Reduction of N terminal-pro-brain (B-type) natriuretic peptide levels with exercise-based cardiac rehabilitation in patients with left ventricular dysfunction after myocardial infarction.

Abstract

Introduction: N-terminal-pro-brain (B-type) natriuretic peptide (NT-pro-BNP) is a peptide hormone released from ventricles in response to myocyte stretch. The aim of the study was to investigate the influence of exercise training on plasma NT-pro-BNP to verify if this parameter could be used as a biological marker of left ventricular remodelling in myocardial infarction patients undergoing an exercise training programme.

Methods: Forty-four patients after myocardial infarction were enrolled into a cardiac rehabilitation programme, and were randomized in two groups of 22 patients each. Group A patients followed a 3-month exercise training programme, while group B patients received only routine recommendations. All patients underwent NT-pro-BNP assay, and cardiopulmonary exercise test before hospital discharge and after 3 months.

Results: In Group A, exercise training reduced NT-pro-BNP levels (from 1498+/-438 to 470+/-375 pg/ml, P=0.0026), increased maximal (VO2peak+4.3+/-2.9 ml/kg per min, P<0.001; Powermax+38+/-7, P<0.001) exercise parameters and work efficiency (Powermax/VO2peak+1.3+/-0.4 Power/ml per kg per min, P<0.001); there was also an inverse correlation between changes in NT-pro-BNP levels and in VO2peak (r=-0.72, P<0.001), E-wave (r=-0.51, P<0.001) and E/A ratio (r=0.59, P<0.001). In group B, at 3 months, no changes were observed in NT-pro-BNP levels, exercise and echocardiographic parameters.

Conclusion: Three months exercise training in patients with moderate left ventricular systolic dysfunction after myocardial infarction induced a reduction in NT-pro-BNP levels, an improvement of exercise capacity and early left ventricular diastolic filling, without negative left ventricular remodelling. Whether the reduction of NT-pro-BNP levels could be useful as a surrogate marker of favourable left ventricular remodelling at a later follow-up remains to be further explored.