Impact of Restrictive Transfusion Strategy in CABG: A Retrospective Analysis

Background: Restrictive Transfusion Strategies (RTS) are widely used in perioperative care, showing superior outcomes. However, their use in cardiac surgery sparks debate due to conflicting trials reporting increased mortality. In our institution, RTS was adopted for Coronary Artery Bypass Grafting (CABG) in 2014. We retrospectively analyzed 3,178 CABG cases from 2010 to 2018, assessing mortality, kidney function, delirium, stroke, and infections. This study aims to evaluate the safety and effectiveness of a RTS in CABG.

Methods: We conducted a retrospective analysis of 3178 consecutive CABG patients, excluding combined and pediatric cases. The study comprised two groups: Pre-RTS (pRTS, n = 1533) from January 2010 to February 2014, and RTS group (RTS, n = 1,645) from 03/2014 to 09/2018. Starting March 2014 on, our institution lowered the minimum acceptable hemoglobin (Hb) level from 9.0 g/dL to 7.0 g/dL, unless hemodynamic instability was present. The primary endpoint encompassed in-hospital mortality, stroke, severe delirium, acute kidney injury and wound infections. Secondary endpoints included transfusion frequency and perioperative minimum Hb values, evaluating RTS effectiveness.

Results: The implementation of the RTS yielded noteworthy findings. There was a reduction of 0.6 red blood cell (RBC) units per patient (4.3 vs. 3.7 units, p = 0.04), along with an increase in patients not requiring transfusion (45% vs. 55%, p < 0.001). The primary endpoint, a composite measure, showed a significant reduction (38.7% vs. 34.6%, p = 0.017). Moreover, a notable 7% decrease in Acute Kidney Injury (AKI) was observed (33.7% vs. 26.4%, p < 0.001). However, there were significant increases in mortality (2.4 vs. 3.8%, p = 0.031) as well as postoperative wound infections. In a regression model for perioperative mortality, RTS showed a consistent effect size, although the p-value did not reach significance (OR 1.56, p = 0.051) after adjusting for risk factors.

Conclusion: The introduction of RTS correlated with a decreased primary endpoint, primarily driven by reduced perioperative acute kidney injury. Although there was a slight but significant increase in mortality within the RTS group, this effect did not reach significance in the regression model. Our findings align with the TITTRE and TRICCS trials, highlighting the positive impact on kidney injury achievable in practical clinical settings. However, the increased incidence of certain adverse events suggests a need for cautious application of a restrictive transfusion strategy in high-risk patients.

Publication History

Article published online:
13 February 2024