Comparison of Left Ventricular Ejection Fraction by Cine Computed Tomography and Single Plane Right Anterior Oblique Ventriculography

Abstract

Cine Computed Tomography (CCT) is a minimally invasive technique which offers high temporal (50 msec scan time) and spatial (2 line pairs) resolution. Left ventricular ejection fraction (LVEF) has been determined by this technique in dogs and normal subjects but no comparison has been made with contrast left ventriculography by cardiac catheterization. Ten patients, 9 male and 1 female, mean age 61 (range 46-70) had LVEF determined by both single plane RAO left ventriculography and CCT. Patients were studied in the fasting state, on differ ent days without change in medication. LVEF by CCT was determined in the long axis, a new view which has been developed for CCT to be comparable to the RAO view of contrast left ventriculography by catheterization. This view is obtained by positioning the patient head first into the scanner, supine, with a counterclockwise table slew (20°) without table tilt. Contrast is introduced via a median antecubital vein, and injected in a prolonged bolus of 7-13 seconds de pendent on arm to heart circulation time. Scans are performed in the cine mode (17 frames/sec) timed during maximal opacification of the right and left ventri cles. Four or six contiguous levels are imaged as required to slice the entire left ventricular cavity. End-systolic and end-diastolic frames are identified. Left ventricular cavity areas are determined by computerized planimetry after the Hounsfield level number is set at half the difference between the contrast in the cavity and the myocardium and setting the window width at one giving a black and white image. Left ventricular end-diastolic volume (LVEDV) and end-sys tolic volume (LVESV) per slice are summated to obtain LVEDV and LVESV from which LVEF is desired. LVEF was also determined in the same patients using single plane RAO contrast ventriculography at cardiac catheterization employing the area-length method. The LVEF was the average of two sinus beats not preceded by an extrasystole. Mean LVEF for CCT was 0.61 (range 0.38-0.80) versus catheterization LVEF of 0.56 (range 0.42-0.80). Comparing both methods we found a significant correlation (r = .92). We conclude that CCT determination of LVEF using the long axis view is a reliable method.