4.6 Hypercalcaemic and hypocalcaemic syndromes in children

The calcium-regulating system employs an intricate network of homoeostatic signals and targets in order to meet the body’s mineral demands. Mineral requirements vary considerably throughout progressive stages of development, in large part reflecting the changing mineral demands of skeletal growth, and representing characteristic features of the calcium homoeostatic system during childhood years. As a consequence, this system must be adaptable to the wide-ranging mineral demands occurring throughout the life cycle. Furthermore, the numerous factors involved in calcium homoeostasis allow for compensatory mechanisms to limit the severity of disease when an isolated insult occurs to the system. Indeed, many heritable disorders of mineral homoeostasis become evident in early childhood and are best recognized when viewed in the light of mineral requirements during infancy and childhood. As understanding of the relevant physiology is central to formulating approaches to management of such problems, we review these disorders in the context of physiology specific to childhood to provide the basis for understanding hypocalcaemia and hypercalcaemia in this age group.

The growing fetus must be supplied with sufficient calcium for the formation and growth of a mineralizing skeleton. In addition, the physiological milieu of the fetus must be maintained in an environment appropriate for normal cellular function. Thus adequate extracellular calcium must be provided for normal function of the clotting factors, and avoidance of neuromuscular hyperexcitation. Yet, at the same time, the supply must be appropriately limited to prevent damaging soft tissue calcification or other toxicity to the developing fetus. A critical calcium-dependent process in fetal life is skeletal development. Most of the skeleton is formed by the complex process referred to as endochondral ossification (1). Cartilage templates are organized in concert with the transition of undifferentiated mesenchymal cells to differentiated chondrocytes. The cartilage templates serve as a nidus for eventual development into the skeleton. A system of chondrocyte maturation and proliferation occurs at what will become the ends of long bones, allowing for the continued linear growth of the skeleton. Regulation of this early formative process is dependent upon a variety of local and systemic factors, such as insulin-like growth factors, fibroblast growth factors, parathyroid hormone-related protein (PTHrP), and Indian hedgehog protein (2). Once mature cartilage forms, chondrocytes hypertrophy, and blood vessels penetrate the region, with the appearance of marrow stroma and osteoblasts soon to follow. Mineralization of the newly established skeleton begins, and growth results in a continuing mineral demand in order to effectively mineralize the newly formed tissue. Indeed, the fetus has substantial mineral demands: approximately 21 g of calcium accumulate in the human through a term gestation, and accretion of more than three-quarters of this amount occurs in the third trimester (3). Calcium supply from the maternal circulation must be regulated by specific mechanisms in order to meet these demands throughout the later weeks of gestation.

The maternal circulation is the source of calcium provided to the fetus. An abundance of calcium occurs in the mother primarily as a result of a pregnancy-induced doubling of maternal circulating 1,25-dihydroxyvitamin D (1,25 (OH)₂D) levels, which in turn increases fractional absorption of calcium at the intestine (4). This occurs with no significant increases in circulating levels of parathyroid hormone (PTH) in the mother.

The placenta is the site of transfer of nutrients from the maternal circulation to the fetus. Calcium may be transported by several mechanisms across the placenta; the dominant direction of flow is from maternal to the fetal circulation, requiring active transport. A Ca²⁺-ATPase located in the fetus-directed basement membrane of the syncytial trophoblast cells appears to mediate this important function (5). Although it is not clear exactly when in gestation active calcium transport begins, it is present by the beginning of the third trimester. The fetal circulating calcium level is maintained at a slightly higher concentration than the maternal circulation. Active placental calcium transfer plays an important role in determining fetal circulating calcium level, but other factors may play a role, including PTH and PTHrP. The relative hypercalcaemia in the fetus is ample for normal skeletal growth and development. Placental calcium transfer seems to be mainly regulated by PTHrP, and to a lesser extent PTH (5–7). The mid- and C-terminal portions of the PTHrP molecule are required, whereas the N-terminus (most related to PTH in sequence) does not have activity in this regard (3, 5). PTHrP plays an important role in embryonic growth and development of many tissues, and is produced by multiple tissues. Major sources of PTHrP production are the placenta, and to a lesser extent, the parathyroid glands. In a fetal mouse model, disruption of PTHrP results in hypocalcaemia and severe chondrodysplasia (7, 8). Fetal circulating PTH levels are low, probably due to Ca-sensing receptor (CaSR) mediated suppression of PTH secretion by fetal parathyroid glands (5). Nevertheless, aparathyroid fetal mice develop hypocalcaemia and defective bone mineralization, pointing towards a role for PTH in maintaining normal serum calcium, and thereby perhaps supporting normal bone mineralization (9). PTH may also exert its effect on bone formation, to some extent, via direct interaction with osteoblasts.

Fetal circulating 1,25 (OH)₂D levels are low. This may be related to low PTH and high serum phosphorus levels. 1,25 (OH)₂D does not play a major role in placental calcium transfer or maintenance of serum calcium level as evidenced by the fully mineralized skeleton and normal fetal serum calcium levels at term in vitamin D receptor-null (Vdr-null) mice. In human cases of maternal vitamin D deficiency, skeletal mineralization seems to be unaffected but the newborn will be at risk of developing hypocalcaemia (5) The presence of CaSR in both human and murine placenta suggests a possible role for this membrane receptor in fetal calcium homoeostasis. Some insight has been provided by CaSR knockout mice: fetuses of this strain demonstrate increased PTH levels, reduced placental calcium transport, increased amniotic fluid calcium, and increased markers of bone resorption. This constellation of findings suggests that an increase in resorption of the skeleton can occur, when inadequate calcium levels are sensed by the fetus (9, 10). The fetal kidneys and skeleton may be involved in regulation of fetal calcium levels as well, but their roles are less defined. Excreted calcium is not lost from the fetal unit as it remains in the amniotic fluid.

