Toward Wisdom From Failure
Lessons From Neuroprotective Stroke Trials and New Therapeutic Directions
Background— Neuroprotective drugs for acute stroke have appeared to work in animals, only to fail when tested in humans. With the failure of so many clinical trials, the future of neuroprotective drug development is in jeopardy. Current hypotheses and methodologies must continue to be reevaluated, and new strategies need to be explored.
Summary of Review— In part 1, we review key challenges and complexities in translational stroke research by focusing on the “disconnect” in the way that neuroprotective agents have traditionally been assessed in clinical trials compared with animal models. In preclinical studies, determination of neuroprotection has relied heavily on assessment of infarct volume measurements (instead of functional outcomes), short-term (instead of long-term) end points, transient (instead of permanent) ischemia models, short (instead of extended) time windows for drug administration, and protection of cerebral gray matter (instead of both gray and white matter). Clinical trials have often been limited by inappropriately long time windows, insufficient statistical power, insensitive outcome measures, inclusion of protocol violators, failure to target specific stroke subtypes, and failure to target the ischemic penumbra. In part 2, we explore new concepts in ischemic pathophysiology that should encourage us also to think beyond the hyperacute phase of ischemia and consider the design of trials that use multiagent therapy and exploit the capacity of the brain for neuroplasticity and repair.
Conclusions— By recognizing the strengths and limitations of animal models of stroke and the shortcomings of previous clinical trials, we hope to move translational research forward for the development of new therapies for the acute and subacute stages after stroke.
Neuroprotection for Stroke: A Fantasy Invented by Basic Scientists?
The quest for effective stroke treatments remains an urgent priority. A stroke occurs every 53 seconds in North America,1 and by 2020, cerebrovascular disease is projected to become the fourth-leading burden of disease worldwide, after heart disease, depression, and motor vehicle collisions.2 According to the summary of historical trends in clinical stroke trials by Kidwell et al,3 the 20th century saw the publication of 178 controlled trials of acute stroke therapies in the English-language literature, yet only a few produced “positive” results: the Neurological Institute of Neurological Disorders and Stroke (NINDS) rt-PA trial,4 Prolyse in Acute Cerebral Thromboembolism (PROACT II),5 a low-molecular-weight heparin trial,6 and most recently a trial of ancrod.7 The successful translation of these “vascular approaches” from the animal laboratory to the hospital emergency room has demonstrated that stroke is a treatable disorder in the hyperacute stage and has provided optimism that additional therapies to improve stroke outcome will be possible in the future.
In contrast, neuroprotective drugs that aim to salvage ischemic tissue, limit infarct size, prolong the time window for reperfusion therapy, or minimize postischemic reperfusion injury or inflammation have shown great promise in preclinical testing but disappointment in clinical trials.8–10⇓⇓ Of >49 neuroprotective agents studied in >114 stroke trials, none has proven successful clinically.3 Similarly, neuroprotective therapy has been unsuccessful in clinical trials of head trauma.11–14⇓⇓⇓ With the failure of so many trials, some clinicians may ask, “Is neuroprotective stroke therapy just a fantasy invented by basic scientists?” Will it ever play a clinical role? The answer is unclear. It must be acknowledged that neuroprotection may never be effective for promoting functional recovery after brain injury in humans. However, a lack of evidence of efficacy does not necessarily mean a lack of efficacy. Because of problems in basic science experiments and in clinical trial design, the evidence against neuroprotection is not conclusive, as is discussed below.
The purpose of this article is to provide clinicians and investigators with an up-to-date summary of the complex issues involved in translating stroke-related research from bench to bedside and to argue that experience from recent trials can provide important lessons that can be translated from the bedside back to the bench. Part 1 reviews pitfalls that have arisen in the development of neuroprotective therapies and reinforces recent recommendations regarding preclinical and clinical evaluation of new drugs.15,16⇓ Part 2 highlights emerging concepts in ischemic pathophysiology that should encourage us to think beyond the hyperacute phase of ischemia and consider the design of trials that use multiagent therapy and exploit the capacity of the brain for neuroplasticity and repair.
Part 1: Reconciling the Results of Negative Stroke Trials
Reasons for the failure of so many neuroprotective agents in clinical trials, despite their apparent benefit in animal (mostly rodent) models, have been the subject of intense discussion recently.15–31⇓⇓⇓⇓⇓⇓⇓⇓⇓⇓⇓⇓⇓⇓⇓⇓ It is becoming clear that existing animal models of focal cerebral ischemia are an imperfect representation of human stroke and may be relevant only to a minority of human stroke types.32 Neff33 reminds us that the lissencephalic brain of a rat is about the size of a lacunar infarct in humans, and some humans have infarcts the size of an entire rat. In addition, despite significant similarities between the rodent and human genomes, the differences that do exist are sufficient to remind us that conclusions reached regarding genomic and proteomic characteristics in rodent studies may not apply to human stroke.
Neuroanatomical, pathophysiological, pharmacokinetic, and genetic differences between rodents and humans notwithstanding, there has been a fundamental “disconnect” in the way that the efficacy of putative neuroprotective agents has been assessed in animal studies compared with clinical trials. The dramatic rise in the number of stroke trials over the past several decades has been accompanied by an improving, yet still highly variable, quality in study design.3,34⇓ Some trials could not have been expected to succeed because of conceptual or methodological flaws. Perhaps we have abandoned efficacious treatments prematurely on the basis of results of flawed trials. Other trials have been criticized, in retrospect, for proceeding on the basis of insufficient evidence of efficacy in preclinical studies (eg, only 50% of published animal studies were in favor of nimodipine).35 Other studies, still, have raised safety concerns because of drug toxicity, a danger of accelerating research by combining phase II and phase III trials.36–38⇓⇓
In this section, we highlight 9 pitfalls that have arisen in trying to extrapolate from animals to humans in the investigation of neuroprotective therapy. By understanding how animal models may be made more relevant to human stroke and how the design of clinical trials may be improved, we can move forward translational research for the development of stroke therapies.
Pitfall 1: Preclinical Studies Have Used Very Short Time Windows for Drug Administration, Whereas Clinical Trials Allow Longer Time Windows
Most neuroprotective studies in animals have relied on drug administration either before the ischemic insult or very soon after the onset of ischemia.39,40⇓ In contrast, time windows for entry in acute stroke neuroprotective trials have been longer and highly variable; in studies published between 1995 and 1999, the median time to entry was12 hours (range, 4 hours to 12 days), with a median time to treatment of 14 hours.3 None of the published neuroprotective trials has used a 3-hour window.3,16⇓ Treatment within 3 hours would be expected to have a greater chance of efficacy because more patients would be expected to have potentially reversible ischemic tissue.40–42⇓⇓ The effects of neuroprotective agents in the laboratory are even more time dependent than thrombolytics, leading Jonas et al19 to summarize the failure of neuroprotective trials as a matter of “too little, too late.” Certain “failed” drugs could potentially have clinical value if given at earlier time periods (within 2 hours after ischemia or prophylactically).19
On the other hand, Baron et al43,44⇓ and Fisher et al45 emphasize that we need not stipulate a fixed time limit for neuroprotective therapy because the duration of the ischemic penumbra is highly individualized. For example, PET studies suggest that the window of opportunity may be extended in some patients46; in 1 study, about one third of patients still had evidence of penumbra when assessed at 5 to 18 hours (mean, 10 hours) after stroke onset.44 The PROACT II results further support the fact that salvageable tissue is present up to 6 hours after onset in some patients.5 The rate of progression of the penumbra from reversible to irreversible ischemic injury is dependent on many factors and may be accelerated in the presence of poor collateral circulation, hyperglycemia, and other exacerbating factors.45
Putative neuroprotective drugs should not be advanced into clinical stroke trials until preclinical studies have investigated their effects when administered many hours, not minutes, after ischemia. Clinical trialists must aim for the shortest possible door-to-needle times, particularly given the tendency that physicians have of “waiting until the last minute” of the time window to treat, regardless of when the patient arrives at hospital.47 The NINDS rt-PA study showed that enrollment within 3 hours can be achieved, although 17 324 patients were screened to recruit the 624 subjects eligible for the study, with most excluded because of the time window.4 With increasing public awareness of this issue and improvements in regional organization of stroke services and stroke teams, response times are improving, and at some centers, >25% of patients are reaching hospital within 3 hours.48 Clinical trial protocols should enforce benchmarks for door-to-needle times and stratify patients by time of treatment with appropriate power calculations. If the long time windows are a major reason for the lack of efficacy of neuroprotective therapy in human trials, then the investigation of agents given prophylactically (eg, before surgical procedures with an increased risk of ischemic cerebrovascular events)49–52⇓⇓⇓ or by paramedics in the field (a phase I trial is already underway) may provide the necessary “proof-of-principle” data that are much needed as long as a strict target population can be defined by preplanned posthoc analysis.
