Secrets – Pediatric: Endocrinology

Secrets – Pediatric: Endocrinology

ADRENAL DISORDERS
1. What are the symptoms of adrenal insufficiency?
• Newborns: Nonspecific findings of vomiting, irritability, and poor weight gain; may progress to cardiovascular shock
• Children: Lethargy, easy fatigability, poor weight gain, and vague abdominal complaints; hyperpigmentation (primary insufficiency); symptoms of hypoglycemia (primary or secondary insufficiency); may also exhibit vascular collapse with intercurrent illness
2. What distinguishes primary and secondary adrenal insufficiency?
• Primary: Abnormality of the adrenal gland, low cortisol accompanied by an elevated adrenocorticotropic hormone (ACTH) level; may also have mineralocorticoid deficiency
• Secondary: Hypothalamic or pituitary dysfunction, low cortisol accompanied by an inappropriately normal or low ACTH level; normal mineralocorticoid production; often associated with multiple pituitary deficiencies
3. What is the differential diagnosis of primary adrenal insufficiency?
• Inherited enzymatic defects: Congenital adrenal hyperplasia (multiple enzymatic defects are known), congenital adrenal hypoplasia
• Autoimmune disease: Isolated, autoimmune polyendocrinopathy syndrome (APS) types 1 and 2; type 2 is also known as Schmidt syndrome
• Infectious disease: Tuberculosis, meningococcemia, disseminated fungal infections
• Trauma: Bilateral adrenal hemorrhage
• Adrenal hypoplasia: Due to inherited defects in the adrenal ACTH receptors
• Iatrogenic: Use of exogenous steroids
4. What are the most common causes of secondary adrenal insufficiency?
Secondary causes can include failure of the hypothalamic and/or pituitary gland axis to develop in the embryonic stage or disruption of the axis as a result of tumor, central nervous system (CNS) trauma, irradiation, infection, or surgery. Prolonged treatment with exogenous glucocorticoids will also suppress the hypothalamic and pituitary parts of the axis.
5. What clinical clues suggest that adrenal insufficiency is a primary rather than a secondary problem?
• Primary adrenal insufficiency: ACTH levels will rise as a result of disruption of the hormonal feedback loop, and these elevated levels often cause hyperpigmentation. Primary deficiency commonly leads to hyponatremia and hyperkalemia. This can present as salt craving or muscle cramping. Mild hypercalcemia may also be found.
• Secondary adrenal insufficiency: ACTH levels are low; therefore, no hyperpigmentation occurs. Furthermore, in secondary insufficiency, function of the zona glomerulosa of the adrenal gland (responsible for aldosterone secretion) remains intact. Therefore, hyperkalemia and volume depletion are distinctly uncommon, but isolated hyponatremia may still occur as a result of decreased capacity to excrete a water load (dilutional hyponatremia). The most important clinical clues come from the history; that is, has the child been exposed to exogenous steroids, or is there a history of CNS insult? Additionally, are there deficiencies of other pituitary hormones?
6. What is the most common form of congenital adrenal hyperplasia (CAH)?
CAH refers to a group of autosomal recessive disorders that result from various enzymatic defects in the biosynthesis of cortisol. Depending on the enzyme involved, the blockade can result in deficiencies and/or excesses in the other steroid pathways (i.e., mineralocorticoids and androgens). 21-Hydroxylase

189

deficiency accounts for more than 90% of cases; the complete (salt-losing, about two-thirds of cases) and partial (simple virilizing) forms occur in about 1 in 12,000 births and have an equal sex distribution. There are substantial differences in prevalence in various racial and ethnic groups. A late-onset or attenuated form (mild deficiency) manifests in adolescent girls with hirsutism and menstrual irregularities.

Zoltan A, Zhou P: Congenital adrenal hyperplasia: diagnosis, evaluation, management, Pediatr REV 30:e49–e57, 2009.

7. In newborns with CAH, why are girls likely to be diagnosed earlier than boys? The most common forms of CAH result in excess androgen production in the fetus; the effects of prenatal androgen excess on the development of the clitoris and labia majora can be easily identified in the newborn period. In boys, androgen excess does not cause any clearly abnormal appearance of the external genitalia. CAH should always be considered in the differential diagnosis of disorders of external sexual development, particularly in infants with a 46,XX karyotype.
8. How do the major steroid preparations vary in potency?
See Table 6-1.

Table 6-1. Potency of Common Steroid Preparations

NAME RELATIVE GLUCOCORTICOID POTENCY RELATIVE DOSING (MG) RELATIVE MINERALOCORTICOID POTENCY
Cortisone 1 100 +
Hydrocortisone 1.25 80 ++
Prednisone 5 20 +
Prednisolone 5 20 +
Methylprednisolone 6 16 0
9a-Fluorocortisol 20 5 +++++
Dexamethasone 50 1 0
Adapted from Donohoue PA: The adrenal cortex. In McMillan JA, DeAngelis CD, et al, editors: Oski’s Pediatrics, Principles and Practice, ed 3. Philadelphia, 1999, JB Lippincott, p 1814.

9. How do physiologic, stress, and pharmacologic doses of hydrocortisone differ?
• Physiologic: Careful studies have shown that adrenal glucocorticoid production in the normal individual is about 7 to 8 mg/m2 per 24 hours. Because 50% to 60% of oral hydrocortisone is absorbed, the recommended oral physiologic replacement is about 12 to 15 mg/m2 per 24 hours.
• Stress: On the basis of studies performed before the development of high-quality radioimmunoassays, the consensus opinion was that production of glucocorticoid increased about threefold when individuals were physiologically stressed. Hence, when the term stress dose is used, it generally means that the dose is at least three times above physiologic replacement, that is, 50 to 100 mg/m2 per 24 hours of hydrocortisone.
• Pharmacologic: Glucocorticoids are extensively used in pharmacologic doses for the treatment of various inflammatory processes and in surgery or trauma to reduce or prevent swelling and inflammation. Doses of glucocorticoid higher than 50 mg/m2 per 24 hours of hydrocortisone that are being used to treat these conditions are referred to as pharmacologic doses; that is, the medication is not being used for adrenal replacement or stress dosing.
10. When does adrenal-pituitary axis suppression occur in prolonged glucocorticoid treatment?
As a general rule, the longer the duration of treatment and the higher the dose of glucocorticoid, the greater the risk for adrenal suppression. If pharmacologic doses of glucocorticoids are used for less than

10 days, there is a relatively low risk for permanent adrenal insufficiency, whereas daily use for more than 30 days carries a high risk for prolonged or permanent adrenal suppression. The reason for glucocorticoid treatment must also be considered; that is, a child with severe head trauma may have initially been on treatment with glucocorticoids to reduce brain swelling but is also at significant risk for secondary pituitary deficiencies.

CALCIUM METABOLISM AND DISORDERS
11. What are the causes of hypercalcemia?
Remember the “High 5-Is” mnemonic: H (hyperparathyroidism) plus the five Is (idiopathic, infantile, infection, infiltration, and ingestion) and S (skeletal disorders).
Hyperparathyroidism:
• Familial
• Isolated
• Syndromic
Idiopathic:
• Williams syndrome
Infantile:
• Subcutaneous fat necrosis
• Secondary to maternal hypoparathyroidism and inadequate transfer of calcium across the placenta.
Infection:
• Tuberculosis
Infiltration:
• Malignancy
• Sarcoidosis
Ingestion:
• Milk-alkali syndrome
• Thiazide diuretics
• Vitamin A intoxication
• Vitamin D intoxication
Skeletal Disorders:
• Hypophosphatasia
• Immobilization
• Skeletal dysplasias
12. An 8-year-old in a spica cast after hip surgery develops vomiting and a serum calcium concentration of 15.3 mg/dL. Has there been error in the order written for intravenous fluids and what should be done?
This child’s extreme hypercalcemia is likely due to immobilization from the full body cast and bone resorption. A serum calcium concentration of more than 15 mg/dL or the presence of significant symptoms (i.e., vomiting, hypertension) constitutes a medical emergency and requires immediate intervention to lower the calcium level. The initial approach of medical treatment is to increase urinary excretion of calcium. This is achieved with isotonic saline at two to three times maintenance rates with normal renal function. Once hydration is adequate, furosemide, 1 mg/kg intravenously every 6 to 8 hours, may be added until calcium decreases to 12 mg/dL. Furosemide and other loop diuretics are potent diuretic and calciuric agents. Meticulous monitoring of input and output and of serum and urinary electrolytes (including serum magnesium) is vital.
Electrocardiographic (ECG) monitoring is mandatory because hypercalcemia can be associated with conduction disturbances including premature ventricular contractions, ventricular tachycardia, prolonged PR interval, prolonged QRS duration, decreased QTc interval, and atrioventricular block. Additional treatment with glucocorticoids to decrease calcium absorption and antihypercalcemic agents to inhibit bone resorption may also be considered. Admission to an intensive care unit for careful monitoring of cardiac status, electrolyte levels, and fluid management may be necessary. If increased mobilization is possible, this will help correct the hypercalcemia.

Kirkland JL: Parathyroid glands. In Crocetti M, Barone MA, editors: Oski’s Essential Pediatrics, ed 2. Philadelphia, 2004, Lippincott Williams & Wilkins, p 551.

13. Is it the Chvostek or Trousseau sign that gets the tap? Both are clinical manifestations of neuromuscular irritability that occur when hypocalcemia or hypomagnesemia are present; normal extracellular calcium concentrations are necessary for muscle and nerve function.
• Chvostek sign: Tapping the parotid gland over the facial nerve in front of the ear results in facial muscle spasm with movement of the upper lip.
• Trousseau sign: Mild hypoxia induced by inflating a blood pressure cuff at pressures greater than systolic for 2 to 5 minutes results in carpopedal spasm in the setting of hypocalcemia.
Of these two signs, Trousseau sign is more specific. It is recommended that both clinical signs be confirmed with measurement with ionized calcium. An easy way to remember the difference is that the ChVOSTEK sign affects part of the cheek.

Cooper M, Gittoes N: Diagnosis and management of hypocalcemia, BMJ 336:1298–1302, 2008.

14. What is hypoparathyroidism?
Parathyroid hormone (PTH) is a calcium regulatory hormone released by the parathyroid glands that increases serum calcium by increasing the resorption of Ca2+ from bone and by increasing gastrointestinal and urinary absorption of calcium through the increasing synthesis of calcitriol. Hypoparathyroidism can result from a developmental defect, destruction by surgery or autoimmune process, or from a biosynthetic defect in hormone production. The result can be acute or chronic hypocalcemia. An intact PTH level should be obtained in all children presenting with hypocalcemia. The result should be interpreted in light of the calcium level; that is, is the PTH appropriately elevated for the degree of hypocalcemia?

Shoback D: Hypoparathyroidism, N Engl J Med 359:391–403, 2008.

15. In what clinical circumstances should hypoparathyroidism be suspected?
• Manifestations of hypocalcemia (e.g., carpopedal spasm, bronchospasm, tetany, seizures)
• Lenticular cataracts (these can also occur with other causes of long-standing hypocalcemia)
• Changing behaviors, ranging from depression to psychosis
• Mucocutaneous candidiasis (seen in familial form)
• Dry and scaly skin, psoriasis, and patchy alopecia
• Brittle hair and fingernails
• Enamel hypoplasia (if hypocalcemia is present during dental development)

16. What are the main causes of hypocalcemia in children?
• Nutritional: Inadequate intake of vitamin D and, in rare instances, severely inadequate intake of calcium and/or excessive intake of phosphate may cause this condition.
• Renal insufficiency: This may be the result of the following: (1) increased serum phosphorus from a decreased glomerular filtration rate with depressed serum calcium and secondary hyperparathyroidism or (2) decreased activity of renal α-hydroxylase, which converts 25- hydroxyvitamin D into the biologically active form, 1,25-(OH)2 D.
• Nephrotic syndrome: With lowered serum albumin, total calcium levels are reduced. Additionally, intestinal absorption of calcium is decreased, urinary losses of cholecalciferol-binding globulin are increased, and urinary losses of calcium are increased with prednisone therapy (standard treatment for minimal change nephrotic syndrome). In patients with hypoalbuminemia, there will be a decrease in total calcium but no decrease in ionized calcium. The corrected calcium is estimated by adding
0.8 mg/dL to the total calcium for every 1-mg decrease in the serum albumin below 4 mg/dL.
• Hypoparathyroidism: In infants, this may result from a developmental defect during embryogenesis (parathyroid gland aplasia or hypoplasia) and may occur in the context of a syndrome such as DiGeorge syndrome caused by a deletion in chromosome 22q11. In older children, it may occur in the context of autoimmune polyglandular syndrome (type 1) or mitochondrial myopathy syndromes.
• Pseudohypoparathyroidism: This is a group of peripheral resistance syndromes in which resistance to PTH results in elevated PTH levels in the setting of normal renal function and subsequent hypocalcemia due to blunted or absent PTH effect in the setting of high serum concentrations of PTH.
• Disorders of calcium sensor genes: Activating mutations of the calcium sensing receptor gene (CaSR) result in calcium being sensed as normal at subphysiologic levels and PTH secretion switched off inappropriately causing hypoparathyroidism.

