non-inherited form of aneurysm and AVM — are very likely to give important clues to the biology of these diseases. There is a need to better understand the effect of genetic polymorphisms on both susceptibility to primary bleeding events, and as modulators of secondary tissue injury. Substantial progress is being made in our ability to rapidly and affordably assess genetic variation, including epigenetics. Both common-variant (GWAS) and rare variant approaches are feasible. Also critical is the need to develop RNA transcript expression or protein biomarkers in biological fluids and circulating cells of aneurysm and AVM patients.


1) Basic science research to better understand how blood in the CNS damages the brain. Basic science research is needed to better understand the special nature of coagulation of blood within the tissue milieu of the CNS, and how it differs from normal intravascular hemostasis. More progress is needed to elucidate the pathophysiology of neuro-hemo-inflammation beyond current insights that have implicated thrombin as an important trigger of this process. Research is needed to determine how blood affects the neurovascular unit, with particular regard to electrical disturbances, cellular signaling, microvascular dysfunction, tissue inflammation, and matrix biology. There remains a pressing need for more relevant models of acute hemorrhage-induced brain injury.

2) Advanced imaging and physiologic studies in humans. Human data using time-based acute phase neuroimaging, multimodality physiological monitoring, tissue/molecular/biomarker analysis and genetic profiling is needed to better define the time window, mechanisms, and clinical impact of potentially-modifiable secondary injury pathways after hemorrhagic stroke. Studies are needed to define the metabolic abnormalities, cellular inflammatory response, and derangements of electrical and molecular intracellular signaling that characterize the perihematomal penumbra and the brain after SAH. Clinical imaging studies are needed to better define the predictability, timing, and pathophysiology of the active bleeding process in ICH. Research is also needed to fully characterize mechanisms of acute brain injury in SAH; processes specific to perihematomal inflammation, BBB disruption, and edema formation; and advanced imaging of brain AVM including flow dynamic patterns, wall stress, and inflammation to assess future hemorrhage risk.

3) Surgical hematoma evacuation. Despite the negative results of the STICH trial, the optimal role of surgical hematoma evacuation remains the most pressing unresolved clinical question in ICH management. A major opportunity to mitigate delayed neuronal injury after ICH remains removal of the “offending pathogen” -- part or all of the hematoma. Surgery is still frequently performed for this disease, but practice patterns vary widely, and the benefit remains uncertain. Trials that remove clinically significant volumes of clot, as determined by translational and human data, are critical to answer the simple question: does clot burden reduction help? More research is needed to better define the role of very early surgical intervention, possibly in combination with heostatic therapy, and to develop novel image-guided minimally-invasive surgical interventions for deep hematoma aspiration. Phase III clinical trials that fully test the volume reduction hypothesis are critical to all other research efforts in brain hemorrhage.



Co-Chairs: Gregory del Zoppo, Maiken Nedergaard, David Pinsky, Jeffrey Weitz

Members: Ulrich Dirnagl, Paula Dore-Duffy, Peter Gross, Antoine Hakim, John Hallenbeck, Richard Keep, James Morrissey, Bruce Ransom, Marc Simard, Lydia Sorokin

NINDS Liaison: Thomas Jacobs

Exciting progress has been made over the last decade in our understanding of the complex events that accompany focal ischemia and hemorrhage; however, numerous problems remain that have yet to be solved.


Broad Conceptual Advances

  • Characteristics of the neurovascular unit and its relevance to focal ischemia and hemorrhage
  • Relationship of the neurovascular unit to hemostasis
  • Innate inflammation

Unresolved Scientific Questions: The neurovascular unit and hemostasis

  • Integrity of the blood-brain barrier
  • Interaction of cellular and structural components of the neurovascular unit/
  • Vascular hemostasis in the central nervous system (CNS)
  • Neuron-vascular relations
  • Innate inflammation

Resources Required for Resolving Outstanding Questions: The neurovascular unit and hemostasis

  • Need for appropriate models for the neurovascular unit and CNS hemostasis
  • Understanding of the components of the neurovascular unit
  • Translational models
  • Introduction of evaluation of comorbidities to problems of neurovascular unit

New Key Research Areas that have Emerged in the Neurovascular Unit

  • Roles of CNS microvasculature in the neurovascular unit
  • Innate inflammation
  • Hemostasis in the CNS and role (s) of CNS hemostasis in hemorrhage


The cerebral microvasculature is an integral part of cerebral tissue that reacts to focal ischemia as swiftly as neurons.

Although the interweaving of cerebral microvessel biology and hemostasis has been exploited clinically for some time, recent focus on the microvasculature enforces the notion that both local and remote networks, at several levels, are involved in the microvessel responses to cerebral injury. Over the ten years of the SPRG, and particularly during the last five years, clear insights into the increasing complexity of these networks have emerged.

To manage this complexity, the concept of the neurovascular unit (NVU) continues to provide an appropriate structural and hypothesis-generating framework with which to understand how the cerebral microvasculature interacts with neurons, glia, and other supporting cells of the central nervous system (CNS). This conceptual framework provides much-needed direction for interpreting tissue responses, experimental research, and clinical trials outcomes that relate the endothelium and associated glial and matrix components of cerebral microvessels to neuron function and the evolving neuronal injury caused by ischemic stroke or hemorrhage. The NVU consists of microvessels (endothelium, basal lamina, and astrocyte end-feet), pericytes, intervening astrocytes, the neurons and their axons, and other supporting cells (e.g. microglia and oligodendroglia). Breakdown of normal endothelial cell-matrix-astrocyte relationships contribute to profound changes in neuron function and in the architecture of the NVU. Consequently, understanding the inter-relationships of the components of the NVU provides a i) conceptual framework for understanding brain function in health and during ischemic and hemorrhagic stroke, and ii) solid foundation for the rational development of treatment strategies aimed at limiting the impact of focal ischemia and hemorrhage in the brain.

Relevance of the neurovascular unit

The NVU links the endothelium and microvessels with the neuropil. Breakdown of normal endothelial cell-matrix-astrocyte relationships contributes to the profound changes in neuron function and in “neurovascular unit” architecture that occur with ischemic stroke or cerebral hemorrhage. Normal endothelial function and microvessel reactivity are essential for maintaining the blood-brain barrier. The cerebral endothelium and microvasculature have unique features that distinguish them from vessels elsewhere. Furthermore, the dependence of the permeability barrier on glial and pericyte participation, the different characteristics of the grey matter and white matter, and the tissue/context specificity of vascular responses in the CNS under normoxia and in response to injury complete the conceptual NVU framework.

There is mounting evidence that the extracellular matrix (ECM) plays a significant role in development and maintenance of microvascular integrity, functions that are perturbed by focal ischemia and hemorrhage. The pluripotential role of pericytes is recognized, but the interaction and contribution of these cells to microvessel stability and response to injury is only beginning to be explored and realized. Perivascular cells, the network of astrocytes and other glial cells, as well as components of the white matter also contribute to cerebral tissue stability and modulate the response to ischemia. Subtle but real differences in the contributions of these various cell types to the structural stability of cerebral tissue are responsible, at least in part, for differences in species-specific responses to focal ischemia. Calcium-mediated signaling processes occur in astrocytes and endothelium. The cross-talk between these two cell compartments, exemplified by their production of the ECM in the cerebral microvasculature, is likely to involve similar signaling processes, including integrin-ECM interactions and tensor processes. How the pericyte contributes to these interactions and modulates them appears to be of major importance. The ECM, matrix-degrading enzymes and local and systemic inhibitors are important regulators of cell signaling events at the basal lamina and the cell surface. The glycocalyx, a specialized ECM, is likely also to be important, but its role in signaling is unknown.

Separately, axons and their transmission are essential conduits that inform the environment of the integrity of the NVU. Injury to axons, neurons, the supporting glia, or the microvasculature can silence the NVU. Understanding the sensitivity and responses of axons to ischemic injury has advanced our appreciation of NVU integrity.

Relationship of NVU to vascular hemostasis

From the hemostasis perspective, the endothelial cell is central to the NVU. However, its function is regulated in complex ways by input from adjacent pericytes, astrocytes, and neurons. Cerebral endothelial cells differ from those in other organs and tissues. For example, thrombomodulin and endothelial protein C receptor (EPCR) expression is lower in the brain, and the distribution of basal lamina components is more variable than in other vascular beds. Even within the brain, regional differences in endothelial cells and their responses have been reported, which probably reflect, at least in part, the impact of signaling between endothelial cells and other components of the NVU.

The apparently homogeneous distribution of tissue factor (TF), a major initiator of coagulation, in cerebral grey matter (in excess of white matter) is a unique feature of the CNS. TF initiates thrombin generation when the blood-brain barrier is compromised and the neuropil is exposed to plasma. Unless regulated, thrombin induces fibrin formation and subsequent thrombotic occlusion of the microvasculature during ischemia (focal “no-reflow”). If the microvessels rupture, hemorrhage occurs until sufficient thrombin is generated to arrest the bleeding. This may be greater in gray matter than in white matter.

In addition to its role in coagulation, thrombin also plays a part in cell signaling, inflammation, and cell proliferation. These diverse activities are mediated, at least in part, by the capacity of thrombin to signal via protease activated receptors (PARs). PAR-1 is not only expressed on platelets, but also is highly expressed on astrocytes of white and grey matter and moderately expressed on neurons. PAR-2, PAR-3, and PAR-4 are also variably expressed during brain development; expression of all of the PARs can be up- or down-regulated in response to cerebral ischemia and hemorrhage. Thrombin activation of PAR-1 on platelets leads to platelet activation and aggregation. Activated platelets promote thrombin generation by serving as a procoagulant surface on which clotting factor complexes assemble, whereas platelet aggregates contribute to the coagulum that occludes the microvasculature and slows hemorrhage. In contrast, thrombin-mediated activation of PAR-1 on astrocytes and neurons trigger calcium-dependent signaling events that can influence inflammation and cell survival. The role of PARs in health and cerebral ischemia or hemorrhage still remains to be fully elucidated.

Recent studies have delineated another mechanism by which the coagulation system can be activated. Polyphosphates released from activated platelets or from other dead or dying cells in the CNS can trigger activation of the contact pathway, which leads to thrombin generation in a TF-independent fashion via the contact pathway.

The results of three recent clinical studies that have compared new oral anticoagulants that either target thrombin (dabigatran) or factor Xa (rivaraoxaban or apixaban) with warfarin, a vitamin K antagonist, for stroke prevention in patients with non-valvular atrial fibrillation (AF) indicate that the frequency of intracranial hemorrhage is lower with the targeted agents than it is with warfarin. These findings provide another indication of differences in the microvascular hemostatic response to ischemic challenge in the CNS, and open the door for molecular studies of mechanism. They also stress the need for animal models in which to assess the risk of CNS bleeding with novel anticoagulants.

The integrity of the endothelium depends on its interaction with platelets and other cellular and plasma components. Normal endothelium has antithrombotic properties that can be perturbed under conditions of ischemia. Reduced expression of anticoagulant proteoglycans, thrombomodulin, and EPCR, increased release of type 1 plasminogen activator inhibitor (PAI-1) and decreased synthesis of plasminogen activators, ectoenzymes, prostacyclin and nitric oxide can promote vasoconstriction, platelet activation and aggregation and fibrin formation. Tethering of leukocytes and/or microparticles to the endothelial cell surface may compound the problem because released free radicals and proteases can activate or damage the endothelium, alter vascular permeability and disrupt other components of the NVU.

There is growing recognition of the importance of transcriptional mechanisms in the responses of NVU cells to ischemia and hemorrhage. Both adaptive and maladaptive responses are involved. Mechanisms driven by the transcription factors, hypoxia inducible factor 1, specificity protein 1 and nuclear factor kappa B, appear to be of particular importance in the contexts of CNS ischemia and hemorrhage. Distinguishing transcriptional mechanisms from post-translational mechanisms is important because, from an operational viewpoint, transcriptional mechanisms occur more slowly. Consequently, new targets that appear subsequent to transcriptional up-regulation (e.g. matrix proteases) are likely to be associated with longer treatment windows than those that appear subsequent to translation. Understanding the role of transcriptional processes in endothelium and other members of the NVU is likely to provide a wealth of potential new treatment targets with favorable treatment windows.


Over the past five years, at least two research areas appear to have receded. First, mounting evidence suggests that microparticles are less relevant for hemostasis and more important for cell-cell communication. For example, microparticle-endothelial interactions may modulate endothelial cell responses to local injury, thereby contributing to the focal “no-reflow” phenomenon of ischemic stroke. Similarly, microparticle-leukocyte interactions may modulate the inflammatory response both systemically and locally at sites where the blood-brain barrier is compromised.

Second, the sophisticated imaging modalities discussed in the interim report continue to provide useful information in select centers, but have yet to reach the level of distribution that is needed to capitalize on their potential. Even with these tools, limitations in resolution restrict their utility to the study of phenomena occurring within the superficial layers of the CNS. However, there is hope that newer methods of ultrasonography may help to image events deeper in the CNS, and several laboratories have developed relatively novel imaging materials to monitor cell activation and signaling events.

Responses to unresolved scientific questions

Several observations have been made that underscore the importance of the neurovascular unit as a framework to advances in cerebral microvascular research: 1) more opportunities for convergence of the various disciplines represented by the neurovascular unit are required, 2) new studies are needed to further define the unique aspects of hemostasis in the CNS, 3) ongoing studies of the matrix biology of the cerebral microvasculature and of the neuropil must be advanced, 4) the significance of axonal transmission for signaling the neurovascular unit to the environment requires further definition, 5) an emphasis on the CNS tissue contributors to innate inflammation and their contributions to the evolution of the ischemic and hemorrhagic injury must be maintained, and 6) the need for appropriate models that are truly translational.

All of these scientific advances/interests must be represented by ongoing translational efforts that identify and test new targets for tissue salvage. Identification of new and appropriate treatment targets requires a deeper understanding of the injury processes at a fundamental level.

General conclusion

Investigational work since the interim report has provided new research momentum and more insights into the NVU in health and disease. The adaptive response of the NVU to focal ischemia has given further impetus for the study of matrix biology, signaling events that occur through cell-matrix interactions, unique features of hemostasis in the CNS, and innate inflammation/immunity in the brain. As will be seen, recent clinical studies have offered new and exciting hypotheses that link hemostasis and hemorrhage risk to the integrity of the neurovascular unit.


Foremost, it has long been held that model systems of focal ischemia do not adequately capture the features of ischemic stroke or of hemorrhagic stroke, and their consequences, that are seen in humans. The implication is that either: i) the known features of model systems that are cognate should be exploited and further employed, or ii) energy or support should focus on molecular and physiologic characterization of those features of the model systems that most closely represent the role of the NVU in the clinical setting. Transformative efforts are essential to render available model systems truly translational. Although it is well-known that after focal cerebral ischemia microvessel responses to injury, and microvessel interactions with neuronal tissue change over time, the temporal course of the factors that regulate these NVU responses to ischemia and their transcriptional control are largely unknown. Differences in the endothelium, heterogeneity among astrocytes in their perivascular distribution, and cross-talk between the endothelium and astrocytes are important targets.

Integrity of the blood-brain barrier

Further fundamental understanding of the blood-brain barrier, capitalizing on mechanisms of vascular biology in other organ systems, is required. Research on the integrity of the blood-brain barrier from the perspective of tight junction protein expression and how it might be modified needs new insights. For example, it is now apparent that, for instance, endothelial cell adhesion receptor-matrix interactions influence tight junction expression and the integrity of the permeability barrier. Matrix composition in CNS vessels varies along the vascular tree. This and related observations require further work to better understand the mechanisms responsible for permeability barrier function, and how to manipulate their features. The blood-brain barrier also possesses an array of xenobiotic efflux transporters that may limit the penetration of many potential stroke therapeutics into the brain. Recent evidence has shown that these transporters undergo substantial regulation (e.g. by inflammatory mediators). Other transporters may also be newly up-regulated in response to injury; some may selectively facilitate influx of RNA and DNA mimetics into the endothelium. More work is needed to determine whether manipulating the regulation of these transporters will enhance drug delivery to the brain.

Cellular components of the NVU

Much has been learned about the function of i) astrocytes, which rapidly transduce information between the microenvironment and other brain cells, and ii) the pericytes which can induce changes at the cerebral capillary level. New methods, such as electrical or optical recordings of calcium transients, have been developed to follow activity in single cells or small groups of cells. Astrocytes are central to the NVU because they participate in microvessel integrity and maintain the neuron microenvironment. Thus, astrocytes metabolize glutamate for consumption by the neuron and signal to each other via calcium-dependent mechanisms that depend on ATP and glutamate. Death of astrocytes proceeds more slowly than that of neurons. Consequently, therapies aimed at astrocyte preservation may attenuate neuronal death during periods of ischemia.

The pericyte has multiple roles in the function of the NVU; it is a unique cell with macrophage and smooth muscle cell properties. The role of pericytes is emerging now that better methods are available for their identification. Because of their proximity to the endothelium, pericytes may influence tight junction integrity and vascular permeability, either by secretion of soluble factors by contributing to the molecular composition of the endothelial basal lamina, or by affecting astrocyte end-feet. Under inflammatory conditions, pericytes release substances, such as cytokines, free radicals, and matrix metalloproteinase-3 (MMP-3), as well as other proteases that can disrupt the NVU. Consequently, like macrophages, pericytes may play a role in innate inflammation. In support of this concept, MMP-3-deficient mice exhibit an attenuated inflammatory response and have fewer infiltrating neutrophils. Pericytes also have properties similar to those of stem cells. If they are pluripotential, they may have a role in the angiogenesis after cerebral ischemia.

Alterations in the microvessel portions of the NVU underlie vascular cognitive impairment. Amyloid deposition disorders can disrupt the microvasculature and neuropil, thereby altering functional properties of the NVU. How ?-amyloid alters microvessel integrity is poorly understood. Alterations in cell-matrix interactions offer one potential explanation.

Vascular hemostasis

Regulation of vascular hemostasis in the CNS is largely unknown. Additionally, how thrombin generation increases blood-brain barrier permeability is not understood. Emerging evidence that the contact system plays a part in this process identifies factor XII or factor XI as novel targets for attenuation of coagulation without disruption of hemostasis. Studies in factor XII or XI deficient mice, key components of the contact pathway, have shown that this pathway contributes to thrombus stability even though it is not important for hemostasis.

The reduction in intracranial hemorrhage observed with targeted inhibitors of thrombin or factor Xa compared with warfarin remains unexplained and offers a unique opportunity for reverse translation; in vitro and animal models are needed to explain these clinical observations. Finally, the impacts of age, hypertension, and diabetes mellitus on NVU responses are largely understudied. Of these, age is likely to be the easiest element most amenable to examination in model systems and in humans.

Neuron-microvessel relationships

Even though spreading depression, which is mostly described in model systems, is a product of ischemia and post-ischemic neuron injury, it has a microvascular connection that is not yet fully defined. As exemplified here, the relationship of model systems to human CNS disease requires further evaluation. Studies exploring these links will advance our understanding of how electrical alterations in the CNS and vascular responses may contribute to the development of CNS injury in patients with stroke and cerebral hemorrhage.

Innate inflammation

Importantly, there is a resurgence of interest in the innate inflammatory responses of the CNS to injury. Inflammation and hemostasis are linked and dependent on cross-talk among all of the components of the neurovascular unit during ischemic injury. The peripheral inflammatory system also contributes to this process (e.g. PMN leukocytes, macrophage/monocytes). Where the inflammatory response ends and recovery begins is unknown. A better understanding of these processes requires fundamental work on the innate (glial) and peripheral (leukocyte) inflammatory processes, CNS lymphatics, investigations into the role of “fingerprinting” of lymphocytes by CNS antigens, and better understanding of CNS tissue to plasma and blood cells in health ad disease.

While information is required on CNS-specific hemostatic mechanisms, a greater understanding of the extravascular role of proteases within the coagulation and the fibrinolytic systems is also needed to better understand the effects of cerebral hemorrhage and microbleeds, and to identify relevant treatments. In doing so, it is important to recognize that these proteases not only influence hemostasis and hematoma resolution, but also edema formation, vascular and neural function, and repair. Knowledge of the extravascular role of these proteases and their interaction with the inflammatory system is important in determining how best to modulate these systems to prevent or treat cerebral hemorrhage.

Model systems

Understanding the interactions of the cerebral microvasculature with the surrounding tissue (neuropil), and the responses of the hemostatic system to injury in the CNS relies heavily upon appropriate models. Unfortunately, currently available model systems for focal cerebral ischemia and hemorrhage do not adequately reflect the human disease. Discrepancies between models and the human disease may be due to fundamental species differences, the “lack of scalability” from small to large brains, and/or other factors. Furthermore, cell culture, ex vivo, and animal models that can be used to consistently and predictably translate fundamental discoveries in humans to models to humans and from humans to models, in the setting of ischemic or hemorrhagic stroke are lacking.

To overcome this problem, we need: i) a clear understanding of the interaction of the components of the neurovascular unit, ii) good translational models with proven records, that demonstrate the temporal sequence of events within the NVU that display/recapitulate the evolution of ischemic and hemorrhagic injury, particularly how the cellular components of the unit interact with each other (e.g. to modulate barrier permeability, affect hemostasis, alter glial responses, mitigate axonal injury, and affect neuron survival), iii) emphasis on the impact of age (and other comorbidities such as hypertension, obesity, and diabetes), and iv) imaging modalities that can distinguish grey and white matter responses and provide real time information including changes in metabolism. These approaches should be convenient, well-established, and well-funded.


Primary among the research areas is the need to further progress in understanding the components of the NVU and the relationships of cerebral microvascular responses to those of the neuropil during focal ischemia. Once thought to be simple conduits of flow, the cerebral microvasculature is now known to be dynamic and pluripotential in its acute and chronic responses to focal ischemia and reperfusion. The CNS microvasculature is unique, and stands as a part of the neurovascular unit; exactly how it differs from other microvessel beds requires further study. How microvascular endothelium and astrocytes communicate with each other, the roles of the intervening ECM, and how these interactions are modified by amyloid deposition, age, and other factors under normoxia and ischemia, provides an exciting avenue of research. This alone is likely to inform new therapeutic directions and explain current research limitations in ischemia, intracerebral hemorrhage, vascular dementia, and amyloid deposition disorders.

A second area of growing importance is innate inflammation in the CNS, under the conditions of health, focal ischemia, and hemorrhage. The role of innate inflammation in the CNS is gaining increased attention, with increased understanding of i) how the peripheral inflammatory network affects the CNS, ii) the early roles of PMN leukocytes, which have been difficult to target, iii) the role of complement components in modulating inflammation, and iv) emerging links between the contact pathway of coagulation and regulation of inflammation and vascular permeability, and the role of innate inflammation in the CNS is gaining increasing attention.

A third important and new avenue of inquiry, CNS hemostasis, is highlighted by the reduced rates of intracerebral hemorrhage observed with the new oral anticoagulants relative to warfarin when used for stroke prevention in patients with AF, and emerging evidence that the contact pathway plays a role in CNS thrombosis. The observation that in models of stroke, ischemic injury is reduced in factor XII-/- or factor XI-/- constructs is mirrored by the relative decrease in stroke incidence in patients with congenital factor XI deficiency. These findings identify new avenues for research. Agents that target factor XIIa or XIa may inhibit coagulation without affecting hemostasis; properties that would render them ideal for the prevention or treatment of stroke. Understanding and exploiting these phenomena requires further investigation into the unique aspects of the CNS microvasculature, better understanding of the development and maintenance of the permeability barrier, and additional identification of the proteases, receptors, ligands and inhibitors that regulate CNS hemostasis. Such research will help to identify new targets to more safely attenuate coagulation without impairing hemostasis.

A better understanding of CNS-specific hemostatic mechanisms (and their link to other processes, e.g. inflammation) is also crucial for identifying new modalities to prevent intracerebral hemorrhage and hematoma expansion, and to best reduce and manage the risk of cerebral hemorrhage in patients on anticoagulants. As well as continuing work developing agents to prevent/reduce cerebral hemorrhage, and to reduce the brain injury due to clot-derived factors, there is new interest in modulating hematoma resolution. The latter requires mechanistic understanding of the interaction between the inflammatory system and different proteases in the coagulation and fibrinolytic systems.

In brief, the three future directions are

  • Further understanding of the components and inter-relationship within the NVU during focal ischemia
  • Innate inflammation under conditions of health, focal ischemia, and hemorrhage
  • First steps at understanding how the brain and its vasculature regulate hemostasis uniquely and the impact of hemostasis in response to injury (focal ischemia and hemorrhage).



Co-Chairs: Ralph Sacco, Lynda Lisabeth, Brett Kissela 

Members: Heather Fullerton, Philip Gorelick, George Howard, Steve Kittner, Judith Lichtman, Matthew Reeves, Philip Wolf, Dan Woo

NINDS Liaisons: Claudia Moy, Salina Waddy


Epidemiologic research is critical for understanding the public health burden of stroke. Through epidemiologic research, population subgroups at the greatest risk of stroke are identified allowing for the prioritization of future research and funding. High quality observational epidemiologic studies are critical in identifying new associations and relationships which can be translated into the development of interventions to reduce the public health impact of stroke through clinical and behavioral intervention trials. Some of the observed reduction in stroke incidence and mortality, as well as improved outcomes in recent stroke clinical trials, has been attributed to more intensive control of risk factors. A great deal of progress has been made in stroke epidemiology since the last SPRG. Staff identified at least 35 grant awards (23 NINDS, 12 non-NINDS) since 2007 that were related to stroke epidemiology. Advances are summarized in a variety of areas:

  • Common Data Elements (Previous Priority #1) – NINDS initiated a Common Data Element (CDE) project to facilitate the combining of epidemiologic data from multiple sources and studies
  • Linkage between Epidemiological Data and Administrative Data (Previous Priority #1) - Several NIH funded grants are linking population-based epidemiologic studies to Medicare data, allowing for the identification of strokes
  • Stroke Trends (Previous Priority #2) – New epidemiologic data on temporal trends in stroke incidence have been published from the Framingham Heart Study, the Greater Cincinnati-Northern Kentucky Stroke Study, and the Brain Attack Surveillance in Corpus Christi (BASIC)
  • Epidemiologic Training (Previous Priority #3) - Increase in training opportunities in clinical research methodology and epidemiology: at least 6 NINDS T32 programs, 19 K23s, 2 K24s, 4 K02s, and 1 F31 have been funded. SPOTRIAS and R25s have also increased research training.
  • Other advances in areas are outlined including: Stroke Geography, Stroke in Women, Pediatric Stroke, Disparities in Stroke Care, Novel Risk Factors, Vascular Cognitive Impairment, Stroke Genetic Epidemiology, and Translational Epidemiology

New stroke research opportunities, emerging topics, and unresolved areas since 2007 are summarized across a variety of topic areas: Race-ethnic and Geographic Disparities, Eliminating stroke health disparities, Post-stroke Outcomes including Cognition, Stroke in Women including HRT, Pediatric Stroke, Stroke Trends, Novel Risk Factors including Sociocultural factors and subclinical disease, and Epidemiologic Training

Three priorities for future directions of stroke epidemiology research:

1. Improve the Understanding of Race and Ethnic Stroke Disparities (TOP PRIORITY)

Despite recent successes from NIH-funded studies providing some insights into the causes of these disparities, the understanding of the contributors to these disparities remains insufficient to guide development of interventions. Epidemiologic studies need to focus on specific gaps in our understanding of the risk, determinants and outcomes of stroke in special populations including women, racial and ethnic subgroups, and children, as well as explanation for geographic variability.

2. Evaluate the usefulness of health IT as a tool for epidemiology research

There has been an explosion of electronic health-related data, including large administrative data sets and data collected in electronic medical records, which will become increasingly available for research studies. A major priority in the next 5 years will be to evaluate the validity and effectiveness of these data sources in providing added value to stroke epidemiology and surveillance studies.

3. Translate Knowledge from Epidemiological Studies into Improved Health

Continued support of epidemiologic stroke studies that monitor trends in stroke burden, fill gaps in knowledge, and discover new associations should be a high priority. Critically, we need to accelerate the translation of the results from epidemiology studies into improved health by informing evidence-based practice recommendations and clinical care, translating findings into behavioral interventions, and providing the fundamental preliminary data needed for randomized clinical trials. ?


Epidemiologic research is critical for understanding the public health burden of stroke including the distribution of stroke in the population as well as its determinants. Through epidemiologic research, population subgroups at the greatest risk of stroke are identified allowing for the prioritization of future research and funding. Epidemiologic research is essential for examining trends in stroke over time, including changes in disease burden such as risk and outcomes, as well as prevalence and control of risk factors, data that are pivotal to identify emerging problems, but also to document our successes in the field. Finally, high quality observational epidemiologic studies are critical in identifying new associations and relationships which can be translated into the development of interventions to reduce the public health impact of stroke through clinical and behavioral intervention trials. A great deal of progress has been made in stroke epidemiology during the interval since the last Stroke Progress Review Group report. Staff identified at least 35 grant awards (23 NINDS, 12 non-NINDS) since 2007 that were related to stroke epidemiology. Some of these advances are described below.

Common Data Elements (Previous Priority #1)

To facilitate the combining of epidemiologic data from multiple sources and studies for more powerful analyses, there is a need for common data elements and definitions. As part of its effort to facilitate high quality research and streamline clinical trial data collection, NINDS initiated a Common Data Element (CDE) project. A stroke CDE working group was assembled and data standards that include recommended CDEs as well as case report forms, which could be used by epidemiologic studies, were developed. The National Human Genome Research Institute (NHGRI) has also developed the PhenX Toolkit, which provides standard measures related to complex diseases, phenotypic traits and environmental exposures relevant to epidemiological studies of stroke.

Linkage between Epidemiological Data and Administrative Data (Previous Priority #1)

Several NIH funded grants are linking population-based epidemiologic studies to Medicare data, allowing for the identification of strokes without the logistical difficulty and expense of medically verified stroke events. As examples – the NIH-funded Chicago Health and Aging Project has been linked to Medicare data and is being used to study psychosocial risk factors for stroke in a biracial population. The Health and Retirement Study has been linked to Medicare and is being used to study factors related to functional and cognitive outcome following stroke. The GWTG-Stroke registry has also been linked with Medicare files. This linkage will provide important opportunities to measure longer term post-discharge outcomes such as 30-day and 1-year mortality and readmission rates.

Stroke Trends (Previous Priority #2)

New epidemiologic data on temporal trends in stroke incidence have been published. The Framingham Heart Study (FHS) reported trends in incidence data showing a significant decline in stroke incidence over the 50-year span of the study. Data from the Greater Cincinnati/Northern Kentucky Stroke Study (GCNKSS) also showed a decline in stroke incidence for whites but not for blacks over the time period 1993-94 through 2005. Importantly, GCNKSS has been funded to continue to examine trends over time. In addition, the Brain Attack Surveillance in Corpus Christi (BASIC) Project has been funded to examine trends in stroke incidence in Mexican Americans, an understudied population with increased stroke risk.

Epidemiologic Training (Previous Priority #3)

An increase in training opportunities in clinical research methodology and epidemiology has occurred since 2007. At least 6 NINDS T32 programs exist that include training opportunities for clinical research methods. At least 19 K23 awards and 2 K24s have been awarded to encourage further clinical research training and mentorship. At least 4 K02 grants support investigators in stroke epidemiology. At least 1 F31 award has also supported a pre-doctoral student investigating stroke epidemiology. The CTSAs at multiple institutions have improved the infrastructure for clinical research and have interfaced with K30 and K12 programs to support more training opportunities. The SPOTRIAS program included a training core in several of the awards. The R25 program was launched to allow residents to enter research training in the PGY4 year and accelerate a more rapid transition into research. A number of workshops/training courses have emphasized clinical trial methodology training, but few have been specifically directed towards observational epidemiology: NIMHD Translational Health Disparities Course http://www.nimhd.nih.gov/hdss/NIMHD_thdc.html, Clinical Trial Methods Course in Neurology http://www.neurologytrials.org/CourseInformation.aspx, The Science of Small Clinical Trials http://small-trials.keenminds.org/.

Stroke Geography

The Reasons for Geographic and Racial Disparities in Stroke (REGARDS) Study published its first stroke incidence results. Consistent with previous studies, this study found higher stroke rates for blacks versus whites, especially at younger ages. The study further showed that the geographic variation in stroke mortality (with highest mortality in the southeastern US, the "stroke belt") is due in part to higher incidence rates in blacks versus whites.

Stroke in Women

New epidemiologic data illustrating the public health impact of stroke in women have emerged. Data from the FHS have demonstrated that women have greater lifetime risk of stroke than men. Data from several population-based studies, including the FHS, have reported on the reversal of the association between sex and stroke with advanced age such that women have an increased risk compared with men in the oldest ages. Several papers have been published documenting worse post-stroke outcomes in women than men, particularly functional outcomes and institutionalization.

Hormone Replacement Therapy and Stroke

New observational epidemiologic data from Europe have emerged suggesting that current use of low dose transdermal estrogen may not increase stroke risk among post-menopausal women.

Pediatric Stroke

New epidemiologic data regarding disparities, risk factors and cost in pediatric stroke have been published. A previously documented race disparity in US childhood ischemic stroke mortality (black>white) has largely resolved, likely due to the implementation of transfusion therapy as per the NINDS funded STOP trial. A UK study demonstrated higher pediatric stroke mortality in boys than girls, consistent with previously documented findings from the US. The International Pediatric Stroke Study also demonstrated a male predominance in childhood ischemic stroke incidence. Elevated testosterone levels were associated with childhood ischemic stroke in a case-control study, suggesting a possible mechanism for sex differences. Recent minor acute infections and head or neck trauma have emerged as risk factors for childhood arterial ischemic stroke. The same case-control study suggested that traditional atherosclerotic risk factors may play a role in stroke in adolescents. A study using administrative data suggested that atherosclerotic risk factors are prevalent in adolescents and young adults admitted for stroke. Direct costs of neonatal and childhood stroke (acute and 5-year costs) have been published and exceed reported costs for adult stroke.

Disparities in Stroke Care

The Get With the Guidelines (GWTG)-Stroke program has grown immensely, with data from over 1.6 million stroke admissions from more than 1,400 US hospitals. The GWTG data have provided important insights into disparities in the quality of stroke care. GWTG-Stroke data on race/ethnic differences in care among ischemic stroke patients found that compared to non-Hispanic white (NHW) patients, black patients had modest but consistently poorer care for several performance measures including tPA use, DVT prophylaxis, lipid therapy, antithrombotic therapy at discharge, anticoagulation therapy among atrial fibrillation patients, and smoking cessation. The quality of care was similar between Hispanic and NHW patients. GWTG-Stroke data examining sex differences among ischemic stroke patients have documented that women were less likely than men to receive defect free care (66% vs. 71%).

Novel Risk Factors

As control of traditional risk factors improves, the importance of novel risk factors may increase. Epidemiologic research from a variety studies including REGARDS and NOMAS has identified several novel stroke risk factors as independent predictors of stroke including obstructive sleep apnea, metabolic syndrome, chronic kidney disease, insulin resistance, high-sensitivity C-reactive protein, lipoprotein-associated phospholipase A2, lipoprotein (a), and infectious burden. Subclinical measures such as carotid plaque area, distensibility, white matter hyperintensities, and silent strokes are related to traditional and novel vascular risk factors and are determinants of stroke and cognitive impairment.

