Published online before print March 18, 2004, doi: 10.1161/01.STR.0000125010.70652.93
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
(Stroke. 2004;35:1216.)
© 2004 American Heart Association, Inc.
Comments, Opinions, and Reviews |
Sample Size Calculations in Acute Stroke Trials: A Systematic Review of Their Reporting, Characteristics, and Relationship With Outcome
From Institute of Neurosciences (C.S.W., P.M.W.B.), Institute of Clinical Research (J.L.-B.), and School of Nursing (F.J.B.-H.), University of Nottingham, UK.
Correspondence to Philip Bath, Division of Stroke Medicine, Institute of Neuroscience, University of Nottingham, City Hospital Campus, Nottingham NG5 1PB, UK. E-mail philip.bath{at}nottingham.ac.uk
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Background and Purpose— Only a few randomized controlled trials in acute stroke have shown a treatment-related benefit. Inadequate trial design, especially low sample size, may partly explain this failure. We investigated sample size calculations (SSCs) in a systematic review of acute stroke trials.
Methods— Full reports of nonconfounded randomized controlled trials that recruited patients within 1 week of stroke onset and were published before the end of 2001 were identified from the Cochrane Library and other bibliographic databases. Information on the SSC and outcome event rates was collected for each trial.
Results— Of 189 identified trial reports, 57 (30%) reported 
1 components of the SSC, phase II 14/129 (11%) versus phase III 43/60 (72%) (P<0.001), with 32 (56%) giving all the required parameters. Significance (
) was mentioned in 54 (96%) reports; 53 used a significance level of =0.05. And 55 (98%) reports gave the power (1–ß) of the study (median [25th and 75th percentile] 0.80 [0.80, 0.90]). The anticipated percentage of control subjects having a primary outcome event was given in 24 (42%) articles: case fatality 21.8% (11.8%, 23.5%, n=4) and combined death or disability/dependency 55.5% (44.5%, 66.3%, n=20); 25 studies used other outcomes and 8 studies gave insufficient information. Four of the 22 trials achieved a control rate within 5% of their prediction. 49 (86%) reports gave the anticipated treatment effect; case fatality: anticipated 9.5% (1.1%, 12.5%, n=6), achieved –0.3% (–4.1%, +2.4%); combined death or disability/dependency: anticipated 13.0% (10.0%, 16.0%, n=25), achieved 1.8% (–0.5%, +5.4%). The median calculated sample size was 600 (198, 995, n=54).
Conclusions— Too few trial publications report the assumptions underlying their SSC. Most trials were underpowered, ie, power <0.90, used inappropriate assumptions for event rates, and were grossly overoptimistic in their expectation of treatment effect. These deficiencies will together have resulted in trials being far too small and reduced their chance of being able to detect real treatment effects.
Key Words: stroke • randomized controlled trials • acute stroke • systematic review
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Although several hundred trials have been completed in acute stroke, only a few assessing efficacy have been positive, eg, relating to aspirin, alteplase, and prourokinase.1–3 Many explanations have been put forward to explain this situation, eg, findings in animals may be irrelevant to humans,4,5 drugs are ineffective6–10 or detrimental,11–13 and trial design may be suboptimal, using inappropriate outcomes4,14 or statistical analyses.15 To address these problems, the Stroke Therapy Academic Industry Roundtable (STAIR) group created guidelines with the aim of improving future experimental and clinical trials.16–18
In an earlier study, we found that the reporting of trials was often suboptimal.19 One result in that systematic review was that components of the sample size calculation (SSC) were only published in 33 of 114 trials,19 and that these studies were mostly small and underpowered. We hypothesized that a further reason why trials might be failing could relate to their design in the context of an inadequate sample size. Conventionally, a trial’s sample size is estimated mathematically from 4 assumptions if a dichotomous outcome measure is being assessed: (1) anticipated treatment effect; (2) event rate in the control group; (3) desired power; and (4) desired significance. Inappropriate choice of values for these parameters would lead to trials being too small. Here, we systematically assess the reporting and parameters used in SSC in trials of acute stroke.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Trial Eligibility
We assessed full and final reports of nonconfounded randomized controlled trials that recruited patients within 1 week of stroke onset. Studies published in a peer-reviewed journal by the end of 2001 were included. Trials that had been published in abstract-form alone or as a chapter in a book, had a crossover design, only compared active treatments, or primarily involved patients with transient ischemic attack, subarachnoid hemorrhage, or nonstroke conditions were excluded.
