Studies in cardiac surgery patients have suggested that allogeneic erythrocyte blood transfusions are associated with an increased long-term mortality
. However, studies in patients undergoing surgery for hip fractures have found no effect of transfusion on short-term mortality
, but did not evaluate the effects on long-term mortality
. The purpose of this study is to analyze the effect of allogeneic erythrocyte blood transfusions on long-term mortality
Charts of all patients undergoing surgery for hip fracture
(International Classification of Disease 820.0-820.9) between January 1, 2003 and December 31, 2005 were reviewed for demographic, comorbidities, laboratory values, use of and age of transfused blood products. Death was determined from the Social Security Death Index. Survival
was analyzed with Cox models and Kaplan-Meier statistics. To control for biases in this retrospective study, a subpopulation was analyzed after propensity matching using Cox modeling.
Thirty-one of the 59 patients (53%) dead at follow-up had received allogeneic erythrocyte transfusions, compared with 59 of 170 survivors (35%) (p
= 0.02). However, the increased risk of death was time dependent. Transfusion became a risk factor for death only after at least 90 days after surgery. By Cox modeling, transfusion was associated with an increased risk of death (relative risk = 3.386, 95% CI = 1.255–4.534, p
= 0.01; c-statistic = 0.612 ± 0.055, p
= 0.03). Seventy-four (32%) of patients were matched using propensity analysis. Similar to the total population, the increased mortality
associated with transfusion did not occur for at least 90 days. Using Cox proportional hazard modeling in propensity-matched patients who survived at least 90 days after surgery, transfusion remained a predictor of death (relative risk = 3.760, 95% CI = 1.216–11.626, p
We found that use of allogeneic erythrocyte transfusions to patients undergoing surgical repair of hip fractures was associated with an increased risk of death. This risk started after 90 days from surgery and persisted the length of follow-up.