Effectiveness and safety of reperfusion strategies within 4.5 hours for large-vessel occlusion stroke: a network meta-analysis

Importance Large-vessel occlusion (LVO) strokes cause substantial morbidity and mortality. Reperfusion within 4.5 hours (intravenous thrombolysis (IVT), endovascular thrombectomy (EVT) or bridging strategies (EVT following IVT)) is critical, but the relative effectiveness and safety of different thrombolytic agents/doses and strategies remain debated. Objective To compare the relative effectiveness and safety of different IVT regimens (alteplase (rt-PA) and tenecteplase (TNK) at commonly studied doses), EVT alone and bridging strategies within 4.5 hours after symptom onset for LVO stroke using a network meta-analysis. Design, setting and participants Frequency-based network meta-analysis of randomised clinical trials (RCTs). We included RCTs enrolling adults (≥18 years) with imaging-confirmed LVO treated within 4.5 hours. Trials were identified through systematic searches of PubMed, Embase and Cochrane CENTRAL (through 27 August 2025). Exposures Treatment strategies categorised as (a) rt-PA 0.9 mg/kg, (b) TNK 0.4 mg/kg, (c) rt-PA 0.6 mg/kg+EVT, (d) rt-PA 0.9 mg/kg+EVT, (e) TNK 0.25 mg/kg+EVT, (f) TNK 0.4 mg/kg+EVT and (g) EVT. Main outcomes and measures Primary outcome: 90-day excellent functional outcome (modified Rankin scale (mRS) 0–1). Secondary outcomes: 90-day independence (mRS 0–2). Safe outcomes: 90-day all-cause mortality, symptomatic intracerebral haemorrhage (sICH) and any intracranial haemorrhage (ICH). Relative effects were estimated as RRs with 95% CIs and treatments were ranked by the surface under the cumulative ranking curve (SUCRA). Results 20 RCTs (5602 patients) were included. By SUCRA ranking for the primary outcome, TNK 0.25 mg/kg+EVT had the highest probability of being best (SUCRA 83.8%), followed by rt-PA 0.9 mg/kg+EVT (72.2%). Single-agent rt-PA 0.9 mg/kg and TNK 0.4 mg/kg ranked lowest for functional benefit (SUCRA 14.4% and 8.5%, respectively). For safety endpoints, rt-PA 0.6 mg/kg+EVT and EVT alone tended to rank better in reducing sICH, any ICH and mortality; TNK 0.4 mg/kg (alone or with EVT) frequently occupied lower safety rankings. Most pairwise comparisons did not achieve statistical significance, and many CIs included the null, indicating uncertainty. Conclusions and relevance In the available RCT network, TNK 0.25 mg/kg+EVT had the highest probability of best functional outcome in the 4.5-hour window, while lower-dose rt-PA strategies or EVT alone may be preferable from a safety perspective. Interpretation is limited by imprecision, heterogeneity between trials and the small number of events for some safety outcomes. Additional large, direct head-to-head RCTs and individual-patient data meta-analyses are warranted.