Prediction of mortality in chronic heart failure from peak oxygen consumption adjusted for either body weight or lean tissue.

BACKGROUND Peak oxygen consumption (pVO2) reflects oxygen extraction from the skeletal muscles, but is routinely corrected for body weight. We hypothesized that correcting pVO2 for lean tissue rather than total body weight would improve the prediction of prognosis in patients with chronic heart failure (CHF). METHODS AND RESULTS A total of 272 CHF outpatients (mean age 61 +/- 12 years, New York Heart Association [NYHA] class 2.3 +/- 0.8) underwent a cardiopulmonary exercise testing and body composition assessment by dual-energy X-ray absorptiometry. During a median follow-up of 608 days (range 8-3656), 75 patients died. Univariate survival analysis showed strong survival prediction from pVO2 adjusted for total weight or lean tissue (chi2 17.7, P < .001; chi2 27.5, P < .0001, respectively). Both predicted survival significantly in bivariate analysis, (chi2 4.6, P = .032; chi2 16.6, P < .0001). The predictive effects were independent of exercise protocol (treadmill versus cycle ergometer) (both P < .001). Multivariate analysis showed that pVO2 adjusted for lean tissue had prognostic importance independently of NYHA class, ejection fraction, and ventilation and carbon dioxide production slope (P < .05 for each). In patients with NYHA class I and II (n = 160), pVO2 adjusted for lean tissue predicted outcome (P = .03). CONCLUSION Adjustment for lean tissue instead for body weight increases the prognostic power of pVO2, particularly in patients with mild heart failure.