Adoption of a Restrictive Transfusion Strategy Is Associated with Reduced Incidence of Acute Kidney Injury in CABG Patients

Objectives: The association between postoperative acute kidney injury (AKI) and red blood cell (RBC) transfusions in cardiac surgery has been noted in multiple observational studies. However, the two large RCTs (TRICCS, TITTRe2) in which a restrictive transfusion strategy was tested against a liberal transfusion regime failed to show a significant effect on the incidence of postoperative AKI. The aim of this study was to investigate the effect of implementing a restrictive transfusion strategy on the incidence of postoperative acute kidney injury on a CABG collective in a retrospective sequential cohort analysis.

Methods: Between 01/2010 and 11/2018, a total of 3,178 patients received isolated CABG surgery of which 1,621 were treated with a liberal transfusion threshold and 1,579 patients belonged to the restrictive transfusion group (RT) after implementation of a patient blood management protocol in 03/2014. For both groups, perioperative renal parameters, hemoglobin values and blood product transfusion records were analyzed and compared as well as comorbidities from the medical history and intraoperative parameters. A logistic regression model was established from initially 27 perioperative parameters to find factors with a significant impact on the incidence of AKI.

Result: Implementation of the restrictive protocol led to a significant decrease in RBC transfusions from an average of 4.3 to 3.7 units (p = 0.044). Minimum hemoglobin values decreased slightly in the RT group (7.9 vs. 7.6 g/dL, p < 0.001). The RT group showed a decrease in the incidence of postoperative AKI from 33.7 to 26.5% (OR: 0.70, p < 0.001). The incidence of stage III AKI was also significantly decreased (14.4 vs. 10.4%, OR: 0.75, p = 0.012). There was no difference in the rate of postoperative hemodialysis. In a logistic regression model, five factors were shown to impact the incidence of AKI: High volume transfusions (> 5 RBC) showed the largest effect size (OR: 1.73, p < 0.0001) while avoiding transfusions was the strongest protective factor (OR: 0.56, p < 0.0001). The other factors were preoperatively reduced kidney function (OR: 1.62, p < 0.0001), reduced ejection fraction (OR 1.48, p = 0.0172) and short ECC time (OR: 0.74, p = 0.0049).

Conclusion: The implementation of a restrictive transfusion regime was effective at reducing the amount of RBCs administered and accompanied by a decrease in the incidence of acute kidney injury. Logistic regression analysis confirmed that this decrease was due to restriction of RBC transfusions.

No conflict of interest has been declared by the author(s).

Publication History

Article published online:
19 February 2021