Pitfall 2: Preclinical Studies Target the Ischemic Penumbra, Whereas Clinical Trials Do Not
As Fisher,45 Baron,53 and others54 have emphasized, the target of current neuroprotective therapy is the penumbra, ischemic tissue that is functionally impaired but whose damage is potentially reversible.55,56⇓ If reversible ischemic tissue is not present at the time of treatment, then neuroprotective therapy cannot be expected to work. Perhaps we have discarded some agents prematurely because clinical trials have not been selective enough in targeting patients with evidence of penumbra. Future trials may need to use stricter entry criteria to target not just cortical strokes but specifically those with a sufficient volume of penumbra.16,45⇓
Patients with potentially salvageable penumbra tissue may be identified by functional neuroimaging.44,45,53,57–59⇓⇓⇓⇓⇓ According to PET studies by Heiss et al57 performed in patients within 3 hours of acute stroke, the penumbra made up 18% (range, 8% to 34%) of the final infarct volume; 70% (range, 51% to 92%) was already critically hypoperfused, and 12% (2 to 25%) was sufficiently perfused. Although such observations imply that on average neuroprotective therapy may be able to salvage only a relatively small fraction of an infarct (supporting the rationale for combination reperfusion-neuroprotective studies), some patients can be identified with larger volumes of penumbra.46,60⇓ A recent PET study suggested that 45% of the final infarct (and in some patients, up to 85%) remained viable for up to 12 hours.60,61⇓
With MRI, a perfusion-diffusion mismatch (perfusion abnormality greater than diffusion abnormality) can be identified in ≈70% of patients studied within 6 hours of stroke onset and may indicate salvageable tissue.62,63⇓ Some trials are now using MRI criteria to improve patient selection64 (eg, a trial of sipatrigine currently underway requires a perfusion-diffusion mismatch at baseline of at least 30%65). Dynamic CT perfusion imaging also provides promise as a method of acute stroke imaging that may allow rapid identification of tissue compartments perfused within predetermined blood flow thresholds.66
In the absence of perfusion imaging, a mismatch between the clinical deficit and imaging findings has been suggested as a way to optimize patient selection (ie, severe clinical deficit with limited early lesion on diffusion-weighted MRI or CT).54 A prediction formula that incorporates the time elapsed, National Institutes of Health Stroke Scale (NIHSS) score, and diffusion-weighted MRI lesion has recently been validated.67 The Alberta Stroke Program Early CT Score, a 10-point score quantifying the signs of early infarction, also aims to improve patient selection.68,69⇓
Therefore, in future trials, there should be a major effort to improve patient selection through the use of imaging criteria, in combination with other descriptors, (1) to select candidates who are expected to benefit from treatment, ie, those who have perfusion abnormality greater than diffusion abnormality, and (2) to exclude inappropriate patients, ie, those with lacunes or large infarcts with no perfusion-diffusion mismatch.44,54,70,71⇓⇓⇓ If imaging analysis cannot be performed online in the acute stage in time for decision making, the scans should be analyzed by prespecified criteria as soon as possible to select the target population; patients with no evidence of penumbra posthoc should then be excluded from an efficacy analysis.
Pitfall 3: Preclinical Studies Have Demonstrated Protection of Gray Matter, Whereas Clinical Trials Frequently Enroll Patients Without Specifying Location of Damage
A particular concern is that preclinical neuroprotective studies have concentrated almost exclusively on the protection of cerebral gray matter from ischemic injury; the effects of neuroprotective therapy on cerebral white matter tracts are largely unknown.23,72⇓ The human brain contains a greater proportion of white matter compared with the rat brain,72 and the failure of some neuroprotective trials may be due to an inability of certain agents to protect against axonal damage.23,32,72⇓⇓ Approximately one third of human strokes are small-vessel lacunes, yet adequate animal models of lacunar stroke are lacking.32,72⇓ Until such data are available, it may not be reasonable to expect lacunes or subcortical white matter infarcts to respond to neuroprotective therapy. The pathophysiology of ischemic injury in white matter is different than in gray matter, and treatment targets likely differ as well.23,72,73⇓⇓ N-methyl-d-aspartate (NMDA) receptors, for example, are preferentially located at synapses rather than along axons.74,75⇓ Blockade of sodium channels, calcium, or alpha-amino-3-hydroxy-5-methyl-4-isoxasole-propionic acid (AMPA) receptors (rather than blockade of glutamate-mediated excitotoxicity) has been hypothesized to be more important for white-matter protection.73,74⇓ The NMDA antagonist MK-801 reduced cortical gray-matter injury but not the amount of axonal damage after middle cerebral artery occlusion in cats.76 Interestingly, however, a recent study showed that axonal and myelin damage could be reduced in rats with the NMDA antagonist CNS 1102 (Cerestat).74 More studies like this that take into account both gray-matter and white-matter pathophysiology are needed if we are to achieve “total brain protection.”72 For now, only cortically based strokes should be enrolled in neuroprotective trials, unless the agent being tested is specifically designed to protect white matter also.16 Diffusion-weighted MRI is beginning to be used in some trials to select patients with cortical involvement and exclude those with lacunar infarction.
Pitfall 4: Optimal Duration of Neuroprotectant Administration Is Unknown
In recent trials, drug administration has varied from a single injection to continuous infusions to multiple doses lasting up to 3 months after stroke.77 Because acute treatment may only delay but not prevent cell death, Dyker and Lees77 advocate continuing neuroprotective therapy for at least the first 72 hours, if not longer. Prolonged elevation of excitatory amino acids after stroke in some patients78,79⇓ and MR spectroscopy evidence suggesting ongoing neuronal loss over many days after stroke80 support the concept of extended treatment. In rats, longer-lasting neuroprotection was achieved when a glutamate antagonist was given for 1 week compared with administration in only the acute phase.31
However, certain drugs may exert different or even opposite actions, depending on the timing of administration. NMDA antagonists, benzodiazepines, or barbiturates may be beneficial if given early after ischemia but may have detrimental effects if given at later times.81–84⇓⇓⇓ For example, GABA-ergic agonists may be neuroprotective when given acutely (a trial of diazepam within the first 3 days after stroke is currently underway85), yet may impair recovery if administered at later stages after stroke.86 Additionally, neurite outgrowth requires both NMDA and voltage-sensitive calcium channel activation.87 Because neurite outgrowth is likely an essential process during recovery, patients maintained on NMDA antagonists and voltage-sensitive calcium channel blockers may suffer impaired recovery. The time point at which the therapeutic transition from neuroprotection to repair occurs merits further study.