Identification of Acute Stroke Trials
Potential trials for inclusion were identified from the Cochrane Controlled Trials Register (Cochrane Library, issue 1, 2002). Further searches were made of the EMBASE and MEDLINE bibliographic databases, a previous review of acute stroke trials,19 and a book detailing trials in acute stroke.20 The key words used in searches were: randomized controlled trial, controlled clinical trial, acute stroke, ischemic stroke, cerebrovascular disorders, and cerebral hemorrhage (see Figure 1).
|
Data Extraction
Each publication was scrutinized and the data were abstracted by 2 authors (C.S.W., J.L.-B.) for information about the trial. Components of the SSC were recorded (Table 1): significance level (), power (1–ß), anticipated rates of primary outcome in control group (p1) and treatment group (p2), and intended absolute treatment effect (P=p2–p1). When necessary, relative or proportional effects were converted to absolute effects. Data were excluded if based on mean or median scores. Achieved sample size, outcome rates in control and treatment groups, and actual treatment effect (positive, neutral, or negative) were also recorded (Table 1). Trials were defined as phase II if they studied safety, tolerability, feasibility, and/or surrogate measures of outcome, and phase III if they stated that their primary intention was to assess efficacy, usually judged on death, combined death and disability/dependency, or impairment. Data were double-entered (C.S.W., J.L.-B.) into a database (FileMaker Pro v4.1; Claris, Santa Clara) and cross-checked to ensure accuracy. Any disagreements about the interpretation of the information in the report were discussed and resolved by P.M.W.B. A questionnaire was sent out to all first-named authors to see if they could provide any further information to the data extracted.
|
Analysis
Nonparametric descriptors (median, 25th to 75th percentile) and tests (
2 with Yates correction, Spearman correlation) were used because most of the data were of an ordered categorical (ordinal) or binary (nominal) type. Analysis was performed using the SPSS statistical package (v10.0.7a for Apple PowerPC; SPSS Corp).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Trials
The searches found 189 trial reports (phase II, n=129; phase III n=60) that fulfilled the inclusion criteria. A total of 82 716 patients had been included, with individual trials enrolling between 12 and 21 106 subjects, median (25th to 75th percentile), 89 (40 to 294). The studies were reported over a period from 195621 to the end of 2001. Trial size had increased during this period (rs=0.387, P<0.001).
Sample Size Calculation
Of the 189 reports, 57 (30%) reported 1 components of the SSC (Table 1); 32/57 (56%) reports gave all the required parameters. Although the first trial in acute stroke was published in 1956, the first to give a SSC was in 1983.22 The proportion of trials reporting a SSC increased after the publication of the CONSORT statement (which recommended that SSC should be reported in trial publications in 199623). Before CONSORT, 27/135 (20%) included SSC; after CONSORT, 29/54 (54%) included SSC (2=19.9 P<0.001; see Figure 2). A higher proportion of phase III trials reported a SSC in comparison with phase II trials: phase II, 14/129 (11%); phase III, 43/60 (72%) (2=70.1, P<0.001). A value for significance () was present in 55 of 57 (96%) reports; 54 of these used =0.05 whereas 1 used =0.025.24 Power (1–ß) was given in 56 of 57 (98%) reports; the median power was 0.80 (25th to 75th percentile 0.80 to 0.90, range 0.75 to 0.95).
|
Trial Results
Of the 57 trials, 8 were positive on their primary outcome (active treatment superior to control), 43 were neutral, and 6 were negative (control superior to active treatment). When assessing just phase III trials (Tables 2 and 3
), 4 were positive: CAST (aspirin), NINDS (alteplase), STAT (ancrod), and PROACT II (pro-urokinase);1–3,25 35 were neutral; and 3 were negative: EAST (enlimomab), MAST-E (streptokinase), and ATLANTIS A (alteplase).13,26,27
|
|
Functional Outcome
The anticipated percentage of control subjects having an event for the trial’s primary outcome was reported in 24 of 57 (42%) articles: case fatality, 21.8% (11.8% to 23.5%, n=4); combined death or disability/dependency, 55.0% (44.5% to 66.3%, n=20). And 25 studies used another outcome, such as change in neurological score (n=11), ordinal comparison of functional outcome (n=7), rate of spontaneous recanalization (n=1), percentage of patients achieving a National Institute of Health Stroke Scale (NIHSS) score of 
1 (n=1), or another approach (n=5). Eight studies gave insufficient information. The median anticipated rate of combined death and dependency/disability was 55.0%, with that for death being 21.8%. Table 4 shows the full results. The achieved rates of death, or combined death and dependency, in control patients tended to be lower than planned; slightly >50% of studies came within 20% of their target. Table 5 shows a more detailed examination. Only 7 (12.5%) trials stated that the control rate estimate was obtained from a previous study.
|
|
Forty-eight of 57 (84%) reports gave the desired absolute treatment effect (P), reported either directly or by calculation of the difference between active and control event rates (p2–p1). The intended reduction in 6 trials using case fatality was 9.5% (1.1% to 12.5%) whereas the achieved result was –0.3% (–4.1% to +2.4%). Similarly, the intended reduction in 24 trials using combined death and disability/dependency was 12.0% (10.0% to 15.3%), with the final result being 1.9% (–0.5% to +5.4%). The remaining 18 studies did not use these primary outcomes.
