Effectiveness of thrombectomy in stroke according to baseline prognostic factors: inverse probability of treatment weighting analysis of a population-based registry

Abstract

Background and purpose: in real-world practice, the benefit of mechanical thrombectomy (MT) is uncertain in stroke patients with very favorable or poor prognostic profiles at baseline. We studied the effectiveness of MT versus medical treatment stratifying by different baseline prognostic factors. Methods: retrospective analysis of 2,588 patients with an ischemic stroke due to large vessel occlusion nested in the population-based registry of stroke code activations in Catalonia from January 2017 to June 2019. The effect of MT on good functional outcome (modified Rankin Score ≤2) and survival at 3 months was studied using inverse probability of treatment weighting (IPTW) analysis in three pre-defined baseline prognostic groups: poor (if pre-stroke disability, age >85 years, National Institutes of Health Stroke Scale [NIHSS] >25, time from onset >6 hours, Alberta Stroke Program Early CT Score <6, proximal vertebrobasilar occlusion, supratherapeutic international normalized ratio >3), good (if NIHSS <6 or distal occlusion, in the absence of poor prognostic factors), or reference (not meeting other groups' criteria). Results: patients receiving MT (n=1,996, 77%) were younger, had less pre-stroke disability, and received systemic thrombolysis less frequently. These differences were balanced after the IPTW stratified by prognosis. MT was associated with good functional outcome in the reference (odds ratio [OR], 2.9; 95% confidence interval [CI], 2.0 to 4.4), and especially in the poor baseline prognostic stratum (OR, 3.9; 95% CI, 2.6 to 5.9), but not in the good prognostic stratum. MT was associated with survival only in the poor prognostic stratum (OR, 2.6; 95% CI, 2.0 to 3.3). Conclusions: despite their worse overall outcomes, the impact of thrombectomy over medical management was more substantial in patients with poorer baseline prognostic factors than patients with good prognostic factors.