Background Surgery in the lesser pelvis is associated with a high complication rate as surgeons are spatially limited by solid anatomic structures and soft tissue borders. So far, only two-dimensional (2D) parameters have been used for risk stratification. Purpose To precisely measure the inner pelvic volume a computed tomography (CT)-based three-dimensional (3D) approach was established and compared to approximations by 2D parameter combinations. Material and Methods Thin-layered multi-slice CT datasets were used retrospectively for slice by slice depiction of the inner pelvic surface. The inner pelvic volume was then automatically compounded. Combinations of two to four 2D dimensions determined in 3D volume rendered reconstructions were correlated with the inner pelvic volume. Pearson's correlation coefficient and Chi square test were used for statistical calculations. Significance level was set at P < 0.05. Results In total 142 patients (91 men, 51 women) aged 64.8 ± 10.6 years at surgery were included in the study. Mean calculated pelvic volume was 1031.13 ± 180.06 cm3 (men, 996.57 ± 172.43 cm3; women, 1093.34 ± 178.39 cm3). Best approximations were obtained by combination of the 2D measurements transverse inlet and pelvic height for men (r = 0.799, P < 0.05) as well as transverse inlet, obstetric conjugate, interspinous distance and pelvic depth for women (r = 0.855, P < 0.05). Conclusion We describe a precise and reproducible CT-based method for pelvic volumetry. A less time consuming but still reliable approximation can be achieved by combination of two to four 2D dimensions.
Keywords: Abdomen/GI; CT spiral; anatomy; image manipulation/reconstruction; pelvis.