The Evolution and Future of Stroke Trials

Introduction

Randomized trials have revolutionized acute treatment, prevention, and recovery after stroke since their initial implementation in the 1960s. The first stroke trials included those of antihypertensive medications for primary stroke prevention,^{1 2} aspirin for secondary stroke prevention,³ and intravenous plasmin for patients with acute ischemic stroke and occlusion of the internal carotid or trunk of the middle cerebral artery on cerebral angiography.⁴ The success of stroke clinical trials has been highly dependent upon better understanding of the pathophysiology underlying brain ischemia and hemorrhage and development of drug and device-based therapies that can slow, halt, or reverse the pathologic processes. As importantly, advances in computed tomographic and magnetic resonance imaging of the brain and non-invasive and invasive imaging of the cerebrovascular tree were critical for development of the first treatment of ischemic stroke, recombinant tissue plasminogen activator (r-tPA), 27 years ago,⁵ and in the past decade, the successful implementation of endovascular therapy for acute ischemic stroke for patients with larger artery occlusions.^6,7 Advances in antithrombotic and anticoagulant medications, as well as diagnostic and treatment devices, have led to improved primary and secondary stroke prevention. While the past decade has seen several large stroke recovery trials, the advances in stroke rehabilitation has lagged behind the successes of stroke prevention and acute stroke therapies. Excitingly, trials of stroke recovery are poised to be at the forefront of stroke therapeutic advances in the next decades as our understanding of brain recovery and experience with larger multi-center trials increases. In this manuscript, the authors outline selected trials that drove innovation in stroke clinical trial methodology and outline the future evolution of clinical stroke trials.

The Ideal Clinical Trial

The basic question that clinical trials try to answer for patients and health care providers is whether a therapy has a positive biologic effect that either prevents disease or improves outcome when a disease occurs. The ideal clinical stroke trial tests a therapy clearly targeting the underlying pathophysiology, is administered within an appropriate time frame where brain function can be preserved or enhanced, has a sufficient sample size to ensure adequate power and minimize confounders and biases, includes all populations that may benefit, and is completed efficiently and expeditiously. Particularly, a trial population should include the patient population that is most affected by the disease, in terms of race, ethnicity, sex and age. Because many environmental, genetic and psychological factors can affect outcome and recurrence in individual patients, demonstrating efficacy of a therapy requires a reasonable balance of these important confounding factors between treatment and non-treatment groups, best achieved by randomization in clinical trials. Preclinical stroke models try to address these confounders by using genetically similar animals in tightly controlled laboratory environment. The National Institute of Neurologic Disorders and Stroke (NINDS) Stroke Pre-Clinical Assessment Network (SPAN) for testing potential acute stroke treatments in mice has tried to better model human variability by including male and female animals of various ages and genetic strains (hypertensive mice), and by conducting experiments in multiple laboratory environments, better mirroring the multi-center approach to clinical stroke trials.⁸ Hopefully, better design of preclinical trials may improve the pipeline of future therapies to be tested in patients with stroke.

Advances in Stroke Trials

Advances in the design of randomized trials have accompanied and accelerated advances in proven stroke therapies. Some of the early innovations in the design and analyses of randomized stroke trials have included double-blinding in the 1963 trial of plasmin for acute ischemic stroke,⁴ factorial design testing two medications alone or together in the secondary-prevention Canadian Aspirin Trial,³ the incorporation of the modified Rankin Score as the primary outcome measure in many acute stroke trials with categorical, ordinal, and sliding dichotomy statistical methodologies,⁹ step-forward randomization to minimize time to treatment in the NINDS t-PA Stroke Trials of recombinant tissue plasminogen activator (rt-PA),¹⁰ and pooled analyses of the initial rt-PA trials in 2004.¹¹

