Evaluation of left ventricular contractile pattern by intravenous digital subtraction ventriculography: comparison with cineangiography and assessment of interobserver variability.

The accuracy and reproducibility of measurements of left ventricular (LV) end-diastolic volume (EDV), end-systolic volume (ESV), ejection fraction (EF) and regional wall motion obtained by digital subtraction ventriculography (DSV) were compared with values of direct cineangiography in 40 patients, 21 of whom were ambulatory. DSV was performed with a 1-second, 30-ml contrast injection, which yielded real-time fluoroimages composed of 512 X 512 pixels at 30 frames/s. Single-plane right anterior oblique LV volumes were calculated by area-length methods for both DSV and cineangiography. Wall motion was assessed as percent area shortening for 12 equal myocardial segments, with results classified as abnormal if greater than 2 standard deviations below the mean of 20 normal values. DSV exhibited close correlation with angiography for EDV (r = 0.88), ESV (r = 0.92) and EF (r = 0.93). Intravenous DSV and direct cineangiography were concordant in classification of LV contractile pattern in 436 of 480 (91%) myocardial segments. Measurements of DSV obtained by 2 observers showed close correlations for EDV (r = 0.88), ESV (r = 0.95) and EF (r = 0.94), and wall motion classification was in agreement in 434 of 480 (90%) LV segments. Artifacts induced by respiratory motion, persistence of contrast in the right ventricle or left atrium, or low cardiac output may have contributed to the discrepancies observed. These data indicate that DSV is accurate in assessing LV volume and EF, correlates well with cineangiography and exhibits good interobserver reproducibility.