With birth the supply of maternal calcium is abruptly withdrawn from the fetus, as well as any placental sources of PTHrP and 1,25 (OH)₂D. A resultant acute decrease in serum calcium of approximately 1 mg/dl occurs in term infants, and slightly more in preterm infants. One study indicates that the decrement in serum calcium and rise in PTH is greater in babies born by caesarean section than in babies born spontaneously by the vaginal route. This decrease in calcium then stimulates secretion of PTH, suppressed during fetal life, which in turn, stimulates the kidney to generate adult normal levels of 1,25 (OH)₂D within the next several days. Levels of PTHrP are reduced; this hormone probably plays a lesser role in postnatal calcium homoeostasis than in utero. The serum calcium gradually increases to normal childhood levels within a few days of the acute postnatal decrement.

The intestine and kidney assume major roles in mineral homoeostasis with this transition. The neonatal skeleton continues to accrue calcium at rates close to that attained in late gestation (averaging 100–150 mg/kg per day). Thus the newborn infant becomes critically dependent upon its nutritional environment for nonmaternal sources of calcium. Renal excretion of calcium increases over the first few weeks of life, as glomerular filtration rate (GFR) increases. As the kidney matures, it begins to play a minor role in regulation of calcium. The newborn infant, however, becomes primarily dependent upon the intestine to maintain its calcium supply. In the first few days to weeks of neonatal life, passive or facilitated calcium transport (not vitamin D-mediated mechanisms) are the dominant means by which calcium is brought into the body. After several weeks, vitamin D appears to be useful in enhancing calcium absorption in term infants. Fractional calcium absorption can be relatively high in infancy, particularly in very low birthweight children, who may develop hypercalcaemia during high calcium intake, as may occur with the administration of breast-milk fortifiers. This phenomenon may occur independently of vitamin D status (with normal circulating levels of 25 (OH)D, and appropriately low circulating PTH and 1,25 (OH)₂D), implying that passive or facilitated, nonvitamin D mediated calcium transport in the immature intestine can be remarkably efficient.

Skeletal growth and mineralization continue at a very rapid pace throughout the first 2 years of life. The growth velocity on average during the first 4 months of life can be annualized to approximately 28 cm/year. From that time on, a child’s growth rate asymptotically decreases from a rate of 1 cm/month (approximately 12.5 cm/year) to about 5–6 cm/year at the time of the pubertal growth spurt, when a rate of about 10 cm/year is transiently achieved prior to the cessation of growth. This linear growth represents the growth of the appendicular skeleton, which must be adequately mineralized; thus rapid growth in infancy places considerable mineral demands on the skeleton. Bone mineral content and areal bone mineral density, as assessed by standard two-dimensional techniques, such as dual energy X-ray absorptiometry proceeds at a steady pace until approximately age 11 in girls and slightly later in boys (11). Specific guidelines for the use and interpretation of bone densitometry in children have recently been published by the International Society of Clinical Densitometry (12).

Although the focus on bone activity during these years is primarily on formation and mineralization, there must also be extremely active turnover in general. The growing bone must be constantly modelled in order to maintain an appropriate structure. Weight-bearing forces begin to correct the physiological bow of childhood, as lower extremity alignment becomes more linear. As metaphyseal long bone segments accrue mineral at growth plate cartilage, extending the length of the long bone calls for an eventual narrowing of the former metaphyseal segment as it assumes a diaphyseal position. These processes require extensive bone resorptive activity. Thus, when investigating disorders of the bone and mineral system, one must recognize this relatively hyperdynamic state of bone turnover. None of the established normal ranges of bone activity apply, and the remarkably high numbers (by adult standards) can be the norm (13). In fact, the normal range of values for such markers as serum osteocalcin, alkaline phosphatase activity, or urinary excretion of deoxypyridinoline cross-links of collagen, or the N-telopeptide of type I collagen are quite wide (Table 4.6.1). Several investigators have compiled normative data on these and other biomarkers of bone turnover throughout childhood and/or adolescent age groups (14). There is a consistent peak in the concentrations of most serum markers of bone formation and resorption in adolescence, and this rise occurs approximately 2.5 years earlier in girls than in boys (14). Values in adolescence for the more widely used markers are shown in Table 4.6.1. Although studies are limited, there appears to be less variation by age with serum tartrate-resistant acid phosphatase (17).

In addition to the rapid growth spurt beginning in early puberty in girls, and later stages of puberty in boys, bone mineral density accrues at an accelerated pace. The rate of increase in bone mineral density in girls between the ages of 11 and 16 years is more rapid than at any time in late childhood or during adult life. The National Academy of Sciences, USA, has set ‘adequate intake’ levels for calcium by age ranges. These levels are 210 mg/day through the first 6 months of life, 270 mg/day for months 6–12, 500 mg/day from years 1–3, and 800 mg/day for ages 4–8. In keeping with the rapid rate of bone accretion in adolescence, calcium ‘adequate intake’ has been set at 1300 mg/day for the 9–18 year-old group (18). Some have thought this number underestimates calcium requirements and have suggested that teenage girls consume 1500 mg of calcium daily.

Commensurate with the pubertal growth spurt are transient rises in the markers of bone formative activity, serum osteocalcin and alkaline phosphatase activity. The bone resorptive markers, which decrease somewhat throughout later childhood, decrease substantially in late puberty (Tanner stages IV and V), reflecting more quiescent bone turnover than in earlier childhood, as described above and in Table 4.6.1. The postpubertal period of elevation in turnover markers persists in males longer than in females, suggesting a longer period in young men of active mineral accrual than in young women.

In addition to the changes in bone markers, geometric properties of long bones change during puberty, and appear to differ between boys and girls. These changes may be reflected in the differential changes in biomarker levels described above. However, the finding of wider long bones of males remains largely unexplained. Male long bones progressively grow in circumferential diameter beyond female growth in this regard, in part due to a prolonged period of generalized prepubertal growth (19). Furthermore, recent data suggests that such sex differences in geometry are evident in prepubertal years, determined by complex genetic traits and environmental stimuli (19).

