Diagnosis and characterization of left ventricular hypertrophy by computerized acoustic cardiography, brain natriuretic peptide, and electrocardiography

BACKGROUND: Using echocardiography as the gold standard to diagnose and classify subtypes of left ventricular hypertrophy (LVH), we compared the diagnostic accuracy of computerized acoustic cardiography, brain natriuretic peptide (BNP), and the Cornell voltage criteria. METHODS: Three hundred fifty-two patients with suspected heart failure had contemporaneous BNP sampling, 12-lead electrocardiography, computerized acoustic cardiography, and echocardiography. Left ventricular hypertrophy was classified as eccentric vs concentric based on echocardiographic relative wall thickness. Computerized acoustic cardiography was used to measure acoustic and automated electrical parameters. RESULTS: Of all models, BNP combined with either computerized acoustic cardiography (c-statistic, 0.78; 95% confidence interval [CI], 0.74-0.78) or Cornell voltage (c-statistic, 0.76; 95% CI, 0.71-0.81) had the best diagnostic performance for LVH detection. For LVH characterization, the computerized acoustic cardiography model outperformed other models (c-statistic, 0.73; 95% CI, 0.66-0.80). CONCLUSIONS: Brain natriuretic peptide combined with either computerized acoustic cardiography or Cornell voltage had the highest diagnostic accuracy for the detection of LVH, compared to Cornell voltage, BNP, or computerized acoustic cardiography alone. Computerized acoustic cardiography outperformed other models for the characterization of LVH subtypes.