Thrombolysis Is Associated With Consistent Functional Improvement Across Baseline Stroke Severity
A Comparison of Outcomes in Patients From the Virtual International Stroke Trials Archive (VISTA)
Background and Purpose—Baseline stroke severity predicts outcomes among thrombolysed patients. The baseline National Institutes of Health Stroke Scale (NIHSS) thresholds are sometimes used to select patients for thrombolysis, clinical trial enrollment, or both. Using data lodged with Virtual International Stroke Trials Archive, we compared adjusted outcomes between thrombolysed and nonthrombolysed patients enrolled in neuroprotection trials (1998–2007) to assess the influence of various levels of baseline NIHSS.
Method—We assessed the association of treatment with outcome, measured across the modified Rankin scale score distribution, in patients categorized by baseline NIHSS in increments of 4. We used an age and baseline NIHSS adjusted Cochran-Mantel-Haenszel test followed by proportional odds logistic regression analysis. We report the Cochran-Mantel-Haenszel P values and estimated odds ratios (OR) for improved modified Rankin scale score distribution with treatment for patients within each baseline NIHSS category.
Results—Data were available for 5817 patients (1585 thrombolysed and 4232 nonthrombolysed). Baseline severity was greater among thrombolysed than nonthrombolysed (median baseline NIHSS, 14 vs 13; P<0.05). An association of treatment with outcome was seen independently and was of similar magnitude within each of the baseline NIHSS categories 5 to 8 (P=0.04; OR, 1.25; 95% confidence interval [CI], 1.0–1.6; N=278/934 thrombolysed/nonthrombolysed), 9 to 12 (P=0.01; OR, 1.3; 95% CI, 1.1–1.6; N=404/942), 13 to 16 (P<0.05; OR, 1.6; 95% CI, 1.3–2.1; N=342/814), 17 to 20 (P<0.05; OR, 1.7; 95% CI, 1.3–2.1; N=311/736), and 21 to 24 (P<0.05; OR, 1.6; 95% CI, 1.1–2.1; N=178/466). No association was observed within baseline NIHSS categories 1 to 4 (P=0.8; OR, 1.1; 95% CI, 0.3–4.4; N=8/161) or ≥25 (P=0.08; OR, 1.1; 95% CI, 0.7–1.9; N=64/179).
Conclusions—In this nonrandomized comparison, outcomes after thrombolysis were significantly better than in untreated comparators across baseline NIHSS 5 to 24. The significant association was lost only at extremes of baseline NIHSS when sample sizes were small and confidence limits were wide.
Intravenous thrombolysis with alteplase is a proven therapy for acute ischemic stroke patients presenting before 4.5 hours of symptom onset.1 However, some patients are denied therapy for fear of poor outcomes.2 European guidelines recommend that patients with baseline stroke severity, National Institutes of Health Stroke Scale (NIHSS) ≥25, and minor/rapidly improving strokes should not be thrombolysed,2,3 because it is believed that many patients who show rapid improvement/have minor strokes would not display residual deficit, and treatment with thrombolytic therapy would expose them to risk of complications, such as cerebral hemorrhage. Similarly, those patients who present with baseline NIHSS ≥25 are also supposed to have poorer outcomes because of excess symptomatic hemorrhages.2–3 Baseline stroke severity (baseline NIHSS) is known to affect outcomes among thrombolysed patients4 and therefore was incorporated for patient selection in the ECASS III trial.1 Although the regulatory authorities have recommended withholding thrombolytic therapy among patients with minor/rapidly improving strokes and for those with severe stroke at baseline, poorer response to therapy in these subgroups has never been demonstrated in randomized, controlled trials.5 Post hoc analyses of the NINDS and ECASS-III trials suggest equal efficacy across severity range, although power to examine subgroups is inevitably lower than chosen for the primary analyses, and patients at extremes of severity were under-represented.6–8 The logistical challenges involved in generating randomized trial evidence for these limited subgroups militate against any prospect for producing a definitive answer in the foreseeable future. Therefore, we must turn to alternative sources of evidence.
