Reproducibility of myocardial contrast echocardiography in human studies

Experimental and clinical studies were performed to assess the ability of myocardial contrast echocardiography for quantitation of regional myocardial blood flow. To evaluate whether myocardial contrast echocardiography is a reproducible technique in humans, 18 nonselected patients undergoing coronary angiography were studied. A total of 107 intracoronary injections into either the left or the right coronary artery were analyzed by computer assisted videodensitometry for peak intensity, contrast decay half-time, and area under the curve. By means of these parameters intraobserver, interobserver, and interinjection variability were determined. Intraobserver measurements showed lowest variability with correlation coefficients of 0.83 for contrast decay half-time, 0.93 for peak intensity, and 0.95 for area under the curve. Mean percent error varied between 6.8% (peak intensity) and 11.2% (area under the curve). The correlation coefficients for interobserver variability ranged from 0.73 for area under the curve to 0.97 for peak intensity. Mean percent error revealed a range between 7.5% for peak intensity and 19% for area under the curve. For interinjection variability, the correlation coefficient for contrast decay half-time was lower (0.56) than for peak intensity (0.73) and area under the curve (0.84). Mean percent error were higher than for intraobserver and interobserver variability (range 24.1% to 34.2%). Thus, intraobserver and interobserver variability for parameters derived from time-intensity curves after intracoronary injection of echo contrast agent in humans are sufficient and comparable to data from animal studies. Interinjection variability, however, showed a higher mean percent error.(ABSTRACT TRUNCATED AT 250 WORDS)