Delays in presentation and mortality among Black patients with mechanical thrombectomy after large-vessel stroke at a US hospital

Abstract

OBJECTIVE The incidence and severity of stroke are disproportionately greater among Black patients. In this study, the authors sought to examine clinical outcomes among Black versus White patients after mechanical thrombectomy for stroke at a single US institution. METHODS All patients who underwent mechanical thrombectomy at a single center from January 1, 2014, through March 31, 2020, were retrospectively analyzed. Patients were grouped based on race, and demographic characteristics, preexisting conditions, clinical presentation, treatment, and stroke outcomes were compared. The association of race with mortality was analyzed in multivariable logistic regression analysis adjusted for potential confounders. RESULTS In total, 401 patients (233 males) with a reported race of Black (n = 28) or White (n = 373) underwent mechanical thrombectomy during the study period. Tobacco use was more prevalent among Black patients (43% vs 24%, p = 0.04), but there were no significant differences between the groups with respect to insurance, coronary artery disease, diabetes, illicit drug use, hypertension, or hyperlipidemia. The mean time from stroke onset to hospital presentation was significantly greater among Black patients (604.6 vs 333.4 minutes) (p = 0.007). There were no differences in fluoroscopy time, procedural success (Thrombolysis in Cerebral Infarction grade 2b or 3), hospital length of stay, or prevalence of hemicraniectomy. In multivariable analysis, Black race was strongly associated with higher mortality (32.1% vs 14.5%, p = 0.01). The disparity in mortality rates resolved after adjusting for the average time from stroke onset to presentation (p = 0.14). CONCLUSIONS Black race was associated with an increased risk of death after mechanical thrombectomy for stroke. The increased risk may be associated with access-related factors, including delayed presentation to stroke centers.