The median calculated sample size was 600 (199 to 995, n=54); 3 trial reports did not give the result of their SSC. The calculated sample size by type of trials was: phase II, 69 (40 to 116, n=11); phase III, 722 (400 to 1200, n=39). For phase III studies, the calculated sample size by the trial’s primary result were: positive, 560 (215 to 15 200, n=4);1–3,28 neutral, 775 (426 to 1200, n=32); and negative 600 (300 to 600, n=3).13,26,27
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The relative failure to find effective treatments for acute stroke has led to much speculation as to the causes. A consistent theme is that trial design has been suboptimal, although this statement has been difficult to quantify. We show here that issues relating to the size of trials may explain some trial failures. In particular, most trial publications failed to report the SSC in its entirety, thereby making it impossible to repeat the calculation and assess the assumptions used. Furthermore, those trials reporting aspects of SSC were often underpowered, assumed unrealistic event rates and intervention effects, or used inappropriate primary outcomes (eg, death).
Alarmingly, most trials did not report a SSC. It is common, although not correct, for small "phase II" trials to not perform or report a SSC on the grounds that dose, safety, feasibility, and/or tolerability are being assessed in a few tens of patient. Nevertheless, such acute stroke trials should, ideally, have a primary outcome (eg, adverse events in a dose escalation trial29 or the effect of treatment on a surrogate measure such as blood pressure30,31); therefore, they should base their sample size on SSC. In some cases, no pilot data exist and the aim of the trial is to develop this for use in the design of a future larger trial, in which case SSC may not be possible. Surprisingly, only three-quarters of "phase III" trials, in which efficacy was being studied, presented SSC, although all the trials stated they had a primary outcome. The 1996 CONSORT statement requires that trial reports reported a SSC,23 a message reinforced in the recently updated CONSORT statement.32 We found that significantly more trials published a SSC after publication of the CONSORT statement.
Many forms of SSCs can be used when designing a clinical trial. An important determinant is whether the outcome is continuous (eg, blood pressure), ordinal (ordered categorical, eg, modified Rankin Scale), or binary (nominal, eg, death) in nature. Continuous and ordinal data can be "collapsed" to produce binary data by inserting cut-off points, eg, combined death or dependency can be defined as a modified Rankin Scale >2, although this reduces statistical power. We found that the majority of acute stroke trials reporting a SSC performed an analysis based on a dichotomous outcome. In this situation, the Trialists needed to choose levels for significance () and power (1–ß) and to make assumptions about the proportion of subjects achieving the outcome in the control group (p1) and the desired treatment effect (p2–p1).33 From these values/numbers, the sample size (N) can be calculated:
|
|
where 
(, ß) is a function of the cumulative distribution function of a standardized normal deviate.33
This formula can be adapted to include adjustments for small sample sizes (Yate correction), multiple active groups, groups of different sizes, and multiple comparisons; all of these extensions lead to an increased sample size. Only 1 of the trials using a binary sample size formula extended this to account for multiple comparisons.24
Statistical convention sets significance at an arbitrary level of =0.0534 and all but 1 trial24 used this figure. Most trials used a power of 0.80, although this ranged from 0.75 to 0.95. Many authorities recommend that a power of 0.90 should be used in clinical efficacy (phase III) studies. The power of the study relates to the probability (ß=1–power) of missing a genuine difference between the treatment and control groups. It is questionable whether it is worthwhile for Trialists, funding sources (whether academic, charity, government, or commercial), or ethics committees to support studies in which there is a 20% chance of missing a significant effect, ie, in which the power is only 0.80, because an effective drug could be discarded and lost forever if a trial was falsely neutral (type-II error).
The most difficult part of setting parameters for the SSC is choosing appropriate event rates for the control and active groups. The control rate (p1) should usually be obtained from recent historical data relating to a group of stroke patients similar to those likely to be enrolled into the planned trial, eg, from stroke registers or similar trials. However, estimating event rates from registers is fraught, because outcome tends to be worse than in trials,35 so analysis of data in registers should be restricted to trial-eligible patients. Either way, Trialists should report and reference how they chose the estimate for control event rate.
The event rate in the treatment group (p2) is usually determined from the control group event rate (p1) and an assumption about the anticipated absolute treatment effect (P, where p2=P–p1). Almost all trials reporting P (or both p1 and p2) grossly overestimated the anticipated absolute treatment effect in reducing combined death and dependency/disability; the median value across the trials reporting SSC was 12%. Of the 2 "proven" treatments for acute stroke, only alteplase achieved this treatment effect while aspirin managed slightly >1%.1,2,36 It is clear that future trials must assume smaller treatment effects and that the chosen figure will depend on a number of factors. First, drugs acting early in the pathophysiological cascade of acute stroke are likely to be more efficacious than agents acting later, eg, thrombolytics versus antithrombotics and neuroprotectants. Second, expensive drugs (costing hundreds or thousands of euro) will probably need to demonstrate larger effects than inexpensive agents (costing a few euro) if they are to be used widely. Which treatment rate is appropriate is unclear, but a survey of stroke physicians suggested that an absolute reduction in combined death and dependency/disability of 5% absolute risk reduction would be worthwhile clinically. In general, efficacy should be judged on the basis of reducing death or dependency rather than death alone, because the former is of more relevance to patients, their carers, and society. Using death and dependency as the primary outcome will also tend to limit sample size as compared with death because, all other things being equal, event rates of 
50% maximize statistical power.