Moe SM: Disorders involving calcium, phosphorus, and magnesium, Prim Care 35:215–237, 2008. Umpaichitra V, Bastian W, Castells S: Hypocalcemia in children: pathogenesis and management, Clin Pediatr 40:305–312, 2001.

17. What is the most likely diagnosis in a child with hypocalcemia who has abnormally shaped fingers?
Albright hereditary osteodystrophy (AHO), a type of pseudohypoparathyroidism, is characterized by short stature, obesity, developmental delay, and brachydactyly, specifically a shortening of the fourth and fifth metacarpals (Fig. 6-1).

Desai N, Kalra A: Short fourth and fifth metacarpals, JAMA 308:1034–1035, 2012.

Figure 6-1. Short fourth and fifth metatarsals of a child (A), which are more readily appreciated on the radiograph (B).
(From Moshang T Jr: Pediatric Endocrinology: The Requisites in Pediatrics. Philadelphia, 2005, ELSEVIER Mosby, p 8.)

CLINICAL SYNDROMES
18. How does the syndrome of inappropriate secretion of antidiuretic hormone (SIADH) develop?
Antidiuretic hormone (ADH) is released from the posterior pituitary gland and serves as a regulator of extracellular fluid volume. The secretion of ADH is regulated by changes in osmolality sensed by the hypothalamus and alterations in blood volume detected by carotid and left atrial stretch receptors. By definition, the secretion of ADH in SIADH is inappropriate; therefore, the person cannot be given this diagnosis if dehydrated.
Intracranial pathology can increase the secretion of ADH directly by local CNS effects, and intrathoracic pathology can increase secretion by stimulating volume receptors. Medications can directly promote ADH release and enhance its renal effects. SIADH is usually asymptomatic until symptoms of water intoxication and hyponatremia develop. Nausea, vomiting, irritability, personality changes, progressive obtundation, and seizures can result. An individual with hyponatremia that has developed over a prolonged period of time is less likely to have symptoms than one in whom the hyponatremia has developed acutely.
19. What is cerebral salt wasting and how is it separated from SIADH?
Cerebral salt wasting (CSW) is defined as excessive urinary sodium losses in individuals with intracranial disease that result in hyponatremia and dehydration. The mechanism is still not clear. CSW typically develops in the first week after brain injury and generally resolves over time. Both CSW and SIADH are associated with hyponatremia. However, individuals with CSW have signs of intravascular volume depletion (e.g., rapid pulse, low blood pressure), whereas children with SIADH have evidence of intravascular volume overload. In SIADH, fluid restriction often leads to an increase in the serum sodium. In contrast, fluid restriction in CSW will not increase serum sodium but will further impair intravascular volume; therefore, it can be dangerous and may result in cardiovascular compromise.
20. What are the five criteria for the diagnosis of SIADH?
1. Hyponatremia with reduced serum osmolality
2. Urine osmolality elevated compared with serum osmolality (a urine osmolality <100 mOsm/dL usually excludes the diagnosis)
3. Urinary sodium concentration excessive for the extent of hyponatremia (usually >20 mEq/L)
4. Normal renal, adrenal, and thyroid function
5. Absence of volume depletion
21. What clinical features suggest diabetes insipidus (DI)?
Because DI is caused by an insufficiency of ADH or the inability to respond to ADH, the signs and symptoms tend to be directly related to excessive fluid loss. The clinical spectrum may vary depending on the child’s age. The infant may present with symptoms of failure to thrive as a result of chronic dehydration, or there may be a history of repeated episodes of hospitalizations for dehydration. There may also be a history of intermittent low-grade fever.
Often, caretakers report a large volume of intake or an inability to keep a dry diaper on the infant. The higher absorbency of disposable diapers may delay the diagnosis in infants. In the young child, DI may appear as difficulty with toilet training. In the older child, the reappearance of enuresis, increasing frequency of urination, nocturia, or dramatic increases in fluid intake may be noted. Frequent urination with large urinary volumes should lead to the suspicion of DI.
22. How is the diagnosis of DI made?
Deprivation of water intake for a limited time and judicious monitoring of physical and biochemical parameters may be required. The diagnosis of DI rests on the demonstration of the following: (1) an inappropriately dilute urine in the face of a rising or elevated serum osmolality; (2) urine output that remains high despite the lack of oral input; and (3) changes in physical parameters that are consistent with dehydration (weight loss, tachycardia, loss of skin turgor, dry mucous membranes). A child who, with
water deprivation appropriately concentrates urine (>800 mOsm/L) and whose serum osmolality remains constant (<290 mOsm/L) is unlikely to have DI. When DI is considered, a pediatric endocrinology consultation is strongly recommended.
If a child meets the criteria for the diagnosis of DI, the water-deprivation test is usually ended with the administration of some form of ADH, such as desmopressin, and the provision of fluids.
If the urine subsequently becomes appropriately concentrated, this confirms the diagnosis of ADH

deficiency (central DI). Failure to concentrate suggests renal resistance to ADH (nephrogenic DI).
DI may often be the first clinical sign of tumor of the hypothalamus or base of the skull (e.g., Wegener granulomatosis). Brain magnetic resonance imaging (MRI) is recommended if a diagnosis of DI is confirmed.

Ranadive SA, Rosenthal SM: Pediatric disorders of water balance, Endocrinol Metab Clin North Am 38:663–672, 2009. Linshaw MA: Congenital nephrogenic diabetes insipidus, Pediatr REV 28:372–379, 2007.

DIABETIC KETOACIDOSIS
23. What is diabetic ketoacidosis (DKA)?
DKA is a state of severe metabolic derangement that results from both severe insulin deficiency and increased amounts of counter-regulatory hormones (catecholamines, glucagon, cortisol, and
growth hormone). The main features are hyperglycemia (glucose > 200 mg/dL), ketone production, and acidosis (venous pH <7.30 or serum HCO3 <15 mEq/L).
24. What percentage of newly diagnosed diabetic patients present with symptoms of DKA?
This is extremely variable from location to location (13% to 80%) depending on access to care; economic status of the community; and other factors, including family history. The early symptoms of DKA are more likely to be missed or misinterpreted in very young children.

Klingensmith GJ, Tamborlane W, Wood J, et al: Diabetic ketoacidosis at diabetes onset: still an all too common threat in youth, J Pediatr 162:330.e1–334.e1, 2013.
Usher-Smith JA, Thompson M, Ercole A, et al: Variation between countries in the frequency of diabetic ketoacidosis at first presentation of type 1 diabetes in children: a systematic review, Diabetologia 55(11):2878–2894, 2012.

25. What are the mainstays of therapy for DKA?
• Adequate initial supportive care (airway maintenance if necessary, supplemental oxygen as needed)
• Volume resuscitation (which should begin before starting insulin therapy)
• Insulin administration (Initial dose of 0.05 to 0.1 unit/kg/hour)
• Frequent monitoring of vital signs, electrolytes, glucose, and acid-base status

Olivieri L, Chasm R: Diabetic ketoacidosis in the pediatric emergency department, Emerg Med Clin North Am
31:755–773, 2013.

26. What should be the initial fluid management in DKA?
The association of the rate of sodium and fluid administration in DKA and the development of cerebral edema remains controversial. The concern is that falling osmolarity might contribute to cerebral edema. The International Society for Pediatric and Adolescent Diabetes (ISPAD) recommends the following: Initial:
• In the rare patient who presents in shock, circulatory volume should be rapidly restored with isotonic saline (or lactated Ringer solution) in 20 mL/kg boluses with reassessment after each bolus.
• In patients who are severely volume depleted but not in shock, the initial volume is typically 10 mL/kg given over 1 to 2 hours.
Subsequent:
• Once a patient is hemodynamically stable, fluid replacement is given more slowly. Replacement of the remainder of the fluid deficit (after subtracting the volume of the boluses that were received) is given over the next 48 hours at a rate not to exceed 1.5 to 2 times the maintenance rate. Generally, DKA is associated with an initial weight loss of 7% to10%.
• Choice of fluid tonicity should be made based on each patient’s clinical status (degree of hyperosmolarity, CNS status, serum sodium trend, etc.). ISPAD guidelines state that “no treatment strategy can be definitively recommended as being superior to another based on evidence.”

Wolfsdorf J, Craig ME, Daneman D, et al: Diabetic ketoacidosis in children and adolescents with diabetes, Pediatr Diabetes 10(Suppl 12):118–133, 2009.

27. Why is a falling serum sodium concentration during the treatment of DKA of concern? Most patients with DKA have a significant sodium deficit of 8 to 10 mEq/kg, which needs to be replaced. After initial fluid boluses, fluids containing 0.5% normal saline or greater are generally required. As a general rule, the serum sodium concentration is low at the outset and rises throughout the course of treatment. An initial sodium concentration of more than 145 mEq/L suggests severe dehydration or hyperosmolarity.
The “corrected” serum sodium should be followed throughout treatment. This value can be calculated using the equation: Corrected sodium Measured sodium (mEq/L)+ 0.016 [serum glucose (mg/dL) – 100]. A corrected serum sodium that begins to fall with treatment merits prompt attention because it indicates either inappropriate fluid management or the onset of SIADH and can signal impending cerebral edema.

Katz MA: Hyperglycemia-induced hyponatremia—calculation of expected serum sodium depression, N Engl J Med 289 (16):843–844, 1973.

28. What is the typical potassium status in children with DKA?
In almost all children with DKA, there is a depletion of intracellular potassium and a substantial total body potassium deficit of 3 to 6 mmol/kg, although the initial measured serum potassium value may be normal or high, in large part because of acidosis. Replacement therapy will be needed. If the patient is hypokalemic, potassium should be given with the initial volume expansion and before insulin administration. Insulin administration results in potassium transport into cells with a further decrease in serum levels. If the initial potassium level is within a normal range, begin potassium replacement (with the concentration in the infusate at 40 mEq/L) after the initial volume expansion and concurrent with starting insulin therapy, provided that urine output can be documented. If the initial potassium measurement is significantly elevated, defer potassium replacement until urine output has been documented and the hyperkalemia abates. Of note, if rapid serum potassium levels are not available, an electrocardiogram (ECG) to look for changes of hypokalemia or hyperkalemia (e.g., T-wave changes) can be valuable in guiding management.

Wolfsdorf J, Craig ME, Daneman D, et al: Diabetic ketoacidosis in children and adolescents with diabetes, Pediatr Diabetes 10(Suppl 12):118–133, 2009.

29. Why do potassium levels fall during the management of DKA?
Correction of acidosis (less K+ exchanged out of cell for H+ as pH rises)
• Insulin administration (increases cellular uptake of K+)
• Dilutional effects of rehydration
• Ongoing urinary losses
Most patients are potassium depleted, although the serum K+ is usually normal or elevated. A low K+ is particularly worrisome because it suggests severe potassium depletion.
30. Should bicarbonate be used for the treatment of children with DKA?
The pros and cons of using sodium bicarbonate are shown in Table 6-2.