The Virtual International Stroke Trials Archive (VISTA) is a repository of data from many rigorously controlled clinical trials.9 Although most of these trials examined putative neuroprotectant agents, use of recombinant tissue plasminogen activator was generally recorded. We planned to use data from VISTA, hypothesizing that clinical practice over the past decade would have been sufficiently diverse to allow analysis of existing rigorously collected clinical data lodged in VISTA to examine the influence of baseline stroke severity on outcomes after thrombolytic therapy.
Patients and Methods
Data Source and Patients
We collated the demographics, clinical data, and measures of functional outcome from neuroprotection trials conducted in the period 1998 to 2007, held within VISTA (www.vista.gla.ac.uk).10 All trials held necessary review board and regulatory approvals, and patients consented to participation; only anonymous data are held by VISTA. We sought VISTA data derived from trials in which the investigational neuroprotection agent was not vasoactive and did not interfered with clotting or from placebo groups. We excluded any patient who had cerebral hemorrhage or stroke of undetermined etiology. To avoid dual publication, we excluded patients who may have been enrolled in SITS-MOST; we determined this from their country and date of enrollment. Finally, we excluded patients lacking our chosen outcome measure, 90-day modified Rankin scale (mRS) score, or secondary outcome, 90-day NIHSS score. Patients who died within 90 days were attributed the mRS score of 6 and categorized separately for NIHSS analysis.
We compared outcome between patients who received thrombolysis and patients who did not receive thrombolysis (controls) among the categories of baseline NIHSS scores (<4, 5–8, 9–12, 13–16, 17–20, 21–24, and ≥25). Note that the reason for withholding thrombolysis in each case was not recorded but will include absence of marketing approval in the region at that time, clinical uncertainty over the use of thrombolysis for stroke generally, absence of treatment facilities for thrombolysis in the hospital at that time, and contraindications to thrombolysis for the individual patient. For each contrast, we compared the overall distribution of all 7 categories of day 90 mRS scores of the 2 groups, ie, from 0 (asymptomatic) through 5 (bed-bound and completely dependent) to 6 (dead). The European Medicines Evaluation Agency Points to Consider for reporting trials allow for use of the full distribution of the mRS but suggest that this may be supported by a secondary analysis of a second outcome measure, such as NIHSS.11 For analysis of our supporting end point, NIHSS, we grouped adjacent scores into categories: 0 (no measurable deficit), 1 to 4, 5 to 8, 9 to 12, 13 to 16, 17 to 20, 21 to 24, and ≥25 (most severe neurological deficit) or dead. The distribution of patients across these categories was then compared between the groups as it was for mRS.
To test for a significant association of outcome distribution with thrombolysis exposure, we used the Cochran-Mantel-Haenszel statistic, adjusting for both age and baseline NIHSS as continuous variables.12,13 This nonparametric approach avoids invoking an assumption of proportional odds in which there should be a common odds ratio across all cut points on the ordinal outcome scale, and we consider that the Cochran-Mantel-Haenszel test provides the most conservative estimate of statistical significance. However, it does not express the extent of the association. For this, we applied logistic regression analysis, also adjusted for age and baseline NIHSS, to estimate the odds ratio under the assumption of proportional odds and its associated 95% confidence interval.
Our choice of baseline factors for adjustment was based on 2 influences. First, age and baseline severity are the 2 most powerful prognostic factors for stroke and are usually included in outcome distribution analyses.13–17 Second, age and NIHSS data were available for our entire sample, whereas other factors of potential interest were incomplete. However, we also undertook a sensitivity analysis in which we adjusted for ECASS III variables diabetes and previous stroke. In addition, we also undertook an adjusted analysis by combining the variables that differed significantly at baseline; however, if this resulted in excessive diminution of our sample, we reported the limitations.