An adjustment may be necessary during the trial limiting recruitment to particular groups of patients, eg, severe or mild impairment, to ensure that the overall outcome rate (average of p1 and p2) approximates to that assumed in the SSC, as performed in the TAIST trial.7 In this respect, it is vital to estimate severity using a standardized and validated stroke impairment scale, eg, Scandinavian Stroke Scale. Ensuring that the SSC assumptions are achieved is further complicated because the relationship between severity and outcome is changing, ie, outcome has improved with time for a given severity, as seen when comparing the ECASS and ECASS II trials.37,38 Unfortunately, most trials in our analysis did not report their assumptions relating to control event rates; when these data were given, rates of 55% for combined death or disability/dependency and 22% for death were assumed on average.
Combining the need for increasing power and reducing the anticipated treatment effect means that future efficacy trials will need to be substantially larger than recent studies. Such studies may need to include 5000 patients, assuming significance of 0.05, power of 0.90, control event rate of 50%, and absolute treatment effect of 5%.39 Because the total cost of future trials cannot increase dramatically in real terms, a balance between recruiting more patients and collecting less data on each patient will need to be found. Modern approaches to randomization and data registration, eg, based on the Internet,40 will also help in reducing costs.
In summary, Trialists should ensure, and editors and referees should insist, that publications give detailed information on the SSC in acute stroke trials. When designing trials, SSC should use appropriate parameters as detailed here. Combining this approach and improving the rigor of other design issues will help increase the success rate of future trials.
|
Acknowledgments |
|---|
The full list of trials included in this systematic review is given at www.nottingham.ac.uk/stroke-medicine/acutetrialslist.htm. We thank those who responded to our request for further information on their trials. P.M.W.B. is Stroke Association Professor of Stroke Medicine. The Division of Stroke Medicine receives core funding from The Stroke Association. P.M.W.B. was a member of the STAIR III group.18 No pharmaceutical company was involved in the design, execution, or interpretation of this study. The work was presented, in part, at the 10th European Stroke Conference, Lisbon, May 2001.41
Received September 10, 2003; revision received January 7, 2004; accepted January 15, 2004.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
1. CAST (Chinese Acute Stroke Trial) Collaborative Group. CAST: randomised placebo-controlled trial of early aspirin use in 20,000 patients with acute ischaemic stroke. Lancet. 1997; 349: 1641–1649.[CrossRef][Medline] [Order article via Infotrieve]
2. The National Institute of Neurological Disorders and Stroke rt-PA Stroke Study Group. Tissue plasminogen activator for acute stroke. N Engl J Med. 1995; 333: 1581–1587.
3. 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 trial. JAMA. 1999; 282: 2003–2011.
4. Grotta J. Why do all drugs work in animals but none in stroke patients? 2. Neuroprotective therapy. J Intern Med. 1995; 237: 89–94.[Medline] [Order article via Infotrieve]
5. del Zoppo GJ. Why do all drugs work in animals but none in stroke patients? 1. Drugs promoting cerebral blood flow. J Intern Med. 1995; 237: 79–88.[Medline] [Order article via Infotrieve]
6. Bath PMW, Iddenden R, Bath FJ. Low molecular weight heparins and heparinoids in acute ischaemic stroke: a systematic review. Stroke. 2000; 31: 311–314.
7. Bath P, Lindenstrom E, Boysen G, De Deyn P, Friis P, Leys D, Marttila R, Olsson J-E, O’Neill D, Orgogozo J-M, Ringelstein B, van der Sande J-J, Turpie AGG, for the TAIST Investigators. Tinzaparin in acute ischaemic stroke (TAIST): a randomised aspirin-controlled trial. Lancet. 2001; 358: 702–710.[CrossRef][Medline] [Order article via Infotrieve]
8. Horn J, Limburg M. Calcium antagonists for ischemic stroke: a systematic review. Stroke. 2001; 32: 570–576.
9. Gandolfo C, Sandercock P, Conti M. Lubeluzole for acute ischaemic stroke. Cochrane Database Syst Rev. 2002: CD001924.
10. Lees KR, Muir KW. Excitatory amino acid antagonists for acute stroke (Cochrane Review). In: The Cochrane Library. Oxford: Update Software; 2002.
11. Davis SM, Lees KR, Albers GW, Diener HC, Markabi S, Karlsson G, Norris J, for the ASSIST Investigators. Selfotel in acute ischemic stroke. Possible neurotoxic effects of an NMDA antagonist. Stroke. 2000; 31: 347–354.
12. Tirilazad International Steering Committee. Tirilazad mesylate in acute ischaemic stroke. A systematic review. Stroke. 2000; 32: 2257–2265.