Epidemiologic research has begun to explore the role of the environment in stroke. Ecologic studies have suggested links between neighborhood disadvantage and density of fast food restaurants and stroke risk. Neighborhood-level social cohesion was recently found to be protective against stroke mortality in a prospective study after accounting for sociodemographics and risk factors. New evidence regarding the association between air pollution and stroke has also emerged.

Vascular Cognitive Impairment

As reviewed in detail in the Vascular Cognitive Impairment section, epidemiologic research has demonstrated that vascular risk factors are increasingly recognized as contributing to cognitive decline and dementia. Data from FHS, Northern Manhattan Study (NOMAS), and Rotterdam have shown that vascular risk factors are associated with subclinical microvascular disease thereby contributing to cognitive impairment associated with aging. Covert cerebral infarcts and microvascular changes predispose to cognitive decline, depression and gait disturbance and to overt brain infarcts and are more frequent in African Americans. It is now known that cerebrovascular disease frequently co-exists with Alzheimer’s Disease pathology and appears to exacerbate the cognitive impairment in patients with Alzheimer’s dementia. Incident cognitive decline occurred more commonly in the stroke belt compared to other geographic regions, possibly related to subclinical strokes and/or precursor or concomitant risk factors for both stroke and cognitive impairment. Neuroimaging studies have identified a markedly increased rate of microhemorrhages among African-American ICH cases.

Stroke Genetic Epidemiology

As explicated in detail in the Genetics section, several advances in stroke genetics have been made. Specific to genetic epidemiology, familial aggregation studies based on confirmed occurrence of stroke in both parents and offspring have more firmly established family history as an independent risk factor for ischemic stroke. In the FHS, parental ischemic stroke by the age of 65 was associated with a 300% increase in ischemic stroke risk in the offspring, even after adjustment for other vascular risk factors. Increased risk conferred by family history was greatest in the subgroup with other established stroke risk factors. Case-control and prospective studies have identified family history of ruptured intracranial aneurysm and smoking as risk factors for rupture of intracranial aneurysm, with important implications for screening and patient management. Meta-analysis of intracerebral hemorrhage studies has shown that Apo-E genotypes are not only associated with lobar intracerebral hemorrhage but also with deep intracerebral hemorrhage. Large GWA studies of non-stroke conditions such as myocardial infarction and atrial fibrillation have led to replicated ischemic stroke subtype-specific genetic associations with atherosclerotic at chromosome 4q25 and cardioembolic stroke at chromosome 16q22, respectively. Large GWA studies of subarachnoid hemorrhage have identified replicated associations at new candidate genes. Precursor subclinical traits such as carotid IMT, carotid plaque, left ventricular mass, and left atrial size have been found to have high heritability and preliminary regions of linkage identified in the Family Study Of Stroke Risk and Carotid Atherosclerosis.

New projects have been initiated to address the genetics of stroke including the NINDS Stroke Genetics Network Study (SiGN), Risk Assessment of Cerebrovascular Events Study (RACE), Ethnic/Racial Variation in Intracerebral Hemorrhage (ERICH), Genetics of Intracerebral Hemorrhage in Anticoagulation (GOCHA), Familial Intracranial Aneurysm Study II (FIA II), and Predictors of Spontaneous Cerebral AVM Hemorrhage.

Translational Epidemiology

Epidemiologic studies have identified key stroke risk factors, led to intervention trials, and provided evidence for guidelines. There are several recent examples that illustrate how epidemiology has informed intervention trials and clinical practice. A few relevant examples are provided. Epidemiologic study from NOMAS and others has demonstrated that patients taking statins at the time of stroke have decreased mortality and improved outcomes. This coupled with the pre-clinical data on the potential neuroprotective properties of statins have led to NINDS funded phase 1 and 2 trials. Observational data on the variation in outcomes for mild and rapidly improving stroke patients who are not treated with IV rtPA have led to the design of proposed trials. Data from the BASIC Project have identified increased stroke risk in Mexican Americans driven in part by increased prevalence of risk factors. This research has been translated into an NINDS funded culturally-sensitive, church-based, multicomponent, behavioral intervention for Mexican Americans and non-Hispanic whites to reduce important behavioral and biological stroke risk factors. Epidemiologic data demonstrating the independent association of obstructive sleep apnea with stroke and cardiovascular disease endpoints have resulted in a clinical trial of sleep apnea treatment among patients with TIA.

In summary, epidemiologic research has made significant progress since the last SPRG report. Research funded by NINDS and other NIH Institutes has resulted in a clearer understanding of the public health burden of stroke in high risk groups, including African Americans, women, children, and those residing in the southeast US. A significant decline in stroke incidence in whites has been reported but a similar trend in African Americans is not apparent. Novel stroke risk factors have been identified, and knowledge gained has been transformed into interventions that may improve the health of the US population.


Race-ethnic and Geographic Disparities

Although one of the two primary goals of the Healthy People 2010 report was the elimination of health disparities, the magnitude of racial and geographic stroke disparities substantially increased between 2000 and 2010. The understanding of the contributors to these disparities remains insufficient to guide the development of interventions to reduce the disparities. For example, traditional risk factors have been shown to contribute less than half of the black-to-white racial disparities in stroke, where the unexplained additional risk may potentially be attributable to sources only now beginning to be investigated.

Eliminating Stroke Health Disparities

Eliminating stroke health disparities will likely require multi-level interventions that address not only the individual factors that are associated with stroke but also the structural conditions that disproportionately expose race-ethnic minorities to elevated risk. Research regarding the role of the environment in stroke has lagged substantially behind progress in heart disease. Research is needed to identify specific features of the social and physical environment, both positive and negative, that influence stroke risk. Understanding the pathways by which environments influence stroke risk could inform the development of novel multi-level interventions to reduce stroke risk and to reduce race-ethnic disparities.

Post-stroke Outcomes including Cognition

Better epidemiologic data regarding post-stroke outcomes, including functional, cognitive, neurologic and quality of life outcomes, are needed. In particular, there is a lack of information on long-term stroke outcomes in representative populations, and there is very limited information about stroke outcomes in race-ethnic minorities. The lack of outcome data in race-ethnic minorities is particularly important given the considerably younger average age at stroke onset in African and Hispanic Americans. The limited availability of stroke outcome data is partially driven by the difficulty in collecting outcome measures 90-days or greater post-stroke.

The number of stroke survivors will increase dramatically with the aging US population. Post-acute rehabilitation care will play an increasingly important role in our health care system. There are significant gaps in our understanding of the use and effectiveness of post-acute care for stroke largely due to difficulty obtaining longitudinal outcome data on survivors treated across different post-acute care settings. There is a need to develop better measures and determinants of long term outcomes post-stroke. The development of proxy measures collected at or soon after discharge that accurately reflect later recovery would assist in research aimed at assessing disparities in outcomes as well as research aimed at evaluating the impact of hospital-based stroke care or post-acute care.

Comparative effectiveness research that identifies post-acute care settings and patterns associated with better post-stroke outcomes could guide future intervention studies designed to develop more effective, equitable and cost-effective approaches to post-acute care in stroke survivors.

Stroke in Women

The stroke burden in women is higher than for men, including an increased lifetime risk and less favorable outcomes after stroke, particularly functional outcomes. This disparity has received little attention compared to other health disparities. The lack of information on the reasons why women have poorer outcomes post-stroke has prohibited the design of interventions aimed at reducing the high burden of stroke disability in women.

Hormone Replacement Therapy and Stroke

Clinical trial data indicate that use of standard dose estrogen plus progestin, as well as estrogen alone, increases stroke risk in healthy post-menopausal women. It is unknown how these findings have influenced use of hormone therapy over time and the subsequent impact on stroke risk in women. In addition, more epidemiologic research is needed to understand the safest and most effective formulation, dose, and duration of hormone therapy that will treat vasomotor symptoms without increasing risk for stroke.

Pediatric Stroke

The NINDS-funded Vascular effects of Infection in Pediatric Stroke (VIPS) study will explore the role of infection in childhood arterial ischemic stroke, particularly cerebral arteriopathy. Studies of the genetics of childhood arteriopathies including genetic susceptibility to the vascular effects of infection are needed. Trauma has emerged as an important risk factor for stroke in the young. Because boys sustain more trauma than girls, and black children more than white children, this risk factor may partly explain observed sex and race disparities in childhood stroke risk but additional research is needed. Clinical prediction rules for stroke after trauma are not available, yet could be useful in identifying a subgroup of high risk trauma victims that could be targeted for prophylactic anti-thrombotic therapy. Recent data suggest that atherosclerotic risk factors may play a role in adolescent stroke. A better understanding of these associations is important for public health measures; it may be that detection and management of these risk factors should start in childhood. Data on rates of recurrent stroke in children are limited, yet critical for the design of secondary stroke prevention studies. The VIPS study will measure recurrence rates, and predictors of recurrence, in a prospective cohort. There are no published utilities for pediatric stroke; hence, cost-utility analyses for pediatric stroke treatments cannot be performed. The majority of childhood stroke research has focused on ischemic stroke. The most common cause of a childhood hemorrhagic stroke is a structural vascular lesion: arteriovenous malformation, cavernous malformation, or aneurysm. Although there are NIH-funded studies on vascular malformations, risk factors for cerebral aneurysms in children have not been studied.

Stroke Trends

The US population is changing rapidly with respect to demographic composition and risk factor prevalence. In addition, our ability to prevent and treat strokes is changing. For these reasons, it is imperative that there is continued funding for epidemiologic studies to collect data on secular and temporal trends in stroke incidence, recurrence, mortality, and functional and cognitive outcomes. Currently, limited data are available to evaluate long term trends in minority populations and less so for hemorrhagic stroke compared to ischemic stroke. Temporal trends of hemorrhagic stroke in high risk populations are also needed to identify high impact areas for potential intervention.

Novel Risk Factors including Sociocultural factors and subclinical disease

Traditional risk factors explain a large proportion of stroke risk, but further studies of novel stroke risk factors including sociocultural and environmental factors and subclinical disease may help refine risk prediction and lead to new prevention and/or intervention approaches. Studies on the progression of subclinical diseases also can aid in defining surrogate measures for potential risk modifying therapies.

Epidemiologic Training

Metrics are needed to measure the success of previously instituted training programs on launching research careers. There is a need to accelerate training in clinical research and epidemiologic methods during residency and fellowship with more training grants (including R25s). Further expansion of short courses to help clinical investigators become more competitive in clinical and translational study design, including courses for observational epidemiology, are needed. Training programs to encourage pre-doctoral students in epidemiology and biostatistics to engage in stroke research are needed.


1. Improve the Understanding of Race and Ethnic Stroke Disparities

Although rates of stroke and mortality are dropping across many population groups, significant disparities continue to exist. The persistence of these disparities has led to high-level goals in the HHS Action Plan to Reduce Racial and Ethnic Health Disparities to “increase the availability, quality, and use of data to improve the health of minority populations” and underscoring the need to “implement a multifaceted health disparities data collection strategy across HHS.” Likewise, the Institute of Medicine statement on the surveillance of cardiovascular and chronic lung diseases stated that “Untangling the effects of environment, income, education, race, ethnicity, and genetics may lead to more precise targeting of preventive, diagnostic, and therapeutic interventions.”

Despite recent successes from NIH-funded studies providing some insights into the causes of these disparities, the understanding of the contributors to these disparities remains insufficient to guide development of interventions to reduce the disparities. Epidemiologic studies need to focus on specific gaps in our understanding of the risk, determinants and outcomes of stroke in special populations including women, racial and ethnic subgroups, and children, as well as explanation for geographic variability. Such studies need to consider disparities in the development of traditional risk factors, differential response to traditional risk factors, variations that are unexplained by traditional risk factors, residual confounding from incomplete characterization of risk factor levels, novel risk factors not currently considered in risk models, socioeconomic and sociocultural factors, the relationship to subclinical disease, and measurement error in quantifying factors. Economic, cultural, and behavioral factors that influence compliance and control of traditional risk factors also need to be explored for their contributions to stroke disparities. The priority of studies should be directed to the ultimate goal of reducing disparities in stroke incidence and outcomes, as well as to characterizing temporal trends in disparities to monitor change. A better understanding of the causes of disparities should lead to subsequent efforts to design and mount studies to assess the effectiveness of interventions to reduce these disparities and their associated striking public health burden.

2. Evaluate the usefulness of health IT as a tool for epidemiology research

There has been an explosion of electronic health-related data, including large administrative data sets and information collected in electronic medical records, which will become increasingly available for public use and for research studies. These data sources will not be a replacement for important ongoing stroke epidemiology studies. There are important limitations to such large databases including limited data collected, unverified quality of data, issues of reliability and standardization of data collection, and lack of population-based sampling. A major priority in the next five years will be to evaluate the validity and effectiveness of these data sources in providing added value to stroke epidemiology and surveillance studies. Once validity is established, innovative methods should be developed to more efficiently utilize these data sources for epidemiologic studies of temporal trends in stroke, including real-time surveillance of stroke and real-time monitoring of population health. This may require new data harmonization, novel statistical methods, validation samples, and innovative ways to link data across disparate data sources to provide longitudinal information.

3. Translate Knowledge from Epidemiological Studies into Improved Health

Continued support of epidemiologic stroke studies that monitor trends in stroke burden, fill gaps in knowledge as articulated above, and discover new associations should be a high priority. Critically, we need to accelerate the translation of the results from epidemiology studies into improved health by informing evidence-based practice recommendations and clinical care, translating findings into behavioral interventions, and providing the fundamental preliminary data needed for randomized clinical trials. This could be accomplished by RFAs for translational epidemiology studies, i.e., epidemiology research that has clear implications for behavioral or clinical intervention trials.

Epidemiological studies have provided pivotal, but underappreciated, support for many evidence based statements in the prevention of first and recurrent stroke. Moreover, epidemiology research is the initial step that assists with hypothesis generation. Observational epidemiologic studies identify new associations and relationships, including identifying novel risk factors (such as subclinical disease), which can then be used as preliminary data for planning of new treatments and randomized controlled trials. Epidemiologic research also provides essential information required by scientific review for the proper determination of the public health impact (such as measuring disparities), cost effectiveness and feasibility of research studies. Epidemiologic research is necessary for explaining trends in population health and risk factors, and variations in treatment (acute stroke treatment, use of medications or procedures for stroke prevention, and trends in medication side effects--for example, anticoagulant associated intracerebral hemorrhage, strokes in the aftermath of anti-thrombotic medication withdrawal). Finally, epidemiology studies increase our understanding of the impact of certain risk factors in circumstances in which they cannot be ethically randomized or are too rare to study in a clinical trial. In summary, support for translational epidemiologic research can provide the fundamental preliminary data necessary to improve health including the development of interventions studies.



Co-Chairs: Jonathan Rosand, James Meschia, Andrew Singleton

Members: Cenk Ayata, Mark Cookson, Frank Faraci, Murat Gunel, Sek Katherisen, Jennifer Majersik, Michael Nalls, Stephen Rich, Owen Ross, Dan Woo, Bradford Worrall, Cara Carty, Steve Pavlakis

NINDS Liaisons: Katrina Gwinn, Tim LaVaute


There have been advances in several areas over the previous period. There is a general and plausible belief that a key to understanding the genetics of stroke lies in large collaborative genetic analyses. A key limitation in such an endeavor lies in the harmonization of phenotypes and measures across datasets; significant progress has been made in this regard with the development and integration of classification algorithms and large consortia. These consortia and other independent studies have enabled progress in several avenues; first in ischemic stroke, the identification of replicated risk loci at chromosomes 9p21 and 16q22; second, the association of APOE e2 alleles as a risk factor in intracerebral hemorrhage; third, the identification of multiple loci containing risk alleles for intracranial aneurysms; and last, the investigation of mitochondrial DNA variants as a modulator for stroke. In the context of monogenic disease, there has been continued interest and progress in developing cohorts of familial stroke, and given the sibling relative risk of this disease, this is considered an important resource for the future. Likewise considerable progress has been made in testing transfusion therapy in the prevention of stroke in sickle cell disease. There has been an effort to understand the genetics of drug response and this bears direct relevance to stroke with the identification of genetic modifiers of response/tolerance to warfarin, clopidogrel and statins. From an mechanistic perspective, genetics has been used to create animal models of disease with mice carrying targeted mutations in NOTCH3, PPAR? and collagen type IV each recapitulating some of the key features of disease.

There remain several critical challenges in the field of stroke genetics. From a gene discovery standpoint, efforts to identify further risk loci for ischemic stroke, pediatric stroke, and intracerebral hemorrhage each require very large numbers of samples from well-characterized subjects, and the support to pursue modern genetic approaches to gene discovery (high density genotyping, targeted sequencing, whole exome sequencing, whole genome sequencing etc). It will also be important to combine many of these genetic studies with longitudinal collection and biobanking of RNA, serum, plasma, and tissue. These studies will be expensive and time consuming; however they offer the best opportunity to delineate the etiology of these complex diseases and to define biomarkers for risk, prognosis and treatment efficacy. Still in this vein, it is important that genetics is incorporated early into clinical trials; not only to define groups with likely differential outcomes or responses, but also to define risk factors. One area that requires increasing emphasis is the investigation of stroke in underrepresented populations, while difficult studies to initiate and maintain, the varied burden of stroke across diverse populations likely means that it is important to establish the extent and genetic basis of population heterogeneity of this disease.

The genetics working group defined a set of 3 research priorities:

1) To perform large multi-center collaborations to recruit cases and controls with excellent and consistent characterization.

2) Integrate genetic studies into current and future supported clinical trials.

3) Using in vivo and in vitro modeling to elucidate the mechanisms by which genetic factors modulate the risk and outcome of cerebrovascular disease and stroke. Clearly this is somewhat limited by the identification of identifiable genetic risk factors for disease.

While the committee noted that substantial hurdles exist to understanding the genetic basis of stroke, we were enthusiastic about the future of this field. Much of the groundwork has been laid, and it is apparent from other complex diseases that many of the proposed strategies work well; thus given appropriate support, these research priorities represent tractable problems over the coming period.


Animal models

Small vessel disease is a result of a complex mix of genetic and vascular risk factors, including age and hypertension. Specific treatments for small vessel disease are lacking. Genetics studies have contributed importantly to understanding the pathophysiology of small vessel disease. Genetic models including mice expressing CADASIL (Cerebral Autosomal Dominant Arteriopathy with Subcortical Infarcts and Leukoencephalopathy) mutations involving the Notch3 gene reproduce key features of the human disease, including reduced resting cerebral blood flow, impaired vasodilator responses, and progressive white matter damage. This work suggests that early microvascular abnormalities cause subsequent perfusion deficits and white matter injury. Other genetic models like mice with mutations in collagen type IV, and PPAR? mimic components of human disease. For example, mice expressing human dominant negative mutations in PPAR? exhibit oxidative stress, microvascular hypertrophy, inward remodeling and impaired vasodilation. Such models help define pathways that underlie human small vessel disease associated with hypertension.

Harmonization of the ischemic stroke phenotype

A major accomplishment has been the international validation of the web-based algorithmic Causative Classification of Stroke (CCS) system, which has now been integrated into the NINDS Stroke Genetics Network (SiGN) study. Several large collaborative groups have been formed, each of which has undertaken large-scale genetic studies of stroke. The International Stroke Genetics Consortium (ISGC) is open to anyone with the resources to contribute to advancing our understanding of the role of genetics in stroke (www.strokegenetics.org). With > 30 centers from around the globe, the ISGC has been responsible for several major discoveries, including identification of risk loci for ischemic stroke related to large-vessel atheromatous disease and of risk loci for lobar intracerebral hemorrhage (ICH). The ISGC has enabled the establishment of several consortia devoted to particular forms of stroke. Most notable of these are the NINDS-funded Stroke Genetics Network (SiGN) study and the NINDS-funded ICH genetics collaborative.

Ischemic stroke: Discovery of novel risk loci

After many groups had found an association between single nucleotide polymorphisms (SNPs) in the chromosome 9p21 region and coronary artery disease, the ISGC found six SNPs associated with stroke, independent of coronary disease risk, with one SNP having a pooled odds ratio of 1.21. For atherosclerotic stroke specifically, the population-attributable risk was estimated at 20.1%. Similarly, in another pooled analysis, SNP rs7193343-T on chromosome 16q22 was found to be associated with atrial fibrillation and cardioembolic stroke. The Wellcome Trust Case Control Consortium-2 has assembled ~4,000 patients with ischemic stroke through the ISGC in the discovery phase; however this sample size remains of only modest statistical power.

Intracerebral Hemorrhage: Discovery of novel risk loci

Three major NIH-funded efforts in this area have resulted in several genetic discoveries. These results are consistent with earlier epidemiologic and pathologic observations supporting the hypothesis that differences in anatomical location of ICH reflect differences in pathobiology. The apolipoprotein E (APOE) e2 and e4 alleles are risk factors for ICH in lobar brain regions; APOE e2 is a risk factor for more severe bleeding in lobar ICH; and CR-1 is a risk factor for lobar ICH.

Intracranial Aneurysms

Completion of two large genome wide association (GWA) studies and a follow-up study discovered 8 independent chromosomal regions that increase risk of intracranial aneurysms. These studies identified risk variants common in the population but conferring modest risk. The 5 published risk loci explain ~10% of genetic risk. Extrapolating from other common diseases in which common variants typically explain 30-40% of the risk, it is hypothesized that there are as yet unidentified loci containing common risk variants, potentially with lower odds ratios. Although the disease-causing variants in identified intervals need to be determined, some of these intervals contain only one gene. Identification of these genes generates unique biological insights and ideas for potentially novel biologically-based treatments.

Cerebrovascular Malformations

The major advance in cerebrovascular malformations has been the generation of animal models of cerebral cavernous malformations (CCMs). These models are aimed at examining the biology of genes implicated in familial forms of CCMs and biochemical, structural and proteomic studies of their encoded proteins. Studies in model organisms demonstrated that the genes involved in familial CCM syndromes are essential for vascular development. The three CCM proteins appear to interact; each also has distinct interacting partners. Studies are ongoing in understanding the genetic basis of other malformations such as arteriovenous malformations (AVMs) and moyamoya variants.

Mitochondrial genetics

The identification of the disorder Mitochondrial myopathy, Encephalopathy, Lactic Acidosis with Stroke-like episodes (MELAS) associated with a number of point mutations in the mtDNA confirmed the role of genetic variation in stroke phenotypes. About 80% of patients carry an A-to-G transition at position 3243 in the tRNALeu (UUR) gene. This variant is associated with OXPHOS Complex I deficiency. Two studies suggest that mtDNA variation can alter the individual susceptibility to sporadic ischemic stroke. It is suggested that variation forming one of the larger mtDNA haplogroups (K) protected against the development of an ischemic stroke. Subsequent work has suggested that rather than a specific haplogroup, an overall score of common variation in the mtDNA genome increased the risk of stroke.

Single gene/familial disorders

Stroke is the sole or major clinical phenotypic expression of several monogenetic disorders. Pivotal trials funded by the NINDS established the striking efficacy of transfusion therapy in preventing stroke in high-risk patients with sickle cell disease. The ongoing Silent Cerebral Infarct Transfusion (SIT) trial will test in children with sickle cell disease and MRI-detected silent infarcts the efficacy of transfusion therapy in preventing silent and overt incident infarcts. Most single-gene disorders that predispose to ischemic or hemorrhagic stroke remain without an effective treatment.

CADASIL is the leading cause of inherited vascular dementia. New mouse models suggest that Notch3 signaling relates to small vessel ischemic disease in CADASIL. A multicenter randomized trial showed that donepezil can improve measures of cognitive performance in CADASIL. This trial provides insight into the effectiveness of cholinesterase inhibitors for sporadic vascular dementia, because CADASIL-associated dementia presents at an age where coexistent Alzheimer pathology is expected to be minimal.

Epidemiological studies support a strong genetic component in risk of stroke, a systematic review of the familial studies generated quantitative estimates of risks suggesting that first degree relatives of stroke patients have increased risk of stroke (odds ratio of 1.76). With the advent of next-generation sequencing, identification of new genes is possible without large expanded pedigrees, which can take many years to assemble. A concerted effort is now required to establish independent cohorts of patients with familial forms of cerebrovascular disease.


Response to drug therapy varies among individuals, and genetic polymorphisms contribute to this variability. More than 10% of the 69 drugs currently with pharmacogenomic tests included in their labeling are commonly used to prevent stroke directly or indirectly through treatment of vascular risk factors. Two mainstays of secondary stroke prevention, warfarin and clopidogrel, have had identified pharmacogenomic modifiers of treatment response or dosing. Pharmacogenomic variation may account for one of the most common treatment-limiting adverse reactions to HMG-coA reductase inhibitors (statins). The Clinical Pharmacogenomics Implementation Consortium of the Pharmacogenomics Research Network has begun to develop and evaluate user-friendly, evidence-based algorithms to facilitate clinical adoption of pharmacogenomics.

Gene variants influence response to warfarin therapy.

The relabeling of warfarin in 2007 represents one of the earliest examples of FDA labeling that included pharmacogenomic data, and the DNA-based drug sensitivity test for both the CYP2C9 (cytochrome P450 2C9) and VKORC1 (vitamin K epoxide reductase complex subunit 1) variants was the first ever approved. More recent clinical trial data demonstrate reductions in hospitalization, bleeding and thromboembolic events in those using genotypic information at the time of warfarin initiation compared to those using standard of care. The significance of warfarin pharmacogenomics has been muted somewhat with the arrival of several direct thrombin inhibitors and anti-Xa inhibitors.

Gene variants influence response to clopidogrel.

The evidence supporting adding pharmacogenomics data regarding the CYP2C19 (cytochrome P450 2C19) polymorphism to the labeling for clopidogrel came from a range of data and sources including post hoc analyses of clinical trials, cohort studies from academic centers, and large-scale epidemiological data using both biomarker and clinical endpoints. A second pharmacogenomic variant due to polymorphisms in ABCB1 (the ATP-Binding Cassette, Subfamily B, Member 1) is also associated with decreased antiplatelet responses and increased ischemic events. Collectively, almost 50% of the clinical population harbors one or more genotypes associated with increased risk vascular events on standard dosing schedules for clopidogrel. An algorithm for using pharmacogenomic information to guide therapy in treatment of interventional cardiology patients has recently been published. To date, no studies have focused primarily on the effect on ischemic stroke. CYP2C19 variants appear not to influence other thienopyridines. However, neither of the newer agents has been specifically tested in ischemic stroke and prasugrel carries a black box warning against use in patients with a history of stroke. Ticlopidine has demonstrated efficacy in preventing stroke but carries risk of thrombotic thrombocytopenic purpura and other blood dyscrasias.

Genetic variation influences tolerance to statin medications.

Although HMG-coA reductase inhibitors (statins) reduce ischemic stroke risk, many individuals cannot tolerate them because of myalgias and myopathy. Individuals harboring variants in SLCO1B1 (the solute carrier organic anion transporter family member 1B1) gene that encodes the hepatic drug transporter have a 4.5-fold increase in the odds of myopathy per copy of the SLCO1B1*5 allele.

II. Key Challenges

Animal models

Designing therapies to protect against SVD remains a major challenge due to our incomplete understanding of genetic factors, signaling pathways, and cell types involved in this form of vascular disease. While studying mice expressing mutations in CADASIL or other relevant genes provides opportunities to advance our understanding, single gene mutations do not fully recapitulate the etiology of SVD in the broader population. Although genetic influences are generally highly cell specific, current models have often failed to take full advantage of available technologies and genetic approaches to understand cell-specific aspects of SVD. While detailed mechanistic studies can be performed in mice expressing human gene mutations, this approach has been used to a limited degree. In most cases, effects of single gene alterations or single risk factors have been studied. Despite the clinical impact, the contribution of aging and risk factors like hypertension to SVD remains understudied. Current and new genetic models provide tools to test interventions that may alter the course of SVD.

Ischemic stroke

Common genetic variants that influence ischemic stroke appear to be subtype-specific. This highlights the need for reliable subtyping systems, such as the CCS that is being employed by the NINDS-SiGN study, but it also raises a substantial challenge for further progress in ischemic stroke genetics: the current shortage of adequately phenotyped cases for genetic studies. Compared to diseases of comparable public health import, the number of sufficiently phenotyped cases of stroke with genome-wide genotyping available or planned is an order of magnitude smaller. The sample size gap only increases when one considers that stroke subtypes appear to be different diseases and therefore must be considered each as separate samples.

Intracerebral Hemorrhage

Like ischemic stroke, ICH appears to be more than one disease, and available sample sizes for genetic study are inadequate. Identification of all of the genetic contribution to disease risk is essential in order to identify all the novel biological pathways that could become drug targets. Additional well characterized samples are needed for replication and studies of molecular mechanisms. These new collections should have more biobanking with not only DNA but also RNA, serum, plasma and, when available, tissue samples.

GWAS studies probe associations of common variants with stroke; future studies are needed to address the rare variant hypothesis, including exome sequencing and whole genome sequencing studies of extremely well-phenotyped ischemic stroke subtypes, intracerebral hemorrhage and subarachnoid hemorrhage.

Pediatric Stroke and Genetics

There are many obstacles to understanding childhood stroke genetics primarily related to small numbers of pediatric stroke cases. Initially it seemed that genes were an important cause of stroke because of the obvious stroke association with monogenetic disorders such as sickle cell disease. However it soon became apparent that not all patients with genetic diseases, which predispose to stroke actually develop cerebral vasculopathy and most children with stroke do not have obvious monogenetic disorders. In other words, the genetic association is more complicated than first assumed.

Today there are challenges and opportunities. First, there are obvious synergies with the adult stroke studies. There is ongoing work at standardizing stoke phenotype which will make inclusion of pediatric stroke in large studies possible. For example SVD is now recognized in pediatric stoke and ischemic stroke is better characterized in a reproducible and reliable fashion allowing for pediatric stroke to be included in the NINDS Genetic Stroke network. This should be an initial priority.

A current focus in pediatric stroke research involves understanding patients with cerebral vasculopathy since such patients seem at highest risk for secondary stroke recurrence. Developing standardized protocols for looking at genetics in this population with vasculopathies might be a fruitful approach. These disorders include moyamoya and other vasculopathies. The impact of the NF-1 mutation on vasculopathy including a recently described genetic–inflammatory interplay may, for example, be a model for some of the vasculopathy seen in childhood. Here, as in many genetic conditions, a clinical diagnosis is often difficult. As such a standard protocol for looking at the genetics of disorders that result in vasculopathy would be helpful to apply to all children with vasculopathy and stroke even if they do not have an obvious genetic disorder. A standardized genetic approach in pediatric research trials is needed.


  • Incorporating genetics at the design stage of clinical trials.
  • Defining the type and levels of evidence required to change clinical practice by using genetic variant information to alter drug or dose choice.
  • Developing ways to obtain a patient’s genotypic information at the point of care.
  • Integrating genetic information into practical treatment algorithms into the electronic medical record and practice guidelines.

Stroke in underrepresented populations

The burden of stroke varies greatly across racial and ethnic groups in the US. Both stroke incidence and mortality in African Americans are nearly twice that of non-Hispanic whites. In Hispanics, stroke incidence and risk of recurrent stroke are also higher than in non-Hispanic whites. In addition, stroke tends to occur at younger ages in African Americans and Hispanics, resulting in greater disability and expense. Diverse populations remain underrepresented in stroke population genetics research. In 2008, the NIH Center for Research on Genomics and Global Health was established to explore the relationship between human genetic variation and population differences in disease with the goal of understanding health inequalities. Similarly, the SNP Health Association Resource (SHARe) was established to pursue genome-wide data collection and analyses in several multi-ethnic NHLBI cohort studies to identify genes underlying cardiovascular and lung disease and other disorders.

Mitochondrial genetics

There has been a scarcity of studies focusing on stroke-related research; analysis of mitochondrial genetics should be built in to studies powered to understand the complex genetic nature of stroke.


Ischemic stroke

Despite the active enrollment of ischemic stroke cases in clinical trials and observational studies without explicit genetic aims, there is no efficient mechanism for ensuring that enrolled cases participate in future genetics studies.

Intracerebral Hemorrhage

NIH-funded (ATACH-2) and international (INTERACT-2) clinical trials for acute ICH will enroll thousands of ICH patients, whose genetic material could be useful for further study. Studies of Alzheimer Disease (AD) have been leveraged to identify risk factors for cerebral amyloid angiopathy (CAA). Similar leveraging of acute trials would be fruitful.

Intracranial Aneurysm and cerebrovascular malformations

The studies aimed at discovering genetic risk for intracranial aneurysms must account for the complex genetic architecture, which includes contribution of common, rare and possibly Mendelian alleles to aneurysm risk. Discovery of Mendelian disease genes will be very useful in understanding the disease biology. Disease-causing variants need to be identified within the association intervals discovered by genome-wide studies. Rare variants with higher effect sizes appear to contribute significantly to intracranial aneurysm risk. Next generation sequencing technologies also offer the greatest opportunity for understanding the genetic basis of the sporadic forms of the cerebrovascular malformations.

Identification of cerebrovascular malformation genes, especially in genes causing CCMs, has already formed the basis of other studies aimed at dissecting the biology of these malformations. Studies that focus on downstream signaling pathways of each of the genes will lead to new insights into mechanisms of disease pathogenesis.

Opportunities in Translational Applications

The discovery of the genetic risk variants will provide insights into pathophysiology of IA formation and rupture and, when considered in combination with environmental and other risk factors such as smoking, hypertension and cerebral architecture, will lead to genotype-phenotype studies aimed at identifying at risk individuals in a cost effective way prior to catastrophic problems such as rupture. Although this clinical application relies on obtaining a comprehensive catalog of most (rare + common) variants, the ones that have already been identified significantly contributed to our thinking regarding the mechanisms underlying aneurysm formation and rupture, forming the basis for future biology based treatments. Further understanding of the genetic basis of cerebrovascular malformations, especially the sporadic ones, will lead to similar genotype-phenotype studies, assisting the clinical treatment decision making process in the future.

Stroke in underrepresented populations

Recent studies suggest that it is not only the higher prevalence of risk factors, but also the overall lower socioeconomic status and health care system challenges that contribute e to stroke health disparities seen among US racial/ethnic groups. It is likely that new methods and study designs will be necessary to disentangle these complex relationships that create health disparities for stroke and diseases that are important risk factors for stroke, like hypertension and diabetes mellitus. A key area of research will be the investigation of the biological (age, sex and genetics) and behavioral (environmental) bases of stroke and stroke risk factor differences in multi-ethnic populations, as well as their potential interactions contributing to stroke outcomes. As the population genetics field shifts its focus from investigating common variants to rare variants, inclusion of minority populations will be critical for exploring disease biology.

Genetic analysis

Large epidemiological studies focusing on meta-analytic and Mendelian randomization analyses of both clinical risk factors combined with genetic factors have shown possible therapeutic targets for stroke prevention. Recent studies have incorporated genetic data in interventions to determine optimal therapeutic patterns that enhance risk reduction with improved efficacy. These studies represent one application of genomics to personalized medicine.