Pitfall 5: Preclinical Studies Have Relied on Infarct Size to Judge Therapeutic Efficacy, Whereas Clinical Trials Rely on Behavioral Outcomes
Traditionally, animal studies have relied on reduction in infarct size within the first few hours after stroke as the primary measure of therapeutic efficacy. In contrast, clinical trials judge efficacy by using neurological and/or functional outcomes, not infarct volume, most often at 3 months after stroke.3,34⇓ As an example, preclinical studies of the antineutrophil adhesion agent Hu23F2G (LeukArrest) and the glycine antagonist GV150526 (gabestinel; Glycine Antagonist in Neuroprotection study)88 assessed infarct volume within hours of occlusion, while patients were evaluated on day 28 or 90, respectively, for neurological and functional outcomes. Such discrepancies suggest that reliance on infarct size measurement alone in animals can be misleading as an indicator of therapeutic efficacy.89,90⇓ Histological end points cannot tell whether surviving neurons are functional or dysfunctional or will go on to die in a delayed fashion, and they are less predictive of long-term histology than early behavioral assessments.90 Moreover, some compounds [eg, basic fibroblast growth factor (bFGF), osteogenic protein-1] have been associated with functional improvement without affecting infarct size in animals, suggesting that they act by other mechanisms, eg, enhancement of neural repair, rather than by neuroprotection.91,92⇓
Therefore, assessment of therapeutic efficacy in preclinical studies should require, in addition to infarct size, demonstration of benefit on functional measures of motor, sensory, or cognitive deficits.89,93⇓ Examples include tests of limb placing, beam walking, grid walking, Rotorod performance, grip strength, balance beam inclined plane performance, prehensile traction, and cognition (eg, Morris water maze, radial maze, 1-trial passive avoidance, T-maze retention test).89,90,94⇓⇓ These measurements are not equivalent in their reliability or predictive value. For this reason, it is better to use a battery of appropriate tests rather than a single measure. The staircase test has been recommended because of its greater sensitivity in detecting persisting deficits in forepaw dexterity months after ischemia, unlike simpler sensorimotor tasks on which animals can recover quickly.89,90⇓ The bilateral sticky tape test may be a useful indicator of poststroke neglect.94 Further development, refinement, and standardization of reliable functional assessments will continue to be a priority.89,90,94⇓⇓
Pitfall 6: Preclinical Studies Have Relied on Early Outcomes, Whereas Clinical Trials Rely on Late Assessments
Preclinical studies of neuroprotective drugs have rarely shown that early therapeutic benefit, when it is achieved, has a lasting impact. That is, of studies that have explored the long-term results when a drug showed early favorable influence on histological outcome, most have concluded that the early reduction in infarct volume does not persist if one continues to observe the animal; ie, most therapeutic attempts delay but do not arrest cell death.26,90,95⇓⇓ For example, the NMDA antagonist MK-801 appeared neuroprotective at day 3, but at 4 weeks, there was no significant difference in infarct size.95 Similarly, the cyclin-dependant kinase inhibitor flavopiridol, AMPA antagonist NBQX, and N-type calcium channel antagonist SNX-111 all appeared neuroprotective histologically when assessed at 1 week, but this was not sustained at 4 weeks.96,97⇓
Such findings demonstrate that reliance on early end points is not sufficient and can be misleading; assessments at extended time points after ischemia are necessary to determine whether there is evidence of sustained neuroprotection.15,26,95⇓⇓ Indeed, histopathological studies in animals show that infarcts evolve over time and may take many days to months to acquire their final appearance.95,98,99⇓⇓ Late consequences of ischemia (eg, inflammation) or slow death mechanisms that are unleashed (eg, apoptosis) may in part explain such findings. Thus, if single-dose neuroprotective treatment only postpones the evolution of an infarct, perhaps multidose, extended treatments or combination therapy will be required for optimal neuroprotection (see below).77,100⇓
Pitfall 7: Experimental Stroke Models Are Homogeneous, Whereas Human Stroke Is Heterogeneous
Another problem confounding the evaluation of neuroprotective therapy is the tremendous variability of human stroke types, recovery patterns, and associated clinical factors. Preclinical studies usually involve middle cerebral artery occlusion in young, healthy animals under anesthesia with tightly controlled temperature, blood pressure, oxygenation, and glucose levels (“interventional homeostasis”54).101 In contrast, clinical trials often permit entry of multiple stroke types (cortical, mixed cortical-subcortical, pure subcortical white-matter strokes, and in some neuroprotective trials, both ischemic and hemorrhagic stroke), and there is a lack of standardized control over physiological parameters.54 Unlike the animal model, stroke patients typically have a multitude of associated variables that may affect prognosis,102–104⇓⇓ including old age, comorbidities, polypharmacy, recurrent ischemia, poor collateral circulation, or prior strokes. Hyperglycemia and other metabolic prognostic markers105 may be particularly important variables to control or adjust for in future trials.106–108⇓⇓
Instead of viewing stroke as a single disease entity, future trials should more appropriately be directed only to specific homogeneous stroke subtypes. Currently, of 178 published acute stroke trials, only 2% specified a target stroke mechanism, and 35% specified a specific stroke territory.3 Although a lack of specificity regarding stroke types in trial entry criteria may be appropriate for thrombolysis (where the pathophysiologic target, ie, clot, is similar regardless of stroke location, perhaps with the exception of lacunes), this may not be appropriate in neuroprotective trials in which efficacy in some patients (eg, cortical stroke), if present, could be diluted by the inclusion of other stroke types (eg, subcortical strokes). Indeed, posthoc analysis of some neuroprotective trials has suggested that a benefit may exist for certain subgroups, eg, patients with large cortical stroke [total anterior circulation stroke in the Clomethiazole Acute Stroke Study (CLASS)].109,110⇓ A follow-up study specifically targeting these patients failed to validate this hypothesis, however, but it set a good example for how we should approach future studies.111
Pitfall 8: Choice of Outcome Measures Can Determine the Success of a Clinical Trial More Than Actual Drug Efficacy
The choice of outcome measures in clinical trials is critical to the success or failure of a putative therapeutic intervention.16,34,54⇓⇓ However, there is a lack of agreement about the most appropriate measures that should be used and about what constitutes “recovery” or “favorable outcome.”34,112–114⇓⇓⇓ For example, on the Barthel Index, cutoff points anywhere between 50 to 95 of 100 have been used to define recovery.115 Kidwell et al3 showed that less than half of published acute stroke trials used a validated outcome measure and that only 17% had a prespecified primary end point. In the review of 51 phase II and III acute stroke trials by Duncan et al,34 there were 14 different impairment level measures, 11 different activity (disability) measures, 8 miscellaneous scales, and only 1 quality-of-life measure. If disability scales are used (eg, Barthel Index, modified Rankin Scale), more patients will be considered recovered; if impairment scales are used (eg, NIHSS) fewer patients will be considered recovered.34 Table 1 shows how one’s impression of recovery is directly dependent on the type of outcome measure chosen.
This variability in outcome assessment has made the stroke literature appear confusing and at times conflicting. Indeed, much of the controversy regarding the thrombolytic trials has resulted from inconsistency in the definition of recovery and differences in end points used among the various trials. In the NINDS study,4 the benefit of tPA at 24 hours did not reach statistical significance on the prespecified NIHSS end point. However, posthoc analysis showed that if recovery is instead defined as an NIHSS score of 0 to 2, a striking difference is found: 24% of tPA-treated patients versus 5% of placebo-treated patients are recovered at 24 hours.116 Moreover, the 24-hour outcome results would have been statistically significant on the predefined primary end point if the recently published modified version of the NIHSS had been used instead.117 Furthermore, in the European Cooperative Acute Stroke Study (ECASS II), which used “favorable outcome” as defined in the NINDS study (modified Rankin scale score of 0 or 1), the result was statistically negative.118 However, when a different dichotomization that classifies outcome in terms of self-care independence (Rankin score of 0 to 2) was used, the study was positive.118
Thus, greater consensus and standardization in outcome measures for acute stroke studies are needed, and this would facilitate meta-analysis. It is recommended that future efficacy trials incorporate outcome assessments that span the spectrum of stroke recovery, ie, impairment, activity limitations (disability), and participation restrictions (handicap).34 As suggested by Duncan et al,34 a single scale likely is inadequate to capture recovery, and dichotomized outcomes should be avoided; inclusion of extended/instrumental activities of daily living assessments, advanced mobility measures, and quality-of-life assessments is recommended.34 Although it may be appropriate in 3-hour thrombolysis trials to aim for neurological cures (eg, NIHSS score 0 or 1) or functional recovery (eg, modified Rankin score of 0 or 1), in trials of neuroprotective agents or longer time windows, our expectations should be different. Grotta40 reminds us that we should be aiming for neuronal protection, not neuronal “reincarnation.” Therapeutic efficacy will likely be of a smaller magnitude, one that may be captured only by using less stringent criteria for recovery (eg, NIHSS <7 or modified Rankin score of 0 to 2), or by measuring shifts in disability states.118a A newer end point, neurological deterioration in hospital, has been proposed for traumatic brain injury trials.119
Different measures of recovery may be necessary, depending on the severity of the stroke population under investigation (ie, mild, moderate, or severe), the particular type of treatment being studied, or the specific neurological function targeted by the intervention. For example, in a recent trial of stem cell implantation for hemiparetic stroke patients,120 the global neurologic deficit scales used (NIHSS and European Stroke Scale) would not be expected to capture meaningful change in motor impairment; instead, a motor-specific impairment scale such as the Fugl-Meyer Stroke Assessment121,122⇓ might be more revealing given the location of the target stroke in the subcortical basal ganglia region. As we learn the limitations of existing scales, newer stroke-specific indexes incorporating quality-of-life measurements are being developed; eg, the Stroke Impact Scale123 and SS-QOL124 are intended to provide more comprehensive and more meaningful outcomes from the patient’s perspective.124 Inclusion of specific scales for aphasia, neglect, or apraxia may reveal benefits in subgroups of patients that are not apparent on global deficit rating scales. The concept of separate “motor-Rankin” and “cognitive-Rankin” scales has been advocated.125 Stratification of patients in clinical trials by initial stroke severity is important. The Orpington prognostic scale, recently shown to have excellent predictive value for stroke recovery, may have value for stratifying patients in treatment trials.126 The optimal time for outcome assessment is debatable, 3 months after stroke (or earlier) according to some16 and 6 months according to others.34 In some studies, even if a therapeutic effect on final outcome (ie, at 3 or 6 months) cannot be demonstrated, it may be desirable to detect whether the intervention was able to accelerate the rate of recovery. Comparison of change scores (the difference between baseline and final scores for each subject) may have the advantage of minimizing interindividual variability in stroke severity and recovery if reliable outcome measures are used.