13. Investigators EAST. Use of anti-ICAM-1 therapy in ischemic stroke. Neurology. 2001; 57: 1428–1434.
14. Hacke W, Bluhmki E, Steiner T, Tatlisumak T, Mahagne M-H, Sacchetti M-L, Meier D. Dichotomized efficacy end points and global end-point analysis applied to the ECASS intention to treat data set. Post hoc analysis of ECASS 1. Stroke. 1998; 29: 2073–2075.
15. Optimising Analysis of Stroke Trials (OAST) Collaborators, Bath PMW. Optimising the statistical analysis of functional outcome in stroke trials. Stroke. 2003; 34: 316. Abstract.
16. Stroke Therapy Academic Industry Roundtable (STAIR). Recommendations for standards regarding preclinical neuroprotective and restorative drug development. Stroke. 1999; 30: 2752–2758.
17. Stroke Therapy Academic Industry Roundtable II (STAIR-II). Recommendations for clinical trial evaluation of acute stroke therapies. Stroke. 2001; 32: 1598–1606.
18. Fisher M, for the Stroke Therapy Academic Industry Roundtable. Recommendations for advancing development of acute stroke therapies. Stroke Therapy Academic Industry Roundtable 3. Stroke. 2003; 34: 1539–1546.
19. Bath FJ, Owen VE, Bath PMW. Quality of full and final publications reporting acute stroke trials. A systematic review. Stroke. 1998; 29: 2203–2210.
20. Bath P, Albers G. Controlled Clinical Trials: Stroke. 1st ed. London: Science Press; 2001.
21. Dyken ML, White P. Evaluation of cortisone in the treatment of cerebral infarction. JAMA. 1956; 162: 1531–1534.
22. Hossmann V, Heiss WD, Bewermeyer H, Wiedemann G. Controlled trial of ancrod in ischemic stroke. Arch Neurol. 1983; 40: 803–808.
23. Statement TC. Improving the quality of reporting of randomized controlled trials. JAMA. 1996; 276: 637–639.
24. Hennerici M, Kramer G, North PM, Schmitz H, Tettenborn D, Nimodipine European Stroke Trial Group. Nimodipine in the treatment of acute MCA ischemic stroke (NEST). Cerebrovasc Dis. 1994; 4: 189–193.
25. Sherman DG, Atkinson RP, Chippendale T, Levin KA, Ng K, Futrell N, Hsu CY, Levy DE, Participants TS. Intravenous ancrod for treatment of acute ischemic stroke. The STAT study: a randomised controlled trial. JAMA. 2000; 283: 2395–2402.
26. The Multicenter Acute Stroke Trial - Europe Study Group. Thrombolytic therapy with streptokinase in acute ischemic stroke. N Engl J Med. 1996; 335: 145–150.
27. Clark WM, Albers GW, Madden KP, Hamilton S. The rtPA (alteplase) 0–6 hour acute stroke trial, part A (A0276g). Results of a double-blind, placebo-controlled, multicenter study. Stroke. 2000; 31: 811–816.
28. Lenzi GL, Grigoletto F, Gent M, Roberts RS, Walker MD, Easton JD, Carolei A, Dorsey FC, Rocca WA, Bruno R, Patarnello F, Fieschi C, Early Stroke Trial Group. Early treatment of stroke with monosialoganglioside GM-1. Efficacy and safety results of the Early Stroke Trial. Stroke. 1994; 25: 1552–1558.[Abstract]
29. The Abciximab in Ischemic Stroke Investigators. Abciximab in acute ischemic stroke. A randomised, double-blind, placebo-controlled, dose-escalation study. Stroke. 2000; 31: 601–609.
30. Bath PM, Pathansali R, Iddenden R, Bath FJ. The effect of transdermal glyceryl trinitrate, a nitric oxide donor, on blood pressure and platelet function in acute stroke. Cerebrovasc Dis. 2001; 11: 265–272.[CrossRef][Medline] [Order article via Infotrieve]
31. Rashid P, Weaver C, Leonardi-Bee JA, Fletcher S, Bath FJ, Bath PMW. The effects of transdermal glyceryl trinitrate, a nitric oxide donor on blood pressure, cerebral and cardiac haemodynamics and plasma nitric oxide levels in acute stroke. J Stroke Cerebrovasc Dis. 2003; 13: 143–151.[CrossRef]
32. Moher D, Schultz KF, Altman DG. The Consort statement: revised recommendations for improving the quality of reports of parallel-group randomised trials. Lancet. 2001; 357: 1191–1194.[CrossRef][Medline] [Order article via Infotrieve]
