Keywords

Introduction

Congenital hyperinsulinism (HI) is a common cause of persistent hypoglycemia in infants and children. This condition is characterized by dysregulated insulin secretion, and delayed treatment is associated with severe hypoglycemia with increased risk of seizures, developmental delay, and permanent brain injury (Palladino and Stanley 2011; Meissner et al. 2003; Steinkrauss et al. 2005; Lord et al. 2015). Causal mutations in eleven genes have been described in congenital HI, with the most severe forms of disease seen in individuals with inactivating mutations affecting the ATP-sensitive potassium channel (KATP) (Vajravelu and De Leon 2018). However, approximately 50% of infants with congenital HI do not have an identified genetic cause of disease (Lord et al. 2013; Lord and De Leon 2013).

Available medical treatments for the management of infants with HI are limited. Currently, two medications are routinely used to suppress insulin secretion and ameliorate hypoglycemia in these infants: diazoxide, a KATP channel activator that maintains the channel in the open state, and octreotide, a somatostatin analog (Lord and De Leon 2013). For children with severe HI, medical management can be ineffective in preventing hypoglycemia and pancreatectomy may be required (Lord et al. 2013). Prior to pancreatectomy, these children require a continuous glucose infusion at high concentrations and often at high fluid volumes. Central venous access may be required to deliver high concentrations of intravenous dextrose, which may be associated with increased risk of infection and intestinal ischemia (Hawkes et al. 2016; Barrington 2000). Increased intravenous fluid volumes in neonates can result in fluid overload and cardiac failure, which can also be exacerbated by diazoxide administration (Lord and De Leon 2013). Glucagon is often used to reduce glucose demands and consequently can reduce these risks in infants with congenital HI (Palladino and Stanley 2011; Lord et al. 2013).

Glucagon opposes insulin action (Quesada et al. 2008) and prevents hypoglycemia by stimulating hepatic gluconeogenesis and glycogenolysis, as well as inhibiting glycogen synthesis and hepatic glucose uptake. Glucagon can also stimulate the uptake of amino acids in the liver as well as increase the release of glycerol from adipocytes, both of which can be used for gluconeogenesis [reviewed in (Unger and Cherrington 2012)]. Despite widespread use of intravenous glucagon in congenital HI, the only published reports describing its use have been in neonates with unspecified hypoglycemia (Charsha et al. 2003; Miralles et al. 2002; Carter et al. 1988). There have been concerns raised regarding a possible association between glucagon treatment and thrombocytopenia or hyponatremia (Belik et al. 2001), but thrombocytopenia was not seen in these retrospective studies including 108 infants, and hyponatremia was considered to be related to excess fluid administration (Charsha et al. 2003; Miralles et al. 2002; Carter et al. 1988).

The aim of this study was to determine the effect of intravenous glucagon on the required glucose infusion rate (GIR) in infants with severe HI awaiting surgical management of their disease. As a secondary outcome, we sought to describe the rates of adverse events seen in infants during glucagon infusion.

Methods

A retrospective chart review of infants with HI who were treated with a continuous intravenous glucagon infusion prior to partial or subtotal pancreatectomy between January 2003 and December 2013 at the Children’s Hospital of Philadelphia was conducted. The Institutional Review Board at The Children’s Hospital of Philadelphia approved this study.

All infants with a diagnosis of focal or diffuse HI who subsequently underwent pancreatic surgery were eligible for inclusion in this study if they were under 6 months of age at the time of glucagon treatment, they received glucagon for greater than 24 h, and glucose measurements in the 24 h prior to glucagon treatment were available. All children treated with glucagon, regardless of age and availability of glucose measurements, were included in the analysis of adverse events during glucagon treatment.

The diagnosis of HI was based on biochemical evidence of insulin excess at the time of plasma glucose <50 mg/dL, as previously described (Ferrara et al. 2016). The time of glucagon initiation was extracted from the electronic health record. Capillary glucose concentrations were monitored during glucagon treatment at least every 3 h, using a point-of-care glucose meter [Lifescan SureStep Pro (Johnson & Johnson, PA, USA) prior to 2012 and Nova StatStrip (Novo Nordisk, Bagsvaerd, Denmark) after 2012].

Glucagon Administration

Glucagon was administered intravenously, either through a peripheral or a central intravenous line. For administration, glucagon was diluted in 5% dextrose to a concentration of 42 mcg/mL. The initial dose for all patients was 1 mg in 24 h, independent of body weight. To prevent precipitation, bags of diluted glucagon and infusion tubing were replaced with freshly prepared glucagon every 24 h.

Statistical Analysis

In analyzing the efficacy of glucagon infusion, the primary outcome was the GIR at 24 h before and after starting glucagon infusion. Secondary outcomes were the frequency of hypoglycemia (plasma glucose <70 mg/dL) and hyperglycemia (plasma glucose >140 mg/dL) in the days prior to and during glucagon treatment. The number of hypo- and hyperglycemia events was averaged for up to 4 days before and during glucagon treatment. Non-normally distributed continuous variables including GIR, hypoglycemia frequency, and hyperglycemia frequency before and during treatment were described using medians and interquartile ranges (IQR) and compared using Wilcoxon signed-rank test. Statistical analyses were performed using SPSS 22.0 (IBM, N.Y., USA). Figures were generated using Prism 5.0 (GraphPad Software Inc., California, USA) and Adobe Illustrator 16.0 (Adobe Systems Inc., California, USA).