Our objective was mainly to undertake ordinal distribution or “shift” analysis, which is an efficient end point analytic technique recommended by European Medicines Evaluation Agency.11,18–20 Shift analysis is considered better than dichotomization of end point measures, although there are differences of opinion.21,22 Dichotomization is criticized for the statistical information it discards, ie, loss of power, and shift analysis is especially useful when the treatment effect is mild and/or uniform across all Rankin categories. Odds ratios in our analysis express the common odds of an improved distribution of outcome in association with alteplase treatment.
Cochran-Mantel-Haenszel and logistic regression analyses were undertaken using SAS 9.2 software (SAS Software Limited, United Kingdom) and other analyses by Stats Direct software (StatsDirect Limited, United Kingdom).
Reliable information on symptomatic hemorrhage was not available because post-treatment imaging was not routinely applied in neuroprotection trials to patients who had not been treated with alteplase.
We collated data on 9665 patients, of whom 5342 (59%) were enrolled from non-European sites. To avoid dual publication with SITS-MOST, we excluded 2789 patients (28%) enrolled from European sites between 2002 and 2006, and 177 patients for whom we lacked information on country. Complete data were available for analysis of mRS for 5817 patients and data on NIHSS were available for 5715 (Figure 1).
All stroke patients were treated as per institutional practice and stroke guidelines acceptable at the point of trial conduct. Monitoring for protocol compliance was undertaken on behalf of sponsors for these trials. This implies that when thrombolysis was administered, this was in accordance with marketing authorization for the relevant country, ie, that treatment commenced within 3 hours of stroke onset; however, the onset to treatment delay is not recorded for thrombolysis in these trials. Our data derived mainly from North American (60%), European (16%), and Australasian (13%) centers. Baseline characteristics are shown in the Table. Of the 5817 patients with mRS outcome data, 1585 (27.2%) received thrombolysis.
Does Baseline Stroke Severity Influence Stroke Outcomes?
In an ordinal logistic regression analysis, we found that baseline severity (P<0.0001), use of recombinant tissue plasminogen activator, and age were significant predictors of outcomes. Then, in an age-adjusted ordinal logistic regression analysis, we found that baseline stroke severity (P<0.0001) and the interaction between severity and use of alteplase (P=0.04) were associated with outcome of stroke, but we did not see an independent effect of alteplase (P=0.65).
Supported by this interaction test, we classified the baseline stroke severity into 7 baseline NIHSS score categories: 1 to 4, 5 to 8, 9 to 12, 13 to 16, 17 to 20, 21 to 24, and ≥25, and undertook tests of association for thrombolysis with outcomes in each of these categories.
Are There Improved Outcomes Across All Baseline Stroke Severity Categories?
Findings from age and baseline NIHSS-adjusted analysis of functional outcomes are shown in Figure 2. This essentially shows a significant association of better outcomes with use of alteplase for patients presenting with baseline NIHSS 5 to 24. The patients with NIHSS <4 at baseline had a mixed distribution of outcomes at 90 days, some Rankin categories appeared to have improved, and others worsened. Patients with baseline NIHSS >24 showed generally improved Rankin distribution with alteplase; however, proportionality of the treatment effect was maintained.
Findings were consistent for the neurological outcomes (by NIHSS on day 90) and also for the sensitivity analyses (ie, unadjusted analysis and analysis adjusting for age, baseline NIHSS, diabetes, and previous stroke). We could not adjust for onset to treatment time because time to initiation of thrombolytic therapy was not recorded within our source neuroprotection trials. Fifty-nine percent of records lacked coding for the variable “antithrombotic” (N=3432), 4.8% lacked coding for atrial fibrillation (N=278), and 3.2% lacked coding for patients with previous strokes (N=186). This limitation to our sample precluded a reliable analysis that was adjusted for all variables that differed at baseline (age, baseline NIHSS, previous use of antithrombotic drugs, previous stroke, and atrial fibrillation).
