Abstract 2700: What Is The Expected Distribution of Ischemic Stroke Severity? Population-based Estimates of NIHSS From The Cincinnati/Northern Kentucky Stroke Study.
Background: There is little data concerning the distribution of stroke severity in contemporary ischemic stroke populations. Changes in the prevalence of stroke risk factors, as well as in the secondary prevention of stroke could be expected to influence the distribution of stroke severity at the population level . The collection of National Institute of Health Stroke Scale (NIHSS) data are frequently missing in clinical registries, and more complete data are often limited to clinical trial settings which typically include a select patient sample. We describe the distribution of NIHSS across major demographic sub-groups in a nationally representative, population-based study of ischemic stroke.
Methods: Within a biracial population of 1.3 million, all strokes among area residents in 2005 were ascertained by screening discharge records at local hospitals and clinics. A sampling scheme was developed to ascertain additional cases in physician offices and nursing homes. All confirmed ischemic stroke cases underwent comprehensive chart abstraction, including generation of a retrospective NIHSS score (range 0 - 42), which described symptom severity on presentation. Descriptive statistics of NIHSS (median, IQR) were generated by demographic subgroup and surveillance location (hospital admission, in-hospital stroke, or out-of-hospital). To account for the sampling design and multiple stroke events within some subjects, statistically significant differences were tested using age-adjusted GEE-based linear models.
Results: There were 2210 ischemic stroke cases identified during the 12-month study period. The overall median NIHSS score was 3.0 (IQR 1.0 - 7.0). The distribution of NIHSS by age, sex, race, and surveillance source are summarized in the Table. NIHSS score was statistically significantly higher in older age groups, and also differed significantly by surveillance location. There were no significant differences in NIHSS by sex or race.
Conclusions: More than half of all ischemic stroke cases have mild symptom severity upon initial presentation (i.e., NIHSS < 5). Ischemic stroke severity increases with age and, as expected, varies greatly depending on the location where cases are first identified. Given the potential for secular changes in the natural history of ischemic stroke severity, the monitoring of NIHSS across population sub-groups and by time represents a legitimate target for population-based surveillance efforts.
- © 2012 by American Heart Association, Inc.