Long Tran1,2, Guri Greiff2,3, Alexander Wahba3,4, Hilde Pleym3,5, Vibeke Videm1,6. 1. Department of Clinical and Molecular Medicine, Faculty of Medicine and Health Sciences, NTNU - Norwegian University of Science and Technology, Trondheim, Norway. 2. Department of Cardiothoracic Anaesthesia and Intensive Care, St. Olavs Hospital, Trondheim, Norway. 3. Department of Circulation and Medical Imaging, Faculty of Medicine and Health Sciences, NTNU - Norwegian University of Science and Technology, Trondheim, Norway. 4. Clinic of Cardiothoracic Surgery, St. Olavs Hospital, Trondheim, Norway. 5. Clinic of Anaesthesia and Intensive Care, St. Olavs Hospital, Trondheim, Norway. 6. Department of Immunology and Transfusion Medicine, St. Olavs Hospital, Trondheim, Norway.
Abstract
OBJECTIVES: Our goal was to investigate long-term mortality associated with red blood cell (RBC) transfusion among patients with anaemia undergoing cardiac surgery when adjusting for known risk factors. METHODS: Adults with preoperative anaemia as defined by World Health Organization criteria undergoing open-heart surgery from 2000 through 2017 were included. Cox regression was performed for long-term mortality (30 days-5 years), comparing patients who received ≥1 unit of RBC with those who did not. Unadjusted and multivariable analyses adjusted for risk factors were performed. RESULTS: The study included 1859 patients, 1525 (82%) of whom received RBC transfusion. A total of 370 (19.9%) deaths were registered between 30 days and 5 years; 88 patients (23.8%) died between 30 days and 1 year. The unadjusted hazard ratio (HR) associated with RBC transfusion was 2.09 (1.49-2.93, P < 0.001) from 30 days to 5 years postoperatively. The HR for RBC transfusion were 4.70 (1.72-12.81, P = 0.002) and 1.77 (1.23-2.55, P = 0.002) for 30 days-1 year and 1-5 years, respectively. Adjusting for perioperative risk factors, which included postoperative complications, the HR decreased to 1.16 (0.80-1.68, P = 0.43), 1.79 (0.63-5.12, P = 0.28) and 1.11 (0.75-1.65, P = 0.61) for observation time from 30 days to 5 years, 30 days to 1 year and 1 to 5 years, respectively. Results were similar when postoperative complications were excluded from the adjustment variables. CONCLUSIONS: No statistically significant association between RBC transfusion and long-term mortality was found when we adjusted for known risk factors. This study suggests that the observed difference in mortality in this patient group is largely due to patient-related risk factors.
OBJECTIVES: Our goal was to investigate long-term mortality associated with red blood cell (RBC) transfusion among patients with anaemia undergoing cardiac surgery when adjusting for known risk factors. METHODS: Adults with preoperative anaemia as defined by World Health Organization criteria undergoing open-heart surgery from 2000 through 2017 were included. Cox regression was performed for long-term mortality (30 days-5 years), comparing patients who received ≥1 unit of RBC with those who did not. Unadjusted and multivariable analyses adjusted for risk factors were performed. RESULTS: The study included 1859 patients, 1525 (82%) of whom received RBC transfusion. A total of 370 (19.9%) deaths were registered between 30 days and 5 years; 88 patients (23.8%) died between 30 days and 1 year. The unadjusted hazard ratio (HR) associated with RBC transfusion was 2.09 (1.49-2.93, P < 0.001) from 30 days to 5 years postoperatively. The HR for RBC transfusion were 4.70 (1.72-12.81, P = 0.002) and 1.77 (1.23-2.55, P = 0.002) for 30 days-1 year and 1-5 years, respectively. Adjusting for perioperative risk factors, which included postoperative complications, the HR decreased to 1.16 (0.80-1.68, P = 0.43), 1.79 (0.63-5.12, P = 0.28) and 1.11 (0.75-1.65, P = 0.61) for observation time from 30 days to 5 years, 30 days to 1 year and 1 to 5 years, respectively. Results were similar when postoperative complications were excluded from the adjustment variables. CONCLUSIONS: No statistically significant association between RBC transfusion and long-term mortality was found when we adjusted for known risk factors. This study suggests that the observed difference in mortality in this patient group is largely due to patient-related risk factors.