Effects of Acute and Chronic Stress on Heart Failure Functional Status, Hospitalization and Mortality

Background: Mental stress is implicated as a trigger of acute cardiovascular (CV) events, but its role in heart failure (HF) outcomes is unclear. We investigated associations between perceived stress and CV hospitalizations or death, functional status, and symptoms in BETRHEART, a prospective study of psychosocial precipitating factors for HF exacerbations. Methods: In 144 HF patients (77% male; 58±12 yrs; LVEF ≤40%), we measured perceived stress (Perceived Stress Scale; PSS), CV hospitalizations or death, HF symptoms (Kansas City Cardiomyopathy Questionnaire; KCCQ), functional status (6min walk test: 6MWT), and BNP at study intake and repeatedly every 2 weeks over a 3 month period. Hospitalizations and death were monitored for an additional 6 months. Linear mixed models were used to examine prospective between- and within-subject associations between perceived stress and outcomes, adjusting for HF risk factors. Results: At follow-up, 42 patients (30.0%) had been hospitalized at least once and 9 (6.4%) had died. Patients with repeated high levels of perceived stress had a greater likelihood of hospitalization or death compared to those with lower stress (OR=1.10, 95% CI=1.04, 1.17). Probability of hospitalization or death over 9 months increased as PSS scores increased from the lowest to the highest third of the PSS distribution (Figure). Within-patient changes in stress over time were not related to hospitalization or death (p=0.89) shortly (≤ 2 weeks) after a stress assessment. However, stress was higher at assessments occurring after a hospitalization (OR=1.09, 95% CI=1.03,-1.16), indicating that hospitalizations increased perceived stress. In the subset of assessments occurring before a first hospitalization, higher stress was associated with CV hospitalization or death (B=2.39, SE=1.22, p=0.05). Within-patient changes in stress were associated with decrements in functional status (6MWT B= -3.32, p≤0.001) and increased HF symptoms (KCCQ B=-0.29, p≤0.001), indicating that stress also affects functional and symptom outcomes. BNP was not related to PSS (B= 0.003, p=n.s). Conclusions: High perceived stress is prospectively associated with CV hospitalizations and death in HF patients and hospitalization also increases stress. Increases in stress are associated with poorer functional status and more symptoms. Thus, stress is both a predictor of, and a consequence of, CV hospitalization. (Figure presented).