Optimal cutoff point of vertebral body cross-sectional area as a morphological parameter for predicting lumbar spondylolysis

Abstract

The relationship between vertebral body cross-sectional area (VBCSA) and spinal conditions associated with increased lumbar lordosis, such as lumbar spondylolysis (LSL), is not known. We investigated the morphological parameter, VBCSA, to predict LSL. The VBCSA on MRI has not been studied for its association with LSL. We hypothesized that VBCSA is an important morphological parameter for LSL prediction. We collected VBCSA data from 90 patients (43 males and 47 females) with LSL and 91 control subjects (44 males and 47 females) who underwent lumbar spine magnetic resonance imaging (LS-MRI). Axial T2-weighted LS-MRI images were obtained from all participants. Using our picture archiving and communications system, we analyzed the VBCSA at the level of the L5 vertebral body by utilizing MRI. The average VBCSA was 2263.51 ± 306.02 mm2 in the male control group and 1820.92 ± 224.89 mm2 in the male LSL group. LSL patients had significantly lower VBCSAs (P < .001) than did the male controls. The average VBCSA was 1985.21 ± 258.05 mm2 in the female control group and 1553.73 ± 250.02 mm2 in the female LSL group, and the LSL patients also had significantly lower VBCSAs (P < .001) than did the female controls. The optimal VBCSA cutoff value in the male group was 2014.69 mm2 with 76.7% sensitivity, 75.0% specificity, and an area under the receiver operating curve (AUC) of 0.89 (95% CI: 0.82–0.95). In the female group, the optimal cutoff score was 1814.11 mm2 with 76.6% sensitivity, 76.6% specificity, and an AUC of 0.88 (95% CI: 0.82–0.95). VBCSA is a sensitive objective morphological parameter for assessing LSL, and a lower VBCSA is associated with a higher possibility of LSL. We believe that these results will be useful in diagnostic radiology for evaluating patients with LSL.