Appropriate diagnosis of disease or monitoring of therapy require an understanding of changes in the biochemical parameters used to facilitate an evaluation of mineral metabolism in children. Figure 4.6.1 illustrates the changes in circulating minerals and related hormones during early infancy. The serum levels of calcium and magnesium do not change significantly after the first few days of life until adulthood. On the other hand urinary excretion of calcium is much greater in infancy than in later childhood and adulthood. A convenient measure for urinary excretion of calcium is the ratio of calcium to creatinine (Ca/Cr) in a random urine sample. Urinary calcium excretion varies with type of feedings, vitamin D nutrition, and gestational age (20). In the older child a fasting urine sample should have a Ca/Cr less than 0.21 (mg/mg). A 24-h urine collection should be confirmed by measurement of total creatinine (which should be 10–20 mg/kg per 24 h in most children), and the calcium should be less than 4 mg/kg per 24 h. Circulating phosphate concentrations decrease considerably throughout the first year of life, and even further throughout later childhood. The normal ranges are substantially greater than that seen in older adults. This change is primarily due to increased reclamation of filtered phosphate in the proximal renal tubule early in life. The confusion in interpretation of age-related normal ranges has continued to result in missed diagnoses and inappropriate interpretation of mineral status. The assessment of urinary phosphate excretion should be performed on a 2-h fasting urine specimen, with a blood sample obtained at the midpoint of the urine collection. The tubular reabsorption of phosphate (TRP) is calculated as:

The %TRP can be plotted on the nomogram of Walton and Bijvoet (21) to obtain the TMP/GFR, or tubular maximum for phosphate expressed per GFR. This value reflects the value of serum phosphate above which one will tend to stop reclaiming phosphate in the tubule. The normal ranges vary with age and approximate the normal phosphate concentrations for age.

Finally, values for circulating PTH do not change after early infancy throughout childhood. Circulating 1,25 (OH)₂D levels tend to be slightly higher in childhood than in later life. This is particularly true during the first 2 years of life. The authors generally have observed values up to 30% in excess of the adult normal range in normal children at his institution. Furthermore, there appears to be less stringent regulation of conversion of 25 (OH)D to 1,25 (OH)₂D in early life.

Total serum calcium is comprised of a free or ionized calcium component, a protein (primarily albumin) bound component, and a small component of filterable calcium that is complexed to other ions such as sulfate, citrate, or phosphate. The ionized and protein-bound components each represent approximately 45–50% of the total calcium. The ionized fraction is the active component, and derangements in this fraction result in symptoms. As discussed elsewhere, serum calcium can be low with a simultaneously normal ionized fraction. This finding is typical of hypoalbuminaemia or acidosis. Various correction factors have been proposed, and are applicable to children as well as adults, however accurate measures of ionized calcium are preferable to calculated corrections.

This section discusses disorders of calcium homoeostasis in childhood, with a primary focus on abnormalities in the maintenance of serum calcium. Several of these disorders, however, primarily affect the skeletal calcium compartment, and bone disease may be a more significant abnormality than perturbations in the serum concentrations. Thus certain disorders are described in which serum calcium levels are often normal in the clinical setting, but at the expense of osteopenic or rachitic abnormalities.

The presentation of hypocalcaemia in the newborn period typically includes facial twitching, limb jitteriness, or other features of neuromuscular irritability, occasionally progressing to focal or generalized convulsions. Poor feeding, hyperacusis, and laryngospasm have been described. On the other hand, nonspecific findings such as apnoea, tachypnoea, tachycardia, cyanosis, or vomiting may be the only features evident. In older children tetany or perioral tingling, presentations more characteristically seen in adults, are more likely to be encountered. As with infants, focal seizures and generalized convulsions may occur. Carpopedal spasm in school-age children has often been attributed to a writer’s cramp. This phenomenon may be exacerbated by the hypomagnesaemia and alkalosis frequently encountered in states of parathyroid insufficiency. Lethargy, vomiting, and other nonspecific signs have also been reported. The electrocardiogram may reveal a prolonged corrected Q–T interval, Q–T_c, which is determined by dividing the Q–T interval by the square root of the EKG cycle. The upper limit of normal in children is 0.44. The musculature may be affected by chronic hypocalcaemia. Serum creatine kinase activity may be elevated in chronic hypocalcaemia; over time actual myopathic changes may occur.

Neonatal hypocalcaemia seen transiently in the first few days of life is commonly referred to as early neonatal hypocalcaemia. This is often seen in preterm infants and has been explained as an exaggeration of the normal postnatal decrease in serum calcium levels. Early neonatal hypocalcaemia appears to occur with greater frequency in asphyxiated babies and in infants of diabetic mothers than otherwise. The hypocalcaemia seen in infants of diabetic mothers is probably multifactorial. Magnesium deficiency has been implicated, as well as alterations in maternal metabolism secondary to poor glucose control throughout gestation. Whether the normal postnatal increase in PTH secretion is blunted is not entirely clear.

Late neonatal hypocalcaemia occurs after 5–7 days of life and is a syndrome more characteristic of the term infant. Late neonatal hypocalcaemia often presents with seizures and is less likely to be transient in nature. Hypoparathyroidism and magnesium deficiency often present in this time frame. Hypocalcaemia in babies with congenital heart disease of many types has been reported as a relatively common finding. Hypocalcaemia related to vitamin D deficiency may present at several weeks of age, however radiographic evidence of rickets is usually not observed until the child is over 2 months old.

One classic situation in which prolonged neonatal hypocalcaemia occurs is in the infant of the hyperparathyroid mother. Presumably the maternal hypercalcaemia results in increased transport of calcium from the maternal to fetal circulation. The resultant excess calcium supply to the fetus is thought to suppress parathyroid responsivity, and prolonged hypoparathyroidism results. Symptomatic hypocalcaemia and hyperphosphataemia are typical biochemical features; hypomagnesaemia may occur as well. The disorder is usually transient, but some cases have been prolonged for months. Unrecognized maternal hyperparathyroidism should be carefully investigated in children that present with the characteristic features of the disorder. Maternal familial hypocalciuric hypercalcaemia (FHH) can result in this syndrome; it is presumed that any cause of chronic maternal hypercalcaemia can result in a similar clinical picture.