Table 6-2. Factors Determining Use of Bicarbonate Treatment in Diabetic Ketoacidosis
PROS CONS
Improved pH enhances myocardial contractility problems rare in children and response to catecholamines Cardiac function problems rare in children
Ventilatory response to acidosis blunted when pH
is <7.0 Ventilatory response well-maintained in children
No adverse effect of bicarbonate on oxygenation has been demonstrated clinically May alter oxygen-binding of hemoglobin, potentially decreasing tissue oxygenation

Table 6-2. Factors Determining Use of Bicarbonate Treatment in Diabetic Ketoacidosis (Continued )
PROS CONS
Questionable relevance of central nervous system acidosis Paradoxical central nervous system acidosis documented in humans
May be useful in the rare patient with hyperkalemia Hypokalemia may result from uptake of K+ as acidosis is corrected; low serum K is six times more common after bicarbonate treatment
May be associated with increased hyperosmolarity and cerebral edema

31. Are there any indications for the use of bicarbonate? Bicarbonate administration for the acidosis in DKA has not been shown to be beneficial in controlled trials. The establishment of an adequate intravascular volume and the provision of sufficient quantities of insulin are far more important in the treatment of DKA than bicarbonate. The decision to initiate bicarbonate therapy should be based on an arterial blood gas level and not a venous blood gas level. Two possible indications include:
• Profound metabolic acidosis (arterial pH< 6.9), which may be compromising cardiac contractility and/or adversely affecting the action of epinephrine during resuscitation
• Life-threatening hyperkalemia with bradycardia, severe muscle weakness

Wolfsdorf J, Craig ME, Daneman D, et al: Diabetic ketoacidosis in children and adolescents with diabetes, Pediatr Diabetes 10(Suppl 12):125, 2009.
Green SM, Rothrock SG, Ho JD, et al: Failure of adjunctive bicarbonate to improve outcome in severe pediatric diabetic ketoacidosis, Ann Emerg Med 31:41–48, 1998.

32. When should glucose be added to the intravenous fluids in patients with DKA? This will depend on the rate at which the serum glucose level is decreasing. Generally, when the glucose level approaches 300 mg/dL, glucose should be added to the intravenous fluid. It is usually wise to order the appropriate glucose-containing fluid in advance to avoid hypoglycemia. Many centers now use the “two-bag” method: they order two bags of intravenous fluid, with identical electrolyte content except for the glucose concentration. One contains 10% or 12.5% glucose, and the other contains no glucose. As the blood sugar approaches 300 mg/dL, glucose is added to the infusate (through a Y tube). With the two-bag system, it is possible to alter the concentration of glucose anywhere between 0% and 12.5%, with a goal of maintaining the blood sugar in the 100- to 200-mg/dL range, thereby avoiding hypoglycemia. It is important to note that if the blood glucose concentration is decreasing too quickly or is too low before the resolution of acidosis, it is preferable to increase glucose levels by adding glucose to the infusate rather than decreasing the rate of insulin infusion.

Poirier MP, Greer D, Satin-Smith M: A prospective study of the “two-bag system” in diabetic ketoacidosis management,
Clin Pediatr 43:809–813, 2004.

KEY POINTS: DIABETIC KETOACIDOSIS
1. The triad of metabolic derangement includes hyperglycemia, ketosis, and acidosis.
2. Abdominal pain can mimic appendicitis; hyperventilation can mimic asthma or pneumonia.
3. Initial bolus of insulin is no longer recommended.
4. Total-body potassium is usually significantly diminished.
5. Cerebral edema is the most common cause of morbidity and mortality in children with DKA.
6. If the sodium level begins to fall with fluid replenishment, beware of secretion of antidiuretic hormone and possible cerebral edema.
7. Bicarbonate therapy is usually not indicated for acidosis.

33. In the past, a bolus of insulin was given at the start of therapy for DKA. Is that still recommended?
No. An initial bolus (traditionally 0.1 U/kg) was previously given before any subsequent insulin. This has been found to be unnecessary and may increase the risk for cerebral edema.

Wolfsdorf J, Glaser N, Sperling MA: Diabetic ketoacidosis in infants, children, and adolescents: a consensus statement from the American Diabetes Association, Diabetes Care 29:1150–1159, 2006.

34. Is continuous or bolus insulin better for the initial treatment of DKA?
Extensive evidence demonstrates that continuous intravenous (IV) insulin (with an initial dose of 0.05 to
0.1 unit/kg/hr) should be the standard of care. The lower insulin dose may be more suitable for younger children, who are more insulin sensitive. Therapy should begin after the initial fluid bolus is complete. Beginning insulin at the start of fluid therapy increases the risk for severe hypokalemia and for decreasing the serum osmolarity too quickly. In general, this infusion should be maintained until the acidosis has significantly improved (pH >7.30, bicarbonate >15 mmol/L, and/or closure of the anion gap). If
continuous IV administration of insulin is not possible, short- or rapid-acting insulin (insulin lispro or
insulin aspart) can be given subcutaneously (SC) or intramuscularly (IM) every 1 to 2 hours if peripheral circulation is not impaired. A recommended initial dose is 0.3 unit/kg SC followed 1 hour later by SC insulin at 0.1 unit/kg every hour or 0.12 to 0.2 unit/kg every 2 hours.

Wolfsdorf J, Craig ME, Daneman D, et al: Diabetic ketoacidosis in children and adolescents with diabetes, Pediatr Diabetes 10(Suppl 12):123–124, 2009.
Wolfsdorf J, Glaser N, Sperling MA: Diabetic ketoacidosis in infants, children, and adolescents: a consensus statement from the American Diabetes Association, Diabetes Care 29:1153–1154, 2006.

35. Which corrects sooner during continuous insulin administration: hyperglycemia or acidosis?
Hyperglycemia. Even though there may be normalization of the serum glucose, a persistent acidosis may be present. Thus, the continuous infusion of insulin should not be reduced until there is resolution of the ketoacidosis. Glucose may be added to the intravenous fluids to prevent hypoglycemia while the acidosis is being corrected.
36. What is the main cause of mortality in DKA? Cerebral edema. Clinically significant cerebral edema occurs in up to 1% of pediatric patients with high mortality rates.
37. What risk factors are associated with the development of cerebral edema?
The pathogenesis of cerebral edema is incompletely understood. Computed tomography (CT) studies have demonstrated that subclinical cerebral edema may occur in a majority of pediatric patients with DKA. The escalation to life-threatening cerebral edema is unpredictable, often occurring
as biochemical abnormalities are improving. It may be sudden in onset or occur gradually, but it typically occurs during the first 5 to 15 hours after therapy begins. Risk factors identified include the following:
• Younger age
• Newly diagnosed patients
• More profound acidosis
• Attenuated rise in serum sodium during therapy
• Greater hypocapnia (after correcting for acidosis)
• Increased blood urea nitrogen (BUN)
• Bicarbonate therapy for acidosis
• Administration of insulin in first hour of fluid treatment
• Higher volumes of fluid given during the first 4 hours

Wolfsdorf J, Craig ME, Daneman D, et al: Diabetic ketoacidosis in children and adolescents with diabetes, Pediatr Diabetes 10(Suppl 12):126, 2009.
Levin DL: Cerebral edema in diabetic ketoacidosis, Pediatr Crit Care Med 9:320–329, 2008.
Glaser NS, Wootton-Gorges SL, Buonocore MH, et al: Frequency of sub-clinical cerebral edema in children with diabetic ketoacidosis, Pediatr Diabetes 7:75–80, 2006.

38. What signs and symptoms suggest worsening cerebral edema during the treatment of DKA?
• Headache
• Vomiting, recurrent
• Change in mental status: increased drowsiness, irritability, restlessness
• Change in neurologic status: cranial nerve palsy, abnormal pupillary responses, abnormal posturing
• Incontinence (age inappropriate)
• Rising blood pressure
• Inappropriate heart rate slowing
• Decreased oxygen saturation

Wolfsdorf J, Craig ME, Daneman D, et al: Diabetic ketoacidosis in children and adolescents with diabetes, Pediatr Diabetes 10(Suppl 12):126, 2009.

DIABETES MELLITUS
39. What are the risks of a child developing type 1 diabetes (T1D) if one sibling or parent is affected?
Most people with T1D have no family history of the disorder, but having a first or second degree relative with T1D increases the risk of T1D. The human leukocyte antigen (HLA) gene cluster is responsible for 40- 60% of a person’s GENETIC risk, with many other genes and unknown environmental factors also playing a role. The major genetic determinants are polymorphisms of class II HLA genes encoding DQ and DR with specific haplotypes imparting the highest risk. Current approaches to assessing T1D risk use family history, measurement of T1D associated antibodies and HLA genotyping (DR and DQ).
• Sibling with T1D: 6%
• DR3/4-DQ8 POSITIVE HLA identical sibling with T1D: 55% to 80%
• DR3/4-DQ8 POSITIVE HLA nonidentical sibling with T1D: 5%
• DR3/4-DQ8 NEGATIVE HLA identical sibling with T1D: 25%
• Father with T1D: 4% to 6%
• Mother with T1D: 2% to 4%

Baschal EE, Eisenbarth GS: Extreme genetic risk for type 1A diabetes in the post-genome era, J Autoimmun 31:1–6, 2008.

40. How long does the “honeymoon” period last in patients with newly diagnosed T1D? The honeymoon usually begins within 2 to 4 weeks after the initiation of insulin treatment. It is a period of decreased exogenous insulin requirements due to residual endogenous insulin production. The honeymoon period may last for a few weeks or months, but it is not predictable. Evidence is accumulating that endogenous insulin production may be preserved by the maintenance of “excellent control” and avoidance of hyperglycemia. Cessation of the honeymoon is often heralded by increasing insulin requirements. This is generally gradual but may be more acute, occurring with an intercurrent illness that increases insulin requirements.
41. How do the types of insulin vary in their onset and duration of action?
See Table 6-3.

Table 6-3. Pharmacokinetics of Insulin and Insulin-Like Agents*

INSULIN
ONSET
PEAK EFFECTIVE DURATION
Rapid Acting
Aspart, glulisine, lispro (NovoLog, Apidra, Humalog)
5-15 min
30-90 min
3-5 hr
Short Acting
Regular†

30-60 min
2-3 hr
4-8 hr
Continued on following page

Table 6-3. Pharmacokinetics of Insulin and Insulin-Like Agents (Continued )

INSULIN
ONSET
PEAK EFFECTIVE DURATION
Intermediate Acting
NPH 2-4 hr 4-10 hr 10-16 hr
(Humulin N, Novolin N)
Detemir (Levemir) 2-4 hr 3-8 hr 10-24 hr
Long Acting
Glargine (Lantus)
2-4 hr
No peak
20-24 hr
*Assuming 0.1 to 0.2 U/kg per injection. Onset and duration vary significantly by injection site.
†Regular insulin is available in different strengths; the standard is U-100 (100 units per mL) but U-500 is also available for those with extreme insulin resistance.

42. When should the Somogyi phenomenon be suspected?
The Somogyi phenomenon is rebound hyperglycemia after an incident of hypoglycemia. This rebound is secondary to the release of counterregulatory hormones, which is the natural response to hypoglycemia. As tighter glucose control is maintained, there is an increased likelihood of hypoglycemia and therefore of the Somogyi phenomenon. If the hypoglycemia is recognized and treated promptly, rebound hyperglycemia is less likely to occur. Thus, the Somogyi phenomenon is commonly reported more frequently at night because there is the greater likelihood of unrecognized and untreated hypoglycemia when the child is asleep. The Somogyi phenomenon should be suspected when a
child whose blood sugar is in excellent control begins to have intermittent high blood glucoses in the morning. If that pattern is noted, blood glucose concentration should be checked between 2:00 and 3:00 AM on several nights to determine whether hypoglycemia is occurring. If hypoglycemia can be documented, the dose or type of evening insulin may need to be altered, or the time that the dose is given may need to be changed.
43. What causes the “dawn phenomenon”? The term dawn phenomenon describes a rise in blood glucose concentration that occurs during the early morning hours (between 5:00 and 8:00 AM), particularly among patients who have normal glucose levels throughout most of the night. The rise in glucose is thought to be due to several factors, including the following:
• The cumulative effect of increased nocturnal growth hormone
• The normal increase in the morning cortisol level
Preventing morning hyperglycemia in someone with a pronounced “dawn phenomenon” may be difficult without the use of an insulin pump.