Patients with mild and severe strokes are under-represented in randomized trials and post-marketing analyses. As a result, the European Medicines Evaluation Agency marketing authorization for alteplase in acute ischemic stroke lists minor neurological deficit or symptoms rapidly improving before start of infusion and severe stroke as assessed clinically (eg, NIHSS >25) and/or by appropriate imaging techniques as contraindications.2,23 Such patients do present to hospital services, however, and this places the physician in a dilemma of whether to offer treatment. Some experienced physicians treat such patients. For example, 12% of patients in the SITS-ISTR thrombolysis registry had a baseline NIHSS score in the range of 0 to 4, and 4% had severe stroke with NIHSS ≥25. Many more patients were probably denied treatment. A Canadian series found that 31% of cases were considered too mild or improving too rapidly for treatment,19 and a report from the United States indicates that only 1 in 5 patients with NIHSS scores <8 are treated.6 This cannot be justified on the basis of observed outcomes. In retrospect, 32% of patients with cases considered too mild to be treated had either died or were disabled 90 days later.19 Others7,20–21 report similar findings. Randomized trials to establish the existence or extent of benefit at extremes of baseline severity may be difficult to conduct and delayed in execution. Other sources of evidence must be examined, and high-quality registry data are the obvious choice.
In our present nonrandomized comparison of data held in VISTA, outcomes after thrombolysis were significantly better than in untreated comparators across baseline NIHSS scores 5 to 24. This significant association was lost only at extremes of baseline NIHSS (ie, 1–4 and ≥25). Although the point estimates for both adjusted and unadjusted odds ratios remain favorable in the extreme groups, they are lower than those observed at other levels of stroke severity.
In these extreme groups, the small sample size seriously undermines the power of the statistical tests and, with wide confidence intervals, the true point estimate is not reliably indicated. There is a second statistical issue to consider relating to the outcome measure that we used. By examining the full distribution of the mRS, we have used a test that is less dependent on case-mix than dichotomization. We are able to use the same test for patients with mild stroke as is used severe stroke and may still detect benefit. Even so, at the extremes of baseline severity, outcomes are generally so good or so poor that only a few mRS categories are well-represented in the control groups. Both the Cochran-Mantel-Haenszel test and the proportional odds estimations will be compromised if some categories are not contributing to the analysis. Effectively, the test of treatment effect will be diluted by the noncontributing groups. For Cochran-Mantel-Haenszel, this means that it becomes more difficult to reach statistical significance; however, but for the proportional odds tests, the basic assumption has been breached and the effect is not proportional. There is no easy solution to this problem. If case-mix is altered to deliver a significant result, then patients with mild or severe stroke must be excluded, which is the solution used by the trials. Conversely, if the outcome measure is varied according to the sample case-mix (the sliding dichotomy approach discussed by Murray et al24), then interpretation is rendered difficult. Is an odds ratio for achieving mRS 0 vs 1 to 6 equivalent to an odds ratio for achieving mRS 0 to 5 vs 6, ie, is survival free from symptoms equivalent to survival at any cost?
Here, we have chosen to present 1 analytic approach for all severities of stroke, but we also illustrate the range of outcomes at extremes of severity. From these, although the summary statistics show only a nonsignificant but favorable trend, we can draw further conclusions. Among patients with severe stroke, there are evident trends toward benefit across almost all boundaries of mRS. Among patients with mild stroke, all boundaries except 0 to 1 vs 2 to 6 show benefit, but 4 of the mRS categories are entirely unrepresented. Our data show no reason to withhold treatment from either group of patients but are not in themselves sufficient evidence to justify treatment.
Our findings draw validity from the fact that our source clinical trials rigorously reported concomitant treatments and outcomes and had strict on-site data verification procedures. However, the nonrandom allocation to treatment vs control groups is a significant weakness of our design. We could not determine the degree and cause of exclusion of patients from our database. We can only consider factors known to be associated with prognosis.
We have adjusted statistically for factors that have a large influence on outcome. We can also ‘anchor’ our findings by comparison of treatment associations for patients with moderate stroke severity in our study against known treatment effects in comparable patients from randomized trials. For example, we find an odds ratio for favorable outcome of 1.3 to 1.6 for patients with baseline NIHSS 9 to 12 and 13 to 16; the comparable estimate from treatment within 3 hours of stroke onset in a randomized control trial would be 1.64 and for 3 to 4.5 hours would be 1.34.25 Our estimates are comparable and perhaps conservative.