Hypocalcaemia in the newborn setting may also occur during blood transfusions using citrated blood products. Citrate complexes with ionized calcium, reducing its circulating concentration to a level where neuromuscular hyperexcitability may occur. Total serum calcium is usually not decreased. Hypocalcaemia can occur in the congenital nephrotic syndrome. Persistent hypocalcaemia may present in this time frame as well. Congenital hypoparathyroidism may be present, as in the DiGeorge syndrome. The classic triad of this chromosome 22 deletion syndrome (hypoparathyroidism, athymia, and conotruncal defects of the heart) typically results in long-standing hypoparathyroidism, although ‘partial’ hypoparathyroidism has been described. Mitochondrial diseases also may present as congenital hypoparathyroidism. Severe osteopetrosis may present with hypocalcaemia secondary to impaired mobilization of calcium from bone. Typically PTH levels are elevated in this situation. Severe vitamin D deficiency is generally an acquired condition manifest as hypocalcaemia as early as 2–3 months, but low maternal stores have rarely contributed to its development in an even younger age range.

Osteopenia of prematurity is commonly encountered in premature infants. Poor bone mineralization is evident on radiographs or other measures of bone mineral density. In general, the problem is more severe in children of lower birthweight. With increasing survival of children with birthweights less than 1000 g the severity of this problem is increasing. Classical rachitic changes of flared and frayed epiphyses, craniotabes, and a rachitic rosary may develop over the first months of life. The histological pattern of bone is thought to be a combined lesion with components of osteomalacia and osteoporosis. This disorder is a consequence of premature withdrawal of the maternal mineral supply. The enteral route, even with maximum feeding delivery, cannot provide for the mineral demands of the skeleton as it rapidly grows and mineralizes throughout the latter weeks of gestation. The problem often occurs in the setting of normocalcaemia. In preterm infants fed solely with breast milk, a phosphate deficiency syndrome may occur, as the phosphate content of breast milk is considerably less than that of commonly used cow’s milk formulas. Although breast milk phosphate is adequate for the growth of the term infant’s skeleton, human breast milk fortifiers are routinely used to increase the mineral intake of the preterm infant. One caveat regarding the use of such fortifiers: calcium intake, if excessive, can result in hypercalcaemia, as its absorption is not tightly regulated in early infancy, and the fractional absorption of calcium can be very high in a low birthweight premature infant.

Symptomatic infants are replaced with calcium, but there is controversy regarding treatment of hypocalcaemic infants who are asymptomatic. The emergency treatment of neonatal hypocalcaemia consists of the intravenous administration of 1 ml/min of 10% calcium gluconate, which should not exceed 2.0 ml/kg. This may be repeated three to four times in 24 h. After acute symptoms have been managed, 5.0 ml/kg of 10% calcium gluconate may be given with intravenous fluids over 24 h. Calcium supplements may be introduced orally if tolerated. In persistent cases, the load of dietary phosphate should be lessened with a formula such as Similac PM 60/40. When hypomagnesaemia is identified, it can be treated with 0.1–0.2 ml/kg of a 50% solution of magnesium sulfate (MgSO₄ •7H₂O).

A wide variety of hypocalcaemic syndromes occur in children due to abnormalities in parathyroid synthesis or secretion. These syndromes provide classic examples of the critical role PTH plays in protecting the organism from acute decreases in the serum calcium level. The clinical manifestations of hypocalcaemia described above are the typical presenting features of hypoparathyroidism. Biochemical features usually include low blood total and ionized calcium levels, and an elevated blood phosphate level. The serum magnesium level may be low, and alkalosis may be present.

Hypoparathyroidism may result from a variety of causes (Box 4.6.1). A number of genetic disorders may cause agenesis of parathyroid glands, disrupt PTH synthesis, processing, and secretion, and/or result in autoimmune destruction of parathyroid glands. As noted above, agenesis of the parathyroid glands occurs in the classic DiGeorge (OMIM 530000) triad of hypoparathyroidism, athymia, and conotruncal heart defects. This syndrome is now known to be part of the larger spectrum of disease referred to as CATCH 22, a sequence of contiguous microdeletion syndromes localized to chromosome 22q11.2. These mutations are most notable in the TBX1 gene, a transcription factor which plays an important role in development of thymus and parathyroid glands (22, 23). Hypoparathyroidism usually occurs in those disorders most related to the classic DiGeorge’s syndrome, but has also been described in the velocardiofacial syndrome, another CATCH-22 pattern of anomalies. Fluorescent in situ hybridization using DNA probes that hybridize to the 22q11.2 locus are helpful in establishing the diagnosis.

Mutations of the preproPTH gene, resulting in disruption of PTH secretion (24), or processing and translocation of PTH (OMIM 146200) from the endoplasmic reticulum, (25) can cause familial isolated hypoparathyroidism (24). Likewise, mutations of the transcription factor glial cells missing 2 (GCM2; also referred to as glial cell missing B (GCMB)), resulting in loss of activity, will result in familial isolated hypoparathyroidism (26–29). A form of X-linked hypoparathyroidism has been reported in patients with mutation in the gene for the transcription factor Syr box 3 (SOX3) (30).

A number of other patients with congenital hypoparathyroidism have concomitant involvement of other organ systems. The syndrome of hypoparathyroidism, sensorineural deafness, and renal anomalies (HDR syndrome) (OMIM 146255) is an autosomal dominant disorder due to mutations or deletions of the gene for the transcription factor GATA3 on chromosome 10 (31–34). Loss-of-function mutations in tubulin chaperone E (TBCE) gene cause the autosomal recessive syndrome of hypoparathyroidism, mental retardation, and dysmorphism (Sanjad–Sakati syndrome) (OMIM 241410) (35–37) and Kenny–Caffey syndrome (hypoparathyroidism, dwarfism, medullary stenosis of the long bones, and eye abnormalities) (OMIM 244460). Mitochondrial gene defects, ranging from deletions to mutations and rearrangement, have been associated with hypoparathyroidism in patients with Kearns–Sayre syndrome (OMIM 53000) (external ophthalmoplegia, pigmentary retinopathy, cardiomyopathy, diabetes, and hypoparathyroidism), MELAS (OMIM 540000) syndrome (diabetes and hypoparathyroidism), and MTPDS syndrome (peripheral neuropathy, retinopathy, hypoparathyroidism) (24, 38, 39). Other mitochondrial disorders associated with hypoparathyroidism include mitochondrial trifunctional protein deficiency (OMIM 609015), long-chain 3-hydroxyacyl-coenzyme A dehydrogenase deficiency (LCHAD) (OMIM 609016), and propionic acidaemia (OMIM 606054). We have recently observed hypoparathyroidism in a mitochondrial DNA deletion syndrome initially diagnosed as Pearson’s syndrome (OMIM 557000).

