Preoperative anemia in patients undergoing coronary artery bypass grafting predicts acute kidney injury

Objectives: Recent authoritative studies suggested that low preoperative hemoglobin concentration may affect cardiac surgery outcomes. This study aimed, primarily, to investigate whether preoperative anemia is an independent determinant of adverse events after coronary artery bypass grafting and, secondarily, to evaluate the potential dose responsiveness between anemia severity and primary end points. Methods: This single-center prospective study investigated 1214 consecutive patients undergoing coronary artery bypass grafting between January 2004 and June 2007, collecting 100 variables per patient. In 1047 patients (median age 64 years, 18.8% female, 38.9% diabetic, 31.9% urgent/emergency, 15.3% with low preoperative left ventricular ejection fraction) who underwent on-pump procedures and received no preoperative transfusion, the prevalence of preoperative anemia (according to World Health Organization definition) and its unadjusted and adjusted relationships with in-hospital death, cardiac morbidity, and acute kidney injury (AKI-RIFLE [Risk, Injury, Failure, Loss, End-stage kidney disease] criteria) were obtained. Results: The prevalence of preoperative anemia was 28%. In-hospital death averaged 3.9%, cardiac morbidity 7.3%, and acute kidney injury 4%. Unadjusted odds ratios (Ors) for in-hospital death, cardiac morbidity, and acute kidney injury were 3.8 (95% confidence interval [CI] 2.0-7.3), 1.7 (95% CI 1.1-2.8), and 4.0 (95% CI 2.1-7.6), respectively. Adjusting for anemia in confounders proved an independent predictor of acute kidney injury (OR 2.06; 95% CI 1.14-3.70), whereas the cardiac morbidity and in-hospital mortality were independently predicted by kidney function. No dose-response relationship emerged between anemia severity and acute kidney injury. Conclusions: Preoperative anemia is independently associated with acute kidney injury after coronary artery bypass grafting. Further studies are warranted to determine whether preoperative low hemoglobin concentration is a marker of severity of illness or a modifiable risk factor.