Mitochondrial genetics

Next-generation technologies provide a potential avenue of investigation through whole mtDNA genome sequencing. In addition, identification and quantification of levels of mtDNA deletions and heteroplasmy within disease-related tissue is yet to be performed. A concerted collaborative effort to generate large datasets of mtDNA variation is required, including the parameters mentioned above.

Functional understanding of etiology

Top priorities for the future include studies with greater focus on defining the cell-specific impact of genetic factors in SVD. The need for development and use of models that better mimic the human condition, including better incorporation of risk factors and issues related to gender and aging, remain. Models that incorporate newer advances in human genetics will be critical in unraveling the pathogenesis and impact of SVD. One area that remains unresolved is the etiological relevance of the MCAO model to the various types of stroke seen in humans. Although this is a robust model and has been used to develop some therapies that have moved forward into clinical use, how well it can be leveraged to address genetic risk factors is unclear.

GWAS and other studies provide identified loci that may alter risk for stroke or of differences in outcome between people. However, identifying a locus does not a priori identify a mechanism. One way to address the proximal effect of gene variation is to look at the effects of nominated SNPs on gene expression. This has been done in a series of eQTL (expression quantitative trait loci) mapping experiments, including in brain. Some of these datasets are publically available and can be mined to nominate candidates for functional effects of risk, which may be helpful in designing animal experiments and in identifying new targets for therapy. However, in the context of stroke, the ongoing pathological process in the brain may modulate genetically controlled gene expression. One missing element is to look at genotype: gene expression relationships in the context of subtypes of stroke and of stroke outcomes.


Research on prevention of first and recurrent stroke

Investigation of group and individual responses to therapies has the potential to both improve mechanistic understanding of drug response and develop individualized treatment for stroke prevention. Regardless of a trial’s outcome, pharmacogenomic and biomarker data that provide insight into mechanism of effect, differential treatment response, and identification of susceptible and resistant subpopulations have the potential to amortize the NINDS investment in testing of specific agents. Incorporating genetic and biomarker analyses into every clinical trial would take advantage of the rich phenotypic data captured in the trial and would be an efficient use of scarce resources by compounding the return on investment. Furthermore, critical determinants of drug action are often poorly defined when a drug reaches market. Thus, having incorporated sample collection in the clinical trial populations provides an efficient and scientifically sound opportunity to test new theories, assess the impact of genetic factors (using Mendelian randomization), and compare in vitro surrogates to the gold standard clinical outcome.

Research on improving outcomes from stroke

Little is currently known about pharmacogenomics of rehabilitative therapies. Similar to investigation of acute treatments, studies of agents that ameliorate the devastating consequences of stroke or enhance recovery should also include collection of genetic and biomarker data. 

Research on reducing the burden of stroke among underrepresented populations

The differential responses across racial and ethnic groups to drugs used to prevent and treat stroke warrant further investigation. The recent NIH initiatives partnering with African investigators may provide unique opportunities to investigate genetic and environmental determinants of stroke risk in ancestral and Diaspora populations and allow meaningful intra- and inter-racial investigation of pharmacogenomics.


1) Large, multi-center collaborations to recruit new cases and appropriate controls are needed to build on current work that is utilizing nearly every available sample. New recruitment can effectively address the power/sample size demands of sequencing, the heterogeneity of stroke as a phenotype, and the importance of uniform phenotyping and radiographic and risk factor ascertainment.

2) Develop a mechanism whereby investigators can efficiently engraft parallel pharmacogenomic studies onto NINDS-supported clinical trials meeting appropriate criteria. Pharmocogenomics has the potential to attach the heterogeneity problem that plagues many trials, which dilutes the effects of the experimental treatment and reduces chances of observing a significant difference among treatment groups.

3) Elucidate the mechanisms by which genetic factors modulate the risk and outcome of cerebrovascular disease and stroke. Achieving this goal will critically depend on the development and use of animal models that incorporate newly discovered human genetic factors, while modeling the impact of risk factors, gender and aging, to better mimic the human condition.



Co-Chairs: Mark Alberts, Larry Goldstein, Amytis Towfighi

Members: Eric Cheng, Edward Jauch, Dana Leifer, David Bruce Matchar, Marilyn Rymer, Linda Williams


The effective implementation of stroke-related advances in medical science remains a top priority as it has the potential to rapidly improve the delivery of care and outcome of patients with cerebrovascular disease. These advances include development of criteria for Stroke Centers, effective EMS triage and diversion, the use of various tele-technologies to extend clinical expertise to underserved regions, and similar system-based changes.

There were few grants specifically addressing the health services implementation (HSI) priorities from the last Stroke PRG. Of those grants identified in the HSI group, many were either training grants or were sub-studies of larger SPOTRIAS projects.

New programs that emphasize ‘pay for performance’ may accelerate the adoption and implementation of best practices for healthcare systems, hospitals, physician groups, individual practitioners, and other healthcare professionals.

Outcome measures and reimbursement should be appropriately adjusted for stroke severity and medical complexity to accurately reflect quality of care.

Based on these accomplishments and unmet needs, we suggest the following three priorities for future research support and focus:

1) Evaluate strategies to improve the identification, treatment and control of vascular risk factors across the spectrum of care. These strategies should include environmental changes and development of effective interventions for improving medication compliance and lifestyle changes.

2) Identify and address barriers to the widespread use of hypothermia to treat post-cardiac arrest patients.

3) Identify strategies to improve patient access to comprehensive rehabilitation services that have been shown to enhance outcomes and quality of life post-stroke.


HSI advances over the past 4-5 years have centered on several areas that primarily focus on institutional and system changes, as opposed to individual effectors. Some of the major changes have not been due to formal ‘clinical trials’, but rather the result of an evolution in the organization of care.

Research Advances

Studies in the U.S. and Europe show the benefits of Primary and Comprehensive Stroke Centers.1, 2

Studies using GWTG database prove the effectiveness of such programs for improving the process of care.3

Several studies using telemedicine and similar technologies show these systems are safe and effective for extending medical expertise, reducing delays in care, and expanding the use of IV tPA for acute ischemic stroke.4, 5

A growing body of evidence suggests that a ‘pay for performance’ model may promote improved care and outcomes.

Results from recently completed studies (i.e., ECASS III supporting the efficacy and safety of IV tPA within a 4.5 hour treatment time window) have been incorporated into treatment guidelines and clinical care.6

There is an overall trend for increased use of IV tPA in appropriate patients.7

Studies such as Fast Mag show that administration of a putative neuroprotective agent in an ambulance setting by EMS personnel is feasible.

Data support the efficacy and safety of hypothermia as a treatment for selected patients after cardiac arrest.8

Studies reveal several disparities in stroke care.

Studies show the efficacy of some new and innovative rehabilitation methods and approaches9, 10

Advances in Stroke Care Systems

Plans for uniform criteria for the independent certification of Comprehensive Stroke Centers.

Revisions to criteria for Primary Stroke Centers.11

EMS systems in several states and regions preferentially routing or triaging patients with known or suspected acute strokes to the nearest most appropriate stroke hospitals (i.e. a certified stroke center).12, 13

Public reporting of outcomes in stroke care, with an emphasis on hospital-based outcomes.

An emphasis by CDC, AHA, and others on the identification and treatment of basic stroke risk factors for primary prevention.14


This section will begin with a discussion of emerging topics and unresolved issues as a foundation to define potential new stroke research opportunities related to HSI.

One broad area is the concept of Stroke Systems of Care. As the CSC certification program emerges and is implemented, the Brain Attack Coalition (BAC) will publish guidelines for the establishment of Acute Stroke Ready Hospitals. These two new tiers of care and service will complete the range of stroke care facilities that most patients will access during the acute care process. Going forward, it will be important to assess and track how patients enter this system of care, are transferred from one level of care to another, and how this changing care paradigm affects both short term and long term outcomes (i.e. rehabilitation, nursing home, return to work) and healthcare costs. This will require compiling data from EMS, hospitals, and rehabilitation facilities.

The triage of stroke patients by EMS remains an important issue. While many parts of the country now have laws or regulations that mandate preferential routing of patients with a suspected acute stroke to the nearest stroke center facility, this is not universally accomplished in a timely manner, if at all. An emerging concept is to determine the feasibility of EMS triage based on stroke severity, whereby patients with more severe strokes (or suspected hemorrhage strokes) are routed to a CSC. There are a number of scales and models that could be tested in this regard to determine the accuracy and effectiveness of such a screening and triage process. One exciting opportunity is to use ambulance-based telemedicine technologies to enhance triage in the field. Pilot programs of ambulance-based telemedicine show promise.

Within the context of care at a Stroke Center, it is very important that outcomes be assessed using risk adjusted and severity measures. This is key because PSCs and CSCs will be caring for the sickest of patients, and as a result, their outcomes will be skewed due to the relatively high acuity and greater number of co-morbid conditions in their patient populations. Various algorithms have been developed to address this issue, but many lack the most important variable, namely initial stroke severity. Future studies should include such data for all types of strokes in an attempt to more accurately assess comparative outcomes.

The Stroke Center concept is not without flaws and limitations. Processes of care at some Stroke Centers are suboptimal due to a variety of factors. The geographic distribution of PSCs and CSCs is often not based on optimizing patient access, due mostly to demographic and economic factors. Staffing models and the availability of services on a 24/7 basis is problematic in some settings. The costs associated with stroke center certification will pose a challenge to some facilities, especially as reimbursement levels are reduced in many areas. All of these issues should be addressed to ensure that the level of care is equitably distributed throughout the country. Future research should determine if such care paradigms are cost-effective.

Despite these potential limitations, we believe that stroke center based care has shown and will continue to show improved outcomes. It should also be emphasized that in terms of identifying and recruiting patients for clinical research studies, the stroke centers provide a valuable resource of patients, personnel, and expertise that might further enhance clinical research. With NIH and NINDS having a focus on the timely completion of clinical trials, the use of Stroke Centers should be a high priority when appropriate.

The FDA is poised to approve an expanded label indication for IV tPA with a 4.5 hour treatment time window. With the expanded window (now included in national guidelines and used in many but not all stroke care facilities), it will be important to determine how this affects the total number of treated patients, changes in the door to needle time epoch, and overall outcomes. Many clinicians are concerned that an expanded time window will lead to a ‘treatment time drift’, meaning that the door to needle time, or the onset to treatment time may expand. Preventing this will require not just careful research but ongoing education of patients, EMS personnel, and hospital-based health care professions.

An area of future research involves improving the identification and control of vascular risk factors. Although programs such as “Get with the Guidelines” have improved the initiation of medications for secondary stroke prevention in the inpatient setting, and the Primary Stroke Center certification program assesses the use of these interventions, little research has been completed regarding improving adherence to these medications and initiating lifestyle changes after hospital discharge.15 It will be important to investigate patient- and provider-level barriers to vascular risk factor control, in both primary and secondary prevention settings. On the patient level, these barriers include socioeconomic status, poor access to outpatient care, poor compliance with medications, and the inherent difficulty in making and adhering to lifestyle changes (particularly in the setting of neurologic disability and economic hardships). On the provider level, barriers include lack of outpatient performance measures, inadequate interventions aimed at improving compliance and instituting lifestyle change, and physicians’ lack of knowledge regarding the importance of intensive vascular risk factor control. Future research should have an emphasis on reducing socioeconomic and sex disparities in stroke care.

Organized, multidisciplinary post-stroke rehabilitation improves patient outcomes and quality of life. Yet many stroke patients do not receive any significant rehabilitation due to a number of factors such as lack of available rehabilitation resources, financial limitations, and suboptimal assessments for comprehensive rehabilitation. Future studies and efforts should address disparities in access to rehabilitation and develop strategies and solutions to correct or overcome these barriers. This area of research could also reduce societal costs related to long-term care of stroke patients and perhaps improve the return of such patients to home and/or a work environment.

Patient and public education about the need for accurate recognition of stroke symptoms and rapidly seeking medical care remains a vexing issue despite over 15 years of various efforts to address these problems. The time delay in presentation remains a major factor that severely limits the implementation of acute therapies and ultimately leads to poorer outcomes. Future efforts should focus on unique messaging content and delivery strategies that have the ability to alter perceptions, change actions, and accelerate the presentation of such patients for medical care. (Perhaps another Stroke PRG subgroup has also identified this issue as a priority).

Although some of the focus of the above topics is hospital based, there is a clear recognition that much of this research and care is based (or should be based) in physician offices, clinics, and other care venues. Thus, a significant portion of these research efforts and patient care initiatives should be implemented in a way that directly reaches all levels of health care professionals in their local care settings, not just large medical centers. This will require innovative research approaches, perhaps largely Internet based, with efficient and user-friendly data collection tools, as well as streamlined protocols and patient consent procedures (i.e. centralized IRBs).

Hypothermia after cardiac arrest is a proven treatment that greatly improves outcomes and reduces disabilities in selected patients. Yet, several recent reports show that this therapy is under-utilized at many hospitals, some of which are major medical centers. Because there is a lack of alternative therapies in this setting, the failure to use hypothermia is a vexing question that has not been addressed to any great degree in terms of research studies and remedial efforts. This area is clearly in need of further detailed research with the hope of changing care paradigms.


As noted above, there is no shortage of important clinical topics related to HSI, many of which could be further enhanced by new research data and related findings. Below are three that our group feels hold the most promise for changing care and improving outcomes in the next 4-5 year time frame:

1) Research programs that determine the causes and remedies for suboptimal risk factor control, including limited access to outpatient care, lack of standardized outpatient performance measures, poor medication adherence and difficulty in instituting lasting lifestyle changes. Improving risk factor control could have far-reaching effects, reducing initial and recurrent strokes rates, improving outcomes, preventing readmissions, and reducing health care costs throughout the spectrum of care.

2) Aggressive efforts to identify barriers and limitations to using hypothermia to improve outcomes after cardiac arrest. Considering that this is a class I level I recommendation, it should be widely implemented. This area has the potential to improve outcomes and reduce the utilization of long-term and expensive health care and related resources.

3) Increasing access to and use of rehabilitation services for post-stroke patients. Although there have been significant advances in rehabilitation services over the past 10 years, they are not uniformly applied nor are they easily accessible to all patients in all parts of the country. Considering the impact that comprehensive post-stroke rehabilitation can have on improving outcomes and improving the quality of life, this is a high impact area of focus.


1. Xian Y, Holloway RG, Chan PS, Noyes K, Shah MN, Ting HH, Chappel AR, Peterson ED, Friedman B. Association between stroke center hospitalization for acute ischemic stroke and mortality. JAMA. 2011;305:373-380

2. Meretoja A, Roine RO, Kaste M, Linna M, Roine S, Juntunen M, Erila T, Hillbom M, Marttila R, Rissanen A, Sivenius J, Hakkinen U. Effectiveness of primary and comprehensive stroke centers: Perfect stroke: A nationwide observational study from finland. Stroke. 2010;41:1102-1107

3. Cumbler E, Murphy P, Jones WJ, Wald HL, Kutner JS, Smith DB. Quality of care for in-hospital stroke: Analysis of a statewide registry. Stroke. 2011;42:207-210

4. Schwamm LH, Holloway RG, Amarenco P, Audebert HJ, Bakas T, Chumbler NR, Handschu R, Jauch EC, Knight WAt, Levine SR, Mayberg M, Meyer BC, Meyers PM, Skalabrin E, Wechsler LR. A review of the evidence for the use of telemedicine within stroke systems of care. Stroke. 2009;40:2616-2634

5. Audebert HJ, Kukla C, Vatankhah B, Gotzler B, Schenkel J, Hofer S, Furst A, Haberl RL. Comparison of tissue plasminogen activator administration management between telestroke network hospitals and academic stroke centers: The telemedical pilot project for integrative stroke care in bavaria/germany. Stroke. 2006;37:1822-1827

6. Hacke W, Kaste M, Bluhmki E, Brozman M, Davalos A, Guidetti D, Larrue V, Lees KR, Medeghri Z, Machnig T, Schneider D, von Kummer R, Wahlgren N, Toni D. Thrombolysis with alteplase 3 to 4.5 hours after acute ischemic stroke. N Engl J Med. 2008;359:1317-1329

7. Adeoye O, Hornung R, Khatri P, Kleindorfer D. Recombinant tissue-type plasminogen activator use for ischemic stroke in the united states: A doubling of treatment rates over the course of 5 years. Stroke. 2011;42:1952-1955

8. Bernard SA, Gray TW, Buist MD, Jones BM, Silvester W, Gutteridge G, Smith K. Treatment of comatose survivors of out-of-hospital cardiac arrest with induced hypothermia. N Engl J Med. 2002;346:557-563

9. Wolf SL, Winstein CJ, Miller JP, Taub E, Uswatte G, Morris D, Giuliani C, Light KE, Nichols-Larsen D. Effect of constraint-induced movement therapy on upper extremity function 3 to 9 months after stroke: The excite randomized clinical trial. JAMA. 2006;296:2095-2104

10. Miller EL, Murray L, Richards L, Zorowitz RD, Bakas T, Clark P, Billinger SA. Comprehensive overview of nursing and interdisciplinary rehabilitation care of the stroke patient: A scientific statement from the american heart association. Stroke. 2010;41:2402-2448

11. Alberts MJ, Latchaw RE, Jagoda A, Wechsler LR, Crocco T, George MG, Connolly ES, Mancini B, Prudhomme S, Gress D, Jensen ME, Bass R, Ruff R, Foell K, Armonda RA, Emr M, Warren M, Baranski J, Walker MD. Revised and updated recommendations for the establishment of primary stroke centers: A summary statement from the brain attack coalition. Stroke. 2011;42:2651-2665

12. Schwamm LH, Pancioli A, Acker JE, 3rd, Goldstein LB, Zorowitz RD, Shephard TJ, Moyer P, Gorman M, Johnston SC, Duncan PW, Gorelick P, Frank J, Stranne SK, Smith R, Federspiel W, Horton KB, Magnis E, Adams RJ. Recommendations for the establishment of stroke systems of care: Recommendations from the american stroke association's task force on the development of stroke systems. Stroke. 2005;36:690-703

13. Gropen T, Magdon-Ismail Z, Day D, Melluzzo S, Schwamm LH. Regional implementation of the stroke systems of care model: Recommendations of the northeast cerebrovascular consortium. Stroke. 2009;40:1793-1802

14. Goldstein LB, Bushnell CD, Adams RJ, Appel LJ, Braun LT, Chaturvedi S, Creager MA, Culebras A, Eckel RH, Hart RG, Hinchey JA, Howard VJ, Jauch EC, Levine SR, Meschia JF, Moore WS, Nixon JV, Pearson TA. Guidelines for the primary prevention of stroke. Stroke. 2011;42:517-584

15. Bushnell CD, Olson DM, Zhao X, Pan W, Zimmer LO, Goldstein LB, Alberts MJ, Fagan SC, Fonarow GC, Johnston SC, Kidwell C, Labresh KA, Ovbiagele B, Schwamm L, Peterson ED. Secondary preventive medication persistence and adherence 1 year after stroke. Neurology. 2011;77:1182-1190



Co-Chairs: Gregory Albers, Ting Lee, Steven Warach

Members: Colin Derdeyn, John Detre, Chelsea Kidwell, Maartin Lansberg, Michael Lev, David Liebeskind, Randolph Marshall, Michael Moseley, William Powers, Howard Rowley, Keith St. Lawrence, Max Wintermark, Ona Wu, Gregory Zaharchuk

NINDS Liaison: Scott Janis


Numerous major advances in stroke imaging research have occurred over the past 5 years. Identification of penumbral tissue using both MRI and CT techniques has markedly advanced. In fact, neuroimaging is now being used for selection, monitoring, and testing of a number of different therapeutic interventions. The last few years have also seen more functional imaging markers being developed that could advance our knowledge of the underlying mechanisms of brain injuries and serve as surrogate outcomes for clinical trials. Molecular and cellular neuroimaging are new technologies being used to define pathophysiologic mechanisms. Because neuroimaging methods are noninvasive and can visualize brain structure and function in both patients and preclinical models, they offer the potential to elucidate mechanisms of recovery and serve as biomarkers for predicting recovery and monitoring treatment responses.

Key initiatives for future stroke imaging research, in priority order, are:

1) Serial imaging studies from acute to chronic timeframes using multimodal imaging, cerebrovascular reserve studies, and computational modeling to better understand the impact of cerebral hemodynamics, collateral flow, oxygenation, and brain metabolism upon tissue survival and function.

2) Randomized placebo controlled trial of IV tPA beyond 4.5 hours selected by ‘penumbral’ mismatch imaging.

3) Creation of an acute stroke imaging repository including a supporting infrastructure to enhance collaborative research to standardize and validate imaging protocols and processing methods.

4) Determine the most cost-effective imaging work-up for patients with both ischemic stroke and intracerebral hemorrhage (ICH) based on which modalities lead to treatment decisions that have been proven to affect outcomes.

5) Routine, practical, clinical metabolic imaging (e.g., CMRO2) added to the multimodal acute stroke imaging exam.

6) Investigate imaging markers that reflect injury to large and/or small blood vessels and their consequences.

7) Development of PET ligands and novel imaging techniques to allow imaging of synaptogenesis and other neuroplastic processes in humans.


Diagnostic Markers

To develop relevant diagnostic markers, it is critical to focus imaging research toward the better understanding of physiological changes associated with acute ischemic and hemorrhagic stroke.

Great progress has been made using multimodal CT or MRI (which includes anatomic imaging, noninvasive angiography, perfusion, and permeability maps); these methods have become routine in many centers for detecting structural and functional aspects of cerebrovascular disorders across a continuum from prevention to acute and subsequent recovery phases. In addition to multimodal CT and MRI, other modalities, such as transcranial Doppler ultrasonography, digital subtraction angiography (DSA), and PET will play a prominent role in broadening our understanding of cerebral ischemia and hemorrhage.

Vascular lesions of the aortic arch, neck, and intracranial segments of the cerebral circulation are now frequently detailed in sufficient detail with CTA or MRA, thus avoiding risk of invasive DSA. Architecture of atherosclerotic plaques, aneurysms, and arteriovenous malformations can be imaged with advanced approaches such as dynamic or time-resolved CT/MRI sequences, quantitative MRA (such as “4D-flow”), and selective arterial spin label (ASL) techniques that quantify flow across these lesions.

Penumbral markers and predictors of tissue fate

Penumbral imaging and development of imaging markers of tissue fate in acute stroke continue unabated as active research areas in neuroimaging. Notable advances in these areas include:

  • DEFUSE and EPITHET has shown the importance of mismatch (penumbra) in identifying patients who may benefit from intravenous tPA in extended therapeutic window
  • More perfusion processing software is using deconvolution and comparison between different deconvolution-based processing software was reported
  • Steps toward standardization of image acquisition and processing protocol and creation of a central Stroke Imaging Repository were initiated
  • MR Arterial Spin Labeling (ASL) for the measurement of cerebral perfusion and hence penumbra

Imaging markers of brain injury and prognosis

  • Further refinement of dynamic contrast enhanced CT and MR imaging techniques to study blood-brain barrier disturbances in both ischemic stroke and intracerebral hemorrhage
  • Susceptibility-weighted imaging (SWI) has become a powerful clinical tool to visualize venous structures and iron in the brain and to study diverse pathologic conditions including microbleeds.
  • Studies employing diffusion-weighted imaging have begun to characterize the frequency and clinical significance of silent “microinfarcts” in cerebrovascular diseases including intracerebral hemorrhage, ischemic stroke and cerebral amyloid angiopathy
  • Progress has been made to use 17O-MRI and T2* to assess oxygen extraction fraction of the brain
  • Resting state fMRI as early imaging markers of neurodegeneration
  • Correlative positron emission tomographic (PET) and CT exams with hybrid PET/CT scanners to study relationship of perfusion to brain metabolism
  • Development of MR gradient and spin echo sequences in dynamic contrast enhanced MR imaging studies to quantitatively map cerebral microvascular structure
  • BOLD MRI with hypercapnic (CO2) challenge for the measurement of cerebrovascular reserve
  • Imaging of hypoxia in the penumbra with F-19 FMISO PET imaging was demonstrated
  • PET/MR and PET/CT imaging of arterial plaque characteristics to identify vulnerable plaque
  • Development of targeted nanoparticle probes to investigate atherosclerotic plaque and injuries to vessel wall

Neuroimaging for selection, monitoring, and testing of therapeutic interventions

Biological case definition by neuroimaging has been increasingly used to identify patients eligible for inclusion into acute stroke and preventive randomized controlled trials. Examples of such trials that have been reported in the last 5 years include: Desmoteplase In Acute Stroke - 2 (DIAS-2) used computed tomography (CT) or magnetic resonance (MR) perfusion imaging to identify tissue at risk; three trials of decompressive hemicraniectomy HAMLET, DESTINY, DECIMAL used acute lesion size by CT or diffusion weighted MRI (DWI); Factor Seven for Acute Hemorrhagic Stroke (FAST) trial eligible patients had intracerebral hemorrhage on CT, and the NINDS-sponsored trials: Carotid Occlusion Surgery Study (COSS) randomized patients with higher risk of recurrent stroke identified by increased oxygen extraction fraction on PET; Carotid Revascularization Endarterectomy vs. Stenting Trial (CREST) defined carotid stenosis by angiography, ultrasonography, CT angiography (CTA) or MR angiography (MRA); Stenting vs. Aggressive Medical Management for Preventing Recurrent Stroke in Intracranial Stenosis (SAMMPRIS) selected patients with intracranial stenosis by transcranial Doppler, MRA, or CTA; Secondary Prevention of Small Subcortical Strokes (SPS3) selected patients with demonstrably small infarcts on MRI.

Definitive evidence in a large representative sample of patients presenting with possible stroke established a 3 to 5-fold superior sensitivity of DWI over non-contrast CT for acute stroke diagnosis. Multimodal MRI and CT (including vascular and perfusion imaging) are increasingly utilized in evaluating acute stroke, even within proven time windows for thrombolysis, but cost-effective and clinical effectiveness studies are needed to assess whether this conveys added benefit.

There has been progress toward defining and refining concepts of the ischemic core and penumbra (mismatch) volumes that are targets for patient selection into thrombolytic and recanalization therapies with MRI and to a lesser degree with CT.

Molecular and Cellular Neuroimaging: Develop new technologies to define pathophysiologic mechanisms

  • MRI tracking of cells (e.g., inflammatory, stem) labeled with iron oxide
  • 11C-PIB and 18F-florbetapir beta amyloid imaging agents to investigate vascular dementia/Alzheimer's Disease interactions and overlap
  • 11 C-PBR28 microglia ligand for neuro-inflammation in ischemic stroke
  • Development of MRI methods for quantitative CMRO2 in normals
  • Hybrid imaging platforms of PET/CT and PET/MRI.

Functional imaging of patients to clarify mechanisms of recovery of function and enhance recovery

Nearly all stroke patients would benefit from interventions that enhance neural recovery, and even modest gains could have an enormous public health and economic impact.

Functional MRI (fMRI). Examining patterns of brain activation following stroke was an early clinical application of blood oxygenation dependent (BOLD) fMRI, and demonstrated intralesional, perilesional, and contralesional activation. Over the last 5 years, serial fMRI and correlation of fMRI findings with behavioral outcomes have led to both mechanistic hypotheses and predictive models based on observed activation patterns, though many questions remain. BOLD fMRI has shown both regional activation and widely dispersed activity among networks of brain regions. Persistence of atypical activity in secondary motor and language regions including the opposite hemisphere generally correlates with poorer recovery. Multivariate statistical approaches have shown distributed motor activation in the first few days after stroke that correlate with subsequent motor recovery. Domain-specific network activity can also be detected with the patient at rest, thereby avoiding a performance confound that is often problematic in patients with neurological deficits.

Neuromodulation. A persistent post-stroke imbalance of activity between the ipsilesional and contralesional hemisphere has supported the use of transcranial magnetic stimulation (TMS) and transcranial direct current stimulation (tDCS) as physiological interventions to rebalance hemispheric activity or to target specific regions of abnormal activation. TMS has also demonstrated changes in cortical excitability that suggests specific (early) time windows of enhanced neuroplasticity. The use of fMRI-guided TMS can also provide a means of testing mechanistic hypotheses.

Structural MRI. Significant advances have been made using structural neuroimaging to study stroke recovery. Voxel-based lesion-symptom (VSLM) mapping leverages the statistical methods developed for fMRI to explore the behavioral consequences of stroke lesions with high spatial resolution. Diffusion tensor imaging (DTI) allows in vivo identification of white matter tracts, providing greater anatomical specificity of strokes involving the white matter. DTI has suggested that integrity of certain white matter tracts may correlate with motor recovery, and response to therapy. A growing body of evidence also suggests that use-dependent plasticity after stroke is reflected by changes in gray matter volume, gray matter density, and white matter anisotropy.


Despite the tremendous progress in imaging research outlined above, there remain many unresolved issues. Future imaging studies will be important to translate promising preclinical findings into new therapies and to establish imaging biomarkers that may complement peripheral blood markers. Novel noninvasive imaging may monitor oxygenation, temperature, pH changes, osmotic shifts, metabolic alterations with standard and hyperpolarized MR spectroscopy. These novel imaging strategies will also be useful to monitor arteriogenesis and neurogenesis with restorative therapies and to predict long term recovery. In particular, the emergence of simultaneous MR-PET systems will likely lead to a richer understanding of molecular and oxygenation changes in ischemia, and this is predicted to be a fertile area for identification and validation of novel biomarkers. Below is a list of research areas that are considered to be particularly promising:

  • Cellular and molecular imaging. Although neuroplasticity at the cellular level is presumed to underlie stroke recovery, there remain few cellular and molecular neuroimaging methods that can measure these processes directly and noninvasively in vivo. MRI or radionuclide imaging has been used to monitor the migration of labeled stem cells, both in preclinical models and in human patients. Changes in synaptic activity have been imaged using two-photon microscopy in an adult animal model, demonstrating that dendritic spine turnover is rapid and dynamic during ischemia and reperfusion. In humans, multi-modality imaging with (11)C-flumazenil PET and MR or CT perfusion may provide a marker of selective neuronal loss although the clinical correlation with post-stroke recovery remains unproven.
  • The use of multimodality imaging to improve on prediction models including fMRI multivariate techniques, DTI imaging and physiological probes such as cortical excitability indices with TMS.
  • Computational fluid dynamics approaches are rapidly evolving to image flow patterns in complex and diminutive intracranial lesions such as aneurysms and AV malformations, and may help predict how cerebrovascular hemodynamics will change based on revascularization, such as with bypass or vessel sacrifice. Future research will likely combine computational fluid dynamic models with imaging data to elucidate the hemodynamic impact of stenoses and collateral flow patterns that may stratify risk and explain therapeutic effects of revascularization techniques such as stenting, endarterectomy, or medical strategies.
  • Vessel wall abnormalities such as atherosclerosis and dissection are being evaluated using high-field MRI, ultrasound, PET, and novel molecular probes or contrast agents such as iron-oxide nanoparticles. Future studies may help explain why percent luminal stenosis is not always predictive of stroke risk.
  • Imaging of collateral circulation with CTA, MRI and DSA provides insight on compensatory mechanisms such as ischemic preconditioning and prognostic information that may be used to identify high-risk clinical populations.
  • Cerebrovascular reserve may be investigated with pharmacological means (such as acetazolamide), CO2 inhalation, or breath-hold to predict risk of subsequent events, supplemented by novel dynamic autoregulation testing. Similar to functional testing in cardiac disease, development of less invasive MRI-based measures (e.g. ASL, BOLD, OEF) will be important.
  • Microvascular lesions causative in silent strokes and hemorrhage have recently been advanced with MRI, including gradient-echo based sequences, such as SWI and susceptibility mapping. Cerebral microbleeds and macrobleeds continue to provide new data on small vessel disease, inflammation, and risk of ICH. Longitudinal studies of patients with silent strokes and leukoaraiosis using MRI will be able to gauge risk of symptomatic stroke and cognitive decline.
  • Serial use of multimodal CT or MRI will be pivotal in elucidating the dynamic aspects of perfusion in TIA, acute and subacute stroke, chronic hypoperfusion states, and vasospasm. CT and MRI perfusion studies have recently highlighted important hemodynamic changes of recanalization and reperfusion beyond the acute go-no go decision for revascularization. Serial ASL and other perfusion techniques will clarify the impact of collateral perfusion, hyperperfusion, and permeability changes in the blood-brain barrier (BBB) that may initiate hemorrhage. Imaging can be used to shed light on the complex interactions of emboli and hypoperfusion in various stroke subtypes.
  • Advanced imaging can now be used to map lesion topography and heterogeneity with fusion and 3D techniques, image clot composition, pinpoint bleeding with the CTA spot sign, and track stem cells and inflammation within the brain. This wide array of imaging techniques has potential to chronicle effects of neuroprotection, hypothermia, thrombolysis, and endovascular therapies in ongoing clinical trials.
  • Functional MRI and transcranial magnetic stimulation (TMS) have provided novel and promising approaches to deducing mechanisms of functional reorganization during stroke recovery.

Standardization of image acquisition protocol and validation of different processing software remains incomplete.

  • The Acute Stroke Imaging Roadmap established some recommendations and guidelines for MRI (diffusion-weighted and perfusion-weighted imaging) and perfusion CT protocols, however it is unclear the extent by which these recommendations have been adopted by stroke imaging centers.
  • Stroke Imaging Repository (STIR) is not widely subscribed
  • Do perfusion CT and MR studies produce equivalent results?
  • Which perfusion parameter(s) is(are) optimal for identifying the penumbra (infarct), Tmax, MTT, CBF, CBV, collaterals or combinations thereof?
  • Which one(s), among core, penumbral and vascular imaging, improve(s) clinical outcome and/or extend the time window for reperfusion treatment?
  • What is the optimal time for follow-up imaging to validate imaging markers of penumbra and tissue fate?

Although imaging markers for hemorrhagic transformation subsequent to ischemic stroke have been developed, for instance, blood-brain barrier permeability surface product, markers for symptomatic versus asymptomatic intracranial hemorrhage are not available.