KEY POINTS: DIABETES MELLITUS TYPE 1
1. Destruction of pancreatic islet cells causes an absolute insulin deficiency.
2. Classic triad of symptoms includes polyuria, polydipsia, and polyphagia.
3. Tighter glucose control substantially lowers complication rates of retinopathy, nephropathy, and neuropathy.
4. Obtaining a hemoglobin A1C (glycosylated hemoglobin) level is a standard way to assess average control during the previous 2 to 3 months.
5. Puberty is a time of increased insulin resistance, thereby requiring increased dosing.

44. How rapidly can renal disease develop after the onset of diabetes mellitus? Microscopic changes in the glomerular basement membrane may be present by 2 years after the diagnosis of diabetes. Microalbuminuria often may be detected as early as 5 years after the diagnosis of T1D. Patients with diabetic nephropathy account for more than 25% of those receiving long-term renal dialysis in the United States. Progression can be substantially delayed by meticulous attention to glycemic control.

45. How is hemoglobin A1C (HbA1C) helpful for monitoring glycemic control? The hemoglobin A1C level (also known as glycosylated hemoglobin), is a hemoglobin-glucose combination formed nonenzymatically within the cell. Initially, an unstable bond is formed between glucose and the hemoglobin molecule. With time, this bond rearranges to form a more stable compound in which glucose is covalently bound to the hemoglobin molecule. The amount of the unstable form may rise rapidly in the presence of a high blood glucose level, whereas the stable form changes slowly and provides a time- average integral of the blood glucose concentration through the 120-day life span of the red blood cell. Thus, glycohemoglobin levels provide an objective measurement of averaged diabetic control over time.

Cooke DW, Plotnick L: Type 1 diabetes mellitus in pediatrics, Pediatr REV 29:374–384, 2008. Rewers M, Pihoker C, Donaghue K, et al: Assessment and monitoring of glycemic control in children and adolescents with diabetes, Pediatr Diabetes 8:408–418, 2007.

46. What are the goals for hemoglobin A1C?
See Table 6-4.

Table 6-4. Hemoglobin A1C Goals for Children and Adolescents with Type 1 Diabetes
AGE HBA1C GOAL*

<19 years <7.5%
≤19 years <7.0%
*According to 2013 International Society of Pediatric and Adolescent Diabetes and 2014 American Diabetes Association Recommendations.

47. What pathophysiologic process characterizes type 2 diabetes (T2D)? T2D is characterized by a resistance to insulin action accompanied by a relative insulin secretory defect, in the absence of autoimmune markers.
48. Is the prevalence of T2D in children increasing?
Previously rare in pediatrics, T2D in children is increasing in incidence and prevalence. In the United States, there was a 30% increase in the prevalence of T2D in children less than 19 years of age between 2001 and 2009. It is most common in non-white children and adolescents with a strong family history of T2D. It is rarely seen before the onset of puberty.

Dabelea D, Mayer-Davis EJ, Saydah S, et al: Prevalence of type 1 and type 2 diabetes among children and adolescents from 2001 to 2009, JAMA 311:1778–1786, 2014.

49. What historical and clinical features suggest type 2 rather than type 1 diabetes?
• Obesity is very common in children with T2D; it is much less common in children with T1D at diagnosis.
• Age of onset: T2D only rarely presents before the onset of puberty; T1D commonly presents in prepubertal as well as pubertal children.
• Racial and ethnic minority groups, particularly black, Mexican Americans, and Native Americans, are at higher risk for T2D.
• Family history is usually strongly positive when a child develops T2D; more than 50% of affected children have at least one first-degree relatives with T2D.
• Acanthosis nigricans, a marker of insulin resistance, is present in 90% of T2D cases, most commonly on the posterior neck.
• Hyperandrogenism in girls is associated with insulin resistance and obesity. This is common in girls and young women with T2D.
• Differing symptoms: Unlike patients with T1D, most children and adolescents with T2D will present without ketonuria (although up to 33% of children with T2D will present with ketonuria).

Liu L, Hironaka K, Pihoker C: Type 2 diabetes in youth, Curr Probl Pediatr Adolesc Health Care 34:254–272, 2004.

50. What laboratory features are helpful to distinguish T1D from T2D? Although classification can usually be made on the basis of clinical characteristics, measurement of levels of fasting insulin and C-peptide (low in T1D; normal or elevated in T2D) or islet cell autoantibodies (positive in T1D; absent in T2D) may be useful to distinguish T1D from T2D. Be mindful that there can be overlap in the laboratory evaluation.
51. What is acanthosis nigricans? Acanthosis nigricans is hyperpigmented and often highly rugated patches that are found most prominently in intertriginous areas, especially on the nape of the neck (Fig. 6-2). This is a marker of insulin resistance.

Figure 6-2. Acanthosis nigricans in an adolescent male. (From Schachner LA, Hansen RC, editors: Pediatric Dermatology,
ed 3. Edinburgh, 2003, Mosby, p 915.)

52. How is T2D diagnosed? The diagnosis of diabetes is based on blood glucose level cutoffs and the levels used are the same for T1D and T2D. The diagnosis is made when any of the following criteria are met:
• Random glucose concentration of 200 mg/dL or higher (if accompanied by classic symptoms: polyuria, polydipsia, weight loss)
• Fasting (>8 hours) glucose concentration of more than 125 mg/dL
• Abnormal oral glucose tolerance test defined as a glucose concentration of more than 200 mg/dL
measured 2 hours after drinking 1.75 g/kg of glucose (with a maximum dose of 75 g)
• Hemoglobin A1C≤ 6.5%
American Diabetes Association: Standards of medical care in diabetes—2014, Diabetes Care 37(Suppl 1):S14–S80, 2014.

53. Which pediatric patients should be screened for T2D?
The American Diabetes Association recommends screening beginning at 10 years of age (or earlier if puberty initiates before age 10 years). Screening should be performed using a fasting plasma glucose, oral glucose tolerance test, or hemoglobin A1C for patients with the following risk factors:
• Body mass index more than 85th percentile for age and sex, plus
• Any two of following risk factors: positive family history in first- or second-degree relative; high-risk race/ethnicity (Native American, black, Hispanic, or Asian or Pacific Islander); presence of associated conditions (acanthosis nigricans, hypertension, dyslipidemia, polycystic ovarian syndrome)
• Maternal history of diabetes or gestational diabetes during the child’s gestation

American Diabetes Association: Standards of medical care in diabetes—2014, Diabetes Care 37(Suppl 1):S14–S80, 2014.

54. What hemoglobin A1C level is sufficient to diagnose diabetes?
A level 6.5% on two occasions using a laboratory method is sufficient for the diagnosis. Levels between
5.7 and 6.4 place a person at increased risk for diabetes.

KEY POINTS: DIABETES MELLITUS TYPE 2
1. Pathophysiology includes progressive insulin secretory defect on the background of insulin resistance.
2. Incidence is rising rapidly in association with increased rate of pediatric obesity.
3. Acanthosis nigricans (altered skin pigmentation and texture) associated with insulin resistance is common (found in 90% of cases).
4. Diagnosis is based on detecting hyperglycemia: fasting (≤125 mg/dL), random with symptoms
5. (≤200 mg/dL), or postprandial glucose challenge (≤200 mg/dL), or via HbA1C of > 6.5%.
Screen patients based on known risk factors (obesity, ethnicity, family history).

55. When should oral hypoglycemic agents be considered as part of therapy?
If glucose control is not achieved with dietary adjustments and exercise within 2 to 3 months, oral hypoglycemic agents should be considered. Data on children and adolescents are limited. Metformin (Glucophage) is the best studied and is recommended as initial therapy by many experts. A daily multivitamin is required with metformin therapy because it can interfere with vitamin B12 and folic acid absorption. Insulin is often added to the regimen when patients are unable to meet glycemic targets using lifestyle and metformin alone.

Dileepan K, Feldt MM: Type 2 diabetes mellitus in children and adolescents, Pediatr REV 34:541–548, 2013.

GROWTH DISTURBANCES
56. How do the growth rates of boys and girls differ?
In both boys and girls, the rate or velocity of linear growth begins to decelerate right after birth. In girls, this deceleration continues until the age of about 11 years, at which time the adolescent growth spurt begins. For boys, the deceleration continues until the age of about 13 years. The peak rate of increase in boys occurs at 14 years of age. Growth and growth rate charts are readily available from the Centers for Disease Control and Prevention website (www.cdc.gov/growthcharts/).
57. What is the best predictor of a child’s eventual adult height?
Midparental height. This is an estimate of a child’s expected genetic growth potential based on parental heights (preferably measured rather than by history).
For girls: ([father’s height – 13 cm]+[mother’s height])/2.
For boys: ([mother’s height+ 13 cm]+[father’s height])/2.
This gives the range ( 5 cm) of expected adult height. The predicted height can be compared with the present height percentile, and any significant deviation can be a clue to an abnormal growth pattern in a child. It is important to remember that some forms of growth hormone deficiency are inherited, so one should not automatically assume that the short child with short parents has familial short stature.

58. When have most children achieved the height percentile that is consistent with parental height?
By the age of 2 years. Taking a boy’s length at age 2 years and a girl’s length at age 18 months and doubling them can obtain rough estimates of ultimate adult height.
59. When is a detailed evaluation for short stature warranted?
• Severe height deficit (<1st percentile for age)
• Abnormally slow growth rate (<10th percentile for bone age)
• Predicted height is significantly different from midparental height
• Body proportions are abnormal

Allen DB, Cuttler L: Short stature in childhood—challenges and choices, N Engl J Med 368:1220–1128, 2013.

60. Name the major categories of causes of short stature
• Familial (for short children, 3 standard deviations, with very short parents, consider genetic forms of short stature)
• Constitutional delay (“late bloomer”)
• Chronic disease/treatment (e.g., inflammatory bowel disease, chronic renal failure, renal tubular acidosis, cyanotic congenital heart disease)
• Chromosomal/syndromic (e.g., Turner [45,X], 18q-, Down, achondroplasia)
• Endocrine (e.g., hypothyroidism, growth hormone deficiency, hypopituitarism, hypercortisolism [endogenous and exogenous])
• Psychosocial (e.g., chaotic social situation, orphanage)
• Intrauterine (e.g., small for gestational age)
Genetic patterns and constitutional delay account for the largest percentage of known causes.
61. In a child with short stature, what rate of growth makes an endocrine abnormality unlikely?
In general, heights should be measured over at least a 6-month interval to calculate an accurate rate because growth rates are not completely linear, and measurement is relatively imprecise. Rates of growth are also highly dependent on the age and pubertal status of the child. Growth velocity charts are available at http://www.cdc.gov/growthcharts. Growth rates that are consistently below the 25th percentile or crossing percentiles downward after the age of 2 years warrant careful consideration and possible investigation.
62. When evaluating a short child, why should you ask when the parents reached puberty?
The age at which puberty occurred in other family members may help identify children with constitutional delay because this entity tends to run in families. Most women will remember their age at menarche, and this age can be used as a reference for the age at which other pubertal events occurred. The strongest association for pubertal delay is between father and son. The most useful reference point for adult males is the age at which they reached adult height because almost all normal males will have reached their adult height by the age of 17 years (before high school graduation). Significant growth beyond this age suggests a history of pubertal delay.
63. When does the pubertal growth spurt occur?
For children with an average growth rate, pubertal growth begins earlier in girls. Mean age at the initiation of this spurt is 11 years for boys and 9 years for girls. Peak height velocity occurs at 13.5 years for boys and 11.5 years for girls. Peak velocity occurs at Tanner breast stage II to III for girls and Tanner testis stage III to IV for boys. Girls generally stop growing at an average of 14 years of age, but boys continue to grow until 17 years of age. The major hormone affecting growth cessation is estradiol in both girls and boys. The timing of the pubertal growth spurt may be earlier in certain ethnic groups and in very obese children. Assessment of short stature requires a determination of Tanner pubertal staging. On average, boys will grow 20 to 30 cm following the onset of puberty and girls between
15 and 25 cm.

Cheetham T, Davies JH: Investigation and management of short stature, Arch Dis Child 99:767–771, 2014. Rogol AD, Roemmich JN, Clark PA: Growth at puberty, J Adolesc Health 31(Suppl):192–200, 2002.

