Abstract
Hyperinsulinaemic hypoglycaemia in small-for-gestational age infants usually presents in the first two postnatal days. We present a preterm, small-for-gestational age infant who had hyperinsulinaemic hypoglycaemia on day 13 of life. A female twin infant weighing 1390 g was born at 32+6 weeks of gestation. Her glycaemic profile was normal till day 13 of life, after which she was noted to be lethargic and hypoglycaemic and had hyperinsulinism, hypoketonaemia and hypofattyacidaemia, requiring high glucose infusion rate to maintain normoglycaemia, while negative for septic markers and metabolic screen. Initially, there was no response to diazoxide and the genetic studies for ABCC8 and KCNJ11 gene mutations were negative. Delayed response to diazoxide was followed by complete resolution of hypoglycaemia in 5 months. This case highlights the importance of glucose monitoring in small-for-date infants for hypoglycaemia till they achieve full feeds and gain weight. Early recognition and appropriate management of hypoglycaemia in this group of infants have important implications for neurodevelopmental outcome.
Background
Poor glycogen stores, impaired gluconeogenesis and hyperinsulinism are the most common causes of hypoglycaemia in the neonatal and infancy periods. Hyperinsulinaemic hypoglycaemia (HH) occurs as a consequence of unregulated insulin secretion from pancreatic β cells. Three forms of HH have been recognised. The transient form, seen in infants of diabetic mothers, presents soon after birth and responds well to increasing glucose administration until the β-cell insulin secretion is normalised. Persistent HH, more often known as congenital hyperinsulinism (CHI), is increasingly being reported with identifiable genetic mutations, especially in populations with high degree of consanguinity.1 Prolonged HH, typically seen in small-for-gestational age (SGA) infants, usually presents within 48 h of life and most of them respond well to high glucose infusion rate (GIR). Some of these infants need diazoxide and hypoglycaemia resolves over a few weeks to months.2 Rapid diagnosis and appropriate management of infants with HH are essential to prevent brain injury due to hypoglycaemia. We describe an SGA infant with HH, symptomatic on day 13 of life and required diazoxide for 5 months to maintain normoglycaemia. Delayed onset of hypoglycaemia in this group of SGA infants may go unnoticed in special care areas where glucose monitoring is not mandatory while on feeds.
Case presentation
A baby girl weighing 1390 g was born to a 35-year-old primigravida mother at 32+6 week gestation by caesarean section for non-reassuring fetal status. She was the first twin of a dichorionic diamniotic natural conception, of non-consanguineous parents. There was no family history of diabetes or hypoglycaemia. Antenatal ultrasound scan showed growth restriction but Doppler studies were normal. The Apgar scores were 7 and 9 at 1 and 5 min, respectively. She was not dysmorphic but was growth retarded with weight, head circumference and length below the third centile. The second twin was appropriate for gestational age.
She was admitted after birth to the intensive care unit with respiratory distress requiring surfactant and ventilation for 36 h. She attained a feed volume of 120 mL/kg/day by day 12 and the blood sugar levels (BSLs) were in normoglycaemic (3.5–5.5 mmol/L) range since birth. On day 13 of life, she became lethargic and was noted to be hypoglycaemic (1.9–2.1 mmol/L). The twin sister remained stable on feeds.
Investigations
During a hypoglycaemic episode, serum insulin and cortisol levels were 111 mU/L (normally undetectable with hypoglycaemia) and 370 nmol/L (46–389 nmol/L), respectively. Serum ketones were negative and acyl carnitine levels were low. Septic markers were unremarkable. The screening tests for inborn error of metabolism were negative. Serum ammonia and lactate levels were normal. Although transient, there was a rise in glucose levels following glucagon administration. In view of the initial diazoxide unresponsiveness, genetic studies were performed at Exeter, UK for the baby, the twin sister and the parents. Blood samples were negative for ABCC8 and KCNJ11 mutations. Karyotyping was normal, 46 XX.
