Potential pitfalls of the nuclear stethoscope.

Eleven patients (three with coronary artery disease, one with mitral valve prolapse, one with atrial septal defect, two with rheumatic mitral valve disease, one with rheumatic aortic valve disease, and three normal individuals) underwent determination of global left ventricular ejection fraction (LVEF) by 1) nuclear stethoscope (cardiac probe), 2) Tc-99m-labeled-erythrocyte gated nuclear angiocardiographic gamma camera-computer techniques, and 3) single plane RAO contrast left ventriculography--all within a 6-hour period without concurrent drug intervention. In addition, global right ventricular ejection fraction (RVEF) was determined by a similar gated technique with gamma camera-computer techniques. LVEF as determined by camera-computer techniques correlated well with that of contrast ventriculography (r = 0.95) in this heterogenous group of subjects. LVEF as determined by cardiac probe correlated somewhat less well with that of contrast ventriculography (r = 0.59). Two cases are described in which, although cardiac probe data were highly reproducible, gross overestimation of the true global LVEF was thought to occur due to the malpositioning of the probe over either the right ventricle or the most contractile portion of the left ventricle in patients with coronary artery disease.