Major advances in stroke clinical trials over the past 10+ years include national and global clinical trial infrastructures;^12,13 advances in trial design such as adaptive trials (DAWN: DWI or CTP Assessment with Clinical Mismatch in the Triage of Wake-Up and Late Presenting Strokes Undergoing Neurointervention with Trevo,¹⁴ DEFUSE 3: Endovascular Therapy Following Imaging Evaluation for Ischemic Stroke 3,^15,44 ENRICH: The Early Minimally Invasive Removal of Intracerebral Hemorrhage,¹⁶ MOST: Multi-Arm Optimization of Stroke Therapy,¹⁷ registry-based trials (AcT: Intravenous tenecteplase compared with tissue plasminogen activator for acute ischemic stroke in Canada),¹⁸ stepped-wedge cluster trials (INTERACT 3: The Third Intensive Care Bundle with Blood Pressure Reduction in Acute Cerebral Haemorrhage Trial),¹⁹ pre-hospital acute trials (e.g. FAST-MAG: The Field Administration of Stroke Therapy–Magnesium,²⁰ RACE-CAT: Transfer to the Closest Local Stroke Center vs Direct Transfer to Endovascular Stroke Center of Acute Stroke Patients With Suspected Large Vessel Occlusion in the Catalan Territory,²¹ and RIGHT-2: Prehospital Transdermal Glyceryl trinitrate in Patients with Presumed Ultra-acute Stroke Trial),²² pragmatic trials (SATURN: Statins in Intracerebral Hemorrhage Trial - https://classic.clinicaltrials.gov/ct2/show/NCT03936361, Early versus Late initiation of direct oral Anticoagulants in post-ischemic stroke patients with atrial fibrillation-ELAN)²³; minimization of time from stroke onset to randomization in acute trials using mobile stroke units (BEST MSU (Mobile Stroke Unit)²⁴, Berlin, Germany MSU Study,²⁵ Comparison of tenecteplase with alteplase for the early treatment of ischaemic stroke in the Melbourne Mobile Stroke Unit- TASTE-A),²⁶ direct admission to angiography suite for brain imaging,²⁷ and increasing use of emergency consent and exception from informed consent (EFIC) in acute stroke trials.^27–29 Other advances include increased emphasis on stroke trials in children and infants (TIPS: Thrombolysis in Pediatric Stroke,³⁰ I-ACQUIRE: A Multi-site RCT of Intensive Infant Rehabilitation,^13,31 FOCUS: Focal Cerebral Arteriopathy Steroid Trial)³²; transition in stroke recovery trials from many single center trials to larger multi-center trials with standardized interventions (e.g. TELEREHAB Trial: Efficacy of Home-Based Telerehabilitation versus In-Clinic Therapy After Stroke in Adults);³³ Efficacy and safety of very early mobilisation within 24 h of stroke onset-AVERT),³⁴ patient engagement in study design and implementation (I-ACQUIRE)³¹; and globalization of trials.³⁵

National Stroke Trial Infrastructures

National infrastructures for stroke trials supported by government funding have revolutionized stroke trials in the past 20 years. Every Phase II and especially Phase III multi-center trials require a robust infrastructure to be successful but building an infrastructure de novo to initiate and implement each new trial, such as the North American Symptomatic Carotid Endarterectomy (NASCET) Trial,³⁶ takes substantial time and resources. Having an existing infrastructure makes initiation of new trials more rapid and implementation more efficient. As an example, the first NIH funded stroke network was the Specialized Programs of Translational Research in Acute Stroke (SPOTRIAS, 2000–2013),³⁷ although it focused on Phase II trials of acute stroke at only 8 regional centers. The NIHR Stroke Research Network was established in 2005 and operated across England as one of six topic (disease- related) research networks including stroke until 2014 when they were merged into a single comprehensive clinical research network structure with speciality leads (Gary Ford – written communication 12-14-2023). In 2013, the NIH expanded the scope and size of the successful SPOTRIAS network into the NIH StrokeNet (2013- present)^12,13,37 which included 25 (now 27) regional centers, eventually 500+ potential recruitment sites, a National Coordinating Center at the University of Cincinnati, a National Data Management and Statistical Center at the Medical University of South Carolina, a central institutional review board, a centralized research pharmacy, and imaging and educational cores. The NIH StrokeNet went from 0 trials in 2014 to 20 phase 2 and 3 trials and ancillary studies by 2023 that included acute, prevention and recovery trials. The Canadian Stroke Trials for Optimized Results (CaSTOR) network provided clinical trial infrastructure for a 5-year period from 2015–2020) and was made possible by a grant from Canadian Institutes of Health Research (CIHR) Institute of Circulatory and Respiratory Health (ICRH) Community Development Grant program (Andrew Demchuk, written communication, 12-14-2023). A number of nationally-supported networks have also emerged in other countries over the past 10 years. The Global Alliance of Independent Networks focused on Stroke trials (GAINS - https://www.globalstroketrials.org/) was created to facilitate, strengthen and nurture collaboration across these stroke trial networks and stroke trials worldwide. GAINS has been a valuable forum for data sharing, consideration of proposed trials, and global expansion of proposed and ongoing trials.

