Objectives: It is unclear whether age-related increases in NT-proBNP represent a normal physiological process - possibly affecting the prognostic power of NT-proBNP - or reflect age-related sub-clinical pathological changes. Consequently, we aimed to determine the effect of age on the short-term prognostic value of NT-proBNP in patients with chronic heart failure (CHF).
Design: Prospective observational study with inclusion and matching of consecutive patients aged >65 years (73.1±6.0 years) to patients <65 years (53.7±8.6 years) with respect to NT-proBNP, NYHA, sex and aetiology of CHF (final n=443).
Setting: University Hospital Outpatient's in the UK and Germany.
Patients: Chronic stable heart failure due to systolic left ventricular dysfunction.
Outcome Measure: all-cause mortality.
Results: In both age-groups, NT-proBNP was a significant univariate predictor of mortality, and independent of age, sex and other established risk markers. The prognostic information given by NT-proBNP was comparable between the two groups, as reflected by the one year mortality of 9% in both groups. The prognostic accuracy of NT-proBNP as judged by the area under the receiver operating characteristics curve for the prediction of 1-year-mortality was comparable for elderly and younger patients (0.67 vs. 0.71; p=0.09).
Conclusion: NT-proBNP reflects disease severity in elderly and younger patients alike. In patients with chronic stable heart failure, the NT-proBNP value carries the same one year prognostic information regardless of the age of the patient.