Brain natriuretic peptide for prediction of Cheyne-Stokes respiration in heart failure patients.

ABSTRACT

Cheyne-Stokes respiration (CSR) is indicative of adverse outcome in patients with chronic heart failure (CHF). We evaluated the use of brain natriuretic peptide (BNP) plasma levels to predict CSR in CHF patients. In this cross-sectional study, overnight polygraphy and cardiac work-up were performed and neurohumoral activation was determined in 102 consecutive CHF patients (25-82 years). Demographic characteristics did not significantly differ among patients with (n=38) or without CSR (n=64); BNP (median 377 vs. 142 pg/ml, p<0.001) and norepinephrine levels (459+/-283 vs. 346+/-204 pg/ml, p=0.02) were significantly increased in patients with CSR. BNP concentrations were significantly associated with the central apnoea/hypopnoea index (y=253+/-5.3x; r=0.26, p=0.01). The area under the ROC curve that used BNP to predict CSR was 0.780 (95% CI 0.688 to 0.873). Using established cut-off limits of BNP plasma levels, heart failure patients with BNP levels >500 pg/ml displayed a 13 fold increased risk of CSR (95% CI 2.34-73.50; p=0.03) compared to patients with BNP levels <100 pg/ml. In multiple logistic regression analysis p(a)CO2 (point estimate 0.84, 95% CI 0.72 to 0.98; p=0.02) and higher BNP class (point estimate 3.14, 95% CI 1.38-7.144; p=0.006) emerged as parameters independently predicting the presence of CSR in our cohort of CHF patients. In conclusion, CSR is associated with neurohumoral activation in CHF patients. Specifically, BNP levels are associated with the severity of cardiac and sleep-related disease, and may be helpful in the diagnosis of CSR and more appropriate use of polysomnography in CHF patients.