Physiology

Glucose is the body’s main fuel and the brain is the principal consumer of circulating glucose. The brain can neither synthesize glucose nor store more than a few minutes’ supply as glycogen and depends on a continuous supply of glucose. Blood glucose concentrations are normally maintained within a relatively narrow range. Recurrent or persistent hypoglycaemia during the period of rapid brain growth and differentiation in infancy can cause permanent neurological sequelae, psychomotor retardation, and seizures. Prevention of hypoglycaemia and its prompt recognition, accurate diagnosis, and vigorous treatment are essential to prevent its devastating cerebral consequences.

Hypoglycaemia most often occurs in the newborn period and is the most common metabolic disorder of the newborn (see Table 2.1 showing neonates at increased risk for hypoglycaemia). The ratio of surface area to body mass of a full‐term newborn baby is approximately three times that of an average adult, necessitating a high rate of energy expenditure to maintain body temperature. Also, the infant brain is large relative to its body mass (and hepatic glycogen stores) and its energy requirement is primarily derived from oxidation of circulating glucose. To meet the high demand for glucose, the rate of glucose production in newborn infants and young children (5–8 mg per kg per minute) is two to three times that of older children and adults (2.3 mg per kg per minute). As a consequence, until feeding is well established, maintenance of glucose homeostasis in the newborn period is more precarious than later in life. During infancy and childhood, hypoglycaemia most often occurs when night‐time feeding is discontinued or when an intercurrent illness interrupts a child’s normal feeding pattern, resulting in a period of relative starvation.

During prolonged fasting, infants and young children cannot sustain the high rate of glucose production. Compared to older children, adolescents, and adults, plasma glucose decreases and plasma ketone concentrations increase more rapidly.

In response to feeding, blood glucose concentrations increase, which stimulates insulin and inhibits glucagon secretion, to maintain plasma glucose concentrations below 7–9 mmol/L [126–162 mg/dL]. Insulin increases glucose uptake in skeletal muscle and adipose tissue by increasing GLUT4 expression, promotes hepatic and skeletal muscle glycogen synthesis, and inhibits glycogenolysis and gluconeogenesis. During fasting, plasma glucose concentrations decrease, insulin secretion is suppressed, and increased secretion of the counter‐regulatory hormones glucagon, adrenaline (epinephrine), noradrenaline (norepinephrine), cortisol, and growth hormone (GH) prevent hypoglycaemia. The endocrine and metabolic responses involved in the transition from the fed to the fasted state are shown in Figure 2.1.

In healthy children, glycogen stores are largely depleted within approximately 12 hours of fasting. Thereafter, gluconeogenesis (from amino acids, glycerol, and lactate) is the sole source of glucose. Lipolysis is activated, increasing the supply of free fatty acids, an alternative energy source for most tissues, which reduces glucose consumption. The brain does not oxidize free fatty acids. However, water‐soluble acetoacetate and ß‐hydroxybutyrate (ketones), produced by hepatic ß‐oxidation of fatty acids, readily cross the blood‐brain barrier and become a major source of energy for the brain during prolonged fasting.

Symptoms and signs of hypoglycaemia

During progressively severe hypoglycaemia, a hierarchy of physiological responses occurs, from sympathoadrenal activation to neuroglycopenia. Symptoms and signs are non‐specific and reflect responses of the central nervous system to glucose deprivation (Table 2.2). Neurogenic (autonomic) symptoms result from the central nervous system‐mediated sympathoadrenal discharge triggered by hypoglycaemia. Neuroglycopenic symptoms are caused by brain dysfunction from impaired energy metabolism, i.e. the direct result of brain glucose deprivation. Symptoms are particularly difficult to identify in newborns, who are at greatest risk of developing hypoglycaemia. Neonates with low blood glucose concentrations are frequently asymptomatic and hypoglycaemia is usually detected by routine screening of at‐risk infants or as an incidental laboratory finding. With suspected hypoglycaemia, it is essential to accurately measure a confirmatory plasma or blood glucose concentration and demonstrate that administration of glucose promptly relieves the symptoms and signs (Whipple’s triad, see section ‘Definition’).

Additional signs of hypoglycaemia in newborns may include apnoea, bradycardia, cyanosis, and hypothermia. In older individuals, typical adrenergic symptoms include palpitations, tremor, and anxiety; cholinergic symptoms include sweating, hunger, and paresthesias. Typical signs include pallor, diaphoresis, tachycardia, and elevated systolic blood pressure.

Definition

The numerical definition of hypoglycaemia continues to be controversial, most especially in newborns, related to uncertainty regarding the definition of euglycaemia and the glucose threshold resulting in brain injury. In the absence of a consensus, several methods are employed to pragmatically define hypoglycaemia, including statistical, metabolic, neurophysiological, and neurodevelopmental (i.e. evidence that a particular blood glucose concentration is associated with long‐term neurodevelopmental sequelae).

