Submaximal effort tolerance as a predictor of all-cause mortality in patients undergoing cardiac rehabilitation

Background: Submaximal effort tolerance is routinely available during cardiac rehabilitation, but its prognostic value in relation to underlying referral diagnosis is not known. Hypothesis: Treadmill effort capacity during submaximal exercise training predicts all-cause mortality after cardiac rehabilitation. Methods: We followed 600 consecutive patients (450 men and 150 women, mean age 65 years) who were referred to a 12-week outpatient program of cardiac rehabilitation; 37% had a prior myocardial infarction (MI), 44% had a recent percutaneous intervention (PCI), and 39% had history of coronary artery bypass surgery (CABG). Results: There were 48 deaths during a mean follow-up period of 1603 +/- 822 days. By multivariate Cox analysis, exit MET activity was the most significant predictor of all-cause mortality. In this model, each 1 MET increase in exit submaximal effort tolerance was associated with a 34% decrease in mortality (hazard ratio [HR] 0.66, 95% confidence interval 0.56-0.77) alone and 28% decrease after adjustment for age (HR = 0.72, confidence interval 0.60-0.85). Enty MET level also had predictive value. Subgroup analysis revealed that the predictive value of exit METs was limited to patients after recent CABG and with MI. None of the variables predicted death after PCI, in whom mortality was significantly lower than in the other groups. Conclusions: Submaximal effort tolerance at completion of cardiac rehabilitation, and also at entry, is a strong and age-independent predictor of mortality in patients who have had either recent CABG or MI without intervention, but not in patients after recent PCI.