Treatment
She required increasing GIR (maximum of 22 mg/kg/min) for maintaining normoglycaemia. Oral diazoxide (20 mg/kg/day), hydrochlorothiazide, hydrocortisone and glucagon infusion were given in an attempt to normalise BSL. A surgical opinion for possible pancreatectomy was made in light of the high GIR requirement and initial diazoxide unresponsiveness while continuing with the aggressive medical management, pending genetic study report. However, BSL was controlled by day 22 and her medications except diazoxide were weaned over 3 days.
Outcome and follow-up
Her feed tolerance improved and was graded up to more than half of the fluid requirement as milk by day 30. Thereafter, her stay was uneventful. She was discharged on day 80 of life at a corrected age of 1 month with a weight of 2000 g, on a diazoxide dose of 10 mg/kg/day with home blood glucose monitoring. She was followed up by the neonatal team and was off diazoxide at the corrected age of 5 months. In the last neurodevelopmental follow-up at the age of 23 months, she was assessed to be doing well.
Discussion
HH has an estimated incidence of 1 in 27 000–50 000 live-births.3 HH is defined as a plasma glucose level of less than 3 mmol/L, with an inappropriately elevated plasma insulin concentration in a baby receiving a glucose-infusion rate of more than 8 mg/kg/min, suppressed ketone bodies and free fatty acids and a positive glycaemic response (rise in BSL > 1.5 mmol/L) to parenteral glucagon.1
The aetiology of prolonged HH in SGA infants could be due to a lack of exogenous substrate supply, depletion of hepatic glycogen stores, defective gluconeogenesis, hyperinsulinism and increased sensitivity to insulin or adrenocortical insufficiency.4 5 In this reported case the infant had severe HH lasting for weeks and the mechanism could be a transient alteration in the β-cell regulation of insulin secretion. The underlying possible genetic or epigenetic aetiology of prolonged HH due to growth restriction in SGA infants is poorly understood.6
In infants with HH, diazoxide unresponsiveness is an absolute indication to perform genetic studies as seen in this reported case.7 The aetiology of CHI can be due to two major defects known as channelopathies and metabolopathies. In channelopathies, the commonest genetic cause is recessive mutations in the genes ABCC8 and KCNJ11 encoding the SUR and Kir 6.2 of the pancreatic KATP channels. Once confirmed as congenital HH, identifying focal lesions using 18F-fluoro-L-DOPA positron emission tomography to make the surgical decision is essential.1 7 In metabolopathies, the commonest cause is Hyperinsulinism-Hyperammonaemia syndrome, a dominant missense mutation in glutamate dehydrogenase enzyme encoded by the GLUD1 gene.7 The inborn error of metabolism screen was negative and so the cause was unlikely to be a metabolopathy. Arya et al6 recently published the genotypic characteristics of SGA babies with HH, where no mutations were identified in ABCC8 and KCNJ11 genes, supporting the non-genetic aetiology of prolonged HH in SGA infants.In our case, a non-genetic aetiology is supported by the normal genetic studies, response to diazoxide and spontaneous resolution with time.
Infants with hypoglycaemia most often present during the first 24–48 h of life. Holtrop8 found that the average time for finding low glucose levels in SGA infants was 6.1 h (range 0.8–34.2 h). By contrast, Hoe et al reported a median age of presentation of 13 days (range 2–180 days) in a group of infants with prolonged HH. In neonates, typically the presentation of HH is more severe than in infancy requiring higher concentrations of glucose to maintain the BSL. Symptoms of hypoglycaemia are mostly non-specific such as lethargy, poor feeding, apnoea, seizures and coma.9 Late presentation of hypoglycaemia as seen in our case could be missed as infants at these gestational ages are usually nursed in low dependency neonatal areas while on feeds with infrequent or no BSL monitoring.
Diazoxide, a KATP channel agonist, remains the mainstay of medical treatment in prolonged HH.7 Prolonged HH resolves by 6 months as seen in this reported case as well. Rarely, it may persist up to 1 year of age.2 6 Diazoxide is usually combined with hydrochlorothiazide in neonates to counteract its fluid retention side effects. Hypertrichosis seen in infants on diazoxide usually resolves on discontinuation.7 10 Recent reports of large symptomatic pericardial effusion in infants on diazoxide warrant meticulous cardiovascular monitoring while on treatment.11
Neonates who are treated for HH need a long-term developmental and visual follow-up because of the high risk of neurodevelopmental delay.12
Learning points.
