Longitudinal Auxological recovery in a cohort of children with Hyperinsulinaemic Hypoglycaemia

Abstract

Background: Hypoglycaemia due to hyperinsulinism (HI) is the commonest cause of severe, recurrent hypoglycaemia in childhood. Cohort outcomes of HI remain to be described and whilst previous follow up studies have focused on neurodevelopmental outcomes, there is no information available on feeding and auxology.

Aim: We aimed to describe HI outcomes for auxology, medications, feeding and neurodevelopmental in a cohort up to age 5 years.

Method: We reviewed medical records for all patients with confirmed HI over a three-year period in a single centre to derive a longitudinal dataset.

Results: Seventy patients were recruited to the study. Mean weight at birth was - 1.0 standard deviation scores (SDS) for age and sex, while mean height at 3 months was - 1.5 SDS. Both weight and height trended to the population median over the follow up period. Feeding difficulties were noted in 17% of patients at 3 months and this reduced to 3% by 5 years. At age 5 years, 11 patients (15%) had neurodevelopmental delay and of these only one was severe. Resolution of disease was predicted by lower maximum early diazoxide dose (p = 0.007) and being born SGA (p = 0.009).

Conclusion: In a three-year cohort of HI patients followed up for 5 years, in spite of feeding difficulties and carbohydrate loading in early life, auxology parameters are normal in follow up. A lower than expected rate of neurodevelopmental delay could be attributed to prompt early treatment.

Keywords: Congenital Hyperinsulinism; Diazoxide; Height; Hypoglycaemia; Natural history; Neurodevelopment; Outcomes; Weight.

Fig. 1

Graphical representation of median and interquartile range for weight and height values from raw and imputed data. The graphs above demonstrate the reliability of linear interpolation used for data imputation in this study. As can be seen, there are no changes in the trends for weight (a) and height (b) when imputed data (i) were used compared with raw data (ii) alone

Fig. 2

Flowchart of natural history outcomes of HI. Frequency of patients in each category determined by time of presentation, mutation status, transient/persistent and surgical outcomes

Fig. 3

a Box and whisker plot analysis of median weight SDS by age. This demonstrates a lower than average birth weight with a trend towards the population median weight at 24 months. Each box extends from the lower quartile to the upper quartile of the data used with the horizontal line in the box indicating the sample median. The whiskers extend to the smallest and largest observations within 1.5*IQR from the lower and upper quartiles, respectively. Any data points plotted beyond the ends of either whisker (as circles) are regarded as possible outliers. b. Box and whisker plot analysis of median height SDS by age. This demonstrates an initially lower than average height, in keeping with weight, with a trend back to population median height. Each box extends from the lower quartile to the upper quartile of the data used with the horizontal line in the box indicating the sample median. The whiskers extend to the smallest and largest observations within 1.5*IQR from the lower and upper quartiles, respectively. Any data points plotted beyond the ends of either whisker (as circles) are regarded as possible outliers

Fig. 4

Percentage of children with feeding difficulties by age. Demonstration that the percentage of children with HI and feeding difficulties reduced from a maximum of 17% at 3 months of age to only 3% at 48 months of age. Feeding difficulties were classified as requirement for nasogastric or gastrostomy feeding

Fig. 5

Percentage of children with persistent disease. Children either requiring medication or who had undergone surgery were classed as having persistent disease. This demonstrates the steady improvement in spontaneous resolution of patients with HI