Abstract
BACKGROUND: Postmenopausal women face a heightened risk of developing new vertebral compression fractures (NVCFs) following percutaneous kyphoplasty (PKP) for osteoporotic vertebral compression fractures (OVCFs). This study aimed to develop and validate a visual nomogram model capable of accurately predicting NVCF occurrence post-PKP to optimize treatment strategies and minimize occurrence. METHODS: This retrospective study included postmenopausal women diagnosed with OVCF who underwent PKP at the Affiliated Hospital of Shandong University of Traditional Chinese Medicine between January 2016 and January 2021. Patient data, including basic information, surgical details, imaging records, and laboratory findings, were collected. The patients were categorized into two groups based on NVCF occurrence within 2 years post-PKP: the NVCF group and the non-NVCF group. Following the utilization of least absolute shrinkage and selection operator (LASSO) regression for feature selection, a nomogram was constructed. Model differentiation, calibration, and clinical applicability were evaluated using receiver operating characteristic (ROC), calibration, and decision (DCA) curve analyses. RESULTS: In total, 357 patients were included in the study. LASSO regression analysis indicated that cement leakage, poor cement diffusion, and endplate fracture were independent predictors of NVCF. The nomogram demonstrated excellent predictive accuracy and clinical applicability. CONCLUSIONS: This study used LASSO regression to identify three independent predictors of NVCF and developed a predictive model that could effectively predict NVCF occurrence in postmenopausal women. This simple prediction model can support medical decision-making and is feasible for clinical practice.