Using BNP to develop a risk score for heart failure in primary care.

AIMS Chronic heart failure is a common condition with high mortality. Accurate diagnosis in primary care is difficult. Elevated B-type natriuretic peptide (BNP) is associated with left ventricular systolic dysfunction and increased mortality. Prognostic scoring systems using BNP may help to stratify risk in primary care patients. The aim of this research was to establish the independent variables which predict mortality in a primary care population-prescribed loop diuretics and to generate and validate a scoring system for heart failure in general practice. METHODS AND RESULTS Five hundred and thirty-two patients were followed up for a mean of 6.4 years after attending a research clinic for clinical assessment, electrocardiogram (ECG), echocardiography, and BNP. Multivariate analysis was used to establish independent prognostic variables and to generate a prognostic scoring system. The score generated was [0.50 x BNP+5 x age+50 x (CVA+sex+diabetes+ECG)]. The cut-off scores for risk groups were; 25th percentile, 411; 50th percentile, 475; 75th percentile, 524; Harrell's c=0.75. CONCLUSION Developing prognostic scoring systems provides a means of risk stratifying patients without relying on a single cut-off diagnostic value for BNP. Further validation of such scoring systems may improve future management of community heart failure patients.