Glycemic response to 24 hour fast in normal children and children with ketotic hypoglycemia

Idiopathic Pathological Ketotic Hypoglycemia: Finding the Needle in a Haystack

Sick children often have a decreased appetite and experience vomiting and diarrhea; however, hypoglycemia (plasma glucose concentration ≤50 mg/dL or 2.8 mmol/L) is rare. Ketotic hypoglycemia (KH) is the most common cause of hypoglycemia presenting to an Emergency Department in a previously healthy child between 6 months and 6 years of age. Ketosis and hypoglycemia are now well understood to be normal physiologic responses of young children to prolonged fasting.There is now substantial evidence that the term KH describes a variety of conditions including both the lower end of the normal distribution of fasting tolerance in young children as well as numerous rare disorders that impair fasting adaptation. Recent advances in molecular genetic testing have led to the discovery of these rare disorders. Idiopathic pathological KH is a diagnosis of exclusion that describes rare children who have abnormally limited fasting tolerance, experience recurrent episodes of KH, or develop symptoms of hypoglycemia despite elevated ketone levels, and in whom an explanation cannot be found despite extensive investigation. This review provides an approach to distinguishing between physiological KH and pathological KH and includes recommendations for management.

Starvation Ketosis in the Pediatric Emergency Department: A Prospective Estimate of Prevalence

OBJECTIVES

Starvation ketosis may occur in children during intercurrent illnesses due to metabolic adaptation to fasting, resulting in significant ketonemia and sometimes ketoacidosis. Also known as accelerated starvation, common symptoms are vomiting, lethargy, and seizures. Previous studies found the prevalence of ketotic hypoglycemia to be 4 per 100,000 presentations to the emergency department (ED). We hypothesized that the prevalence had been underestimated due to the retrospective nature and restricted definitions of previous studies. Our aim was to determine if a prospective study would confirm a higher prevalence of starvation ketosis in pediatric patients presenting to the ED.

METHODS

A prospective observational study of consecutive patients was performed in an urban pediatric ED utilizing point-of-care testing of capillary blood glucose and ketones in symptomatic children.

RESULTS

This study found the prevalence of clinically significant ketosis (defined as beta-hydroxybutyrate 2.5 mmol/L or more) to be approximately 1800 per 100,000 ED presentations, with more than 170 per 100,000 also being hypoglycemic. Affected patients were 3 months to 9 years of age. Fifty-five percent of the presentations were boys, and 56% were under the 50th centile for weight.

CONCLUSIONS

This prospective observational study using a targeted testing regimen found the prevalence of both starvation ketosis and hypoglycemia to be much greater than previous studies which used retrospective analyses. A simple capillary test for ketones should be considered in unwell children younger than 10 years who present with vomiting or lethargy, as this may identify the need for specific therapy to resolve ketosis.

Clinical and laboratory characteristics and follow up of 62 cases of ketotic hypoglycemia: a retrospective study

Introduction

Idiopathic ketotic hypoglycemia (KH) is the most common cause of hypoglycemia in non-diabetic children ages 0.5–6 years old and typically occurs after a period of poor food intake. There are no large studies looking at the value of common laboratory testing in children presenting with KH or how often other diagnoses are made.

Objectives

To examine the clinical presentations and the value of laboratory testing done in a cohort of children clinically diagnosed with KH.

Methods

Billing records were searched from 2008 to 2017 for patients seen by the endocrine service for “hypoglycemia, not otherwise specified”. Charts were reviewed to determine age, sex, presenting symptoms and testing ordered at the time of the consult. Through chart reviews after the event and parent phone calls, diagnoses other than idiopathic KH were searched.

Results

Of 150 charts reviewed, 62 had sufficient information to make a clinical diagnosis of KH (32 males 30 females, mean age 2.9 years). Most had a history of gastrointestinal illness or prolonged fasting but 29% had no apparent precipitating event. Laboratory testing was quite variable and while low serum CO₂ was seen in over half, no routine hormone testing, metabolic testing or supervised fasting resulted in a relevant diagnosis. We identified 4 out of 62 (6.5%) with relevant diagnoses which explained KH, including one child with failure to thrive found to have growth hormone (GH) deficiency and 3 by genetic testing, including one case of GSD type 9α, but all had atypical presentations.

