Prognostic impact of peak oxygen uptake and percentage of predicted value of peak oxygen uptake in patients with left ventricular dysfunction
Timóteo, A.T.; Mendes, D.E.; Ribeiro, M.A.; Queirós, M.d.C.ção.; Mendes, M.; Seabra-Gomes, R.
Revista Portuguesa de Cardiologia Orgao Oficial da Sociedade Portuguesa de Cardiologia 22(2): 203-211
2003
ISSN/ISBN: 0870-2551 PMID: 12769000 Accession: 050036375
Comparing the prognostic value of peak oxygen uptake (VO2max) in absolute values and percentage of predicted VO2max (%PredVO2) in terms of prognostic stratification, contradictory references are found in the medical literature. In theory, the measurement of VO2max relative to a normal subject should improve its predictive capacity, because VO2max is influenced by gender, weight, physical activity and age. To compare the prognostic value of VO2max and %PredVO2 (Jones equation) in an adult population (> or = 20 years) of patients with left ventricular systolic dysfunction and to study the influence of age on these parameters. We review data from the first cardiopulmonary stress test performed in 295 consecutive patients with left ventricular systolic dysfunction (ejection fraction < or = 40%); mean age was 52 +/- 10 years, 81% were male, and 76 aged > or = 60 years. In 52%, the etiology was ischemic. We evaluated demographic and clinical variables, the baseline ECG and the medication used before the test. We compared VO2max (ml/kg/min) and %PredVO2 in terms of prognostic stratification for the combined endpoint (death and/or heart transplant) for 2 years of follow-up. We detected 60 patients with events (42 deaths and 18 transplants). ROC curve analysis was used to determine the best cut-off value for VO2max and %PredVO2 to identify patients with events. The best cut-off for VO2max was < or = 19 ml/kg/min and < or = 49% for %PredVO2 in the total population, and in the event-free survival analysis, by log-rank test, we obtained p < 0.001 for both cut-offs. Considering the two extreme age groups (20-40 years and > or = 60 years), we found different cut-off values. In the youngest group we obtained VO2max of < or = 23 ml/kg/min and %PredVO2 of < or = 59%, and in the elderly < or = 12.5 ml/kg/min and < or = 64% respectively. For the intermediate group, the cut-off values were the same as for the total population. In the youngest patients, the new cut-offs did not offer an improvement in accuracy. In the elderly, for VO2max, we obtained an improvement in specificity from 28 to 86% and a reduction in sensitivity from 95 to 50%, with an improvement in accuracy from 46.8 to 75.9% (p = 0.05). With regard to %PredVO2, specificity fell from 86 to 61% and sensitivity increased from 50 to 82%, with similar accuracy (75.9 to 67%, p = 0.78). In the extreme age groups, we obtained different cut-off values, but only for the elderly (VO2max < or = 12.5 ml/kg/min and %PredVO2 < or = 64%). A benefit in terms of accuracy was only obtained in respect of VO2max, because for %PredVO2, there is already a correction for age through the Jones equation.