Transfusion requirements in hip fractures: Analysis of predictive factors

Background: The primary objective of this study was to identify factors which determine blood transfusion requirements in patients with hip fractures. Predicting a target population at a higher risk of requiring blood transfusion would enable a selective cross-matching policy and also help to establish the appropriate prophylactic measures. Methods: It was a retrospective study including the time period between February 2003 and February 2005. All patients admitted with fracture neck of femur were included in the study. The following variables were noted perioperative Hb levels, type of fracture and surgery, age and gender and blood transfusion in the perioperative period. Results: There were 238 female and 72 male patients. The mean age was 82.7 years; range 53-100. Out of a total of 310 patients, 49 patients required a postoperative blood transfusion. The mean preoperative Hb of patients who required transfusion was 11; S.D.1.49 while those who did not require a transfusion it was 12.5; S.D.1.42. There were 168 hemiarthroplasties and 142 DHS fixations. Patients undergoing a DHS had a mean preoperative Hb level of 12.1 and a mean postoperative Hb of 9.1. Transfusion was required in 33 (23%) patients. Patients having hemiarthroplasty had a mean preoperative Hb level of 12.4 and a mean postoperative Hb of 10. Transfusion was requried in 16 (9.5%) patients. The univariate analysis showed a significant relationship between postoperative transfusion and the preoperative Hb level (P = 0.0001) and the type of fracture (P = 0.001). However no relationship was found between transfusion and the age (P = 0.423) and the gender of the patient (P = 0.611). Preoperative Hb and type of fracture were found to be statistically significant on multivariate analysis as well. Conclusions: The results of our study indicate that the most important factor in the prediction for blood transfusion in fracture neck of femur is the preoperative level of haemoglobin although the type of fracture is also important. Restrictive cross-matching for the high risk cases would improve costs in healthcare delivery and prevent unnecessary use of blood resources.