Abstract 208: Detailed Analysis of the 30-day Outcomes in the Stenting Arm of the SAMMPRIS Trial

Background: Enrollment in first multicenter, randomized trial of Stenting and Aggressive Medical Management for Prevention of Recurrent Ischemic Stroke (SAMMPRIS) was stopped early owing to an excess 30-day event rate in the stenting group relative to the medical arm and a futility analysis. The purpose of this study was to carefully examine the peri-procedural events that occurred in the trial and to correlate them with any baseline or procedural factors that might be associated with an increased risk for an adverse outcome. Methods: 224 of 451 patients were randomized to the stent arm. 33 suffered an independently-adjudicated endpoint event (stroke or death) within 30 days. The present analysis was limited to patients that underwent the complete per-protocol stenting procedure (n=209). 30 total endpoint events occurred in these 209 patients. Two asymptomatic hemorrhages also occurred and were included in the analysis of peri-procedural complications. Study records, including case report forms, procedural reports, and baseline and procedural imaging for these 32 patients were reviewed. 30-day events were further categorized as wire perforation, reperfusion hemorrhage, and thrombo-embolic ischemic stroke. The ischemic strokes were further categorized as perforator or embolic, based on imaging findings. Baseline and procedural factors in these patients were compared to the remaining 177 patients for predictors of the 30-day events. Results: There were seven wire perforations, six resulting in an endpoint. Five were subarachnoid hemorrhages (SAH). One was treated with nBCA and resulted in an ischemic stroke. One additional asymptomatic SAH occurred as a result of a wire perforation. Six intra-parenchymal hemorrhages (IPH) occurred. Five of the six were definitely delayed hours or days after a successful revascularization procedure, likely representing reperfusion hemorrhage. One was asymptomatic. The remaining 19 30-day endpoints were ischemic strokes. 12 were perforator strokes and seven were distal embolic. Two of these were related to delayed stent thrombosis days after the procedure. No correlation between operator experience or volume in the trial and 30-day events was found. Baseline risk factors and procedural factors were also not correlated with the risk for a procedural complication. Conclusions: The 30-day events in the SAMMPRIS trial were multi-factorial, with no single factor accounting for the majority of events. Further research on angioplasty and stenting for intracranial stenosis will need to address several different mechanisms for this procedure to be proven effective.