EXPRESS: Direct mechanical thrombectomy without intravenous thrombolysis versus bridging therapy for acute ischaemic stroke: a meta-analysis of randomized controlled trials.

Background Direct mechanical thrombectomy (dMT) may result in similar outcomes compared to a bridging approach with intravenous thrombolysis (IVT+MT) in acute ischaemic stroke. Recent randomised controlled trials (RCTs) have varied in their design and non-inferiority margins (NIM).AimWe sought to meta-analyse accumulated trial data to assess the difference and non-inferiority in clinical and procedural outcomes between dMT and bridging therapy.Summary of reviewWe conducted a systematic review of electronic databases following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. Random effects meta-analyses were conducted for the pooled data. The primary outcome was good functional outcome at 90 days (modified Rankin Scale (mRS)â¤2). Secondary outcomes included excellent functional outcome (mRSâ¤1), mortality, any intracranial haemorrhage (ICH), symptomatic ICH, successful reperfusion (TICIï³2b) and procedure-related complications. Four RCTs comprising 1633 patients (817 dMT, 816 bridging therapy) were included. There were no statistical differences for the 90-day good functional outcome (OR=1.02, 95%CI 0.84-1.25, p=0.54, I2=0%), and the absolute risk difference was 1% (95% CI â4% to 5%). The lower 95% CI falls within the strictest NIM of -10% among included RCTs. dMT reduced the odds of successful reperfusion (OR=0.76, 95%CI 0.60-0.97, p=0.03, I2=0%) and any ICH (OR=0.65, 95%CI 0.49-0.86, p=0.003, I2=38%). There was no difference in the remaining secondary outcomes. The risk of bias for all studies was low.ConclusionThe combined trial data assessing dMT versus bridging therapy showed no difference in improving good functional outcome. The wide non-inferiority thresholds set by individual trials are in contrast with the clinical consensus on minimally important differences. However, our pooled analysis indicates non-inferiority of dMT with a 4% margin of confidence. The application of these findings is limited to patients presenting directly to MT-capable centres and real-world workflow times may differ against those achieved in a trial setting.