Pitfall 9: Small Trials Are Trying to Answer Questions That Only Large Trials Can Answer
Have some stroke trials been negative because of a lack of efficacy or because of a lack of statistical power? To detect efficacy of neuroprotective compounds, which are likely to have small rather than large treatment effects, we need large trials (thousands of patients, according to some experts) to prevent type 2 statistical error.54,127⇓ The mean sample size of neuroprotective trials has been 186 (median, 69).3 Only 2% of acute stroke efficacy trials have had sufficient statistical power to demonstrate a 5% absolute clinical benefit, and only 7% of trials have been powered to detect a 10% benefit.3 The use of “adaptive randomization” in future trials may reduce sample size requirements.128 This Bayesian statistical technique, being used in a neuroprotective trial currently underway,129 aims to maximize the number of patients assigned to the dose(s) that appear most efficacious; outcome data from each patient provides feedback to the randomization computer as the trial proceeds to optimize the chance that the correct drug dose will be studied.54
Part 2: New Therapeutic Frontiers
Most trials to date have attempted to modulate the early metabolic events in ischemia, particularly those involving glutamate activation of the calcium cascade. With a growing understanding of the pathophysiology of ischemic brain injury in the acute phase of stroke, as well as progress in understanding the mechanisms that underlie functional recovery in the subacute stages, newer therapeutic strategies are emerging.81,130–134⇓⇓⇓⇓⇓ The concept of a single narrow time window for intervention is being replaced by the potential for multiple overlapping therapeutic windows and the possibility of multiagent chemotherapy “cocktails” administered at selected time periods after stroke (Table 2).130 In this section, we discuss a few selected examples of strategies that may form the basis for future trials, either alone or in combination with traditional neuroprotective approaches.
Delayed Neuronal Death After Ischemia: A Role for Antiapoptotic Therapy?
Observations of delayed neuronal death after ischemia have suggested the possibility of an “apoptosis-necrosis” continuum. Depending on the degree and duration of ischemia, brain cells may die by an ionic cascade, rapidly (necrotic cell death), or by a molecular cascade, slowly (apoptotic cell death).131,132,135–137⇓⇓⇓⇓ At one end of the spectrum, severe focal ischemia produces infarction through excitotoxic necrosis usually evident in rats by 6 hours and maximal at 24 hours. At the other end of the spectrum, when ischemia is mild and short lasting, the resulting cell death may be apoptotic; ie, neurons appear to be spared initially but later go on to die slowly.138 For example, after transient focal ischemia in rats Du et al138 found no infarction at 24 hours; however, by 3 days, a small infarct had developed, and remarkably by 14 days, the infarct had progressed to the same volume as that induced by severe ischemia. Although excitotoxic necrosis is considered the predominant mechanism of ischemic cell death in most cases, apoptosis may occur in penumbral neurons that escape excitotoxic death.132 Importantly, if necrosis is attenuated by therapy (ie, by reperfusion or antiexcitotoxic agents), then apoptosis may be unmasked or even promoted.132,136,138,139⇓⇓⇓
For these reasons, it is short-sighted to continue only with antinecrosis therapies without taking into account the role that apoptosis may play in ischemic cell death. Antiapoptotic agents (eg, cycloheximide, caspase inhibitors) have shown neuroprotective effects in animals in terms of both infarct volume reduction and functional improvement. Furthermore, the neuroprotection appears long lasting.140,141⇓ The therapeutic window for antiapoptotic neuroprotection is longer than that for most other neuroprotective agents (eg, 9 hours for the caspase inhibitor ZVAD-fmk and zDEVD-fmk after brief ischemia and up to 12 hours if combined with an NMDA antagonist).142 The role of apoptosis in human stroke, however, and the clinical relevance of antiapoptotic therapy are not yet known and await further investigation. If further studies confirm the occurrence of apoptosis in the penumbra region, then cerebral blood flow measurement may become a clinically accessible surrogate marker for this outcome.
The Need for Polytherapy
On the basis of the complexity of events in cerebral ischemia and the disappointing results from single-agent trials, it may not be realistic to expect that 1 neuroprotective drug will have lasting benefit. Rather, effective neuroprotection may require “rational” polytherapy that combines drugs with different mechanisms of action, perhaps administered at different poststroke intervals, to maximize efficacy and/or extend the window for reperfusion, minimize reperfusion injury or hemorrhage, or inhibit delayed cell death.16,130,143,144⇓⇓⇓ Furthermore, because the failure of several neuroprotective trials has been attributed to dose-limiting toxicity,145 combination therapy may permit lower doses of each agent and minimize adverse effects. Combinations with nonpharmacological (physiological) neuroprotective strategies such as hypothermia, insulin, and blood pressure control should also be subjected to clinical trials.103,146⇓
Combined thrombolysis-neuroprotective approaches have shown promise in animal studies and are beginning to be investigated in clinical trials. For example, synergistic effects have been demonstrated in animals when thrombolysis is combined with citicoline,147 an AMPA antagonist,148 an NMDA antagonist,149 or other agents.150,151⇓ Administration of antileukocytic adhesion antibodies has been shown to extend the therapeutic window for thrombolysis.152,153⇓ Recently, 2 trials have demonstrated the feasibility and safety of intravenous tPA treatment followed by neuroprotectant administration: the CLASS-T trial of clomethiazole154 and a study of lubeluzole.155
In animals, synergy has been demonstrated by the combination of 2 neuroprotective agents with different actions; some examples include the NMDA antagonist MK-801 in combination with a GABA agonist,156–158⇓⇓ a free radical scavenger,159 a calcium antagonist,160 citicoline,161 or bFGF.162 Similarly, synergy has been observed with the combination of the antioxidant tirilazad and magnesium,163,164⇓ the combination of 2 different antioxidants,165 and citicoline combined with bFGF.166
Synergistic effects on infarct size and therapeutic window have also been found in animals through the use of antiexcitotoxic agents in combination with antiapoptotic agents, eg, the NMDA antagonist dextrorphan plus the protein synthesis inhibitor cyclohexamide,167 or MK 801 with the caspase inhibitor ZVAD-fmk.168 Caspase inhibitors given with bFGF extended the therapeutic window and lowered the required doses for neuroprotection.169
Beyond Neuroprotection: Exploiting the Repair Mechanisms of the Brain
The poststroke recovery period (days, weeks, and months after stroke) represents another target for more active therapeutic development. Our preoccupation with hyperacute and acute intervention, while critically important, should not lead us to neglect the testing of “restorative” interventions that might enhance recovery in the subacute and chronic stages after stroke.81,170⇓ Unlike neuroprotective therapy that targets ischemic tissue and aims to limit infarct size, “recovery-enhancing drugs” and other novel interventions for stroke aim to promote functional recovery after a completed stroke by stimulating repair mechanisms.
The Changing Milieu of the Brain After Stroke
Within minutes of ischemia, a loss of dendritic spines can be observed at excitatory synapses; reestablishment of dendritic spine synapses in surviving neurons can occur rapidly and represents a potential substrate for functional recovery.171 Ischemic cortical injury induces the expression of growth factors in peri-infarct regions, and behavioral recovery is accompanied by increased dendritic branching and synaptogenesis that peaks 2 to 4 weeks after stroke in the rat.172,173⇓ Mechanisms of neuroplasticity postulated to underlie recovery include unmasking of latent connections, redundancy that allows recruitment of alternate parallel pathways to take over lost functions, axonal sprouting from surviving neurons with formation of new synapses, and possibly even neurogenesis.134,174–176a⇓⇓⇓⇓
Intracisternal administration of bFGF beginning 24 hours after stroke or osteogenic protein-1 beginning 3 days after stroke promoted recovery in animals without affecting infarct size and stimulated new neuronal sprouting and synapse formation.91,92,177,178⇓⇓⇓ Clinical trials of bFGF administered within 6 hours have been underway, but 1 trial was terminated early because of safety concerns.179,180⇓
Proteins that are normally expressed in the developing brain and suppressed in the adult brain become upregulated in response to ischemia; stroke “recapitulates ontogeny,” as summarized by Cramer and Chopp.181 Cell-cycle genes are re-expressed in response to ischemia; ie, neurons are activated to “divide,” but they are hard wired not to. Activation of the cell cycle cascade results in cell death, and interrupting this cascade results in cell survival.182 This may be another fruitful therapeutic avenue in ischemia. Stem cell therapy is being investigated for enhancement of stroke recovery. There are suggestions in the literature that bone marrow stromal cells or umbilical cord blood cells injected intravenously or intra-arterially in rats can migrate toward the infarct, stimulate growth factors, and promote functional recovery.183–185⇓⇓ Recently, human experimentation has begun with a phase I trial of transplantation of cultured neuronal cells into basal ganglia infarcts in stroke patients.120 A great deal more animal work is needed to provide a firmer basis for intervention.
