The risk of cardiovascular death in elderly patients with possible heart failure. Results from a 6-year follow-up of a Swedish primary care population

Abstract

Little is known about the prognosis and clinical variables influencing the prognosis among elderly patients in primary health care with mild to moderate heart failure. Aim: To evaluate the risk of cardiovascular mortality in elderly patients with symptoms of heart failure with respect to systolic and diastolic function, and functional impairment. To evaluate prognostic determinants and to risk-stratify the patients. Methods: A cardiologist examined 510 patients, out of 548 invited, attending primary care for symptoms of dyspnoea, fatigue and/or peripheral oedema and assessed New York Heart Association (NYHA) functional class. Examination by Doppler echocardiography was done in 454 patients, 56 patients being excluded because of, e.g., atrial fibrillation. Abnormal systolic function was defined as ejection fraction <40%. The diastolic function was evaluated using the mitral inflow and pulmonary venous flow variables. Different clinical and echocardiographic variables were analysed using a Cox regression analysis to identify those most influencing the risk of cardiovascular mortality. Conclusion: Abnormal systolic and/or diastolic function was found in 219 patients (48% of the 454 patients who could be echocardiographically completely investigated). The follow-up period was 6 years. Total mortality was 20%, and cardiovascular mortality was 14% (70% of total mortality). Cardiovascular mortality was high in patients with severely impaired functional capacity and ejection fraction <40% at the start of the study. Risk variables identified were male gender, diabetes mellitus, impaired functional capacity and abnormal cardiac function by echocardiography. A prognostic score model using simple clinical variables (gender, NYHA class, cardiac function) was developed to assess the risk of cardiovascular death in order to identify patients with high, moderate or low risk. In a ROC curve analysis, the AUC for clinical variables was only 0.75, whereas the AUC for clinical variables and echocardiography was 0.78, indicating that the additional prognostic information obtained by Doppler echocardiography was rather small.