The decay of benefit across later onset to treatment times raises a second issue. We do not have information on the onset to treatment delay for alteplase in our current analysis. Because the patients were permitted only 1 investigational drug in the participating VISTA trials, with alteplase being used as standard of care, and because these trials were closely monitored by their sponsors, we assume that patients were largely treated within 3 hours of stroke onset. We also assume that the onset-to-treatment time is comparable to those from the CASES and SITS-MOST registries (155 [130–175] minutes and 140 [115–165] minutes, respectively; n=6483). Unfortunately, the latency between stroke onset and recording of initial severity differed between our treatment group (3.7 hours) and controls (5.1 hours). Severity is associated with onset to hospital arrival time: patients with more severe stroke present earlier.26 We adjusted our analyses for stroke severity, but it is conceivable that residual bias persists. Such a bias would cause underestimation of true initial severity among our controls and through the baseline adjustment would lead us to overestimate treatment effect. It will influence all patients across our severity range but may be less evident at extremes of severity: the NIHSS criterion will be responsible for discouraging use of alteplase, and so the proportion of patients who are treated with alteplase will have extremes of NIHSS.
With these caveats, it would be desirable to replicate our findings. Supporting evidence could come from a comparison of SITS-ISTR data against VISTA controls, a collaborative analysis that is underway. We lack data on symptomatic hemorrhages because patients who are not treated with thrombolysis generally do not undergo follow-up cerebral imaging for routine detection of hemorrhagic transformation. However, the outcome measure that we use takes into account effects of hemorrhage or other adverse events on function.
We adjusted for age and baseline severity because these are the established most important variables known to influence outcomes.27 We could not adjust for all age, baseline NIHSS, previous use of antithrombotic drugs, previous stroke, and atrial fibrillation data together because 1 of the contributing trial programs did not record pretreatment medications. However, we were able to undertake an adjusted analysis for the variables that were found significant in ECASS III, namely diabetes and previous stroke, and our estimates remained consistent.
Some of the patients in our study received an investigational medicinal product. Each contributing trial has already tested for, and excluded, a significant interaction of that product with alteplase, both in vitro and in vivo.
In conclusion, our findings imply that patients with extremes of NIHSS scores recorded at baseline may still benefit from treatment but the supporting evidence remains weak.
VISTA steering committee: A. Alexandrov, P.W. Bath, E. Bluhmki, L. Claesson, J. Curram, S.M. Davis, G. Donnan, H.C. Diener, M. Fisher, B. Gregson, J. Grotta, W. Hacke, M.G. Hennerici, M. Hommel, M. Kaste, K.R. Lees, P. Lyden, J. Marler, K. Muir, R. Sacco, A. Shuaib, P. Teal, N.G. Wahlgren, S. Warach, and C. Weimar.
The analyses reported in this article were based on a research proposal approved by the VISTA Steering Committee and were undertaken by N.K.M. at University of Glasgow, UK. N.K.M. is supported by a British ORS Scholarship and the University of Glasgow scholarship. VISTA has received financial support from the European Stroke Organisation in the form of an unrestricted grant and contributions toward data extraction and capacity building from the Universities of Glasgow, California San Diego, Nottingham, Edmonton, Calgary, Texas, and Massachusetts; from commercial groups, including Brainsgate, Novartis, Boehringer Ingelheim, and the Vertical Group; and from grant agencies and charities, including the UK Stroke Association. N.K.M. and K.R.L. designed and interpreted the analyses and drafted the manuscript; both had access to the VISTA data. P.L. and J.G. contributed data, reviewed the outline proposal, commented on the manuscript, and approved the final version. All authors take full responsibility for the content. The manuscript was reviewed and approved by the VISTA steering committee (www.vista.gla.ac.uk). No commercial organization was involved in the origination, execution, or reporting of this work. P.L. has received research support from the American Heart Association, the National Institutes of Health, and the Veteran’s Affairs Medical Research Department; consulting fees from Mitsubishi, Co-Axia, Benechill, and Photothera; and research contracts from Astra-Zeneca, Bayer, and Innercool. J.G. has received consulting compensation from Lundbeck. K.L. has received honoraria (modest) from Boehringer Ingelheim, Lundbeck, Thrombogenics, and Talecris.
