Zeroual and colleagues report the results of a single-centre, single-blinded randomised controlled study of cardiac surgery patients comparing two strategies for transfusion of red blood cells. The control group followed a standard regime with transfusion of red blood cells if the haemoglobin level fell < 9 g/dL. In the intervention group, transfusion of red blood cells was given if the haemoglobin level fell < 9 g/dL and the central venous oxygen saturation was below 65%. The study comprised 50 cardiac surgery patients in each group. Primary outcome was frequency of transfusion in ICU the two study groups. Not surprisingly, the intervention group requiring fulfilment of two criteria had significantly less transfusion. Secondary outcomes were frequency of transfused patients until discharge (84% vs. 100%) and cumulative units transfused during hospital stay (96 units vs. 124 units) were both significantly lower in the SvO2 dependent group. Morbidity and mortality were comparable between groups in the present study.
The authors raise an important question as to when transfusion is indicated, but there are no clear arguments as to why a SvO2 below 65% should be a relevant threshold. The fact that transfusion of red blood cells with a SvO2 above 65% will not increase the SvO2 further is not equivalent to a critical organ perfusion if SvO2 is below 65%, especially when one considers that the TRICS III trial did not shown any difference in mortality or morbidity between a transfusion threshold of 7.5 g/dL and 9 g/dL. The choice of primary outcome is somewhat surprising since it contains a self-fulfilment where the control group patient requires only one indication, whereas the intervention group patient requires two criteria to receive a red blood cell transfusion.
The need for more physiological criteria in a balanced approach of transfusion strategy is highlighted in the accompanying editorial by Weiskopf and Cook. The commentary gives a nice overview of the recent clinical studies and the challenges associated with prophylactic treatment to avoid organ ischaemia and choice of physiological variables for guidance in this setting. The authors also underline the importance of acknowledging that well-designed randomised clinical studies “yield the most rigorous means of testing clear hypotheses. However, they yield population-based measures of intervention effects, and do not give insights into how best to treat specific patients”. The editorial is certainly worthwhile reading as it provides insight into the area of interest beyond the study by Zeroual and colleagues.
– Hanne Berg Ravn
- Zeroual N, Blin C, Saour M, et al. Restrictive transfusion strategy after cardiac surgery. Anesthesiology 2021;134:370-380.
- Weiskopf RB, Cook RJ. Known and unknown unknowns in making erythrocyte transfusion decisions. Anesthesiology 2021;134:359-362.