  • Besides detection of vulnerable plaques in extracranial vessels, imaging markers for injury to vessel wall, small vessel disease, amyloid angiopathy are needed
  • Role of visualizing venous structures and microbleeds with high resolution susceptibility weighted imaging in the management of stroke
  • Role of collaterals in the determination of tissue fate in reperfusion treatment
  • FLAIR/DWI mismatch for reperfusion treatment selection in “wake up” and other strokes of unknown onset
  • Non-contrast CT (NCCT) and CTA source image mismatch for selection of patients for reperfusion treatment; does material decomposition with dual energy CT enhance utility of this method?
  • Practical guidelines in IV and endovascular tPA triage at ALL time points concerning the role of NCCT alone versus NCCT plus CTA +/- PCT penumbral imaging.
  • The potential importance of cerebral microbleeds defining subcategories of cerebrovascular disease (e.g., amyloid angiopathy) and intracranial bleeding risk
  • CTA spot sign for acute ICH extension; now being used in trials.
  • Defining the Spectrum of Reversible Cerebral Vasoconstriction Syndrome by MRI
  • MR Diffusion Tensor Imaging (DTI) has emerged into clinical practice to study white matter tracts related to cognitive and motor outcomes and plasticity. Although, still largely a research clinical sequence this seems to have great potential.
  • Resting state connectivity with fMRI may have potential to study functional deficits and recovery in stroke and vascular cognitive impairment
  • The advent of good quality, relatively rapid whole brain arterial spin labeling (ASL) sequences as an alternative and potentially quantifiable alternative to gadolinium based MRI perfusion methods. Especially important since recognition of gadolinium toxicity (nephrogenic systemic fibrosis) has limited the use of MR perfusion in patients with renal failure.
  • Whole brain CT Perfusion with 256 row CT scanners may overcome many of the spatial limitations of CT Perfusion for mismatch imaging
  • One emerging area is that of simple Radiology economics – cost effective, downsizing of resources, medical costs, and reimbursements.
  • High resolution MR imaging of the components of atherosclerotic plaque to define risk more accurately than percent stenosis. This could stimulate the next wave of preventive trials of intra or extra-cranial atherosclerosis, especially in asymptomatic patients
  • Identifying acute stroke patients with imaging evidence of stroke within tPA time window (e.g., normal FLAIR) despite unknown onset time (the wake-up stroke)
  • Consensus on the optimal perfusion parameters and thresholds to select patients for trials or interventions is lacking and awaits direct comparisons of competing models on large datasets and in prospective, randomized controlled trials.
  • Clinical PET/CT & PET/MR. Besides providing a way to validate MR CMRO2 measurements with the acute diffusion and perfusion exam, it offers the ability to combine molecular targets, such neuroinflammation, with hemodynamic and structural imaging.
  • Continued development of MRI measurements of CMRO2
  • Human 7 Tesla and higher field strength brain imaging
  • Radiological phenotyping – using a multi-parameter MR exam to predict genetic information. An MR will best identify the genetic background of not only a stroke, but the vascular/cognitive/neuronal fingerprints.


1) Serial imaging studies from acute to chronic timeframes using multimodal imaging, cerebrovascular reserve studies, and computational modeling to better understand the impact of cerebral hemodynamics, collateral flow, oxygenation, and brain metabolism upon tissue survival and function.

2) Randomized placebo controlled trial of IV tPA beyond 4.5 hours selected by ‘penumbral’ mismatch imaging.

3) Creation of an acute stroke imaging repository including a supporting infrastructure to enhance collaborative research to standardize and validate imaging protocols and processing methods.



Co-Chairs: Eng Lo, David Greenberg, Midori Yenari

Members: Stephen Back, Jun Chen, Rona Giffard, Patricia Hurn, Costantino Iadecola, Paul Rosenberg, Michael Tymianski, Raghu Vemuganti, Berislav Zlokovic

NINDS Liaisons: Francesca Bosetti, Tim LaVaute



  • Dissection of cell-cell signaling in the “neurovascular unit” involving crosstalk between neuronal, glial and vascular elements in the CNS.
  • Understanding of crosstalk between the neurovascular unit and systemic responses in the body (non-CNS vascular systems, circulating blood as well as immune systems).
  • Emerging appreciation that neurovascular unit-systemic responses apply not only to stroke, but may also be important for the pathophysiology of dementia and neurodegeneration.

Unresolved Areas and Opportunities:

  • How do non-neuronal responses (e.g. microglia and pericytes) contribute to injury and repair? And do these mechanisms differ in gray vs white matter?
  • How are cellular mechanisms linked to emerging molecular pathways involving mitochondrial energetics, micro RNA (miRNA), and post-translational modifications?
  • How are all these pathways influenced by gender, age and co-existing disease state (co-morbidities), all factors that could impact stroke outcome in patients?
  • Can we use in vivo imaging/optogenetic tools together with genomic/proteomic/metabolomic approaches in novel transgenics to study the entire neurovascular unit simultaneously?
  • Need to understand how responses in the neurovascular unit transition from initial injury during acute stroke into repair and neurovascular remodeling during stroke recovery, as well as neurodegeneration more broadly.

Recommendations and Priorities:

  • Conceptual Development: Need to better understand how interactions in all elements of the neurovascular unit along with systemic immune responses underlie the transition from injury into repair after stroke.
  • Model Development: Need to build better models (molecular, cellular, animal) that capture pathophysiology as it is influenced by altered neuro-cerebro-vascular baselines (e.g. age, gender, inflammation and other diseased or states).
  • Translational Hurdles: In addition to timing issues that need to be clarified in stroke models, there needs to be a better understanding of how these timelines translate into humans. For example, if neuroprotection is observed within a 6 hour window in rodent stroke, what does that mean in stroke patients? How does dose, timing in rodents translate into humans? What preclinical standards should we expect to see prior to embarking on human studies?


There have been many significant advances in stroke research. Clinical advances have included the characterization of stroke genes and risk factors, the development of potent new anti-thrombotics, the emerging lengthening of the time window for tPA, the increasing use of mechanical reperfusion, and new approaches to enhance stroke recovery. From a basic science perspective, perhaps the most influential conceptual shift has been the continued development of the “neurovascular unit”, which has provided a framework for “looking beyond the neuron” and assessing integrated responses between all cell types and systems after stroke.

Cell-cell interactions within the neurovascular unit have been described by many groups. From a functional perspective, this idea makes sense. Without crosstalk between astrocytes, pericytes and cerebral endothelium, the blood-brain barrier phenotype cannot be elaborated. Without crosstalk between astrocytes and neurons, release-reuptake kinetics of various neurotransmitters cannot be controlled. Without crosstalk between neurons and cerebral endothelium, the hemodynamic coupling between metabolism, energy demand and blood flow cannot be properly regulated.

In the context of stroke, dysfunctional crosstalk between various elements of the neurovascular unit is now known to underlie the various pathomechanisms that lead to brain cell injury and death. Matrix metalloproteinases from endothelium and astrocytes degrade blood-brain barrier integrity, leading to edema and hemorrhage. Reversal in glutamate transporters in astrocytes may exacerbate neuronal excitotoxicity. Loss of endogenous free radical scavenging systems in astrocytes may worsen oxidative stress. Activation of microglia may enhance inflammation and tissue damage. Altogether, stroke pathophysiology is best investigated as an integrated response in all cell types in the brain.

In addition to cell-cell signaling between neurons, glia and the cerebral blood vessels, a major recent advance comprises the realization that the central neurovascular unit also communicates with the systemic responses in blood and immune systems. Genetic responses in various circulating blood cells may provide both novel mechanisms as well as biomarkers. Responses in the immune system may profoundly influence the way the stroke patient evolves over time. And finally, plasticity and recovery in the CNS cannot be properly interpreted without understanding the systemic responses in circulating endothelial progenitor cells.

Beyond stroke, the idea of the neurovascular unit may also apply to many other CNS disorders as well. Early blood-brain barrier leakage has been documented in animal models of ALS and vascular dementia. Sick astrocytes from ALS patients have been shown to be directly neurotoxic to wildtype motoneurons. Matrix metalloproteinases have been shown to contribute to the mechanisms of dopaminergic neuronal death in models of Parkinsons disease. Dysfunctional astrocytes have been found surrounding amyloid plaques in models of Alzheimers disease. Taken together, emerging data strongly suggest that the neurovascular unit may be a powerful concept for investigating not only stroke but also broader spectrums of CNS disorder and neurodegeneration.

Whereas the neurovascular unit has clearly provided a conceptual framework for examining the mechanisms of acute stroke pathophysiology, emerging data now also suggest that cell-cell signaling between all elements of this unit might also underlie stroke recovery. Indeed, the overarching theme here may be considered to be the biphasic nature of neurovascular mediators in general. The same mediators or responses that induce injury during acute stroke may conversely promote neurovascular repair and remodeling during stroke recovery. The classical example involves NMDA excitotoxicity. Over-activation of the NMDA receptor by excess glutamate during acute stroke is neurotoxic. But without NMDA signaling, synaptic and dendritic plasticity cannot take place during stroke recovery. More recently, it has also been shown that proper movement of newborn neuroblasts within the rostral migratory stream required NMDA cross-talk between the guiding sheath of astrocytes and the migrating neuroblasts. The same holds true for matrix metalloproteinases (MMPs). MMPs induce neurovascular damage and blood-brain barrier leakage during the early stages of cerebral ischemia. But without MMPs, one cannot have proper angiogenesis and vasculogenesis during the endogenous repair after stroke. Even well known “bad actors” such as reactive oxygen and nitrogen species may play biphasic roles. Large amounts of radicals will certainly cause cellular damage. But lower homeostatic levels of radicals may be essential for cell signaling.

In summary, within the context of neuro-cerebro-vascular degeneration (the charge for our section), a major advance may be the emerging understanding of cell-cell signaling in the “neurovascular unit” involving crosstalk between neuronal, glial and vascular elements in the CNS. And along with systemic responses in the body (non-CNS vascular systems, circulating blood as well as immune systems), a further dissection of how these mechanisms are regulated may allow stroke pathophysiology to be targeted at multiple levels. And perhaps most importantly, as proposed in the section below, it will be critical to truly understand how various responses and cell-cell signals within the neurovascular unit transition from initial injury into delayed repair and remodeling during stroke recovery.


We now know that there is important crosstalk within the “neurovascular unit” as well as between the “neurovascular unit” and systemic responses, but the detailed mechanisms of these interactions remain poorly understood. How do non-neuronal responses (e.g. microglia and pericytes) contribute to neurovascular injury and repair? Do mechanisms in white matter differ from those in gray matter? How are cellular mechanisms linked to emerging molecular pathways involving mitochondrial energetics, micro RNA (miRNA), and post-translational modifications? How are all these pathways influenced by gender, age and co-existing disease state (co-morbidities), all factors that could impact stroke outcome in patients? And how do these neurovascular unit responses help us link common pathways in stroke and neurodegeneration?

The vast majority of brain tissue in the cerebral hemispheres is white matter, but the vast majority of stroke research is focused on gray matter. The role of white matter injury in outcomes from stroke in either adults or neonates remains under-studied. The mechanisms of white matter injury are emerging as markedly different from gray matter injury. White matter injury in both human and adult neonates involves aberrant regeneration and repair mechanisms, whereby oligodendrocyte progenitors acutely degenerate but surviving glia are able to proliferate/ regenerate. The regenerated progenitors survive but in an arrested state of development where they fail to differentiate to oligodendrocytes. These emerging cellular and molecular mechanisms have the potential to lead to new therapies to promote regeneration and repair of injured white matter.

With the new emphasis on the neurovascular unit protection, it may be important to examine the responses of all cell components of the neurovascular unit to the neuroprotective strategy tested. Well controlled “in-parallel” studies should be conducted to compare the efficacy of therapies that aim at a common target shared by multiple components of the neurovascular unit or at a cell-specific target.

A substantial amount of new research opportunities also exist in the understanding of immunological factors in the brain, including the invading peripheral factors via BBB disruption. The questions still remain as to how blood immune cells and possibly the activated platelet modulate the cellular components of BBB upon ischemic injury. Soluble factors and cell-cell signaling molecules that are required for peripheral immune cells to cross the barrier remain to be identified. Since those peripheral factors are early responders to ischemic injury and easily therapeutically accessible, they may represent potential diagnostic biomarkers and clinically viable targets for treatment. Similarly, responses in circulating blood cells may also offer new windows into mechanisms and putative biomarkers. The rapid development of in vivo imaging tools together with novel approaches such as optogenetics in transgenic animal models may finally allow us to study the entire neurovascular unit in its entirety.

Even if one were to develop the optimal therapeutic drug that targets all elements of the entire neurovascular unit, challenges may still remain in terms of drug delivery. How do we reach all cells we need to reach? The same may apply for the development of cell-based therapies. How do we know what cells are effective? How can we best deliver the cells? What are the target mechanisms? When is the optimal time for therapy, taking into consideration the biphasic nature of neurovascular signals and the complexities of host-cell interactions? There remain many unknowns but these are all opportunities for exciting research.

Ultimately, one unifying theme may be the fact that “timing is everything” in stroke evolution. The same mediators that induce damage during the acute phase may surprisingly promote repair during the recovery phase. For example, inflammation is known to be biphasic, but exactly how and when it may be detrimental or beneficial need to be better defined. We need to fully understand how responses in the neurovascular unit transition from initial injury during acute stroke into repair and neurovascular remodeling during stroke recovery.


1) Conceptual Development: Need to better understand how interactions in all elements of the neurovascular unit along with systemic/systemic immune responses underlie the transition from initial injury into repair and remodeling after stroke.

2) Model Development: Need to build better models (molecular, cellular, animal) that capture pathophysiology as it is influenced by altered neuro-cerebro-vascular baselines (e.g. age, gender, inflammation and other diseased or states) as they may pertain to stroke specifically, as well as neurodegeneration more broadly.

3) Translational Hurdles: In addition to timing issues that need to be clarified in stroke models, there needs to be a better understanding of how these timelines translate into humans. For example, if neuroprotection is observed within a 6 hour window in rodent stroke, what does that mean in stroke patients? How does dose, timing in rodents translate into humans? What preclinical standards should we expect to see prior to embarking on human studies?



Co-Chairs: Patricia Hurn, Kyra Becker, Raymond Swanson

Members: Dennis Choi, Ulrich Dirnagl, Marc Fisher, Mark Goldberg, Louise McCullough, Thaddeus Nowak, Jr, Sean Savitz, Richard Traystman

NINDS Liaisons: Francesca Bosetti, Erik Runko


The working group identified six major research advances related to neurovascular protection, many of which were positively influenced by the SPRG process. One key new opportunity for the field was focused interactions between the injured CNS and the so-called super-systems, particularly the immune system. Five major unresolved areas were identified and analyzed: the fundamental question of whether NvPr is still a useful and productive area of stroke research; the repetitive but unresolved problem that pre-clinical testing of pharmaceutical compounds, biotechnologies or devices remains sub-optimal; the observation that the hypothesis of neurovascular protection has not been tested in clinical trials, the need for new methods to close the gap between bench stroke research and subsequent clinical studies; and the directive that NvPr research cannot be focused solely on brain. Priorities were identified for future stroke research directions: 1) the improvement of the cellular-animal study-clinical trial interface, 2) expansion of our current clinical trial repertoires to include approaches such as specific pathology-focused, exclusive patient trials.


The concept and language of “neurovascular unit” is now strongly a part of our field. This is an important advance in that it has created a paradigm shift away from “neuron-centric” neuroprotection.

The formation of SPOTRIAS has advanced the field with high impact. This advance is also an outcome of SPRG discussions and recommendations.

The role of sex differences in stroke pathobiology and epidemiology has become increasingly clear at many levels of analysis.

Publication of stroke research now requires a higher standard for acceptance in key peer-reviewed journals that specialize in stroke. Examples include the journal Stroke and the Journal of Cerebral Blood Flow and Metabolism.

The recent demonstration of peri-infarct depolarizations in human stroke, including population studies, is highly valuable and demonstrates an important interface with the neurovascular unit components.

Another important advance is the increased understanding of inflammatory signaling and immune cell mechanisms in neurovascular protection. This has led to an initial understanding that brain protection is not enough (see unresolved area discussion below). A key concept is that peripheral organs and overall systemic responses to the injured brain is a causal key to stroke pathology and recovery.


New Opportunities

1) An exciting opportunity is to further understand the role of non –central nervous systems in stroke, the so called “super-systems” (e.g. immune system). This area will not only contribute important basic science to the field, but will provide a new opportunity to advance therapy in an integrated way relevant to patients. 2) Along similar lines of discussion, new studies are needed at the bench and in clinic that examine the realities of systemic therapies used in patients. An example would be a study of a candidate drug’s efficacy when tested along with/ in interaction with antibiotics or beta receptor antagonists that patients require to control co-morbid conditions during the stroke recovery period.

Unresolved Areas

Why has neuroprotection not been successfully moved from bench to clinic?

Pre-clinical studies have firmly established the validity of the concept of neurovascular protection – i.e., that pharmacologic and other interventions can prevent death of cells that would otherwise die as a result of cerebral ischemia. There has also been some success, albeit limited, in the clinical arena of neuroprotection, with hypothermia now being standard of care for patients who are comatose after cardiac arrest and for neonatal asphyxia. However, no pharmacologic agents have yet achieved success in clinical trials, and this is seen as a major disappointment in the field. It is also widely documented that numerous pharmacological agents can reduce ischemic cell death by biologically plausible mechanisms in cell culture and animal models of cerebral ischemia. It is thus of paramount importance to identify why this laboratory success has not translated into any clinical success. We identified three factors that have likely contributed to this failure.

The first is that clinical trials have extrapolated too far from pre-clinical data. There are essentially no pre-clinical studies showing NvPr at time points greater than 6 hours after ischemia onset, yet many earlier clinical trials used inclusion criteria that extended well beyond 6 hours. The pre-clinical data suggest that it may not be realistic to expect neurovascular protectants to have efficacy at time points beyond 3- 6 hours, unless they are targeting inflammation, peri-infarct depolarizations, or events that occur at longer timepoints after cerebral ischemia onset. Clinical trials have also in some cases evaluated drug dosing in ranges not supported by the pre-clinical data. Second, clinical trials of NvPr agents should be centered at shorter time points after ischemia, and/or targeting viable tissue (e.g. DWI/perfusion mismatch on MRI imaging). It is recommended that drug development be focused on targeting pathogenic events that occur at longer time points after ischemia.

Conversely, many pre-clinical animal models of stroke differ in significant respects from human clinical stroke. For example, the widely used 30 -60 minutes of middle cerebral artery (MCA) occlusion, with test drugs administered immediately or shortly after reperfusion, does not model a common stroke scenario. In addition, very few studies have examined efficacy as “add-ons” to thrombolytic therapy. While not a novel recommendation, we again feel obligated to recommend that researchers emphasize longer ischemic intervals and longer delays to treatment in animal models. Similarly, pre-clinical testing of agents in combination with thrombolytics should be encouraged, with the aim of extending the thrombolytic therapeutic window and/or reduce hemorrhage. Pre-clinical studies beyond large vessel occlusion are needed to address the entire spectrum of stroke; e.g. lacunar infarcts, white matter disease, hypertensive hyalinosis, and others. Further discussion of these points is in the sections below.

The third comment is that the field has largely avoided the use of aged animal models with co-morbidities, not because of lack of understanding of the importance of these factors but because of the cost or need to focus on pathobiological mechanisms of interest to the investigator. The vast majority of pre-clinical animal stroke studies continue to use healthy young males, despite our awareness that the human population most subject to stroke is aged, with co-morbidities such as hypertension and diabetes. On the other hand, there was much discussion amongst the working group that there is at present no clear evidence that the use of aged animals or those with co-morbidities will make significant impact on the success of NvPr. It might be that assumptions are necessary in the use of animal models and that these assumptions are based on the desire to avoid large sets of studies that are repeated over complex variable combinations of age, hypertension, diabetes and others co-morbidities. The literature now emphasizes the role of biological sex in NvPr, yet the importance of gender in designing therapy is unclear. The field needs to create accepted, evidence-based paradigms that are attainable and productive regarding when and to what depth factors like age, sex and co-morbidities are to be studied in NvPr.

A repeating theme: pre-clinical testing is still not optimal

There has been considerable progress in the application of more rigorous standards in preclinical study design and data analysis, apparent in the current editorial policies of some major journals in the stroke field. The updated STAIR recommendations of 2009 appropriately place increased emphasis on physiological monitoring, including assessment of perfusion, and note the potential variability of occlusion efficacy by intravascular approaches. However, there is probably not yet adequate consensus regarding the criteria by which to assess technical issues related to model execution, or the quality of the stroke models themselves.

In the context of a definitive protection study, some level of physiological monitoring must be expected throughout the interval during which pathology evolves, and no parameter can be ignored. For example, peri-infarct depolarizations that contribute directly to infarct expansion persist for a day or more in awake animals, and have been documented clinically for up to several days following stroke. Perfusion responses to such events are directly impacted by blood pressure and respiratory status, and their metabolic impact will be determined by temperature and glucose levels. Unrecognized treatment effects on such parameters, if not also clinically relevant, confound interpretation of study results. Preliminary evaluations can reasonably be expected to include more limited assessment. However, continuous telemetric monitoring of all parameters of interest is becoming increasingly feasible, and intervention studies in fully instrumented awake animals should be targeted as the ideal preclinical test.

Not all models are created equal. Although perhaps technically simpler than direct surgical approaches, commonly used intraluminal filament occlusions introduce a number of confounds. 1) Details of the method vary markedly among laboratories. 2) The devices variably obstruct numerous arteries branching from the internal carotid artery, producing ischemia in territories beyond that supplied by the MCA, e.g., involvement of the hypothalamus impacting temperature regulation. 3) To limit the mortality associated with this larger ischemic territory, the approach is most often applied in strains with robust collateral perfusion, perhaps less relevant to the clinical situation. 4) Comparatively short occlusion durations are often intended, but the potential endothelial damage produced by filament insertion and removal is rarely considered. The resulting pro-thrombotic state increases the risk of persistent secondary occlusion, thereby extending the apparent window for intervention in such models. More emphasis should be placed on alternative embolic models that limit these complications.

Testing the NvPr hypothesis

In addition to the need for further refinement of animal studies, clinical trials in NvPr have been flawed in many respects. Therefore, it is reasonable to conclude that the NvPr hypothesis has been tested in the face of many technical and conceptual restrictions. Going forward, we must consider if a NvPr drug with a favorable preclinical evaluation package is to be moved into clinical development, how could the clinical trial be better implemented? Firstly, it should be recognized that even a modestly effective and safe NvPr drug could have a substantial impact on the global treatment of ischemic stroke. For some hospitals in the United States and worldwide, especially in less developed countries that lack the infrastructure and personnel to give i.v. tPA, it is possible that a neuroprotective drug could be easier to initiate, so acute stroke therapy could be expanded to locales where tPA is not available. Secondly, very early initiation of a NvPr drug, as is being tested in the current FAST-MAG trial of i.v. magnesium, could provide more patients with potentially salvageable ischemic tissue and these patients could have larger volumes of such tissue than those not given the agent. If this hypothesis, which has been tested in animals, is correct, then a larger number of patients could ultimately be treated with reperfusion therapies such as i.v. tPA, i.a. tPA and/or clot-removing devices at potentially later time windows. A third consideration is that NvPr may be an adjunctive therapy after successful reperfusion. For example, reperfusion may induce secondary injury, so-called reperfusion injury, and NvPr could be envisioned to ameliorate this process if given after reperfusion occurs. Another concern is that tPA in addition to its beneficial effect of inducing clot dissolution also has deleterious effects on the brain parenchyma. NvPr drugs targeted toward these deleterious effects and given after i.v. tPA could improve brain tissue protection and ultimately clinical outcome.

Several types of future neuroprotection trials can be envisioned. The first consideration would be to perform a trial that confirms that a drug prevents some portion of the ischemic penumbra from progressing into infarction. Such a proof of concept trial would be done in phase II. Patients with MRI-confirmed penumbra on diffusion/perfusion imaging would only be enrolled and not allowed to receive reperfusion therapy that would confound the data analysis. Therefore patients would need to be enrolled beyond the current 4.5 hour time window recommended for i.v. tPA. The goal of the trial would be to compare in the treated group versus the placebo group the percentage of “at risk” tissue in the diffusion/perfusion mismatch region that is shown to be infarcted on a subsequent FLAIR MRI study days after stroke onset. Current estimates suggest that if at least a 50% mismatch is present, pretreatment sample size of approximately 100 per group should be sufficient to document a significant treatment effect on ischemic tissue salvage. The next step after such a positive proof of concept trial would be a large phase III trial comparing this drug to placebo with a clinically relevant primary outcome measure such as the Modified Rankin Scale. This trial should again exclude concomitant reperfusion therapy, to avoid becoming a 4-armed trial where patients are randomized to the neuroprotective drug or placebo but also some percentage received i.v. tPA beforehand. The use of i.v. tPA in some patients will confound the primary outcome assessment and also reduce statistical power as compared to 2-armed trial of drug versus placebo. It is problematic to consider how to perform a monotherapy neuroprotection trial in many centers in the US, Europe and other developed regions around the world. However, there are good centers in these regions and in less developed countries where i.v. tPA is used sparingly or not at all that could be employed in such a monotherapy trial. These centers would address the important question of if a NvPr drug with a robust preclinical evaluation package and a phase II proof of concept trial demonstrating tissue salvage can improve clinical outcome. Such a phase III trial will be difficult to perform but it could be done if it is agreed that it will advance therapy for acute ischemic stroke, especially in underserved areas.

A clinical trial to evaluate the hypothesis that NvPr could extend and enhance the utility of reperfusion therapy has not yet been performed. Indirectly, the FAST-MAG trial may provide some tantalizing data to support the hypothesis, but it was not designed to truly test the hypothesis. In such a complementary trial, a safe drug or placebo would be initiated pre-hospital in the ambulance, as in the FAST-MAG trial. Upon arrival in the hospital, the patients would undergo a rapid clinical assessment, allowing for exclusion of stroke mimics or subjects with intracerebral hemorrhage on baseline imaging. In the target population of ischemic stroke patients, penumbral imaging with diffusion/perfusion MRI will then be done to determine the presence and volume of the mismatch. The basic hypothesis to be tested is that the volume of mismatch will be larger in the patients who receive the neuroprotective drug as compared to the placebo patients. It would also be hypothesized that patients with a mismatch of at least 20% would be greater in the active treatment group. After the imaging assessment, the treating physician would be able to initiate other treatment, i.e. i.v. tPA or i.a. therapy. Long term outcome data would be collected but not used as the primary outcome assessment, because these subsequent therapies would profoundly influence these clinical outcome measures. The primary assessment for the neuroprotective drug would be an imaging one. It is acknowledged that such a trial of penumbral extension by a NvPr drug would be difficult to implement. If early initiation of such an agent can extend the time window for reperfusion therapy and improve its effectiveness, then the impact on acute stroke therapy would be substantial.

Lastly, NvPr, as an adjunct to successful reperfusion to ameliorate the deleterious effects of reperfusion, has also not been tested in the past. In such a trial, a NvPr drug with a mechanism of action relevant to reperfusion injury such as an antioxidant or modulator of inflammation would be evaluated. Only patients with imaging results that confirm reperfusion would be randomized to treatment or placebo after completion of the reperfusion therapy. The goal of the trial would be to improve delayed clinical outcome with the NvPr drug plus reperfusion versus reperfusion alone. The sample size required for this trial would likely be quite large because demonstrating additional clinical benefit beyond that provided by the reperfusion therapy will be difficult. Another consideration for such a trial would be to evaluate the effects of the agent on safety outcome measures such as the rate of intracerebral hemorrhage associated with reperfusion. This outcome assessment will also require a large sample size.

New ways to address the gap between the bench and subsequent clinical studies

The emergence of the concept of “Translationally Relevant Research” over the past decade has influenced the design of both pre-clinical and clinical trials. It is increasingly recognized that the time to move a potentially efficacious therapy from the bench into the clinical setting is unacceptably long, averaging up to 15 years by some estimates. It is apparent to many that a major translational gap remains: a relative lack of support for exploratory studies in clinical populations. We often consider moving from successful pre-clinical agents directly into a Phase IIa trial, with minimal evaluation of the relevance of the therapeutic target in patients. Grants that have attempted to move basic science into the clinic in early phase studies are often considered overly ambitious or lacking in focus. Often criticisms arise regarding the study design and analysis for small early exploratory efforts that are included within the scope and funding of a “basic” RO1. These include early assessments of biological targets in humans. For example, can the biomarkers of injury used in animal studies be measured from human samples such as blood, CSF, or neuropathological samples? Do they change as expected with injury? Can therapeutic targets be monitored with imaging? Clinical trials that begin with safety testing and overlook biological targets are therefore not designed in a way to advance the bench data. Whether the target is relevant in disease isn’t assessed until Phase II. Prior to embarking on incredibly expensive clinical trials, more groundwork should be done to move from the animal to the patient to select the most appropriate therapeutic agents, and the most appropriate patient to give them to. Currently there are few funding opportunities for small scale studies in patients that directly emerge from animal studies. These types of “translational” studies are especially relevant for the field of neuroprotection. This type of funding would also be an important avenue from which to support physician-scientists and foster collaboration between basic and clinical stroke researchers.

Brain protection is not enough

There has recently been much progress in the concept that NvPr cannot be focused solely on the brain. For example, multiple studies (both experimental and clinical) have described key elements in support of this statement: 1) there are large changes in the peripheral immune response following stroke; 2) stroke-induced changes in the systemic immune response may predispose to infection, and post-stroke infection is independently associated with worse stroke outcome; 3) new evidence is emerging that lymphocytes and the post-ischemic immune response contribute to brain injury; and 4) transplantation of autologous BMCs and umbilical cord blood modulates the immune response and improve stroke outcome. Importantly, the post-ischemic immune response can be manipulated (at least in experimental animals) to improve outcome.

The opportunity to be highlighted is that modulation of the immune response to improve clinical outcome needs to be translated to the clinical setting. The mechanistic relationships between the brain and peripheral immune organs such as the spleen and bone marrow require further study both in rodent models and patients with stroke. We need to explore the hypothesis that peripheral immune tissues and cells are novel targets to attenuate secondary brain injury after stroke. Additional hypotheses of import include: enhancing the immune response to prevent infection in the recovering individual may lead to more inflammatory brain injury or conversely, that attenuating the post-ischemic inflammatory response may lead to more infection. Or do ancillary therapies (ie. beta-blockers, antibiotics, IV fluids) affect post-stroke outcomes at the level of the brain or the immune system?

Interactions between the immune system and the CNS are just one example; the endocrine system, the cardiovascular system, and other systems must be considered. Ultimately we are interested in preserving or restoring quality of life for individuals with stroke, and too narrow a focus on the CNS has not proved to be useful. Additional examples of the importance of extending neurovascular protection to other supersystems relate to the “off-target” effects of commonly used drugs. For example, it is clear that infection is common following stroke necessitating antibiotics use in this patient population. Some antibiotics/classes of antibiotics are now thought to have neuroprotective properties (e.g. minocycline and ?-lactams), while other classes of antibiotics, such as fluoroquinolones, may be neurotoxic. Hypertensive therapy, statins, and other common agents are known to have a variety of effects that alter the course of post-stroke recovery. As NvPr moves forward, these factors must be considered and point to the concept that brain protection is not enough. A logical extension of this statement is that “personalized therapy” may be the next generation of NvPr in stroke.


1) Improvement of the cellular-animal study-clinical trial interface. Research topics should consider this interface and the integration of emerging data, rather than disparate directions of inquiry. It is suggested that new funding mechanisms for research in this integrated direction must be developed and implemented.

2) The second priority is to include in our current clinical trial repertoire an additional approach. Specifically, the field should design highly exclusive trials in patients that are selected by their linkage to data arising from animal models and other experimental systems.



Co-Chairs: Valina Dawson, Mingming Ning, Roger Simon

Members: Alison Baird, Christopher Beecher, Frank Sharp, An Zhou

NINDS Liaisons: Katrina Gwinn, Erik Runko


‘Omics refers to both unbiased and targeted, high throughput large scale analysis of biologic samples with the most developed technologies being applied to genomics and proteomics. Metabolomics is a later entry to the Omics approach but the technology is rapidly developing. For the purpose of this report we focus on Genomics/transcriptomic, Proteomics and Metabolomics as applied to Stroke research.

In the Genomics field increased availability, reduced base costs, standardization of analysis software and approaches and improved annotation has permitted application of this ‘Omic approach to clinical stroke research leading to a number of important advances made in the field of stroke. Research teams have shown differential gene expression in ischemic stroke subtypes showing that different types of stroke and causes of can be differentiated. These findings reveal potential for new diagnostic tools that could improve the treatment of stroke patients. Additionally studies comparing human and animal models have shown similarities suggesting the relevance of animal models to human disease. The technology applied to Genomics is advancing from microarray approaches to elegant next generation and deep sequencing strategies. To take advantage of these new approaches there will be a need for quantitative biologists to analyze and manage large data sets and the need for access to human material as the power of these technologies is in the number of samples that can be analyzed.

Proteomics was an emerging field when the SPRG was first initiated. Over 10 years it has rapidly evolved and developed as a technology – making important progress in stroke basic and pre-clinical research and now just entering the translational and clinical realm. A step closer to the ultimate phenotypic expression, proteomics has advanced from characterizing protein numbers to more sophisticated protein-protein interactions in order to probe important mechanistic questions. Emerging methodology to surveying the proteome in real time to distinguish confounding biological noise from important primary death and survival pathways are important next steps. In the future, application of proteomics technology holds enormous potential for unmet needs in stroke by understanding the biological significance of post-translational modifications, dissecting subproteomic/organelle interactions, and quantifying the precise stoichiometry of cellular metabolism, in order to investigate cellular mechanistic interactions and accurately monitor and triage therapeutic responses.

Metabolomics is in the true sense of an ‘Omic approach in development and currently at the stage where Genomics and Proteomics were 10-15 years ago. However, considering the acute and chronic metablolic changes that occur during and following stroke, it may be a most revealing technology when applied to stroke studies. There is tremendous promise in significantly understanding how metabolic dysfunction leads to pathologic processes as well as developing new tools for diagnosis of different types of stroke and monitoring the efficacy of therapeutic intervention in the field of metabolomics. The limiting factors at this time are access to the technologies, standardization and the current cost of entry to this ‘Omic field.

In summary the promise of ‘Omic technologies has been realized in some contexts and continues to be developed in others. The new challenges revolve around analysis and handling of very large data sets and access to clinical samples with sufficient power to derive meaningful conclusions. The use of these technologies has moved from basic science studies to translational and clinical research with the promise of development of clinical diagnosis and therapeutic monitoring.


‘Omics refers to the unbiased and targeted, high throughput large scale analysis of biologic samples with the most developed technologies being applied to genomics and proteomics. Metabolomics is a later entry to the Omics approach but the technology is rapidly developing. For the purpose of this report we focus on Genomics/transcriptomic, Proteomics and Metabolomics as applied to Stroke research.

Genomics research advances include:

Greatly improved availability, reduction in overall cost to run samples, standardization of analysis software and approaches and improved annotation has allowed this ‘Omic approach to broader applications. Thus there have been a number of important advances made in the field of stroke. (Sharp FR. Transl Stroke Research. 2010)

Improved reproducibility of gene expression signatures and finding of differential gene expression in ischemic stroke subtypes have promoted ‘proof of principle’ studies in patients indicating that cardioembolic and large vessel atherosclerotic causes of stroke can be differentiated and lacunar strokes can be differentiated from non-lacunar strokes using gene expression profiles in blood; and that these profiles can be used to differentiate different causes of small deep infarcts in humans. This is important because cardiac and large vessel atherosclerosis cause small deep infarcts that look like lacunes on an MRI. (Jickling et al. Ann Neurol. 2011)

‘Proof of Principle’ studies indicate that RNA expression profiles in blood can be used to assess the causes of cryptogenic strokes, which account for over 30% of all strokes. This is important for providing the most appropriate treatment. (Liao et al.BMC Med Genomics. 2009).

It was discovered that gene expression in the blood of patients with high volume white matter hyperintensities differs from those with low volume white matter hyperintensities. The genes are associated with inflammatory and oxidative stress pathways in blood.

It was demonstrated that gene expression profiles in the blood of humans and non-human models correlate with brief durations of cerebral ischemia termed Transient Ischemic Attacks (TIAs). These data validate the experimental usefulness of non-human models to develop approaches to the human condition.