Results

Eighty-seven children diagnosed with congenital HI were treated with continuous intravenous glucagon infusion prior to pancreatectomy at the Children’s Hospital of Philadelphia between January 2003 and December 2013. Of these, 40 were eligible for inclusion in the analysis of glucagon efficacy (7 were over 6 months of age, 31 commenced treatment with glucagon prior to transfer to our center, and 9 were treated with glucagon immediately on admission and did not have 24 h of glucose measurements available prior to treatment with glucagon). All 87 infants were included in the analysis of adverse events during glucagon treatment.

Efficacy of Glucagon Infusion in Children with Congenital HI

Of the 87 infants treated with intravenous glucagon, 40 had data available of glycemic control before and after glucagon treatment. The median (IQR) age at the time of starting glucagon treatment was 29 (23, 56) days of age, and duration of glucagon treatment was 5 (3, 9) days. Demographic and clinical data for these infants are shown in Table 1.

Table 1 Demographics and clinical characteristics
Full size table

Of the 40 eligible children, 34 had a decreased glucose infusion rate (GIR) 24 h after starting glucagon treatment. Four children had an increased GIR, and two had no change. Intravenous glucose was completely stopped in two infants after starting glucagon. Overall, there was a statistically significant reduction in the median (IQR) GIR during the 24 h following initiation of continuous glucagon infusion compared to 24 h before initiation (18.5 (12.9, 22.8) to 11 (6.6, 17.5) mg/kg/min, p < 0.001) (Fig. 1). Starting glucagon was also associated with a reduction in the median (IQR) frequency of hypoglycemia (1.9 (1.3, 2.9) to 0.7 (0.3, 1.2) episodes per day, p < 0.001) (Fig. 2a) without a change in the frequency of hyperglycemia (0.8 (0.3, 1.9) to 1 (0.5, 1.7) episodes per day, p = 0.3) (Fig. 2b).

Fig. 1
figure 1

Glucose infusion rate (GIR, in mg/kg/min) in children with congenital hyperinsulinism 24 h prior to and 24 h after starting continuous glucagon infusion. Box represents 25th and 75th percentiles, and whiskers represent 10th and 90th percentiles

Full size image
Fig. 2
figure 2

Frequency of (a) hypoglycemia episodes (plasma glucose <70 mg/dL) and (b) hyperglycemia episodes (plasma glucose >140 mg/dL) per day in children with congenital hyperinsulinism up to 4 days prior to and 4 days after starting continuous glucagon infusion. Data presented as median (range)

Full size image

Adverse Events

Thirty-five of the 87 children (41%) treated with continuous glucagon infusion experienced adverse events during treatment. Multiple adverse events were reported during treatment. Vomiting (n = 11, 13%), rash (n = 2, 2%), and respiratory distress (n = 15, 19%) are reported side effects of glucagon treatment and were seen in these patients. One patient had thrombocytopenia prior to starting glucagon treatment, but no patient developed thrombocytopenia during glucagon treatment.

Discussion

Continuous intravenous glucagon is effective in the acute management of infants with congenital HI. We have shown that this treatment is associated with a median reduction in GIR of 7.5 mg/kg/min within 24 h of starting treatment and a significant reduction in the frequency of hypoglycemia without an increase in hyperglycemia. This reduction in GIR can reduce the need for hyperosmolar concentrated glucose or high-volume intravenous infusions. Although numerous adverse events were seen during glucagon treatment, it is not possible to determine if these are related to glucagon infusion, and we generally believe that this is a safe treatment in this population.

When compared with previous studies of glucagon infusion in infants with hypoglycemia, we have observed similar reductions in GIR and hypoglycemia during treatment. However, our study is the first to describe this only in infants with severe HI requiring surgical management. These infants are at particularly high risk of unfavorable developmental outcomes due to the severe and persistent hypoglycemia and often require extremely high concentrations of infused glucose to maintain euglycemia. In infants with congenital HI, endogenous glucagon secretion is blunted during hypoglycemia (Hussain et al. 2005), making exogenous replacement a physiologically appropriate treatment.

There have been concerns regarding the association between treatment of neonates with intravenous glucagon and the development of complications including hyponatremia and thrombocytopenia (Charsha et al. 2003; Miralles et al. 2002; Belik et al. 2001). In this study, we have evaluated the safety of glucagon infusion in the largest reported cohort of treated infants with HI. Only one child in this study had thrombocytopenia, but this was present prior to the initiation of glucagon infusion. Hyponatremia was not seen in any infant included in this study. Vomiting, rash, and respiratory distress are reported adverse effects that may be associated with glucagon administration (Food and Drug Administration n.d.) and, while these were seen in some of the patients included in this study, it is difficult to ascertain that this was associated with glucagon treatment.

The large number of infants included in this study provides an opportunity to describe the effect of glucagon administration in this patient population and to observe the safety of this treatment. This study is limited by its retrospective design and absence of a control group for comparison. However, by describing only infants who were scheduled for surgical management of their HI, only infants with severe disease and high risk of hypoglycemia were included. Although it is conceivable that GIR prior to starting glucagon may have been unnecessarily high due to insufficient weaning of intravenous glucose, this is unlikely. The frequency of episodes of hypoglycemic was higher prior to starting glucagon, suggesting that it was not possible to wean the GIR further prior to glucagon treatment.

In conclusion, this study demonstrates that continuous glucagon infusion is an effective treatment in severe congenital HI. Starting a glucagon infusion allows for a reduction in GIR in most infants, which will facilitate a reduction in intravenous volume and/or glucose concentration administered. Furthermore, we have not demonstrated significant safety concerns in using glucagon but recommend close observation as these infants are at risk of complications due to their underlying disease, high fluid requirements, and coexisting medications.