The accurate measurement of cardiac chamber volume is of major importance in assessing cardiac performance. Accurate equilibrium radionuclide volume estimations are difficult to obtain, due to the geometry of the chambers, and the physical characteristics of the imaging system. The purpose of this study was to examine the effects of imaging projections on relative cardiac chamber volumes, indexes, and stroke volume ratios. Twenty-two male patients, free of clinical evidence of disease, were studied. A series of four 2-minute acquisitions were made with the patient successively imaged in the anterior, 30 degrees left anterior oblique (LAO), 45 degrees LAO, and 60 degrees LAO projections with 15 degrees of caudal inclination. Filtered stroke volume and original images were used by the operator to assign right ventricular (RV), left ventricular (LV), and a combined right and left ventricular (TOT) regions-of-interest. From the data we determined end-diastolic counts (EDC), end-systolic counts (ESC), stroke counts (SC), ejection fractions (EF), and R/L stroke count ratios. The following changes were observed as the projection was moved from the anterior to 60 degrees LAO: 1) all RV parameters decreased in value, including, RVEDC (P less than .001), RVESC (P less than .01), RVESC (P less than .01) and RVEF (P less than .001); 2) LVEDC and LVESC (both P less than .01) increased while LVEF decreased (P less than .004); and 3) the R/L stroke count ratio decreased (P less than .001). Variability could be explained by 1) chamber overlap and geometry; 2) patient variability; and 3) intrachamber, interchamber and chest wall photon attenuation and scatter. We suggest that close attention to detail, with computer assistance, to optimally position the patient may reduce the effect of inherent limitations in radionuclide volumetric measurements, thus improving the reliability and usefulness of existing studies.