Multivariable Analysis of Outcome Predictors and Adjustment of Main Outcome Results to Baseline Data Profile in Randomized Controlled Trials

• DOI: 10.1161/strokeaha.107.510768
• Publication Date: 2008-10-17T01:37:30Z
• Authors: Nils Wahlgren; Niaz Ahmed; Niclas Eriksson; Franz Aichner; Erich Bluhmki; Antoni Dávalos; Terttu Erilä; Gary A. Ford; Martin Grond; Werner Hacke; Michael G. Hennerici; Markku Kaste; Martin Köhrmann; Vincent Larrue; Kennedy R. Lees; Thomas Machnig; Risto O. Roine; Danilo Toni; Geert Vanhooren
• Keywords: Stroke; Fibrinolytic agent

Background and Purpose— The Safe Implementation of Thrombolysis in Stroke-MOnitoring STudy (SITS-MOST) unadjusted results demonstrated that intravenous alteplase is well tolerated the effects were comparable with those seen randomized, controlled trials (RCTs) when used routine clinical practice within 3 hours ischemic stroke onset. We aimed to identify outcome predictors adjust outcomes SITS-MOST baseline characteristics RCTs. Methods— study population was (n=6483) pooled RCTs (n=464) patients treated Multivariable, backward stepwise regression analyses (until P ≤0.10) performed for SITS-MOST. Variables appearing either final multivariable model or differing ( <0.10) between included prediction adjustment outcomes. Main measures symptomatic intracerebral hemorrhage, defined as National Institutes Health Stroke Scale deterioration ≥1 7 days any hemorrhage (RCT definition), mortality, independency by modified Rankin Score 0 2 at months. Results— adjusted proportion 8.5% (95% CI, 7.9 9.0) versus 8.6% (6.3 11.6) RCTs; mortality 15.5% (14.7 16.2) 17.3% (14.1 21.1); 50.4% (49.6 51.2) 50.1% (44.5 54.7), respectively. In analysis, older age, high blood glucose, score, current infarction on imaging scans related poor all parameters. Systolic pressure, atrial fibrillation, weight additional hemorrhage. Current smokers had a lower rate Disability before (modified 5), diastolic antiplatelet other than aspirin, congestive heart failure, new centers, male sex Conclusions— from almost identical relevant reinforce conclusion drawn previously analysis. identified several important better suitable thrombolysis.