[Clinical significance of the acoustic detection of coronary artery stenosis].

OBJECTIVES

The clinical usefulness was investigated of phonocardiography of coronary artery disease based on stenosis sounds.

METHODS

Heart sounds were recorded in 128 patients undergoing coronary angiography and 40 normal young men. Acoustic recordings were made with the patient supine using five sensors affixed to the skin on the left sternal border at the second (L2), third (L3) and fourth (L4) intercostal spaces, and the right sternal border at the second (R2) and third (R3) intercostal spaces. Approximately 400 msec of waveforms immediately following the second heart sound were passed through a band pass filter (400-1,500 Hz) and subjected to frequency analysis using the maximum entropy method. The ratio of heart sound amplitudes within the 400-700 Hz band to those within the 400-1,500 Hz band was designated as the power ratio.

RESULTS

The relationship between the degree of stenosis and the stenosis sounds suggested that the power ratio was greater in mild or moderate (50-75%) stenosis. The patients were divided into the stenosis [American Heart Association (AHA) 50-75%], severe stenosis (AHA 90-100%), and normal groups. The power ratio was significantly greater (p < 0.005) in the stenosis group than in the normal and severe stenosis groups for patients with stenosis of the left anterior descending artery or the left main trunk (single-vessel disease). Using a cut-off level of 45%, the sensitivity was 71%, specificity was 65%, and accuracy was 66%. The greatest power ratio difference was recorded at L4, and represented noninvasive detection of mild to moderate stenosis in the left anterior descending artery or left main trunk. No other site showed any statistically significant difference. Detection was impossible in multivessel disease or in cases complicated with valvular disease.

CONCLUSIONS

Phonocardiography may allow reliable detection of coronary artery stenosis with mild (AHA 50-75%) stenosis.