JOA - 2026-06-24 - Journal Article
Risk Stratification in Periprosthetic Hip Fracture Surgery: The Risk Analysis Index is Superior to the Five-Item Modified Frailty Index for Predicting 30-Day Complications.
Jowkar N, Ruffner M, Koltenyuk V, Sokolik ID, Horsley H, Rohl M, Williamson T, Weick JW
Topics
Key Takeaway
RAI outperformed mFI-5 in predicting 30-day mortality (AUC 0.69 vs 0.53) and non-home discharge (AUC 0.71 vs 0.57) in 824 patients undergoing periprosthetic femoral fracture surgery.
Summary Depth
Choose how much analysis to show on this article page.
Summary
This study compared the predictive validity of RAI versus mFI-5 for 30-day complications in patients undergoing rTHA and/or ORIF for periprosthetic femoral fractures using a national surgical database. Multivariable logistic regression with AUC comparison via DeLong's test showed RAI was significantly superior for mortality (AUC 0.69 vs 0.53, P=0.033) and non-home discharge (AUC 0.71 vs 0.57, P<0.001). Neither index demonstrated significant predictive superiority for major/minor complications, readmission, reoperation, VTE, or wound complications.
Key Limitation
Both indices showed only moderate discriminative ability (AUC ≤0.71), limiting their standalone utility for individual surgical decision-making without integration of fracture-specific and implant-specific variables.
Original Abstract
BACKGROUND
Periprosthetic femoral fractures (PFFs) account for up to 16% of all revision total hip arthroplasties (rTHAs) with one- and two-year mortality rates of 23.4 and 29.2%, respectively. Given the substantial clinical and financial burden, accurate preoperative risk assessment is essential. This study evaluated the predictive performance of the Five-Item Modified Frailty Index (mFI-5) and Risk Analysis Index (RAI) in patients undergoing surgery for PFs, aiming to improve preoperative assessment, inform surgical decision-making, and facilitate postoperative planning.
METHODS
A total of 824 patients at least 18 years old who underwent rTHA and/or open reduction internal fixation for PFF were identified in a national surgical database. The RAI and mFI-5 were calculated for each patient. Outcomes were major and minor complications, readmission, reoperation, non-home discharge, wound complications, venous thromboembolisms, and mortalities. Multivariable logistic regression models for each index-outcome were fitted, controlling for body mass index (BMI), operative time, procedure type, and smoking status. The ability of RAI and mFI-5 to predict postoperative outcomes was assessed using the area under the receiver operating curve (AUC) with DeLong's test used to compare differences between correlated AUCs.
RESULTS
The RAI demonstrated significantly better predictive ability for mortality (AUC: 0.69 versus 0.53, P = 0.033) and non-home discharge (AUC: 0.71 versus 0.57, P < 0.001) compared with mFI-5. In adjusted models, RAI was significantly associated with mortality (OR [odds ratio] 1.08, 95% CI [confidence interval] 1.00 to 1.17, P = 0.040) and non-home discharge (OR 1.16, 95% CI 1.12 to 1.20, P < 0.001).
CONCLUSION
The RAI was superior to the mFI-5 in predicting mortality and non-home discharge for patients undergoing surgical management of PPFs. Our findings support the implementation of RAI as a tool for preoperative risk stratification and may be impactful in surgical planning and counseling of geriatric orthopaedic patients and their families.