Neurocognitive Outcomes at Age 2 Years After Neonatal Hypoglycemia in a Cohort of Participants From the hPOD Randomized Trial

Key Points Question Is neonatal hypoglycemia associated with neurocognitive outcomes in otherwise well late preterm and full-term term neonates born at risk of hypoglycemia? Findings In this cohort study of 1197 participants in a randomized clinical trial, the prevalence of neurosensory impairment at 2 years was significantly higher in those who experienced hypoglycemia than in those who did not (23% vs 18%). The risk was greater in children who experienced severe episodes (28%). Meaning The findings of this study suggest that children who experienced neonatal hypoglycemia are at increased risk of neurosensory impairment at corrected age 2 years, even if screened and treated to maintain blood glucose concentrations greater than or equal to 47 mg/dL.


eTable 1. Maternal and infant characteristics of those included and not included in this analysis
.70 Abbreviations: RD, risk difference; MD, mean difference; CI, confidence interval; RR, risk ratio. An episode of hypoglycemia was defined as ≥1 episode of consecutive blood glucose concentrations of <47 mg/dL. The severity of hypoglycemia was defined as none (all blood glucose concentrations ≥47 mg/dL), mild (≥1 episode of ≥36 and <47 mg/dL), and severe (≥1 episode of <36 mg/dL).
a Models compare relative risks of none vs any episode of hypoglycemia, none vs mild hypoglycemia, and none vs severe hypoglycemia. Adjustments are for study site, primary reason for risk of hypoglycemia, socioeconomic decile at birth and multiple births. b First P value reported in the outcome row is for the overall association between severity of hypoglycemia and outcome. P values below in the any, mild and severe rows are for the comparison between none vs any or none vs mild or none vs severe. c Low performance defined as a score of ≤2, range 0-6, higher scores indicate better performance. a The frequency of hypoglycemia was defined as none (all blood glucose concentrations ≥47 mg/dL), 1-2 episodes of consecutive blood glucose concentrations of <47 mg/dL, and recurrent (≥3 episodes of consecutive blood glucose concentrations of <47 mg/dL).
b Adjusted for study site, primary reason for risk of hypoglycemia, socioeconomic decile at birth and multiple births. c First P value reported in the outcome row is for the overall association between frequency of hypoglycemia and outcome. P values below in 1-2 and ≥3 rows are for the comparison between none vs 1-2 episodes or none vs ≥3 episodes.
d Defined as any of: blindness, hearing impairment requiring aids, moderate or severe cerebral palsy (not walking yet or permanently non-ambulant) or moderate or severe developmental delay. e One child in the hypoglycemia group was assigned a score of 49 for the Bayley-III cognitive, language and motor scales because the child was unable to finish the assessment due to severe delay. Language, cognitive and motor scales, mean 100, SD 15, higher scores indicate better performance.
f Defined as an executive function total score more than 1.5 SD below the mean.
g Total score range 0-24 points, higher scores indicate better performance.
h Low performance defined as a score of ≤2, range 0-6, higher scores indicate better performance.

eTable 4. Interaction between treatment and the association between frequency of hypoglycemia and neurocognitive outcomes at 2 years
No./total (%) ≥3 episodes (N = 63) An episode of hypoglycemia was defined as ≥1 episode of consecutive blood glucose concentrations of <47 mg/dL. Nine children who received treatment but did not have recorded episodes of hypoglycemia were excluded. a Generalized linear models adjusted for study site, primary reason for risk of hypoglycemia, socioeconomic decile at birth and multiple births.
b One child in the hypoglycemia group was assigned a score of 49 for the Bayley-III cognitive, language and motor scales because the child was unable to finish the assessment due to severe delay. c Poor executive function was defined as an executive function total score more than 1.5 SD below the mean. a An episode of hypoglycemia was defined as ≥1 episode of consecutive blood glucose concentrations of <47 mg/dL. b Generalized linear models adjusted for study site, primary reason for risk of hypoglycemia, socioeconomic decile at birth and multiple births.
c One child who became hypoglycaemic was assigned a score of 49 for the Bayley-III cognitive, language and motor scales due to severe delay.
d Poor executive function defined as an executive function total score more than 1.5 SD below the mean. Abbreviations: RD, risk difference; CI, confidence interval; RR, risk ratio; Bayley-III, Bayley-Scales of Infant and Toddler Development, Third Edition (mean=100, SD=15, range=40-160, higher scores indicate better performance); SD, standard deviation.
First language English = excluded 425 children (166 in the hypoglycemia group and 259 in the normoglycemia group).
a An episode of hypoglycemia was defined as ≥1 episode of consecutive blood glucose concentrations of <47 mg/dL. b Generalized linear models adjusted for study site, primary reason for risk of hypoglycemia, socioeconomic decile at birth and multiple births.
c One child who became hypoglycaemic was assigned a score of 49 for the Bayley-III cognitive, language and motor scales due to severe delay.
d Poor executive function defined as an executive function total score more than 1.5 SD below the mean. e Composite score range 0-24 points, higher scores indicate better performance.
© 2022 Edwards T et al. JAMA Network Open. eTable 10. The primary outcome of neurosensory impairment in children who did and did not experience hypoglycemia, excluding children randomized to prophylactic buccal dextrose gel (sensitivity analysis) No./total (%) RD (95% CI)