Conclusions

In the typical setting of a healthy 0.5–6 year-old child with an uncomplicated episode of KH following poor food intake and a normal exam including growth, hormonal and metabolic testing can safely be deferred. However, frequent recurrences and atypical features should prompt further investigation.

Trial registration

Not needed for a retrospective chart review study.

How to use a controlled fast to investigate hypoglycaemia

Controlled fasts can play a valuable role in the diagnosis and management of hypoglycaemia in paediatric clinical practice, but are no substitute for the collecting of appropriate critical samples at the time of hypoglycaemia for metabolic and endocrine studies. Fatty acid oxidation defects, hyperinsulinism and adrenal insufficiency should always be excluded prior to organising controlled fasts. Controlled fasts are safe if conducted in an experienced setting with strict protocols in place. Failure to adhere to protocol can defeat the purpose of the study and can potentially be dangerous. Proper planning in conjunction with the laboratory and close supervision by staff experienced in controlled fasts is crucial to ensure the best quality information is yielded from these procedures.

Fasting adaptation in idiopathic ketotic hypoglycemia: a mismatch between glucose production and demand

In order to study the pathophysiology of hypoglycemia in idiopathic ketotic hypoglycemia (KH), glucose kinetics during fasting in patients with KH were determined. A fasting test was performed in 12 children with previously documented KH. Besides determination of glucoregulatory hormones, plasma ketones, FFA and alanine, the rates of endogenous glucose production (EGP), glucose uptake, gluconeogenesis (GNG) and glycogenolysis (GGL) were quantified using the [6,6-(2)H(2)] glucose isotope dilution method and the deuterated water method. The five youngest subjects (age 2.5-3.9 years) became hypoglycemic (glucose <3.0 mmol/l) during the test. Mean differences in glucose kinetics between overnight fasting and the end of the test in the hypoglycemic vs. the normoglycemic subjects were: EGP: -31.9% vs. -17.9% (p = 0.007), GGL: -66.2% vs. -50.8% (p = 0.465) and GNG 6.8% vs. 19.5% (p = 0.465). Plasma alanine levels were significantly lower (p = 0.028) at the end of the test in the hypoglycemic subjects. Plasma ketones and FFA levels were in the normal range for fasting duration in all subjects. We conclude that hypoglycemia in KH is caused by the inability to sustain an adequate EGP during fasting in view of the higher glucose requirement in young children. The decrease in GGL is not accompanied by a significant increase in GNG, possibly because of a limitation in the supply of alanine. Our results support the hypothesis that KH represents the lower tail of the Gaussian distribution of fasting tolerance in children.

Postprocedural effects of gastrointestinal endoscopy performed as a day case procedure in children: implications for patient and family education

A prospective design that included a survey tool, nursing care records, and telephone interview was used to determine postprocedural effects experienced by children and families following gastrointestinal endoscopy performed as a day procedure. One hundred twenty-one children attending a pediatric gastroenterology unit for endoscopy under general anesthesia participated in the study. Physical symptoms, day care/school attendance, behavioral issues, and economic factors in the 72 hours post procedure were identified. Over half the children (n = 69, 57%) experienced pain in the hospital post procedure. Pain was reported by 73 children (60%) at home on the day of the procedure, by 55 children (45%) on Day 1 post procedure, and by 37 children (31%) on Day 2 post procedure. The throat was the most common site of pain. Nausea or vomiting was experienced by 37 children (31%) at some time following their procedure but was not associated with procedure type, age, or fasting time. Over half the children (n = 53, 51%) who usually attended day care or school did not attend the day following their procedure. Twenty-four parents (40%) who would normally have worked on the day after the procedure did not attend employment. These findings have been used to improve the preprocedural information and discharge management of patients treated in a pediatric gastroenterology ambulatory setting.

Glucose and leucine kinetics in idiopathic ketotic hypoglycaemia

AIMS

To investigate glucose and leucine kinetics in association with metabolic and endocrine investigations in children with ketotic hypoglycaemia (KH) in order to elucidate the underlying pathophysiology.

METHODS

Prospective interventional study using stable isotope tracer in nine children (mean age 4.23 years, range 0.9-9.8 years; seven males) with KH and 11 controls (mean age 4.57 years, range 0.16-12.3 years; four males).

