Background: Patients with anemia frequently undergo surgery, as it is unclear at what threshold clinicians should consider delaying surgery for preoperative anemia optimization. The primary objective of this study was to determine whether there is an association of varying degrees of anemia and transfusion with 30-day mortality.
Methods: This is a retrospective cohort study using the American College of Surgeons National Surgical Quality Improvement Program database from 2011 to 2013. Cohorts were analyzed based on preoperative hematocrit range-patients with: (1) no anemia, (2) hematocrit ≥33% and <36% in females or <39% in males, (3) hematocrit ≥30% and <33%, (4) hematocrit ≥27% and <30%, (5) hematocrit ≥24% and <27%, and (6) hematocrit ≥21% and less than 24%. Multivariable logistic regression was used to analyze the association of anemia and transfusion with 30-day in-hospital mortality.
Results: The odds for 30-day mortality increased incrementally as the hematocrit ranges decreased, in which preoperative hematocrit between 21 and 24% had the highest odds for this outcome (odds ratio [OR] 6.50, p < 0.0001) compared to the reference group (no anemia). The use of transfusion increased the odds of mortality even further (OR 5.57, p < 0.0001). Among patients that received an intra-/postoperative transfusion, preoperative anemia was not predictive of mortality.
Conclusions: Healthcare providers making preoperative clinical decisions for patients undergoing elective surgery should consider the degree of preoperative anemia and likelihood of perioperative transfusion.