33. Pocock SJ. Clinical trials: a practical approach. Chichester: Wiley; 1983.
34. Sterne JAC, Smith GD. Sifting the evidence - what’s wrong with significance tests? BMJ. 2001; 322: 226–231.
35. Barer DH, Ebrahim SB, Mitchell JRA. The pragmatic approach to stroke trial design: stroke register, pilot trial, assessment of neurological then functional outcome. Neuroepidemiology. 1988; 7: 1–12.[Medline] [Order article via Infotrieve]
36. International Stroke Trial Collaborative Group. The International Stroke Trial (IST); a randomised trial of aspirin, subcutaneous heparin, both, or neither among 19435 patients with acute ischaemic stroke. Lancet. 1997; 349: 1569–1581.[CrossRef][Medline] [Order article via Infotrieve]
37. Hacke W, Kaste M, Fieschi C, Toni D, Lesaffre E, von Kummer R, Boysen G, Bluhmki E, Hoxter G, Mahagne M-H, Hennerici M, Group. ES. Intravenous thrombolysis with tissue plasminogen activator for acute hemispheric stroke. The European Cooperative Acute Stroke Trial (ECASS). JAMA. 1995; 274: 1017–1025.
38. Hacke W, Markku K, 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.[CrossRef][Medline] [Order article via Infotrieve]
39. Bath P, On behalf of the TAIST Advisory Committee. Early anticoagulation in acute ischaemic stroke. Lancet. 2002; 359: 524.[Medline] [Order article via Infotrieve]
40. Bath PM, Young S, Watson G, Weaver C. Improving trials in acute stroke using Internet randomisation and data collection. Cerebrovasc Dis. 2003; 16: 100. Abstract.
41. Weaver C, Leonardi-Bee J, Bath FJ, Bath PMW. Sample size calculations in acute stroke trials: their reporting and relationship with outcome. Cerebrovasc Dis. 2001; 11: 77. Abstract.
42. Duke RJ, Bloch RF, Turpie AGG, Trebilcock RG, Bayer N. Intravenous heparin for the prevention of stroke progression in acute partial stable stroke. Ann Intern Med. 1986; 105: 825–828.
43. Norris JW, Hachinski VC. High dose steroid treatment in cerebral infarction. BMJ. 1986; 292: 21–23.
44. Italian Acute Stroke Study Group. Haemodilution in acute stroke: results of the Italian Haemodilution Trial. Lancet. 1988; 1: 318–321.[Medline] [Order article via Infotrieve]
45. Scandinavian Stroke Study Group. Multicenter trial of hemodilution in acute ischemic stroke. I. Results in the total patient population. Scandinavian Stroke Study Group. Stroke. 1987; 18: 691–699.
46. Hsu CY, Norris JW, Hogan EL, Bladin P, Dinsdale HB, Yatsu FM, Earnest MP, Scheinberg P, Caplan LR, Karp HR. Pentoxifylline in acute nonhemorrhagic stroke. A randomized, placebo-controlled double-blind trial. Stroke. 1988; 19: 716–722.
47. Barer DH, Cruickshank JM, Ebrahim SB, Mitchell JR. Low dose beta blockade in acute stroke ("BEST" trial): an evaluation. BMJ. 1988; 296: 737–741.
48. TRUST Study Group. Randomised, double-blind, placebo-controlled trial of nimodipine in acute stroke. Trust Study Group. Lancet. 1990; 336: 1205–1209.[CrossRef][Medline] [Order article via Infotrieve]
49. American Nimodipine Study Group. Clinical trial of nimodipine in acute ischemic stroke. Stroke. 1992; 23: 3–8.
50. Yu YL, Kumana CR, Lauder IJ, Cheung YK, Chan FL, Kou M, Chang CM, Cheung RTF, Fong KY. Treatment of acute cerebral hemorrhage with intravenous glycerol. A double-blind, placebo-controlled randomised trial. Stroke. 1992; 23: 967–971.
51. Kaste M, Fogelholm R, Erila T, Palomaki H, Murros K, Rissanen A, Sarna S. A randomised, double-blind, placebo-controlled trial of nimodipine in acute ischemic hemispheric stroke. Stroke. 1994; 25: 1348–1353.[Abstract]
52. The RANTTAS Investigators. A randomized trial of tirilazad mesylate in patients with acute stroke (RANTTAS). Stroke. 1996; 27: 1453–1458.
53. Franke CL, Palm R, Dalby M, Schoonderwalt HC, Hantson L, Eriksson B, Lang-Jenssen L, Smakman J. Flunarizine in stroke treatment (FIST): a double-blind, placebo-controlled trial in Scandinavia and the Netherlands. Acta Neurologia Scand. 1996; 93: 56–60.
54. Donnan GA, Davis SM, Chambers BR, Gates PC, Hankey GJ, McNeil JJ, Rosen D, Stewart Wynne EG, Tuck RR. Streptokinase for acute ischemic stroke with relationship to time of administration. JAMA. 1996; 276: 961–966.
55. Grotta J, The US and Canadian Lubeluzole Ischemic Stroke Study Group. Lubeluzole treatment of acute ischemic stroke. Stroke. 1997; 28: 2338–2346.
56. De Deyn P, De Reuck J, Debertdt W, Vlietinck R, Orgogozo JM. Treatment of acute ischemic stroke with piracetam. Stroke. 1997; 28: 2347–2352.
