Can Use of Cerebral Oxygenation Predict Developmental Outcomes in Preterm Infants With NEC?

In this issue of Pediatrics, Howarth et al¹  report on a completed longitudinal study of 46 preterm infants born at <30 weeks’ gestational age in which they recorded cerebral oxygenation using near-infrared spectroscopy (NIRS) for 1 hour per week until hospital discharge or transfer. The 6 infants who developed necrotizing enterocolitis (NEC) had consistently lower levels of cerebral oxygenation (mean −6.6%) and small increases in cerebral fractional tissue oxygen extraction (mean 0.08) throughout the study. Using multilevel mixed-effects linear models nested within each infant, the authors adjusted for a large number of confounding variables, including patent ductus arteriosus, sex, ethnicity, volume of enteral feeding, hemoglobin, and gestational age. Although the small number of infants who developed NEC limited the study, the study provides preliminary evidence that lowered cerebral oxygenation may underpin the worse neurodevelopmental outcomes that have been reported in preterm infants with NEC.^2–4 

Worldwide, >15 million infants are born preterm (<37 weeks’ gestation) each year, and these rates are increasing. Improvements in the care of these infants have dramatically improved survival. However, contrary to expectations, recent data, in which outcomes in children born at <28 weeks’ gestational age in 1991–1992, 1995, and 2005 at 8 years of age were compared, revealed that rates of major neurosensory disability had not improved and that academic performance was poorer at early school age in 2005 than in earlier eras.⁵  There is obviously an urgent need to improve neurodevelopmental outcomes of the growing number of infants surviving preterm birth.

There is mounting evidence that low cerebral oxygenation during the neonatal period is related to later neurodevelopmental outcomes. In newborn animal models, cerebral oxygenation of <55% for >2.5 hours predicted neurologic injury.⁶  In preterm infants, cerebral oxygenation <50% on the first day of life (median duration 1 minute; range 1–50 minutes) and during the first 2 weeks of life (median duration 3 minutes; range 1–74 minutes) was negatively associated with total and gross motor skills at 2 to 3 years of age.⁷  Similarly, a cerebral oxygenation of <50% for a total time of 4.5 hours over the first 3 days of life resulted in a 5-point lower developmental quotient at 18 months of age.⁸  Low cerebral oxygenation in preterm infants has been defined as <55% in the recent SafeBoosC II trial,⁹  and this level was recently reported to be associated with unfavorable neurocognitive outcomes in preterm infants born at <32 weeks’ gestational age.¹⁰ 

In the study by Howarth et al,¹  infants who developed NEC had a mean cerebral oxygenation of 56.9% (confidence interval 53.2%–60.6%), compared with those infants without NEC who had a mean cerebral oxygenation of 63.9% (confidence interval 62.2%–65.5%). Infants were studied for a median of 7.3 weeks (range 1–13 weeks), indicating that the persisting low cerebral oxygenation levels likely would result in significant neurologic impairment, although the infants were not followed-up to confirm this. It will be important to repeat this study in a larger number of infants and to include neurodevelopmental follow-up so that these findings can be confirmed. Despite these limitations, the study by Howarth et al¹  provides new insights into the effects of NEC on the brain. As the authors suggest, continuous cerebral oxygenation monitoring with NIRS has the potential to identify infants at risk for cerebral hypoxia so that interventions can be promptly implemented to minimize the risk of long-term neurologic deficits in these vulnerable infants.

This study adds to the growing body of research in which NIRS has been used in preterm infants to investigate the effects of clinical practices and treatments on cerebral oxygenation. Recent studies have revealed that sleeping preterm infants in the prone position, a practice commonly used in neonatal intensive care settings on the understanding that it improves respiratory function, can have a deleterious effect on cerebral fractional tissue oxygen extraction in the first week of life.¹¹  In other recent studies, researchers have identified that a loading dose of caffeine temporarily reduces cerebral oxygenation and increases cerebral tissue oxygen extraction in preterm infants.¹²  Antenatal magnesium sulfate treatment is routinely administered to reduce the risk of cerebral palsy in preterm infants; however, a recent study was unable to demonstrate any effect on cerebral oxygenation of preterm infants in early postnatal life that could be attributed to antenatal neuroprotective magnesium sulfate treatment.¹³  The aim of the current SafeBoosC III¹⁴  trial is to investigate the benefits and harms of treatment based on NIRS monitoring compared with treatment as usual for extremely preterm infants. Similarly, the aim of the COSGOD phase III trial¹⁵  is to investigate whether it is possible to increase survival without cerebral injury in preterm neonates <32 weeks’ gestation by targeting cerebral tissue oxygen saturation during the immediate transition period after birth (the first 15 minutes) in addition to routinely monitoring arterial oxygen saturation and heart rate. These international studies have the potential to provide the necessary evidence to promote routine use of NIRS monitoring in the delivery suite and neonatal unit, which hopefully will decrease the risk of death or survival with severe brain injury in preterm infants in the future.

NEC

necrotizing enterocolitis

NIRS

near-infrared spectroscopy