RESULTS

Plasma insulin levels were significantly lower in KH compared to subjects in the non-KH group. Plasma ketone body levels were significantly higher in KH than in non-KH. Basal metabolic rate was significantly higher in subjects with KH (45.48+/-7.41 v 31.81+/-6.72 kcal/kg/day) but the respiratory quotients were similar in both groups (KH v non-KH, 0.84+/-0.05 v 0.8+/-0.04. Leucine oxidation rates were significantly lower in children with KH (12.25+/-6.25 v 31.96+/-8.59 micromol/kg/h). Hepatic glucose production rates were also significantly lower in KH (3.84+/-0.46 v 6.6+/-0.59 mg/kg/min).

CONCLUSIONS

KH is caused by a failure to sustain hepatic glucose production rather than by increased glucose oxidation rates. Energy demand is significantly increased, whereas leucine oxidation is reduced.

Investigation of the mechanisms of therapy-related hypoglycaemia in children with acute lymphoblastic leukaemia

AIM

To determine the mechanisms of fasting hypoglycaemia occurring during maintenance therapy (MT) for childhood acute lymphoblastic leukaemia (ALL).

METHODS

Thirty-five children and adolescents with ALL, aged 2.4-17.4 y, were fasted for up to 16 h during MT. Nineteen of the children developed hypoglycaemia after 11 to 16 h of fasting. Blood samples for determination of metabolic changes were taken on completion of fasting. Nineteen patients underwent a glucagon stimulation test after 4 to 16 h of fasting during MT. Erythrocyte concentrations of the metabolites of methotrexate (E-MTX) and 6-mercaptopurine (E-TGN) were measured at the time of fasting. Fifteen out of 19 patients who became hypoglycaemic were re-studied 3 to 4 mo after cessation of therapy.

RESULTS

In the hypoglycaemia group, plasma levels of gluconeogenic amino acids alanine and glutamine were lower (medians 117 vs 190 micromol L(-1), p = 0.009, and medians 396 vs 448 micromol L(-1), p = 0.031, respectively) than in the normoglycaemia group. Serum levels of free carnitine were lower (medians 20.3 vs 29.8 micromol L(-1), p = 0.027), free fatty acids higher (medians 3.09 vs 1.23 mmol L(-1), p < 0.001) and marked dicarboxylic aciduria was more common in the patients with hypoglycaemia (in 14/16 vs in 2/14, p < 0.001). Impaired responses to glucagon stimulation occurred in 36% (4/11) in the hypoglycaemia group and in 12.5% (1/8) in the normoglycaemia group (p = 0.243). No significant differences were detected in E-MTX and E-TGN between the groups. Most of the metabolic abnormalities returned to normal after cessation of chemotherapy.

CONCLUSIONS

Low levels of gluconeogenic amino acids, especially of alanine, are associated with hypoglycaemia. Reduced hepatic glycogen stores may also be involved in the aetiology.

Sweet dreams?--nocturnal hypoglycemia in children with type 1 diabetes

Hypoglycemia is an inevitable consequence of the treatment of type 1 diabetes in childhood. Nocturnal hypoglycemia is often considered as merely the submerged part of this serious complication yet there are reasons to believe that hypoglycemia occurring during sleep may be different in physiological terms. Glucose homeostasis during fasting, delayed effects of exercise and alterations in sleep physiology, itself, may not only affect the risk of nocturnal hypoglycemia but may influence the ability to correct glucose concentration as it falls, leading to episodes of hypoglycemia which are both profound and prolonged. The etiology and potential repercussions are incompletely understood and the most appropriate defense remains unclear. A greater understanding of this enigmatic phenomenon is essential before appropriate methods for hypoglycemia avoidance can be developed.

