Can regional cerebral oxygen saturation guide red blood cell transfusion in high risk cardiac surgery?

Cardiac surgery (CS) is associated with a significant use of blood products. The aim of this study was to evaluate the effect of red blood cells concentrates (RBC) transfusion on regional cerebral oxygen saturation (SrO ) in patients undergoing CS with additive EuroSCORE I > 6. This is a prospective descriptive study. Patients with additive EuroSCORE I > 6 undergoing CS with extracorporeal circulation were included in the study. The demographic values, perioperative complications, hospital/ICU length of stay (LOS), as well as the preinduction baseline SrO values, the lowest SrO value during surgery, number of blood products transfused, hemoglobine (HB) and pre and postransfusional SrO2 values were recorded, and events of significant decrease in SrO below 20% of basal value or decrease below 50%. We considered responders those who had an increase post-transfusion SrO at least 20% of pre-transfusion value, an increase of standar deviation (7.9) or an increase up to basal SrO . Data from 57 patients were collected. The average additive EuroSCORE I was 7.4 (SD 2.6) and the EuroSCORE II was 6.1 (SD 7.4). 52% were male. 35.1% of patients received intraoperative transfusion of at least one unit of RBC. The overall mortality was 8.7% (N = 5). During surgery 29.8% of the overall sample presented a decrease of more than 20% of baseline SrO2 or a value lower than 50%. Patients with a significant decrease in SrO2 presented a higher rate of perioperative complications (P=0.04) and longer ICU-LOS 4.3 (SD 3.6) vs. 6.8 (SD 8.2) days (P=0.01) and hospital LOS 10.1 (SD 3.1) vs. 14.2 (SD 9.4) days (P=0.01). Pretransfusional HB was 7.4 (SD 0.8) mg/dl and postransfusional value was 8.4 (SD 0.8) (P =0.00). Pretransfusional SrO2 was 59 (SD 8.6) and increased non- significantly after RBC transfusion to 61.1 (SD 7.9) (P=0.1). Only 6 patients out of 21 could be considered responders. There were no significant differences in morbidity, mortality or LOS between responders and non-responders. In our population a non statistically significant increase in SrO2 was observed after RBC transfusion. When considering responders few patients were identified by SrO . In conclusion SrO might not be reliable triger to decide transfusion.