Intravenous thrombolysis before endovascular treatment versus endovascular treatment alone for patients with large vessel occlusion and carotid tandem lesions: individual participant data meta-analysis of six randomised trials.

Background For patients with large vessel occlusion and carotid tandem lesions, the best treatment approach is not clear. Intravenous thrombolysis in addition to endovascular treatment might increase the risk of intracranial haemorrhage and decrease effectiveness in this cohort, particularly for patients receiving acute carotid stenting. In an individual participant data meta-analysis of six randomised controlled trials (RCTs), no clear benefit was seen of intravenous thrombolysis in patients with large-vessel occlusion stroke who were eligible for direct endovascular treatment. We aimed to assess whether the presence of carotid tandem lesions would modify the safety and efficacy of intravenous thrombolysis in patients who could directly undergo endovascular treatment, in a prespecified secondary subgroup analysis of this individual participant data meta-analysis.Methods We previously did a systematic review and individual participant data meta-analysis of six RCTs comparing intravenous thrombolysis plus endovascular treatment with endovascular treatment alone in patients with anterior circulation stroke presenting directly at centres capable of endovascular treatment. The principal investigators of the six identified trials provided individual participant data for 2313 patients, which we pooled. The primary outcome was functional outcome, as measured by 90-day modified Rankin Scale score. Heterogeneity of treatment effect was assessed in the intention-to-treat population using ordinal regression models, with interaction terms for treatment and carotid tandem lesions, followed by a mixed-effects meta-analysis. A sensitivity analysis included only patients who received acute carotid stenting. The study is registered with PROSPERO, CRD42023411986.Findings Of the 2313 patients who were included in the individual patient data meta-analysis, 2267 (98%) had data for carotid tandem lesions, of whom 1136 were assigned intravenous thrombolysis plus endovascular treatment and 1131 were assigned endovascular treatment alone. 340 patients had carotid tandem lesions (161 intravenous thrombolysis plus endovascular treatment, 179 endovascular treatment alone) and 1927 did not have tandem lesions (975 intravenous thrombolysis plus endovascular treatment, 952 endovascular treatment alone). The median age of patients was 71 years (IQR 62-78); 1003 (44·2%) patients were female and 1264 (55·8%) were male. Compared with endovascular treatment alone, the addition of intravenous thrombolysis did not improve functional outcome in patients with tandem lesions (adjusted common odds ratio [acOR] 1·00, 95% CI 0·62-1·62) or in those without tandem lesions (1·17, 0·99-1·37). No significant heterogeneity of treatment effect was observed between patients with tandem lesions and those without (ratio of odds ratios 0·81, 95% CI 0·48-1·37; pinteraction=0·44). Intracranial haemorrhage rates in patients with tandem lesions were similar for those receiving intravenous thrombolysis plus endovascular treatment (58 [37%] of 155) and for those receiving endovascular treatment alone (65 [38%] of 172; acOR 0·95, 95% CI 0·59-1·54). Rates of symptomatic haemorrhage in patients with tandem lesions were also similar (six [4%] of 159 for those receiving intravenous thrombolysis plus endovascular treatment vs ten [6%] of 179 for those receiving endovascular treatment alone; 0·81, 0·28-2·30). The sensitivity analysis including only patients who received acute carotid stenting showed similar results to the primary analysis.Interpretation The findings of this prespecified secondary analysis of an individual patient data meta-analysis show that, in patients with carotid tandem lesions, the addition of intravenous thrombolysis to endovascular treatment was not associated with an increase in the risk of bleeding or with modification of functional outcome. These data suggest that the presence of tandem lesions should not solely influence the decision to administer intravenous thrombolysis to patients who can directly undergo endovascular treatment.Funding Stryker, Boehringer Ingelheim, and Amsterdam University Medical Centers, University of Amsterdam.