Clinical hypoglycaemia is defined as a plasma glucose concentration low enough to cause symptoms and/or signs of impaired brain function. This glucose concentration varies from one patient to another and depends on the availability of alternative fuels and previous blood glucose concentrations. Hypoglycaemia may be difficult to recognize, especially in neonates and young children in whom the symptoms and signs are not specific. Plasma glucose concentrations are also influenced by several factors, including when the sample is obtained relative to the last meal, the source of blood (arterial, capillary, or venous), and whether measurements are performed on whole blood, serum, or plasma (approximately 12–15% higher than whole blood). Furthermore, a single low plasma glucose concentration, especially when measured on a point‐of‐care device (glucose meter), may be a measurement artefact. Confirmation of Whipple’s triad is valuable in older children: (i) symptoms and/or signs consistent with hypoglycaemia; (ii) an accurately measured low plasma glucose concentration; and (iii) relief of symptoms and signs when plasma glucose concentration is restored to normal. Because reliance on Whipple’s triad is not possible in infants and young children, recognition of hypoglycaemia may require confirmation by repeated measurements of plasma glucose concentration and, in certain circumstances, by performing a controlled fasting test.

In adults, the glycaemic threshold for a decrease in brain glucose metabolism is normally <54 mg/dL [3.0 mmol/L]. During acute insulin‐induced hypoglycaemia, autonomic symptoms appear at a plasma glucose of approximately 60 mg/dL [3.3 mmol/L] and impaired brain function occurs at approximately 50 mg/dl [2.8 mmol/L]. In infants and children, functional changes in the central nervous system (brainstem and somatosensory evoked potentials) occur when the venous plasma glucose falls below 47 mg/dL [2.6 mmol/L]. These observations suggest that the physiologic threshold is a plasma glucose in the range 50–60 mg/dL [2.8–3.3 mmol/L]. For clinical care of children, a venous plasma glucose of ≥60 mg/dL [3.3 mmol/L] may be regarded as normoglycaemia and levels <50 mg/dL [2.8 mmol/L] as hypoglycaemia.

When considering management decisions, the above numerical definitions are useful; however, plasma glucose concentrations should also be evaluated in the context of the availability of other metabolic fuels (lactate and ketones). For example, a plasma glucose concentration <50 mg/dL [2.8 mmol/L] is not necessarily harmful if the blood ketone concentration is markedly elevated because infants and children can utilize ketones for cerebral metabolism. Note, however, that free fatty acid and ketone levels are suppressed by insulin, thus accounting for the increased risk of brain injury in infants with hyperinsulinaemic hypoglycaemia.

Aetiology

Hypoglycaemia occurs when the rate of blood glucose utilization exceeds the rate of glucose production and, beyond the newborn period, most commonly presents when an intercurrent illness is associated with reduced calorie consumption caused by anorexia (prolonged fasting), vomiting, and diarrhoea. Hypoglycaemia may also occur after a period of less protracted (e.g. overnight) fasting when blood glucose production is limited by an impaired counter‐regulatory response resulting from cortisol and/or growth hormone deficiency, an inborn error of metabolism affecting glycogenolysis or gluconeogenesis, or when an increased rate of glucose utilization occurs together with decreased hepatic glucose production (hyperinsulinism, malaria, sepsis). The causes of hypoglycaemia in infants and children are shown in Table 2.3.

Preliminary examination and investigation

History and physical examination

When hypoglycaemia occurs in a neonate, the following details should be obtained:

• Pregnancy (duration, maternal health history, gestational or pre‐existing diabetes, hypoglycaemia).
  
• Mode of delivery (breech delivery occurs more often in infants with hypopituitarism).
  
• Birth weight (may be increased in an infant whose mother had gestational diabetes or may be consistent with foetal growth restriction).
  
• Relationship between hypoglycaemia and feeding (hypoglycaemia despite adequate calorie intake is strongly suggestive of hyperinsulinism).

In the older child, clarify the following details:

• Feeding regimen and the maximum length of time that fasting has been tolerated without symptoms suggestive of hypoglycaemia.
  
• Symptoms suggestive of hypopituitarism (e.g. prolonged jaundice in the neonatal period).
  
• Use of oral or inhaled glucocorticoids.
  
• Consider the possibility of accidental ingestion or factitious symptoms following administration to the child of oral glucose‐lowering medications or insulin.
  
• A careful family history: parental consanguinity or unexplained infant death may suggest an inborn error of metabolism; congenital hyperinsulinism may be inherited in an autosomal dominant or recessive fashion.
  
• Hypoglycaemic or other symptoms and signs, such as vomiting, diarrhoea, jaundice, hepatomegaly, or failure to thrive following consumption of lactose or foods containing fructose and sucrose, suggest galactosaemia or a disorder of fructose metabolism, respectively.

Individuals presenting with hypoglycaemia usually do not have abnormalities on clinical examination. Nonetheless, evaluation of growth (obtain all available growth records), weight gain, and a careful physical examination may identify abnormalities that will inform the cause of the child’s hypoglycaemia. Table 2.4 lists the most common signs and associated disorders (see Figure 2.2).

