International Normalized Ratio (INR) Sample Rejection in Neck of Femur Fracture Patients: A Retrospective Closed-Loop Study From a Major UK Trauma Centre.

Background Patients presenting with neck-of-femur (NOF) fractures often require urgent surgery, as prolonged delays beyond 36 hours are associated with increased morbidity, mortality, and length of hospital stay, whereas shorter time-to-surgery intervals have been shown to improve outcomes. Many of these patients are elderly and on anticoagulant therapy; therefore, making accurate International Normalized Ratio (INR) assessment is crucial for determining surgical readiness and anaesthetic safety. The INR reflects the extrinsic pathway of coagulation and is prolonged in patients on warfarin or who have underlying coagulopathies. Inaccurate or rejected INR samples delay operative clearance, prolong fasting, and increase bed occupancy and cost of treatment. A frequent pre-analytical cause of INR rejection is underfilling of sodium citrate tubes, which alters the required 9:1 blood-to-anticoagulant ratio. Objective To improve the rejection rate of INR samples through a simple phlebotomy intervention involving staff education and the appropriate use of a discard tube before citrate collection. Methods A retrospective two-cycle closed-loop audit was conducted at Heartlands Hospital, part of the University Hospitals Birmingham (UHB) NHS Foundation Trust. The first cycle included NOF fracture patients admitted between July and August 2023, during which 399 INR samples were analysed. Of these, 66 (16.5%) were rejected, 62 (94%) due to underfilling and four (6%) due to haemolysis. Following targeted interventions, including staff education on correct discard-tube use with butterfly systems and the introduction of shorter-tubing blood collection sets, a second audit cycle was performed and included NOF patients admitted between July and August 2025. In this cycle, 261 INR samples were reviewed, of which 29 (11.1%) were rejected, 27 (93%) for underfilling and two (7%) for haemolysis. Rejection proportions were compared between cycles, and absolute and relative changes were calculated. Statistical significance of the observed difference was assessed using a two-proportion z-test (two-sided, α = 0.05). Data collection and review were performed using the UHB trust Prescribing Information and Communication System (PICS) electronic system. Results INR sample rejection rate decreased from 16.5% in cycle 1 to 11.1% in cycle 2, an absolute reduction of 5.4% and relative reduction of ~33% (p ≈ 0.11). Among patients taking some form of blood thinner (e.g. warfarin, direct oral anticoagulants, low molecular weight heparin), 132 (33%) in cycle 1 and 56 (25%) in cycle 2, INR rejection occurred in 17.4% and 19.6%, respectively. Most rejections were due to underfilling the INR sample tube. Conclusions In trauma patients, particularly those awaiting urgent NOF surgery, preventing INR sample rejection can significantly reduce avoidable operative delays. This closed-loop audit demonstrated that a simple, low-cost intervention focused on correct tube filling, discard-tube use, and appropriate equipment selection led to a clinically meaningful reduction in INR sample rejection rates. Most remaining rejections remain preventable, underscoring the importance of continuous education, reinforcement of best practice, and regular re-audit to sustain long-term improvement.