Impact of electrocardiographic interpretability on outcome in patients referred for stress testing

Abstract

Background: There is limited insight into the association of electrocardiographic interpretability with outcome in patients referred for stress testing.

Methods: Exercise echocardiography was performed in 8226 patients with known or suspected coronary artery disease. Electrocardiograms were considered uninterpretable in the presence of left bundle-branch block (LBBB), left ventricular hypertrophy (LVH) with strain, repolarization abnormalities because of digitalis therapy, ventricular paced rhythm, preexcitation or ST depression ≥ 0.1 mV because of other causes. End points were all-cause mortality, cardiac death and hard cardiac events (i.e. cardiac death or nonfatal myocardial infarction).

Results: A total of 2450 patients had uninterpretable electrocardiograms. During a follow-up period of 4.1 ± 3.5 years, there were 1011 deaths (of which 478 were cardiac deaths) and 1069 patients experienced a hard cardiac event. The 5-year rates of death, cardiac death and hard cardiac events were, respectively, 18.7%, 10.9% and 18.8% in patients with uninterpretable ECGs, compared with 9.5%, 4.1% and 10.9% in those with interpretable ECGs (P < 0.001). After covariate adjustment, lack of ECG interpretability remained an independent predictor of all-cause mortality (hazard ratio [HR] 1.25, 95% confidence interval [CI] 1.08-1.44, P = 0.002), cardiac death (HR 1.63, 95% CI 1.32-2.01, P < 0.001) and hard cardiac events (HR 1.28, 95% CI 1.11-1.47, P < 0.001). When the specific ECG abnormalities were included as covariates, LBBB, LVH and digitalis therapy remained predictors of cardiac death; LBBB and LVH were predictors of hard cardiac events, and LVH remained predictive of all-cause mortality.

Conclusion: Uninterpretable ECGs portend a worse prognosis in patients referred for stress testing.