Hypoglycemia presenting as acute respiratory failure in an infant

Hypoglycemia, a common metabolic abnormality seen in the pediatric population, is most often easily diagnosed and rapidly treated with satisfactory outcome. If not recognized and treated in prompt fashion, however, hypoglycemia may cause irreversible central nervous system injury or expose the patient to unnecessary procedures; it rarely results in death. The classic emergency department (ED) presentation of hypoglycemia, the diabetes mellitus patient using hypoglycemic therapy, is frequently encountered and adequately managed with excellent outcome. Alternatively, the patient may present to the ED in a fashion suggestive of a situation other than hypoglycemia. For example, the patient with an altered sensorium following a traumatic event, with a focal neurologic finding, or with bradycardia--all situations in which hypoglycemia is the causative issue--may not be immediately recognized as such a metabolic problem. This report presents a case of a 9-month-old boy who presented with acute respiratory failure and mental status change; the initial ED impression was one of pneumonia with sepsis. Further evaluation uncovered the actual reason for the mental status change and respiratory insufficiency: hypoglycemia was noted on laboratory analysis; no clinical evidence of pneumonia was found after thorough ED evaluation and a prolonged hospital stay. His mental status improved and his respiratory insufficiency resolved after glucose therapy. No other explanation for the respiratory failure was found during the hospital admission. It is imperative that the emergency physician consider hypoglycemia in all patients with any degree of mental status abnormality, even when the findings seem to be explained initially by other etiologies.

Hypoglycemia in childhood: a clinical approach

A child with hypoglycemia presents very frequently an urgent diagnostic and therapeutic challenge. The main causes of hypoglycemia and the clinical approach and laboratory investigations are reviewed underlining that the relationship between the hypoglycemic event and the last meal can aid in diagnosis, in particular for metabolic diseases. Only a correct and precise diagnosis can allow adequate and appropriate treatment and prevention of future hypoglycemic events.

Hypoglycemia in infancy and childhood

Hypoglycemia is a medical emergency in the newborn nursery. The requirements for glucose homeostasis, definition of hypoglycemia, classification of hypoglycemia (increased glucose utilization versus decreased glucose production), diagnostic evaluation and management will be presented. Clinical cases will be used to illustrate important disorders.

Evaluation of fasts for investigating hypoglycaemia or suspected metabolic disease

AIM

To assess the value and safety of fasts for investigating hypoglycaemia or suspected metabolic disease.

STUDY DESIGN

Review of all diagnostic fasts performed over a 2.5 year period.

SETTING

The neonatal intensive care unit and programmed investigation unit at a tertiary referral centre for endocrinology and metabolic disease.

RESULTS

138 diagnostic fasts were performed during the study period. Hypoglycaemia (< 2.6 mmol/l) occurred in 54 cases but in only four did the blood glucose concentration fall below 1.5 mmol/l. One patient became unwell as a result of a fast, but prompt treatment averted any sequelae. Specific endocrine or metabolic defects were identified in 30 cases, the most common being hyperinsulinism and beta-oxidation defects.

CONCLUSIONS

Fasting is safe if conducted on an experienced unit with appropriate guidelines. It continues to provide useful information for diagnosis and management, particularly in cases of hyperinsulinism. Diagnoses should, however, be established by lower risk procedures whenever possible. Thus specimens for metabolic and endocrine studies should be obtained during the presenting episode and blood acylcarnitine species should be analysed prior to fasting.

Ketotic hypoglycemia: a case report

Ketotic hypoglycemia (KH) is recognized in the pediatric literature as the most common cause of childhood hypoglycemia. Since the first manifestation of the syndrome is classically that of new onset seizures, initial emergency department presentation is likely. We describe a representative case and emphasize the importance of routine serum glucose and urine ketone measurements in young children with new onset seizures.

An unusual case of symptomatic hypoglycaemia in a child

An unusual case of pre-operative, symptomatic hypoglycaemia in an otherwise fit, well developed 6-year-old boy is reported, and highlights the unpredictable effect of pre-operative starvation on children admitted on the day of surgery. The presentation and incidence of hypoglycaemia before operation in children is also reviewed.

CSF concentration and CSF/blood ratio of fuel related components in children after prolonged fasting

In order to obtain information about blood and cerebrospinal fluid (CSF) concentrations, and CSF/blood ratio data of fuel related substrates at the end of a prolonged fast in children, we have selected biochemical data from fasting test procedures in 11 control children aged 3-5 yr, fasted 24 h, and 58 control children aged 6-15 yr, fasted 40 h. There was a good correlation between blood and CSF concentrations for glucose, acetoacetate and beta-hydroxybutyrate. The relation with age and sex has been analyzed only in the older children. CSF and blood values for glucose are positively related with age, and both ketones are negatively related with age. Lactate, pyruvate and alanine concentrations in blood and CSF are not related with age, except for CSF pyruvate. With respect to the CSF/blood ratio for the above mentioned components, only the value for acetoacetate is sex and age related. The calculated median caloric values for the sum of glucose, lactate, pyruvate and ketones in CSF are independent of age at the end of a 40-h fast. The diminished glucose contribution on the CSF caloric homeostasis in younger children is fully compensated by the ketone bodies.