64. Are upper to lower body ratios helpful for the diagnosis of growth problems? Disproportionate short stature generally refers to an inappropriate ratio between truncal length and limb length (upper to lower segment ratio). Lower segment (limb length) is the distance from the superior border of the pubic bone to the floor surface. Height minus the lower segment
gives the height of the upper segment (truncal length). In an infant, the head and trunk are quite long relative to the limbs, so the ratio of truncal length to limb length is about 1.7. Throughout childhood, this ratio declines, so that by 7 to 10 years of age this ratio is about 1.0. The adult ratio is 0.9.
An increased ratio is seen in bony dysplasias (e.g., achondroplasia, hypochondroplasia), hypothyroidism, gonadal dysgenesis, and Klinefelter syndrome (the patients are then tall in adolescence). Decreased ratios are seen in certain syndromes (e.g., Marfan syndrome), spinal disorders (e.g., scoliosis), and children who have been exposed to specific types of therapy (e.g., spinal irradiation).

Halac I, Zimmerman D: Evaluating short stature in children, Pediatr Ann 33:170–176, 2004.

65. What laboratory studies should be obtained when evaluating short stature? Extensive laboratory tests are generally not indicated unless the growth velocity is abnormally low. Laboratory testing may include any or all of the following: complete blood count, urinalysis, chemistry panel, sedimentation rate, thyroxine, thyroid-stimulating hormone, insulin-like growth factor-1 (IGF-1), and IGF-binding protein-3 (IGFBP-3). Depending on the ethnic background of the child or the clinical history, testing might also be done for celiac disease, inflammatory bowel disease, renal tubular acidosis, or other occult conditions.
Random growth hormone levels are of little value because they are generally low in the daytime, even in children of average height. IGF-1 mediates the anabolic effects of growth hormone, and levels correlate well with growth hormone status. However, IGF-1 can also be low in nonendocrine conditions (e.g., malnutrition, liver disease).
IGFBP-3, which is the major binding protein for IGF-1 in serum, is also regulated by growth hormone. IGFBP-3 levels generally indicate growth hormone status and are less affected by nutritional factors than IGF-1. Most endocrinologists now use IGF-1 and IGFBP-3 as their initial screening tests for growth hormone deficiency.

Cheetham T, Davies JH: Investigation and management of short stature, Arch Dis Child 99:767–771, 2014.

66. In a very obese child, how does height measurement help determine whether an endocrinopathy might be the cause?
In children with simple obesity (e.g., familial), linear growth is typically enhanced; in children with endocrinopathies (such as Cushing’s syndrome or hypothyroidism), it is usually impaired. If the height of a child is at, or greater than, the midparental height percentile, an endocrine cause of the obesity is unlikely. In some children with craniopharyngiomas, significant obesity with good linear growth can be seen despite documented growth hormone deficiency.

67. How does a growth chart help determine the diagnosis of failure to thrive?
If an infant is demonstrating deceleration of a previously established growth pattern or growth that is consistently less than the 5th percentile, the pattern of growth of head circumference, height, and weight can help establish the likely cause (Fig. 6-3). There are three main types of impaired growth:
• Type I: Retardation of weight with near-normal or slowly decelerating height and head circumference; most commonly seen in undernourished patients
• Type II: Near-proportional retardation of weight and height with normal head circumference; most commonly seen in patients with constitutional growth delay, genetic short stature, endocrinopathies, and structural dwarfism
• Type III: Concomitant retardation of weight, height, and head circumference; seen in patients with in utero and perinatal insults, chromosomal aberrations, and central nervous system abnormalities

3 6 9 12 15 18 21 24 27 30 33 36 3 6 9 12 15 18 21 24 27 30 33 36

(a)

Months

(b)

Months

3 6 9 12 15 18 21 24 27 30 33 36
(c) Months
Figure 6-3. Types I, II, and III of impaired growth. CNS, Central nervous system. (From Roy CC, SILVERMan A, Alagille DA:
Pediatric Clinical Gastroenterology, ed 4. St. Louis, 1995, Mosby, pp 4–8.)

68. How can one track growth in children who have spinal cord abnormalities or severe scoliosis?
There is an excellent 1:1 correlation between span (longest fingertip to longest fingertip measured across the nape of the neck) and height. Thus, span is a useful proxy measure for height/length if it is not possible to get an accurate height. Height and rate of growth, when determined in this way, can be plotted on standard growth and velocity charts.
69. What is bone age?
Bone age is a measure of somatic maturity and growth potential. Standards of normal skeletal radiographic maturation are available, and these are based on the progression of ossification centers that occur at particular ages. A radiograph of the left hand and wrist is taken and compared with those standards to determine a patient’s bone age. This result can be compared with chronologic age to gauge the remaining potential for growth. The interpretation of bone ages can be somewhat difficult and dependent on the pediatric experience of the radiologist.
70. Why is a bone age determination helpful for evaluating short stature? A single bone age is of value for differentiating familial short stature and genetic diseases (in which bone age is normal) from other causes of short stature. A delayed bone age (>2 standard deviations below the mean) that correlates with the child’s height age (age on growth chart at which child’s height would be at the 50th percentile) is suggestive of constitutional delay, whereas a markedly delayed bone age is
suggestive of endocrine disease. Serial bone ages determined every 6 to 12 months are often helpful because, in both the normal child and the child with constitutional delay, the bone age will advance in parallel with the chronologic age. In endocrine disease, the bone age falls progressively further behind the chronologic age. Bone age may be normal or delayed in patients with chronic disease, depending on the severity of disease, its duration, and the type of treatment used.
71. What features suggest constitutional delay as a cause of short stature?
• No signs or symptoms of systemic disease
• Bone age delayed up to 2 to 4 years but consistent with height age (age at which individual’s height would plot on the 50th percentile)
• Period of poorest growth often occurring between the ages of 18 and 30 months, with steady linear growth thereafter (i.e., normal rate of growth for bone age)
• Parental or sibling history of delayed physical development
• Height prediction consistent with family characteristics.

KEY POINTS: GROWTH DISTURBANCES
1. Bone age is used as a diagnostic key: genetically determined short stature (bone age¼ chronologic age) versus constitutional delay (bone age< chronologic age).
2. Midline defects (e.g., single maxillary incisor, cleft lip/palate) and short stature suggest hypopituitarism.
3. Random growth hormone levels are usually not helpful (due to pulsatile delivery during sleep); provocative testing is more reliable.
4. Family history is key. Use growth data about family to establish a pattern.
5. Short stature with overweight suggests endocrinopathy (adrenal, thyroid or growth hormone deficiency).
6. Growth hormone deficiency that appears during the first year of life is associated with hypoglycemia; after the age of 5 years, it is associated with short stature.

72. How is constitutional delay managed? If the results of history, physical examination, and clinical laboratory evaluation are unremarkable, the child is seenonce every 3 to 6 months for accurate heightmeasurementsand determination ofgrowth velocity. A bone age test may bedoneyearly to assess the progression of bony maturation. In patients with constitutional delay, the rate of bone maturation should keep pace with the chronologic age. In children who are of mid- to late pubertal ages (girls >13 years; boys >14 years) but showing minimal or no signs of puberty, selective use of
estrogen or testosterone supplementation to initiate puberty or additional assessment may be indicated.
73. Should growth hormone therapy be given to the normal short child?
This is an area of controversy in pediatric endocrinology. Human growth hormone is most effective when administered subcutaneously on a daily basis. It does increase growth rate and modestly improves adult height (1.2 to 2.8 inches). The safety profile has been good and the risk of short-term adverse events (such as intracranial hypertension or glucose intolerance) is very low. Long-term safety remains under study.

Opponents argue that short stature is not a disease and appropriate therapeutic goals are ill defined. The therapy is expensive, estimated in 2006 dollars to be about $35,000 to $50,000 per inch of height gained.

Allen DB, Cuttler L: Short stature in childhood—challenges and choices, N Engl J Med 368:1220–1128, 2013.
Allen DB: Growth hormone therapy for short stature: is the benefit worth the burden? Pediatrics 118:343–348, 2006.

74. What are the clinical manifestations of growth hormone excess?
Before puberty, the cardinal manifestations are an increase in growth velocity with minimal bone deformity and soft tissue swelling—a condition called pituitary gigantism. Hypogonadotropic hypogonadism and delayed puberty often coexist with growth hormone excess, and affected children exhibit eunuchoid body proportions. If the growth hormone excess occurs after puberty (after epiphyseal closure), the more typical features of acromegaly occur, including coarsening of the facial features and soft tissue swelling of the feet and hands. Growth hormone excess is rare in children.

HYPOGLYCEMIA
75. How is hypoglycemia defined?
A serum glucose concentration of less than 50 mg/dL is defined as hypoglycemia in childhood. Some argue for lower levels being used for term and preterm infants; however, these arguments are based on population sampling data rather than on physiology. Hypoglycemia is a laboratory finding, and its presence should always lead to a diligent search for the underlying pathology. A common cause of a falsely reported abnormal glucose is that the plasma is not quickly separated from the red blood cells. The red cells continue to metabolize glucose, thus lowering the glucose, often into an abnormal range. This should be suspected when the glucose result is reported as part of a chemistry panel, especially if the child was reported to be asymptomatic.
76. What are the clinical findings associated with hypoglycemia?
Neuroglycopenic symptoms include irritability, headache, confusion, unconsciousness, and seizure. Adrenergic signs include tachycardia, tremulousness, diaphoresis, and hunger. Any combination of the above signs and symptoms should lead to the measurement of the blood glucose level.
77. What are the causes of childhood hypoglycemia?
No single cause predominates in any age group. Therefore, the entire differential diagnosis must be considered in any child who presents symptoms of hypoglycemia. Hypoglycemia often occurs as a result of a combination of two or more of the problems listed in Table 6-5 (e.g., prolonged fasting during an

Table 6-5. Differential Diagnosis of Childhood Hypoglycemia
Decreased Glucose Utilization
Hyperinsulinism: focal (adenoma) or diffuse hyperplasia, oral hypoglycemic agents, exogenous insulin
Decreased Glucose Production
Inadequate glycogen reserves: Enzymatic defects in glycogen synthesis and glycogenolysis Ineffective gluconeogenesis: Inadequate substrate (e.g., ketotic hypoglycemia), enzymatic defects
Diminished Availability of Fats
Depleted fat stores
Failure to mobilize fats (e.g., hyperinsulinism)
Defective use of fats: Enzymatic defects in fatty acid oxidation (e.g., medium-chain acyl CoA dehydrogenase deficiency)
Decreased Fuels and Fuel Stores
Fasting, malnutrition, prolonged illness, malabsorption
Increased Fuel Demand
Fever, exercise
Inadequate Counterregulatory Hormones
Growth hormone or cortisol deficiency, hypopituitarism

illness coupled with fever in medium-chain acyl-CoA dehydrogenase deficiency). Specific genetic testing is now available for a number of these entities.
78. An unconscious 3-year-old girl is brought to the emergency department with a serum glucose concentration of 26 mg/dL. What other laboratory tests should be performed?
Blood and urine samples are critically important. The first urine specimen obtained after the presentation of the child is of significant value, even if this cannot be gotten for several hours after the acute event. The blood sample, however, should be drawn before dextrose is administered. The principal laboratory evaluations should include the measurement of the following: (1) the metabolic compounds associated with fasting adaptation; (2) the hormones that regulate these processes; and (3) drugs that can interfere with glucose regulation. It is strongly recommended that an extra purple top and red top tube of blood be drawn, if at all possible. The extra tubes of blood should be kept for additional analyses once the first battery of tests described below is available or after specific recommendations by a metabolic specialist.
Blood can be sent for measurement of the following:
• Markers of the principal regulatory hormones: insulin, growth hormone, and cortisol
• Markers of fatty acid metabolism: ketones (β-hydroxybutyrate and acetoacetate), free fatty acids, and total and free carnitine
• Markers of gluconeogenic pathways: lactate, pyruvate, and alanine
Urine can be sent for measurement of the following:
• Ketones
• Metabolic by-products associated with known causes of hypoglycemia (e.g., organic acids, amino acids)
• Toxicology screen, especially for alcohol and salicylates
Taken together, these tests provide valuable clues as to the cause. For example, low levels of ketones and free fatty acids suggest that fat was not appropriately mobilized. As a consequence, ketones were not formed by the liver. Those biochemical abnormalities are seen in hyperinsulinemic states and can be confirmed by documenting a high level of circulating insulin. Low urinary ketones also suggest an enzymatic defect in fatty acid oxidation.

