Recanalization as a Surrogate Outcome Measure in Thrombolytic Clinical Trials: A Meta-Analysis

초록

Background & Purpose: For a biomarker to serve as a surrogate outcome measure, it must be tightly linked to functional clinical outcome. Vessel recanalization is a potential surrogate outcome marker for functional outcome in thrombolytic acute stroke therapy, but the correlation of recanalization and functional outcome has not been previously systematically reviewed. Methods: Through Medline, we reviewed all the published paper with data regarding vessel recanalization (spontaneous or therapeutically-induced) and functional outcome. Recanalization data was also categorized by degree of residual stenosis and the interval from the onset. Results: From 1985 to 2002, a total 62 articles encompassing 2284 cases were identified to assess recanalization in acute ischemic stroke. Most of the studies were case series but 5 randomized controlled trials and 3 case-control studies were also included. Method of recanalization was intra-arterial thrombolysis in 30, intravenous thrombolysis 14, angioplasty 1, embolectomy 1, spontaneous 4, and any combination of these in 12 studies. Recanalization was assessed mostly by catheter angiography in 54, by TCD in 7 and by MR Angiography in 1 study. SPECT was also used in 3 studies combined with above methods. Among these, only 29 studies had a clinical outcome data of 781 patients according to the recanalization. The numbers of the good outcome were 272 from 466 recanalized patients (58.4%) and 65 from 315 non-recanalized patients (20.6%). The odds ratio for the good outcome was 5.4 (95% CI 3.9~7.5) in favor of recanalization. The numbers of the death were 48 and 107 respectively (odds ratio 4.2, 95% CI 2.8~6.2 for recanalization better). Conclusions: These results clearly support the recanalization hypothesis of thrombolytic therapy. High odds ratio for the both clinical outcome and mortality suggest the possible usefulness of recanalization as the surrogate outcome of thrombolytic therapy.