ACTA ENDOCRINOLOGICA (BUC)

Background. Heart failure (HF) is a disease with a high prevalence in the elderly population. Problems posed by HF diagnosis and management within this special category validate the interest in studying ways to improve the quality and duration of life in these patients. Aim. To determine the prognostic accuracy of certain clinical and laboratory parameters and to predict mortality rates in elderly patients with HF. Study design. It is a cohort, analytical, prospective, observational, crosssectional study conducted between 2006- 2008. Data regarding patient survival over a period of six months were recorded from the moment of enrollment. Subjects and methods. The study group consisted of 177 patients of over 65 years of age admitted to geriatrics, internal medicine and cardiology units, diagnosed with HF. A series of clinical and laboratory data from these patients has been examined, including NT-proBNP and CRP. Results. CRP levels higher than 3.57 mg/dL determined a hazard ratio (HR) of 3.9 (CI95% 1 - 15.2, p=0.04), urea levels higher than 79 mg/dL determined a HR of 8.8 (CI95% 1.1 - 37.4, p=0.003) and NT-proBNP levels higher than 9370 pg/mL determined a HR of 18.6 (CI95% 5.2 - 66.2, p<0.001). After multiple comparison adjustments (Bonferroni correction), only urea and NT-proBNP remained independent variables predicting 6-month mortality in elderly patients with HF. Conclusion. High NT-proBNP and urea levels were independent predictors for mortality in elderly patients with heart failure.