Gain-of-function mutations of the recently discovered calcium-sensing receptor (CaSR) on parathyroid cells have been found to cause an autosomal dominant variety of hypocalcaemia. The CaSR has increased affinity for ionized calcium, such that relatively low concentrations of this ion effectively suppress PTH secretion, and a steady state serum calcium level in the hypocalcaemic range is established. One family with previously diagnosed autosomal dominant hypoparathyroidism has been shown to actually have this disorder. It may be possible to distinguish autosomal dominant hypocalcaemia from other forms of parathyroid insufficiency because of the relative hypercalciuria that occurs. Circulating PTH levels may not be undetectable in the untreated state (40). This distinction from hypoparathyroidism has important consequences regarding long-term management. That is, the standard treatments for hypoparathyroidism, vitamin D or its metabolites, and calcium, can further exaggerate this hypercalciuria such that nephrocalcinosis, nephrolithiasis, and renal impairment may result, particularly when serum calcium levels are kept in the usual normal range. The CaSR and associated disorders are discussed in Chapter 4.4.

A major cause of acquired hypoparathyroidism is due to the autoimmune polyglandular syndrome (APS) type I. The disorder is also referred to by other names such as autoimmune polyendocrinopathy, candidiasis, and ectodermal dystrophy (APECED) (41). The primary manifestations of this disorder include hypoparathyroidism, primary adrenal insufficiency, and mucocutaneous candidiasis. A variety of other autoimmune phenomena may occur, often grouped with the endocrine, gastrointestinal, or dermatological systems (Box 4.6.2). Defects in cellular or humoral immunity may occur. The presentation typically begins in early childhood with candidiasis. In later childhood the onset of hypocalcaemic symptoms related to hypoparathyroidism occurs, and Addison disease often presents during adolescence. There is considerable variability, however, and often patients do not develop the classic triad. The elevation in serum calcium that may occur during an acute adrenal crisis in an undiagnosed individual may result in an increase of the serum calcium from a hypocalcaemic to a normocalcaemic value, thus masking a coexistent presentation of hypoparathyroidism. Serum calcium should be determined during the initial presentation of suspected Addison’s disease, but also shortly after recovery from the acute crisis.

APS 1 syndrome is due to mutations in the gene encoding the autoimmune regulator protein AIRE (42–44). Mutations resulting in a single amino acid substitution at position 257 (arginine→glutamic acid) were present in more than three-quarters of the affected Finnish cases. The same mutation has been identified in the majority of cases in another series from Italy and Switzerland. Autoantibodies directed to parathyroid cells have been described in patients with APS 1, but the frequency of these findings vary greatly dependent upon the series studied (45). In one recent series, antibodies to the parathyroid CaSR were detected in the sera of approximately half the affected individuals.

In addition to the usual measures taken in the management of hypoparathyroidism, treatment of this disorder may require aggressive mineral replacement, due to the often complicating issue of malabsorption. Acute illness can be associated with such severe impairment of gastrointestinal absorption of calcium and magnesium that parenteral replacement of these minerals may be required. Continuous nocturnal nasogastric calcium supplementation may be a useful temporary measure in the affected child unable to tolerate standard bolus feeding, and where prolonged parenteral infusions are not practical.

The long-term prognosis of this condition has improved greatly over the past generation. Early cases succumbed to such problems as unrecognized adrenal crises or diabetic ketoacidosis. Although numerous complications of this disorder are recorded, including overwhelming Candida sepsis, oesophageal carcinoma, and chronic active hepatitis, these severe features are rare. One review records a 50-year survival of greater than 75%.

Surgical hypoparathyroidism is rarely encountered in childhood. Standard guidelines for thyroid surgery include identification and preservation of parathyroid tissue. The use of radioactive iodine to ablate any thyroid remnants following surgery for cancer has resulted in a less aggressive approach to thyroid surgery. Transient hypocalcaemia in the 36 hours acutely following thyroid surgery is common, although no mechanism has been clearly established which accounts for this finding. Early (1 hour) postoperative intact PTH measurement has been proposed as a sensitive way of identifying patients who are at risk of becoming hypocalcaemic (46, 47).

Hypoparathyroidism may occur as a complication in disorders related to metal toxicity. Thalassaemia has been shown to result in a variety of endocrinopathies related to iron deposition. Although hypoparathyroidism was recognized as an occasional complication in most clinics managing such patients, the incidence of this complication has decreased over the past 20 years with increasing use of chelating therapies such as desferoxamine (48). Hypoparathyroidism has been reported to occur in Wilson’s disease presumably related to copper deposition in the parathyroid glands (49).

Functional hypoparathyroidism occurs in severe magnesium deficiency. Both impairment of parathyroid secretion and resistance to PTH activity at the renal tubule have been described in the setting of chronic magnesium deficiency. In classic studies by Anast et al. (50), the dependence of the parathyroid glands on magnesium for secretion of PTH was described in a girl with a congenital magnesium wasting syndrome. It appears that more subtle defects in PTH secretion may occur with less severe magnesium depletion. The serum magnesium level may not reflect total body magnesium status, as magnesium is predominantly an intracellular ion. However, magnesium deficiency associated with a serum level of greater than 1.3 mg/dl is unlikely to result in clinically significant changes in parathyroid secretion. In children, chronic magnesium deficiency occurs in familial hypomagnesaemia (51). Familial hypomagnesaemia (OMIM 602014) with secondary hypocalcaemia is an autosomal recessive disease due to mutations in the TRPM6 ion channel, resulting in electrolyte abnormalities in the newborn period (52). This disorder may present with hypocalcaemic/ hypomagnesaemic seizures in the first 2 months of life. If diagnosed early, severe neurologic impairment may be prevented. Mutations in paracellin (encoded by CLDN16), a renal tubular paracellular transport protein of the claudin family, may also cause hypomagnesaemia, hypocalcaemia, and hypercalciuria (53) (OMIM 248250). Another member of this family, CLDN19, may also cause a similar syndrome (54) (OMIM 248190). Gitelman’s syndrome (OMIM 263800) is an autosomal recessive disorder of magnesium and potassium wasting with metabolic alkalosis and hypocalciuria, due to mutations in SLC12A3, which encodes a thiazide-sensitive Na–Cl cotransporter (55).

