Advances in the development and validation of physiological, accessible, practical and reliable markers to guide RBC transfusions are expected. Near-infrared spectroscopy (NIRS) is a non-invasive technique that uses the differential absorption properties of oxyhaemoglobin and deoxyhaemoglobin to measure the tissue oxygenation status that had previously been used to evaluate the influence on tissue oxygenation response produced by the RBC storage age.

In 2017, Leal- Noval et al. were the first group to explore in a randomised controlled trial the utility of regional cerebral oxygen saturation (rSO2) measured by NIRS compared with predefined Hb thresholds to guide RBC transfusion in adult neurocritical patients who had moderate anaemia (Hb values between 7 and 10 g/dL). Patients were randomly assigned to Hb-guided (to maintain Hb levels between 8.5 and 10 g/dL) or rSO2-guided RBC transfusion (to maintain forehead rSO2 >60%).

The introduction of an rSO2 cutoff into the RBC transfusion protocol resulted in a decreased number of transfused RBC units per patient (mean difference, 0.51 U [95% CI -1.008 to -0.002]; P = 0.04), although it did not reduce the RBC transfusion rate. In the rSO2 group, RBC transfusion was associated with an increase in Hb (77 ± 4 to 88.7 ± 7 g/dL) parallel to an increase in the rSO2 values (55 ± 4 to 58 ± 4%), suggesting a direct relationship between RBC transfusion and rSO2 (although not in the same proportion).

This innovative study had several limitations. It would also have been interesting to know if this rSO2 increase occurred in all patients, if it was homogeneous in all patients, if it was maintained over time, and if it was influenced by the age and volume of the transfused RBC, as shown by the same group in prevoius studies. In addition, the inclusion of patients with a wide range of Hb concentrations (from 7.1 to 9.9 g/dL) could result in including patients that may not be comparable, since patients with Hb close to 10 g/dL have almost 40% more erythrocyte mass. Presumably, the low number of patients included prevented a subgroup analysis based on the Hb level at inclusion. Also, it would have been interesting to report the rSO2 values in the Hb-guided group to determine whether RBC transfusion in patients with rSO2 values greater than those set in the intervention group were associated with increases in rSO2.

Although there were no differences in mortality or hospital stay between the two transfusion strategies (the study was not powered to evaluate these outcomes), and despite the limitations of this study, Leal-Noval et al. point directly to tissue oxygenation, the ultimate goal of RBC transfusion, and present physiological alternatives that could complement or replace arbitrary Hb thresholds to guide RBC transfusions.

Despite the multiple clinical trials and meta-analyses comparing two predefined transfusion thresholds (“restrictive” vs. “liberal” strategy), it would seem preferable to individualise the decision to transfuse by integrating physiological and individual clinical factors with the Hb concentration, avoiding the use of simple laboratory value as an indication of transfusion. As stated by the AABB Clinical Practice Guidelines on Red Blood Cell Transfusion Thresholds and Storage, “it is good practice to consider the haemoglobin level, the overall clinical context, patient preferences, and alternative therapies when making transfusion decisions regarding an individual patient” (Carson JL et al. JAMA 2016;316:2025-35). We need physiological, accessible, practical and reliable markers to guide the oxygen supply need and RBC transfusion effectiveness focused on the individual patient. This evaluated technology could help to optimise to use of blood: the right volume, to the right patient, at the right moment, in the right conditions, and with the right assessment.

– José A. García-Erce

Part of this commentary will be published: Ripollés-Melchor J, García-Erce JA, Vincent JL. Transfusion thresholds and red blood cells transfusion focused on tissue oxygenation (Editorial). Rev Esp Anestesiol Reanim 2018; in press. Acknowledgment: Dr. Ripollés-Melchor.