Right ventricular ejection fraction validation of first-pass radionuclide studies by contrast angiography.

The accuracy of first-pass radionuclide angiocardiography (FPRNA) in the assessment of right ventricular ejection fraction (RVEF) using a multicrystal scintillation camera in the right anterior oblique view has not been reported. To test the validity of this technique, RVEFs derived from first-pass time-activity curves with and without background correction were compared with those obtained from contrast ventriculography in 36 patients. Background regions-of-interest tested included tricuspid valve, free wall, and horseshoe-shaped approaches. The tricuspid valve approach yielded a mean RVEF of 0.485 +/- 0.100 (SD) which slightly underestimated the contrast mean value (0.553 +/- 0.099, p less than 0.05) but correlated well (r = 0.88). The horseshoe approach yielded a mean value of 0.548 +/- 0.100 which approximated the contrast mean value and also correlated well (r = 0.81). The free wall approach yielded a mean of 0.502 +/- 0.095 which did not differ from contrast data (p greater than 0.05) but correlated less well (r = 0.77). With the method without background correction, a much lower mean value (0.387 +/- 0.081, p less than 0.01) and less correlation (r = 0.77) were obtained. Thus, FPRNA using a multicrystal camera in the right anterior oblique view is a reliable technique for measuring RVEF when a tricuspid valve or horseshoe-shaped background approach is employed, but is less accurate if the time-activity curve without background correction is applied.