Enhancing Recruitment, Retention, and Globalization of Stroke Trials

Enrollment of participants is the lifeblood of stroke trials and is the rate-limiting step for testing an exciting array of new therapies to improve outcomes of stroke patients.^{28, 38,39} Ongoing approaches to enhance recruitment include greater inclusion of elderly participants and those with some pre-existing disability, technological tools such as remote/electronic consent to mirror the changing clinical work-flow of frequently evaluating participants via telemedicine, increasing use of emergency consent in acute stroke trials, incorporation of study design into standard work-flow as much as possible, and the increased expectation by physicians and participants that clinical research should be a standard consideration for every patient in all clinical settings. However, training, growing, and maintaining a workforce of clinical research professionals and new clinical trial investigators will be critical to match the pace of clinical trials with the number of emerging therapies as well as to insure the diversity of participants.^{28, 40,41}

Finally, stroke trials are becoming increasingly global which enhances the ability to recruit the needed number of participants more quickly and provides greater generalizability of study findings. Certain countries may have greater numbers of stroke subtypes, such as intracerebral hemorrhage in Asia. For example, in the ongoing rFVIIa for Acute Hemorrhagic Stroke Administered at Earliest Time (FASTEST) trial^28,29 that enrolls ICH patients within 2 hours of onset, 14 sites in Japan without emergency consent have enrolled 136 participants as compared to 148 participants in 65 sites with associated MSUs in Canada, Spain, United Kingdom, and the U.S. (personal written communication – Joseph Broderick – PI of FASTEST Trial, 10/17/2023). The major challenges for globalization of investigator-initiated stroke trials include the complexity of clinical trial agreements that involve data sharing across national boundaries, legal jurisdiction of the trial agreements, and co-funding of important trial questions by multiple countries. Pharmaceutical and device companies that have a legal and physical presence in multiple countries are often able to accomplish industry-sponsored multi-centered trials more expeditiously. But the coronavirus disease (COVID) global pandemic and the coordinated global response of vaccinations and trials of various therapies demonstrate that these barriers can be overcome when there is political will and sufficient financial resources.⁴² As stroke therapies become increasingly more effective, we will need global studies with large sample sizes to detect smaller and smaller benefits. Ideally, with increasing sample sizes these trials will be appropriately powered to detect differences in treatment effects according to sex and age.

Novel Trial Methodology

In addition to ever improving therapies and diagnostic testing, important advances in trial methodology have revolutionized stroke trials. The most notable advances in trial methodology in stroke include adaptive designs. Such trial adaptations have come in various forms. For example, adaptive randomization (e.g. SHINE: Stroke Hyperglycemia Insulin Network Effort Trial⁴³ and MOST: Multiarm Optimization of Stroke Thrombolysis¹⁷) increases the chances that the participant will be randomized to a better performing arm which is arguably a more ethical way to conduct a three-arm trial. A trial can adapt their intended population based on pre-specified rules, for example DEFUSE 3⁴⁴ and ENRICH,¹⁶ to maximize detection of treatment effects. Adaptations using interim analyses allow trials to conclude based on pre-specified futility or efficacy rules to maximize finite research resources use (e.g. ARCADIA: AtRial Cardiopathy and Antithrombotic Drugs In prevention After cryptogenic stroke Trial⁴⁵ and MOST¹⁷). Adaptive seamless phase 2–3 trial designs also serve a similar purpose by “weeding out” ineffective early therapies and allowing sample size benefits in phase-3 for the promising therapies.

Stroke trials have increasingly used pragmatic design components. For example, the BEST-MSU²⁴ trial used pragmatic approach in defining enrollment population by including all participants with suspected stroke, increasing the generalizability of the findings. Comparative effectiveness trials, that compare two treatment strategies otherwise widely used in clinical practice, use a pragmatic approach to study implementation. This approach has become common in stroke prevention trials (SATURN - https://classic.clinicaltrials.gov/ct2/show/NCT03936361, ELAN: Early versus Later Anticoagulation for Stroke with Atrial Fibrillation Trial,²³ ASPIRE: Anticoagulation for Stroke Prevention and Recovery After ICH Trial - https://classic.clinicaltrials.gov/ct2/show/NCT03907046). AcT trial used a pragmatic, registry-based approach to data collection, minimizing the burden of research procedures on coordinators by de-duplicating the work done as routine clinical care.¹⁸ Harmonizing the data elements across the trials, for example by using the NIH Common Data Elements, will further increase efficiency of data collection and allow comparisons and pooled analyses across trials.⁴⁶ Using clinically available electronic-medical record data and learning healthcare systems as sources of clinical trial data are pragmatic trial approaches in the broadest sense but remain under-utilized in stroke.

