Abstract TP62: Predictive Factors For Depressive Symptoms In The Stroke Recovery Follow-up Period

Abstract

Introduction: Although SSRIs are no longer widely prescribed for post-stroke motor recovery, fluoxetine demonstrated beneficial effects on post-stroke depression (PSD). Given potential side effects of SSRIs, targeted initiation among individuals at highest risk for PSD warrants consideration. While previous studies have identified stroke severity and psychiatric history as factors associated with PSD, the predictability of PSD remains unknown. The purpose of this study is to investigate inpatient predictive factors to better identify individuals who might derive the most benefit from targeted initiation of SSRIs. Methods: All adult patients presenting to our tertiary academic stroke center with acute ischemic stroke between 2016-2020 were enrolled in a prospectively-collected stroke registry. Patients were seen 4-6 weeks post-discharge and a subset were administered the PHQ-9 (Patient Health Questionnaire-9) to screen for PSD (PHQ-9 ≥5). Demographics, history of depression, stroke severity, and inpatient PHQ-9 scores were abstracted. Logistic regression was used to determine factors associated with PSD and an ROC analysis determined the predictability of PSD in the inpatient setting. Results: 307 individuals were administered the PHQ-9 at follow-up. This group was younger with less severe strokes, but had no difference in history of depression or inpatient PHQ-9 scores than those without follow-up screening. The mean age of the cohort was 65.5 years; 52% were female. Individuals with history of depression (OR=4.22, 95% CI: 2.40-7.42), inpatient depression (OR=2.82, 95% CI: 1.80-4.43), higher stroke severity (OR=1.05, 95% CI: 1.01-1.10), and those living alone (OR=1.86, 95% CI: 1.01-3.40) had increased odds of PSD at follow-up. Marital status, employment, and outpatient therapy were not associated PSD. The ROC curve using a positive inpatient PHQ-9 achieved an area under the curve (AUC) of 0.62 (95% CI:0.57-0.67), while the AUC was 0.69 (0.64-0.74) after adding history of depression. Conclusion: History of depression and a positive inpatient PHQ-9 appear to be most strongly associated with long-term PSD. Initiating SSRIs only in those individuals at higest risk for PSD may help reduce the burden of stroke recovery in this targeted population.