Low vertebral hounsfield unit values predict nonunion after single-level posterior lumbar interbody fusion

Study Design: Retrospective case-control study. Objective: To evaluate whether L1 vertebral body Hounsfield unit (HU) values measured on preoperative computed tomography (CT) can predict nonunion following posterior lumbar interbody fusion (PLIF). Background: HU values have been associated with bone condition and implant-related complications in spine surgery. However, their predictive value for bony nonunion after PLIF remains unclear. Materials and Methods: Patients aged older than 60 years who underwent single-level PLIF with preoperative CT and dual-energy x-ray absorptiometry (DXA) were included. Patients were divided into union and nonunion groups based on one-year postoperative CT and dynamic radiographs. Variables analyzed included demographic and clinical factors, lumbar spine and hip bone mineral density (BMD) T-scores, L1-HU value, and osteoporosis diagnosis. Multivariable Firth logistic regression identified predictors of nonunion. Receiver operating characteristic (ROC) curve analysis was used to determine the L1-HU threshold. The correlation between the L1-HU value and the lumbar spine BMD T-score was assessed using Pearson correlation coefficient (r) and linear regression. Results: Among the 79 patients, 66 met the inclusion criteria (mean age, 74.3 yr). At one year, 17 (25.8%) had nonunion. Multivariable Firth logistic regression identified the L1-HU value as the only significant independent predictor of nonunion ( P = 0.0047; odds ratio, 0.96; 95% CI, 0.93–0.99). ROC analysis yielded an area under the curve of 0.79 with a cutoff of 99.0 HU. Although the mean lumbar spine BMD T-score of all patients was within the normal range, the corresponding average L1-HU value was within the osteoporotic range. The L1-HU value showed a moderate correlation with lumbar spine BMD T-score (r = 0.55). Conclusions: The preoperative L1-HU value was the only independent predictor of nonunion after PLIF. The CT-based HU assessment can be a practical tool for preoperative evaluation and risk stratification in patients undergoing PLIF.