Allogeneic blood transfusion after hip fracture: risk factors and associated outcomes.

BACKGROUND

Acute blood loss anemia is a common complication of hip fractures. Due to this complication, the rate of allogenic blood transfusions remains high in this patient population. More recently, surgeons have utilized intraoperative tranexamic acid as a strategy to minimize blood loss.

PURPOSE: The purpose of this study was to identify risk factors and associated outcomes of blood transfusions in hip fracture patients.

STUDY

DESIGN

Retrospective cohort study.

METHODS

A trauma database from one academic medical center was queried for hip fracture patients (OTA/AO 31A or 31B). Demographic, clinical, quality, and cost data were obtained for each patient. A Score for Trauma Triage in Geriatric and Middle Aged (STTGMA), a validated risk predictive and matching tool, was calculated for each patient. Patients receiving blood transfusions during their hospitalization were identified. Multivariate logistic regression analysis identified independent risk factors for transfusion during admission. Patients receiving transfusions were matched to patients not receiving a transfusion based on STTGMA. Comparative analyses of matched cohorts were performed.

RESULTS

In total, 1,344 hip fracture patients (29.8% male and 70.2% female) with mean age 81.04 ± 10.18 were identified. Four hundred ninety-nine (37.1%) patients received a transfusion during their admission. Risk factors for transfusion included higher STTGMA scores (P = .006), lower hemoglobin levels on admission (P < .001), anticoagulation or antiplatelet use on admission (P = .019), and fracture repair with a long cephalomedullary nail (P = .005). Trauma risk score-matched cohort outcomes show transfusions correlated with increased mortality, complications, readmission, and longer, more costly hospital stays.

CONCLUSION

Surgeons should be cognizant of transfusion risk factors such as repair with long cephalomedullary nail and poorer matched outcomes correlated with patients receiving transfusions.

LEVEL OF EVIDENCE: Level 3.