Regional cerebral oxygen saturation variability and brain injury in preterm infants

Objective: To examine whether variation of regional cerebral oxygen saturation (rScO₂) within three days after delivery predicts development of brain injury (intraventricular/cerebellar hemorrhage or white matter injury) in preterm infants.

Study design: A prospective study of neonates <32 weeks gestational age with normal cranial ultrasound admitted between 2018 and 2022. All received rScO₂ monitoring with near-infrared spectroscopy at admission up to 72 h of life. To assess brain injury a magnetic resonance imaging was performed at term-equivalent age. We assessed the association between rScO₂ variability (short-term average real variability, rScO_2ARV, and standard deviation, rScO_2SD), mean rScO₂ (rScO_2MEAN), and percentage of time rScO₂ spent below 60% (rScO_2TIME<60%) during the first 72 h of life and brain injury.

Results: The median [IQR] time from birth to brain imaging was 68 [59-79] days. Of 81 neonates, 49 had some form of brain injury. Compared to neonates without injury, in those with injury rScO_2ARV was higher during the first 24 h (P = 0.026); rScO_2SD was higher at 24 and 72 h (P = 0.029 and P = 0.030, respectively), rScO_2MEAN was lower at 48 h (P = 0.042), and rScO_2TIME<60% was longer at 24, 48, and 72 h (P = 0.050, P = 0.041, and P = 0.009, respectively). Similar results were observed in multivariable logistic regression. Although not all results were statistically significant, increased rScO₂ variability (rScO_2ARV and rScO_2SD) and lower mean values of rScO2 were associated with increased likelihood of brain injury.

Conclusions: In preterm infants increased aberration of rScO₂ in early postdelivery period was associated with an increased likelihood of brain injury diagnosis at term-equivalent age.