Nomogram for Osteoporosis Risk Using LDCT Trabecular Parameters

Background: Low-dose computed tomography (LDCT) offers a unique opportunity to assess osteoporosis risk during routine lung cancer screening. This study aims to develop an integrated prediction model using trabecular bone features from LDCT and clinical factors to identify high-risk individuals early. Methods: This retrospective observational cohort study included 429 adults who underwent both DEXA and LDCT scans within one week at Kaohsiung Veterans General Hospital (2018–2022). Clinical data, including demographics, lifestyle factors, and comorbidities, were extracted from electronic records. Participants were categorized into osteoporotic (T-score ≤ −2.5) and non-osteoporotic groups. Trabecular bone morphometry was assessed at the T12 vertebra using QUIBIM Precision® software, analyzing parameters such as BV/TV, Tb.Th, Tb.N, Tb.Sp, D2D, D3D, and QTS. ROI placement and measurements followed standardized protocols. Ethical approval was obtained, and informed consent was waived. Statistical analyses included t-tests, ROC curves, logistic regression, and Delong tests to compare clinical and trabecular predictors of osteoporosis using SPSS v22. Results: In this study of 429 individuals, osteoporosis was significantly associated with female gender, older age, lower BMI, and smaller waist circumference. Trabecular bone morphometry revealed that osteoporotic individuals had significantly thinner trabeculae (lower Tb.Th), higher trabecular number (Tb.N), and more complex trabecular architecture (higher D2D/D3D), with lower QTS. Logistic regression showed that Model 1—the model combining clinical factors and Tb.N—showed a slightly higher predictive performance (AUC 0.738) than Model 2 (AUC 0.711), although the improvement was modest (p = 0.022). A nomogram based on age, sex, BMI, waist circumference, and Tb.N effectively estimated osteoporosis probability, providing a clinically useful tool for risk stratification. Conclusions: In conclusion, combining trabecular bone morphometry (Tb.N) from routine LDCT with age, sex, BMI, and waist circumference enhances osteoporosis risk prediction, enabling personalized assessment without extra radiation during standard lung cancer screening.