Small-for-gestational age (SGA) infants usually present with hypoglycaemia during the first 2 days of life. Late onset hypoglycaemia can be missed if blood sugar level (BSL) is not frequently monitored when these SGA infants are cared for in low-dependency areas.
Hyperinsulinaemic hypoglycaemia (HH) is increasingly being diagnosed today. Differentiation of transient, prolonged and persistent HH is paramount, as unrecognised low BSL will lead to significant neurodevelopmental morbidity.
If glucose infusion rate is above 8 mg/kg/min, basic workup for HH including serum insulin, ketones and acylcarnitine levels should be performed and a response to glucagon will help with the diagnosis.
Prolonged HH seen in SGA infants are diazoxide responsive. Diazoxide unresponsiveness in a baby with HH warrants genetic studies for common mutations.
There are no clear guidelines on how long to monitor BSL in SGA infants. Our experience and published literatures support to continue BSL monitoring for 2–3 weeks from birth to identify delayed onset of hypoglycaemia in SGA infants.
Acknowledgments
The authors express their gratitude to Dr Khalid Hussain, Molecular Genetics Unit, Institute of Child Health, UCL, London for giving advice while this case was worked up and also to his genetics department team for getting the genetics study performed for the patient and family.
Footnotes
Contributors: CJH wrote the case report. SC wrote the discussion section. PA edited the case report. VSR edited the discussion section.
Competing interests: None.
Patient consent: Obtained.
Provenance and peer review: Not commissioned; externally peer reviewed.
References
- 1.Hussain K. Diagnosis and management of hyperinsulinemic hypoglycaemia of infancy. Horm Res 2008;69:2–13 [DOI] [PubMed] [Google Scholar]
- 2.Hoe FM, Thornton PS, Wanner LA, et al. Clinical features and insulin regulation in infants with syndrome of prolonged neonatal hyperinsulinism. J Pediatr 2006;148:207–12 [DOI] [PubMed] [Google Scholar]
- 3.Glaser B, Thornton PS, Otonskoski T, et al. Genetics of neonatal hyperinsulinism. Arch Dis Child 2000;82:F79–86 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Collins JE, Leonard JV, Teale D, et al. Hyperinsulinaemic hypoglycaemia in small for date babies. Arch Dis Child 1990;65:1118–20 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Collins JE, Leonard JV. Hyperinsulinism in asphyxiated and small for date infants with hypoglycemia. Lancet 1984;2:311–13 [DOI] [PubMed] [Google Scholar]
- 6.Arya VB, Flanagan SE, Kumaran A, et al. Clinical and molecular characterization of hyperinsulinaemic hypoglycaemia in infants born small-for-gestational age. Arch Dis Child Fetal Neonatal Ed 2013;98:F356––8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Senniappan S, Arya VB, Hussain K. The molecular mechanisms, diagnosis and management of congenital hyperinsulinism. Indian J Endocrinol Metab 2013;17:19–30 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Holtrop PC. The frequency of hypoglycemia in full-term large and small for gestational age newborns. Am J Perinatol 1993;10:150–4 [DOI] [PubMed] [Google Scholar]
- 9.Aynsley-Green A, Hussain K, Hall J, et al. Practical management of hyperinsulinism in infancy. Arch Dis Child Fetal Neonatal Ed 2000;82:98–107 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Arnoux JB, Lonlay PD, Ribeiro MJ, et al. Congenital hyperinsulinism. Early Hum Dev 2010;86:287–94 [DOI] [PubMed] [Google Scholar]
- 11.Avatapalle B, Banerjee I, Malaiya N, et al. Echocardiography monitoring for diazoxide induced pericardial effusion. BMJ Case Rep. Published online: 3 Jul 2012.10.1136/bcr.03.2012.6110 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Kumaran A, Kar S, Kapoor RR, et al. The clinical problem of hyperinsulinemic hypoglycemia and resultant infantile spasms. Pediatrics 2010;126:e1231–6 [DOI] [PubMed] [Google Scholar]