Remapping and Rehabilitation
Cortical reorganization after stroke is promoted by rehabilitation and an enriched environment.134,186,187⇓⇓ Nudo et al188,189⇓ showed that in monkeys who do not receive rehabilitative training after a small focal injury in the hand motor cortical region, the surrounding intact cortical representation undergoes shrinkage; however, with repetitive training in the form of motor skill acquisition, behavioral recovery is promoted, and the cortical hand representation is maintained or expanded. Thus, physical therapy may derive its effectiveness from “teaching” the brain to learn. Improved recovery in humans may be achieved with increased intensity of rehabilitation (more hours and greater frequency of therapy).190 Such dose-response effects suggest the concept that “more is better” applies to stroke recovery. One intensive physical therapy regimen that is gaining popularity is “forced use” or constraint-induced movement therapy, which can be applied if the weak arm has enough strength to at least move against gravity.191 In this protocol, the unaffected arm is constrained in a sling, forcing the patient to use the affected arm as much as possible in meaningful daily activities. The aim is to minimize the development of learned helplessness resulting from overreliance on the “good arm.” Based on reported success in chronic stroke patients, this therapy is now under investigation in multicenter trials as a rehabilitation intervention in the earlier stages after stroke. Bilateral arm use assisted by passive movement may enhance activation in the peri-infarct region of a stroke192 and may have a role in early rehabilitation. Body weight–supported treadmill training is a promising technique being investigated to promote recovery of ambulation after stroke.193
Pharmacological Manipulation After Stroke
Recovery after stroke may also be modulated pharmacologically.81,178,194–196⇓⇓⇓⇓ Accumulating evidence suggests that the recovery process is dynamic and vulnerable to neurotransmitter modulation.82 For example, pharmacological studies in animals have emphasized the importance of central noradrenergic transmission in mediating some forms of recovery after focal cortical injury. Drugs augmenting noradrenergic activity (eg, dextroamphetamine) enhance functional recovery when coupled with symptom-relevant experience, whereas drugs decreasing noradrenergic activity impair recovery and can reinstate deficits after recovery has taken place.197–200⇓⇓⇓ (For a review of additional studies, see Reference 170.) Similar findings have been demonstrated for other classes of drugs and may relate to their ability to facilitate (or impair) long-term potentiation.82,201⇓ Histologically, dextroamphetamine administration after cortical infarction in rats has been associated with upregulation of neural sprouting and synaptogenesis in the peri-infarct cortex and contralesional cortex, correlating with behavioral recovery.202,203⇓ In animal models, chronic amphetamine administration, coupled with another stimulus, has been shown to increase cortical responsiveness possibly through upregulation of CREB, a protein transcription factor, and induction of genes mediating other molecular changes.204 Experiments from Leonardo Cohen’s laboratory investigating the effects of various drugs on cortical plasticity have shown that dextroamphetamine administration facilitates the induction, magnitude, and retention of use-dependent plasticity in humans during performance of a motor training task.205
The concept of rehabilitation pharmacology, which dates back >20 years,206 proposes that conventional physical, occupational, or speech/language therapy might be augmented if coupled with pharmacotherapy to enhance activity-dependent plasticity.207 On the basis of small clinical studies showing some promise,208–210⇓⇓ we are conducting a clinical trial in Canada to investigate the effects on motor recovery of dextroamphetamine versus placebo coupled with physiotherapy after hemiparetic stroke,211 and in the United States, a multicenter trial of amphetamine-facilitated recovery is underway. A recently published study, however, showed no benefit of amphetamine on motor recovery in a randomized, controlled trial of 36 patients treated with intermittent doses of drug coupled with physical therapy beginning 5 to 10 days after stroke.212 However, racemic amphetamine 10 mg was used instead of dextroamphetamine as in previous studies, and their patients were older than in previous studies (average age, 80 years). Stroke type and neuroimaging characteristics that might influence recovery or treatment response (eg, lesion size, location) were not reported, and patients with moderate hemiparesis were not analyzed separately in comparison to those with severe hemiparesis. The effects of dextroamphetamine coupled with speech/language therapy are also being investigated in ongoing studies by Walker-Batson et al213–215⇓⇓ and have shown some promise for enhancing aphasia recovery. Trials of other noradrenergic agonists (methylphenidate, L-DOPS), levodopa, and fluoxetine have been conducted recently and provide further proof of concept for the strategy of poststroke rehabilitation pharmacotherapy.216–219⇓⇓⇓
Goldstein et al195,220,221⇓⇓ have drawn attention to the observation that several drugs shown to be detrimental in the laboratory are commonly prescribed to hospitalized patients after stroke and head trauma and may have similar detrimental effects on recovery in humans. These include the antihypertensives clonidine and prazosin (α-noradrenergic antagonists), haloperidol and other dopamine antagonists, benzodiazepines, phenytoin, and phenobarbital.194,195⇓ Retrospective reports suggest that exposure to such drugs is associated with poorer motor recovery, independent of the severity of initial deficit or comorbid conditions.222,223⇓ Aphasia recovery may also be impaired by certain drugs.206,224⇓ In a prospective study of chronic stroke patients, administration of increasing doses of a benzodiazepine reinstated previously recovered focal deficits (hemiparesis, aphasia, neglect).86
Clinical rehabilitation trials are promising for many reasons and may avoid some of the methodological obstacles encountered in many acute stroke trials.125 (Table 3) Preclinical studies of restorative therapies necessarily rely more on behavioral outcomes and extended follow-up periods (weeks after stroke), making these animal models potentially more relevant to the human condition than models of acute neuroprotective therapy. Without the constraints of a narrow time window, effective restorative therapies may be able to reach many more patients than would qualify for acute treatment. The challenge remains, however, to design these trials well, select appropriate outcome measures, determine the clinically important difference, and power the samples accordingly.
Researchers and clinicians must become more cognizant of the pitfalls and paradoxes that have arisen in attempting to translate the results of animal studies into clinical trials of neuroprotective stroke therapy. Much needed recommendations to improve the quality of preclinical and clinical drug development have been published recently by the Stroke Therapy Academic Industry Roundtable (STAIR)15,16⇓ and others20 and should be followed. Preclinical evaluation of therapeutic efficacy based solely on measuring infarct volume in the early phase is no longer adequate. Clinical trials should be based on preclinical evidence demonstrating improved functional outcomes at long-term end points measured on standardized batteries of validated behavioral tests. Assessment of neuroprotection should rely more on delayed time windows for drug administration, longer durations of ischemia, and models that take into account the protection of cerebral white matter in addition to gray matter.15,16,72⇓⇓ Ideally, there should be converging evidence from >1 laboratory, different stroke models, and >1 animal species.15
Previous neuroprotective trials have not been selective enough in targeting homogeneous patients with cortical stroke and evidence for the presence of ischemic penumbra. Future trials need to use stricter entry criteria to target not just cortical strokes but specifically those with a sufficient volume of penumbra as determined by imaging. Although functional imaging is currently restricted to specialized centers, we urgently need proof of principle that neuroprotectives can work in humans if administered to the appropriate patient populations. Perfusion imaging with CT is worthy of further investigation because of the availability and practicality of CT for acute stroke imaging. With improved methods of patient selection, it is anticipated that future treatments will more appropriately be customized according to an individual’s “penumbra window” rather than a rigid time window. Patient selection should be “penumbra-specific” for acute treatment interventions and “deficit-specific” for restorative/rehabilitative interventions. Greater consensus on outcome measures is needed because the choice of outcome measure remains highly variable and can determine the success or failure of a putative therapeutic agent.
If current neuroprotective approaches only delay rather than arrest cell death, future strategies may require “rational” polytherapy that combines drugs with different mechanisms of action, perhaps administered at different poststroke intervals and in combination with reperfusion strategies. The concept of the therapeutic window is evolving from a single rigid and narrow time period to multiple potential overlapping and sequential windows, spanning minutes, hours, days, weeks, and months after stroke. Although most trials have concentrated on excitotoxicity, newer targets, including inhibition of inflammatory reactions and apoptosis, are being explored. The time point at which the therapeutic transition from neuroprotection to repair occurs merits further study.
Stroke is a chronic condition, and we must not ignore opportunities for therapeutic intervention in the subacute and chronic stages. With a growing understanding of the mechanisms that underlie recovery, research directed at enhancement of neural repair and rehabilitation should be a high priority.
We wish to acknowledge financial support from the Canadian Institutes of Health Research and Heart and Stroke Foundation of Ontario. Dr Gladstone received a research fellowship from the Canadian Institutes of Health Research and Heart and Stroke Foundation of Canada and is a trainee in the Royal College of Physicians and Surgeons of Canada Clinician-Investigator Program in the Department of Medicine and Institute of Medical Sciences, University of Toronto.
- Received November 15, 2001.
- Revision received April 2, 2002.
- Accepted May 1, 2002.
- 1.↵American Heart Association. 2001 Heart and Stroke Statistical Update. Dallas, Tex: American Heart Association; 2001.
- 3.↵Kidwell CS, Liebeskind DS, Starkman S, Saver JL. Trends in acute ischemic stroke trials through the 20th century. Stroke. 2001; 32: 1349–1359.