57. The Publications Committee for the Trial of ORG 10172 in Acute Stroke Treatment (TOAST) Investigators. Low molecular weight heparinoid, ORG 10172 (danaparoid), and outcome after acute ischemic stroke. JAMA. 1998; 279: 1265–1272.
58. Yamaguchi T, Sano K, Takakura K, Saito I, Shinohara Y, Asano T, Yasuhara H. Ebselen in acute ischemic stroke. Stroke. 1998; 29: 12–17.
59. Aichner FT, Fazekas F, Brainin M, Polz W, Mamoli B, Zeiler K. Hypervolemic hemodilution in acute ischemic stroke: the multicentre Austrian hemodilution trial (MAHST). Stroke. 1998; 29: 743–749.
60. Diener HC. Multinational randomised controlled trial of lubeluzole in acute ischaemic stroke. Cerebrovasc Dis. 1998; 8: 172–181.[CrossRef][Medline] [Order article via Infotrieve]
61. Wahlgren NG, Ranasinha KW, Rosolacci T, Franke CL, Erven van PMM, Ashwood T, Claesson L. Clomethiazole acute stroke study (CLASS): Results of a randomised controlled trial of clomethiazole versus placebo in 1360 acute stroke patients. Stroke. 1999; 30: 21–28.
62. Clark WM, Williams BJ, Selzer KA, Zweifler RM, Sabounjian LA, Gammans RE. A randomised efficacy trial of citicoline in patients with acute ischaemic stroke. Stroke. 1999; 30: 2592–2597.
63. Clark WM, Wissman S, Albers GW, Jhamandas JH, Madden KP, Hamilton S. Recombinant tissue-type plasminogen activator (alteplase) for ischemic stroke 3 to 5 hours after symptom onset. The ATLANTIS study: a randomized controlled trial. JAMA. 1999; 282: 2019–2026.
64. Lees KR, Asplund K, Carolei A, Davis SM, Diener HC, Kaste M, Orgogozo JM, Whitehead J. Glycine antagonist gavestinel in neuroprotection (GAIN International) in patients with acute stroke: a randomised controlled trial. Lancet. 2000; 355: 1949–1954.[CrossRef][Medline] [Order article via Infotrieve]
65. Diener H, Cortens M, Ford G, Grotta J, Hacke W, Kaste M, Koudstaal P. Lubeluzole in acute ischaemic stroke treatment. Stroke. 2000; 31: 2543–2551.
66. Clark WM, Raps EC, Tong DC, Kelly RE. Cervene (nalmefene) in acute ischemic stroke. Final results of a phase III efficacy study. Stroke. 2000; 31: 1234–1239.
67. Berge E, Abdelnoor M, Nakstad PH, Sandset PM. Low molecular-weight heparin versus aspirin in patients with acute ischaemic stroke and atrial fibrillation: a double-blind randomised study. Lancet. 2000; 355: 1205–1210.[CrossRef][Medline] [Order article via Infotrieve]
68. Horn J, deHaan R, Vermeulen M, Limburg M. Very early nimodipine use in stroke (VENUS). A randomised, double-blind, placebo-controlled trial. Stroke. 2001; 32: 461–465.
69. Sacco RL, DeRosa JT, Haley EC, Levin B, Ordronneau P, Phillips SJ, Rundek T, Snipes RG, Thompson JLP, For the GAIN Americas Investigators. Glycine antagonist in neuroprotection for patients with acute stroke. GAIN Americas: a randomized controlled trial. JAMA. 2001; 285: 1719–1728.
70. Clark WM, Wechsler LR, Sabounjian LA, Schwiderski UE. A phase III randomized efficacy trial of 2000mg citicoline in acute ischemic stroke patients. Neurology. 2001; 57: 1595–1602.
71. Albers G, Goldstein LB, Hall D, Lesko LM. Aptiganel hydrochloride in acute ischemic stroke. JAMA. 2001; 286: 2673–2682.
72. Matthews WB, Oxbury JM, Grainger KM, Greenhall RC. A blind controlled trial of dextran 40 in the treatment of ischaemic stroke. Brain. 1976; 99: 193–206.
73. Geismar P, Marquardsen J, Sylvest J. Controlled trial of intravenous aminophylline in acute cerebral infarction. Acta Neurol Scand. 1976; 54: 173–180.[Medline] [Order article via Infotrieve]
74. Mulley G, Wilcox RG, Mitchell JR. Dexamethasone in acute stroke. BMJ. 1978; 2: 994–996.
75. Ohtomo E, Araki G, Itoh E, Tougi H, Matuda T, Atarashi J. Clinical efficacy of urokinase in the treatment of cerebral thrombosis. Multi-center double-blind study in comparison with placebo. Clin Eval. 1985; 13: 711–751.
76. Tazaki Y, Sakai F, Otomo E, Kutsuzawa T, Kameyama M, Omae T, Fujishima M, Sakuma A. Treatment of acute cerebral infarction with a choline precursor in a multicenter double-blind placebo-controlled study. Stroke. 1988; 19: 211–216.