Another important innovation is the mandatory use of single institutional review board by the federally funded multi-center research.⁴⁷ While there is no published data about efficiencies, the authors’ anecdotal experience suggests that most of the efficiency is gained during the protocol amendments and continuing reviews.

Other innovations, such as the use of telephone/video follow-up, allowing study personnel travel to participant’s residence, wearable technology, and central outcome assessments may improve trial retention. While some of these innovations are being increasingly utilized, stroke prevention trials that require longer follow up will particularly benefit from increased adoption.

Future trials in stroke, following the example of trials in oncology, community-acquired pneumonia,⁴⁸ treatment of COVID-19,⁴⁹ and pain,⁵⁰ will utilize multiple approaches mentioned above in parallel. For example, the NIH StrokeNet Thrombectomy Platform trial https://reporter.nih.gov/search/jfB3EbaWqEmbvuDKotY7QA/project-details/10547985, and the proposed AcT GLOBAL Platform (Bijoy Menon – written communication – 10-12-2023) are randomized, multifactorial, adaptive platform trials. STEP uses registry-leveraged data collection using the AHA Get With The Guidelines and Neurovascular Quality Initiative Quality Outcomes Database registries. The platform approach will not only drastically decrease the time and cost of translating novel therapies into patient care but will also provide insight into real-world implications by allowing testing of multiple therapies in combinations (using multifactorial design).

Minimization of Time to Randomization in Acute Trials

Time from participant presentation to randomization remains an important challenge for acute stroke therapies, where time is critical for efficacy. Progress made since the approval of r-tPA 27 years ago^5,51 to overcome this challenge will serve as a launching platform for future trials that seek to minimize the time to randomization/treatment administration.^20,24 First, acute stroke trials have increasingly strived towards embedding the clinical trial processes within the acute stroke clinical workflow. By minimizing additional steps such as specialized imaging (e.g. The Efficacy and Safety of Thrombectomy in Stroke with extended lesion and extended time window or TENSION trial⁵² required only non-contrast brain CT prior to enrollment), laboratory testing (e.g. AcT trial only required standard of care lab studies prior to rt-PA/tenecteplase administration), and clinical examination, participants can be randomized and experimental therapy can be initiated promptly.

The second innovation is the process of informed consent and participant-investigator interactions. Trials have incorporated innovations such as focused consent form with an additional participant information sheet (e.g., MOST),^17,53 that allow reduction in the time to consent a participant while maintaining all essential components of informed consent in a succinct, short consent form. A separate, comprehensive participant information sheet can be reviewed by the participant when they are not pressed for time. Emergency/deferred consent in countries outside of the U.S. and exception from informed consent in the U.S. (e.g. FAST-MAG,²⁰ FASTEST trial²⁹) are aimed towards initiating the experimental therapy as soon as possible in the participant’s best interest if all criteria for such approaches are met. The latter is likely to serve as a learning model for future acute, pivotal trials in stroke within the US and globally. The use of telemedicine to expedite the interaction between investigator and participant/legally authorized representative and use of a remote consent form (sent via email or text message) are other options to facilitate enrollment within acute stroke trials, particularly after standard business hours.²⁸

The last but important innovation in acute stroke trials is bringing research to the participant’s doorstep. Trials using the mobile stroke units have revolutionized our ability to enroll stroke patients in the ultra-early time window.²⁴ Being able to administer study medication within this “golden hour” following onset of stroke symptom may substantially contribute to the futility or success of many acute stroke therapies.

