Reproducibility and Accuracy of Pelvic Tilt in Predicting the Difference Between Pelvic Incidence and Lumbar Lordosis Value

Abstract

We investigated the pelvic morphologic factors that determine the degree of pelvic incidence (PI)-lumbar lordosis (LL) mismatch. Overall, 306 patients were included. The regional and global sagittal parameters were measured. Linear regression analyses were performed for 4 pelvic parameters and PI-LL mismatch. E1 and E2 were defined as linear regression equations between pelvic tilt (PT) and PI-LL mismatch and PI and PI-LL mismatch, respectively. The patients were categorized by cluster analysis using the hierarchal method for the 4 pelvic parameters. E1 and E2 showed statistical significance; however, the coefficient of determination of E1 was higher than that of E2 (R2= 0.675 vs. 0.238; P < 0.01). Sex, LL, E1, and E2 showed significant differences in the regional parameters. The T1 pelvic angle (TPA), spinosacral angle (SSA), and incidence angle of inflection points (IAIPs) showed significant differences in global parameters (P < 0.01). The IAIPs and TPA were low in the anteverted pelvis group and high in the retroverted pelvis group (P < 0.001). The SSA was low inthe small pelvis group and high in the large pelvis group (P< 0.001). The proportion of women in the large pelvis group (93%) was significantly higher than that in the other groups (P < 0.01). The individual differences between the PI and LL values can be more accurately determined using the individual PT, and the optimal PT amount will differ depending on the pelvis shape. The increase in the TPA and IAIPs corresponded to the PT, and the SSA increased in accordance with the pelvic size.