Construction and validation of a risk prediction model for postoperative recurrent fracture in patients with osteoporotic vertebral fracture based on serum IGF-1/CTX/P1NP and other indicators

ZHANG Cheng-yan; LI Jun-jie; RAO Yao-jian

doi:10.3969/j.issn.2095-252X.2026.01.011

Abstract：Objective To construct and validate a risk prediction model for postoperative recurrent fracture in patients with osteoporotic vertebral fracture(OVF)based on serum insulin-like growth factor-1(IGF-1),cross-linked C-terminal peptide of type Ⅰ collagen(CTX)and total procollagen type Ⅰ N-terminal propeptide(P1NP)and other indicators.Methods The subjects were OVF patients who underwent vertebroplasty or posterior fusion in our hospital from May 2020 to May 2024,and a total of 215 patients were included.According to whether there was recurrent vertebral fracture during the 12-month follow-up after surgery,the subjects were divided into a non-recurrent fracture group(n=173)and a recurrent fracture group(n=42),and the subjects were divided into a training set and a validation set at a ratio of 7:3.LASSO regression and binary logistic regression analysis were used to explore the risk factors for recurrent fracture after surgery in patients with OVF.The static nomogram model was constructed using the"rms"package of R R 4.4.1 language and a web-based risk calculator was developed.The model was internally validated using the Bootstrap method(repeated sampling 1000 times),and the validity,calibration and discrimination of the model were evaluated by drawing decision curves,calibration curves and ROC curves.Results The results of multivariate logistic regression analysis showed that body mass index(BMI),bone mineral density(BMD),P1NP,IGF-1,CTX and osteoporosis course were independent risk factors for recurrent fracture after surgery in patients with OVF(P<0.05).Based on the six risk factors above,a nomogram model for risk prediction of recurrent fracture after surgery in patients with OVF was constructed.The ROC curve showed that the AUC of the model was 0.724[95%CI(0.628,0.821)],and the consistency index was 0.824;the decision curve indicated that the prediction model had a net benefit when the probability threshold was 0.050-0.950;the calibration curve of the model basically coincided with the ideal curve.Conclusions This study constructs and verifies a nomogram model for risk prediction of recurrent fracture after surgery in patients with OVF based on serum IGF-1,CTX,P1NP and clinical characteristics(BMI,BMD,osteoporosis course),and the model has good prediction performance.The developed web-based risk calculator can realize the real-time dynamic monitoring of the risk of recurrent fracture in patients after OVF surgery.