- 5.↵Furlan A, Higashida R, Wechsler L, Gent M, Rowley H, Kase C, Pessin M, Ahuja A, Callahan F, Clark WM, Silver F, Rivera F. Intra-arterial prourokinase for acute ischemic stroke: the PROACT II study: a randomized controlled trial: Prolyse in Acute Cerebral Thromboembolism. JAMA. 1999; 282: 2003–2011.
- 15.↵Stroke Therapy Academic Industry Roundtable (STAIR). Recommendations for standards regarding preclinical neuroprotective and restorative drug development. Stroke. 1999; 30: 2752–2758.
- 16.↵Stroke Therapy Academic Industry Roundtable II. Recommendations for clinical trial evaluation of acute stroke therapies. Stroke. 2001; 32: 1598–1606.
- 17.↵Grotta J. Neuroprotection is unlikely to be effective in humans using current trial designs. Stroke. 2001; 33: 306–307.
- 18.↵Lees KR. Neuroprotection is unlikely to be effective in humans using current trial designs: an opposing view. Stroke. 2001; 33: 308–309.
- 22.↵Lees KR. Neuroprotection. Br Med Bull. 2000; 56: 401–412.
- 23.↵Muir KW, Grosset DG. Neuroprotection for acute stroke: making clinical trials work. Stroke. 1999; 30: 180–182.
- 27.↵Wardlaw JM, Warlow CP, Sandercock PA, Dennis MS, Lindley RI. Neuroprotection disappointment yet aGAIN. Lancet. 2000; 356: 597.
- 30.↵Morgenstern LB. What have we learned from clinical neuroprotective trials? Neurology. 2001; 57: S45–S47.
- 31.↵Slikker W, Saran J, Auer RN, Palmer GC, Narahashi T, Youdim MBH, Maynard KI, Carbone KM, Trembly B. Neuroprotection: past successes and future challenges. Ann N Y Acad Sci. 2001; 939: 465–477.
- 32.↵Small DL, Buchan AM. Animal models. Br Med Bull. 2000; 56: 307–317.
- 33.↵Neff SR. Rodent models of stroke. Arch Neurol. 1997; 54: 350–351.
- 34.↵Duncan PW, Jorgensen HS, Wade DT. Outcome measures in acute stroke trials: a systematic review and some recommendations to improve practice. Stroke. 2000; 31: 1429–1438.
- 35.↵Horn J, de Haan RJ, Vermeulen M, Luiten PGM, Limburg M. Nimodipine in animal model experiments of focal cerebral ischemia: a systematic review. Stroke. 2001; 32: 2433–2438.
- 38.↵Use of anti-ICAM-1 therapy in ischemic stroke: results of the Enlimomab Acute Stroke Trial. Neurology. 2001; 57: 1428–1434.
- 40.↵Grotta JC. Acute stroke therapy at the millennium: consummating the marriage between the laboratory and bedside: the Feinberg lecture. Stroke. 1999; 30: 1722–1728.
- 41.↵Zivin JA. Factors determining the therapeutic window for stroke. Neurology. 1998; 50: 599–603.
- 42.↵Marler JR, Tilley BC, Lu M, Brott TG, Lyden PC, Grotta JC, Broderick. JP, Levine SR, Frankel MP, Horowitz SH, Haley EC Jr, Lewandowski CA, Kwiatkowski TP. Early stroke treatment associated with better outcome: the NINDS rt-PA stroke study. Neurology. 2000; 55: 1649–1655.
- 43.↵Baron JC, von Kummer R, del Zoppo GJ. Treatment of acute ischemic stroke. challenging the concept of a rigid and universal time window. Stroke. 1995; 26: 2219–2221.
- 45.↵Fisher M. Characterizing the target of acute stroke therapy. Stroke. 1997; 28: 866–872.
- 46.↵Marchal G, Beaudouin V, Rioux P, de la Sayette V, Le Doze F, Viader F, Derlon JM, Baron JC. Prolonged persistence of substantial volumes of potentially viable brain tissue after stroke: a correlative PET-CT study with voxel-based data analysis. Stroke. 1996; 27: 599–606.
- 48.↵Barber PA, Zhang J, Demchuk AM, Hill MD, Buchan AM. Why are stroke patients excluded from tPA therapy? An analysis of patient eligibility. Neurology. 2001; 56: 1015–1020.
- 49.↵Arrowsmith JE, Harrison MJG, Newman SP, Stygall J, Timberlake N, Pugsley WB. Neuroprotection of the brain during cardiopulmonary bypass. Stroke. 1998; 29: 2357–2362.
- 50.↵Fisher M, Jonas S, Sacco RL. Prophylactic neuroprotection for cerebral ischemia. Stroke. 1994; 25: 1075–1080.
- 52.↵Grieco G, d’Hollosy M, Culliford AT, Jonas S. Evaluating neuroprotective agents for clinical anti-ischemic benefit using neurological and neuropsychological changes after cardiac surgery under cardiopulmonary bypass: methodological strategies and results of a double-blind, placebo-controlled trial of GM1 ganglioside. Stroke. 1996; 27: 858–874.
- 53.↵Baron JC. Perfusion thresholds in human cerebral ischemia: historical perspective and therapeutic implications. Cerebrovasc Dis. 2001; 11 (suppl 1): 2–8.
- 56.↵Hakim AM. Ischemic penumbra: the therapeutic window. Neurology. 1998; 51: S44–S46.
- 57.↵Heiss WD, Thiel A, Grond M, Graf R. Which targets are relevant for therapy of acute ischemic stroke? Stroke. 1999; 30: 1486–1489.
- 58.↵Schlaug G, Benfield A, Baird AE, Siewert B, Lovblad KO, Parker RA, Edelman RR, Warach S. The ischemic penumbra: operationally defined by diffusion and perfusion MRI. Neurology. 1999; 53: 1528–1537.
- 60.↵Baron JC. Mapping the ischaemic penumbra with PET: a new approach. Brain. 2001; 124: 2–4.
- 61.↵Heiss WD, Kracht LW, Thiel A, Grond M, Pawlik G. Penumbral probability thresholds of cortical flumazenil binding and blood flow predicting tissue outcome in patients with cerebral ischaemia. Brain. 2001; 124: 20–29.
- 63.↵Barber PA, Davis SM, Darby DG, Desmond PM, Gerraty RP, Yang Q, Jolley D, Donnan GA, Tress BM. Absent middle cerebral artery flow predicts the presence and evolution of the ischemic penumbra. Neurology. 1999; 52: 1125–1132.
- 65.↵Orgogozo JM. Sipatrigine in Stroke Trial: a European multicenter study to establish the maximum tolerated dose of sipatrigine and to assess the efficacy of that dose by measuring the change in volume of ischaemic lesion by magnetic resonance imaging. Stroke. 2001; 32: 1455.Abstract.
- 66.↵Nabavi DG, Cenic A, Henderson S, Gelb AW, Lee TY. Perfusion mapping using computed tomography allows accurate prediction of cerebral infarction in experimental brain ischemia. Stroke. 2001; 32: 175–183.
- 68.↵Pexman JH, Barber PA, Hill MD, Sevick RJ, Demchuk AM, Hudon ME, Hu WY, Buchan AM. Use of the Alberta Stroke Program Early CT Score (ASPECTS) for assessing CT scans in patients with acute stroke. Am J Neuroradiol. 2001; 22: 1534–1542.
- 70.↵Warach S. New imaging strategies for patient selection for thrombolytic and neuroprotective therapies. Neurology. 2001; 57: S48–S52.
- 71.↵Baird AE, Warach S. Using pathophysiology in acute stroke trials. Stroke. 1999; 30: 1293.Letter.
- 74.↵Schabitz WR, Li F, Fisher M. TheN-methyl-d-aspartate antagonist CNS 1102 protects cerebral gray and white matter from ischemic injury following temporary focal ischemia in rats. Stroke. 2000; 31: 1709–1714.
- 77.↵Dyker AG, Lees KR. Duration of neuroprotective treatment for ischemic stroke. Stroke. 1998; 29: 535–542.
- 79.↵Bullock R, Zauner A, Woodward J, Young HF. Massive persistent release of excitatory amino acids following human occlusive stroke. Stroke. 1995; 26: 2187–2189.
- 80.↵Saunders DE, Howe FA, van den Boogaart A, McLean MA, Griffiths JR, Brown MM. Continuing ischemic damage after acute middle cerebral artery infarction in humans demonstrated by short-echo proton spectroscopy. Stroke. 1995; 26: 1007–1013.
- 81.↵Goldstein LB. Restorative Neurology: Advances in Pharmacotherapy for Recovery After Stroke. New York, NY: Futura Publishing; 1998.
- 82.↵Goldstein LB. Pharmacology of recovery after stroke. Stroke. 1990; 21: 139–142.
- 83.↵Lodder J. Stroke recovery. Neurology. 1996; 46: 1187.Letter.