- Received May 3, 2010.
- Accepted July 28, 2010.
Hacke W, Kaste M, Bluhmki E, Brozman M, Davalos A, Guidetti D, Larrue V, Lees KR, Medeghri Z, Machnig T, Schneider D, von Kummer R, Wahlgren N, Toni D, ECASS Investigators. Thrombolysis with alteplase 3 to 4.5 hours after acute ischemic stroke. N Engl J Med. 2008; 359: 1317–1329.
European Stroke Organisation. Guidelines for Management of Ischaemic Stroke and Transient Ischaemic Attack 2008. http://www.eso-stroke.org/pdf/ESO08_Guidelines_Original_english.pdf. Accessed March 11, 2010.
Larrue V, von Kummer R, del Zoppo G, Bluhmki E. Hemorrhagic transformation in acute ischemic stroke. Potential contributing factors in the European Cooperative Acute Stroke Study. Stroke. 1997; 28: 957–960.
Wahlgren N, Ahmed N, Eriksson N, Aichner F, Bluhmki E, Davalos A, Erilä T, Ford GA, Grond M, Hacke W, Hennerici MG, Kaste M, Köhrmann M, Larrue V, Lees KR, Machnig T, Roine RO, Toni D, Vanhooren G, Safe Implementation of Thrombolysis in Stroke-MOnitoring STudy Investigators. Multivariable analysis of outcome predictors and adjustment of main outcome results to baseline data profile in randomized controlled trials: Safe Implementation of Thrombolysis in Stroke-MOnitoring STudy (SITS-MOST). Stroke. 2008; 39: 3316–3322.
Baumann CR, Baumgartner RW, Gandjour J, von Budingen HC, Siegel AM, Georgiadis D. Good outcomes in ischemic stroke patients treated with intravenous thrombolysis despite regressing neurological symptoms. Stroke. 2006; 37: 1332–1333.
Gonzales NR, Albright KC, Noser EA, Khaja AM, Shaltoni HM, Romman A, Zimbaldi N, Baraniuk MS, Moyé LA, Grotta JC. “Mild stroke” should not exclude patients from thrombolytic therapy. Abstracts From the International Stroke Conference. Stroke. 2006; 37: 621.
Virtual International Stroke Trials Archive V, University of Glasgow, Scotland. VISTA. http://www.vista.gla.ac.uk/vista_publications.aspx. Accessed January 29, 2010.
Ali M, Bath PM, Curram J, Davis SM, Diener HC, Donnan GA, Fisher M, Gregson BA, Grotta J, Hacke W, Hennerici MG, Hommel M, Kaste M, Marler JR, Sacco RL, Teal P, Wahlgren NG, Warach S, Weir CJ, Lees KR. The Virtual International Stroke Trials Archive. Stroke. 2007; 38: 1905–1910.
The European Agency for the Evaluation of Medicinal Products. Committee for proprietary medicinal products (CPMP) at London, 20 Sep 2001, Reference: CPMP/EWP/560/98 (Points to consider on clinical investigation of medicinal products for the treatment of acute stroke. http://www.tga.gov.au/docs/pdf/euguide/ewp/056098en.pdf) and letter reference Number: CPMP/3346/02-EN, dated 21 November 2002 (available at www.ema.europa.eu/pdfs/human/referral/Actilyse/334602en.pdf) accessed on 10th January, 2010.
Stokes ME, Davis CS, Koch GG. Categorical Data Analysis Using the SAS System (2nd ed.), Cary, NC: SAS Publishing, BBU Press and John Wiley Sons Inc. 2000.
