Author(s)

Myzoon Ali, Sari Atula, Philip M. W. Bath, James Grotta, Werner Hacke, Patrick Lyden, John R. Marler, Ralph L. Sacco, Kennedy R. Lees

ISBN

0039-2499

Publication year

2009

Periodical

Stroke

Periodical Number

1

Volume

40

Pages

35-40

Author Address

Full version

Background and Purpose-Stroke incidence and outcome vary widely within and across geographical locations. We examined whether differences in index stroke severity, stroke risk factors, mortality, and stroke outcome across geographical locations remain after adjusting for case mix.Methods-We analyzed 3284 patients from the Virtual International Stroke Trials Archive (VISTA). We used logistic regression to examine the incidence of mild index stroke, functional, and neurological outcomes after accounting for age, medical history, year of trial recruitment, and initial stroke severity in the functional and neurological outcome analyses. We examined mortality between geographical regions using a Cox proportional hazards model, accounting for age, initial stroke severity, medical history, and year of trial recruitment.Results-Patients enrolled in the USA and Canada had the most severe index strokes. Those recruited in Austria and Switzerland had the best functional and neurological outcomes at 90 days (P<0.05), whereas those enrolled in Germany had the worst functional outcome at 90 days (P=0.013). Patients enrolled in Austria, Switzerland, Belgium, Netherlands, Finland, Germany, Greece, Israel, Spain, and Portugal had a significantly better survival rate when compared with those enrolled in USA and Canada. Patients enrolled in trials after 1998 had more severe index strokes, with no significant difference in outcome compared with those enrolled before 1998.Conclusion-We identified regional variations in index stroke severity, outcome, and mortality for patients enrolled in ischemic stroke clinical trials over the past 13 years that were not fully explained by case mix. Index stroke severity was greater in patients enrolled after 1998, with no significant improvement in outcomes compared to those enrolled before 1998. (Stroke. 2009;40:35-40.)