Stroke in the stroke unit: recognition, treatment and outcomes in a single-centre cohort

Background and purpose: In-hospital strokes (IHS) are associated with longer diagnosis times, treatment delays and poorer outcomes. Strokes occurring in the stroke unit have seldom been studied. Our aim was to assess the management of in-stroke-unit ischaemic stroke (ISUS) by analysing ISUS characteristics, delays in diagnosis, treatments and outcomes. Methods: Consecutive patients from the Acute Stroke Registry and Analysis of Lausanne (ASTRAL), from January 2003 to June 2019, were classified as ISUS, other-IHS or community-onset stroke (COS). Baseline and stroke characteristics, time to imaging and time to treatment, missed treatment opportunities, treatment rates and outcomes were compared using multivariate analysis with adjustment for relevant clinical, imaging and laboratory data available in ASTRAL. Results: Amongst the 3456 patients analysed, 138 (4.0%) were ISUS, 214 (6.2%) other-IHS and 3104 (89.8%) COS. In multivariate analysis, patients with ISUS more frequently had known stroke onset time than other-IHS (adjusted odds ratio [aOR] 2.44; 95% confidence interval [CI] 1.39–4.35) or COS (aOR 2.56; 95% CI 1.59–4.17), had fewer missed treatment opportunities than other-IHS (aOR 0.22; 95% CI 0.06–0.86) and higher endovascular treatment (EVT) rates than COS (aOR 3.03; 95% CI 1.54–5.88). ISUS was associated with a favourable shift in the modified Rankin Scale at 3 months in comparison with other-IHS (aOR 1.73; 95% CI 1.11–2.69) or COS (aOR 1.46; 95% CI 1.00–2.12). Conclusion: In-stroke-unit ischaemic stroke more frequently had known stroke onset time than other-IHS or COS, fewer missed treatment opportunities than other-IHS and a higher EVT rate than COS. This readiness to identify and treat patients in the stroke unit may explain the better long-term outcome of ISUS.