Retrospective analysis and prediction model for vascular and nerve injuries in elderly patients following total knee arthroplasty.

OBJECTIVE

This study aimed to investigate the risk factors associated with vascular and nerve injury following total knee arthroplasty (TKA) in elderly patients and to develop, as well as validate, a corresponding predictive model.

METHODS

A cohort of 380 elderly patients who underwent TKA at Weifang People's Hospital was included. Participants were categorized into a vascular/nerve injury group (n=34) and a non-injury group (n=346). The incidence of postoperative vascular or nerve injury was documented. Risk factors were identified using chi-square tests and multivariate logistic regression analysis. The model's discrimination and calibration were assessed by the receiver operating characteristic (ROC) curve and the Hosmer-Lemeshow (H-L) test, respectively.

RESULTS

The incidence of vascular and nerve injury after TKA was 8.95%. Multivariate analysis identified the following independent risk factors: deformity correction ≥20° (OR=20.063, P<0.001), knee valgus ≥12° (OR=7.717, P=0.004), tourniquet time ≥2 h (OR=5.451, P=0.038), epidural anesthesia (OR=7.525, P=0.003), preoperative neuropathy (OR=24.906, P<0.001), and improper use of double-click electrocoagulation (OR=6.175, P=0.013). The H-L test indicated good calibration of the predictive model (P>0.05). The model's AUC was 0.806, with a sensitivity of 79.47% and specificity of 75.39%.

CONCLUSION

In elderly patients undergoing TKA, close attention should be paid to several risk factors, including significant deformity correction (≥20°), knee valgus (≥12°), preoperative neuropathy, improper use of double-click electrocoagulation, epidural anesthesia, and prolonged tourniquet application (≥2 h). These factors are associated with an increased risk of vascular and nerve injury.