Comparison of Clinical and Electrocardiographic Predictors of Ischemic and Nonischemic Cardiomyopathy During the Initial Evaluation of Patients With Reduced (≤40%) Left Ventricular Ejection Fraction

Abstract

Invasive coronary angiography is routinely performed during the initial evaluation of patients with suspected cardiomyopathy with reduced left ventricular function. Clinical and electrocardiographic (ECG) data may accurately predict ischemic cardiomyopathy (IC). Medical records of adults referred for coronary angiography for evaluation of left ventricular ejection fraction ≤40% from 2010 to 2014 were retrospectively reviewed. Patients with myocardial infarction (MI), previous coronary revascularization, cardiac surgery, or left-sided valvular disease were excluded. IC was defined as ≥70% diameter stenosis of the left main, proximal left anterior descending, or involvement of ≥2 epicardial coronary arteries. A risk model was developed from logistic regression coefficients, with a dichotomous cut-point based on the maximal Youden's index from the receiver-operating characteristic curve. A total of 273 patients met study inclusion criteria. Mean age was 56.8 ± 11.6 and 68.1% were men. IC was identified in 41 patients (15%). Patients with IC were more likely to have ECG evidence of Q-wave MI (34% vs 13%, p <0.001) and less likely to have left bundle branch block (2% vs 15%, p= 0.03) than non-IC. A model including age, hypertension, diabetes mellitus, tobacco use, ECG evidence of ST or T-wave abnormalities concerning for ischemia, and previous Q-wave MI, yielded a 95% negative predictive value for IC. In conclusion, at an urban referral hospital, the prevalence of IC was low. Left bundle branch block on electrocardiography was rarely associated with IC. A risk score incorporating clinical and ECG abnormalities identified patients at a low likelihood for IC.