The role of gene silencing in neuroprotection, miRNA as putative therapeutic targets or therapies and epigenetic regulation of stress responses have been applied to stroke research. (Stamova et al. BMC Med Genomics 2009)

Proteomics research advances include:

Proteomics-led discovery of a novel mechanism of neuroprotection against ischemic brain injury, which is a new ‘Omic direction. (Stapels et al., 2010, Sci. Sig., Polycomb group proteins as epigenetic mediators of neuroprotection in ischemic tolerance). The study incorporated both comprehensive quantitative proteomic analyses (LC-MS/MS, exact mass spectometry-based quantitation) of multiple focal cerebral ischemic conditions (allowing identification of tolerant-specific proteomic signature), and a battery of follow up studies (in vivo and in vitro, cell biology, molecular biology and physiology). The findings reconcile the genomics, proteomics and physiology of brain ischemic tolerance and indicate epigenetic mechanisms.

Proteomics reveals the underlying mechanism of a neuroprotective reagent. (Wu et al., 2009, Circulation Ligand-activated peroxisome proliferator-activated receptor-gamma protects against ischemic cerebral infarction and neuronal apoptosis by 14-3-3 epsilon upregulation). The study also incorporated multiple approaches to depict the underlying mechanism of rosiglitazone/PPAR-gamma-mediated protection. The proteomic comparison (2DE MS/MS, spot intensity-based quantitation) of rosiglitazone-treated brains with controls played an instrumental role in revealing the key player in the anti-apoptosis cascade.

Proteomic screening of human plasma samples was used to reveal stroke risk factors. (Prentice et al., 2010, Genome Med., Novel proteins associated with risk for coronary heart disease or stroke among postmenopausal women identified by in-depth plasma proteome profiling). A sizable WHI study on plasma proteomics (LC-MS/MS, isotopic labeling-based quantitation) of post menopause women. Comparisons were made between those who developed cardiovascular disease or stroke (800 cases each) and matched controls. The study signifies the feasibility and dimension of efforts to develop diagnostic tools for stroke, as well as to understand the risk factors.

Understanding of the study of stroke mechanism and cellular signaling. The study of pADPr binding proteins has advanced the field to more sophisticated understanding of pADPr related protein-protein interaction by building an interactome (Gagne/Dawsons et al. Nuc Acid Res 2008). The study of neuronal cell culture nascent proteome established novel bioinformatic comparison of various post ischemic conditions (Zhou/Simon et al. Int J Comp Biol Drug Des 2011).

Proteomics has had an important role in advancing translational stroke research. Proteomic screening in less complex mixture such as endothelial cell culture, in essence building an endothelial vascular “secretome” map, has guided more complex clinical plasma exploration to develop markers in monitoring disease progression. (Ning et al. Stroke. 2010) And vice versa, proteolytic markers of interest found from proteomic clinical blood samples post tPA administration are taken back to animal and cell culture model to better understand thrombolytic mechanisms. (Ning/Lo et al. Tranl Stroke Res. 2010)

Over the last decade, the field has advanced from characterizing protein number to more sophisticated mapping of protein-protein interactions. While gaining a global view, the most elegant approach has been stimulus response - either in cell culture/animal models or bedside therapeutic monitoring of specific therapeutics. Using bench side screening to guide bedside investigation and vice versa, innovative work has been done to push the field forward. The studies above and many others have addressed the previous SPRG important priorities such as “stroke-associated network/pathway analysis and new information on cellular signaling…if made accessible can eventually contribute to new discoveries and the development of new assays…”


Metabolomics is a relatively new field and there are very few instances of metabolomic investigations of stroke. None-the-less, indications of success seem apparent in the targeted analyses of two chemical classes, arginine methylation products, and lyso-phosphatidylcholines. In addition, an NMR-based study suggested the potential to predict patients with cerebral infarctions, but could not definitively name the biomarker compounds involved. Given the success of the targeted analyses, it is unfortunate that no broad-based, unbiased metabolomic studies have been reported. Given the chemical complexity of the brain and its environment, such studies would likely provide insights into the biochemical and physiological events associated with stroke.

There have been a few studies involving metabolism or metabolites but none deploying metabolomics. Of the ‘Omic approaches, this is a rapidly evolving technology with limited availability. There is tremendous power in metabolomic approaches, in particular for the study and treatment of stroke and there is hope this technology will be advanced in the next decade.


There are many opportunities for the application of these technologies to better understand the human condition. The major roadblocks for ‘Omics are accessibility of advanced instrumentation with the accompanying expertise in design and conduct of the studies and the need for advanced analysis and quantitative biology support. There are needs and opportunities to develop searchable repositories for the large data sets that are generated. There is also a strong bias against these approaches in the scientific and clinical community supported by the concern that these studies are ‘fishing’ expeditions and that there is no reasonable way to extract meaningful biologic and clinical information from the data sets. Building on recent successful studies and future work there is hope this opinion will change as the technologies provide valuable information about the total landscape of biology and pathobiology that is not achievable with a single candidate approach and thus lead to new research concepts.


  • The complexity of conducting an entire experimental, clinical and/or translational study with the need for advanced bioinformatics and quantitative biology support, remains an issue.
  • The specificity of gene expression signatures and the cellular sources of gene expression alterations in the peripheral blood for ischemic stroke is still uncertain.
  • There is an unmet opportunity to validate and characterize findings in the successful genomic studies that has yet to occur.
  • There is an unmet opportunity to refine and obtain genomic profiles from peripheral blood that differentiate ischemic stroke from intracerebral hemorrhage and subarachnoid hemorrhage as well as predict individuals at high risk of stroke prior to any vascular events, after TIAs, after a previous stroke, and cryptogenic strokes that could lead to future Point of Care tests for ruling in ischemic stroke and ruling out hemorrhage for tPA treatment, identify patients at risk, or direct other emerging therapies.
  • The transcriptome for all known RNA species and all cell types in peripheral blood following stroke including alternative splicing variants, RNA editing, alternative promoters, and non-coding RNAs has yet to be defined. This RNA biology should then be related to the unresolved and new research opportunities.
  • The identification of genomic profiles for ischemic strokes correlated to gender, age, race, stroke volumes and other clinical features of stroke may lead to markers that can predict or account for improved or worsened outcomes.
  • Combined genomic, proteomic and metabolomic studies of stroke in animals and humans could be designed and used to compare animal models to humans, and to develop animal models that better model the human condition.
  • It is not yet known if there is an association between genomic profiling in brain and periphery and whether current findings are relevant to vascular biology or neurobiology.
  • As genomics have grown, the need for assessment of very large data sets remains a limitation. The field of quantitative biology is emerging and the approaches to analyzing these data sets is evolving. There appears to be a shortfall of qualified individuals who can span the computational and the biological to extract meaningful information.
  • Do our ‘Omics data sets provide insights into mechanisms of injury and/or neuroprotection or are these data more useful as biomarkers of disease?


  • Blood proteomics is a priority. It is the one of the only accessible fluids for stroke patients at the bedside. And this is especially of interest to monitor circulatory cell-cell signaling, gauge therapeutic response to individualize medicine, and follow efficacy in clinical trials. This was noted previously as a goal, but now the advance in proteomic technology are making possible this real translatable “blood test for stroke” potential.
  • Premises of ‘Omic studies are fundamentally different, in important ways, from traditional approaches. Pragmatically, the traditional sense of “reproducibility” may not be feasible for proteomics – the proteome present in a given system will differ not only form one individual to the next, but also from one moment to the next - it will be difficult to streamline costly instrumentation, various cell culture conditions, animal model phenotype, and patient populations. As the previous SPRG alluded to, specific stimulus response in experimental conditions or therapeutic efficacy would be important. Perhaps unique biologic signatures need to be described in relation to both space and time – that is, the precise interaction may need to be expressed as a function of many “phenotypic dimensions” over time. If ‘Omic technology can afford us the sensitivity of studying a multitude of markers at the same time, can we leverage this power to study a smaller number of well phenotyped individuals over time, across specific pathways, interactions, or disease states, utilizing each individual as their own control, thereby reducing the caveat of complexity/dimensionality/confounders?
  • There are similar opportunities and unmet needs as genomics in regards to the association between proteomic profiling in brain and periphery, the need for assessment of very large data sets and the development of computational biologists with expertise in stroke to mine the data.
  • The methodology and technology providing for quantitation and absolute quantitation is a maturing but evolving area. Although there publications are on the rise, quantitative MS studies still consist only a small portion of published studies, and even less for absolute quantitation to produce stoichiometric information. This is due in part to the standardization of these approaches and the access to expensive, high tech equipment. The time required to run each quantitative sample limits sample size or necessitates multiple pieces of equipment. However, quantitative and absolute quantitative information is vital in bring forth a precise description of cellular mechanisms.
  • High throughput, targeted/chip-based technology would greatly benefit the field, as there is currently a lack of chips that are both robust and comprehensive
  • There is a difference between hypothesis driven vs. hypothesis generating approaches, with more weight given to hypothesis driven studies, although both approaches have power. There is a lack of tailored experimental design, follow up studies, and a lack of yielding ‘Omics-wide hypotheses. What would be a testable hypothesis in ‘Omic terms? Perhaps future studies should incorporate complimentary “discovery/unbiased” and “targeted/hypothesis driven” modes?
  • Availability of samples from living human subjects (not post mortem specimens) is limited.
  • An emerging opportunity in its infancy is MetaboProteomics (or ProteoMetabolomics) in which tissue specific, time dependent experiments allowing kinetic analysis to capture and characterize critical windows of protein-metabolite interaction in a living organism.
  • Kinetics of proteomic changes during stroke injury development or the exhibition of injury resistance, focusing on the nascent proteomes to capture the initiation of must-have or must-stop events.
  • Emerging proteomic functional investigations from basic, translational, and clinical study of post-translational modifications (PTM’s), subproteomic analysis, and quantitative proteomics are exciting and promising areas of exploration for the next decade.


  • Considering the acute and chronic biologic changes that occur during and following stroke there is tremendous promise in significantly understanding the pathologic process as well as developing new tools for diagnosis in the field of metabolomics. The limiting factors are access to the technologies and the cost of entry to this ‘Omic field.

Emerging Opportunities and Unmet Needs in ‘Omics in General:

‘Omics analyses – The complexity and the poorly understood etiology of many neurological diseases, make it important to search broadly for clues to their developmental characteristics and mechanisms. NINDS could best accomplish this by using the unbiased genomics, transcriptomics, proteomics, and metabolomics approaches. Traditionally these approaches have been difficult to fund as they are not hypothesis-driven, but their potential to yield novel hypotheses, and define clinically relevant biomarkers make them quite attractive. Because the brain is such a chemically complex organ, metabolomics should hold particular value for short-term diagnostic potential, while genomics holds the most potential for long-term mechanistic understanding. Transcriptomics and proteomics will critically fill in the mid-terms of disease progression. The analysis of multiple aliquots of the same samples across multiple platforms and at multiple ‘Omics levels, and the development of bioinformatics capabilities which can integrate all levels of ‘Omics analysis is an important direction for research. Given the current funding for genomic, transcriptomic and proteomic studies, an emphasis on metabolomics studies may yield significant benefit per dollar spent. These high throughput studies can process statistically appropriate sample sets in relatively short periods of time, and yield chemically definitive quantitative data.

‘Omics datasets are large and complex – The potential value of ‘Omics datasets would be tremendously increased if any and all ‘Omics projects funded by NINDS were required to make their data publically available so that the data mining of these large complex datasets could be accomplished from the broadest possible perspective. In order to do this not only should the NIH policy of making data public upon publication be strongly enforced, but possibly NINDS should consider providing a mechanism for the storage, and distribution of all such generated data. Such a resource would be of significant value to the entire research community, within NINDS and well beyond. It is critical to recognize that the analysis of data is a separate function from the generation and analysis and banking of data. In order to assure the quality of both the analysis of the data and maximize the data-mining capacity, split funding should be considered, whereby platform analysis experts generate and deposit data and multiple separate data-mining experts get quick access to the data.

Biospecimen availability – An additional potential for NINDS would be the creation of collections of well-authenticated biological materials that could be administered by NINDS committees, or committees overseen by NINDS. These materials could be acquired without much additional cost if the regularly funded NINDS trials and research programs were encouraged to contribute excess materials after their own analyses were completed. Since the cost of sample procurement is often a significant component of any project this efficiency would not only reduce the cost of additional sample collection for follow-on projects, but also provide materials for unrelated projects, and provide a mechanism for multpile analyses of the same sample. The material in such a biorepository could be made available through a grant-driven process.

Studies in humans need to integrate immunology with genomics, proteomics and metabolomics – including studies of specific cell types and immune cell function in blood of patients with ischemic and hemorrhagic stroke. This could potentially be correlated with changes in brain for those patients who die. These data will relate to variable stroke outcomes as a function of age, gender, race, prior medical conditions including diabetes, hypertension and others. Genomics, proteomics, metabolomic and immunology approaches could be used to predict high risk of stroke in asymptomatic patients, TIA patients, and patients with recent stroke. Such knowledge could be used to prevent strokes – rather than treat them.


1) Facilitate access and cross-talk of genomic/transcriptomic, proteomic, metabolomic and bioinformatics technology and develop methods for interaction and standardization of design, data analysis, and interpretation related to stroke.Promoting functional studies harnessing advances in multiple ‘Omics strategies are exciting and promising areas of exploration for the next decade.

2) Create a platform for defining stroke at a molecular and mechanistic level. This should include the creation of methods and protocols for the collection of biological samples for ‘Omic studies with access to biospecimens linked to current and future National Institutes of Health (NIH) trials. With basic research as a template, identify molecular markers or profiles of stroke and stroke risk in patients.

3) Recruit geneticists and molecular biologists into collaborations with stroke researchers (both clinical and basic science), with the objective of achieving near-term progress in capitalizing on current technologies. In parallel, enhance the recruitment of young and mid-career scientists into existing training programs.



Co-Chairs: S. Claiborne Johnston, Karen Furie, Bruce Ovbiagele

Members: Marc Chimowitz, Mitchell Elkind, Heather Fullerton, Scott Kasner, Walter Kernan, Laura Sauerbeck, Barbara Vickrey

NINDS Liaisons: Scott Janis, Salina Waddy


In the last 5 years, measured progress has been made in research on prevention of first and recurrent stroke. Among other advances, several major clinical trials have helped to clarify appropriate antiplatelet use, the role of stenting in carotid and intracranial atherosclerosis, and brought new, safer and more effective anticoagulants to the market for stroke prevention in patients with atrial fibrillation. Several of the prior recommendations of this Stroke PRG subgroup were implemented or will come on line soon, including NEURO-NEXT and a U54 for stroke prevention research centers. The committee identified particularly exciting opportunities in five areas: population-based primary stroke prevention, perinatal stroke prevention, personalized stroke prevention, plaque hemorrhage and rupture, and technology as a tool to enhance patient and physician adherence. In addition, three priorities were recommended:

1) Implementation research

Improve implementation of existing, proven stroke prevention interventions by: (1) identifying barriers to such implementation by assessing the individual, healthcare providers, and the healthcare system, (2) studying methods of overcoming these barriers, including through technologies, systems-based interventions, and incentives, and (3) supporting the development of research evaluating the effectiveness of innovative initial and recurrent stroke prevention interventions, especially in underserved populations and minority racial/ethnic groups.

2) Risk factor assessment and risk prediction

Track the prevalence and impact of stroke risk factors in the US, and develop and examine the effectiveness of quantitative risk factor assessment tools that can identify stroke-prone individuals who need aggressive risk factor management and recurrent stroke prevention intervention, with particular emphasis on underserved populations and minority racial/ethnic groups.

3) Innovative prevention strategies

Support research designed to identify and evaluate innovative stroke prevention treatments and strategies, with particular attention to dietary, lifestyle, and population-directed interventions that are inexpensive and cost-effective. The committee felt that NINDS had been very responsive to its prior recommendations.


  • Statin therapy with intensive lipid-lowering effects reduces the risk of stroke and cardiovascular events among patients with ischemic stroke or TIA. 1
  • Aggressive blood sugar control does not reduce the risk of stroke (pre-specified secondary outcome) in diabetics 2
  • Combination extended release dipyridamole plus aspirin may be superior to aspirin (ARR 1%/yr) but is not non-inferior to clopidogrel.3-4
  • Rosuvastatin significantly reduced the incidence of stroke in healthy subjects with elevated CRP, although whether this reflected an effect on inflammation remains uncertain. 5
  • High-dose extended-release niacin given in addition to statin therapy in patients with a history of cardiovascular disease, high triglycerides (TG), and low levels of HDL cholesterol offers no additional benefits, and may be associated with an increase in ischemic stroke. 6
  • Cilostazol is non-inferior (and might be superior) to aspirin for prevention of stroke after an ischemic stroke, and is associated with fewer hemorrhagic events.7
  • Ten risk factors are associated with 90% of the risk of stroke. Interventions that specifically reduce blood pressure and smoking, and promote physical activity and a healthy diet, could substantially reduce the burden of stroke 8
  • Combination antiplatelet therapy with clopidogrel and aspirin, compared to aspirin alone, reduces the risk of recurrent stroke and major cardiovascular events, but increases the risk of hemorrhagic complications in patients with AF who cannot take warfarin. 9
  • Warfarin is superior to combination antiplatelet therapy in patients with atrial fibrillation.10
  • Dabigatran, apixaban, and rivaroxaban are non-inferior or superior to warfarin for the prevention of stroke and systemic thromboembolism in patients with AF.11-12
  • Prolonged cardiac outpatient rhythm monitoring detects previously undiagnosed atrial fibrillation in a substantial proportion of patients with unexplained stroke.13
  • Aggressive medical management alone is associated with a lower short–term risk of stroke and death than angioplasty combined with stenting with the Wingspan stent in patients with recent symptoms and high grade intracranial arterial stenosis.14
  • CEA or CAS are reasonable choices for symptomatic patients at average or low risk of complications associated with endovascular intervention when the diameter of the lumen of the internal carotid artery is reduced by more than 70% by noninvasive imaging or more than 50% by catheter angiography.15
  • In carotid occlusion, carotid artery bypass surgery is no better than medical therapy for the secondary prevention of ischemic stroke. 16
  • One report indicates that PFO closure is not superior to antithrombotic therapy in the prevention of recurrent stroke/TIA in young patients with cryptogenic stroke or TIA. 17
  • In children with sickle cell disease, discontinuation of prophylactic blood transfusions (used to prevent stroke) increases the risk of silent brain infarction. 22
  • The NHLBI discontinued the Stroke With Transfusions Changing to Hydroxyurea (SWiTCH) trial because of evidence that the new treatment, hydroxyurea, was unlikely to prove better than blood transfusions for prevention of recurrent stroke in children with sickle cell disease (http://public.nhlbi.nih.gov/newsroom/home/GetPressRelease.aspx?id=2709).
  • Several initiatives have been supported that support the advancement of research in stroke prevention.
    • NEXT: Neuroscience Clinical Trials Network Workshop and RFA
    • Preventing Stroke and Heart Disease: Connecting Traditional and Emerging Approaches to Change Behavior Workshop
    • Workshop: The CNS and Glycemic Control
    • Announcement of an upcoming RFA for a U54 in stroke prevention, focusing on disparities.


Key areas of opportunity

1. Population-based Primary Stroke Prevention

NINDS has funded several large epidemiological studies geared at promoting an understanding of predisposing risk factors for stroke. The knowledge garnered from these large-scale studies can now be used to identify persons at high risk for first time stroke. Individuals could be selected using multivariable global risk profiles for stroke, such as with the Framingham stroke risk score, and then treated using aggressive evidence-based multimodal therapeutic approaches (pharmacological and non-pharmacological) with a goal of mitigating the incidence of future stroke as well as reducing overall vascular risk. Major advances in stroke prevention may require community-wide interventions that promote physical activity, salt restriction, calorie restriction, and consumption of foods known to be associated with lower risk for vascular disease.

2. Perinatal Stroke Prevention

Efforts to define the relative contribution of maternal and placental problems as well as fetal and neonatal disorders to the occurrence of perinatal arterial ischemic stroke and sinovenous thrombosis are minimal. Large natural history prospective studies of perinatal stroke risk factors including hematologic and autoimmune disorders, as well as other factors affecting the maternal-fetal environment could be designed with the goal of improving understanding and fostering potential therapeutic avenues (such as the use of antithrombotic agents in prevention) for modifying perinatal stroke risk.

3. Personalized Stroke Prevention

Prevailing clinical practice for selecting stroke prevention treatments is generally based on the presence or absence of comorbidities, issues of cost, side effects and other non-genetic characteristics of individual patients. However, recent successes in stroke genomics including the discovery of potential risk factor genes have brought us closer to developing individualized approached to reducing stroke risk. Pharmacogenomics studying the interactions between secondary prevention drugs and gene variants could permit the selection of high-responder versus low-responder populations to various drug regimens in the future. Further studies to determine optimal primary and secondary prevention based on genetic polymorphisms are needed.

4. Plaque Rupture and Hemorrhage

Recent studies in coronary, carotid, and large intracranial arteries suggest that plaque rupture is an important precipitant of acute thrombosis and may cause ischemic stroke. Studies have characterized several inflammatory and lipid-related pathways that contribute to the risk of plaque rupture. Although risk of plaque rupture and hemorrhage appears to be related to the overall burden of atherosclerosis, higher risk morphologies of specific plaques have been identified. Strategies to more directly image plaque and target plaque stabilization should be identified in animal models and brought into human trials. Strategies identified in extracranial vessels should be brought to bear in intracranial vessels.

5. Technology as a Tool to Enhance Patient and Physician Adherence

With the rise of online social networks, mobile technologies, monitoring devices, and electronic health records, there is tremendous opportunity to capitalize on the instant availability of information to improve adherence. Without specific investment in research to assess efficacy, such tools may not yield optimal benefits.

Specific unresolved issues

  • Optimal medical management of established vascular risk factors, particularly elevated blood pressure, remains inadequate. Successful treatment of stroke risk factors could reduce risk by as much as 80%. In order to prevent initial and recurrent stroke, resources need to be devoted to developing and implementing strategies for risk factor intervention.
  • There are sub-populations of Americans at increased risk for stroke. Race, ethnicity, geography, and socioeconomic status contribute to this risk. Increased screening and aggressive risk factor management may help reduce morbidity and mortality in these groups, but ideally targeted and tailored interventions have not been defined. 18
  • Lifestyle factors contribute to the development of vascular risk factors and the subsequent risk of stroke. Their role in secondary stroke prevention, particularly in an elderly population, remains unclear. Research into potential risk/benefits for lifestyle interventions would endorse and clarify recommendations.
  • For patients with atrial fibrillation, several treatment options now exist. Improved methods for detection of atrial fibrillation have been developed. Recent studies show that atrial fibrillation is commonly detected during prolonged monitoring after stroke.21
  • Further research is needed to determine which patients benefit from prolonged cardiac monitoring and how long monitoring should be performed to optimize detection.
  • The role of new approaches to identification of atrial fibrillation in patients with unexplained stroke (outpatient telemetry, implanted devices) remains uncertain.
  • Optimal strategies for managing atrial-fibrillation detected by prolonged monitoring need to be developed.
  • Beyond the management of hypertension, there are no proven preventive therapies for intracerebral hemorrhage. A treatment for cerebral amyloid angiopathy or intervention to reduce anticoagulant-related hemorrhage, or improved methods to detect those at risk, could reduce the incidence of this devastating disease.
  • Optimal management of arterial dissection, an important cause of stroke in the young, remains unclear.
  • Prediction rules for stroke after head or neck trauma, an important cause of arterial dissection, are needed to develop cost-effective strategies for primary stroke prevention in this setting.
  • Management of patients with stroke due to PFO remains unclear. Ongoing and future studies must identify those patients at highest risk of recurrence who may be most likely to benefit from PFO intervention.
  • The appropriate timing of secondary prevention after acute stroke remains uncertain. For example, data on safety of acute lowering of blood pressure in patients with stroke are lacking. The importance of blood pressure variability has been identified.19-20
  • Brain and cerebrovascular imaging has advanced, providing information about white matter hyperintensity, silent brain infarction, and early vascular pathology, and their functional consequences. However, the implications of abnormal findings for therapy have not been determined. Demonstrating the utility of a specific diagnostic imaging battery would improve and standardize the evaluation of patients at risk for TIA/stroke. Ideally, imaging factors could be integrated into conventional stroke risk models.
  • The management of small unruptured aneurysms and AVMs remains controversial.
  • Prevention of stroke in children remains understudied. The risk of recurrence is very high in this population but even basic questions about antiplatelet therapy have not been addressed.
  • The rate of stroke recurrence in clinical trials has declined dramatically over the last decade, possibly due to a background of more aggressive prevention interventions. Although this is a positive trend, a consequence is that clinical trials in stroke prevention require large sample sizes and multiple high volume centers. The lengthy start up time and recruitment challenges have resulted in extended timelines and delays in the emergence of new treatments.
  • Surgical and endovascular treatments of both asymptomatic and symptomatic extracranial carotid stenosis, which were established as standard of care treatments before modern day medical management was established, have yet to be compared with aggressive medical management alone
  • Persons with coronary artery disease, peripheral arterial disease and chronic kidney disease are at substantially higher risk of primary stroke than the general population. Identifying ways to screen and optimize primary stroke preventive therapy in these high-risk populations is needed.
  • Inclusion of stroke among outcomes in risk prediction models
  • The relationship of cerebrovascular disease to dementia, including Alzheimer disease, remains uncertain, both clinically and mechanistically. The benefits on cognitive function of cerebrovascular prevention strategies, including blood pressure control and lipid reduction, remain unclear.
  • Primary stroke prevention should probably begin in childhood with efforts to prevent obesity, promote an active lifestyle and moderate salt intake. Such efforts may be directed toward changes in policies that impact risk factors.
  • Stroke is a major complication of surgery, particularly cardiothoracic and vascular surgery. However, no validated method exists to estimate an individual's risk of perioperative stroke. Moreover, interventions to prevent peri-operative strokes have been understudied.


1) Implementation research: Improve implementation of existing, proven stroke prevention interventions by: (1) identifying barriers to such implementation by assessing the individual, healthcare providers, and the healthcare system, (2) studying methods of overcoming these barriers, including through technologies, systems-based interventions, and incentives, and (3) supporting the development of research evaluating the effectiveness of innovative initial and recurrent stroke prevention interventions, especially in underserved populations and minority racial/ethnic groups.

Barriers to implementation of proven prevention therapies remain a major public health problem that limits the impact of scientific advancements from clinical trials. For example, anticoagulation for patients with atrial fibrillation, carotid revascularization for symptomatic stenosis, and treatment of hypertension are all interventions proven in NIH sponsored randomized trials and implementation of all these therapies has been slow and incomplete. Reasons for this have not been well studied and interventions to improve implementation have rarely been studied systematically.

Embracing research on implementation as part of the mission of the NINDS, inherent in its goal to promote health, is essential to make progress in this area, progress that has been particularly slow in the last decade. Other institutions are not well positioned to fund this research because their resources are limited, and the Patient-Centered Outcomes Research Institute (PCORI) may not identify stroke prevention as an initial priority, in spite of its high burden.

Actions Needed

  • Target funding toward implementation research through an RFA/RFP. Although needs in stroke prevention are obvious, such an RFA/RFP could more broadly encompass neurological therapeutics, since implementation in other neurological diseases is also a problem. Research on patient adherence to medications for neurological indications, physician adherence to guideline recommendations, and system-level approaches to improving implementation would all be acceptable areas of research.
  • Convene a summit of potential partners (eg, CMS, PCORI, other NIH institutes and centers, disease-based organizations, health insurers) and academics to identify research synergies in the area of implementation of prevention strategies in adults and children. The goal of the summit would be to produce a plan to accelerate collaborative research in this area, explicitly discussing opportunities for cost sharing and for working together to define appropriate priorities. Support for the development of a national stroke registry should also be considered as part of the agenda; given the expense of this endeavor, rapid evolution in the availability of electronic records, and experience of other institutions in registries (eg, CDC, CMS, AHA, large healthcare plans), collaboration in promoting a registry is crucial.

2) Risk factor assessment and risk prediction: Track the prevalence and impact of stroke risk factors in the US, and develop and examine the effectiveness of quantitative risk factor assessment tools that can identify stroke-prone individuals who need aggressive risk factor management and recurrent stroke prevention intervention, with particular emphasis on underserved populations and minority racial/ethnic groups.

There is no standard for stratifying risk of first or recurrent stroke, although some tools have been proposed. Furthermore, instruments specific for stroke subtypes (eg, ischemic stroke or intracerebral hemorrhage) or for use in distinct populations (eg, children or those with cerebral microbleeds) are lacking. Such instruments, once validated, are likely to be important tools for clinical care, and also for future randomized trials and prospective cohort studies.

Actions Needed

  • Convene a working group to assess the value of collaboratively supporting a large nationally representative cohort to track prevalence of stroke risk factors, incidence, and outcomes taking advantage of electronic health records to streamline data collection and increase efficiency. Such a database could substantially reduce the costs of studies on stroke prevention strategies and provide early warning signs of trends. The working group should include representation from the CDC, CMS, PCORI, private insurers, and other institutes that could leverage the database to address other risk factors and diseases.
  • Support initiatives that increase access to existing datasets from completed studies through dissemination of de-identified datasets, standardization of variable terms, and through support of retrieval and analysis (as does the NHLBI: see http://grants1.nih.gov/grants/guide/pa-files/PA-06-239.html). Secondary analyses of existing databases remain an important and efficient source of preliminary evidence that is essential in identifying risk factors for diseases and in deriving risk factor assessment tools. Limiting access to such datasets reduces the potential for new discovery. De-identification is always feasible and is adequate to address concerns about confidentiality, privacy, and HIPAA regulations. Policies that require publication of datasets derived from NINDS funding in a timely manner should be considered.
  • Convene a working group to explore development and validation of standardized stroke risk prediction rules (eg, Framingham risk score) for various settings (eg, primary prevention, TIA, secondary stroke prevention, perioperative stroke risk assessment) and populations (adult and pediatric) to be used in future prevention trials and, possibly for health screening and clinical care. Similar to consensus conferences for developing standard disease definitions or research terms, such a working group may identify preferred tools already developed and may help to set priorities for the research required to establish and validate such tools.
  • Continue to support efforts to collect and provide open access to biospecimens as part of clinical trials and observational studies, and consider more efficient mechanisms for reviewing proposals for ancillary studies to clinical trials.

3) Innovative prevention strategies: Support research designed to identify and evaluate innovative stroke prevention treatments and strategies, with particular attention to dietary, lifestyle, and population-directed interventions that are inexpensive and cost-effective.

Although there has been notable progress in the development of stroke prevention treatments, the overall age-adjusted incidence of stroke has decreased little during the last decade and the burden of stroke has increased as the population continues to age. Progress by industry and the NIH has been modest and new models for collaboration could accelerate development. Interventions directed to populations (such as policies that encourage exercise or reduce salt intake) may be particularly effective. With rates of diabetes and obesity increasing dramatically, strategies that impact their prevalence and burden are also required. Clearly, more effective, well-tolerated, and cost-effective interventions must be sought.

Actions Needed

  • Increase interaction and collaboration between pharmaceutical companies and NINDS to optimize development of new therapeutics and share costs and benefits. First, a conference devoted to development of stroke prevention therapeutics should be convened, bringing together representatives from academia, the NINDS, the FDA, and interested pharmaceutical, biotech, and device companies. Such a conference could work toward establishing standards for pre-clinical testing, set standard definitions and analytical techniques for trials, encourage study and acceptance of appropriately validated surrogate outcomes, identify barriers and potential solutions to common problems such as subject recruitment, and encourage richer collaborations. The STAIR conference has done this for acute stroke and it has been very influential in the development of new therapies in this area.

    In addition, the NINDS should explore more fluid approaches to negotiating for cost-sharing and access to potential targets with specific companies. Before a decision to fund a trial is made, representatives from the NINDS should meet with any companies that hold patents for studied agents. Ultimate NIH funding of such a trial might be contingent on acceptable negotiation for cost sharing based on an assessment of the potential risks and gains for all parties. Once fully funded by the NIH, a company is no longer motivated to support a study. Such a strategy is only feasible if the decision to fund a trial is based first on a more stringent review of the proposal in light of its clinical and scientific impact in comparison to other potential funding priorities, as proposed above.
  • Develop an institute-initiated research agenda designed to address the most promising interventions in terms of public health benefit and societal net cost, as a supplement to traditional investigator-initiated mechanisms. Based on organized input from medical practitioners and academics with broad representation, the NINDS should identify the most relevant prevention therapies for evaluation in clinical trials. Using the mechanism described above, RFP/RFAs would then be issued to request initial proposals for such studies.
  • Complete the funding of a SPOTRIAS-type program (as a recent announcement of a new U54 mechanism suggests) focused on primordial, primary, and secondary prevention and underserved populations. Institutions focused on stroke prevention are not necessarily the same ones focused on acute stroke interventions. Reducing the barriers to prevention studies, in particular trials, could accelerate development.
  • Consider supporting research to identify markers of subclinical vascular disease—such as white matter disease, silent infarction, carotid intima-media thickness, high resolution MRI of vascular plaque, impaired cognition and neuropsychological testing, and serum biomarkers—and validate them for use as surrogate outcomes for future prevention trials. Surrogate outcomes have greatly accelerated development of therapies in many disease areas, such as for antihypertensives, statins, cancer therapeutics, and even multiple sclerosis. Additional research is required to validate such surrogates for use in clinical trials of stroke prevention therapies. Including these measures in clinical trials could provide the crucial link and validation.
  • More specifically, evaluate the potential for allowing brain infarction as a surrogate for ischemic stroke in prevention trials. Such a surrogate could rapidly increase progress in establishing preventive therapies by increasing power and reducing necessary sample size. In particular, if two trials are required for FDA approval, the first might evaluate the impact of therapy on new infarction with a much smaller necessary sample size, confirmed later by a larger trial with a pure clinical outcome. Such a development pathway would reduce cost and risk. Ask the Foundation for NIH’s Biomarker Consortium to convene a group of academics, industry leaders, and the FDA to consider the steps necessary to establish brain infarction is an acceptable surrogate. Much of the necessary research has already been done and barriers may be more imagined than real, particularly given new definitions of ischemic stroke.
  • Create an RFP directed at funding prevention research in currently understudied areas, such as pediatric stroke, intracerebral hemorrhage, and subarachnoid hemorrhage. Industry is not motivated to provide prevention therapies for rarer conditions. This is clearly part of the mission of NINDS, but such proposals may not fare as well in peer review because of the perceptions that recruitment is a greater problem for more rare diseases and that these questions are not as important as those directed toward more common diseases. An RFP could help provide more balance in the research portfolio.
  • Target funding through an RFA/RFP toward the development and assessment of behavioral interventions and incentives after stroke. Behavioral interventions might target risk factor mitigation, particularly diet, physical activity, and weight optimization. While surgical and pharmacological approaches to secondary prevention remain important, behavioral approaches have great potential but have received relatively little attention from the research community.


1. Amarenco, P., et al. Results of the Stroke Prevention by Aggressive Reduction in Cholesterol Levels (SPARCL) trial by stroke subtypes. Stroke 40, 1405-1409 (2009).