Josefson J, Zimmerman D: Hypoglycemia in the emergency department, Clin Pediatr Emerg Med 10:285–191, 2009.

79. In patients with acute hypoglycemia, what are the treatment options?
The principal acute treatment is the provision of glucose orally or intravenously. If the patient is alert, 4 to 8 ounces of a sugar-containing liquid (e.g., orange juice, cola) may be given. If the patient is obtunded, intravenous glucose (2 to 3 mL/kg of D10W or 1 mL/kg of D25W) should be administered rapidly. If venous access cannot be achieved promptly, glucose can be provided through a nasogastric tube because glucose is rapidly absorbed. The risk for prolonged hypoglycemia far outweighs the risk associated with the passage of a nasogastric tube in an obtunded patient. Subsequently, the blood sugar should be monitored closely and, if necessary, maintained by the constant infusion of glucose (6 to 8 mg/kg/min).
Ten percent dextrose and water in an electrolyte solution given at about 1.5 times maintenance dose approximates 6 to 8 mg/kg/min. Larger quantities may be necessary, and the blood sugar concentration should be closely followed.
Glucagon promotes glycogen breakdown. In settings in which glycogen stores have not been depleted (e.g., insulin overdose), 0.03 mg/kg (max dose of 1 mg) of glucagon IM or SC will raise blood glucose levels.
Glucocorticoids should not be used routinely. Their only clear indication is in known primary or secondary adrenal insufficiency. In other settings, they have little acute value and may cloud the diagnostic process. The decision to use glucocorticoids is somewhat dependent on the child’s medical history (e.g., reasonable to use in the context of a history of prior central nervous system irradiation).

HYPOTHALAMIC AND PITUITARY DISORDERS
80. What clinical signs or symptoms suggest hypothalamic dysfunction? The signs and symptoms of hypothalamic dysfunction are as variable as the processes controlled by the hypothalamus, ranging from disorders of hormonal production to disturbances of thermoregulation. Precocious or delayed sexual maturation represent the most common presentations of a hypothalamic endocrine abnormality in childhood. Diabetes insipidus, behavioral and cognitive disturbances, and excessive sleepiness are found in about one-third of all patients with hypothalamic dysfunction and may

be the first manifestation of disease. Eating disorders (obesity, anorexia, bulimia) and convulsions are also reported. Dyshidrosis and disturbances of sphincter control (e.g., encopresis, enuresis) are occasionally seen.
81. List the intracranial processes that can interfere with hypothalamic-pituitary function
• Congenital: Inherited deficiencies of gonadotropin-releasing factor, growth hormone–releasing hormone; syndromic (Laurence-Moon-Biedl and Prader-Labhart-Willi syndromes)
• Structural: Craniopharyngioma, Rathke pouch cyst, hemangioma, hamartoma
• Infectious: Meningitis and encephalitis
• Tumors: Glioma, dysgerminoma, ependymoma, Wegener granulomatosis, histiocytosis X
• Idiopathic
82. Why is it bad to have a “Turkish saddle” that is too large?
The sella turcica derives its name from the Latin words for Turkish saddle. The name reflects the anatomic shape of the saddlelike prominence on the upper surface of the sphenoid bone in the middle cranial fossa, above which sits the pituitary gland. A variety of conditions can lead to sellar enlargement, including tumors of the pituitary or functional hypertrophy of the pituitary, which may occur in primary hypothyroidism or primary hypogonadism. Modern imaging techniques have supplanted the skull series as a tool for searching for pituitary or hypothalamic disease; however, an enlarged sella may be noted in children for whom skull series are obtained for other reasons (e.g., head trauma).
83. Which tests are useful for studying suspected hypothalamic and pituitary malfunction? Either MRI or CT is required to rule out structural pathology before searching for functional abnormalities.
Studies of the pituitary-hypothalamus may include any or all of the following:
• Prolactin: Random levels tend to be elevated in the presence of hypothalamic lesions. A normal level does not rule out CNS pathology. An elevated level may occur in an anxious or stressed child during venipuncture.
• Growth hormone production tests (see question 65): These tests are generally indicated only if the child’s growth rate is subnormal. Growth hormone–releasing factor is now available for testing pituitary responsiveness. It has proved useful, in some instances, for delineating pituitary causes of growth hormone underproduction from primary hypothalamic disease.
• Gonadotropin-releasing hormone analogue (GnRHa) provocative test: Random levels of luteinizing hormone and follicle-stimulating hormone are not generally helpful if one is searching for pituitary hypofunction. The results of the GnRHa test must be correlated with the age of the child because there are developmental changes in the response to GnRHa.
• ACTH stimulation testing (Cortrosyn): This test of adrenal production of cortisol is often used in determining whether there has been adrenal destruction or to demonstrate more subtle abnormalities in adrenal steroid hormonogenesis. The hypothalamic-releasing hormone, corticotrophin-releasing factor, is also available and can be used to examine the production of ACTH by the pituitary.
• Simultaneous urine and serum osmolalities: A normal serum osmolality and a concentrated urine osmolality tend to rule out diabetes insipidus. If these results are equivocal, a water deprivation test may be required.
• Thyrotropin-releasing hormone (TRH) is no longer available for provocative testing.

SEXUAL DIFFERENTIATION AND DEVELOPMENT
84. An infant is born with “ambiguous genitalia.” What features of the history and physical examination are key in the evaluation?
Of note, the term ambiguous genitalia is largely antiquated. The contemporary terminology is disorder of sexual differentiation (DSD). This term is thought to more accurately suggest causation rather than consequence and to be less pejorative in discussions with families and nonmedical lay people.
History: One should search for evidence of maternal androgen excess (hirsutism during pregnancy) or androgen ingestion (rare now, but common in the 1960s with certain progestational agents),
other hormonal use (e.g., for infertility or endometriosis), alcohol use, parental consanguinity, previous neonatal deaths, or a family history of previously affected children.

Physical examination: The presence of a gonadal structure in the labioscrotal fold strongly implies the presence of some form of testicular tissue. Gonads containing both ovarian and testicular components (ovotestes) have been found in the inguinal canal. However, it is rare to find an ovary in the inguinal canal. In the absence of a palpable gonad, no conclusions can be drawn regarding probable chromosomal sex. The size of the phallic structure and the location of the urethral meatus provide no information about genetic makeup. However, phallic size and function may be important considerations when determining the sex the child will be assigned.
The presence of midline abnormalities (e.g., cleft palate) suggests hypothalamic or pituitary dysfunction, whereas congenital anomalies such as imperforate anus suggest structural derangements. A digital rectal examination will confirm the patency of the anus and may allow palpation of the uterus. In infants and young children, ultrasound is the more definitive approach to exploring intra-abdominal structures and can often be helpful in confirming the presence or absence of m€ullerian structures
and gonads. Other anomalies should be noted because disorders of genital development are often associated with other developmental disorders in syndromes.

Shomaker K, Bradford K, Key-solle M: Ambiguous genitalia, Contemp Pediatr 26:40–56, 2009. Wolfsdorf J, Padilla A: Goodbye intersex.. .hello DSD, Int Pediatr 23:120–121, 2008.

85. What are the causes of a DSD?
Undervirilized male (XY karyotype):
• Androgen resistance: Complete (testicular feminization)
• Partial defects of androgen synthesis: 3-β-hydroxysteroid dehydrogenase deficiency, 5-α-reductase deficiency
Virilized female (XX karyotype):
• Excess androgen: Congenital adrenal hyperplasia, 21-hydroxylase deficiency, 3-β-hydroxysteroid dehydrogenase deficiency
• Maternal androgen exposure: Medication, virilizing adrenal tumor
Intersex (mosaic karyotypes; e.g., XO/XY) Structural abnormalities

Houk CP, Lee PA: Consensus statement on terminology and management: disorders of sex development, Sex DEV 2:172– 180, 2008.
MacLaughlin DT, Donahoe PK: Sex determination and differentiation, N Engl J Med 350:367–378, 2004.

86. Which studies are essential for the evaluation of a DSD?
• Ultrasonography: This test is the most helpful for identifying internal structures, particularly the uterus and occasionally the ovaries. The absence of a uterus suggests that testes were present early in gestation and produced m€ullerian-inhibiting factor, thereby causing regression of the m€ullerian- derived ducts and thus the uterus. The injection of contrast medium into the urethrovaginal openings will often demonstrate a pouch posterior to the fused labioscrotal folds. Occasionally, the cervix and cervical canal will be highlighted by this study as well.
• Chromosomal analysis: Obviously, this is useful for predicting gonadal content. There are a number of highly specialized and sensitive genetic tests to confirm the presence or absence of X or Y chromosomal material. A geneticist should always be consulted in infants with a DSD.
• Measurement of adrenal steroids (17-hydroxyprogesterone, 11-deoxycortisol,
17-hydroxypregnenolone): 17-Hydroxyprogesterone is the precursor that is elevated in the most common variety of congenital adrenal hyperplasia associated with a DSD (21-hydroxylase deficiency).
• Measurement of testosterone and dihydrotestosterone: It is very important to have input from staff with expertise in this area, including a geneticist, a pediatric endocrinologist, and a pediatric urologist. It is also essential that this group synthesize information after all data are available and that it be communicated to the family by a single spokesperson.

Hiort O, Bimbaum W, Marshall L, et al: Management of disorders of sex development, Nat REV Endocrinol
10:520–529, 2014.
Lee PA, Houk CP, Ahmed SF, et al: Consensus statement on management of intersex disorders, Pediatrics 118: e488–e500, 2006.

87. What major criteria are used to define a micropenis?
To be classified as a micropenis, the phallus must meet two major criteria:
1. The phallus must be normally formed, with the urethral meatus located on the head of the penis and the penis positioned in an appropriate relationship to the scrotum and other pelvic structures. If these features are not present, then the term micropenis should be avoided.
2. The phallus must be more than 2.5 standard deviations below the appropriate mean for age. For a term newborn, this means that a penis less than 2 cm in stretched length is classified as a micropenis.
It is essential that the phallus be measured appropriately. This entails the use of a rigid ruler pressed firmly against the pubic symphysis, depressing the suprapubic fat pad as much as possible. The phallus is grasped gently by its lateral margins and stretched. The measurement is taken along the dorsum of the penis. Note should also be made of the breadth of the phallic shaft. Micropenis must be recognized early in life so that appropriate diagnostic testing can be done.

Lee PA, Mazur T, Danish R, et al: Micropenis. I. Criteria, etiologies and classification, Johns Hopkins Med J
146:156–163, 1980.

88. What are the main concerns to be addressed during the initial evaluation of a 1-month-old infant with micropenis?
1. Is there a defect in the hypothalamic-pituitary-gonadal axis? Specific tests include the measurement of testosterone, dihydrotestosterone, luteinizing hormone, and follicle-stimulating hormone. Because circulating levels of these hormones are normally quite high during the neonatal period, the measurement of random levels during the first 2 months of life may be useful for identifying diseases of the testes and pituitary. Beyond 3 months of age, the tests are generally not useful because the entire axis becomes quiescent and remains so until late childhood. Depending on the patient’s age, provocative tests may be necessary, including
the following: (1) repetitive testosterone injection to evaluate the ability of the penis to respond to hormonal stimulation; (2) the use of human chorionic gonadotropin as a stimulus for testosterone production by the testes; and (3) leuprolide administration to examine the responsiveness of the pituitary to stimulation. The trial of testosterone therapy is especially important because it indicates whether phallic growth is possible. If it is not, gender reassignment may become a consideration.
2. Does a possible pituitary deficiency involve other hormones? Isolated growth hormone deficiency, gonadotropin deficiency, and panhypopituitarism have been associated with micropenis. The presence of hypoglycemia, hypothermia, or hyperbilirubinemia (e.g., associated with hypothyroidism) in a child with micropenis should lead one to search for other pituitary hormone deficits and structural abnormalities of the CNS (e.g., septo-optic dysplasia).
3. Is there a renal abnormality? Because of the association of genital and renal abnormalities, it is prudent to obtain an abdominal and pelvic ultrasound to better define the internal anatomy.
89. Discuss the terms that denote aspects of precocious sexual development.
The terms used to describe precocious puberty reflect the fact that normal puberty is an orderly process by which female children are feminized and male children masculinized. The development of breast tissue without pubic hair is called premature thelarche. If pubic hair subsequently develops, the term precocious puberty is used. If pubic hair develops without breast tissue, it is premature pubarche. Because pubic hair development in the female is thought to be the result of adrenal androgens, the term premature adrenarche is commonly used. If the pubertal changes are early and appear to proceed
in the orderly fashion of breast budding, pubic hair development, growth spurt, and, finally, menstruation, the term true precocious puberty is used. When some of the changes of puberty are present, but their appearance is isolated or out of normal sequence (e.g., menses without breast development), the term pseudoprecocious puberty is used.
90. Boys or girls: Who is more likely to have an identifiable cause for precocious puberty?
Although precocious puberty occurs much more frequently in girls (80% of cases are girls), boys are more likely to have identifiable pathology. As a general rule, the younger the child and the more rapid the onset of the condition, the greater the likelihood of detecting pathology.

