(B-025) Association Between Perioperative Hemoglobin Transfusion Thresholds and Renal Adverse Events in Cardiac Surgery: A Regression Discontinuity Cohort Study of 10,337 Patients
Monday, August 25, 2025
11:30 AM - 1:00 PM ET
Background: Red blood cell (RBC) transfusion thresholds in cardiac surgery significantly impact postoperative organ function. While a restrictive transfusion strategy (Hb < 7.0 g/dL) has demonstrated advantages over a liberal strategy (Hb < 10.0 g/dL), evidence regarding its impact on renal outcomes remains limited. The optimal transfusion threshold for patients with hemoglobin levels near 7.0 g/dL, particularly in intensive care unit (ICU) and complex intraoperative settings, remains unclear. Further research is needed to refine transfusion strategies and mitigate renal complications in this high-risk population.
Objectives: This study aims to: 1) delineate the distribution of hemoglobin transfusion thresholds across various countries among cardiac surgery patients; 2) evaluate whether a hemoglobin threshold of < 7 g/dL during and after cardiac surgery serves as the optimal transfusion standard for minimizing the risk of postoperative acute AKI and AKD in-hospital. Additionally, the research seeks to investigate the impact of red blood cell transfusion at differing hemoglobin thresholds (7.0 g/dL, 8.0 g/dL, 9.0 g/dL) on mortality rates in patients undergoing cardiac surgery.
Methods: We conducted a retrospective cohort study of patients undergoing cardiac surgery at three medical centers in eastern China between January 2015 and September 2023. A total of 10,337 eligible patients were analyzed using regression discontinuity analysis, logistic regression, survival analysis, and Cox proportional hazards models. A 'fuzzy' regression discontinuity approach was applied to compare outcomes among patients with hemoglobin levels just above and below the 7.0 g/dL threshold while adjusting for potential confounders. The primary outcome was acute kidney injury (AKI). Secondary outcomes included the requirement for continuous renal replacement therapy (CRRT) and in-hospital mortality.
Results: Baseline characteristics were comparable between transfused and non-transfused patients at the 7.0 g/dL threshold. However, in the intention-to-treat (ITT) analysis, routine transfusion at this threshold was associated with:
A 25% increased risk of AKI (95% CI: 3%–48%). A 2.4% higher likelihood of requiring post-operative CRRT (95% CI: 0.9%–3.9%). A 2.7% increase in in-hospital mortality (95% CI: 1.5%–3.9%).
Conclusions: Perioperative RBC transfusion at an Hb threshold of 7.0 g/dL in cardiac surgery was associated with increased AKI incidence, higher CRRT requirements, and elevated in-hospital mortality. These findings highlight the need for a reassessment of transfusion strategies to minimize renal complications, even in patients with Hb < 7.0 g/dL.