2. Riddle, M.C. Effects of intensive glucose lowering in the management of patients with type 2 diabetes mellitus in the Action to Control Cardiovascular Risk in Diabetes (ACCORD) trial. Circulation 122, 844-846 (2010).

3. Halkes, P.H., van Gijn, J., Kappelle, L.J., Koudstaal, P.J. & Algra, A. Aspirin plus dipyridamole versus aspirin alone after cerebral ischaemia of arterial origin (ESPRIT): randomised controlled trial. Lancet 367, 1665-1673 (2006).

4. Diener, H.C., et al. Effects of aspirin plus extended-release dipyridamole versus clopidogrel and telmisartan on disability and cognitive function after recurrent stroke in patients with ischaemic stroke in the Prevention Regimen for Effectively Avoiding Second Strokes (PRoFESS) trial: a double-blind, active and placebo-controlled study. Lancet Neurol 7, 875-884 (2008).

5. Ridker, P.M., et al. C-reactive protein levels and outcomes after statin therapy. N Engl J Med 352, 20-28 (2005).

6. FDA. http://www.fda.gov/Drugs/DrugSafety/PostmarketDrugSafetyInformationforPatientsandProviders/ucm256841.htm. (2011).

7. Shinohara, Y., et al. Cilostazol for prevention of secondary stroke (CSPS 2): an aspirin-controlled, double-blind, randomised non-inferiority trial. Lancet Neurol 9, 959-968 (2010).

8. Tu, J.V. Reducing the global burden of stroke: INTERSTROKE. Lancet 376, 74-75 (2010).

9. Connolly, S.J., et al. Effect of clopidogrel added to aspirin in patients with atrial fibrillation. N Engl J Med 360, 2066-2078 (2009).

10. Connolly, S., et al. Clopidogrel plus aspirin versus oral anticoagulation for atrial fibrillation in the Atrial fibrillation Clopidogrel Trial with Irbesartan for prevention of Vascular Events (ACTIVE W): a randomised controlled trial. Lancet 367, 1903-1912 (2006).

11. Connolly, S.J., et al. Dabigatran versus warfarin in patients with atrial fibrillation. N Engl J Med 361, 1139-1151 (2009).

12. Patel, M.R., et al. Rivaroxaban versus Warfarin in Nonvalvular Atrial Fibrillation. N Engl J Med (2011).

13. Tayal, A.H., et al. Atrial fibrillation detected by mobile cardiac outpatient telemetry in cryptogenic TIA or stroke. Neurology 71, 1696-1701 (2008).

14. USNLM. http://www.nlm.nih.gov/databases/alerts/intracranial_arterial_stenosis.html. (2011).

15. Brott, T.G., et al. Stenting versus endarterectomy for treatment of carotid-artery stenosis. N Engl J Med 363, 11-23 (2010).

16. Grubb, R.L., Jr., Powers, W.J., Derdeyn, C.P., Adams, H.P., Jr. & Clarke, W.R. The Carotid Occlusion Surgery Study. Neurosurg Focus 14, e9 (2003).

17. Furlan, A.J., et al. Study design of the CLOSURE I Trial: a prospective, multicenter, randomized, controlled trial to evaluate the safety and efficacy of the STARFlex septal closure system versus best medical therapy in patients with stroke or transient ischemic attack due to presumed paradoxical embolism through a patent foramen ovale. Stroke 41, 2872-2883 (2010).

18. Cruz-Flores, S., et al. Racial-ethnic disparities in stroke care: the American experience: a statement for healthcare professionals from the American Heart Association/American Stroke Association. Stroke 42, 2091-2116 (2011).

19. Rothwell, P.M. Limitations of the usual blood-pressure hypothesis and importance of variability, instability, and episodic hypertension. Lancet 375, 938-948 (2010).

20. Rothwell, P.M., et al. Prognostic significance of visit-to-visit variability, maximum systolic blood pressure, and episodic hypertension. Lancet 375, 895-905 (2010).

21. Gaillard, N., et al. Detection of paroxysmal atrial fibrillation with transtelephonic EKG in TIA or stroke patients. Neurology 74, 1666-1670 (2010).

22. Charles H. Pegelow, MD; Winfred Wang, MD; Suzanne Granger, MS; Lewis L. Hsu, MD; Elliott Vichinsky, MD; Franklin G. Moser, MD; Jacqueline Bello, MD; Robert A. Zimmerman, MD; Robert J. Adams, MD; Donald Brambilla, PhD; for the STOP Trial. Silent Infarcts in Children With Sickle Cell Anemia and Abnormal Cerebral Artery Velocity Arch Neurol. 2001;58:2017-2021.



Co-Chairs: Steve Cramer, Pam Duncan, Anna Barrett

Members: John Chae, Leonardo Cohen, Bruce Crosson, Leigh Hochberg, Rebecca Ichord, Albert Lo, Randy Nudo, Randall Robey, R. Jarrett Rushmore, Sean Savitz, and Robert Teasell with assistance from Norine Foley MSc

NINDS Liaisons: Francesca Bosetti, Daofen Chen, Scott Janis


Recovery of function after stroke arises on the backbone of biological events that are therapeutic targets. Thus preclinical studies and early human clinical trials have provided evidence for a large number of potential therapies to improve patient outcomes after stroke, including cell-based therapies, drugs new and old, a host of devices, cognitive approaches, and activity-based approaches.

Recovery of function after stroke also involves many factors that extend beyond brain biology, factors that distinguish us as humans, such as social role, family role, vocational competence, participation in sports and leisure, affective state, and more.

Most strokes affect many neural systems in parallel. Thus the most robust understanding of the stroke impact and restorative therapy effects is obtained by considering the individual components of a stroke such as effects on attention, balance, coordination, executive function, speech and language, memory, mobility, motor function, pain, mood, sexual function, somatosensory function, spatial cognition, swallowing, vision, and more.

Understanding stroke recovery requires modality-specific endpoints. Many preventative, reperfusion, and neuroprotective therapies treat the brain as a single target. In contrast, therapies that target stroke recovery must treat the brain as a collection of injured neural systems, some of which are potential therapeutic targets and some of which are not. Global outcome measures, while of sure value in stroke, may lack the granularity to resolve some of these important distinctions.

During the weeks following a single stroke, patients often are moved from one setting to the next, receiving care from numerous providers along the way. Maximum attention to the continuity of care can improve behavioral outcomes and also reduce medical complications.

Substantial data exist as to how to maximize stroke rehabilitation effects, but this information is often incompletely applied in practice. This could be improved with healthcare policy research and reform.

Stroke in children differs from stroke in adults in numerous ways from pathogenesis to developmental state at time of injury. A number of major questions remain in stroke recovery and rehabilitation among children, who have the largest number of lifetime years affected by stroke.

There have been few human drug or cell-based trials targeting stroke recovery despite many promising preclinical study results. Calls are needed for specific NIH-directed mechanisms targeting the successful translation of novel drug and biological therapies from animals to stroke patients. More resources should be applied to implementing and directly supporting the recommendations of the SPRG if the enormous potential of recovery and rehabilitation interventions is to be realized.


Introduction: The Stroke PRG 2011 Recovery and Rehabilitation Group examined past and future research issues in several categories. These span from basic science and mechanistic research to a range of new therapies. The organization of stroke care is as important as any of the findings that will be implemented within these systems. The motor exam dominates the NIHSS but most of the brain is not part of the motor system, and so language and neglect were given special attention, as were pediatric issues. Overall, research since 2007 has generated enormous excitement, as theories of neuroplasticity and findings from preclinical studies have been examined in human studies, and many results to date are favorable. The remarkable progress in the field of stroke recovery is good news: it remains true that many patients with stroke have enduring disability, as only a minority of patients with ischemic stroke receive an acute reperfusion therapy and many so treated have enduring deficits, and there are no approved therapies for acute intracerebral hemorrhage. Therapies based on brain repair have the potential to help a large fraction of patients with stroke. The main advances since 2007 in this regard are summarized below, by category.

Translating preclinical findings to human studies

  • Substantial progress in mechanisms underlying endogenous neurogenesis and cell migration
  • Demonstration of positive behavioral effects of exogenous cellular and growth factors and their interactions with behavioral training
  • Evidence from pre-clinical models that compensatory learning with the less-impaired limb worsens the function of the more paretic limb as well as underlying mechanisms
  • Substantial advances in response of the aged brain to stroke in pre-clinical models

Drugs and biological therapies to promote motor recovery and rehabilitation

  • Fluoxetine with physiotherapy significantly improved motor recovery[1].
  • Agents that directly stimulate neurogenesis and angiogenesis in rodents have been brought forward to initial human clinical studies [2], [3].
  • Robot-assisted therapy did not differ from usual care or from an intensive comparison therapy in chronic stroke, though some secondary analyses suggested superior long-term benefits from the robotic therapy.
  • Cellular products with therapeutic activity in animal models of stroke are being actively developed for clinical testing.
  • Small studies have reported safety of intravenous administration of autologous bone marrow derived mononuclear cells or mesenchymal stem cells (MSCs) in stroke patients [4, 5] [6]. The application of autologous bone marrow derived cells to stimulate brain repair is therefore ripe for further development.

Devices to promote recovery and rehabilitation

  • Rehabilitation robotics: Demonstration of effectiveness; cost-benefit needs continued study.
  • Neuromodulation technologies: Revealing the potential for rapid restoration of lost function.
  • Intracortically-based Brain-Computer Interfaces: Promising early demonstrations

Activity-based therapies (ABT) to promote recovery and rehabilitation

  • Since 2007, larger, randomized, ABT clinical trials have been completed for upper extremity paresis and gait impairment (4,5,6).
  • These clinical trials have included acute to chronic stages of stroke and those with severe impairment, and have examined robot, treadmill and constraint-induced movement interventions.
  • All were comparative efficacy studies.
  • These studies also partially analyzed the role of dosing, timing and clinical characteristics (4,5,6).
  • Interestingly all the above studies showed that the comparison between active experimental treatment and active control treatment were equally effective, rather than showing the experimental treatment to be superior.
  • The LEAPS study suggested that early initiation of ABT may accelerate recovery.
  • The VA ROBOTICS study suggested that the intensity of practiced movements yields results whether delivered by a robot or traditional methods; whereas in the LEAPS study, the lack of intensity-matching did not seem to negatively affect the treatment effect.
  • The VECTORS study indicated a possible interaction between timing and intensity, where greater intensity delivered early may be less beneficial than lower intensity.
  • Other clinical studies have pursued indirect means to promote the effects of ABT, such as the SIRROWS trial, which used daily feedback to reinforce walking function (3).
  • Smaller studies have examined ABT for post-stroke aphasia (2,6,7).

Mechanisms of recovery in humans: spontaneous and therapy induced

  • Development of improved clinical trial designs for recovery and rehabilitation.
  • Improved understanding of mechanisms of neuroplasticity, particularly brain regions and networks engaged.
  • Development of new algorithms to predict recovery based on neuroimaging, physiological and genetic tools.
  • Identification of new techniques, such as central or peripheral nervous system stimulation, which may become adjuvant strategies to repetitive training.
  • Important advances in the ability to characterize--at a systems level--structural and functional neuroplasticity in patients with stroke.

Recovery in aphasia and neglect

  • The efficacy and lasting benefit of non-invasive brain stimulation for neglect and aphasia as a treatment modality has been demonstrated in small studies.1-11
  • Treatment studies have increased in number, moved from case- to small-group and occasional randomized controlled trials (RCT) studies, with some focusing on theory-driven behavioral and mechanistic interventions.12-22
  • Imaging advances have increased the resolution of lesion-symptom mapping as well as the functional, network-based and anatomical bases of these syndromes and their recovery.23-35
  • Studies of language in normal aging have shown right-hemisphere activity to be the norm for older as opposed to younger persons during language tasks. 36-38
  • Studies have begun to examine the intersection between the neural networks affected by stroke and the networks activated by rehabilitation.39-43

Organization of Stroke Care Across the Continuum

  • Early supported discharge (ESD) from inpatient stroke care improves functional and disposition outcomes.6-11
  • NIH driven development of outcome measures may be used to assess the effectiveness of organized stroke care across the continuum (PROMIS,12, 13 Neuro-QoL,14, 15 and Toolbox.16).

Pediatric issues in stroke recovery and rehabilitation

  • Improved understanding of determinants of outcome after pediatric stroke (Domi, 2009; DeVetten, 2010; Beslow, 2010)
  • Preliminary evidence supporting the potential benefit for constraint-induced movement therapy in pediatric stroke (Taub, 2007)
  • Successful application for grants supporting intensive and novel rehabilitation approaches in children with chronic motor deficits following early life stroke (Bastian, NICHD NCT01288040)
  • Demonstration of feasibility of transcranial magnetic stimulation as a component of rehabilitation in children recovering from stroke (Kirton, 2010)
  • Improved understanding of mechanisms of neuroplasticity relative to rehabilitative intervention in surviving pediatric stroke (Kirton, 2010), and NIH-supported research further exploring this topic (NCT00100503)

Conclusion: Basic science advances, preclinical findings, and an improved understanding of neural plasticity in humans are together beginning to yield the fruit of positive human studies based on therapies that promote brain repair.


Translating preclinical findings to human studies

Since 2007, a host of putative restorative therapies have demonstrated improved recovery in preclinical stroke models. Most prominently, the mechanisms underlying endogenous neurogenesis and cell migration, as well as the positive behavioral effects of exogenous cellular therapies and growth factors have been made. It has been demonstrated in preclinical models that compensatory learning with the less-impaired limb worsens the function of the more paretic limb. Thus, inappropriate motor habits may interfere with recovery. These new findings are critical for understanding the changes in human brains that lead to learned non-use and learned disuse after stroke, and the potential negative impact of learning compensatory strategies. Substantial advances have been made in understanding the response of the aged brain to stroke in pre-clinical models, including greater blood-brain barrier disruption, increased edema, inflammatory cell infiltration and neuronal damage, as well as reduced neurogenesis and differences in molecular mechanisms controlling axonal sprouting.

While several studies have now focused on age as an important factor in behavioral outcomes and the responsiveness of rehabilitative training, they still represent a small proportion of all pre-clinical stroke studies. Further, relatively little attention has yet been paid to comorbidities or gender. It is important to better understand the mechanisms underlying the interactions between these factors and rehabilitative therapy. New, high-resolution imaging modalities such as multiphoton confocal microscopy combined with transgenic models expressing fluorescent proteins with light-sensitive channels (optogenetics) have recently been brought to bear on structural changes after stroke in animal models. These technological advances for in vivo imaging and stimulation now need to address questions directly relevant for recovery and rehabilitation.

Drugs and biological therapies to promote motor recovery and rehabilitation

A large number of therapies appear promising in preclinical studies and early human trials. Larger drug studies modulating neurotransmission and enhancing brain repair are urgently needed to address dose response and tolerability, and to provide more definitive evidence of efficacy. Trials focused on recovery and rehabilitation are distinct from other stroke trials in many ways. For example, the choice of endpoints must match the biological model and the goals of therapy. Many restorative therapies require concomitant experience to shape treatment effects. Research into the science of brain plasticity and repair is key to maximizing the impact of this emerging field of restorative therapeutics.

Devices to promote recovery and rehabilitation

Devices to promote recovery and rehabilitation after stroke can be divided into (1) those that stimulate (either activating or inhibiting) the central nervous system (CNS) with the goal of promoting beneficial effects through neuromodulatory or neuroplastic mechanisms; (2) those that stimulate the peripheral nervous system (PNS) to create movement; (3) rehabilitation robotics; and (4) devices that record signals from the central nervous system (brain-computer interfaces).

Multiple small studies and case studies have demonstrated that CNS neuromodulating technologies (rTMS, tDCS) may be useful in promoting motor recovery or in improving language function, particularly when combined with the patient’s attempt to perform an impaired activity1-3. Neuromuscular stimulation technologies (NMES) are now commercially available. Rehabilitation robotics have been studied in a large cohorts of patients with chronic stroke 4,5. Considerable improvements in function may be seen despite years between stroke and intervention. Cost analyses suggest that, with anticipated reductions in the cost of technology over time, robotic-assisted therapy may become cost efficient. One of the most exciting emerging research areas is the development of devices that record signals from the central nervous system and harness these signals for the control of external devices, so-called “brain-computer interfaces”6,7.

Activity-based therapies (ABT) to promote recovery and rehabilitation

ABTs are impairment-oriented therapies that force practice of a desired behavior, and contrasts with compensatory strategies. ABTs have been most applied to motor rehabilitation, but other examples exist. Including a health economic analysis for ABT can provide important cost-benefit information such as demonstrating that additional rehabilitation may be a cost neutral yet medically beneficial proposition when all healthcare utilization was taken into consideration (10). There has been very little progress on the influences of gender or health disparities, nor much recent work for pediatric patients, in the area of ABT.

Mechanisms of recovery in humans: spontaneous and therapy induced

Stroke leads to structural and functional damage that is susceptible to plastic changes relevant to rehabilitation 1, 2. With standard neurorehabilitation, many stroke patients remain disabled 3. Pharmacologic, biologic, and electrophysiologic techniques are under study to augment training-induced plasticity 4 5. An additional arm of neurorehabilitation uses biomedical and tissue engineering to promote neural repair through innate or exogenous stem cell biology 6, and to develop neuroprostheses to bypass injury 7, 10, 11. Structural12, 13 and metabolic imaging14-17 and neurophysiology18-21 have identified neuroplasticity 26, 27 at local and network levels 22, and may be useful for predicting recovery. 23-25 Neuroimaging and neurophysiological techniques have identified plasticity 26 after stroke 27, 28, 26, 29 that predict outcomes 30 20.

The optimal strategies and techniques to accomplish these goals remain unknown 31, 32 , 33, as do the relationship to outcomes, activities of daily living, and participation and quality of life . Neuropsychological and social/personality dimensions in stroke recovery remain incompletely explored. 35, 36 Studies using a range of interventions are not only opening the door to new treatments, they also provide insights into mechanisms of recovery37,38,60,39, 40, 41, 42,43, 44, 64, 65, 66, 47,67,62,55,56,57.

Recovery in aphasia and neglect

Research demonstrates that different symptoms of neglect and aphasia and circuits underlying recovery and treatment are discretely mapped to specific neural systems13,30-34,44-46. However, much work remains to establish high-resolution lesion-symptom linkages between optimal outcomes, treatment protocols, and subpopulations. A major priority is to generate an improved understanding of the neural basis of therapies that target aphasia or neglect.

Substantial evidence now demonstrates that non-invasive brain stimulation can be useful in this context1-11; however, further work needs to optimize stimulation parameters, accrual and generalizability of effects, and to identify the conditions under which different patients benefit from stimulation.

The role of aging has emerged as a major factor in interpreting the effects of lesion, treatment, and recovery in discrete neural systems36-38. Further research is needed to better understand how language and spatial cognitive circuits change with aging, the degree to which neuroplastic capacity changes with aging, and how the neural response to lesion and treatment may fundamentally differ with age.47

Organization of Stroke Care Across the Continuum

Approximately 87% of patients survive the first month post-stroke and on average live several years thereafter. Stroke care can thus be seen as along a continuum: acutely, the focus is on factors such as reperfusion, brain edema, and stabilization; subacutely, the focus is on recovery, rehabilitation, and preventing medical complications; and chronically, the focus is on factors such as managing enduring symptoms and community reintegration. Stroke care can thus be appreciated as being organized along a continuum.

Much of what we know about organized stroke care across the continuum comes to us from Europe. Inpatient units that provide interdisciplinary combined acute neurological and rehabilitation services (or rehabilitation services only) reduce mortality and improve function and quality of life (QoL).1-5 As noted below, there is also now strong evidence of effectiveness of Early Supported Discharge (ESD).

The successful European experience with organized stroke care has not translated to the US healthcare system. The prevailing system of stroke care in the US should be rigorously evaluated, including the emphasis on acute management relative to post-acute care. Principles of organized stroke care developed in Europe should be translated for implementation in the US. Such a system should be patient-centric, in contrast to the usual provider, institution or payer centered approaches.

Numerous secondary complications can arise during the chronic phase of stroke recovery and negatively impact long-term function and QoL. Principles of organized rehabilitation management for these complications are needed. Finally, the increasing contraction of healthcare resources suggests further attention to the cost-effectiveness of organized stroke rehabilitation services throughout the continuum of care.

Meta-analyses of Existing Data in Stroke Rehabilitation

As of 2010, the Stroke Rehabilitation Evidence-Based Review (SREBR) (13th edition), identified 968 randomized controlled trials RCTs; 805 RCTs investigated therapies, technologies, models of care or medications used in stroke rehabilitation and 163 evaluated strategies related to the secondary prevention of stroke. However, the large number of research studies in stroke rehabilitation is offset by heterogeneity in methodology and outcome measures, and by small numbers. Of 32 meta-analyses pertaining to stroke rehabilitation published after 2007 in the Cochrane library, only 12 were identified as “strong” meaning a large number of subjects/studies and limited heterogeneity.

Knowledge Translation, i.e., clinical implementation of best practices, has the potential to significantly improve patient care.

Important and unresolved issues that are priority areas for future research in stroke rehabilitation include clarifying (1) optimal timing and dosing of therapeutic interventions, (2) the full value of cognitive rehabilitation, which has the potential to improve function in a number of domains such as executive function and memory, and (3) long-term rehabilitation of stroke, as in some cases patients might derive further benefit from continuation of rehabilitation therapy for periods of time that are much longer than current practice.

Pediatric issues in stroke recovery and rehabilitation

The Stroke Progress Review Group report of 2006 provided a comprehensive overview and recommendations about stroke recovery and rehabilitation research – in adults. While most of the points raised are relevant to children, the 2006 Stroke PRG did not specifically address these issues for pediatric stroke.

Much remains to be learned about the fundamentals of outcome in childhood stroke – the prevalence and types of impairments and long-term neurologic sequelae including epilepsy; determinants and predictors of outcomes; and the effects of rehabilitation interventions. These problems are compounded dramatically by factors unique to stroke in childhood. Methods to assess stroke outcome in children are incompletely developed, and must address the difficulty of measuring function in a constantly changing and growing child from birth to adolescence. Progress has been made in defining outcome predictors, including the validation of a pediatric acute stroke scale (Ichord, 2011), and neuroimaging markers (Domi, 2009; DeVetten, 2010; Beslow, 2010). Much more work is needed to evaluate outcome determinants in childhood stroke, including patient characteristics (gender, age at stroke onset, race/ethnicity); injury characteristics (stroke subtype, lesion volume & location); comorbid conditions (heart disease, persistent/progressive vasculopathy, genetic syndromes); and psychosocial factors (family function, socioeconomic status, behavioral/psychiatric disorders). Substantial progress in all of these fundamental research questions is an absolute pre-requisite to undertaking intervention trials on a large scale.

Despite these major limitations and knowledge gaps, innovations and progress have been made in pediatric stroke recovery and rehabilitation, with benefit shown for constraint-induced movement therapy (Taub, 2007); and feasibility of transcranial magnetic stimulation to probe plasticity after stroke in children (Kirton, 2010). Emerging therapies currently in development or in trial in adult stroke could eventually be evaluated in children, where enhanced plasticity and lower prevalence and severity of medical comorbidity should increase the odds of success. These include: virtual reality and related technology-computer augmented function; cognitive rehabilitation; aerobic training; treadmill training; behavioral/cognitive training; stimulant drugs; stem-cell therapies.

A number of health care delivery systems issues affect recovery and rehabilitation for children with stroke. Foremost among these is the fact that post-acute rehabilitation services for children are largely delivered through, and intimately integrated with, the school system. As public education resources are subject to wide regional disparities in funding levels, and are increasingly the target of government spending cutbacks, there will be fewer and lower quality community-based resources for implementing research advances in childhood stroke recovery. In an era where fiscal limitations have spawned a much-needed focus on cost-effectiveness research, it would be important for some of the research budget devoted to recovery and rehabilitation to address these health systems issues. As with adult stroke, spending millions of dollars on advanced technological, cell-based and pharmacological rehabilitation interventions in pediatric stroke will be wasted if not matched by a firm commitment to maintain and strengthen basic health, education, and vocational systems that are needed to sustain and aid the development of children recovering from stroke and related childhood brain injuries.


1) Understanding and harnessing clinical brain plasticity: A need exists for improved methods to measure brain plasticity after stroke. Studies identifying valid, reliable, affordable, and accessible measurements of neuroplasticity, both functional and structural, are thus needed. New methods are emerging that have great promise in this regard but require further study to realize their full potential. Measures of neuroplasticity in the lesioned brain should be better validated with outcomes of high clinical priority that include functional recovery, and, as possible, return to activities of daily living or reinsertion into work environments. A better understanding is needed regarding how these measures of brain plasticity can be used to guide and individualize rehabilitation/restorative therapies in order to achieve best patient outcomes. Issues specific to the aging or the developing brain are important to consider in order to achieve maximum effect across all persons affected by stroke.

2) Understanding the experience-dependent nature of post-stroke plasticity: Substantial data suggest that brain plasticity after stroke, whether spontaneous or treatment-induced, is shaped by experience. This feature distinguishes restorative therapies from other classes of stroke therapy such as prevention, reperfusion, and neuroprotection. Studies are critically needed to understand which experiences are most important, what dose of experience is needed to maximize outcomes, and how to measure these experiences, bearing in mind that in this context, experience includes standard therapies such as speech therapy or physiotherapy as well as psychological and socioeconomic issues. These goals may be aided by an improved understanding of biomarkers of recovery and restorative therapies.

3) Translating restorative post-stroke therapies: The burgeoning knowledge base in the basic science of post-stroke brain repair suggests an enormous opportunity for translating new restorative therapies in humans who otherwise would be destined to years of post-stroke disability. Increased knowledge is needed on many fronts to successfully translate these findings, such as how to match the right patients with the right therapies, how to generalize therapeutic effects across differences such as in age and gender, choice of biomarkers to maximize translation of restorative therapies; and how to combine therapies once each is better understood individually. Successful translation of stroke recovery therapeutics must be seen as a team effort, from bench to bedside to health policy reform, for delivery of appropriate treatment. To advance restorative therapy trials in humans, implementation of Specialized Programs of Translational Stroke Research in Recovery (SPOTS-R2), akin to the SPOTRIAS Centers that were created to foster acute stroke care, are a priority.


Translating preclinical findings to human studies

The 2002 SPRG Report, as other SPRG reports, certainly raised awareness of specific topics in translating preclinical findings to human studies. In addition, scientists in the field often cited this report to help justify the importance of particular topics, both in grant applications and in published reports. Assuming that NIH incorporates the content of SPRG reports into its general mission, either formally or informally, then it is reasonable to conclude that the SPRG report has functioned as a set of guideposts for the field at large, encouraging researchers to direct their efforts toward those issues that have been identified by leaders in the field. This process can be improved in a most direct way by targeting funding opportunities that address one or another of the priorities outlined in the current report.

Drugs and biological therapies to promote motor recovery and rehabilitation

Since 2007, there have been very few NIH funded clinical trials testing drugs or biologics to enhance stroke recovery. Only two such studies examine drugs - the acetylcholine esterase inhibitor, donepezil, is currently being studied to assess whether it promotes functional recovery after stroke; a current study funded to examine the effect of pioglitazone on incidence of recurrent cardiovascular events includes a secondary outcome examining the effect of the drug on cognitive recovery (cognitive decline). With regards to cell therapies, there has been only one NIH funded study - testing autologous bone marrow derived mononuclear cells in patients with acute ischemic stroke. This situation suggests that the SPRG process has not substantially helped translation of promising preclinical results into human restorative stroke trials.

We suggest that the SPRG needs to highlight the lack of funding in drugs and biologics for stroke recovery since the last interim report, and call for specific NIH-directed mechanisms targeting the successful translation of novel drug and biological therapies from animals to stroke patients. While intense investigations in stroke research have focused on the acute window for recanalization therapy, a large body of research indicates that the brain in the first few days to weeks after stroke is primed to respond to both drug and biological based therapies to enhance repair. Partnering with the brain repair committee of the SPRG would help to identify therapeutic approaches ready for testing in clinical trials.

Devices to promote recovery and rehabilitation

The 2002 report identified robotic technologies as potentially valuable in stroke rehabilitation, and the 2007 interim report highlighted the uses of emerging technologies to understand cortical physiology and plasticity. In recognizing the potential for these technologies, the SPRG provided a platform for the marked expansion in technologies for both studying and promoting functional restoration after stroke.

Activity-based therapies (ABT) to promote recovery and rehabilitation

The issues identified by the 2007 SPRG recommendations for future studies on ABTs for stroke rehabilitation and recovery have not been fully resolved. It is significant that ABTs have matured to the point where they are being tested in larger rehabilitative multi-site trials, because these larger studies have the statistical power and design structure to show with some confidence, the relative efficacy of ABTs. These clinical trials are in many regard first generation trials, and as such the sophistication of their design will need to evolve. Although the trial results have generally supported beneficial treatment effects from ABTs, they have not shown the clear results expected from animal data or preliminary human studies. The larger randomized ROBOTICS and LEAPS clinical trials and those clinical trials that have included active comparison groups have not shown dramatic differential efficacy between active treatments. It is noteworthy that pre-post longitudinal improvements do occur, so there is an effective component within the active therapy. What is perplexing is why the experimental ABT protocol of interest does not show convincing superiority. What are key contributing factors within the process of ABT, such as timing, dose, frequency, or target population, that makes the difference? SPRG could contribute more to addressing these pivotal questions but has not.

Recovery in aphasia and neglect

The 2007 SPRG report was well conceptualized, and much research since that time has addressed the priorities set forth in that document. However, it is difficult to discern the direct impact of the 2007 report, and research over the past 5 years has only begun to scratch the surface as far as 2007 SPRG priorities are concerned. Our highest priority recommendation is that more resources should be applied to implementing and directly supporting the recommendations of the SPRG to producing immediate scientific and clinical advances. These resources are necessary to bring NINDS sponsored research to the forefront of global research advances.

Organization of Stroke Care Across the Continuum

The 2002 Recovery and Rehabilitation Group identified as one of their priorities the evaluation of organization of rehabilitation services. Specific objectives under this priority included 1) Investigate determinants of variation in access to stroke rehabilitation services and their quality, intensity and duration; 2) Explore and evaluate new options for timing and sequencing of rehabilitation care; 3) Evaluate relationships between stroke rehabilitation structures, processes, and outcomes; 4) Create common outcome measures across the continuum of stroke care; and 5) Evaluate the cost-effectiveness of rehabilitation services.

There has been minimal progress in objective 1, and 5. There has been considerable progress in objective 2 since 200217 but not since 2007; progress here was likely influenced by SPRG. There has been considerable progress in objective 3, especially with respect to early supported discharge; however, studies were conducted in Europe and it is unlikely that SPRG had a significant role. There has been significant progress in objective 4, which have been driven almost exclusively by the NIH, and SPRG may have had a significant role. SPRG may improve outcomes by helping NINDS appreciate the full continuum of recovery trajectory, and identify priorities that have high clinical and societal relevance with plausible potential for translation to actual clinical practice.

Pediatric issues in stroke recovery and rehabilitation

There has been no direct impact of the SPRG on pediatric rehabilitation after stroke because previous versions of the process did not address pediatric stroke rehabilitation. There may have been benefits for pediatric stroke-related topics in general because the larger SPRG process in 2006 did endeavor to include pediatric stroke in many other topics. As a result of inclusion of pediatric stroke investigators in the 2006 SPRG, there has been increased awareness in the stroke research community in general of the need to support research in pediatric stroke, likely contributing to the successful application by pediatric stroke research community for three R01-level original investigator-initiated grants in topics related to epidemiology (“Vascular Effects of Infection in Pediatric Stroke, Heather Fullerton PI, clinical trial methods (“Validation of the Pediatric NIH Stroke Scale” R. Ichord PI), and acute treatment (“Thrombolysis in Pediatric Stroke: A safety and dose-finding Trial”, C Lefond PI); as well as the successful application by junior pediatric stroke investigators for K23 career development awards in topics related to pediatric stroke (Timothy Bernard, Lori Jordan).


Translating preclinical findings to human studies

Adkins DL, Hsu JE, and Jones TA. 2008. Motor cortical stimulation promotes synaptic plasticity and behavioral improvements following sensorimotor cortex lesions. Exp Neurol 212(1):14-28.

Alaverdashvili M, and Whishaw IQ. 2010. Compensation aids skilled reaching in aging and in recovery from forelimb motor cortex stroke in the rat. Neuroscience 167(1):21-30.

Allred RP, Cappellini CH, and Jones TA. 2010. The "good" limb makes the "bad" limb worse: experience-dependent interhemispheric disruption of functional outcome after cortical infarcts in rats. Behav Neurosci 124(1):124-132.

Allred RP, and Jones TA. 2008. Maladaptive effects of learning with the less-affected forelimb after focal cortical infarcts in rats. Exp Neurol 210(1):172-181.

Andrews EM, Tsai SY, Johnson SC, Farrer JR, Wagner JP, Kopen GC, and Kartje GL. 2008. Human adult bone marrow-derived somatic cell therapy results in functional recovery and axonal plasticity following stroke in the rat. Exp Neurol 211(2):588-592.

Bao X, Wei J, Feng M, Lu S, Li G, Dou W, Ma W, Ma S, An Y, Qin C et al. . 2011. Transplantation of human bone marrow-derived mesenchymal stem cells promotes behavioral recovery and endogenous neurogenesis after cerebral ischemia in rats. Brain Res 1367:103-113.

Brown CE, Wong C, and Murphy TH. 2008. Rapid morphologic plasticity of peri-infarct dendritic spines after focal ischemic stroke. Stroke 39(4):1286-1291.

Buga AM, Dunoiu C, Balseanu A, and Popa-Wagner A. 2008. Cellular and molecular mechanisms underlying neurorehabilitation after stroke in aged subjects. Rom J Morphol Embryol 49(3):279-302.

Chen C, Cheng Y, and Chen J. 2011. Transfection of Noggin in bone marrow stromal cells (BMSCs) enhances BMSC-induced functional outcome after stroke in rats. J Neurosci Res 89(8):1194-1202.

Chen Y, and Sun FY. 2007. Age-related decrease of striatal neurogenesis is associated with apoptosis of neural precursors and newborn neurons in rat brain after ischemia. Brain Res 1166:9-19.

Cui X, Chopp M, Zacharek A, Roberts C, Lu M, Savant-Bhonsale S, and Chen J. 2009. Chemokine, vascular and therapeutic effects of combination Simvastatin and BMSC treatment of stroke. Neurobiol Dis 36(1):35-41.

Dinapoli VA, Benkovic SA, Li X, Kelly KA, Miller DB, Rosen CL, Huber JD, and O'Callaghan JP. 2010. Age exaggerates proinflammatory cytokine signaling and truncates signal transducers and activators of transcription 3 signaling following ischemic stroke in the rat. Neuroscience 170(2):633-644.

DiNapoli VA, Huber JD, Houser K, Li X, and Rosen CL. 2008. Early disruptions of the blood-brain barrier may contribute to exacerbated neuronal damage and prolonged functional recovery following stroke in aged rats. Neurobiol Aging 29(5):753-764.

Eisner-Janowicz I, Barbay S, Hoover E, Stowe AM, Frost SB, Plautz EJ, and Nudo RJ. 2008. Early and late changes in the distal forelimb representation of the supplementary motor area after injury to frontal motor areas in the squirrel monkey. J Neurophysiol 100(3):1498-1512.