The concentration of blood components related to fuel metabolism during prolonged fasting in children

In order to study the relationship between sex, age and glucose, and the concentrations of various fuel related blood substrates in children during prolonged fasting, we have selected data of fasting procedures in 13 control children aged 3-5 yr, fasted 24 h, and 58 control children aged 6-15 yr, fasted 40 h. Compared to the blood results after overnight fast, glucose is decreased, and lactate, pyruvate, ketones and non-esterified fatty acids (NEFA's) are all clearly increased at the end of fast. The concentrations of alanine and triglycerides remain unchanged. The relation with sex, age and glucose has only been analyzed in the older children group. A sex-dependency is indicated for the ketones. Ketones are negatively related with age. NEFA's pyruvate and alanine are not age-related, whereas glucose, lactate and triglycerides are moderately age-dependent. Ketones are negatively related with glucose, whereas pyruvate, NEFA's and triglycerides are not glucose-related. Lactate and alanine are weakly related to glucose. The data demonstrate diminished glucose homeostasis and increased ketogenesis in younger children compared to older ones during prolonged fasting.

Reference values of blood components related to fuel metabolism in children after an overnight fast

The interrelation between blood components, involved in fuel metabolism, and age, sex and glucose was studied in 72 control children (26 girls and 46 boys, aged between 3 and 15 yr) after an overnight fast (14 h). Glucose, lactate, pyruvate, triglycerides and cholesterol are age-independent. Alanine is positively correlated, whereas beta-hydroxybutyrate, acetoacetate and NEFA's are negatively correlated with age. With respect to blood sugar, acetoacetate, NEFA's and cholesterol are glucose-independent. Lactate, pyruvate, alanine and triglycerides are positively correlated with glucose, and beta-hydroxybutyrate--and total ketone bodies--are negatively correlated with glucose. Except for triglycerides, no differences in the concentrations of the above mentioned blood substrates are seen between boys and girls. These data demonstrate that after an overnight fast lipolysis and ketogenesis already are active in young children, probably related to inadequate gluconeogenesis and improvement of carbohydrate regulation with age.

Plasma glucose levels in the peri-operative period in children

Peri-operative plasma glucose levels were studied in one hundred children under 5 years of age and under 20 kg. All underwent procedures of less than 30 minutes. Of 32 cases under the third percentile for weight, ten developed low plasma glucose during the pre-operative and/or postoperative phase. This occurred despite receiving 5% dextrose by mouth 4 hours before induction. Even in children of normal weight there was a low incidence of peri-operative hypoglycaemia. Routine intravenous glucose-containing fluids are thus recommended during the peri-operative period.

Chronic fructose intoxication after infancy in children with hereditary fructose intolerance. A cause of growth retardation

In two unrelated boys, 5.3 and 3.8 years of age with hereditary fructose intolerance, apparently isolated growth retardation (-2.71 S.D. and -2.40 S.D.) occurred after infancy, even though acute symptomatic fructose intoxication was prevented by restriction of dietary fructose. When more stringent restriction of dietary fructose was instituted (approximately 40 mg per kilogram of body weight per day), growth velocity increased from the 25th to the 97th percentile in one child and from well below the 3d to above the 75th percentile in the other. When restriction of dietary fructose was experimentally relaxed (from 10 to 250 mg per kilogram per day), neither boy had symptoms, hypoglycemia, or evidence of hepatic or renal dysfunction, but both had sustained hyperuricemia and hyperuricosuria and increases in the plasma concentration and urinary excretion of magnesium. We conclude that in patients with hereditary fructose intolerance, clinically important chronic fructose intoxication can occur after infancy without causing symptoms of acute fructose intoxication and can be expressed as an apparently isolated, reversible retardation of somatic growth with a continuing disorder of adenine nucleotide metabolism, characterized in part by recurrently increased rates of degradation of adenine nucleotides.

