Numerous studies have established that plasma levels of brain natriuretic peptide (BNP) can predict morbidity and mortality in patients with chronic heart failure (HF), hence the growing use of BNP level testing in clinical diagnosis; however, studies have also found heterogeneity in BNP levels. It is, therefore, necessary to establish what stimulates BNP release in patients with HF in the clinical setting.OBJECTIVE
To determine what triggers BNP secretion in HF patients with a range of etiologies, including systolic and diastolic HF.DESIGN AND INTERVENTION
From October 2003 to December 2004, this prospective study enrolled consecutive patients with congestive HF who had been referred for treatment at the study center. Before discharge from hospital, patients underwent cardiac catheterization and echocardiographic assessment; on the day before discharge, BNP levels of all patients were recorded. Patients were not included in the analysis if they were not assessed by catheterization and echocardiography for any reason, or if they had serum creatinine levels of more than 2 mg/dl (176.8 μmol/l) indicating renal dysfunction. Echocardiographic and hemodynamic data were used to calculate end-diastolic and systolic meridional wall stress.OUTCOME MEASURES
The main outcomes were serum BNP level and end-diastolic and systolic meridional wall stress.RESULTS
In total, 160 patients were included in this study (mean age 66.8±1.0 years, 69% men). Of these patients, 98 patients had a left ventricular ejection fraction of 50% or less (the systolic HF group) and 62 had a left ventricular ejection fraction of more than 50% (the diastolic HF group). There were no significant demographic differences between the two groups. Analysis of overall echocardiographic and hemodynamic data found that BNP levels correlated with left ventricular systolic wall stress (correlation coefficient [r2]=0.277); however, there was a much stronger correlation between BNP level and left ventricular end-diastolic wall stress (r2=0.887; P<0.001). This correlation was also observed when only patients with diastolic HF were analyzed (BNP and left ventricular diastolic wall stress, r2=0.704). Notably, patients with diastolic HF had significantly higher BNP levels than patients with systolic HF (267 pg/ml [interquartile range 136-583 pg/ml] vs 105 pg/ml [interquartile range 64-146 pg/ml]; P<0.001). Furthermore, patients with systolic HF had significantly higher end-diastolic wall stress than patients with diastolic HF (P<0.001).CONCLUSION
Iwanaga et al. have established that, compared with other previously reported parameters, plasma BNP levels are strongly correlated with left ventricular end-diastolic wall stress in both patients with systolic HF and patients with diastolic HF. They highlight that the link between plasma BNP and end-diastolic wall stress could help future diagnosis and management of HF.