Lees KR, Davalos A, Davis SM, Diener HC, Grotta J, Lyden P, Shuaib A, Ashwood T, Hardemark HG, Wasiewski W, Emeribe U, Zivin JA, SAINT I Investigators. Additional outcomes and subgroup analyses of NXY-059 for acute ischemic stroke in the SAINT I trial. Stroke. 2006; 37: 2970–2978.
Opitmising the Analysis of Stroke Trials (OAST) Collaboration, Gray LJ, Bath PM, Collier T. Should stroke trials adjust functional outcome for baseline prognostic factors? Stroke. 2009; 40: 888–894.
Saver JL, Yafeh B. Confirmation of tPA treatment effect by baseline severity-adjusted end point reanalysis of the NINDS-tPA stroke trials. Stroke. 2007; 38: 414–416.
Mann J. tPA for acute stroke: balancing baseline imbalances. CMAJ. 2002; 166: 1651–1652;author reply 2–3.
Opitmising the Analysis of Stroke Trials (OAST) Collaboration, Bath PM, Gray LJ, Collier T, Pocock S, Carpenter J. Can we improve the statistical analysis of stroke trials? Statistical reanalysis of functional outcomes in stroke trials. Stroke. 2007; 38: 1911–1915.
Brown DL, Coffey CS. Stroke trials: a shift to shift analysis? Neurology. 2009; 72: 1292–1293.
Saver JL, Gornbein J. Treatment effects for which shift or binary analyses are advantageous in acute stroke trials. Neurology. 2009; 72: 1310–1315.
Savitz SI, Lew R, Bluhmki E, Hacke W, Fisher M. Shift analysis versus dichotomization of the modified Rankin scale outcome scores in the NINDS and ECASS-II trials. Stroke. 2007; 38: 3205–3212.
Saver JL. Novel end point analytic techniques and interpreting shifts across the entire range of outcome scales in acute stroke trials. Stroke. 2007; 38: 3055–3062.
Ahmed N, Mishra NK, Kobayashi A, Käll TB, Ringleb P, Lees KR, Wahlgren N. Improved outcomes after thrombolysis across baseline NIHSS scores 5–24: controlled comparison of the SITS international registry (SITS-ISTR) versus Virtual International Stroke Trials Archive (VISTA). Cerebrovasc Dis. 2010;49,Suppl 2. www.eurostroke.org. Accessed 28th May, 2010.
Murray GD, Barer D, Choi S, Fernandes H, Gregson B, Lees KR, Maas AI, Marmarou A, Mendelow AD, Steyerberg EW, Taylor GS, Teasdale GM, Weir CJ. Design and analysis of phase III trials with ordered outcome scales: the concept of the sliding dichotomy. J Neurotrauma. 2005; 22: 511–517.
Lees KR, Bluhmki E, von Kummer R, Brott TG, Toni D, Grotta JC, Albers GW, Kaste M, Marler JR, Hamilton SA, Tilley BC, Davis SM, Donnan GA, Hacke W, Allen K, Mau J, Meier D, del Zoppo G, De Silva DA, Butcher KS, Parsons MW, Barber PA, Levi C, Bladin C, Byrnes G. Time to treatment with intravenous alteplase and outcome in stroke: an updated pooled analysis of ECASS, ATLANTIS, NINDS, and EPITHET trials. Lancet. 2010; 375: 1695–1703.
Azzimondi G, Bassein L, Fiorani L, Nonino F, Montaguti U, Celin D, Re G, D'Alessandro R. Variables associated with hospital arrival time after stroke: effect of delay on the clinical efficiency of early treatment. Stroke. 1997; 28: 537–542.
Weimar C, Konig IR, Kraywinkel K, Ziegler A, Diener HC, German Stroke Study Collaboration. Age and National Institutes of Health Stroke Scale Score within 6 hours after onset are accurate predictors of outcome after cerebral ischemia: development and external validation of prognostic models. Stroke. 2004; 35: 158–162.