Obesity Surgery · Nov 14, 2026
Development and Internal Validation of a Nomogram Model for Identifying Preoperative Osteoporosis Among Metabolic Bariatric Surgery Candidates Using Routinely Available Clinical Variables
By Mo Z, Yang Y, Liu J et al.
Background Osteoporosis (OP) is a common metabolic bone disease, and metabolic bariatric surgery (MBS) candidates may have impaired bone health. This study aimed to develop a nomogram model based on routinely available clinical variables for identifying preoperative OP among MBS candidates. Methods This single-center cross-sectional study included 738 MBS candidates who underwent preoperative evaluation between July 2020 and December 2023. All participants had available preoperative dual-energy X-ray absorptiometry (DXA) measurements. Candidate predictors were initially selected using least absolute shrinkage and selection operator (LASSO) regression and subsequently entered into multivariable logistic regression to develop the final nomogram model. Model discrimination, calibration, and clinical utility were evaluated using the area under the receiver operating characteristic curve (AUC), calibration measures, and decision curve analysis (DCA), respectively. Internal validation was performed using 1,000 bootstrap resamples. Results Among 738 MBS candidates, 100 (13.6%) met the criteria for OP. Multivariable logistic regression identified seven variables independently associated with OP status, including age, body mass index (BMI), alanine aminotransferase (ALT), albumin (Alb), total bile acid (TBA), lymphocyte count, and sex, which were incorporated into the nomogram. The model achieved an AUC of 0.793 (95% CI: 0.747–0.838), with an optimism-corrected AUC of 0.763 after bootstrap validation. The calibration curve showed overall agreement between predicted and observed probabilities, and DCA suggested potential clinical utility within a range of threshold probabilities. Conclusion A nomogram based on routinely available preoperative clinical variables may facilitate the identification of preoperative OP among MBS candidates and demonstrate acceptable discrimination and calibration performance.