Long-Term Survival and Function After Stroke: A Longitudinal Observational Study From the Swedish Stroke Register.

Abstract

Background and Purpose- Longitudinal long-term prognostic data after stroke based on large cohorts are sparse. We report recent survival and functional outcome data on ischemic stroke (IS) and intracerebral hemorrhage (ICH) for up to 5 years poststroke from the Swedish Stroke Register (Riksstroke). Methods- Beyond Riksstroke's regular follow-up surveys at 3 and 12 months, additional surveys were conducted in 2016 on 2 one-year cohorts with stroke 3 and 5 years earlier. Functional dependency was defined as modified Rankin Scale ≥3. Mortality data of the original cohorts were obtained from the Swedish Causes of Death Register. Multiple imputation was used to estimate functional status in nonresponders. Results- The study included 22 929 patients, 87.5% with IS and 12.5% with ICH. Loss to follow-up in the 4 surveys was 12.8% to 21.2%. Thirty-day mortality was higher for ICH than for IS (30.7% versus 11.1%; P<0.01), whereas for 30-day survivors, 5-year mortality did not differ significantly (P=0.858). Functional outcome was less favorable for ICH at all follow-up points. At 5 years, poor outcome (death or dependency) was 79% in ICH and 70.6% in IS (including imputed data; P<0.01). Favorable outcome was less common with increasing age and in patients with prestroke functional dependency. Conclusions- Despite advances in stroke care, long-term prognosis remains a cause for concern. At 5 years after stroke over 2 in 3 patients with IS, and over 3 in 4 patients with ICH, were dead or dependent. We present robust long-term prognostic data to serve as a reference for further development of healthcare and research in stroke.