Defining a blood glucose treatment threshold for neonatal hypoglycemia

The transition from supported intrauterine to independent extrauterine life at birth is associated with many physiological and metabolic adaptations some of which are potentially harmful for the newly born baby. Reduced blood glucose (hypoglycemia) is the most common metabolic problem during the first 72 hours of life when blood glucose concentration can typically range from 27 to 110 mg/dL (1.5-6.1 mmol/L) in full-term healthy neonates before rising to the normal range for infants, children and adults (60-100 mg/dL (3.3-5.5 mmol/L)) from day 3 of life.

Since this so-called “transitional neonatal hypoglycemia” is often asymptomatic at the time, but associated with long-term risk of permanent brain damage if not treated, national guidelines for the care of neonates recommend screening all at-risk neonates for hypoglycemia during the first few hours after birth. Study has identified four common risk factors for transitional neonatal hypoglycemia: late preterm birth (i.e. gestational age 35-37 weeks); increased birthweight for gestational age; reduced birthweight for gestational age; and maternal diagnosis of diabetes.

Whilst there is broad consensus on which neonates should be screened for hypoglycemia, there remains controversy surrounding the precise blood glucose concentration threshold that should be used to diagnose asymptomatic hypoglycemia and prompt its treatment. Traditionally, a plasma glucose of <47 mg/dL (2.6 mmol/L) has been used as this diagnosis/treatment threshold, but lower thresholds in the range 36-47 mg/dL (2.0-2.6 mmol/L) have been proposed.

This recently published highlighted study from the Netherlands was designed to address the controversy and determine if use of lower diagnostic/treatment threshold (36 mg/dL, 2.0 mmol/L) is inferior to the use of the traditional diagnostic/treatment threshold (47 mg/dL, 2.6 mmol/L) in terms of subsequent neurodevelopment.

The study population comprised 689 healthy newborns delivered at 17 hospitals across the Netherlands between October 2007 and April 2011. Inclusion in the study required that they had one or other of the previously mentioned risk factors for neonatal hypoglycemia, so that although in all other respects healthy, they required routine hypoglycemia screening and as a result of the screening were found to have moderate hypoglycemia (i.e. plasma glucose in the range 36-46 mg/dL (2.0-2.6 mmol/L)).

Each of the 689 neonates were randomly assigned to one of two treatment groups: lower threshold group (n=348) and traditional threshold group (n=341). Those in the lower threshold group only received treatment if plasma glucose was <37 g/dL and those in the traditional treatment received treatment if plasma glucose was <47 g/dL. The aim was to maintain, with glucose supplements if necessary, plasma glucose at or above the relevant threshold.

The primary outcome was psychomotor development at 18 months as measured by the Bayley Scales of Infant and Toddler Development, third edition. This is widely considered to be the gold standard tool for assessment of infants and toddlers covering the following five developmental aspects: cognition, language, motor, adaptive, and social-emotional. In this study two developmental aspects – cognition and motor – were scored.

Cognitive and motor outcome scores at 18 months were similar between the two groups and a prespecified inferiority limit was not reached. The authors were able to conclude that use of the lower glucose threshold (36 mg/dL, 2.0 mmol/L) is noninferior to the traditional threshold (47 mg/dL, 2.6 mmol/L) in terms of psychomotor development at 18 months.

The study has thus provided more evidence in support of the notion that it is safe to use a low threshold (36 mg/dL, 2.0 mmol/L) or at least no less safe than using a higher threshold of 47 mg/dL, 2.6 mmol/L. Clearly, widespread adoption of this lower threshold would significantly reduce the number of neonates who are diagnosed and treated for asymptomatic transitional hypoglycemia.