Fang PC, Barbay S, Plautz EJ, Hoover E, Strittmatter SM, and Nudo RJ. 2010. Combination of NEP 1-40 treatment and motor training enhances behavioral recovery after a focal cortical infarct in rats. Stroke 41(3):544-549.

Gillani RL, Tsai SY, Wallace DG, O'Brien TE, Arhebamen E, Tole M, Schwab ME, and Kartje GL. 2010. Cognitive recovery in the aged rat after stroke and anti-Nogo-A immunotherapy. Behav Brain Res 208(2):415-424.

Guerra-Crespo M, Gleason D, Sistos A, Toosky T, Solaroglu I, Zhang JH, Bryant PJ, and Fallon JH. 2009. Transforming growth factor-alpha induces neurogenesis and behavioral improvement in a chronic stroke model. Neuroscience 160(2):470-483.

Han Q, Li B, Feng H, Xiao Z, Chen B, Zhao Y, Huang J, and Dai J. 2011. The promotion of cerebral ischemia recovery in rats by laminin-binding BDNF. Biomaterials 32(22):5077-5085.

Iaci JF, Ganguly A, Finklestein SP, Parry TJ, Ren J, Saha S, Sietsma DK, Srinivas M, Vecchione AM, and Caggiano AO. 2010. Glial growth factor 2 promotes functional recovery with treatment initiated up to 7 days after permanent focal ischemic stroke. Neuropharmacology 59(7-8):640-649.

Keiner S, Wurm F, Kunze A, Witte OW, and Redecker C. 2008. Rehabilitative therapies differentially alter proliferation and survival of glial cell populations in the perilesional zone of cortical infarcts. Glia 56(5):516-527.

Kerr AL, Cheng SY, and Jones TA. 2011. Experience-dependent neural plasticity in the adult damaged brain. J Commun Disord 44(5):538-548.

Lee HJ, Lim IJ, Lee MC, and Kim SU. 2010. Human neural stem cells genetically modified to overexpress brain-derived neurotrophic factor promote functional recovery and neuroprotection in a mouse stroke model. J Neurosci Res 88(15):3282-3294.

Leu S, Lin YC, Yuen CM, Yen CH, Kao YH, Sun CK, and Yip HK. 2010. Adipose-derived mesenchymal stem cells markedly attenuate brain infarct size and improve neurological function in rats. J Transl Med 8:63.

Li S, Overman JJ, Katsman D, Kozlov SV, Donnelly CJ, Twiss JL, Giger RJ, Coppola G, Geschwind DH, and Carmichael ST. 2010. An age-related sprouting transcriptome provides molecular control of axonal sprouting after stroke. Nat Neurosci 13(12):1496-1504.

Lim DH, Alaverdashvili M, and Whishaw IQ. 2009. Nicotine does not improve recovery from learned nonuse nor enhance constraint-induced therapy after motor cortex stroke in the rat. Behav Brain Res 198(2):411-419.

Liu XS, Chopp M, Zhang RL, Tao T, Wang XL, Kassis H, Hozeska-Solgot A, Zhang L, Chen C, and Zhang ZG. 2011. MicroRNA Profiling in Subventricular Zone after Stroke: MiR-124a Regulates Proliferation of Neural Progenitor Cells through Notch Signaling Pathway. PLoS One 6(8):e23461.

Maldonado MA, Allred RP, Felthauser EL, and Jones TA. 2008. Motor skill training, but not voluntary exercise, improves skilled reaching after unilateral ischemic lesions of the sensorimotor cortex in rats. Neurorehabil Neural Repair 22(3):250-261.

Mochizuki N, Takagi N, Kurokawa K, Onozato C, Moriyama Y, Tanonaka K, and Takeo S. 2008. Injection of neural progenitor cells improved learning and memory dysfunction after cerebral ischemia. Exp Neurol 211(1):194-202.

Molina-Luna K, Pekanovic A, Rohrich S, Hertler B, Schubring-Giese M, Rioult-Pedotti MS, and Luft AR. 2009. Dopamine in motor cortex is necessary for skill learning and synaptic plasticity. PLoS One 4(9):e7082.

Moon SK, Alaverdashvili M, Cross AR, and Whishaw IQ. 2009. Both compensation and recovery of skilled reaching following small photothrombotic stroke to motor cortex in the rat. Exp Neurol 218(1):145-153.

Moore TL, Killiany RJ, Pessina MA, Moss MB, Finklestein SP, and Rosene DL. 2011. Recovery from ischemia in the middle-aged brain: a nonhuman primate model. Neurobiol Aging.

Morris DC, Zhang ZG, Wang Y, Zhang RL, Gregg S, Liu XS, and Chopp M. 2007. Wnt expression in the adult rat subventricular zone after stroke. Neurosci Lett 418(2):170-174.

Muller HD, Hanumanthiah KM, Diederich K, Schwab S, Schabitz WR, and Sommer C. 2008. Brain-derived neurotrophic factor but not forced arm use improves long-term outcome after photothrombotic stroke and transiently upregulates binding densities of excitatory glutamate receptors in the rat brain. Stroke 39(3):1012-1021.

Ploughman M, Attwood Z, White N, Dore JJ, and Corbett D. 2007. Endurance exercise facilitates relearning of forelimb motor skill after focal ischemia. Eur J Neurosci 25(11):3453-3460.

Popa-Wagner A, Carmichael ST, Kokaia Z, Kessler C, and Walker LC. 2007. The response of the aged brain to stroke: too much, too soon? Curr Neurovasc Res 4(3):216-227.

Shehadah A, Chen J, Cui X, Roberts C, Lu M, and Chopp M. 2010. Combination treatment of experimental stroke with Niaspan and Simvastatin, reduces axonal damage and improves functional outcome. J Neurol Sci 294(1-2):107-111.

Shimamura M, Sato N, Sata M, Kurinami H, Takeuchi D, Wakayama K, Hayashi T, Iida H, and Morishita R. 2007. Delayed postischemic treatment with fluvastatin improved cognitive impairment after stroke in rats. Stroke 38(12):3251-3258.

Sigler A, and Murphy TH. 2010. In vivo 2-photon imaging of fine structure in the rodent brain: before, during, and after stroke. Stroke 41(10 Suppl): S117-123.

Soderstrom I, Strand M, Ingridsson AC, Nasic S, and Olsson T. 2009. 17beta-estradiol and enriched environment accelerate cognitive recovery after focal brain ischemia. Eur J Neurosci 29(6):1215-1224.

Takasawa M, Beech JS, Fryer TD, Jones PS, Ahmed T, Smith R, Aigbirhio FI, and Baron JC. 2011. Single-subject statistical mapping of acute brain hypoxia in the rat following middle cerebral artery occlusion: a microPET study. Exp Neurol 229(2):251-258.

Teng H, Zhang ZG, Wang L, Zhang RL, Zhang L, Morris D, Gregg SR, Wu Z, Jiang A, Lu M et al. . 2008. Coupling of angiogenesis and neurogenesis in cultured endothelial cells and neural progenitor cells after stroke. J Cereb Blood Flow Metab 28(4):764-771.

Tsai SY, Papadopoulos CM, Schwab ME, and Kartje GL. 2011. Delayed anti-nogo-a therapy improves function after chronic stroke in adult rats. Stroke 42(1):186-190.

Wang T, Wang J, Yin C, Liu R, Zhang JH, and Qin X. 2010. Down-regulation of Nogo receptor promotes functional recovery by enhancing axonal connectivity after experimental stroke in rats. Brain Res 1360:147-158.

Wurm F, Keiner S, Kunze A, Witte OW, and Redecker C. 2007. Effects of skilled forelimb training on hippocampal neurogenesis and spatial learning after focal cortical infarcts in the adult rat brain. Stroke 38(10):2833-2840.

Zai L, Ferrari C, Dice C, Subbaiah S, Havton LA, Coppola G, Geschwind D, Irwin N, Huebner E, Strittmatter SM et al. . 2011. Inosine augments the effects of a Nogo receptor blocker and of environmental enrichment to restore skilled forelimb use after stroke. J Neurosci 31(16):5977-5988.

Zhang C, Chopp M, Cui Y, Wang L, Zhang R, Zhang L, Lu M, Szalad A, Doppler E, Hitzl M et al. . 2010. Cerebrolysin enhances neurogenesis in the ischemic brain and improves functional outcome after stroke. J Neurosci Res 88(15):3275-3281.

Zhang ZH, Wang RZ, Li GL, Wei JJ, Li ZJ, Feng M, Kang J, Du WC, Ma WB, Li YN et al. . 2008. Transplantation of neural stem cells modified by human neurotrophin-3 promotes functional recovery after transient focal cerebral ischemia in rats. Neurosci Lett 444(3):227-230. Zhao C, Wang J, Zhao S, and Nie Y. 2009. Constraint-induced movement therapy enhanced neurogenesis and behavioral recovery after stroke in adult rats. Tohoku J Exp Med 218(4):301-308.

Drugs and biological therapies to promote motor recovery and rehabilitation

1. Chollet, F., et al., Fluoxetine for motor recovery after acute ischaemic stroke (FLAME): a randomised placebo-controlled trial. Lancet Neurol. 10(2): p. 123-30.

2. Cramer, S.C., et al., The beta-hCG+erythropoietin in acute stroke (BETAS) study: a 3-center, single-dose, open-label, noncontrolled, phase IIa safety trial. Stroke. 41(5): p. 927-31.

3. Silver, B., et al., Sildenafil treatment of subacute ischemic stroke: a safety study at 25-mg daily for 2 weeks. J Stroke Cerebrovasc Dis, 2009. 18(5): p. 381-3.

4. Honmou, O., et al., Intravenous administration of auto serum-expanded autologous mesenchymal stem cells in stroke. Brain. 134(Pt 6): p. 1790-807.

5. Lee, J.S., et al., A long-term follow-up study of intravenous autologous mesenchymal stem cell transplantation in patients with ischemic stroke. Stem Cells. 28(6): p. 1099-106.

6. Savitz, S.I., et al., Intravenous autologous bone marrow mononuclear cells for ischemic stroke. Ann Neurol, 2011. 70(1): p. 59-69.

Devices to promote recovery and rehabilitation

Bolognini, N., Pascual-Leone, A. & Fregni, F. Using non-invasive brain stimulation to augment motor training-induced plasticity. J Neuroeng Rehabil 6, 8, doi:1743-0003-6-8 [pii]
10.1186/1743-0003-6-8 (2009).

2 Dimyan, M. A. & Cohen, L. G. Contribution of transcranial magnetic stimulation to the understanding of functional recovery mechanisms after stroke. Neurorehabil Neural Repair 24, 125-135, doi:1545968309345270 [pii]
10.1177/1545968309345270 (2010).

3 Galletta, E. E., Rao, P. R. & Barrett, A. M. Transcranial magnetic stimulation (TMS): potential progress for language improvement in aphasia. Top Stroke Rehabil 18, 87-91, doi:HL8342402761J029 [pii]
10.1310/tsr1802-87 (2011).

4 Lo, A. C. et al. Multicenter randomized trial of robot-assisted rehabilitation for chronic stroke: methods and entry characteristics for VA ROBOTICS. Neurorehabil Neural Repair 23, 775-783, doi:1545968309338195 [pii]
10.1177/1545968309338195 (2009).

5 Lo, A. C. et al. Robot-assisted therapy for long-term upper-limb impairment after stroke. N Engl J Med 362, 1772-1783, doi:NEJMoa0911341 [pii] 10.1056/NEJMoa0911341 (2010).

6 Simeral, J. D., Kim, S. P., Black, M. J., Donoghue, J. P. & Hochberg, L. R. Neural control of cursor trajectory and click by a human with tetraplegia 1000 days after implant of an intracortical microelectrode array. J Neural Eng 8, 025027, doi:S1741-2560(11)64722-8 [pii] 10.1088/1741-2560/8/2/025027 (2011).

7 Pancrazio, J. J. & Peckham, P. H. Neuroprosthetic devices: how far are we from recovering movement in paralyzed patients? Expert Rev Neurother 9, 427-430, doi:10.1586/ern.09.12 (2009).

Activity-based therapies to promote recovery and rehabilitation

1. Cramer, S. C. "Stratifying patients with stroke in trials that target brain repair." Stroke 41(10 Suppl): S114-6.

2. de Jong-Hagelstein, M., W. M. van de Sandt-Koenderman, et al. "Efficacy of early cognitive-linguistic treatment and communicative treatment in aphasia after stroke: a randomised controlled trial (RATS-2)." J Neurol Neurosurg Psychiatry 82(4): 399-404.

3. Dobkin, B. H., P. Plummer-D'Amato, et al. "International randomized clinical trial, stroke inpatient rehabilitation with reinforcement of walking speed (SIRROWS), improves outcomes." Neurorehabil Neural Repair 24(3): 235-42.

4. Dromerick, A. W., C. E. Lang, et al. (2009). "Very Early Constraint-Induced Movement during Stroke Rehabilitation (VECTORS): A single-center RCT." Neurology 73(3): 195-201.

5. Duncan, P. W., K. J. Sullivan, et al. "Body-weight-supported treadmill rehabilitation after stroke." N Engl J Med 364(21): 2026-36.

6. Fridriksson, J., J. M. Baker, et al. (2009). "Treating visual speech perception to improve speech production in nonfluent aphasia." Stroke 40(3): 853-8.

7. Lee, J., R. Fowler, et al. "IMITATE: An intensive computer-based treatment for aphasia based on action observation and imitation." Aphasiology 24(4): 449-465.

8. Lo, A. C., P. D. Guarino, et al. "Robot-assisted therapy for long-term upper-limb impairment after stroke." N Engl J Med 362(19): 1772-83.

9. Manheim, L. M., A. S. Halper, et al. (2009). "Patient-reported changes in communication after computer-based script training for aphasia." Arch Phys Med Rehabil 90(4): 623-7.

10. Wagner, T. H., A. C. Lo, et al. "An Economic Analysis of Robot-Assisted Therapy for Long-Term Upper-Limb Impairment After Stroke." Stroke.

Mechanisms of recovery in humans: spontaneous and therapy induced

1. Nudo RJ, Wise BM, SiFuentes F. Neural substrate for effects of rehabilitation on motor recovery following focal ischemic infarct. Neurosci. Abstr. 1995;21:517.

2. Taub E, Uswatte G, Elbert T. New treatments in neurorehabilitation founded on basic research. Nat Rev Neurosci. Mar 2002;3(3):228-236.

3. Lloyd-Jones D, Adams R, Carnethon M, et al. Heart disease and stroke statistics--2009 update: a report from the American Heart Association Statistics Committee and Stroke Statistics Subcommittee. Circulation. Jan 27 2009;119(3):480-486.

4. Buonomano DV, Merzenich MM. Cortical plasticity: From synapses to maps. Annual Review of Neuroscience. 1998;21:149-186.

5. Payne BR, Lomber SG. Reconstructing functional systems after lesions of cerebral cortex. Nat Rev Neurosci. Dec 2001;2(12):911-919.

6. Lindvall O, Kokaia Z. Stem cells for the treatment of neurological disorders. Nature. Jun 29 2006;441(7097):1094-1096.

7. Wolpaw JR, Birbaumer N, McFarland DJ, Pfurtscheller G, Vaughan TM. Brain-computer interfaces for communication and control. Clin Neurophysiol. Jun 2002;113(6):767-791.

8. Romero JR, Babikian VL, Katz DI, Finklestein SP. Neuroprotection and stroke rehabilitation: modulation and enhancement of recovery. Behav Neurol. 2006;17(1):17-24.

9. Khaja AM. Acute ischemic stroke management: administration of thrombolytics, neuroprotectants, and general principles of medical management. Neurol Clin. Nov 2008;26(4):943-961, viii.

10. Dimyan MA, Cohen LG. Neuroplasticity in the context of motor rehabilitation after stroke. Nat Rev Neurol. Feb 2011;7(2):76-85.

11. Cramer SC, Riley JD. Neuroplasticity and brain repair after stroke. Curr Opin Neurol. Feb 2008;21(1):76-82.

12. Stinear CM, Barber PA, Smale PR, Coxon JP, Fleming MK, Byblow WD. Functional potential in chronic stroke patients depends on corticospinal tract integrity. Brain. Jan 2007;130(Pt 1):170-180.

13. Schaechter JD, Perdue KL, Wang R. Structural damage to the corticospinal tract correlates with bilateral sensorimotor cortex reorganization in stroke patients. Neuroimage. Feb 1 2008;39(3):1370-1382.

14. Ward NS, Frackowiak RS. The functional anatomy of cerebral reorganisation after focal brain injury. J Physiol Paris. Jun 2006;99(4-6):425-436.

15. Johansen-Berg H. Functional imaging of stroke recovery: what have we learnt and where do we go from here? Int J Stroke. Feb 2007;2(1):7-16.

16. Prabhakaran S, Zarahn E, Riley C, et al. Inter-individual variability in the capacity for motor recovery after ischemic stroke. Neurorehabil Neural Repair. Jan-Feb 2008;22(1):64-71.

17. Marshall RS, Perera GM, Lazar RM, Krakauer JW, Constantine RC, DeLaPaz RL. Evolution of cortical activation during recovery from corticospinal tract infarction. Stroke. Mar 2000;31(3):656-661.

18. Butefisch CM, Netz J, Wessling M, Seitz RJ, Homberg V. Remote changes in cortical excitability after stroke. Brain. Feb 2003;126(Pt 2):470-481.

19. Manganotti P, Acler M, Zanette GP, Smania N, Fiaschi A. Motor cortical disinhibition during early and late recovery after stroke. Neurorehabil Neural Repair. Jul-Aug 2008;22(4):396-403.

20. Swayne OB, Rothwell JC, Ward NS, Greenwood RJ. Stages of motor output reorganization after hemispheric stroke suggested by longitudinal studies of cortical physiology. Cereb Cortex. Aug 2008;18(8):1909-1922.

21. Murase N, Duque J, Mazzocchio R, Cohen LG. Influence of interhemispheric interactions on motor function in chronic stroke. Ann Neurol. Mar 2004;55(3):400-409.

22. Wang L, Yu C, Chen H, et al. Dynamic functional reorganization of the motor execution network after stroke. Brain. Apr 2010;133(Pt 4):1224-1238.

23. Cramer SC, Parrish TB, Levy RM, et al. Predicting functional gains in a stroke trial. Stroke. Jul 2007;38(7):2108-2114.

24. Marshall RS, Zarahn E, Alon L, Minzer B, Lazar RM, Krakauer JW. Early imaging correlates of subsequent motor recovery after stroke. Ann Neurol. May 2009;65(5):596-602.

25. Lindenberg R, Renga V, Zhu LL, Betzler F, Alsop D, Schlaug G. Structural integrity of corticospinal motor fibers predicts motor impairment in chronic stroke. Neurology. Jan 26 2010;74(4):280-287.

26. Newton JM, Ward NS, Parker GJ, et al. Non-invasive mapping of corticofugal fibres from multiple motor areas--relevance to stroke recovery. Brain. Jul 2006;129(Pt 7):1844-1858.

27. Fridman EA, Hanakawa T, Chung M, Hummel F, Leiguarda RC, Cohen LG. Reorganization of the human ipsilesional premotor cortex after stroke. Brain. Apr 2004;127(Pt 4):747-758.

28. Ward NS, Newton JM, Swayne OB, et al. Motor system activation after subcortical stroke depends on corticospinal system integrity. Brain; a journal of neurology. Mar 2006;129(Pt 3):809-819.

29. Johansen-Berg H, Rushworth MF, Bogdanovic MD, Kischka U, Wimalaratna S, Matthews PM. The role of ipsilateral premotor cortex in hand movement after stroke. Proc Natl Acad Sci U S A. Oct 29 2002;99(22):14518-14523.

30. Stinear CM, Barber PA, Smale PR, Coxon JP, Fleming MK, Byblow WD. Functional potential in chronic stroke patients depends on corticospinal tract integrity. Brain. January 1, 2007 2007;130(1):170-180.

31. Baron JC, Cohen LG, Cramer SC, et al. Neuroimaging in stroke recovery: a position paper from the First International Workshop on Neuroimaging and Stroke Recovery. Cerebrovasc Dis. 2004;18(3):260-267.

32. Milot MH, Cramer SC. Biomarkers of recovery after stroke. Curr Opin Neurol. Dec 2008;21(6):654-659.

33. Cheeran B, Cohen L, Dobkin B, et al. The future of restorative neurosciences in stroke: driving the translational research pipeline from basic science to rehabilitation of people after stroke. Neurorehabil Neural Repair. Feb 2009;23(2):97-107.

34. Harris-Love ML, Morton SM, Perez MA, Cohen LG. Mechanisms of Short-Term Training-Induced Reaching Improvement in Severely Hemiparetic Stroke Patients: A TMS Study. Neurorehabil Neural Repair. Feb 22 2011.

35. Scheidtmann K, Fries W, Muller F, Koenig E. Effect of levodopa in combination with physiotherapy on functional motor recovery after stroke: a prospective, randomised, double-blind study. Lancet. Sep 8 2001;358(9284):787-790.

36. Sonde L, Lokk J. Effects of amphetamine and/or L-dopa and physiotherapy after stroke - a blinded randomized study. Acta Neurol Scand. Jan 2007;115(1):55-59.

37. Altman J. Are new neurons formed in the brains of adult mammals? Science. Mar 30 1962;135:1127-1128.

38. Benowitz LI, Carmichael ST. Promoting axonal rewiring to improve outcome after stroke. Neurobiol Dis. Feb 2010;37(2):259-266.

39. Popa-Wagner A, Stocker K, Balseanu AT, et al. Effects of granulocyte-colony stimulating factor after stroke in aged rats. Stroke. May 2010;41(5):1027-1031.

40. Belayev L, Khoutorova L, Zhao KL, Davidoff AW, Moore AF, Cramer SC. A novel neurotrophic therapeutic strategy for experimental stroke. Brain Res. Jul 14 2009;1280:117-123.

41. Ploughman M, Windle V, MacLellan CL, White N, Dore JJ, Corbett D. Brain-derived neurotrophic factor contributes to recovery of skilled reaching after focal ischemia in rats. Stroke. Apr 2009;40(4):1490-1495.

42. Muller HD, Hanumanthiah KM, Diederich K, Schwab S, Schabitz WR, Sommer C. Brain-derived neurotrophic factor but not forced arm use improves long-term outcome after photothrombotic stroke and transiently upregulates binding densities of excitatory glutamate receptors in the rat brain. Stroke. Mar 2008;39(3):1012-1021.

43. Ehrenreich H, Weissenborn K, Prange H, et al. Recombinant human erythropoietin in the treatment of acute ischemic stroke. Stroke. Dec 2009;40(12):e647-656.

44. Cramer SC, Fitzpatrick C, Warren M, et al. The beta-hCG+erythropoietin in acute stroke (BETAS) study: a 3-center, single-dose, open-label, noncontrolled, phase IIa safety trial. Stroke. May 2010;41(5):927-931.

45. Liebetanz D, Nitsche MA, Tergau F, Paulus W. Pharmacological approach to the mechanisms of transcranial DC-stimulation-induced after-effects of human motor cortex excitability. Brain. Oct 2002;125(Pt 10):2238-2247.

46. Reis J, Robertson EM, Krakauer JW, et al. Consensus: Can transcranial direct current stimulation and transcranial magnetic stimulation enhance motor learning and memory formation? Brain Stimul. Oct 2008;1(4):363-369.

47. Fritsch B, Reis J, Martinowich K, et al. Direct current stimulation promotes BDNF-dependent synaptic plasticity: potential implications for motor learning. Neuron. Apr 29 2010;66(2):198-204.

48. Lotze M, Markert J, Sauseng P, Hoppe J, Plewnia C, Gerloff C. The role of multiple contralesional motor areas for complex hand movements after internal capsular lesion. J Neurosci. May 31 2006;26(22):6096-6102.

49. Gerloff C, Bushara K, Sailer A, et al. Multimodal imaging of brain reorganization in motor areas of the contralesional hemisphere of well recovered patients after capsular stroke. Brain. Mar 2006;129(Pt 3):791-808.

50. Nair DG, Hutchinson S, Fregni F, Alexander M, Pascual-Leone A, Schlaug G. Imaging correlates of motor recovery from cerebral infarction and their physiological significance in well-recovered patients. Neuroimage. 2007 Jan 1 2007;34(1):253-263.

51. Grefkes C, Nowak DA, Eickhoff SB, et al. Cortical connectivity after subcortical stroke assessed with functional magnetic resonance imaging. Ann Neurol. Feb 2008;63(2):236-246.

52. Nowak DA, Grefkes C, Ameli M, Fink GR. Interhemispheric Competition After Stroke: Brain Stimulation to Enhance Recovery of Function of the Affected Hand. Neurorehabil Neural Repair. Jun 16 2009.

53. Kinsbourne M. Hemi-neglect and hemisphere rivalry. Adv Neurol. 1977;18:41-49.

54. Ameli M, Grefkes C, Kemper F, et al. Differential effects of high-frequency repetitive transcranial magnetic stimulation over ipsilesional primary motor cortex in cortical and subcortical middle cerebral artery stroke. Ann Neurol. Sep 2009;66(3):298-309.

55. Shin HI, Han TR, Paik NJ. Effect of consecutive application of paired associative stimulation on motor recovery in a rat stroke model: a preliminary study. Int J Neurosci. Jun 2008;118(6):807-820.

56. Castel-Lacanal E, Marque P, Tardy J, et al. Induction of cortical plastic changes in wrist muscles by paired associative stimulation in the recovery phase of stroke patients. Neurorehabil Neural Repair. May 2009;23(4):366-372.

57. Celnik P, Paik NJ, Vandermeeren Y, Dimyan M, Cohen LG. Effects of Combined Peripheral Nerve Stimulation and Brain Polarization on Performance of a Motor Sequence Task After Chronic Stroke. Stroke. Mar 12 2009.

58. Heilman KM, Watson RT and Valenstein E, 2011. Neglect and Related Disorders. In: Clinical Neuropsychology, 4th edition. New York: Oxford University Press, pp. 314-315, 332.

59. Parkinson BR, Raymer A, Chang YL, Fitzgerald DB, Crosson B. Lesion characteristics related to treatment improvement in object and action naming for patients with chronic aphasia. Brain and Language Aug 2009; 110(2):61-70.

Recovery in aphasia and neglect

1. Barwood CHS, Murdoch BE, Whelan B-M, Lloyd D, Riek S, O’sullivan J, Coulthard A, Wong A, Aitken P, Hall G. (2011) The effects of low frequency Repetitive Transcranial Magnetic Stimulation (rTMS) and sham condition rTMS on behavioural language in chronic nonfluent aphasia: Short term outcomes. Neurorehabilitation 28:113-128.

2. Hamilton, R.H., Chrysikou, E.G., Coslett, B. (2011) Mechanisms of aphasia recovery after stroke and the role of noninvasive brain stimulation. Brain and Language, 118, 40-50.

3. Monti, A., Cogiamanian, F., Marceglia, S., Ferrucci, R., Mameli, F., Mrakic-Sposta, S., et al. (2008). Improved naming after transcranial direct current stimulation in aphasia. Journal of Neurology, Neurosurgery and Psychiatry, 79, 451–453.

4. Kakuda, W., Abo, M., Momosaki, R., Morooka, A. (2011) Therapeutic application of 6-Hz-primed low-frequency rTMS combined with intensive speech therapy for post-stroke aphasia. Brain Injury (e-pub ahead of print).

5. Schlaug, G., Marchina, S., Wan, C.Y. (2011). The Use of Non-invasive Brain Stimulation Techniques to Facilitate Recovery from Post-stroke Aphasia. Neuropsychology Review, 21, 288–301.

6. You, D.S., Kim, D.-Y., Min Ho Chun, M.H., Jung, S.E., Park, S.J. (2011). Cathodal transcranial direct current stimulation of the right Wernicke’s area improves comprehension in subacute stroke patients. Brain & Language, 119, 1–5.

7. Sparing R, Thimm M, Hesse MD, Küst J, Karbe H, Fink GR. Bidirectional alterations of interhemispheric parietal balance by non-invasive cortical stimulation. Brain. 2009 Nov;132(Pt 11):3011-20.

8. Song W, Du B, Xu Q, Hu J, Wang M, Luo Y. Low-frequency transcranial magnetic stimulation for visual spatial neglect: a pilot study. J Rehabil Med. 2009 Feb;41(3):162-5.

9. Lim JY, Kang EK, Paik NJ. Repetitive transcranial magnetic stimulation to hemispatial neglect in patients after stroke: an open-label pilot study. J Rehabil Med. 2010 May;42(5):447-52.

10. Shindo K, Sugiyama K, Huabao L, Nishijima K, Kondo T, Izumi S. Long-term effect of low-frequency repetitive transcranial magnetic stimulation over the unaffected posterior parietal cortex in patients with unilateral spatial neglect. J Rehabil Med. 2006 Jan;38(1):65-7.

11. Ko MH, Han SH, Park SH, Seo JH, Kim YH. Improvement of visual scanning after DC brain polarization of parietal cortex in stroke patients with spatial neglect. Neurosci Lett. 2008 Dec 26;448(2):171-4.

12. Crosson, B., Fabrizio, K. S., Singletary, F., Cato, M. A., Wierenga, C. E., Parkinson, R. B., Sherod, M. E., Bacon Moore, A., Ciampitti, M., Holiway, B., Leon, S., Rodriguez, A., Kendall, D. L., Levy, I. F., Gonzalez Rothi, L. J. (2007). Treatment of naming in nonfluent aphasia through manipulation of intention and attention: A phase 1 comparison of two novel treatments. Journal of the International Neuropsychological Society, 13, 582-594.

13. Crosson B, Moore AB, McGregor KM, Chang YL, Benjamin M, Gopinath K, et al. (2009) Regional changes in word-production laterality after a naming treatment designed to produce a rightward shift in frontal activity. Brain and Language, 111,73–85.

14. Raymer, A. M., Beeson, Holland, A. L., P. Kendall, D., Maher, L. M., Martine, N., Murray, L., Rose, M., Thompson, C. K., Turkstra, L., Altman, L. Boyle, M., Conway, T., Hula, W., Kearns, K., Rapp, B., Simmons-Mackie, N., & Gonzalez Rothi, L. J. (2008). Translational Research in Aphasia: From Neuroscience to Neurorehabilitation. Journal of Speech, Language, and Hearing Research, 51, S259-S275.

15. Robey, R. R. (2011). Treatment Effectiveness and Evidence-Based Practice. In L. LaPointe (Ed), Aphasia and Related Neurogenic Language Disorders (4th ed.). New York: Theime.

16. Bays PM, Singh-Curry V, Gorgoraptis N, Driver J, Husain M. Integration of goal- and stimulus-related visual signals revealed by damage to human parietal cortex. J Neurosci. 2010 Apr 28;30(17):5968-78.

17. Reinhart S, Schindler I, Kerkhoff G. Optokinetic stimulation affects word omissions but not stimulus-centered reading errors in paragraph reading in neglect dyslexia. Neuropsychologia. 2011 Jul;49(9):2728-35.

18. Thimm M, Fink GR, Küst J, Karbe H, Willmes K, Sturm W. Recovery from hemineglect: differential neurobiological effects of optokinetic stimulation and alertness training. Cortex. 2009 Jul-Aug;45(7):850-62.

19. Làdavas E, Bonifazi S, Catena L, Serino A. Neglect rehabilitation by prism adaptation: different procedures have different impacts. Neuropsychologia. 2011 Apr;49(5):1136-45.

20. Fortis P, Goedert KM, Barrett AM. Prism adaptation differently affects motor-intentional and perceptual-attentional biases in healthy individuals. Neuropsychologia. 2011 Jul;49(9):2718-27.

21. Mizuno K, Tsuji T, Takebayashi T, Fujiwara T, Hase K, Liu M. Prism Adaptation Therapy Enhances Rehabilitation of Stroke Patients With Unilateral Spatial Neglect: A Randomized, Controlled Trial. Neurorehabil Neural Repair. 2011 Jun 23.

22. Serino A, Angeli V, Frassinetti F, Làdavas E. Mechanisms underlying neglect recovery after prism adaptation. Neuropsychologia. 2006;44(7):1068-78.

23. Holland, A. L. (2008). Recent Advances and Future Directions in Aphasia Therapy. Brain Impairment, 9, 179-190.

24. Meinzer, M., Harnish, S., Conway, T., Crosson, B. (2011). Recent developments in functional and structural imaging of aphasia recovery after stroke. Aphasiology, 25, 271-290.

25. Vuilleumier P, Schwartz S, Verdon V, Maravita A, Hutton C, Husain M, Driver J. Abnormal attentional modulation of retinotopic cortex in parietal patients with spatial neglect. Curr Biol. 2008 Oct 14;18(19):1525-9.

26. Ptak R, Schnider A. The dorsal attention network mediates orienting toward behaviorally relevant stimuli in spatial neglect. J Neurosci. 2010 Sep 22;30(38):12557-65.

27. Committeri G, Pitzalis S, Galati G, Patria F, Pelle G, Sabatini U, Castriota-Scanderbeg A, Piccardi L, Guariglia C, Pizzamiglio L. Neural bases of personal and extrapersonal neglect in humans. Brain. 2007 Feb;130(Pt 2):431-41.

28. Koch G, Oliveri M, Cheeran B, Ruge D, Lo Gerfo E, Salerno S, Torriero S, Marconi B, Mori F, Driver J, Rothwell JC, Caltagirone C. Hyperexcitability of parietal-motor functional connections in the intact left-hemisphere of patients with neglect. Brain. 2008 Dec;131(Pt 12):3147-55.

29. He BJ, Snyder AZ, Vincent JL, Epstein A, Shulman GL, Corbetta M. Breakdown of functional connectivity in frontoparietal networks underlies behavioral deficits in spatial neglect. Neuron. 2007 Mar 15;53(6):905-18.

30. Verdon V, Schwartz S, Lovblad KO, Hauert CA, Vuilleumier P. Neuroanatomy of hemispatial neglect and its functional components: a study using voxel-based lesion-symptom mapping. Brain. 2010 Mar;133(Pt 3):880-94.

31. Shirani P, Thorn J, Davis C, Heidler-Gary J, Newhart M, Gottesman RF, Hillis AE. Severity of hypoperfusion in distinct brain regions predicts severity of hemispatial neglect in different reference frames. Stroke 2009 Nov; 40(11):3563-6.

32. Lee BH, Suh MK, Kim EJ, Seo SW, Choi KM, Kim GM, Chung CS, Heilman KM, Na DL. Neglect dyslexia: frequency, association with other hemispatial neglects, and lesion localization. Neuropsychologia. 2009 Feb;47(3):704-10.

33. Medina J, Kannan V, Pawlak MA, Kleinman JT, Newhart M, Davis C, Heidler-Gary JE, Herskovits EH, Hillis AE. Neural substrates of visuospatial processing in distinct reference frames: evidence from unilateral spatial neglect. J Cogn Neurosci. 2009 Nov;21(11):2073-84.

