Academic Background

Preoperative anemia and red blood cell (RBC) transfusions are closely associated with poorer surgical outcomes. Although RBC transfusions can increase oxygen delivery, they are also linked to an increased risk of infectious complications, thromboembolic events, pneumonia, sepsis, cardiac events, and mortality. Therefore, when studying the impact of preoperative anemia on surgical outcomes, it is essential to consider both the benefits and potential harms of RBC transfusions. However, establishing a causal relationship between RBC transfusions and adverse clinical outcomes is challenging, as patients receiving RBC transfusions are often sicker and undergo more complex surgeries.

Few studies have assessed the mediating effect of transfusions on preoperative anemia and postoperative outcomes. Some research suggests that RBC transfusions may mitigate the impact of preoperative anemia on postoperative complications by increasing oxygen delivery. However, age-related pathophysiological changes may alter anemia tolerance. This study aims to investigate the effect of preoperative hemoglobin levels on 1-year major adverse cardiovascular and cerebrovascular events (MACCE) and all-cause mortality in patients undergoing major non-cardiac surgery, while considering the mediating effect of perioperative RBC transfusions.

Source of the Paper

This paper was co-authored by Fraser J. D. Morris, Rasmus Åhman, Alison Craswell, Helen Didriksson, Carina Jonsson, Manda Gisselgård, Henrik A. Andersson, Yoke-Lin Fung, and Michelle S. Chew. The authors are affiliated with the School of Health at the University of the Sunshine Coast, Australia, and the Department of Anaesthesia and Intensive Care at Linköping University, Sweden. The paper was published online ahead of print on October 16, 2024, in the British Journal of Anaesthesia, Volume 133, Issue 6, pages 1183-1191.

Study Design and Methods

This study is a prospective multicenter observational study involving patients aged ≥50 years undergoing elective major non-cardiac surgery at four Swedish hospitals. The primary endpoints were 1-year MACCE and all-cause mortality, while the secondary endpoint was a composite of 30-day outcomes, including mortality, MACCE, acute kidney injury (AKI), pulmonary embolism, anastomotic leakage, and postoperative infection.

The study employed causal mediation analysis, with preoperative hemoglobin levels as the exposure variable and RBC transfusion as the mediator. The hypothesis was that preoperative hemoglobin levels affect postoperative cardiovascular complications through perioperative RBC transfusions. The study design was based on a directed acyclic graph (DAG), which accounted for confounders such as age, sex, surgery duration, American Society of Anesthesiologists (ASA) physical status, and blood loss.

Results

A total of 1060 patients were included, with a mean age of 70 years, 45% of whom were female. Within one year, 16.1% of patients experienced MACCE, and 9.9% died. Preoperative hemoglobin levels were significantly associated with 1-year MACCE (β=-0.015, p=0.041) and all-cause mortality (β=-0.028, p<0.001). However, the volume of RBC transfusions was not directly associated with these outcomes and did not mediate the relationship between preoperative hemoglobin levels and 1-year MACCE (β=-0.001, p=0.451) or all-cause mortality (β=-0.002, p=0.293).

For the 30-day composite outcome, RBC transfusions had a significant mediating effect between preoperative hemoglobin and the 30-day outcome, but no direct association was observed (β=0.006, p=0.554). Further analysis revealed that age ≥65 years significantly moderated the relationship between RBC transfusions and the 30-day composite outcome.

Conclusions

Preoperative hemoglobin levels were significantly associated with 1-year MACCE and all-cause mortality, but this effect was not mediated by perioperative RBC transfusions. The findings suggest that while RBC transfusions may influence certain short-term outcomes, they do not significantly alter the impact of preoperative anemia on long-term prognosis. Further research is needed to confirm these findings and explore other potential mediating factors.

Key Findings

1. Important Findings: Preoperative hemoglobin levels were significantly associated with 1-year MACCE and all-cause mortality, but RBC transfusions did not mediate this relationship.
2. Methodological Innovation: The study employed causal mediation analysis to systematically evaluate the mediating effect of RBC transfusions on the relationship between preoperative anemia and postoperative outcomes.
3. Clinical Implications: The results suggest that while RBC transfusions may influence certain short-term outcomes, they do not significantly alter the impact of preoperative anemia on long-term prognosis. This provides new insights for clinicians managing anemia and transfusion strategies in the perioperative period.

Research Value

This study offers a new perspective on understanding the relationship between preoperative anemia and postoperative outcomes, particularly by revealing the limited role of RBC transfusions through causal mediation analysis. The findings have significant implications for clinical practice, suggesting that the management of preoperative anemia should consider multiple factors rather than relying solely on RBC transfusions. Additionally, the study provides direction for future research, particularly regarding the impact of age on the effectiveness of RBC transfusions and the assessment of long-term outcomes.

Other Valuable Information

The study also explored the moderating effect of age ≥65 years on the relationship between RBC transfusions and the 30-day composite outcome, finding that age significantly influenced this relationship. This finding suggests that transfusion strategies for elderly patients in the perioperative period may require special attention to optimize short-term outcomes.

Through systematic causal mediation analysis, this study elucidates the relationship between preoperative hemoglobin levels and long-term postoperative outcomes, highlighting the limited role of RBC transfusions in this relationship. The results provide important references for clinical practice and point the way for future research.