Executive function and behaviour problems in school-age children born at risk of neonatal hypoglycaemia

Abstract

Aim: To examine the relationship between neonatal hypoglycaemia and specific areas of executive function and behaviour in mid-childhood.

Method: Participants in a prospective cohort study of infants born late preterm or at term at risk of neonatal hypoglycaemia were assessed at 9 to 10 years. We assessed executive function using performance-based (Cambridge Neuropsychological Tests Automated Battery) and questionnaire-based (Behavior Rating Inventory of Executive Function) measures and behaviour problems with the Strengths and Difficulties Questionnaire. Data are reported as adjusted odds ratio (aOR) with 95% confidence intervals, and standardized regression coefficients.

Results: We assessed 480 (230 females, 250 males; mean age 9 years 5 months [SD 4 months, range 8 years 8 months-11 years 0 months]) of 587 eligible children (82%). There were no differences in performance-based executive function between children who did and did not experience neonatal hypoglycaemia (blood glucose <2.6 mmoL/L). However, children who experienced hypoglycaemia, especially if severe or recurrent, were at greater risk of parent-reported metacognition difficulties (aOR 2.37-3.71), parent-reported peer (aOR 1.62-1.89) and teacher-reported conduct (aOR 2.14 for severe hypoglycaemia) problems. Both performance- and questionnaire-based executive functions were associated with behaviour problems.

Interpretation: Neonatal hypoglycaemia may be associated with difficulties in specific aspects of parent-reported executive functions and behaviour problems in mid-childhood.

Figure 1:

Relationships between (a) frequency and (b) severity of neonatal hypoglycaemia, and questionnaire based executive function and behaviour problems at 9 years of age. Odds ratios and 95% confidence intervals (CI) are shown using the normoglycaemia group as the comparator and are adjusted for New Zealand Deprivation Index (NZDep), sex, primary risk factor for neonatal hypoglycaemia and IQ at 4.5 years. Hypoglycaemia is defined as ≥1 hypoglycaemic event, defined as the sum of nonconcurrent hypoglycaemic and interstitial episodes more than 20 minutes apart; a hypoglycaemic episode is defined as ≥1 consecutive blood glucose concentration <2.6 mmol/L and an interstitial episode as Interstitial glucose concentrations <2.6 mmol/L for 10 minutes; mild hypoglycaemia event is defined as mild hypoglycaemic events ≥2.0 to 2.5 mmol/L only; severe hypoglycaemia event defined as ≥1 severe hypoglycaemic event <2.0 mmol/L; clinically undetected hypoglycaemia, defined as ≥1 hypoglycaemic events but no hypoglycaemic episodes.

Figure 2:

Mediation models of the indirect effect of frequency and severity of hypoglycaemic events on behaviour problems through executive function without adjusted for FSIQ at 4.5 years. Measures of the indirect effect include the bootstrapped 95% confidence intervals (CI). B = unstandardised regression coefficient, SE = standard error. Hypoglycaemia defined as ≥1 hypoglycaemic event, defined as the sum of nonconcurrent hypoglycaemic and interstitial episodes more than 20 minutes apart; a hypoglycaemic episode is defined as ≥1 consecutive blood glucose concentration <2.6 mmol/L and an interstitial episode as Interstitial glucose concentrations <2.6 mmol/L for 10 minutes; mild hypoglycaemia event is defined as mild hypoglycaemic events ≥2.0 to 2.5 mmol/L only; severe hypoglycaemia event defined as ≥1 severe hypoglycaemic event <2.0 mmol/L. Adjusted for New Zealand Deprivation Index (NZDep), sex, and primary risk factor for neonatal hypoglycaemia.