34. Vossel S, Eschenbeck P, Weiss PH, Weidner R, Saliger J, Karbe H, Fink GR. Visual extinction in relation to visuospatial neglect after right-hemispheric stroke: quantitative assessment and statistical lesion-symptom mapping. J Neurol Neurosurg Psychiatry. 2011 Aug;82(8):862-8.

35. Bartolomeo P, Thiebaut de Schotten M, Doricchi F. Left unilateral neglect as a disconnection syndrome. Cereb Cortex. 2007 Nov;17(11):2479-90.

36. Meinzer M, Flaisch T, Wilser L, Eulitz C, Rockstroh B, Conway T, Rothi, LJG, Crosson B. (2009). Neural signatures of semantic and phonemic fluency in young and old adults. Journal of Cognitive Neuroscience, 21, 2007-2018.

37. Meinzer, M., Seeds, L., Flaisch, T., Harnish, S., Cohen, M.L., McGregor, K., Conway, T., Benjamin, M., Crosson, B. (2010), Impact of changed positive and negative task-related functional activity on word-retrieval in aging. Neurobiology of Aging, e-pub ahead of print.

38. Wierenga CE, Benjamin M, Gopinath K, Perlstein WM, Leonard CM, Rothi LJG, Conway T, Cato MA, Briggs RW, Crosson B. (2008) Age-related changes in word retrieval: Role of bilateral frontal and subcortical networks. Neurobiology of Aging, 29, 436-451.

39. Fridriksson, J. (2010) Preservation and modulation of specific left hemisphere regions is vital for treated recovery from anomia in stroke. Journal of Neuroscience, 30, 11558–11564.

40. Fridriksson, J., Bonilha, L., Baker, J.M., Moser, D., & Rorden, C. (2010). Activity in preserved left hemisphere regions predicts anomia severity in aphasia. Cerebral Cortex, 20, 1013–1019.

41. Thimm M, Fink GR, Küst J, Karbe H, Willmes K, Sturm W. Recovery from hemineglect: differential neurobiological effects of optokinetic stimulation and alertness training. Cortex. 2009 Jul-Aug;45(7):850-62.

42. Thimm M, Fink GR, Küst J, Karbe H, Sturm W. Impact of alertness training on spatial neglect: a behavioural and fMRI study. Neuropsychologia. 2006;44(7):1230-46

43. Hassa T, Schoenfeld MA, Dettmers C, Stoppel CM, Weiller C, Lange R. Neural correlates of somatosensory processing in patients with neglect. Restor Neurol Neurosci. 2011;29(4):253-63.

44. Breier, J.I., Juranek, J., Maher, L.M., Schmadeke, S., Men, D., & Papanicolaou, A.C. (2009). Behavioural and neurophysiologic response to therapy for chronic aphasia. Archives of Physical and Medical Rehabilitation, 90, 2026–2033.

45. Meinzer, M., Flaisch, T., Breitenstein, C., Wienbruch, C., Elbert, T., & Rockstroh, B. (2008) Functional re-recruitment of dysfunctional brain areas predicts language recovery in chronic aphasia. Neuroimage 39:2038–2046.

46. Warren, J.E., Crinion, J.T., Lambon Ralph, M.A,, Wise, R.J. (2009) Anterior temporal lobe connectivity correlates with functional outcome after aphasic stroke. Brain, 132, 3428–3442.

47. Simmons-Mackie, N., Conklin, J., & Kagan, A. (2008). Think Tank Deliberates Future Directions for the Social Approach to Aphasia. Perspectives on Neurophysiology and Neurogenic Speech and Language Disorders, 18, 24-32.

Organization of Stroke Care Across the Continuum

1. Collaboration SUTs. Organised inpatient (stroke unit) care for stroke. Stroke Unit Trialists' Collaboration. Cochrane Database Syst Rev 2000(2):CD000197.

2. Collaboration SUTs. How do stroke units improve patient outcomes? A collaborative systematic review of the randomized trials. Stroke Unit Trialists Collaboration. Stroke 1997;28(11):2139-44.

3. Collaboration SUTs. Collaborative systematic review of the randomised trials of organised inpatient (stroke unit) care after stroke. Stroke Unit Trialists' Collaboration. Bmj 1997;314(7088):1151-9.

4. Database C. Organised inpatient (stroke unit) care for stroke. Cochrane Database Syst Rev 2007;17(4):CD000197.

5. Langhorne P, Trialists SU. The effect of different types of organized inpatient (stroke unit) care. Cerebrovasc Dis 2005;19(Supplement 2).

6. Langhorne P, Bernhardt J, Kwakkel G. Stroke rehabilitation. Lancet 2011;377(9778):1693-702.

7. Saka O, Serra V, Samyshkin Y, McGuire A, Wolfe CC. Cost-effectiveness of stroke unit care followed by early supported discharge. Stroke 2009;40(1):24-9.

8. Langhorne P, Holmqvist LW. Early supported discharge after stroke. J Rehabil Med 2007;39(2):103-8.

9. Fisher RJ, Gaynor C, Kerr M, Langhorne P, Anderson C, Bautz-Holter E et al. A consensus on stroke: early supported discharge. Stroke 2011;42(5):1392-7.

10. Fjaertoft H, Rohweder G, Indredavik B. Stroke unit care combined with early supported discharge improves 5-year outcome: a randomized controlled trial. Stroke 2011;42(6):1707-11.

11. Rousseaux M, Daveluy W, Kozlowski R. Value and efficacy of early supported discharge from stroke units. Ann Phys Rehabil Med 2009;52(3):224-33.

12. Pilkonis PA, Choi SW, Reise SP, Stover AM, Riley WT, Cella D. Item banks for measuring emotional distress from the Patient-Reported Outcomes Measurement Information System (PROMIS(R)): depression, anxiety, and anger. Assessment;18(3):263-83.

13. Gershon RC, Rothrock N, Hanrahan R, Bass M, Cella D. The use of PROMIS and assessment center to deliver patient-reported outcome measures in clinical research. J Appl Meas;11(3):304-14.

14. Gershon RC, Lai JS, Bode R, Choi S, Moy C, Bleck T et al. Neuro-QOL: quality of life item banks for adults with neurological disorders: item development and calibrations based upon clinical and general population testing. Qual Life Res;2011:27.

15. Perez L, Huang J, Jansky L, Nowinski C, Victorson D, Peterman A et al. Using focus groups to inform the Neuro-QOL measurement tool: exploring patient-centered, health-related quality of life concepts across neurological conditions. J Neurosci Nurs 2007;39(6):342-53.

16. Nowinski CJ, Victorson D, Cavazos JE, Gershon R, Cella D. Neuro-QOL and the NIH Toolbox: implications for epilepsy. Therapy;7(5):533-40.

17. Teasell R, Bitensky J, Salter K, Bayona NA. The role of timing and intensity of rehabilitation therapies. Top Stroke Rehabil 2005;12(3):46-57.

Meta-analyses of Existing Data in Stroke Rehabilitation

Teasell R, Foley N, Salter K, Bhogal S, Jutai J, Speechley M. The Stroke Rehabilitation Evidence Based Review: 13th edition. Canadian Stroke Network, October 2010, www.ebrsr.com

Pediatric issues in stroke recovery and rehabilitation

Beslow LA, Licht DJ, Smith SE, Storm PB, Heuer GG, Zimmerman RA, Feiler AM, Kasner SE, Ichord RN, Jordan LC. “Predictors of outcome in childhood intracerebral hemorrhage: a prospective consecutive cohort study.” Stroke. 2010 Feb;41(2):313-8. Epub 2009 Dec 17.

DeVetten G, Coutts SB, Hill MD, Goyal M, Eesa M, O'Brien B, Demchuk AM, Kirton A; MONITOR and VISION study groups. “Acute corticospinal tract Wallerian degeneration is associated with stroke outcome.” Stroke. 2010 Apr;41(4):751-6. Epub 2010 Mar 4.

Domi T, deVeber G, Shroff M, Kouzmitcheva E, MacGregor DL, Kirton A. “Corticospinal tract pre-wallerian degeneration: a novel outcome predictor for pediatric stroke on acute MRI.” Stroke. 2009 Mar;40(3):780-7. Epub 2009 Jan 8.

Ichord RN, Bastian R, Abraham L, Askalan R, Benedict S, Bernard TJ, Beslow L, Deveber G, Dowling M, Friedman N, Fullerton H, Jordan L, Kan L, Kirton A, Amlie-Lefond C, Licht D, Lo W, McClure C, Pavlakis S, Smith SE, Tan M, Kasner S, Jawad AF. “Interrater reliability of the Pediatric National Institutes of Health Stroke Scale (PedNIHSS) in a multicenter study.” Stroke. 2011 Mar;42(3):613-7. Epub 2011 Feb 11.

Kirton A, Deveber G, Gunraj C, Chen R. “Cortical excitability and interhemispheric inhibition after subcortical pediatric stroke: plastic organization and effects of rTMS.” Clin Neurophysiol. 2010 Nov;121(11):1922-9.

Taub E, Griffin A, Nick J, Gammons K, Uswatte G, Law CR “Pediatric CI therapy for stroke-induced hemiparesis in young children.” Dev Neurorehabil. 2007 Jan-Mar;10(1):3-18.

Additonal reading

1. Cramer SC, Sur M, Dobkin BH, O'Brien C, Sanger TD, Trojanowski JQ, et al. Harnessing neuroplasticity for clinical applications. Brain : a journal of neurology. 2011;134:1591-1609



Co-chairs: Steven Greenberg, Sandra Black, Philip Gorelick

Members: Cenk Ayata, Hugues Chabriat, Chelsea Kidwell, Jin-Moo Lee, Vincent Mok, David Nyenhuis, Bruce Reed, Gary Rosenberg, John Sled, Eric Smith

NINDS Liaison: Thomas Jacobs


Top Research Advances since 2007

Animal Models

  • New and better transgenic models of CADASIL and CAA
  • New experimental systems for studying chronic brain ischemia
  • New noninvasive neuroimaging tools, with potential for human translation

Novel Human Biomarkers

  • Incorporation of neuroimaging into VCI diagnostic criteria
  • Expanded focus on cerebral microbleeds and microinfarcts as key markers
  • ß-amyloid imaging for detection of plaque (AD) and vascular (CAA) amyloid
  • Newly identified genetic causes of hereditary VCI

Outcome Markers and Clinical Trial Design

  • Improved VCI trial outcome markers
  • New clinical trial data on VCI prevention
  • New data on symptomatic treatments applied to VCI

Opportunities, Emerging Topics, and Key Challenges

Animal Models

  • Translational potential of animal models still limited and unproven.
  • The SHR/SP spontaneously hypertensive/stroke prone rat may be underutilized.

Novel Human Biomarkers

  • Insufficient studies of VCI imaging across racial/ethnic/demographic groups.
  • Most clinically meaningful VCI imaging biomarkers still not established.
  • Emerging methods for measuring small vessel integrity/function and detecting chronic microinfarcts appear promising.

Outcome Markers and Clinical Trial Design

  • VCI prevention may be achievable with existing methods, but hard to demonstrate practically, e.g. if long-term treatment is required.
  • Potential strategies might be to perform VCI prevention trials using high-risk groups or to build them on top of ongoing acute stroke trials.

Priorities for Future Research

1) Animal Models: Develop animal models that mimic the full range of tissue damage of human VCI in the setting of common vascular risk factors.

2) Novel Human Biomarkers: Fully incorporate ß-amyloid imaging into studies of VCI and small vessel-related neuroimaging markers into studies of AD, with focus on unraveling the links between small vessel pathology, neurodegenerative disease, and neurologic impairment.

3) Outcome Markers and Clinical Trial Design: Develop large-scale clinical trials of dementia-free but high-risk subjects for intensive vascular risk factor modification. ?


Animal Models

1. New and better characterized transgenic mouse models of CADASIL and CAA have provided key clues to the pathophysiologic mechanisms of small vessel brain disease.

The reported CADASIL transgenic mouse models express varying disease-associated mutations in human NOTCH3 or rodent Notch3 (e.g., C428S, R169C, R1031C, C455R) with differing transgenic strategies, expression levels (50% to 400% of endogenous Notch3), presence or absence of endogenous Notch3, and study endpoints. Despite these methodological heterogeneities, the mice have reproduced some of the ultrastructural and functional hallmarks of CADASIL, such as age-dependent GOM and NOTCH3 ectodomain deposition at the vascular smooth muscle membranes and vasomotor dysfunction in cerebral and systemic resistance arteries. Across the mutant strains, the onset and extent of pathology appear to be related to expression level of the mutant transgenes, including (in Notch3 R169C) white matter degeneration associated with reduced capillary density and resting blood flow. Enhanced cortical excitability has also been demonstrated, possibly explaining the migraine with aura and seizure phenotypes associated with CADASIL.

Two of the cerebral amyloid angiopathy (CAA) mouse models (APP23 and Tg2576) with increased wild-type ß-amyloid peptide (Aß) production develop amyloid plaques, cerebrovascular amyloid, and microhemorrhages. Two other models (APPDutch and Tg-SwDI) that express mutant forms of Aß (Dutch and Iowa) develop CAA without amyloid plaques; APPDutch also develop microhemorrhages. Studies across mouse strains implicate a series of factors in determining amyloid deposition in vessels versus plaques: the Aß40 to Aß42 ratio, the perivascular drainage pathways for Aß clearance, and the additional catabolic role of perivascular macrophages. Other key advances to emerge from the transgenic models are improved understanding of downstream actions of vascular amyloid, including vascular dysfunction and attenuation of vasoreactivity (likely due to smooth muscle cell loss) and induction of matrix metalloproteinase (MMP) activity and inflammatory cascades. Attenuation of inflammation with minocycline in Tg-SwDI mice improved behavioral performance without affecting CAA accumulation. Further preclinical testing using CAA animal models and relevant endpoints may provide candidate molecules for human trials.

2. New candidate experimental systems for chronic brain ischemia have emerged.

The 2007 SPRG took note of the absence of a definitive animal model for the chronic ischemia that results from widespread small vessel disease. Several potentially important approaches have emerged in the interim. One is to generate chronic hypoperfusion by bilateral carotid occlusion (BCAO) in the rat. BCAO produces progressive injury to the white matter with gliosis, death of oligodendrocytes, disruption of blood-brain barrier (BBB), and cognitive impairment. An inflammatory reaction is generated in the damaged white matter, characterized by activated microglia and macrophages and increased MMP expression. Although BCAO cannot be reliably used in mice because of poor collateral circulation, placing a small coil around both carotids produces white matter lesions and behavioral impairments similar to those seen with rat BCAO. Transgenic mice can be studied with the use of the coils, making it an attractive model for mechanistic studies. The major drawback is the small amount of white matter in the mouse.

Since BCAO is typically performed on young normotensive rats, extrapolating to the elderly hypertensive population is difficult. A potentially more appropriate model is the spontaneously hypertensive/stroke prone rat (SHR/SP). These animals develop hypertension around 6 weeks of life that progresses over months, yielding changes in white matter similar to those in patients with subcortical vascular dementia. Various manipulations of diet and salt can accelerate progression of these white matter changes. SHR/SP have cognitive impairment reminiscent of that in VCI patients. Since there is sufficient white matter for imaging studies and immunohistochemistry correlation, SHR/SP represents a promising animal model for testing potential long-term treatments.

3. A plethora of new/emerging neuroimaging tools has emerged for detecting small vessel disease-related brain injury in living animals, with potential for direct human translation.

Imaging methods are of great value for studying animal models of VCI, both to address spatial distribution and to follow animals longitudinally. MRI in particular is non-invasive, allows for diverse contrast mechanisms, and readily visualizes white matter, vasculature, and deep brain structures. Growing availability of high-field animal magnets with optimized imaging protocols and hardware have made high quality animal MRI widely accessible.

Among available MRI applications for neuroimaging in animal models are 1) detection of microbleeds by susceptibility-weighted imaging (SWI), 2) use of high-resolution 3D protocols with post-processing techniques such as deformation-based morphometry for detection of structural changes (particularly for inbred mouse lines with low background variability), 3) sensitive detection of demyelination by protocols such as diffusion tensor and magnetization transfer, 4) measurement of metabolites with magnetic resonance spectroscopy, such as marking of axons by N-acetyl-aspartate, and 5) electron spin resonance in combination with oxygen sensitizing agents for mapping of free oxygen. A variety of imaging methods are also available for assessing cerebrovascular physiology. Among these are 1) visualization of BBB disruption by intravascular contrast agents, 2) blood oxygenation dependent contrast (BOLD) for measurement of functional activation (rare in mice because of limited spatial resolution), 3) measurement of cerebral perfusion under baseline or physiologically challenged conditions with exogenous contrast-based techniques such as dynamic contrast enhanced imaging or non-invasive methods such as arterial spin labeling, and 4) measurement of cerebral blood volume in vivo by MRI with intravascular paramagnetic iron particles or ex-vivo by micro-CT. Finally, a variety of methods are available for angiographic imaging of small animals. These include visualization of vessels down to 30 to 50µ diameter by in vivo contrast-enhanced MRI or CT or to finer resolution (e.g. 1µ) using ex-vivo methods such as synchrotron CT methods of intact whole brains.

Optical imaging methods paired with skull thinning or cranial window preparations are another rapidly emerging approach for animal models with cortical pathology sufficiently close to the brain surface. 3D multi-photon fluorescence imaging provides direct visualization of the cortical microvasculature and has been used to describe vascular changes in CAA models. Optical techniques can also be used to assess perfusion, functional hyperemia, and blood oxygenation. Optical intrinsic signaling (for blood oxygenation) and laser speckle coherence (for red blood cell velocity) capability can be incorporated into the same microscope used for multi-photon imaging. Optical techniques can effectively be combined with reporter gene technology to visualize other cell types, e.g. neuronal remodeling. All of these angiographic techniques can also be paired with post-processing methods (such as vessel tracking) for quantitative assessment of vascular architecture. Truly non-invasive optical techniques have the advantage of circumventing potential confounding effects of acute or chronic cranial surgery in small animals.

Novel Human Biomarkers

4. Neuroimaging has been explicitly incorporated as supporting evidence for the diagnosis of VCI

Criteria for the diagnosis of VCI proposed in the recent American Heart Association Scientific Statement (Gorelick et al Stroke 2011;42:2672) are the most noteworthy new research criteria for VCI since the 1993 publication of the NINDS-AIREN criteria. The new criteria state that “a clear relationship in the severity and pattern of cognitive impairment and the presence of diffuse, subcortical cerebrovascular disease pathology” can be used to support a diagnosis of Vascular Dementia or Vascular Mild Cognitive Impairment; however, operational biomarker criteria for “diffuse, subcortical vascular pathology” have not yet been defined. These diagnostic advances come in the context of growing appreciation, based on a wealth of neuropathology studies, confirming that cerebrovascular disease is a major contributor to the risk of cognitive impairment and dementia, often co-occurring with other brain pathologies such as Alzheimer disease (i.e., mixed dementia).

5. Cerebral microbleeds and microinfarcts have emerged as common VCI-associated lesions and markers of underlying small vessel pathology

Cerebral microbleed (CMB) detection via T2*-weighted gradient recalled echo (GRE) MRI is now considered a standard component in the evaluation of dementia, ischemic stroke and hemorrhagic stroke. More sensitive techniques such as 3D GRE, SWI, and high magnetic field strengths are available, though their clinical application is still unclear. Identification of CMB in the deep grey matter or brainstem suggests underlying arteriolosclerosis or hypertensive vasculopathy, whereas isolated lobar CMB are usually related to CAA, sometimes in association with Alzheimer disease (AD). Using clinical GRE techniques, CMB prevalence ranges from around 5-10% in normal elders (depending on age), 20-30 % in AD and ischemic stroke, and 60% in cerebral hemorrhage. Identification of CMB in normal elders appears to confer increased risk of future ischemic and hemorrhagic stroke.

Cerebral microinfarts (CMI), foci of brain infarction too small (diameter typically < 1mm) to be visible on gross inspection or conventional structural imaging, are harder to detect than CMB but may be substantially more numerous. Given that only a small area of brain is microscopically sampled in autopsy protocols, it is striking that CMI are detected in ~20 to 40% of elderly autopsied brains. Further, presence of these lesions appears to correlate with impaired cognitive performance during life, even after controlling for other important covariates such as age, sex, education, AD, Lewy bodies, and macroscopic infarcts. Detecting CMI during life is challenging because of their small size and lack of associated iron or other salient neuroimaging property. Acute, large CMI may be transiently visible as punctate diffusion-restricted lesions observed in the setting of advanced CAA, hemorrhagic or ischemic stroke, or CADASIL. An exciting advance in this field has been detection of pathologically proven cortical CMI by post-mortem high-field (7T) structural MRI. Given the likelihood that the few CMI found by random pathological sampling are just the “tip of the iceberg,” these clinically invisible lesions may be numerous enough to serve as key mediators of small vessel disease-related cognitive/ neurologic dysfunction.

6. ß-Amyloid imaging has opened a new window for identifying both Alzheimer and CAA pathology

Amyloid PET imaging using C11 or F18 ligands detects fibrillar amyloid, whether deposited in plaques or vessels. These techniques, together with cerebrospinal fluid biomarker measurements, have been explicitly incorporated into research criteria for the diagnosis of probable AD and pre-Alzheimer states. From the standpoint of VCI, the ability to detect Alzheimer pathology in vivo opens the important possibility of teasing apart various degrees of “pure” and “mixed” vascular processes. An analysis of 45 Korean patients meeting criteria for vascular dementia, for example, identified 14 with positive amyloid imaging and 31 with apparently pure vascular disease, the latter demonstrating more lacunar lesions, less hippocampal atrophy, and relatively better performance on memory tasks. Another multimodal study combined amyloid imaging, structural and functional MRI, using set-shifting and inhibitory control tasks that activated frontal-parietal attentional networks. The study found that white matter lesions but not amyloid burden correlated with failure to modulate attentional networks, whereas amyloid burden was associated with less ability to suppress the default mode network. These multimodal imaging approaches to teasing apart the contributions of small vessel and amyloid/neurodegeneration to cognitive impairment appear extremely promising.

PET detection of vascular amyloid in advanced CAA demonstrates a similar pattern as AD, though with relatively greater occipital signal. CAA represents a further potential contributor to VCI; for example, a systematic review of population autopsy studies found 55-59% CAA in dementia vs 37-43% in non-demented subjects. Another clinical rationale for noninvasive detection of CAA is the possible hemorrhage risk from antithrombotic agents.

7. New genetic causes of hereditary VCI and polymorphisms linked with sporadic small vessel disease have been identified.

Joining Notch3, the genetic source of CADASIL (the most common hereditary VCI), mutation of HTRA1 has been identified as the cause of CARASIL and mutations of COL4A1 for cases of intracerebral hemorrhage associated with white-matter lesions and other clinical manifestations. For sporadic small vessel disease, linkages have been established for chromosomes 4 and, in a GWAS meta-analysis large enough to allow for replication, for 6 gene candidates on chromosome 17.

Outcome Markers and Clinical Trial Design

8. Outcome markers for VCI have been identified and refined.

Our understanding of the cognitive and behavioral profile of VCI comes in the context of the very high prevalence of mixed vascular plus Alzheimer dementia. With some exceptions (e.g., CADASIL), small vessel cerebrovascular disease may have its largest effects on cognition when it occurs together with concurrent AD pathology. This conceptual framework predicts that cognitive profiles of VCI and AD will often overlap. There is now substantial evidence based on clinical and neuropathological studies that executive dysfunction, for example, may be as common in AD as in VCI and may occur together with impairment of episodic memory. VCI, conversely, may affect multiple cognitive domains. Thus, even when the primary consequences of vascular lesions affect executive abilities and cognitive speed, it may be difficult to differentiate VCI from AD. Vascular depression, defined by a constellation of features including deep white matter lesions (WML), executive dysfunction, and depressive symptoms, is increasingly recognized to overlap with VCI. Depression may be more common in VCI than in AD and can also involve other cognitive domains including episodic memory.

Scales designed to measure daily function in dementia, such as Disability Assessment for Dementia, Functional Rating Scale, and Functional Assessment Staging Tool, can be used to capture longitudinal decline in VCI. A newly developed scale designed to measure loss of daily function in specific cognitive domains, Everyday Cognition scale (ECog), has good psychometric properties, is sensitive to functional loss in mild cognitive impairment (MCI), and has been adopted for use in ADNI studies of MCI and AD. Recent advances have also led to new neuropsychological tests and protocols that are well suited to the study of VCI and are in various stages of development. These include test instruments from the NIH Toolbox Initiative, EXAMINER (Executive Abilities: Methods and Instruments for Neurobehavioral Examination and Research) project, NINDS VCI harmonization protocol, and the Montreal Cognitive Assessment (MoCA; mocatest.org).

9. Secondary analyses of clinical trials have explored the preventive effects of various interventions.

The cognitive effect of blood pressure (BP) lowering has been explored in several secondary analyses. The HYVET trial showed no difference in dementia incidence with BP lowering in persons ? age 80 over relatively short mean follow-up (2.2 years). PRoFESS also showed no cognitive benefit of telmisarten in ischemic stroke. Like HYVET, PRoFESS entailed relatively short mean follow-up (2.4 years) and a small BP difference (systolic / diastolic BP -3.8/-2mmHg) between treatment groups. Neither study was designed with cognition as primary endpoint. Although there is evidence that BP lowering is effective for reducing the risk of post-stroke dementia (e.g. PROGRESS), meta-analyses of BP lowering agents in persons with or without a history of cerebrovascular disease have been equivocal.

Two studies designed to reduce atherosclerotic risk in middle-aged or elderly persons showed no major primary beneficial effect on cognition with aspirin administration. In PRoFESS, aspirin plus extended-release dipyridamole was not superior to clopidogrel in prevention of post-stroke cognitive decline. Trials using statins among dementia-free elderly (PROSPER) or mild to moderate AD (LEADe, CLASP) failed to show benefits in slowing of cognitive decline. In the ROCAS sub-studies where MRI measures were used as surrogate markers, statins appeared to reduce progression of confluent WML and lacunar infarcts among stroke-free subjects with asymptomatic large artery disease over 2 years.

Among shorter-term clinical trials, FISS-tris showed that functional improvement with anticoagulant therapy was greater than in the placebo group at 6 months, possibly caused by better cognition. An exploratory study of an antidepressant suggested that selective serotonin reuptake inhibitor administration within 3 months of stroke improved cognition at 1 year.

10. Symptomatic treatments used in AD may also offer modest benefits in VCI.

Clinical trials of approximately 6-months of the acetylcholinesterase inhibitors galantamine, donepezil, and rivastigmine showed modest cognitive benefit of uncertain clinical relevance. These findings are concordant with meta-analysis of VCI trials of cholinesterase inhibitors and memantine. There may be subgroups that preferentially respond to cholinesterase inhibitors such as patients with hippocampal atrophy/mixed dementia or the large artery disease subtype of VCI.


Key Challenges

Animal Models

1. Despite impressive improvements in the breadth of available animal models, each has important flaws and drawbacks.

Key features of human VCI have remained elusive. It is notable, for example, that lacunar infarcts have not been observed to date in any of the transgenic CADASIL or CAA models. To model lacunar infarcts, stereotactic endothelin injections into the subcortical white or deep gray matter have created small discrete infarcts in internal capsule, corpus callosum or caudoputamen, an approach that reproduces the ischemic injury but not the true vascular pathology. Transgenic mouse models are also notoriously limited by their short lifespans, low volumes of white matter, and differences in the vascular/brain anatomy that preclude direct correlation between the pathology and cognitive deficits in mouse vs man. The experimental models of chronic ischemia suffer from limited reproducibility.

2. The ability of the noninvasive imaging methods to translate findings from animal models to human disease remains unproven.

Although translation to humans remains a distinct challenge in VCI (as throughout the stroke field), it is encouraging that many of the noninvasive imaging methods for animals have counterparts in human brain imaging with potential for direct application of data from bench to bedside.

3. Noninvasive imaging cannot yet detect most pathophysiologic pathways at the cellular or molecular level.

Among molecular and cellular pathways implicated in small vessel disease-related tissue injury are apoptotic death of oligodendrocytes, generation of intermediate inflammatory molecules such as cycloxygenases, MMPs, and free radicals, microglial activation, and increased expression of hypoxia-inducible factors. These pathways are generally detectable only by tissue sectioning and staining, precluding most longitudinal designs.

Novel Human Biomarkers

4. Vascular biomarker studies of VCI (e.g. CMB, CMI, WML, carotid wall thickness) characterized and contrasted by race have not been sufficiently pursued.

Several recent series suggest these imaging findings do in fact differ by race or demographic group. In patients with primary hemorrhagic stroke, for example, African Americans may have greater brain atrophy, a higher WML burden, and more CMB compared to Caucasians. Another study suggested that African Americans were more likely to have non-amnestic forms of MCI with MR imaging findings indicative of vascular disease. While these differences may arise from health care disparities or the higher prevalence of hypertension and other traditional vascular risk factors, they could also reflect as yet unknown environmental or genetic factors or contribution from non-arteriolosclerotic small vessel processes such as CAA. Further studies of racially/demographically diverse populations could help identify high-risk populations and reveal environmental or genetic influences not apparent in homogeneous samples.

5. Establishing the precise causal links between small vessel-related brain lesions and VCI is difficult.

This problem is particularly troublesome for biomarkers that are nearly ubiquitous with aging, e.g. WML (present to some degree in >90% of elders). Other imaging biomarkers, e.g. cortical or hippocampal atrophy, may correlate well with cognitive impairment, but be etiologically difficult to assign as due to neurodegenerative disease vs. vascular disease. The challenge, therefore, is to determine what biomarker patterns of small vessel disease are most clinically meaningful. Current studies tend to emphasize (or overemphasize) the volume and number of lesions rather than severity of underlying vascular pathology or anatomic location. From this standpoint, continuous measures of axonal damage such as magnetic resonance spectroscopy or diffusion tensor imaging (DTI) may have methodological advantages over “all or nothing” measures such as T2 or FLAIR hyperintensity. Further, the research criteria for the diagnosis of mixed dementia with a vascular contribution remain underdeveloped, and how the relationship between cerebrovascular disease and cognitive impairment is modified by AD and other brain pathologies remains undetermined. A recent successful approach to this question used DTI to correlate fractional anisotropy in regions of the corpus callosum with white matter lesions, grey matter volumes, and cognitive performance, with results suggesting differential region-specific associations of neurodegenerative and vascular pathologies. Further studies will be needed to develop and validate biomarker profiles to discriminate “malignant” small vessel disease causative of VCI from more benign patterns.

6. Studies of VCI biomarkers largely focus on the consequent tissue damage rather than the small vessel disease itself.

The affected cerebral vasculature is typically too small to visualize directly; the underlying pathology is therefore usually inferred from parenchymal injury (e.g. “non-specific” patchy WML vs. focal hyperintensities associated with hypointensities on T1 or FLAIR MRI). Although speculative, it is plausible to think that biomarkers of small vessel pathology might potentially allow for early detection of VCI risk before parenchymal injury occurs. A related defect in biomarker data is that most have emerged from single-center, cross-sectional studies. Replication and validation in larger, demographically diverse cohorts with longitudinal follow-up is essential.

Outcome Markers and Clinical Trial Design

7. Although VCI may well be preventable with existing strategies, such successful strategies might be very difficult to demonstrate in practically achievable clinical trials.

Among key questions difficult to address practically via randomized clinical trial: At what point in life (e.g., midlife, older age, very elderly) might it be optimal and feasible to treat vascular risk factors to prevent cognitive impairment or dementia? Epidemiological evidence suggests that some vascular risk factors elevate risk for late-life cognitive decline when present at midlife but not when present in late life. Designing clinical trials that can evaluate effects of risk reduction when the effects lag the exposure by decades is daunting. Does long-term control of vascular risk factors or long-term administration of aspirin or other antiplatelet therapy reduce the risk of cognitive impairment or dementia in a primary prevention setting? Does longer-term (i.e., longer than prior published trials) control of vascular risk factors or antiplatelet therapy reduce the risk of post-stroke cognitive impairment or dementia?

8. The cognitive scales typically incorporated as secondary cognitive measures in clinical trials (e.g. the Mini Mental Status Exam) are insensitive to mild degrees of cognitive impairment and to some cognitive domains such as executive function.

Preferable alternatives such as the Montreal Cognitive Assessment battery have been developed and come into increasing use.

Unrealized Opportunities

Animal Models

Given the limitations of mouse models, the SHR/SP spontaneously hypertensive/stroke prone rat may be an underutilized resource for understanding white matter changes associated with chronic hypertension.

A growing slate of noninvasive imaging tools (MRI-, CT-, and optically-based) are now available for systematic application to longitudinal analysis of the appearance and progression of small vessel-related brain injuries.

Novel Human Biomarkers

Candidate biomarkers of vascular integrity and function (e.g. vascular permeability imaging, retinal vascular imaging, blood flow regulation in response to brain activity, endothelial progenitor cell counts, matrix metalloproteinase activity) offer important potential complements to current studies of small vessel-related brain lesions and should be pursued.

Comparative studies of biomarkers of vascular disease across racial, ethnic, and demographic groups, analyzed in correlation with cognitive outcomes, post-mortem findings, and known genetic factors (e.g. Apolipoprotein E), offer substantial opportunities for elucidating how genetic and non-genetic inter-group differences affect cerebral vasculopathies and their interactions with other risk factors.

Recently reported findings with high-field MRI highlight the potential for noninvasive imaging of chronic cerebral microinfarcts and differentiating them from dilated perivascular spaces and other microcavities. If pathologically validated, such a noninvasive method would be a key tool for elucidating the epidemiology, causes, effects, and prevention strategies of these otherwise invisible lesions.

Outcome Markers and Clinical Trial Design

Identifying a high-risk vascular profile (based on clinical risk factors, imaging, or other biomarkers) predictive of subsequent cognitive impairment or dementia would be an important step towards improving the feasibility of large-scale prevention trials.

Acute stroke treatment trials offer an excellent opportunity for assessing the preventive value of early-phase interventions (as well as the predictive value of novel biomarkers) on post-stroke cognitive impairment or dementia.


1) Animal Models: Continue to seek animal models that mimic the range of tissue damage of human VCI (white matter lesions, lacunar infarcts, microinfarcts, microbleeds) in the setting of common vascular risk factors (age, hypertension, diabetes). Develop noninvasive tools for detecting the key molecular and cellular pathways and translating to human trials.

2) Novel Human Biomarkers: Fully incorporate ß-amyloid imaging into studies of VCI, and small vessel-related neuroimaging markers into studies of AD, recognizing that mixed cognitive impairment/dementia is very common. Focus future biomarker development on markers of small vessel disease severity and location, integrating these into multimodal studies of existing imaging markers, biochemical and genetic risks, and epidemiologic factors (including variations related to racial/ethnic/demographic group), aimed ultimately at unraveling the precise connections between vascular pathology, neurodegenerative pathology, and neurological impairment.

3) Outcome Markers and Clinical Trial Design: Develop large-scale clinical trials of dementia-free subjects with substantial small vessel disease for intensive vascular risk factor modification versus usual vascular risk factor modification for the prevention of cognitive impairment or dementia. Incorporate VCI biomarkers (e.g. structural MRI, novel vascular imaging methods, and ß-amyloid detection) as well as recent advances in neuropsychological and functional outcome measures into such trials.?


Last updated March 10, 2014