Hypomagnesaemia in children occurs more frequently in children with the use of chemotherapeutic agents such as cisplatin, and with aminoglycoside diuretics such as tobramycin. Hypomagnesaemia has been reported with ibuprofen overdosage in a 21-month old child simultaneously treated with furosemide (56), and following ingestion of ammonium bifluoride-containing automobile wheel cleaner (57). Recent data have suggested that moderate decreases in serum magnesium levels accompany the hypocalcaemia encountered during rehabilitation from burn injuries in children (58).

The mainstay of the management of chronic hypoparathyroidism is replacement with oral calcium supplements and an active metabolite of vitamin D, such as 1,25 (OH)₂D₃ (calcitriol) or 1α (OH)D₃. Various liquid forms of calcium carbonate (often sold as antacids for children) are useful for the small child unable to take tablets. The doses are titrated to maintain the serum calcium in the slightly low to low-normal range, with care not to render the child hypercalciuric. This is especially a concern in patients with autosomal dominant hypocalcaemia due to activating mutations in the CaSR. Teraparatide (parathyroid hormone (1–34)) and calcilytic agents (antagonists of CaSR) are currently being studied as potential therapeutics for these disorders.

Resistance to parathyroid hormone has been termed pseudohypoparathyroidism (PHP). The classical form of the disease, PHP type 1A is due to mutations in the alpha subunit of the heterotrimeric guanine nucleotide binding protein, G_s. The change in conformation of this protein, induced by interaction of PTH with its receptor, activates membrane adenylate cyclase, thus initiating the protein kinase C signal transduction pathway. Patients with PHP 1a fail to transduce PTH (and various other peptide hormone) signals in target tissues. The syndrome represents a fascinating aspect of hormone receptor biology, and is discussed in detail in Chapter 4.5. For our purposes here, we will describe only a few of the clinical features pertinent to children with the disorder.

The major clinical features of PHP 1a (OMIM 103580) are: (1) resistance to PTH, manifest by hypocalcaemia and hyperphosphataemia; (2) a specific phenotype of short stature, round facies, and other skeletal features such as the presence of shortened fourth metacarpal bones, collectively referred to as Albright’s hereditary osteodystrophy; and (3) generalized resistance to peptide hormones that require intact G_sα for signal transduction. PHP may be suspected in child within a family because of other affected members or individuals that manifest the PHP 1a phenotype, but have no biochemical abnormalities or other endocrine deficiencies. Presentations in infancy may include short stature and congenital hypothyroidism, but manifestations of hypocalcaemia usually appear in later in childhood.

PHP 1b (OMIM 603233) is also linked to the G_s-α gene on chromosome 21, but isolated PTH resistance occurs with none of skeletal features manifest in type 1a Associated endocrine abnormalities are usually mild or absent. PHP 1c (OMIM 612462) refers to syndromes of PTH resistance, with other associated endocrine abnormalities, but without Albright’s hereditary osteodystrophy. Evidence as been presented indicating that PHP types 1a and 1b are subject to expression by genomic imprinting (59–62).

Type II PHP (OMIM 203330) refers to PTH resistance mediated by a pathway not resulting from interference with the generation of cAMP by adenylate cyclase. No specific phenotype, or associated hormone resistance, occurs.

Treatment of PHP is similar to that for primary hypoparathyroidism, although there is usually little risk of hypercalciuria in the classic (1a) form of the disease. Other features of the disease may require therapy, such as hypothyroidism, and monitoring of subcutaneous ectopic ossification, which if severe may require surgical removal.

Hypocalcaemia may result from a deficiency of vitamin D, usually related to limited dietary content of vitamin D or limited exposure to ultraviolet light, critical in the early steps of endogenous vitamin D production in skin. Sufficient UV light is necessary for the production of previtamin D in the stratum spinosum of the dermis. UVB light of wavelength 290–315 nm provides the energy to disrupt the 9–10 C–C bond in the B ring of the steroid nucleus of 7-dehydrocholesterol (Fig. 4.6.2). Previtamin D is then rapidly isomerized in the skin to vitamin D. Thus vitamin D deficiency is most frequent in parts of the world where sunlight exposure is limited or exposure to sunlight is prevented. In North America, there have been numerous recent reports of vitamin D deficiency rickets in breast-fed children not supplemented with vitamin D. Breast milk contains little vitamin D, unless the mother is taking pharmacological doses of the vitamin. Pigmented individuals are at higher risk for this problem, as melanin absorbs UV light external to the layer of skin where vitamin D synthesis occurs. There is a greater incidence of the problem in the late winter as compared to other times of the year. In some children, vitamin D deficiency is compounded by the coincident problem of calcium deficiency. This problem may occur in lactose intolerant children avoiding dairy products, or when the diet has a very high phytate content (as with certain grains and cereals) which can limit the bioavailability of ingested calcium.

Vitamin D deficiency may often present with rachitic bone disease. Symptomatic hypocalcaemia generally occurs late in the course of development of vitamin D deficiency. The initial decreases in intestinal calcium absorption which result from vitamin D deficiency are readily compensated for by the resultant secondary elevations in PTH. When frank hypocalcaemia with tetany or seizures occurs due to vitamin D deficiency, substantial chronicity of vitamin D deficiency has usually been present.

Vitamin D is further metabolized to its most abundant circulating metabolite, 25 (OH)D, and its best-known active metabolite, 1,25 (OH)₂D. The critical mechanism of action for this involves binding to its receptor, a DNA binding protein which is part of the large superfamily of steroid/ thyroid/ retinoid receptors. Pathophysiology similar to vitamin D deficiency may result from defects in the 1α hydroxylase enzyme instrumental in the synthesis of 1,25 (OH)₂D, or in mutations in the vitamin D receptor.

