Left Ventricular Ejection Fraction Change with Exercise Versus Ejection Fraction at Rest in Coronary Artery Disease: Implications for Using Ejection Fraction Variations in Making Therapeutic Decisions

Previous studies have differed regarding the prognostic importance of the change (Delta) in left ventricular ejection fraction (LVEF) with exercise among patients with known or suspected coronary artery disease (CAD). Data suggest that these discrepancies may be owing to patient selection, including wide interstudy variations in the range of LVEFrest at study entry; however, the impact of LVEFrest on LVEF exercise response has not been adequately addressed. To test the hypothesis that magnitude and variability in DeltaLVEF are systematically related to LVEFrest, we analyzed data from 2655 patients who underwent rest/exercise radionuclide cineangiography for evaluation of clinically evident CAD, stratified into 5 successive LVEFrest subgroups: <30% (n = 205), 30%-44% (n = 563), 45%-59% (n = 1529), 60%-75% (n = 324), and >75% (n = 34). The standard deviation of DeltaLVEF among patients with LVEFrest <30% was found to be half that among patients in the higher LVEFrest subgroups (P < 0.00001, global). The average magnitude of the rise and fall in LVEF with exercise also varied markedly among LVEFrest subgroups (P < 0.0001, global), being smallest among patients with LVEFrest <30%. These findings may explain differences in predictive accuracy of DeltaLVEF noted among various study populations. Further study is needed to determine whether LVEFrest should be used in selecting exercise-based prognostic descriptors in individual patients.