Accuracy of subxiphoid echocardiography for assessing left ventricular size and performance.

SUMMARYNo data are available comparing subxiphoid M-mode echocardiography with left ventricular (LV) cineangiography, although in some patients an LV echogram can only be obtained by this approach. Therefore, we studied 41 patients without coronary artery disease and with symmetric ventricular contraction documented by cineangiography. Twenty-three patients had adequate LV echograms recorded from both the subxiphoid and standard left sternal border (LSB) transducer positions and were analyzed further. Estimations of end-diastolic volume, end-systolic volume and ejection fraction from the subxiphoid echogram compared favorably with the angiographic determinations (r = 0.92, 0.98 and 0.78, respectively). End-diastolic volume by subxiphoid echo averaged 138 ± 94 ml (SD), not statistically different from the angio volume of 135 ± 91 ml. Subxiphoid echo overestimated ejection fraction at 71 ± 15% compared with 58 ± 16% by angio (p < 0.001), as did the standard LSB echo technique at 69 ± 17% (p < 0.001). Mean LV measurements by both echo methods were nearly identical: end-diastolic dimension, subxiphoid 5.0 ± 1.0 cm and LSB 5.0 ± 1.0 cm (r = 0.94); end-systolic dimension, subxiphoid 3.3 ± 1.2 cm and LSB 3.4 1.3 cm (r = 0.99); percent dimensional shortening, subxiphoid 35 ± 10% and LSB 34 ± 11% (r = 0.95); mean normalized rate of dimensional shortening, subxiphoid 1.4 ± 0.4 sec−1 and LSB 1.3 ± 0.4 sec−1 (r = 0.90). Therefore, subxiphoid echocardiography, like the standard LSB method, is a valid technique for assessing LV size and performance in patients without LV dyssynergy.