Background: Whether brain natriuretic peptide (BNP), a neurohormone marker of ventricular dysfunction, correlates with an enhanced ventilatory response (EVR) during cardiopulmonary exercise test, a well-known predictor of prognosis, in systolic heart failure (HF) is currently unknown.
Material/methods: Resting BNP was measured in 134 consecutive stable outpatients aged 69 +/- 11 years with mild to moderate HF and LV ejection fraction (LVEF) < 40% who performed a maximal exercise test. EVR was assessed as the slope of the relation between minute ventilation and carbon dioxide production (VE/VCO2 slope) > or = 35.
Results: LVEF averaged 33 +/- 7%, BNP 350 +/- 396 pg/ml, and the VE/VCO2 slope 36 +/- 8. Fifty-six of 123 patients (45%) had EVR. BNP correlated with VE/VCO2 slope (r = 0.453; p < 0.01). By multivariate logistic regression, plasma BNP was the only independent predictor of EVR (RR: 1.004 per unit increment, 95% CI: 1.002-1.006, p < 0.0001). A BNP > or = 160 pg/ml had 86% sensitivity, 67% specificity, and 76% overall accuracy for the prediction of EVR (chi square: 37.4, RR 12.2, 95% CI: 4.96-30.3, p < 0.0001, AUC 0.815 (95%CI. 0.738-0.892)).
Conclusions: In systolic HF, plasma BNP is related to an enhanced ventilatory response to exercise and offers a simple and reliable alternative to the cardiopulmonary exercise test in patients with inability or contraindications to exercise.