Abstract W P302: Impact of Post-Stroke Medical Complications on 30-Day Readmission Rate
Background: While many predictors of 30-day readmission after stroke are non-modifiable, some factors may be modifiable. We assessed the hypothesis that medical complications predict 30-day readmission.
Methods: A single-center prospective cohort study of ischemic stroke and transient ischemic attack (TIA) patients was conducted. We identified patients admitted between August 2012 and June 2013 and who survived to 30-day follow-up or died during a readmission within 30 days. Patients readmitted within 30 days of discharge were identified by telephone assessment. We evaluated the association between 30-day readmission and 15 pre-specified post-stroke medical complications in addition to baseline characteristics, in-hospital course and treatments, and discharge status using univariable and multivariable statistics. A p-value < 0.05 was considered significant.
Results: Among 415 patients (mean age, 67.5 ± 15 years; 47% female; 65% white; 78.1% ischemic stroke), 89 (21.4%) patients had at least 1 medical complication during hospitalization. The most common complications were urinary tract infection (6.7%), venous thromboembolism (4.8%), and pneumonia (4.1%). Initial NIHSS score was strongly associated with medical complications (p<0.001). Sixty-three (15.2%) patients were readmitted within 30 days. Readmitted patients were more likely to have had medical complications (33.3% vs. 19.3%, p=0.013). On multivariable logistic regression analysis, cardioembolic or large-artery atherosclerotic subtype (adj. OR 2.24, 95% CI 1.29-3.89) and initial NIHSS score (adj. OR 1.07, 95% CI 1.03-1.11) predicted readmission. Of potentially modifiable factors (not including initial NIHSS score or stroke subtype), any medical complication was the only predictor of readmission (adj. OR 2.09, 95% CI 1.61-3.76).
Conclusions: While stroke severity and stroke subtype are non-modifiable predictors of 30-day readmission, medical complications after ischemic stroke or TIA may mediate that risk and identify a subset for targeted intervention. Specific approaches may include prevention of infections and venous thromboembolism and improving post-discharge transitions of care.
Author Disclosures: S.V. Shah: None. C. Corado: None. D. Bergman: None. Y. Curran: None. R. Bernstein: Research Grant; Modest; Boehringer Ingelheim, Medtronic, Pfizer-BMS, Athersys. Speakers' Bureau; Modest; Medtronic. Consultant/Advisory Board; Modest; Janssen. Speakers' Bureau; Significant; Boehringer Ingelheim, BMS-Pfizer. A. Naidech: None. S. Prabhakaran: None.
- © 2014 by American Heart Association, Inc.