Abstract P305: Post-Stroke Aphasia as a Predictor of 30-Day Readmission Associated With Infection: A Retrospective Chart Review

Abstract

Introduction: Studies have demonstrated that aphasia may negatively impact morbidity and mortality among ischemic stroke (IS) patients. However, the association between post-stroke aphasia and readmission with infection (RI) is poorly understood. We sought to assess the impact of aphasia on post-stroke RI. We hypothesized that aphasic patients are at increased risk of infection in the 30-day post-stroke period. Methods: We performed retrospective chart review of the Mount Sinai Hospital IS patients with 30-day all cause readmission from January 2016 - December 2019. All variables were abstracted from the index admission (IA) electronic medical records except for aspects related to the readmission (RA). Aphasia was present if a neurologist diagnosed the patient with acquired language dysfunction during IA. We performed chi square and logistic regression analyses to compare readmitted patients with and without aphasia at IA. Our fully adjusted model controlled for age, sex, medical comorbidities, NIHSS ≥ 8, IA LOS > 7, IA infection, discharge to facility. We completed all analyses with SPSS. Results: During IA, 36% (n=42) were diagnosed with aphasia. At IA, there were no significant differences in age (dichotomized at 65), sex, or medical comorbidities between aphasic and non-aphasic cohorts. However, more aphasic patients had admission NIHSS ≥ 8 (89% vs 35%, p<0.0001), LOS > 7 (76% vs 42%, p=0.0004), discharge to facility (79% vs 49%, p=0.0016), and RI (52% vs 19%, p=0.002). The presence of aphasia predicted RI in both unadjusted (OR=4.6, p<0.001) and adjusted (OR= 3.3, p=0.014) multivariate analyses. The Kappa inter-reliability ranged from 0.7-1.0 for the key variables included in our adjusted model. Conclusions: The adjusted odds of 30-day readmission with infection were significantly greater in those with diagnosis of aphasia at the time of index admission compared to those without. Our study provides preliminary evidence that the presence of aphasia may have negative consequences on a patient’s health beyond the language disturbance. Further study is needed to better understand the reasons and risk reduction strategies in this vulnerable population.