Abstract W P284: Frailty Is an Independent Predictor of Functional Status at Hospital Discharge in Acute Ischemic Stroke Patients
Introduction: Frailty, a state of cumulative decline in multiple systems, is commonly evaluated in geriatrics outcome but rarely for stroke. We sought to determine if frailty is an independent predictor of functional status after accounting for age, pre-stroke function, stroke severity and other factors.
Methods: Clinical data from ischemic stroke admissions in the Cincinnati/Northern Kentucky Stroke Study during 2005 were obtained. Functional status was measured using the modified Rankin Scale (mRS) with categories ≤1, 2, 3, ≥4. Using established methods, we developed a frailty score using 35 age-related deficits that included comorbidities, symptoms, clinical and lab values (range 0-35 points). We modeled mRS at discharge using an ordinal logistic regression with proportional odds, adjusting for age, race, sex, stroke severity (NIHSS score > 5 vs ≤ 5), pre-stroke functional status (mRS) and EMS (ambulance) arrival. The independent association between the frailty score (expressed as a 1 deficit or point change) and poorer functional status at discharge (higher mRS) was determined.
Results: A total of 1906 ischemic strokes surviving to discharge were included. The median age was 72 years, 23% black, 55% female, median NIHSS 3 (IQR 2, 6) and median frailty score 6 (IQR 3, 8). The distribution of mRS ≤1, 2, 3, or ≥4 at discharge was 19%, 17%, 27%, and 37%, respectively. The final model (Table) illustrates that frailty score, age, stroke severity, pre-stroke function, and EMS use were all significant predictors. A 1 deficit (point) increase in the frailty score was independently associated with a 10% increase in the odds of a higher discharge mRS.
Conclusions: Frailty was independently associated with poorer functional status at discharge even after accounting for age, stroke severity and pre-stroke function. These data suggest that a measure of frailty could improve the predictive accuracy and clinical utility of stroke outcome models.
Author Disclosures: M.J. Reeves: None. H. Sucharew: None. J. Khoury: None. K. Alwell: None. C. Moomaw: None. B. Kissela: None. D. Woo: None. M. Flaherty: None. O. Adeoye: None. P. Khatri: Research Grant; Modest; Dr. Khatri’s Dept of Neurology receives support for her role DSMB member from Biogen, Inc.. Research Grant; Significant; Dr. Khatri’s Dept of Neurology receives support for her roles as: (1) Lead PI of the PRISMS trial from Genentech, Inc and (2) Neurology PI of the THERAPY trial from Penumbra, Inc.. S. Ferioli: None. D. Kleindorfer: None.
- © 2015 by American Heart Association, Inc.