Post-90-day survival after stroke according to 90-day functional outcome in the 2013 and 2020 cohorts: a Swedish Stroke Register study.

Introduction Ninety-day functional outcome is a strong predictor of long-term survival after stroke, but whether survival differs across calendar-year cohorts within a national stroke population at similar levels of post-stroke disability remains unclear. We therefore examined whether long-term survival from 90 days to 3 years after stroke differed between 2013 and 2020, stratified by functional outcome at 90 days. Methods In this nationwide register-based cohort study using the Swedish Stroke Register (Riksstroke), we included patients with ischemic stroke (IS) or intracerebral hemorrhage (ICH) admitted in 2013 or 2020 who were prestroke independent (mRS 0-2) and survived 90 days. The primary outcome was all-cause mortality between day 90 and 3 years after stroke. Multivariable Cox regression models estimated adjusted hazard ratios (aHRs) and assessed whether the association between calendar-year cohort and mortality differed by 90-day mRS. Results Among 27 452 patients (24 574 IS and 2 878 ICH), adjusted 3-year survival improved for 90-day survivors in 2020 compared with 2013 after IS (aHR 0.86, 95% CI 0.80-0.92), whereas no improvement was observed after ICH (aHR 0.98, 95% CI 0.80-1.19). In IS, higher survival in 2020 was observed both among functionally independent patients (mRS 0-2; aHR 0.76, 95% CI 0.68-0.86; 3-year survival 93.4% vs 91.9%) and those with dependency at 90 days (mRS 3-5; aHR 0.84, 95% CI 0.77-0.91; 76.3% vs 71.7%). In contrast, survival was not significantly different after ICH (mRS 0-2: 90.7% vs 91.1%; mRS 3-5: 78.5% vs 77.8%). Across both stroke subtypes, 90-day disability remained strongly associated with long-term mortality. Conclusion Post-90-day survival after IS was higher in the 2020 than the 2013 cohort and was observed across the full spectrum of post-stroke disability. Findings suggest that gains in survival are not limited to patients with favorable recovery but extend across levels of post-stroke disability, potentially reflecting improvements in post-acute stroke care and secondary prevention. In contrast, no corresponding improvement was observed across functional outcome levels after ICH. Further studies are needed to confirm these findings and investigate their underlying mechanisms.