91. A 7 ½ -year-old girl develops breast buds and pubic hair. Is this normal or precocious?
Precocious puberty is the appearance of physical changes associated with sexual development earlier than normal. Traditionally this has been the development of feminine characteristics in girls who are younger than 8 years and masculine characteristics in boys who are younger than 9 years. In 1997, an office-based study of 17,000 healthy 3- to 12-year-old girls revealed that puberty was occurring on average 1 year earlier in white girls and 2 years earlier in black girls and suggested a revision of guidelines for the ages at which precocious puberty should be investigated. Many experts now recommend that an evaluation for precocious puberty of girls need not be undertaken for white girls older than 7 years or black girls older than 6 years with breast and/or pubic hair development. However, this remains controversial and a subject of ongoing debate and data collection. The recommendations for boys remain that investigations for pathologic etiologies be undertaken if pubertal changes begin before the age of 9 years.

Euling SY, Herman-Giddens ME, Lee PA, et al: Examination of US puberty-timing data from 1940 to 1994 for secular trends: panel findings, Pediatrics 121(Suppl):S172–S191, 2008.
Kaplowitz PB, Oberfield SE: Reexamination of the age limit for defining when puberty is precocious in girls in the United States: implications for evaluation and treatment, Pediatrics 104:936–941, 1999.
Herman-Giddens ME, Slora EJ, Wasserman RC, et al: Secondary sexual characteristics and menses in young girls seen in office practice: a study from the Pediatric Research in Office Settings network, Pediatrics 99:505–512, 1997.

92. Breast buds are noted on a 2-year-old girl. Is this worrisome? Premature thelarche, or the development of breast buds, is the most common variation of normal pubertal development. A form of mild estrogenization, it typically occurs between the ages of 1 and 3 years. It is usually benign and should not be associated with the onset of other pubertal events. Precocious puberty, rather than simple premature thelarche, should be suspected if the following occur:
• Breast, nipple, and areolar development reach Tanner stage III (i.e., continued progression is of concern).
• Androgenization with pubic and/or axillary hair begins.
• Linear growth accelerates.
Ongoing parental observation and periodic reexamination are all that are required if there are no signs of progression.
93. Which aspects of the physical examination are particularly important when evaluating a patient with precocious puberty?
• Evidence of a CNS mass: Examination of optic fundus for possible increased intracranial pressure; visual field testing for evidence of optic nerve compression by a hypothalamic or pituitary mass
• Evidence of androgenic influence: Presence of acne and facial and axillary hair; increased muscle bulk and definition; extent of other body or pubic hair; in boys, increased scrotal rugation accompanied by thinning and pigmentation and penile elongation; in girls, clitoromegaly
• Evidence of estrogenic influence: Size of breast tissue and nipple and areolar contouring; vaginal mucosa color (increased estrogen causes cornification of vaginal epithelium with a color change from prepubertal shiny red to a more opalescent pink); labia minor (become more prominent and visible between the labia majora as puberty progresses)
• Evidence of gonadotropic stimulation: Testicular enlargement of greater than 2.5 cm in length or more than 4 mL in volume (preferably measured using a Prader orchidometer of labeled volumetric beads); pubertal development without testicular enlargement usually suggests adrenal pathology
• Evidence of other mass: Asymmetrical testicular enlargement; hepatomegaly; abdominal mass
94. Which radiologic and laboratory tests are indicated for the evaluation of precocious puberty?
Radiologic evaluation
• Bone age: This study helps determine the duration of exposure to the elevated sex hormone. A significantly advanced bone age compared with the chronologic age suggests long-term exposure.
• Abdominal and PELVIC ultrasound: In boys, this test identifies possible adrenal or hepatic masses; in girls, it identifies adrenal masses, ovarian masses, or cysts. Increased uterine size and echogenicity suggest endometrial proliferation in response to circulating estrogen.
• Head CT or MRI: This evaluation is useful in identifying pituitary or hypothalamic abnormalities.
Laboratory evaluation
• Obtain luteinizing hormone, follicle-stimulating hormone, estradiol, and testosterone levels.

• Adrenal steroid levels (17-hydroxyprogesterone, androstenedione, cortisol): More extensive testing may be needed in a virilized child if the initial studies are normal.
• Use provocative testing of the hypothalamic-pituitary axis (using a synthetic GnRHa) or of the adrenal gland (using a synthetic ACTH), especially in the child with slight but progressive pubertal changes.
95. When does the male voice begin to crack?
Voice “breaking” has traditionally been regarded as one of the harbingers of puberty. However, sequential voice analysis reveals that it is usually a late event in puberty, typically occurring between Tanner stages III and IV.

Harries ML, Walker JM, Williams DM, et al: Changes in the male voice at puberty, Arch Dis Child 77:445–447, 1997.

THYROID DISORDERS
96. Which thyroid function tests are “standard”?
Diseases of the thyroid represent a heterogeneous group of disorders. As such, there are no “standard” thyroid function studies that are appropriate for all children with suspected thyroid disease. The choice of laboratory tests is based on the results of a careful history and physical examination.
Biochemical findings that suggest hyperthyroidism: A thyroid-stimulating hormone (TSH) level and a thyroxine level (total T4 or free T4) should be obtained. Compared to total T4, the free T4 is the biologically active component and theoretically is a better measure of thyroid function. TSH suppression is probably the most sensitive indicator of hyperthyroid status. If the patient is symptomatic and has a suppressed TSH level with a normal T4 level, it will be necessary to obtain a triiodothyronine (T3) radioimmunoassay because cases of T3 thyrotoxicosis do occur. If the patient is asymptomatic but has an elevated T4 level, some measure of binding capacity should be obtained (e.g., a T3 uptake).
Biochemical findings that suggest hypothyroidism: The laboratory evaluation consists of the quantitation of T4 (total T4 or free T4) and TSH. A low T4 level and an elevated TSH level are diagnostic of hypothyroidism.
97. What signs and symptoms in an infant suggest congenital hypothyroidism?
See Table 6-6.

Table 6-6. Symptoms and Signs of Hypothyroidism in Infancy
SYMPTOMS SIGNS
Lethargy Hypotonia, slow reflexes
Poor feeding Jaundice (prolonged)
Constipation Mottling
Poor weight gain Distended abdomen
Cold extremities Acrocyanosis
Hoarse cry Coarse features
Large fontanels, wide sutures

98. What causes congenital hypothyroidism?
• Primary: Agenesis or dysgenesis, ectopic, dyshormonogenesis
• Secondary: Hypopituitarism, hypothalamic abnormality
• Other: Transient, maternal factors (e.g., goitrogen ingestion, iodide deficiency)
The most common cause of permanent primary congenital hypothyroidism is thyroid dysgenesis, or failure of the gland to develop properly. Ectopic thyroid gland location accounts for two-thirds of thyroid dysgenesis followed by aplasia or hypoplastic gland. The second most common cause is thyroid dyshormonogenesis, which is a defect in thyroid hormone production. Thyroid dysgenesis accounts for 85% of permanent primary congenital hypothyroidism; inborn errors of thyroid hormone biosynthesis comprise 10% to 15% of cases.

LaFranchi S: Approach to diagnosis and treatment of neonatal hypothyroidism, J Clin Endocrinol Metab 96:2959– 2967, 2011.

99. How effective are screening programs for congenital hypothyroidism?
Screening programs correctly identify 90% to 95% of children who are affected with congenital hypothyroidism. Screening programs are most likely to miss infants with large ectopic glands, those with partial defects in thyroidal hormone biosynthesis, and those with secondary (pituitary or hypothalamic) disease. If an infant presents with a clinical picture of hypothyroidism and has had a normal newborn screen, it is important to realize that the false-negative rate of the screening is up to 10%.

Gr€uters A, Krude H: Detection and treatment of congenital hypothyroidism, Nat REV Endocrinol 8:104–113, 2011.

100. Discuss the risks of delaying treatment for congenital hypothyroidism.
Therapy should begin as early as possible because outcome is related to the time treatment is started. Because less than 20% of patients will have distinctive clinical signs at 3 to 4 weeks of age, screening is now performed on all newborns in the United States at 2 to 3 days of age, and most affected children are started on therapy before they are 1 month old. Many pediatricians and screening programs undertake a second screen at 2 weeks of age to ensure that children with treatable conditions are not missed. The prognosis for intellectual development is directly related to the amount of time from birth to the initiation of therapy, and there is an inverse relationship between age of diagnosis/treatment and intelligence quotient (IQ). In a literature review of 11 studies that evaluated starting treatment at an earlier age (12 to 30 days of life)
compared with a later age (>30 days of life), infants started at an earlier age averaged 15.7 IQ points higher than infants started at a later age.

LaFranchi SH, Austin J: How should we be treating children with congenital hypothyroidism? J Pediatr Endocrinol 2007, (5):559–578.

101. A suspected goiter (diffuse enlargement of the thyroid gland) is noted during a routine examination of an asymptomatic 7-year-old boy. What should be the course of action?
The evaluation of a child with goiter (Fig. 6-4) is generally straightforward. In the absence of signs of thyroidal disease, history should be obtained regarding recent exposure to iodine or other halogens. A family history should be obtained regarding thyroidal disease because thyroiditis tends to run in families. The initial laboratory evaluation typically includes T4, TSH, and antithyroidal antibodies. If there is discrete nodularity within the thyroid or the gland is rock hard or tender, further diagnostic evaluation (ultrasound, CT) may be indicated. Parathyroid enlargement or lymphoma may be misdiagnosed as goiter.

Figure 6-4. Goiter. Note the enlarged thyroid gland in a patient with Hashimoto thyroiditis, easily visualized with neck extension.
(From Zitelli BJ, McIntire SC, Nowalk AJ: Atlas of Pediatric Physical Diagnosis, ed 6. Philadelphia, 2012, Saunders, pp 369–400.)

102. What is the most common cause of acquired hypothyroidism in childhood? The most common cause is chronic lymphocytic thyroiditis, also called Hashimoto thyroiditis or autoimmune thyroiditis and usually occurs early to mid puberty. Its incidence during adolescence is about 1% to 2%. The female-to-male ratio is 2:1.

Counts D, Varma SK: Hypothyroidism in children, Pediatr REV 30:251–257, 2009.

103. What is the most common clinical presentation of Hashimoto thyroiditis? Although symptoms of hypothyroidism or hyperthyroidism may be present, most pediatric patients are asymptomatic, and the condition is detected by the presence of goiter. The diagnosis of Hashimoto thyroiditis is primarily based on the demonstration of antithyroidal antibodies. Clinical manifestations may include a linear growth decline, fatigue, constipation, poor school performance, irregular menstrual periods, and cold intolerance.

Counts D, Varma SK: Hypothyroidism in children, Pediatr REV 30:251–257, 2009. Pearce EN, Farwell AP, Braverman LE: Thyroiditis, N Engl J Med 348:2646–2655, 2003.

104. What should a parent be told about the prognosis of a child who has euthyroid goiter caused by chronic lymphocytic thyroiditis?
About 50% of all children who present with symptoms of euthyroid goiter will have resolution of the goiter over several years, regardless of whether thyroxine replacement is given. It is difficult to predict which children will recover completely, which will remain euthyroid with goiter, and
which will become hypothyroid. Large goiters and increased thyroglobulin at presentation, together with an increase in thyroid peroxidase antibody and TSH levels over time, are the most significant predictors for the development of hypothyroidism. Any child identified with thyroid disease should have T4 and TSH values monitored every 4 to 6 months.

