Association of neonatal hypoglycemia with academic performance at mid-childhood

Abstract

Importance: Neonatal hypoglycemia is associated with increased risk of poor executive and visual-motor function, but implications for later learning are uncertain. Objective: To test the hypothesis that neonatal hypoglycemia is associated with educational performance at 9-10 years. Design, Setting and Participants: The Children with Hypoglycaemia and their Later Development (CHYLD) Study is a prospective cohort of moderate-to-late preterm and term infants born at risk of hypoglycemia. Blood and masked interstitial sensor glucose concentrations were measured for up to 7 days. Infants with hypoglycemic episodes (blood glucose concentration <47 mg/dL [*0.0555 mmol/L]) were treated to maintain blood glucose ≥47 mg/dL. Infants (N=614) were recruited at Waikato Hospital, New Zealand 2006-2010 and assessed at 9-10 years (N=480) 2016-2020. Exposure: Hypoglycemia was defined as ≥1 hypoglycemic event: sum of non-concurrent hypoglycemic and interstitial episodes (sensor glucose concentration <47 mg/dL for ≥10 minutes) >20 minutes apart. Main Outcomes and Measures: The primary outcome was low educational achievement, defined as performing below or well below the normative curriculum level in standardized tests of reading comprehension or mathematics. There were 47 secondary outcomes related to executive function, visual-motor function, psychosocial adaptation and general health. Results: Of 587 eligible children (230, 48% female), 480 (82%) were assessed at a mean (SD) age of 9.4 (0.3) years. Children who were and were not exposed to neonatal hypoglycemia were not significantly different for rates of low educational achievement (138/304 [47%] vs. 82/176 [48%]; adjusted risk difference -2%, 95% CI 11%, 8%; adjusted relative risk 0.95, 95% CI 0.78, 1.15). Children who were exposed to neonatal hypoglycemia, compared to those not so exposed, were less likely to be rated by teachers as being below or well below the curriculum level for reading (68/281 [24%] vs. 49/157 [31%]; adjusted risk difference -9%, 95% CI 17%, -1%; adjusted relative risk 0.72, 95% CI 0.53, 0.99; P=0.04). Groups were not significantly different for other secondary endpoints. Conclusions and Relevance: Among participants at risk of neonatal hypoglycemia who were screened and treated if needed, exposure to neonatal hypoglycemia compared with no such exposure was not significantly associated with lower educational achievement in mid-childhood.