- 84.↵Goldstein LB. Stroke recovery. Neurology. 1996; 46: 1187–1188.Letter.
- 85.↵Lodder J. Early GABA-Ergic Activation Study in Stroke (EGASIS). Stroke. 2001; 32: 2450.Abstract.
- 86.↵Lazar RM, Fitzsimmons BF, Marshall RS, Berman MF, Bustillo MA, Young WL, Mohr JP, Shah J, Robinson JV. Reemergence of stroke deficits with midazolam challenge. Stroke. 2002; 33: 283–285.
- 87.↵Maletic-Savatic M, Malinow R, Svoboda K. Rapid dendritic morphogenesis in CA1 hippocampal dndrites induced by synaptic activity. Science. 1999; 283: 1923–1927.
- 88.↵Sacco RL, DeRosa JT, Haley ECJ, Levin B, Ordronneau P, Phillips SJ, Rundek T, Snipes RG, Thompson JL. The Glycine Antagonist in Neuroprotection Americas Investigators: glycine antagonist in neuroprotection for patients with acute stroke: GAIN Americas: a randomized controlled trial. JAMA. 2001; 285: 1719–1728.
- 95.↵Valtysson J, Hillered L, Andine P, Hagberg H, Persson L. Neuropathological endpoints in experimental stroke pharmacotherapy: the importance of both early and late evaluation. Acta Neurochir. 1994; 129: 58–63.
- 96.↵Colbourne F, Li H, Buchan AM, Clemens JA. Continuing postischemic neuronal death in CA1: influence of ischemia duration and cytoprotective doses of NBQX and SNX-111 in rats. Stroke. 1999; 30: 662–668.
- 99.↵Garcia JH, Liu KF, Ho KL. Neuronal necrosis after middle cerebral artery occlusion in Wistar rats progresses at different time intervals in the caudoputamen and the cortex. Stroke. 1995; 26: 636–642.
- 100.↵Coimbra C, Drake M, Boris-Moller F, Wieloch T. Long-lasting neuroprotective effect of postischemic hypothermia and treatment with an anti-inflammatory/antipyretic drug: evidence for chronic encephalopathic processes following ischemia. Stroke. 1996; 27: 1578–1585.
- 103.↵Jorgensen HS, Reith J, Nakayama H, Kammersgaard LP, Houth JG, Raaschou HO, Olsen TS, Copland S, for the Copenhagen Stroke Study. Potentially reversible factors during the very acute phase of stroke and their impact on the prognosis: is there a large therapeutic potential to be explored? Cerebrovasc Dis. 2001; 11: 207–211.
- 105.↵Boysen G, Christensen H. Early stroke: a dynamic process. Stroke. 2001; 32: 2423–2425.
- 107.↵Demchuk AM, Morgenstern LB, Krieger DW, Linda Chi T, Hu W, Wein TH, Hardy RJ, Grotta JC, Buchan AM. Serum glucose level and diabetes predict tissue plasminogen activator-related intracerebral hemorrhage in acute ischemic stroke. Stroke. 1999; 30: 34–39.
- 108.↵Capes SE, Hunt D, Malmberg K, Pathak P, Gerstein HC. Stress hyperglycemia and prognosis of stroke in nondiabetic and diabetic patients: a systematic overview. Stroke. 2001; 32: 2426–2432.
- 109.↵Wahlgren NG, Ranasinha KW, Rosolacci T, Franke CL, van Erven PM, Ashwood T, Claesson L. Clomethiazole Acute Stroke Study (CLASS): results of a randomized, controlled trial of Clomethiazole versus placebo in 1360 acute stroke patients. Stroke. 1999; 30: 21–28.
- 111.↵Lyden P, Shuaib A, Ng K, Levin K, Atkinson RP, Rajput A, Wechsler L, Ashwood T, Claesson L, Odergren T, Salazar-Grueso E, for the CLASS-I/H/T Investigators. Clomethiazole Acute Stroke Study in Ischemic Stroke (CLASS-I): final results. Stroke. 2002; 33: 122–128.
- 112.↵Roberts L, Counsell C. Assessment of clinical outcomes in acute stroke trials. Stroke. 1998; 29: 986–991.
- 114.↵Uchino K, Billheimer D, Cramer SC. Entry criteria and baseline characteristics predict outcome in acute stroke trials. Stroke. 2001; 32: 909–916.
- 115.↵Sulter G, Steen C, De Keyser J. Use of the Barthel Index and modified Rankin Scale in acute stroke trials. Stroke. 1999; 30: 1538–1541.
- 116.↵Broderick JP, Lu M, Kothari R, Levine SR, Lyden PD, Haley EC, Brott TG, Grotta J, Tilley BC, Marler JR, Frankel M. Finding the most powerful measures of the effectiveness of tissue plasminogen activator in the NINDS tPA stroke trial. Stroke. 2000; 31: 2335–2341.
- 117.↵Lyden PD, Lu M, Levine SR, Brott TG, Broderick J, for the NINDS rtPA Stroke Study Group. A modified National Institutes of Health Stroke Scale for use in stroke clinical trials: preliminary reliability and validity. Stroke. 2001; 32: 1310–1317.
- 118.↵Hacke W, Kaste M, Fieschi C, von Kummer R, Davalos A, Meier D, Larrue V, Bluhmki E, Davis S, Donnan G, Schneider D, Diez-Tejedor E, Trouillas P. Randomised double-blind placebo-controlled trial of thrombolytic therapy with intravenous alteplase in acute ischaemic stroke (ECASS II). Lancet. 1998; 352: 1245–1251.
- 119.↵Morris GF, Juul N, Marshall SB, Benedict B, Marshall LF. Neurological deterioration as a potential alternative endpoint in human clinical trials of experimental pharmacological agents for treatment of severe traumatic brain injuries: Executive Committee of the International Selfotel Trial. Neurosurgery. 1998; 43: 1369–1372.
- 120.↵Kondziolka D, Wechsler L, Goldstein S, Meltzer C, Thulborn KR, Gebel J, Jannetta P, DeCesare S, Elder EM, McGrogran M, Reitman MA, Bynum L. Transplantation of cultured human neuronal cells for patients with stroke. Neurology. 2000; 55: 565–569.
- 122.↵Gladstone DJ, Danells CJ, Black SE. The Fugl-Meyer Assessment of motor recovery after stroke: a critical review of its measurement properties. Neurorehabil Neural Repair. (In Press).
- 124.↵Williams LS, Weinberger M, Harris LE, Clark DO, Biller J. Development of a stroke-specific quality of life scale. Stroke. 1999; 30: 1362–1369.
- 125.↵Finklestein S. Molecular and cellular therapies for stroke recovery. Presented at Neuroplasticity: The Key to Stroke Recovery; March 2000; Kananaskis, Alberta, Canada.
- 126.↵Lai SM, Duncan PW, Keighley J. Prediction of functional outcome after stroke: comparison of the Orpington Prognostic Scale and the NIH Stroke Scale. Stroke. 1998; 29: 1838–1842.
- 127.↵Sandercock P, Hennerici MG, Orgogozo JM, Davis SM, Gorelick PB. Mega trials versus small trials in stroke. In: Fisher M, Bogousslavsky J, eds. Current Review of Cerebrovascular Disease. Philadelphia, Pa: Current Medicine, Inc; 2001: 241–246.
- 128.↵Malakoff D. Bayes offers a “new” way to make sense of numbers. Science. 1999; 286: 1460–1464.
- 129.↵Stroke Trials Directory. Stroke Trial - UK-279,276. Available at: http://www.strokecenter.org/trials/list/trialPage188.htm. Accessed February 2002.
- 134.↵Johansson BB. Brain plasticity and stroke rehabilitation: the Willis Lecture. Stroke. 2000; 31: 223–230.
- 140.↵Hara H, Friedlander RM, Gagliardini V, Ayata C, Fink K, Huang Z, Shimizu-Sasamata M, Yuan J, Moskowitz MA. Inhibition of interleukin 1beta converting enzyme family proteases reduces ischemic and excitotoxic neuronal damage. Proc Natl Acad Sci U S A. 1997; 94: 2007–2012.
- 142.↵Schulz JB, Weller M, Matthews RT, Heneka MT, Groscurth P, Martinou JC, Lommatzsch J, von Coelln R, Wullner U, Loschmann P, Beal MF, Dichgans J, Klockgether T. Extended therapeutic window for caspase inhibition and synergy with MK-801 in the treatment of cerebral histotoxic hypoxia. Cell Death Differ. 1998; 5: 847–857.
- 144.↵Sacchetti ML, Toni D, Fiorelli M, Argentino C, Fieschi C. The concept of combination therapy in acute ischemic stroke. Neurology. 1997; 49: S70–S74.