Radetti G: The natural history of euthyroid Hashimoto’s thyroiditis in children, J Pediatr 149:827–832, 2006.

105. What other autoimmune diseases are associated with chronic lymphocytic thyroiditis?
Adrenal insufficiency, diabetes mellitus, juvenile idiopathic arthritis, systemic lupus erythematosus, rheumatoid arthritis, myasthenia gravis, idiopathic thrombocytopenia purpura, and autoimmune polyendocrine syndrome (type II)

106. What does a normal T4 and an elevated TSH suggest?
The diagnosis of hypothyroidism is based on finding both a low T4 level and an elevated TSH level. However, on occasion, the T4 level can be maintained in a normal range by increased stimulation of the thyroid gland by TSH. This combination of laboratory values is suggestive of a failing thyroid and is referred to as compensated hypothyroidism. Because TSH is the most useful physiologic
marker for the adequacy of a circulating level of thyroid hormone, an elevated TSH level is an indication for thyroid replacement therapy. If the TSH level is only minimally elevated and the child is asymptomatic, it is worthwhile to wait 4 to 6 weeks and repeat the T4 and TSH tests before instituting therapy.

107. What is the most common cause of hyperthyroidism in children?
More than 95% of hyperthyroidism cases are due to Graves disease, a multisystem disease that is characterized by hyperthyroidism; infiltrative ophthalmopathy; and occasionally, an infiltrative dermopathy. The features of this disease may occur singly or in any combination. In children, the ophthalmopathy appears to be less severe, and the dermopathy is rare; the full syndrome may never develop. There has been a tendency to use the terms GRAVES disease, thyrotoxicosis, and hyperthyroidism interchangeably, but there are other causes of hyperthyroidism in childhood
(e.g., factitious).

Léger J: Graves’ disease in children, Endocr DEV 26:171–182, 2014.
Brown RS. Autoimmune thyroid disease: unlocking a complex puzzle. Curr Opin Pediatr 2009:21(4):523–528.

108. In addition to Graves disease, what conditions may cause hyperthyroidism?
• Excess TSH: TSH-producing tumor (these are extraordinarily rare in children)
• Thyroid autonomy: Adenoma, multinodular goiter, activating mutations of G proteins (e.g., McCune-Albright syndrome)
• Thyroid inflammation: Subacute thyroiditis, Hashimoto thyroiditis
• Exogenous hormone: Medication, ectopic production
109. Describe the typical features of hyperthyroidism that occur as a result of Graves disease.
• History: The onset of symptoms is usually gradual, with increasing emotional lability, shortened attention span, and deteriorating school performance. Sleep disturbance, nervousness, headache, and weight loss despite increased appetite may be noted, as may easy fatigability and heat intolerance. Observation of the child’s behavior while the history is being obtained from the parent is often instructive.
• Physical examination: Weight may be low for height, and many children will be tall for age and genetic potential. Some children will have an acceleration in growth rate at the same time that their behavior begins to deteriorate. The pulse rate is usually inappropriately high for age. A widened pulse pressure or an elevated blood pressure is often noted, although this is a more variable finding in children than in adults.
110. What causes Graves disease?
Graves disease is an autoimmune disorder in which TSH receptor antibodies bind to the TSH receptor, thereby resulting in the stimulation of thyroid hormone production and subsequent hyperthyroidism. Most thyroid receptor antibodies belong to the IgG class. The general name used for these antibodies is human thyroid-stimulating immunoglobulins (HTSI or TSI). These were formerly called long-acting thyroid stimulators (LATS).

KEY POINTS: THYROID DISORDERS
1. Midline neck masses usually involve the thyroid gland or thyroid remnants, such as a thyroglossal duct cyst.
2. Neck extension improves visualization and palpation of thyroid masses, especially with swallowing.
3. About 20% to 40% of solitary thyroid nodules in adolescents are malignant; expedited evaluation is needed.
4. Chronic lymphocytic thyroiditis is the most common cause of pediatric goiter in the United States.
5. Chronic lymphocytic thyroiditis most commonly appears as an asymptomatic goiter, thereby reinforcing the need for thyroid palpation (an often overlooked examination feature).
6. The best initial screening studies for hypothyroidism and hyperthyroidism are total T4 and thyroid- stimulating hormone.

111. Why does exophthalmos occur in Graves disease?
Exophthalmos is a bulging of the eye anteriorly (Fig. 6-5). The reason is unknown, but several facts suggest an autoimmune process:
• Histologic studies reveal lymphocytic infiltration of the retrobulbar muscles.
• Circulating lymphocytes are sensitized to an antigen that is unique to the retrobulbar tissues.
• The thyroglobulin-antithyroglobulin antibody complexes found in patients with Graves disease bind specifically to the extraorbital muscles. There may be a separate class of antibodies that is responsible for changes in the retrobulbar muscles.

Bahn RS: Graves’ ophthalmopathy, New Engl J Med 362:726–738, 2010.

112. What treatment options are available for children with Graves disease?
The three types of therapy are antithyroid medication, radioactive (131I) ablation, and subtotal thyroidectomy.

Léger J: Graves’ disease in children, Endocr DEV 26:171–182, 2014.
Bauer AJ: Approach to the pediatric patient with Graves’ disease: when is definitive therapy warranted? J Clin Endocrinol Metab 96:580–588, 2011.

Figure 6-5. Exophthalmos in a young woman with Graves disease. (From Moshang T Jr: Pediatric Endocrinology: The Requisites in Pediatrics. Philadelphia, 2005, ELSEVIER Mosby, p 7.)

113. Describe the principal modes of actions and the side effects of medications used to treat Graves disease.
The thioamide derivatives—propylthiouracil and methimazole—have historically been the keystones of long-term management. However, their effective onset of action is slow because they block the synthesis but not the release of thyroid hormone. Propranolol is useful for treating many of the
β-adrenergic effects of hyperthyroidism. It is used during the acute management of Graves disease but should be discontinued when the thyroid disease is controlled. Iodide (which can transiently block thyroid hormone release) and glucocorticoids are useful stopgap medications while awaiting the inhibitory effects of the thioamide; they are generally used only when the patient is acutely symptomatic (i.e., thyroid storm).
The thioamides are associated with side effects, the most serious of which have been a lupuslike syndrome involving the lungs or liver, thrombocytopenia, neutropenia, agranulocytosis, and hepatitis with elevated transaminase levels. The association of propylthiouracil and severe liver failure led to the issuing of a black box warning from the Food and Drug Administration in 2010, with methimazole now the preferred antithyroid medication option.

Rivkees SA: Pediatric Graves’ disease: management in the post-propylthiouracil era, Int J Pediatr Endocrinol
2014:10, 2014.

114. Has radioactive iodide fallen into disfavor as a treatment option for Graves disease?
On the contrary, radioactive iodide (131I) is increasing in popularity. In some pediatric endocrinology centers, this is now considered the first line of therapy. Concern had been voiced about the possible risk

for thyroid carcinoma, leukemia, thyroid nodules, or genetic mutations, but as the individuals treated with 131I during childhood have been followed for prolonged periods, experience suggests that children are not at a significantly increased risk for developing these conditions.

Rivkees SA: Pediatric Graves’ disease: management in the post-propylthiouracil era, Int J Pediatr Endocrinol
2014:10, 2014.

115. During a routine physical examination, a solitary thyroid nodule is palpated on an asymptomatic 10-year-old child. Can a wait-and-see approach
be taken?
Absolutely not. In children with a solitary nodule, about 20% to 40% have a carcinoma, 20% to 30% have an adenoma, and the remainder will have thyroid abscess, thyroid cyst, multinodular goiter, Hashimoto thyroiditis, subacute thyroiditis, or nonthyroidal neck mass. Given the relatively high incidence of carcinoma, a thyroidal mass demands prompt evaluation. Previous irradiation to the head or neck is associated with a significantly increased incidence of thyroid carcinoma. A family history of thyroid disease increases the likelihood of chronic lymphocytic thyroiditis or Graves disease. The presence of tenderness on palpation or high titers of antithyroid antibodies points away from a malignant process. However, in all cases, radiologic studies should be undertaken; in many cases, surgical exploration is required.

116. How should this solitary thyroid nodule be investigated?
The principal tools used in the investigation of a thyroid mass include 123I scanning and ultrasound. Ultrasound is useful for delineating the size of the mass, its anatomic relationship to the rest of the thyroid, and the presence of cystic structures. 123I imaging that reveals a single nonfunctioning mass (“cold” nodule) suggests a carcinoma or adenoma and is a clear indication for surgery. Patchy uptake is more characteristic of chronic lymphocytic thyroiditis, whereas a poorly functioning lobe may be found in a subacute thyroiditis. Fine-needle aspiration or biopsy is another approach to the investigation of a thyroid mass with current recommendations for aspiration of thyroid nodules 1 cm with ultrasound guidance improving diagnostic accuracy.

Gupta A, Ly S: A standardized assessment of thyroid nodules in children confirms higher cancer prevalence than in adults,
J Clin Endocrinol Metab 98:3238–3245, 2013.
Mehanna HM, Jain A, Morton RP, et al: Investigating the thyroid nodule, BMJ 338:705–709, 2009.

117. How is the euthyroid sick syndrome diagnosed?
Euthyroid sick syndrome, also called nonthyroidal illness syndrome, is an adaptive response to slow body metabolism often seen in critical illness. It is also called the low T3 syndrome because the most consistent finding is a depression of serum T3. Reverse T3, a metabolically inactive metabolite, is increased, although this is rarely measured. T4 and thyroid binding globulin levels may be low or normal; free T4 and TSH levels are normal. In sick preterm infants, the clinical picture is often confusing because levels of T4, free T4, and T3 are naturally low. Infants and children with the euthyroid sick syndrome generally revert to normal as the primary illness resolves.

Marks SD: Nonthyroidal illness syndrome in children, Endocrine 36:355–367, 2009.

Acknowledgment
We would like to thank Dr. Daniel E. Hale for his significant contributions to this text as one of the original authors of the chapter.

BONUS QUESTIONS
118. How do antiepileptic medications affect thyroid function tests? Cytochrome P450 complexes consist of enzymes responsible for oxidative and reducing reactions. Some of these enzymes are induced by antiepileptic drugs such as phenytoin, phenobarbital, and carbamazepine that can produce marked reductions in thyroid hormone levels through the induction of hepatic drug metabolism.
In addition, medications that bind to albumin, such as phenytoin (Dilantin), can cause changes in thyroid assays because some will displace T4. Thus, the total T4 will be low. However, this confusion has been ameliorated somewhat by the wide availability of reliable free T4 assays.

Thalmann, S, Meier CA: Effects of drugs on TSH secretion, thyroid hormones absorption, synthesis, metabolism, and action. In Braverman LE, Cooper DS, editors: Werner & Ingbar’s the Thyroid: A Fundamental and Clinical Text, ed 10. Philadelphia, 2012, Lippincott Williams & Wilkins, p 193.

119. Of what value is the T3 resin uptake (T3RU) test?
The T3RU test is a measure of serum thyroid-binding capacity. Because T4 is primarily protein bound, only a small amount exists in the unbound (free) state. Physiologically, the free T4 is the metabolically active compound, but historically it was technically complex to assay directly. For many years, the primary strategy for determining the free T4 was simply to measure T3RU and total T4 and calculate the unbound T4 (free T4). Contemporary assays for free T4 are fairly consistent and reliable. As a consequence, T3RU is less frequently measured. In patients with primary thyroidal disease, the T3RU and the T4 should go in the same direction (i.e., both increase or both decrease). If they go in opposite directions, it is probably a binding problem.
120. How common is goiter (thyroid enlargement) in newborns with congenital hypothyroidism?
Congenital goiter is rare in newborns with congenital hypothyroidism. Maternal ingestion of antithyroid medications, iodides, and goitrogens; congenital thyroid dyshormonogenic defects; and congenital hyperthyroidism are associated with palpable thyromegaly. Goiter in the newborn is difficult to recognize because of the infant’s relatively short neck and increased subcutaneous fat. Palpation of the neck is often overlooked during newborn examinations. On occasion, there may be sufficient posterior extension of the goiter to cause airway obstruction in infants, especially in a mother on goitrogens (e.g., thioamides).