Exercise-Induced ST-Segment Elevation in ECG Lead aVR Is a Useful Indicator of Significant Left Main or Ostial LAD Coronary Artery Stenosis

Abstract

The authors tested the hypothesis that exercise treadmill testing (ETT)-induced ST-segment elevation (STE) in electrocardiographic lead aVR is an important indicator of significant left main coronary artery (LMCA) or ostial left anterior descending coronary artery (LAD) stenosis. Although STE in lead aVR is an indicator of LMCA or very proximal LAD occlusion in acute coronary syndromes, its predictive power in the setting of ETT is uncertain. Rest and stress electrocardiograms, clinical and stress test parameters, and single photon-emission computed tomographic myocardial perfusion imaging (MPI) data, when available, were obtained in 454 subjects (378 with MPI) who underwent cardiac catheterization and standard Bruce ETT ≤ 6 months before catheterization. Patients were selected for LMCA or ostial LAD disease (≥ 50% stenosis) with or without other coronary artery disease (CAD), CAD (≥ 70% stenosis) without significant LMCA or ostial LAD, or no significant CAD. Univariate followed by multivariate logistic regression analyses of clinical, electrocardiographic, stress test, and single photon-emission computed tomographic MPI variables were used to identify significant correlates of LMCA or ostial LAD stenosis. Bayesian analysis of the data also was performed. LMCA (n = 38) or ostial LAD (n = 42) stenosis occurred in 75 patients (5 patients had both). The remainder had CAD without LMCA or ostial LAD stenosis (n = 276) or no CAD (n = 103). In multivariate analysis, the strongest predictor was stress-induced STE in lead aVR (p < 0.0001, area under the curve 0.82). Both left ventricular ejection fraction (after stress) and percent reversible LAD ischemia on single photon-emission computed tomographic MPI also contributed significantly in multivariate analysis (p < 0.005 and p < 0.05, respectively, areas under the curve 0.60 and 0.64, respectively). Although additional electrocardiographic, stress test, and MPI variables were significant univariate predictors, none was statistically significant in multivariate analysis. At 1-mm STE in lead aVR, sensitivity for LMCA or ostial LAD stenosis was 75%, specificity was 81%, overall predictive accuracy was 80%, and post-test probability increased nearly 3 times from 17% to 45%. Stress (ETT)-induced STE in lead aVR is an important indicator of significant LMCA or ostial LAD stenosis and should not be ignored.