Pre-operative starvation in children. The role of alanine in blood glucose homeostasis

Ninety-two children aged between 8 months and 8 years were fasted overnight prior to surgery. Plasma glucose and alanine concentrations were measured after the induction of anaesthesia. Of 30 children under four years, 23% had plasma glucose concentrations less than 3.3 mmol/litre. Underweight children were particularly at risk. The mean plasma alanine was low (0.17 mmol/litre) indicative of a shortfall in this essential substrate for gluconeogenesis. This probably reflects the greater obligatory glucose need of young children relative to their body protein mass.

Hypoalaninemia and ketotic hypoglycemia: cause or consequence?

A shortage of alanine for gluconeogenesis is believed responsible for various forms of hypoglycemia and in particular ketotic hypoglycemia (KH). We examined the glucose-alanine relationship in two groups of fasting children, 18 with KH and 44 controls. Glucose levels declined in both groups but significantly more in KH; to 1.98 +/- 0.20 versus 3.26 +/-0.13 mM (mean +/- SEM; P less than 0.001). Alanine also fell in both groups, the concentrations correlating significantly with the concomitant glucose levels (KH: r = 0.64, P less than 0.001, and controls: r = 0.50, P less than 0.001). The relationship of alanine to glucose gave virtually identical regression equations, y = 0.054x + 0.063 for KH and y = 0.054x + 0.050 for controls. The differences in alanine levels between the two groups were too small to account for the greater decline in glucose in KH. The results indicate that hypoalaninemia rather than causing hypoglycemia results from it.

Diet induced ketosis in epilepsy and anaesthesia. Metabolic changes in three patients on a ketogenic diet

Ketogenic diets have a high fat and low carbohydrate and low protein content to induce ketosis which is monitored by daily urine testing. Lapses in diet are frequently associated with loss of anticonvulsant control. There has, as yet, been no report of children maintained on a ketogenic diet subject to anaesthesia and surgery: this paper records the changes in metabolic variables observed in three patients undergoing simple inhalation anaesthetics for minor surgery.

Competitive inhibition of rat brain hexokinase by 2-deoxyglucose, glucosamine, and metrizamide

The effects of metrizamide on the kinetics of rat brain hexokinase were compared in vitro with those of 2-deoxyglucose and glucosamine. Although metrizamide, 2-deoxyglucose, and glucosamine are known to be competitive inhibitors of approximately equal potency for glucose of yeast hexokinase (Ki approximately 0.7 mM for all three), metrizamide is a much weaker competitive inhibitor (Ki about 20 mM) of rat brain hexokinase than either 2-deoxyglucose or glucosamine (Ki about 0.3 mM for both). This indicates a greater active site specificity of rat brain hexokinase than of yeast hexokinase. Rat brain hexokinase activity is enhanced approximately threefold in the presence of 0.05, 0.2, and 0.8 mg/ml bovine serum albumin, while yeast hexokinase is only enhanced by 50% under these conditions. Despite the high Ki value for metrizamide, interference with glucose metabolism may occur whenever metrizamide is present in much higher concentrations than glucose. Myelography in humans is one such situation.

Variation in plasma ketone bodies during a 24-hour fast in normal and in hypoglycemic children: relationship to age

The variations in blood ketone bodies, blood glucose, and insulin were studied in 19 normal and 14 hypoglycemic children, 4 months to 13 years of age, during a 24-hour fast. Except in four patients (two with hyperinsulinism and two with congenital defect in ketogenesis), a significant increase in blood ketone bodies was observed in both controls and patients. A progressive decrease in glucose concentrations was observed up to but not after 20 hours. A highly negative correlation between blood ketone bodies and blood glucose was found, with a large dispersion of blood ketone bodies, especially for those corresponding to the blood glucose between 45 and 65 mg/dl. This dispersion was consistently reduced in a homogenous age group of 4 to 6 years with similar glucose values. There was a positive correlation between age and blood glucose from hour 21 on, and an inverse relationship between age and blood ketone bodies from hour 15 on. The same high inverse relationship between age and blood ketone bodies was again observed when the variable of glucose concentration was factored out, demonstrating that the variation in blood ketone bodies is indeed related to age. These findings need to be taken into account in the interpretation of fasting blood ketone bodies, especially when used as an aid in the diagnosis of the various forms of childhood hypoglycemia, and of hypoketotic states.

