Variance ECG detection of coronary artery disease--a comparison with exercise stress test and myocardial scintigraphy.

Variance electrocardiogram (ECG) is a newly developed method by which resting ECG is registered with 24 leads during 220 beats. The temporal beat-to-beat QRS microamplitude variability is computed and a nondimensional diagnostic variance ECG coronary artery disease (CAD) index is derived from it. Consecutive outpatients (n = 160) were referred to myocardial scintigraphy (SPECT) investigation for the evaluation of angina pectoris. The variance ECG CAD index was compared with a symptom-limited exercise stress test and SPECT during and after the exercise test and with coronary angiography (n = 67). Discriminant accuracy was tested with receiver-operating characteristics (ROC). Relative to angiographic coronary pathology (prevalence 0.85), diagnostic information for the variance ECG CAD index and for SPECT were both p < 0.001, while the outcome of the exercise stress test was non-contributory. Prevalence of persistent or transient perfusion defects at SPECT was 0.59. The exercise stress test had a diagnostic capacity of p < 0.01 for transient perfusion defects and variance ECG CAD index showed a high diagnostic performance (p < 0.001) for persistent perfusion defects. Overall pathology at SPECT was better (p < 0.05) identified by variance ECG CAD index than by symptom-limited exercise stress test. It was concluded that in this high prevalence population the variance ECG CAD index has a diagnostic capacity at least as good as that of SPECT and better than that of the exercise stress test. The variance ECG CAD index was strongly diagnostic for persistent perfusion defects while exercise stress test was slightly diagnostic for transient perfusion defects. Therefore, the two tests provide complementary diagnostic information.

Electrocardiogram signal variance analysis in the diagnosis of coronary artery disease--a comparison with exercise stress test in an angiographically documented high prevalence population.

Variance electrocardiography (variance ECG) is a new resting procedure for detection of coronary artery disease (CAD). The method measures variability in the electrical expression of the depolarization phase induced by this disease. The time-domain analysis is performed on 220 cardiac cycles using high-fidelity ECG signals from 24 leads, and the phase-locked temporal electrical heterogeneity is expressed as a nondimensional CAD index (CAD-I) with the values of 0-150. This study compares the diagnostic efficiency of variance ECG and exercise stress test in a high prevalence population. A total of 199 symptomatic patients evaluated with coronary angiography was subjected to variance ECG and exercise test on a bicycle ergometer as a continuous ramp. The discriminant accuracy of the two methods was assessed employing the receiver operating characteristic curves constructed by successive consideration of several CAD-I cutpoint values and various threshold criteria based on ST-segment depression exclusively or in combination with exertional chest pain. Of these patients, 175 with CAD (> or = 50% luminal stenosis in 1 + major epicardial arteries) presented a mean CAD-I of 88 +/- 22, compared with 70 +/- 21 in 24 nonaffected patients (p < 0.01). Variance ECG provided a stochastically significant discrimination (p < 0.01) which was matched by exercise test only when chest pain variable was added to ST-segment depression as a discriminating criterion. Even then, the exercise test diagnosed single-vessel disease with a significantly lower sensitivity. At a cutpoint of CAD-I > or = 70, compared with ST-segment depression > or = 1 mm combined with exertional chest pain, the overall sensitivity of variance ECG was significantly higher (p < 0.01) than that of exercise test (79 vs. 48%). When combined, the two methods identified 93% of coronary angiography positive cases. Variance ECG is an efficient diagnostic method which compares favorably with exercise test for detection of CAD in high prevalence population.