- 147.↵Andersen M, Overgaard K, Meden P, Boysen G, Choi SC. Effects of citicoline combined with thrombolytic therapy in a rat embolic stroke model. Stroke. 1999; 30: 1464–1471.
- 151.↵Sereghy T, Overgaard K, Boysen G. Neuroprotection by excitatory amino acid antagonist augments the benefit of thrombolysis in embolic stroke in rats. Stroke. 1993; 24: 1702–1708.
- 152.↵Bowes MP, Rothlein R, Fagan SC, Zivin JA. Monoclonal antibodies preventing leukocyte activation reduce experimental neurologic injury and enhance efficacy of thrombolytic therapy. Neurology. 1995; 45: 815–819.
- 153.↵Zhang RL, Zhang ZG, Chopp M. Increased therapeutic efficacy with rt-PA and anti-CD18 antibody treatment of stroke in the rat. Neurology. 1999; 52: 273–279.
- 154.↵Lyden P, Jacoby M, Schim J, Albers G, Mazzeo P, Ashwood T, Norlund A, Odergren T, for the CLASS IHT Investigators. The Clomethiazole Acute Stroke Study in Tissue-Type Plasminogen Activator-Treated Stroke (CLASS-T): final results. Neurology. 2001; 57: 1199–1205.
- 157.↵Lyden PD, Lonzo L. Combination therapy protects ischemic brain in rats. a glutamate antagonist plus a gamma-aminobutyric acid agonist. Stroke. 1994; 25: 189–196.
- 160.↵Uematsu D, Araki N, Greenberg JH, Sladky J, Reivich M. Combined therapy with MK-801 and nimodipine for protection of ischemic brain damage. Neurology. 1991; 41: 88–94.
- 161.↵Onal MZ, Li F, Tatlisumak T, Locke KW, Sandage BWJ, Fisher M. Synergistic effects of citicoline and MK-801 in temporary experimental focal ischemia in rats. Stroke. 1997; 28: 1060–1065.
- 166.↵Schabitz WR, Li F, Irie K, Sandage BW Jr, Locke KW, Fisher M. Synergistic effects of a combination of low-dose bFGF and citicoline after temporary experimental focal ischemia. Stroke. 1999; 30: 427–432.
- 171.↵Hasbani MJ, Schlief ML, Fisher DA, Goldberg MP. Dendritic spines lost during glutamate receptor activation reemerge at original sites of synaptic contact. J Neurosci. 2001; 21: 2393–2403.
- 173.↵Ay I, Ay H, Koroshetz WJ, Finklestein SP. Growth factors and cerebral ischemia. In: Fisher M, Bogousslavsky J, eds. Current Review of Cerebrovascular Disease. Philadelphia, Pa: Current Medicine, Inc; 2001: 25–33.
- 178.↵Fisher M, Finklestein S. Pharmacological approaches to stroke recovery. Cerebrovasc Dis. 1999; 9: 29–32.
- 180.↵Bogousslavsky J, Donnan G. Fibroblast (trafermin) in acute stroke: results of the European-Australian phase II/III safety and efficacy trial. Cerebrovasc Dis. 2000; 10 (suppl 2): 106.Abstract.
- 182.↵Osuga H, Osuga S, Wang F, Fetni R, Hogan MJ, Slack RS, Hakim AM, Ikeda JE, Park DS. Cyclin-dependent kinases as a therapeutic target for stroke. Proc Natl Acad Sci U S A. 2000; 97: 10254–10259.
- 183.↵Li Y, Chen J, Wang L, Lu M, Chopp M. Treatment of stroke in rat with intracarotid administration of marrow stromal cells. Neurology. 2001; 56: 1666–1672.
- 184.↵Chen J, Li Y, Wang L, Zhang Z, Lu D, Lu M, Chopp M. Therapeutic benefit of intravenous administration of bone marrow stromal cells after cerebral ischemia in rats. Stroke. 2001; 32: 1005–1011.
- 185.↵Chen J, Sanberg PR, Li Y, Wang L, Lu M, Willing AE, Sanchez-Ramos J, Chopp M. Intravenous administration of human umbilical cord blood reduces behavioral deficits after stroke in rats. Stroke. 2001; 32: 2682–2688.
- 188.↵Nudo RJ, Wise BM, SiFuentes F, Milliken GW. Neural substrates for the effects of rehabilitative training on motor recovery after ischemic infarct. Science. 1996; 272: 1791–1794.
- 189.↵Nudo RJ, Milliken GW. Reorganization of movement representations in primary motor cortex following focal ischemic infarcts in adult squirrel monkeys. J Neurophysiol. 1996; 75: 2144–2149.
- 192.↵Staines WR, McIlroy WE, Graham SJ, Black SE. Bilateral movement enhances ipsilesional cortical activity in acute stroke: a pilot fMRI Study. Neurology. 2001; 56: 401–404.
- 195.↵Goldstein LB. Potential effects of common drugs on stroke recovery. Arch Neurol. 1999; 55: 454–456.
- 196.↵Dobkin BH. Neurorehabilitation: greater plasticity through chemicals and practice. Neurol Network Commentary. 1998; 2: 171–174.
- 197.↵Feeney DM, Gonzalez A, Law WA. Amphetamine, haloperidol, and experience interact to affect rate of recovery after motor cortex injury. Science. 1982; 217: 855–857.
- 199.↵Goldstein LB, Davis JN. Post-lesion practice and amphetamine-facilitated recovery of beam-walking in the rat. Restorative Neurol Neurosci. 1990; 2: 311–314.
- 202.↵Stroemer RP, Kent TA, Hulsebosch CE. Neocortical neural sprouting, synaptogenesis, and behavioral recovery after neocortical infarction in rats. Stroke. 1995; 26: 2135–2144.
- 203.↵Stroemer RP, Kent TA, Hulsebosch CE. Enhanced neocortical neural sprouting, synaptogenesis, and behavioral recovery with d-amphetamine therapy after neocortical infarction in rats. Stroke. 1998; 29: 2381–2395.
- 207.↵Feeney DM. Rehabilitation pharmacology: noradrenergic enhancement of physical therapy. In: Ginsberg MD, Bogousslavsky J, eds. Cerebrovascular Disease: Pathophysiology, Diagnosis and Management. Oxford, UK: Blackwell Science; 1998: 620–636.
- 209.↵Walker-Batson D, Smith P, Curtis S, Unwin H, Greenlee R. Amphetamine paired with physical therapy accelerates motor recovery after stroke. Stroke. 1995; 26: 2254–2259.
- 210.↵Volpe BT, Shelton F, Krebs HI, Hogan N, Diels C, Reding M. Dextroamphetamine paired with standard and robotic neurorehabilitation changes motor performance one day later. Ann Neurol. 2000; 48: 454.Abstract.
- 211.↵Gladstone DJ, Black SE, Danells CJ, Staines WR, McIlroy WE, Graham SJ, Herrmann N, Szalai JP. Enhancing recovery from hemiplegic stroke: design of a multicentre study of amphetamine-facilitated stroke rehabilitation. Clin Invest Med. 2001; 24: 219.Abstract.
- 213.↵Walker-Batson D, Curtis S, Natarajan R, Ford J, Dronkers N, Salmeron E, Lai J, Unwin DH. A double-bind, placebo-controlled study of the use of amphetamine in the treatment of aphasia. Stroke. 2001; 32: 2093–2098.
- 215.↵Walker-Batson D. Pharmacotherapy in the treatment of aphasia. In: Goldstein LB, ed. Restorative Neurology: Advances in Pharmacotherapy for Recovery After Stroke. New York, NY: Futura Publishing; 1998: 257–270.
- 219.↵Dam M, Tonin P, De Boni A, Pizzolato G, Casson S, Ermani M, Freo U, Piron L, Battistin L. Effects of fluoxetine and maprotiline on functional recovery in poststroke hemiplegic patients undergoing rehabilitation therapy. Stroke. 1996; 27: 1211–1214.
- 220.↵Goldstein LB, Davis JN. Physician prescribing patterns following hospital admission for ischemic cerebrovascular disease. Neurology. 1988; 38: 1806–1809.
- 221.↵Goldstein LB. Prescribing of potentially harmful drugs to patients admitted to hospital after head injury. J Neurol Neurosurg Psychiatry. 1995; 58: 753–755.
- 222.↵Goldstein LB, for the Sygen in Acute Stroke Study Investigators. Common drugs may influence motor recovery after stroke. Neurology. 1995; 45: 865–871.
- 223.↵Graham GD, Ahmed W, Davis LE, Bryniarski E, Hansen MD, Woolson RF, Clark WR, Adams HP. Effects of commonly prescribed medications on stroke recovery: a TOAST study analysis. Stroke. 1999; 30: 236.Abstract.
- 224.↵Porch B, Wyckes J, Feeney DM. Haloperidol, thiazides and some antihypertensives slow recovery from aphasia. Neurosci Abstracts. 1985; 11: 52.