Stroke trials in Children and Infants

Clinical stroke trials in the pediatric population have been challenging given the rare incidence of strokes in children after the neonatal period, the different mechanisms of strokes in infants and children than adults, the limited number of pediatric centers with clinical trial experience, and the limited experience of multi-center stroke trials in the pediatric community. But the expertise in pediatric stroke and the number of trained clinical investigators has grown over the past 10–15 years with increasing standardization of imaging and clinical assessment tools like the pediatric NIH Stroke Scale. Most importantly, in the U.S., the NIH StrokeNet has provided a platform for pediatric stroke trials with the I-ACQUIRE and FOCUS Trials underway.¹³ In particular, stroke recovery trials in infants and children are poised to make substantial advances in the near future.

Multi-center Stroke Recovery Trials

Successful planning and execution of larger multi-center trials of stroke recovery has increased dramatically over the past 5–10 years but is still early in its evolution.⁵⁴ These include constraint therapy for the affected upper extremity⁵⁵, fluoxetine for stroke recovery (FLAME: Fluoxetine for Recovery After Acute Ischaemic Stroke Trial,⁵⁶ FOCUS: Fluoxetine Or Control Under Supervision Trial,⁵⁷ EFFECTS: The Efficacy oF Fluoxetine—a randomisEd Controlled Trial in Stroke)⁵⁸, timing of rehabilitative therapy (AVERT: A Very Early Rehabilitation Trial After Stroke)⁵⁹, telerehabilitation strategies (TELEREHAB)³³, vagal nerve stimulation⁶⁰, and early phase and mechanistic trials to enhance the understanding of stroke recovery (TRANSPORT-II: Transcranial Direct Current Stimulation for Post-Stroke Motor Recovery Trial - https://classic.clinicaltrials.gov/ct2/show/NCT03826030 and VERIFY: Validation of Early Prognostic Data for Recovery Outcome after Stroke for Future, Higher Yield Trials - https://clinicaltrials.gov/study/NCT05338697?cond=stroke&term=verify&rank=1#locations). The knowledge gained in these trials will greatly inform the design of future trials of recovery and rehabilitation after stroke, one of the biggest future opportunities for stroke trials to improve long-term outcome of patients after stroke

Our understanding of how brain cells and neuronal networks interact with another and recover or take on new functions after injury lags substantially behind our understanding of why ischemic and hemorrhagic stroke occurs and how to prevent or remove clots in brain arteries or slow growth of ICH or remove hemorrhage in the brain. But as our understanding increases, new treatments for stroke recovery will almost certainly emerge from well-designed clinical trials. Training and increasing the number of multi-disciplinary research teams of physicians, therapists, imagers, and basic scientists in the design and execution of clinical trials will be critical for their success. Sharing expertise from acute and prevention trialists with current and future leaders in the recovery area is already ongoing in several of the national stroke networks.

Stakeholder Engagement in Clinical Trials

Trials in the past decade have increasingly utilized engagement of key stakeholders, particularly patients, as an important tool for various trial aspects, ranging from study design to recruitment, retention, and results dissemination. Examples of these trials, some with patients or stakeholders in leadership roles, include BEST-MSU,²⁴ the ongoing C3FIT Study: Coordinated, Collaborative, Comprehensive, Family-based, Integrated, Technology-enabled Stroke Care https://classic.clinicaltrials.gov/ct2/show/NCT04000971, TESTED: Treatment with Endovascular Intervention for STroke Patients with Existing Disability Trial -https://clinicaltrials.gov/study/NCT05911568?cond=stroke&term=TESTED&rank=1, and I-ACQUIRE.^13,31 Research with stakeholder engagement is meaningfully informed by clinician and patient preferences and is likely to result in wider implementation compared to research done without such engagement. With increasing awareness regarding patient-engaged research, the NIH StrokeNet formed a patient representative committee to encourage such patient-engagement in NIH/NINDS-funded stroke trials. Other avenues of stakeholder engagement can include, but are not limited to, engagement of key clinicians, nurses, medical and paramedical staff, funders, caregivers, and policymakers.

Future Challenges and Opportunities

The biggest challenge for future stroke trials is the unprecedented progress in effective therapies for the acute treatment and prevention of stroke over the past 25+ years and the incorporation of this progress into standard practice. Every potential additional benefit from new therapies is likely to be smaller and smaller and require larger and larger sample sizes to advance treatment. As an example, there has been no substantial advance in the already effective treatment of acute myocardial infarction over the past two decades despite trials with tens of thousands of participants with multiple promising therapies.