Clinical features of rickets in children include bowing of the lower extremities, craniotabes, and rachitic rosary (hypertrophy of the costochondral junctions). These syndromes are discussed in detail in Chapter 4.10. The laboratory findings of vitamin D-related hypocalcaemic disorders are compared in Table 4.6.2.

Establishment of an appropriate biochemical threshold for the definition of vitamin D deficiency has received considerable attention recently. In the adult population the circulating 25(OH)D level of 32 ng/ml (80 nmol/l) has been suggested as an appropriate target threshold for the definition of vitamin D deficiency (63). Recent reports of various health benefits have been associated with vitamin D status, mostly related to a variety of associations of 25 (OH)D level and prevalence of certain disorders, including diabetes, multiple sclerosis, cancers (particularly colorectal), and obesity (64). These data are epidemiological in nature, and the possibility that vitamin D status serves as a marker for other unidentified contributors to disease remains. Perhaps the most convincing direct evidence of ‘nonclassical’ vitamin D effects (i.e. effects apart from those influencing systemic calcium homoeostasis) comes from immunological studies demonstrating important effects of vitamin D on macrophages (65). In the presence of activated toll-like receptors 1/2, macrophages are able to express their own 1α hydroxlase and its own vitamin D receptor. Thus the macrophage has the capacity to metabolize 25 (OH)D and to use the activated product, 1,25 (OH)₂D, in autocrine fashion, as it can produce this molecule’s receptor. The stimulation of the macrophage in this way leads to the production of a unique antimicrobial peptide, cathelicidin, which is inhibitory to the growth of Mycobacterium tuberculosis. Thus a clear mechanism exists by which vitamin D may play a role in fighting infection. In other studies based on linear regression analysis of small population data, elevations in circulating PTH occur as levels of circulating 25 (OH)D decrease. However the threshold values of 25 (OH)D at which an increase in PTH levels occurs is quite variable (67), suggesting caution in using this measure as a generalizable means of establishing vitamin D deficiency. Nevertheless, these findings, in sum, raise the issue of revising the threshold for 25 (OH)D level as a measure of optimal vitamin D status, and toxicity information in adults appear to indicate that modest increases in supplementation is safe. Data are not yet available to establish a clear benefit to this approach, and the application of such measures to infants and children needs to be carefully examined. Indeed, the administration of vitamin D to infants and children at the increased levels recently suggested for supplementation in adults could be risky. The authors have recently observed hypercalcaemia in an infant given 1400 units of supplemental vitamin D daily, with concomitant circulating 25 (OH)D levels over 225 nmol/l (90 ng/ml). Thus we have continued to support a conservative definition of vitamin D deficiency, using threshold values for 25 (OH)D of 37.5–50 nmol/l (15–20 ng/ml). Likewise, in the normal healthy term infant we advise adherence to current recommendations of a daily vitamin D intake of 400 IU.

Vitamin D in dosages of 1000–2000 IU/day is a standard approach to the initial treatment of vitamin D deficiency. As rachitic lesions heal, the dosage is decreased to 400 IU/day, the generally recognized recommended daily allowance. A single intramuscular dose of 6 00 000 units of vitamin D, or in two oral doses of 3 00 000 units each can be given in the outpatient clinic if the clinical situation would indicate that limited follow-up will occur.

The specialty clinician may happen to evaluate such a patient after a change in season, and sunlight exposure has concomitantly increased since the onset of the disease, or after vitamin supplementation has begun. This situation should be recognized clinically, as low-normal values of the 25 (OH)D level may confuse the diagnosis. Radiographs may demonstrate a thin, dense line of opacity at the metaphyses of long bones, which indicates that recent rapid mineralization has occurred at the edge of the growth plate.

As mentioned above, children with vitamin D deficiency often require supplemental calcium. Some children may manifest hungry bone syndrome, in which mineralization is rapid and serum calcium levels may decrease as the bone mineralizes. Thus supplemental calcium is given in many cases to provide a total daily intake of 30–50 mg/kg of elemental calcium. Vitamin D stores may be depleted rapidly during calcium insufficiency (68), suggesting that dietary calcium deficiency itself may be yet another risk factor for the development of vitamin D deficiency.

Deficiency of 1α hydroxylase is best treated with physiological doses of 1,25 (OH)₂D₃. Hereditary resistance to vitamin D may respond to high dosages of 1,25 (OH)₂D₃, but some patients require parenteral calcium infusions (69).

Hypocalcaemia may also result from rapid loading of phosphate into the circulation. This phenomenon occurs in settings of tissue destruction, such as in tumour lysis syndrome observed during early phases of chemotherapy of large solid tumours. Rhabdomyolysis may decrease serum calcium levels for similar reasons. Several cases of hypocalcaemia and seizures have occurred following high-dose administration of phosphate either by enema or by the oral route. The use of phosphate-based cathartics in infants and small children is contraindicated. We are aware of one case in which severe hyperphosphataemia and hypocalcaemia occurred repetitively in a small child surreptitiously administered oral phosphate by her mother. Hypocalcaemia may also occurs in the setting of pancreatitis due to precipitation of calcium-containing salts in the inflamed pancreatic tissue and it often correlates with the severity of the episode. Children with acute or chronic renal failure will also develop mild hypocalcaemia which is due to a multitude of precipitating factors, such as hyperphosphataemia, and decreased 1α hydroxylation of 25-hydroxyvitamin D. Hypocalcaemia has recently been associated with high volume (1.5 litres or more per week) of soft drinks containing phosphoric acid (70). Hypocalcaemia solely due to low dietary calcium intake has been reported (71). This syndrome has occurred in areas of South Africa in areas where food content is relatively low in calcium, and rickets is the usual presenting feature. The combination of dietary vitamin D and calcium deficiency is more commonly seen in North American children (see above).

Persistent hypercalcaemia is usually attributed to some combination of the following mechanisms: (1) excessive intestinal absorption of calcium; (2) excessive bone resorption of mineral; and (3) abnormal renal retention of calcium. Infants are usually asymptomatic with mild to moderate hypercalcaemia (11.0–13.0 mg/dl). More severe hypercalcaemia may lead to failure to thrive, poor feeding, hypotonia, vomiting, seizures, lethargy, polyuria, and hypertension. Hypercalcaemia is discussed in detail in Chapter 4.2. Several syndromes with specific childhood features are described below.