Systemic carnitine deficiency--a treatable inherited lipid-storage disease presenting as Reye's syndrome

A 3 1/2-year-old boy presented at three months of age with an acute episode of lethargy, somnolence, hypoglycemia, hepatomegaly, and cardiomegaly, which responded poorly to restoration of the blood sugar level to normal. The absence of ketonuria during subsequent episodes of severe hypoglycemia prompted a search for a defect in fatty acid oxidation. Plasma carnitine (2.0 to 5.0 mumol per liter), muscle carnitine (0.01 to 0.02 mumol per gram, wet weight) and liver carnitine (0.021 to 0.065 mumol per gram, wet weight) were all less than 5 per cent of the normal mean. During a 36-hour fast, ketones were barely detectable. Prolonged treatment with oral carnitine over a six-month period resulted in increased muscle strength, a dramatic reduction in cardiac size, relief of cardiomyopathy, partial repletion of carnitine levels in plasma and muscle, and complete repletion in the liver. Systemic carnitine deficiency is an easily treatable cause of recurrent Reye's-like syndrome. Its diagnosis requires measurement of carnitine levels.

Hypoketosis as a cause of symptoms in childhood hypoglycemia

Children with symptomatic hypoglycemia and asymptomatic hypoglycemia-prone children were shown to differ in the degree of ketosis after a 20 h fast. In the latter children the close negative correlation between ketone body levels and glucose levels yielded a regression line against which the former children's data could be compared. Half of the patients were found to be hypoketotic during hypoglycemia. The significance of this hypoketosis in the symptomatology is discussed. The finding of other abnormal responses to fasting, particularly in the patients' nitrogen metabolism, suggests that all these aberrations have a common cause which may be of hepatic origin.

Hypoglycemia in children

The factors that sustain postabsorptive glucose concentrations have been analyzed and the adverse effects of various hypoglycemic disorders on these factors examined. The role of alanine has been reviewed and the importance of glycerol as a precursor of glucose and of ketones as a fuel substitute for glucose emphasized. Finally, we have suggested that fasting functional hypoglycemia replace ketotic hypoglycemia as a descriptive term and that we relinquish the concept of leucine-sensitive hypoglycemia as a specific entity.

Glycemic response to 24-hour fast in normal children: III. Influence of age

A 24-hour fast was performed in 28 normal children-17 boys and 11 girls, 2 to 17 years of age. After the fast, blood was drawn for blood sugar, plasma growth hormone and cortisol, serum free fatty acids and alanine measurements. Blood sugar values ranged between 30 and 77 mg/dl and were significantly correlated to age (R = 0.68, P less than 0.001). Plasma cortisol (R = 0.73, P less than 0.001), GH (R = 0.57, P = 0.01), and FFA (R = 0.76, P less than 0.001) were negatively correlated to age. Serum alanine fasting values ranged between 10 and 36 micrometer/dl and were significantly correlated to age (R = 0.86, P less than 0.001) and to blood sugar values (R = 0.54, P less than 0.01). These data demonstrate that carbohydrate regulation during fast improves with age in children, correlating with higher levels of gluconeogenic substrates and a lower rate of lipolysis.

Ketotic hypoglycaemia of childhood

A patient with recurrent convulsions in childhood and associated ketotic hypoglycaemia is described. Hypoglycaemic attacks started at the age of 3 years and 4 months and continued until 9. At present (aged 15) the patient is mentally retarded, has epilepsy, high tone deafness and a major behaviour disturbance. Prednisone therapy failed to prevent hypoglycaemic convulsions and eventually irreversible brain damage. Intramuscular glucagon and adrenaline were ineffective in raising the blood glucose during acute hypoglycaemic attacks.

Investigations at 3 years and 7 months and at 14 years showed a persistent and markedly abnormal sensitivity to a small dose of exogenous insulin with severe hypoglycaemia with convulsions, absence of clinical hyperadrenalism during hypoglycaemia, and a metabolic block in gluconeogenesis. The demonstration of a persistent biochemical abnormality of glucose metabolism at the age of 14 strongly suggests that ketotic hypoglycaemia of childhood is not another aspect of nutritional deprivation, as recently suggested (Buist, 1974), but the result of a defect in glucose homeostasis.