Peak oxygen consumption (VO2) is traditionally used for risk stratification in chronic heart failure (CHF); however, its predictive value is unknown with carvedilol treatment. Therefore, we sought to investigate the prognostic role of gas-exchange parameters obtained from symptom-limited cardiopulmonary exercise testing (CPX) in patients with CHF that is treated with carvedilol.

A total of 508 consecutive patients (443 men, mean age [± SD] 59 ± 9 years) with a mean left ventricular ejection fraction (LVEF) of 25% ± 7% underwent CPX. The peak VO2 was 13.9 ± 3 mL/kg/min; the rate of increase of minute ventilation per unit of increase of carbon dioxide production (VE/VCO2 slope) was 32 ± 2. Outcomes (cardiovascular death or urgent heart transplantation) were determined when all patients who survived had been observed for a minimum of 6 months.

Patients were divided into groups according to treatment (carvedilol and non-carvedilol); 236 patients were treated with carvedilol (46%), at a mean dose of 25 ±13 mg. The VE/CO2 slope, LVEF, peak VO2, and carvedilol treatment were revealed by means of multivariate analysis to be independent and additional predictors in the total population; VE/VCO2 slope, LVEF, and peak VO2 were revealed to be independent and additional predictors in the patients in the noncarvedilol group (all P <.001); and only peak VO2 was revealed to be an independent and additional predictor in the patients in the carvedilol group (P <.01). In the carvedilol group, mortality rates were 26%, 11%, 10%, and 4% (P <.05) in patients with peak VO2 ≤10 mL/kg/min, >10 to ≤14 mL/kg/min, >14 to18 mL/kg/min, and ≥18 mL/kg/min, respectively. No difference in mortality rates according to peak VO2 or additional outcome indices were identified in the 212 patients with peak VO2 >10 mL/kg/min.

Peak VO2 provides limited predictive information in patients with CHF that is treated with carvedilol, and no additional gas exchange parameter yields supplementary advice.