77. Argentino C, Sacchetti ML, Toni D, Savoini G, D’Arcangelo E, Erminio F, Federico F, Milone FF, Gallai V, Gambi D. GM1 ganglioside therapy in acute ischemic stroke. Italian Acute Stroke Study - Hemodilution + Drug. Stroke. 1989; 20: 1143–1149.
78. German Flunarizine Study Group. Flunarizine in ischemic stroke; a randomised, multicentre, placebo-controlled, double-blind study. Neurol Psych Brain Res. 1993; 1: 173–180.
79. Kramer G, Tettenborn B, Schmutzhard E, Aichner F, Schwartz A, Busse O, Hornig CR, Ladurner G. Nimodipine in acute ischemic stroke. Results of the Nimodipine German Austrian S Trial. Cerebrovasc Dis. 1994; 4: 182–188.
80. Wahlgren NG, MacMahon DG, de Keyser J, Indredavik B, Ryman T, INWEST Study Group. Intravenous Nimodipine West European Stroke Trial (INWEST) of nimodipine in the treatment of acute ischaemic stroke. Cerebrovasc Dis. 1994; 4: 204–210.[CrossRef]
81. Kay R, Wong KS, Yu YL, Chan YW, Tsoi TH, Ahuja AT, Chan FL, Fong KY, Law CB, Wong A, Woo J. Low-molecular-weight heparin for the treatment of acute ischemic stroke. N Engl J Med. 1995; 333: 1588–1593.
82. Haley EC. High-dose tirilazad for acute stroke (RANTTAS II). Stroke. 1998; 29: 1256–1257.
83. Lyden PD, Shuaib A, Ng K, Atkinson R, Ashwood T, Nordlund A, Odergren T. The clomethiazole acute stroke study in hemorrhagic stroke (CLASS-H): final results. J Stroke Cerebrovasc Dis. 2000; 9: 268–275.[CrossRef]
84. Ogun SA, Odusote KA. Effectiveness of high-dose dexamethasone in the treatment of stroke. West African J Med. 2001; 20: 1–6.[Medline] [Order article via Infotrieve]
85. Grotta J. Combination therapy stroke trial. Ann N Y Acad Sci. 2001; 939: 309–310.[Medline] [Order article via Infotrieve]
This article has been cited by other articles:
 |
|
 |
|
  P. Charles, B. Giraudeau, A. Dechartres, G. Baron, and P. Ravaud Reporting of sample size calculation in randomised controlled trials: review BMJ, May 12, 2009; 338(may12_1): b1732 - b1732. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. R. Macleod, M. Fisher, V. O'Collins, E. S. Sena, U. Dirnagl, P. M.W. Bath, A. Buchan, H. B. van der Worp, R. Traystman, K. Minematsu, et al. Good Laboratory Practice: Preventing Introduction of Bias at the Bench Stroke, March 1, 2009; 40(3): e50 - e52. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
The Optimising the Analysis of Stroke Trials (OAST Should Stroke Trials Adjust Functional Outcome for Baseline Prognostic Factors? Stroke, March 1, 2009; 40(3): 888 - 894. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A.-W. Chan, A. Hrobjartsson, K. J Jorgensen, P. C Gotzsche, and D. G Altman Discrepancies in sample size calculations and data analyses reported in randomised trials: comparison of publications with protocols BMJ, December 4, 2008; 337(dec04_1): a2299 - a2299. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. M.W. Bath, C. Geeganage, L. J. Gray, T. Collier, and S. Pocock Use of Ordinal Outcomes in Vascular Prevention Trials: Comparison With Binary Outcomes in Published Trials * Supplemental Appendix I: Statistical Tests Compared * Supplemental Appendix II: Supplementary Analyses * Supplemental Appendix III: Results Stroke, October 1, 2008; 39(10): 2817 - 2823. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
The Optimising Analysis of Stroke Trials (OAST) Co Can We Improve the Statistical Analysis of Stroke Trials?: Statistical Reanalysis of Functional Outcomes in Stroke Trials * OAST Supplemental Appendix I: Statistical Tests Compared (see Table I) * OAST Supplemental Appendix II: Supplementary Analyses * OAST Supplemental Appendix III: Trial Data (see Tables II and III) * OAST Supplemental Appendix IV: Results (see Table IV) Stroke, June 1, 2007; 38(6): 1911 - 1915. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. Song, C. Jerosch-Herold, R. Holland, M. d. L. Drachler, K. Mares, and I. Harvey Statistical methods for analysing Barthel scores in trials of poststroke interventions: a review and computer simulations Clinical Rehabilitation, April 1, 2006; 20(4): 347 - 356. [Abstract] [PDF]
|
|
|
|
|
|
M. Fisher and for the Stroke Therapy Academic Industry Roundtabl Enhancing the Development and Approval of Acute Stroke Therapies: Stroke Therapy Academic Industry Roundtable Stroke, August 1, 2005; 36(8): 1808 - 1813. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||