Comparative effectiveness trials seeking to find optimal management “strategies” for specific patient populations will remain important component of future trials. Acute ischemic stroke trials will likely focus on methods to minimize time to treatment or to buy more physiologic time prior to reperfusion (the holy grail of neuro-protective therapies). Acute medical and surgical treatment of intracerebral hemorrhage may have more immediate opportunities with only some very recent evidence of the first effective treatments as compared to the major advances in ischemic stroke over the past 25 years. Subarachnoid hemorrhage due to ruptured intracranial aneurysms, a much less common disease than other stroke subtypes, represents another opportunity going forward even though there have been limited advances over the past 10–15 years.

Improved and aggressive management of the “four horsemen of the Apocalypse”⁶¹ risk factors for stroke (hypertension, smoking, diabetes, and excess cholesterol), in addition to better exercise and the Mediterranean diet, may make trials of interventional approaches to stenoses of extracranial and intracranial arteries more challenging to demonstrate clear benefit. The dramatic benefits seen with the new anti-diabetic, weight loss medications such as semaglutide, tirzepatide, retatrutide, including recent reports of decreased risk of cardiovascular and cerebrovascular events,⁶² is likely to spark additional research for stroke prevention over the coming decade.

Beyond anticoagulation for atrial fibrillation and multiple anti-platelet agents in patients with mild recent stroke, antithrombotic trials for stroke prevention have struggled to identify new populations that can benefit from new antithrombotic treatments. The use of interventional devices for closure of the atrial appendage in participants who are poor candidates for anticoagulants⁶³ and closure of patent foramen ovale in participants with a high risk of paradoxical emboli⁶⁴ are major advances over the past 10 years. Ongoing trials will further define the populations of patients most likely to benefit from these interventions. Fundamental to finding new effective prevention treatments will be understanding better the specific pathophysiology for a given patient and developing treatments that address the pathophysiology. Regardless, because we will be looking for smaller and smaller benefits and require much larger sample sizes, the need for global trials and pooled data from multiple trials will be needed to advance stroke therapy.

But the biggest future opportunity for stroke trials lies in trials of recovery and rehabilitation after stroke. If the last 25–30 years has been the golden age of stroke prevention and acute stroke treatment, the next 100 years will be the century of recovery after stroke and brain injury. The future is bright, and we look forward to the next generation of innovative, practical, and efficient clinical trials led by talented and collaborative global investigators.

Randomized trials provide information regarding the overall average benefit and safety of a treatment across selected populations. However, within a population, treatment effects can vary considerably. For example, in a post-hoc analysis in women, carotid endarterectomy may reduce the risk of stroke by 72% if done within two weeks of a TIA but have no or small benefit if done more than two weeks after a TIA.⁶⁵ Subgroup analyses attempt to uncover heterogeneous treatment effects like this one, but an individual patient in a clinical setting has many characteristics that can determine outcome and efficacy of the treatment in question. Knowing the treatment effect on an individual patient becomes difficult when one patient fits into multiple subgroups. Artificial intelligence and machine learning can be used to overcome this limitation to better estimate treatment effects in individual patients.⁶⁶ One way to do this is by comparing the characteristics of a new patient with those of the participants enrolled in all of the relevant trials.⁶⁷ These models can weight outcomes from participants similar to your patient more heavily than those who are less similar. This ultimately can help determine the treatment approach with best chances of success for a given patient.

In addition to the impact of artificial intelligence, technology will continue to have a major impact on the future of stroke clinical trials. Examples include neuromodulation devices in stroke recovery, robotic technology as part of endovascular intervention, ongoing advances in automated imaging software for identification of ideal trial population, and hand-held applications for recruitment, retention, and selection in clinical trials, etc.

In summary, with the exponentially increasing number of therapies to be tested, efficient trial designs that weed out ineffective therapies will be increasingly used. These include phase IIb-III and futility designs, multifactorial designs that allow simultaneous testing of therapies while accounting for interactions, and comparative effective trials allowing head-to-head comparison of competing approaches. Platform designs will allow multiple questions to be answered simultaneously given that number of participants available to answer the increasing number of clinical questions will remain limited. Advances in artificial intelligence and technology will drive new treatments and their implementation. Randomized stroke trials have transformed our standard of care over the last 25 years and will continue to do so into the next century.

Non-standard Abbreviations and Acronyms:

ICH

Intracerebral hemorrhage

EFIC

exception from informed consent

rt-PA

Recombinant tissue plasminogen activator