Clinical sign	Possible diagnosis
Optic atrophy	Septo‐optic dysplasia
Cranial midline defects	Growth hormone deficiency
Short stature or decreased height velocity	Growth hormone deficiency
Micropenis	Growth hormone deficiency
Diffusely increased skin or buccal pigmentation	Addison’s disease
Hypotension	Addison’s disease
Underweight or malnutrition	Accelerated starvation
Tall stature or increased height velocity	Hyperinsulinism
Excess weight	Hyperinsulinism
Abnormal ear lobe creases (Figure 2.2)	Beckwith–Wiedemann syndrome
Macroglossia	Beckwith–Wiedemann syndrome
Umbilical hernia	Beckwith–Wiedemann syndrome
Hemihypertrophy	Beckwith–Wiedemann syndrome
Hepatomegaly	Glycogen storage disorder

Investigations

Although hypoglycaemia is a clinical emergency requiring prompt therapy, whenever possible, a blood sample (often referred to as the ‘critical sample’) for investigations should be drawn before administering glucose. Urine should be collected at the earliest opportunity and stored together with blood samples at –20 °C or below. The processing of these samples should be discussed with the laboratory to ensure proper storage of samples. The samples obtained at the time of hypoglycaemia may produce clear biochemical evidence of the cause of the hypoglycaemic episode and thus avoid having to subject the child to further costly and potentially hazardous investigations. Interpretation of the ‘critical’ blood and urine samples obtained at the time of presentation with hypoglycaemia (plasma glucose <50 mg/dL, <2.8 mmol/L), together with a careful physical examination, including assessment of the size of the liver, can frequently determine the specific aetiology of hypoglycaemia (Figure 2.3).

The preliminary investigation of hypoglycaemia is shown in Table 2.5. If blood and urine samples obtained at the time of initial hypoglycaemia do not elucidate its cause (or were not obtained), then additional investigations may be considered. These may include the measurement of intermediary metabolites before and after meals or a prolonged fast performed under controlled conditions, followed by the investigations listed in Table 2.5. Because a prolonged fast may result in severe hypoglycaemia and is potentially dangerous, fasting studies must be performed at a specialist centre. The length of the fast will be determined by the age of the child (Table 2.6). Children undergoing a fasting study should be closely supervised by staff who are experienced in dealing with hypoglycaemia. The timing of the latter stages of the fast when the child is most likely to become hypoglycaemic should be planned to coincide with a time of the day when appropriate staff are available to monitor the patient and assist in any required resuscitation. Intravenous (IV) glucose and hydrocortisone should be available at the bedside. All symptoms should be carefully documented, sample times clearly recorded, and the patient must never be left unsupervised until the fast is complete and the patient is treated with glucose and/or food. Should hypoglycaemia occur during the planned fasting test, the treatment is described below (see section ‘Treatment’).

Sample	Investigation
Blood sample	Glucose
	Bicarbonate or venous pH
	Free fatty acids and β‐hydroxybutyrate
	Lactate
	Insulin and C‐peptide
	Cortisol and ACTH
	Growth hormone
	Liver function tests
	Ammonia
	Urea and electrolytes
	Alanine
	Acylcarnitines; total and free carnitine
Administer glucose
Next urine sample	Ketones
	Glucose and reducing sugars
	Organic acid profile
	Glycine conjugates
	Carnitine derivatives
	Toxicology screena

ACTH, adrenocorticotrophic hormone.

^a Ethanol, salicylates, sulphonylureas.

Age	Duration (hr)
<6 mo	8
6–8 mo	12
8–12 mo	16
1–2 yr	18
2–4 yr	20
4–7 yr	20
>7 yr	24

Investigations may fail to identify a specific cause of hypoglycaemia, which may occur in the context of a normal endocrine counter‐regulatory response, associated with appropriately raised free fatty acids and ketones. Plasma lactate and pyruvate may be normal but alanine low, suggesting a reduced supply of gluconeogenic precursors. This disorder of unknown aetiology is known as ‘accelerated starvation’ or ‘idiopathic ketotic hypoglycaemia’. It is more common in boys than girls and may be associated with a history of foetal growth restriction or a thin physique. Hypoglycaemia may occur after a day of unusually intense physical activity and/or decreased calorie consumption. Episodic hypoglycaemia usually resolves spontaneously by age 7–9 years. There are numerous causes of hypoglycaemia associated with hyperketonaemia (Figure 2.3). Therefore, accelerated starvation or idiopathic ketotic hypoglycaemia should only be diagnosed after other endocrine and metabolic causes of hypoglycaemia have been carefully excluded.

Treatment

Emergency management of acute hypoglycaemia

1 Obtain initial diagnostic blood sample.

2 Give 0.2 grams glucose per kg body weight (2 ml/kg of 10% dextrose) intravenously over 4 to 6 minutes.

3 Start an infusion of IV 10% dextrose at an initial rate of 10 mg per kg body weight per minute (6 ml/kg/h) and adjust according to blood glucose response.

4 If patient remains unconscious, give 25 mg per metre² hydrocortisone intravenously.

5 Collect next urine sample for diagnostic investigations.