Volumetric analysis of right ventricular cadaver models viewed in axially angled projections.

Axial cinematography of 15 cadaver cast models of the right ventricle was performed, using 16 different real and simulated single and biplane axial oblique projections. Heart volumes were then calculated by Dodge's area-length method and Ferlinz' method. In our volumetric studies of models, the smallest positive deviation from real volumes was 3.7% with the simulated long-axis projection (SLP2), evaluated from the frontal plane and calculated by the area-length method. Volumetric determinations of most usefulness, as ranked by mean differences and mean quadratic deviations, were achieved with the simulated long-axis projection, (SLP1--ranked third and SLP2--ranked first), evaluated from the frontal plane and calculated by Dodge's method; with biplane orientation and calculation by Ferlinz' method (SLP1 + SLP1C1-- ranked seventh, deviation from real volumes was 23%, SLP2 + SLP2C2--ranked fourth). We found that single-plane hepatoclavicular projection (HCP--ranked second and fifth), calculated by Dodge's or Ferlinz' method, as well as single-plane sitting-up projection (SUP--ranked sixth), calculated by the area-length method, were also acceptable for right ventricular volume measurements.