Severe neonatal hyperparathyroidism (OMIM 239200) is a rare condition presenting with hypercalcaemic symptoms in the first few days of life. Serum calcium levels may range as high as 15 to 30 mg/dl. The serum phosphate level is usually low, and serum PTH is elevated. The hypercalcaemia is predominantly due to increased bone resorption, but elevated intestinal absorption of calcium, as well as increased renal calcium retention, probably occur. Radiographs of the clavicles typically reveal features of primary hyperparathyroidism. Nephrocalcinosis may be present on ultrasonographic examination. Severe neonatal hyperparathyroidism may occur in families with familial hypocalciuric hypercalcemia (FHH) (OMIM 145980). This autosomal dominant trait is manifest by modest asymptomatic hypercalcaemia with relative hypocalciuria and normal or slightly increased serum PTH levels. A loss-of-function mutation in the CaSR gene on parathyroid cells, acts in a dominant negative manner in FHH; severe neonatal hyperparathyroidism has been shown to result in individuals homozygous for such a mutation (72). Severe neonatal hyperparathyroidism usually requires emergency extirpation of the parathyroid glands. Hypercalcaemia of sufficient severity to warrant surgery has also been described in infants in FHH families that have only one mutant copy of the CaSR gene (73).

In severe Williams syndrome (OMIM 194050), symptoms may be present from the neonatal period, but more frequently recognized later in the first few years of life. Infantile hypercalcaemia may be a presenting feature, in addition to pre- and postnatal growth failure. A characteristic, unusual facies (Fig. 4.6.3) is often present, as well as cardiovascular abnormalities (usually supravalvular aortic stenosis or peripheral pulmonic stenosis), delayed psychomotor development, and selective mental deficiency. A deletion of the elastin gene is found in many cases of Williams’ syndrome. The serum calcium levels may range as high as 12–19 mg/dl. The hypercalcaemia usually subsides spontaneously by the age of 4 years.

The pathogenesis of hypercalcaemia (OMIM 143880) is uncertain, although various disturbances in vitamin D metabolism have been described. Treatment has traditionally consisted of placing the child on a low calcium diet, free of vitamin D. Short-term therapy with corticosteroids may also be necessary. More recently we have found that intravenous bisphosphonate therapy is quite effective in controlling hypercalcaemia in Williams’ syndrome patients. We usually use pamidronate at a dose of 0.25–0.5 mg/kg per dose, and have found that one to three doses have been sufficient to manage this problem permanently.

Milder forms of idiopathic infantile hypercalcaemia have been described with less severe hypercalcaemia and less overt clinical features, however the degree of hypercalcaemia can be quite variable among children with a classic phenotype. Elevations in PTHrP have been reported in some of these individuals.

Subcutaneous fat necrosis is a self-limited disorder which presents in infancy with symptoms of hypercalcaemia and violacious discoloration of the skin. These areas of discoloration consist of a mononuclear cell infiltrate, sometimes coexistent with small calcification. Increased production of 1,25 (OH)₂D has been described. Thus a vitamin D-free, low calcium diet and glucocorticoids have traditionally been used to treat the disorder. More recently, we have employed pamidronate (0.25 mg/kg body weight) to successfully control hypercalcaemia in this condition. Often a single dose is sufficient.

Intoxication with vitamin D or vitamin A should be excluded in the older infant with hypercalcaemia. In vitamin D intoxication, it is important to measure the circulating level of 25 (OH)D, the most abundant circulating vitamin D metabolite. Levels of 1,25 (OH)₂D are usually low. Toxicity may be mediated by the overwhelming large dosages of 25 (OH)D interacting with the vitamin D receptor. Alternatively because 25 (OH)D has a far greater affinity than 1,25 (OH)₂D for the circulating vitamin D binding protein, it has been proposed that the latter, more active metabolite is displaced from vitamin D binding protein, with toxicity resulting from the increase in free levels of 1,25 (OH)₂D. Excess intestinal absorption of calcium is present, and in some cases there is evidence for increased bone resorption.

Vitamin A intoxication results in bone pain, hypercalcaemia, headache, pseudotumour cerebri, and a characteristic erythematous skin rash with exfoliation. Alopecia and profuse ear discharge may be present. The hypercalcaemia is thought to be mediated by bone resorption. Although toxicity is not thought to occur when less than 50 000 units of vitamin A or equivalent is ingested on a daily basis, reports of toxicity with less ingestion have been recorded in children (74). Unrecognized liver disease may decrease the tolerance of vitamin A. In order to establish the diagnosis of vitamin A intoxication, serum retinyl ester levels should be determined in addition to the more common test for serum retinol.

Other conditions in which hypercalcaemia may be manifest in children include Down’s syndrome, skeletal dysplasias (such as Jansen’s), and in osteogenesis imperfecta. Indeed we have observed mild elevations in serum calcium during infancy in association with a variety of skeletal dysplasias. This appears to be a transient phenomenon. Endogenous overproduction of 1,25 (OH)₂D has been described in twins with cat-scratch disease induced granulomata (75). Other major causes of hypercalcaemia include those commonly encountered in adults: immobilization, malignancy, and acquired hyperparathyroidism, including parathyroid adenomas. It may be useful to measure PTHrP levels in the setting of undiagnosed hypercalcaemia. Elevated circulating levels of PTHrP would prompt a careful search for neoplastic disease.

The medical management of acute symptomatic hypercalcaemia consists of the administration of intravenous saline. Additionally, furosemide, in a dose of 1 mg/kg, is frequently given intravenously at 6- to 8-h intervals. Intravenous infusion of pamidronate has also been useful in this setting. Specific long-term therapy depends on the specific hypercalcaemic disorder.

The use of bisphosphonate therapy in children has increased in recent years. Pamidronate has been highly successful in the management of hypercalcaemia associated with childhood cancers (76). We have successfully used this medication for the treatment of hypercalcaemia in Williams’ syndrome and subcutaneous fat necrosis as noted above. Short-term data would suggest that side effects are minimal and that the therapy is safe. One should be aware of the potential complications of electrolyte disturbances, particularly hypocalcaemia, hypophosphataemia, and hypomagnesaemia.