JOA - 2026-08-18 - Journal Article
Predictors for Blood Transfusion Risk in Patients Undergoing Total Hip Arthroplasty for Femoral Neck Fracture: A Two-Center Retrospective Cohort Study.
Zhu J, Yang H, Lou Z, Chen S, Tu M, Zhou Y
Topics
Key Takeaway
A four-variable nomogram (preoperative anticoagulant use, TXA use, surgical duration >120 min, preoperative hemoglobin) predicted post-THA transfusion risk with AUROC 0.90 in a cohort where 20.5% of patients required transfusion.
Summary Depth
Choose how much analysis to show on this article page.
Summary
This two-center retrospective study identified independent predictors of postoperative blood transfusion in older patients undergoing THA for femoral neck fracture, then constructed and internally validated a nomogram. Of 391 patients, 80 (20.5%) required transfusion; multivariate logistic regression identified preoperative anticoagulant use, absence of intraoperative TXA, surgical duration >120 minutes, and lower preoperative hemoglobin as independent predictors. The nomogram achieved AUROC 0.90 with good calibration and demonstrated net clinical benefit on DCA across threshold probabilities of 10–85%.
Key Limitation
The external validation cohort size is not reported, preventing assessment of whether the nomogram's AUROC of 0.90 is reproducible in an adequately powered independent sample.
Original Abstract
BACKGROUND
Postoperative blood transfusion is common in older patients undergoing total hip arthroplasty (THA) for femoral neck fractures, but is associated with an increased risk for complications. This study aimed to identify relevant predictors and develop and validate a nomogram prediction model to guide perioperative blood management and reduce transfusion rates.
METHODS
We retrospectively analyzed older patients who underwent THA for femoral neck fractures at two institutions between January 2017 and August 2023. Of the 391 patients, 80 (20.5%) required postoperative blood transfusion. A dataset with a smaller sample size was used for external validation. Patients' baseline characteristics, surgical data, and preoperative laboratory parameters were collected. We performed univariate and multivariate logistic regression analyses to identify independent predictors for postoperative blood transfusion and then constructed a nomogram based on those predictors. We assessed the model's performance using the area under the receiver operating characteristic curve (AUROC) and calibration curves and evaluated its clinical utility using decision curve analysis (DCA).
RESULTS
Multivariate analysis identified four independent predictors: preoperative anticoagulant use, intraoperative tranexamic acid (TXA) use (protective factor), prolonged surgical duration (greater than 120 minutes), and lower preoperative hemoglobin. The nomogram demonstrated excellent discrimination (AUROC = 0.90) and good calibration. The DCA revealed a substantial net benefit of the nomogram across a wide range of threshold probabilities (10 to 85%) compared to the default strategy of transfusing all or no patients.
CONCLUSIONS
Preoperative anticoagulant use, intraoperative TXA administration, surgical duration, and preoperative hemoglobin levels are key predictors for post-THA blood transfusion in older patients who have femoral neck fractures. Our nomogram should be considered a preliminary tool to be externally validated before clinical implementation. Nevertheless, optimizing modifiable intraoperative factors, especially TXA administration and surgical duration minimization, remains a crucial strategy to reduce unnecessary transfusions in this acute setting.