Risk factors affecting the incidence of postoperative periprosthetic femoral fracture in primary hip arthroplasty patients: a retrospective study.

The purpose of this study was to identify the characteristics and risk factors for postoperative periprosthetic femoral fracture (PFF). This was a retrospective cohort study of 108 patients with and 432 control patients without postoperative PFF. Demographic characteristics, surgery-related information (primary hip disease diagnosed, fixation, femoral stem, method of operation, and bone resorption of the proximal femur), and postoperative patient outcomes (hip function, treatment history, and patients' lifestyle behaviors) were recorded and compared between the groups. PFF characteristics, such as the classification, time, and cause, were also documented, and a Cox regression model was built to identify the independent risk factors for postoperative PFF in these patients. Six independent risk factors for postoperative PFF were identified, namely, advanced age (hazard ratio (HR) = 1.026, 95% confidence interval (CI) = 1.007-1.045), femoral neck fracture as the primary disease (HR = 4.536, 95% CI = 2.955-6.961), osteoporosis (HR = 2.043, 95% CI = 1.234-3.383), hemiarthroplasty (or HA, HR = 2.173, 95% CI = 1.327-3.558), bone resorption of the proximal femur (HR = 1.627, 95% CI = 1.090-2.430), and a standard- or long-stem femoral prosthesis (HR = 2.996, 95% CI = 1.480-6.067). The predictive values for a low risk (estimated incidence ≤ 50%), moderate risk (estimated incidence 51%-89%), and high risk (estimated incidence ≥ 90%) of PFF were ≤ 3.0 points, 3.0-10.0 points, and ≥ 10.0 points, respectively. Most patients with postoperative PFF had Vancouver type B fractures. Six independent risk factors for postoperative PFF were identified: advanced age, hip fracture as the primary disease, osteoporosis, HA, bone resorption of the proximal femur, and a long femoral stem.