New approaches to screening and management of neonatal hypoglycemia based on improved understanding of the molecular mechanism of hypoglycemia

Abstract

For the past 70 years, controversy about hypoglycemia in newborn infants has focused on a numerical "definition of neonatal hypoglycemia", without regard to its mechanism. This ignores the purpose of screening newborns for hypoglycemia, which is to identify those with pathological forms of hypoglycemia and to prevent hypoglycemic brain injury. Recent clinical and basic research indicates that the three major forms of neonatal hypoglycemia are caused by hyperinsulinism (recognizing also that other rare hormonal or metabolic conditions may also present during this time frame). These include transitional hypoglycemia, which affects all normal newborns in the first few days after birth; perinatal stress-induced hypoglycemia in high-risk newborns, which afflicts ∼1 in 1,200 newborns; and genetic forms of congenital hyperinsulinism which afflict ∼1 in 10,000-40,000 newborns. (1) Transitional hyperinsulinism in normal newborns reflects persistence of the low glucose threshold for insulin secretion during fetal life into the first few postnatal days. Recent data indicate that the underlying mechanism is decreased trafficking of ATP-sensitive potassium channels to the beta-cell plasma membrane, likely a result of the hypoxemic state of fetal life. (2) Perinatal stress-induced hyperinsulinism in high-risk infants appears to reflect an exaggeration of this normal low fetal glucose threshold for insulin release due to more severe and prolonged exposure to perinatal hypoxemia. (3) Genetic hyperinsulinism, in contrast, reflects permanent genetic defects in various steps controlling beta-cell insulin release, such as inactivating mutations of the K _ATP-channel genes. The purpose of this report is to review our current knowledge of these three major forms of neonatal hyperinsulinism as a foundation for the diagnosis and management of hypoglycemia in newborn infants. This includes selection of appropriate interventions based on underlying disease mechanism; combined monitoring of both plasma glucose and ketone levels to improve screening for infants with persistent forms of hypoglycemia; and ultimately to ensure that infants at risk of persistent hyperinsulinemic hypoglycemia are recognized prior to discharge from the nursery.

Keywords: brain damage; glucose; insulin; ketones; newborns.

Figure 1

Median plasma glucose and BOHB in normal breastfed newborns. Note the two phases of “neonatal hypoglycemia”: (1) Birth—24 h: Hypoglycemia with hypo-ketonemia due to transient Hyperinsulinism; (2) 36–72 h in breastfed babies: Milder hypoglycemia with hyper-ketonemia = “Fasting” (4).

Figure 2

Neonatal hypoglycemia in normal AGA newborns and high-risk SGA infants. Mean plasma glucose prior to birth is similar to maternal level and drops to range of 55–70 mg/dl during first 12–18 h before rising into normal extra-uterine glucose range between 24 and 48 h. Mean glucose drops lower in high-risk neonates (SGA) and remains below normal range for several days up to a few weeks. Similar prolonged neonatal hypoglycemia occurs in other high-risk neonates due to birth asphyxia, maternal hypertension/toxemia, etc. Redrawn from (14).

Figure 3

Differential diagnosis of hypoglycemia based on plasma metabolic fuel responses. Note that Hyperinsulinism is associated with suppression of ketones and FFA and retention of a large glycemic response to injection of glucagon. Abbreviations: FFA (free fatty acids), GH/Cortisol def (growth hormone and/or cortisol deficiency).

Figure 4

Most common genetic defects causing Genetic Hyperinsulinism. Outlined are the pathways stimulating beta-cell insulin release by glucose and amino acids. Oxidation of these fuels in mitochondria increases ATP/ADP ratio, leading to inhibition of K⁺ efflux via plasma membrane ATP-sensitive KATP-channel, membrane depolarization, Ca⁺⁺ influx, and release of insulin from storage granules. Tolbutamide stimulates insulin release by closing K_ATP-channels; diazoxide inhibits insulin release by opening K_ATP-channels. Genetic Hyperinsulinism can be caused by inactivating mutations of K_ATP-channel subunits (SUR1 or Kir6.2, encoded by ABCC8 and KCNJ11) and by activating mutations of glucokinase (GCK) or glutamate dehydrogenase (GDH, encoded by GLUD1); inactivating mutations of HNF1A or HNF4A cause hyperinsulinism by decreasing gene expression of K_ATP-channel subunits.

Figure 5

Developmental shift in glucose threshold for insulin release in fetal and early neonatal islets. Threshold for glucose stimulated insulin release is lower in fetal and neonatal islets compared to islets from older neonates and adults. Freshly isolated rat islets were perifused with a step-wise ramp stimulation by glucose from 3 mM to 25 mM, followed by maximal insulin release with KCl. Glucose thresholds were defined by the lowest glucose concentration stimulating insulin release greater than baseline glucose-free perifusion. The inset compares the time course of changes in mean plasma glucose concentration and mean glucose threshold for islet insulin release.

Figure 6

Signals controlling K_ATP-channel trafficking from Golgi to plasma membrane. K_ATP channels traffic to the plasma membrane in response to activation of AMP-kinase by the calcium/calmodulin protein kinase 2, (CaMKKß). Abbreviations: PHPT1, protein histidine phosphatase 1; PI3K, phosphatidyl-inositol 3 kinase; TRPC4, transient receptor potential cation channel subfamily C member 4; AMPK, AMP-activated protein kinase; KATP, ATP-sensitive potassium channel.

Figure 7

Hypoxia lowers glucose threshold for insulin release by decreasing K_ATP-channel trafficking. (A) Hypoxia induced by culture of isolated islets for 2 days lowers the glucose threshold for glucose stimulated insulin secretion (GSIS). (B) Islet culture for 2 days reduces K_ATP-channel trafficking compared to P14 islets without culture, similar to P3 islets without culture. (C). Exposure to hypoxia (FiO2 10%) from embryonic day E18 until postnatal day P6 reduces glucose threshold for GSIS. (D) Adaptaquin (AQ) treatment to mimick hypoxia by stabilizing Hypoxia Inducible Factor (HIF) lowers glucose threshold for insulin release in P11 islets.