[Construction and evaluation of a nomogram prediction model for periprosthetic fractures after total hip arthroplasty].

OBJECTIVE

To construct and evaluate nomogram prediction model for periprosthetic fractures in patients undergoing total hip arthroplasty (THA).

METHODS

A total of 538 patients who underwent THA from April 2013 to February 2019 were selected as the research subjects, including 318 males and 220 females, aged 40 to 60 years old with an average age of (50.79±6.37) years old. All patients with THA were divided into non-fracture group (506 patients) and fracture group (32 pathents) according to the 3-year follow-up results. Univariate and multivariate Logistic regression analyses were performed to analyze the influencing factors of postoperative periprosthetic fractures in patients with THA. A nomogram prediction model for periprosthetic fractures in patients undergoing THA was constructed, and the validity and discrimination of the prediction model were evaluated.

RESULTS

The proportion of patients with osteoporosis, trauma history, and hip revision in the fracture group were higher than those in the non-fracture group(<0.05), and the proportion of bone cement prosthesis was lower than that in the non-fracture group(<0.05). The osteoporosis status[=4.177, 95%(1.815, 9.617), <0.05], trauma history[=7.481, 95%(3.104, 18.031), <0.05], and hip revision[=11.371, 95%(3.220, 40.153, <0.05] were independent risk factors for postoperative periprosthetic fractures in patients undergoing THA, cemented prosthesis [=0.067, 95%(0.019, 0.236), <0.05] was an independent protective factor for postoperative periprosthetic fractures in patients undergoing THA(<0.05). Hosmer-Lemeshow goodness of fit test showed that =7.864, =0.325;the area under the curve (AUC) for periprosthetic fractures in patients undergoing THA was 0.892 with a sensitivity of 87.5% and a specificity of 77.7% by receiver operating characteristic(ROC) curve.

CONCLUSION

The nomogram prediction model for periprosthetic fractures after THA constructed in this study has good discrimination, which is beneficial to clinical prediction of periprosthetic fractures in patients undergoing THA, and facilitates individualized fracture prevention.