Effect of low vs high haemoglobin transfusion trigger on cardiac output in patients undergoing elective vascular surgery: Post-hoc analysis of a randomized trial.

Abstract

During vascular surgery, restricted red-cell transfusion reduces frontal lobe oxygen (ScO₂ ) saturation as determined by near-infrared spectroscopy. We evaluated whether inadequate increase in cardiac output (CO) following haemodilution explains reduction in ScO₂ . This is a post-hoc analysis of data from the Transfusion in Vascular surgery (TV) Trial where patients were randomized on haemoglobin drop below 9.7 g/dL to red-cell transfusion at haemoglobin below 8.0 (low-trigger) vs 9.7 g/dL (high-trigger). Fluid administration was guided by optimizing stroke volume. We compared mean intraoperative levels of CO, haemoglobin, oxygen delivery, and CO at nadir ScO₂ with linear regression adjusted for age, operation type and baseline. Data for 46 patients randomized before end of surgery were included for analysis. The low-trigger resulted in a 7.1% lower mean intraoperative haemoglobin level (mean difference, -0.74 g/dL; P < .001) and reduced volume of red-cell transfused (median [inter-quartile range], 0 [0-300] vs 450 mL [300-675]; P < .001) compared with the high-trigger group. Mean CO during surgery was numerically 7.3% higher in the low-trigger compared with the high-trigger group (mean difference, 0.36 L/min; 95% confidence interval (CI.95), -0.05 to 0.78; P = .092; n = 42). At the nadir ScO₂ -level, CO was 11.9% higher in the low-trigger group (mean difference, 0.58 L/min; CI.95, 0.10-1.07; P = .024). No difference in oxygen delivery was detected between trial groups (MD, 1.39 dL_O2 /min; CI.95, -6.16 to 8.93; P = .721). Vascular surgical patients exposed to restrictive RBC transfusion elicit the expected increase in CO making it unlikely that their potentially limited cardiac capacity explains the associated ScO₂ decrease.