Abstract 5638: Distinct Symptom Clusters Predict Event-free Survival In Depressed And Non-depressed Patients With Heart Failure
Heart failure (HF) symptoms likely occur in clusters that may be associated with event-free survival. However, whether symptom clusters affect event-free survival is unknown. The first aim was to exam the impact of symptom cluster on event-free survival. The second aim was to test whether depressive symptoms influence the link between symptom cluster and event-free survival. A total of 309 patients (57% male, 62 ± 14 yrs, 45% NYHA class III/IV) were recruited. HF symptoms and depressive symptoms were measured by the Symptom Status Questionnaire and Beck Depression Inventory. Event-free survival (emergency department visits, hospitalization, and death) was collected for a median follow-up of 185 days and confirmed by review of hospital records. An agglomerative hierarchical clustering approach was used to identify symptom clusters. Hierarchical Cox hazard regression and the Kaplan-Meier technique was used to achive the first and second aim, respectively. Two distinct symptom clusters emerged;
dyspnea on exertion, paroxysmal nocturnal dyspnea, orthopnea, and dyspnea at rest,
ankle swelling, fatigue, anorexia, chest discomfort, and sleep difficulty.
After controlling for age, gender, NYHA functional class, EF, and etiology of HF, the first symptom cluster predicted event-free survival (HR = 1.07, 95% CI = 1.05–1.14). When stratified by depressive symptoms, depressed patients with high distress of the first symptom cluster had the shortest event-free survival (p< .001). Distinct symptom cluster predicted event-free survival. Depressive symptoms influenced the impact of symptom cluster on event-free survival.