Changes in fetal hemoglobin associated with erythrocyte transfusions are clinically relevant in SpO2 targeting: a retrospective cohort observational study.

Background There is a broad awareness of shifts in the oxygen hemoglobin dissociation (ODC) relationship associated with fetal hemoglobin (HbF) changes. However, quantification of the shift has been limited. Objective To quantify the shift of partial oxygen tension (PO2) associated with HbF and with changes after transfusion of adult erythrocytes (TAE) in preterm infants. Methods This is a single center, retrospective observational analysis of blood gas samples. The shifts of ODC and PO2 related to HbF were evaluated in two models. Either HbF or TAE status (0, 1, ≥2) were used as the independent variable. Multivariate analysis was used to correct for confounding effects (gestational age, postnatal age, source of blood gas sample as well as pH, SO2 and PCO2). Results There were 3,452 blood gas observations analyzed from 2,464 infants whose median gestational age was 334 weeksdays (IQR 296-363). With SpO2 between 90-95%, the ODC was shifted to the left (13 mmHg, 1.3 kPa). After adjusting for confounding variables, the number of TAEs (0, 1, ≥2), was highly significantly related to a shift (p<0.001), consistent with the percent fetal hemoglobin level (p<0.001). Based on the multivariate model (i.e., holding confounding parameters constant), with a SpO2 of 92% the PaO2 could be expected to shift markedly higher with 2 or more TAEs in an extremely preterm infant (7.3 mmHg, 0.97 kPa). Conclusion While preliminary, these data suggest that in vulnerable preterm infants a change to a slightly lower SpO2 target range following TAE could maintain equivalent PaO2 exposure.