Unspecified Strokes: Time Trends, Determinants, and Long-Term Prognosis in the General Population
Introduction: In the absence of neuroimaging, a stroke is typically labelled as unspecified. While the majority of clinic-based stroke research focuses on hemorrhagic or ischemic stroke, in the general population, a substantial proportion of strokes remains unspecified. Objective: To investigate time trends in the occurrence and determinants of unspecified strokes and differences in patient characteristics and survival compared to ischemic or hemorrhagic stroke. Methods: We included 1,546 participants from the population-based Rotterdam Study who suffered a first-ever stroke during follow-up (1990-2016). We calculated the proportion of unspecified strokes per year and compared their characteristics between 3 time periods (1990-1999, 2000-2009, and 2010-2016) using a chi-square test, and furthermore investigated differences between unspecified, ischemic, and hemorrhagic stroke in patient characteristics and survival using age- and sex-adjusted survival curves. Results: The occurrence of unspecified stroke among all strokes decreased from 75% in 1990 to 16% in 2016. Compared to the first time period (1991-1999), diagnosis of unspecified strokes was more often done by nursing home physicians (13 vs. 40%) and unspecified stroke patients had more often dementia (30 vs. 43%) in the last time period (2010-2016). Compared to patients with ischemic or hemorrhagic stroke, patients with unspecified stroke were on average older (84.3 vs. 78.5 years) and had more often physical impairments and dementia. Furthermore, patients with unspecified stroke had a lower survival probability up to 10 years after stroke than those with ischemic stroke. Conclusions: The proportion of unspecified strokes decreased drastically from 75 to 16% in the last decades. Patients who do not undergo neuroimaging and therefore are classified as unspecified stroke represent an older, more frail patient group that suffers more often from multimorbidities and poor long-term prognosis than those who do undergo neuroimaging and are thus classified as ischemic or hemorrhagic stroke.
|Cerebral hemorrhage, Cerebral infarction, Cohort studies, Diagnosis, Mortality|
|Organisation||Department of Epidemiology|
Heshmatollah, A. (Alis), Mutlu, U, Rojas-Saunero, L.P. (L. Paloma), Portegies, M.L.P, Wieberdink, R.G, Koudstaal, P.J, … Ikram, M.A. (2020). Unspecified Strokes: Time Trends, Determinants, and Long-Term Prognosis in the General Population. Neuroepidemiology. doi:10.1159/000506130