There is no consensus as to the best projection or correction method for first-pass radionuclide studies of the right ventricle. We assessed the effects of two commonly used projections, 30 degrees right anterior oblique and anterior-posterior, on the calculation of right ventricular ejection fraction. In addition two background correction methods, planar background correction to account for scatter, and right atrial correction to account for right atrio-ventricular overlap were assessed. Two first-pass radionuclide angiograms were performed in 19 subjects, one in each projection, using gold-195m (half-life 30.5 seconds), and each study was analysed using the two methods of correction. Right ventricular ejection fraction was highest using the right anterior oblique projection with right atrial correction 35.6 +/- 12.5% (mean +/- SD), and lowest when using the anterior posterior projection with planar background correction 26.2 +/- 11% (p less than 0.001). The study design allowed assessment of the effects of correction method and projection independently. Correction method appeared to have relatively little effect on right ventricular ejection fraction. Using right atrial correction correlation coefficient (r) between projections was 0.92, and for planar background correction r = 0.76, both p less than 0.001. However, right ventricular ejection fraction was far more dependent upon projection. When the anterior-posterior projection was used calculated right ventricular ejection fraction was much more dependent on correction method (r = 0.65, p = not significant), than using the right anterior oblique projection (r = 0.85, p less than 0.001).(ABSTRACT TRUNCATED AT 250 WORDS)