Effect of lumbosacral transitional vertebrae on sagittal balance of lumbo-pelvic complexity assessed by quantitative whole-body CT imaging.

Abstract

The variation at the lumbosacral junction certainly results in occult alignment changes in the lumbo-pelvic complexity (LPC). This retrospective case-control study aims to investigate the influences of lumbosacral transitional vertebrae (LSTV) on sagittal lumbo-pelvic balance assessment and provide some recommendations for preoperative imaging evaluation. Based on whole-body computed tomography (CT) images, a total of 210 individuals with complete segmentation anomalies of LSTV were included and divided into 23 presacral vertebrae (PSV) (sacralization, n=102), 25 PSV (lumbarization, n=108). The control group with 24 PSV (normal, n=100) was matched by age and gender. Sagittal lumbo-pelvic parameters including pelvic incidence (PI), pelvic tilt (PT), sacral slope (SS), lumbar lordosis (LL), sacral table angle (STA), sacral kyphosis (SK), and pelvic radius (PR) were measured at the ontogenetical S1 (Ontog S1) level and the morphological S1 (Morph S1), respectively. These parameters were compared using t-test, Kruskal-Wallis H test and post hoc test. Spearman's rank correlation coefficient and linear regression were used to investigate the association of lumbo-pelvic parameters with LSTV types and measurement levels. All the parameters at the Ontog S1 differed significantly from those at the Morph S1 (all P<0.001). At the Ontog S1 level, PI, PT, SS, and LL were negatively correlated with vertebrae counts; SK and PR were positively correlated with vertebrae counts (all P<0.001). Instead, reverse results were obtained at the Morph S1 level. The measurement level and vertebrae counts were independent influence factors for the measurement of PI, PT, SS, SK, and PR (all P<0.05). Compared with the measured values of the matched controls, the variability of most lumbo-pelvic parameters (PI, SS, LL, STA, SK, PR values of 25 PSV subgroup, and PI, PT, SS, LL, STA values of 23 PSV subgroup) at the Morph S1 level were significantly smaller than that at the Ontog S1. The measurements of PT, SS, LL, and PR were less influenced by the measurement level and vertebrae counts than those of PI and SK. Morph S1 is more recommended for the measurements of most lumbo-pelvic parameters in patients with LSTV. The parameters (PT, SS, LL, STA, PR) are shown more stable and recommended to help reduce the effects caused by LSTV.