Follow-up of school-age children with bronchopulmonary dysplasia.

OBJECTIVE: To investigate the outcome of school-age children with bronchopulmonary dysplasia (BPD) in terms of nutrition, pulmonary function, and intelligence, and to compare the results with a preterm cohort matched for gestational age and birth weight, and with a term control group. DESIGN: Cross-sectional. SETTING: Follow-up clinic at level III neonatal intensive care unit, university-affiliated hospital, Children's Hospital. SUBJECTS: Twelve children who had BPD as infants and 2 control groups of 12 children each. MAIN OUTCOME VARIABLES: Anthropometric measurements, dietary intake, resting energy expenditure, pulmonary function, body composition measurements by dual energy x-ray absorptiometry, and Weschler intelligence test scores. RESULTS: Children with BPD had decreased forced expiratory volume at 1 second, decreased forced expiratory flow between 25% and 75% of vital capacity, and decreased maximal expiratory flow velocity at 50% of vital capacity compared with age-matched normal inborn subjects (p = 0.025, p = 0.005, and p = 0.0013, respectively). Both children with BPD and matched preterm control children were shorter than infants in the term control group (p = 0.018). There were no significant differences in the other anthropometric parameters studied. The groups did not differ in resting energy expenditure. Lean body mass was lower in the BPD group compared with the term control groups (p = 0.017). Bone mineral content was lower in the BPD group compared with both the preterm and term control infants (p = 0.050 and p = 0.059, respectively). The mean performance intelligence quotient (IQ) and full-scale IQ scores in the BPD group were lower than in the term control group (p = 0.011 and p = 0.029, respectively). The proportion of children with borderline or intellectually deficient scores was significantly higher in the preterm group compared with the term group for verbal, performance, and full-scale IQ scales (p = 0.046, p = 0.018, and p = 0.048 respectively). The proportion of children with BPD who had borderline or deficient performance IQ scores was higher than for the term group (p = 0.046). CONCLUSIONS: The lower height and intelligence scores in children with BPD may be related to prematurity and perinatal events rather than pulmonary disease. Subclinical pulmonary dysfunction in children with BPD persists at school age. The lower amount of lean body mass and possible decrease in bone mineral content in children with BPD may be related to their smaller size.

Calcium, magnesium and phosphorus in the nutrition of the newborn.

Several factors have been suggested to contribute to inadequate bone mineralization in infants. Calcium and phosphorus intakes in preterm infants are below the intrauterine accretion rates. Calcium retention is influenced by the types of calcium salts used and by alterations in dietary phosphorus, fat and carbohydrates. Dietary intakes of vitamin D, and modifications in the protein base of infant formula, e.g., soy base vs cow milk base, may impact bone mineralization. The major hormonal mechanisms involved in the regulation of bone mineralization are parathyroid hormone, calcitonin and vitamin D. From recent animal studies, it has been suggested that parathyroid hormone related peptide (PTH-rp) may also play a role in perinatal calcium homeostasis.

Facilitated tucking: a nonpharmacologic comfort measure for pain in preterm neonates.

OBJECTIVE: To identify the effectiveness of "facilitated tucking," a nonpharmacologic nursing intervention, as a comfort measure in modulating preterm neonates' physiologic and behavioral responses to minor pain. DESIGN: Prospective, repeated measure, random sequencing, and experimental. SETTING: Level III neonatal intensive-care unit of a tertiary care university pediatric hospital. PARTICIPANTS: Thirty preterm neonates, 25-35 weeks gestation. INTERVENTIONS: Heart rate, oxygen saturation, and sleep state were recorded 12 minutes before, during, and 15 minutes after two heelsticks, one with and one without facilitated tucking. HYPOTHESIS: Premature neonates will have less variation in heart rate and hemoglobin oxygen saturation, shorter crying and sleep disruption times, and less fluctuation in sleep states in response to the painful stimulus of a heelstick with facilitated tucking than without. RESULTS: Neonates demonstrated a lower mean heart rate 6-10 minutes post-stick (p < 0.04), shorter mean crying time (p < 0.001), shorter mean sleep disruption time (p < 0.001), and fewer sleep-state changes (p = 0.003) after heelstick with facilitated tucking than without. CONCLUSION: Facilitated tucking is an effective comfort measure in attenuating premature neonates' psychologic and behavioral responses to minor pain.

Evaluation by dual-energy X-ray absorptiometry of age-related changes in body composition of male rats.

The purpose of this study was to determine the feasibility of using a recently developed procedure, dual-energy x-ray absorptiometry, to determine body composition in rats as a function of aging. Results were obtained that were consistent with previous findings for male rats.

Redox cycling and hepatotoxicity of diquat in aging male Fischer 344 rats.

The aim of this study was to determine the influence of aging on diquat-induced redox cycling in liver microsomes and diquat hepatotoxicity in rats. Diquat-stimulated production of superoxide anion radical and NADPH-cytochrome c (P-450) reductase activity were measured in liver microsomes prepared from male Fischer 344 rats at ages representing young adulthood (5-6 months), middle age (15-16 months), and old age (24-27 months). Both activities were decreased substantially (40%) in old rats. Diquat-induced liver damage was assessed 6 hr after the administration of diquat (0.1 mmol/kg, ip) on the basis of serum ALT and sorbitol dehydrogenase activities, hepatic microsomal cytochrome P-450 loss, and histological evaluation. The classical manifestations of hepatotoxicity in diquat-treated rats were as severe in old rats as in young-adult ones, despite the age-associated drop in redox cycling capacity. Diquat treatment also resulted in decreased concentrations of hepatic glutathione and ascorbic acid, increased concentrations of hepatic nonheme iron, and decreased liver weights. The changes in glutathione, nonheme iron, and liver weight were more pronounced in livers of middle-aged and old rats than in those of young-adult rats. These age-dependent differences could not be explained on the basis of plasma diquat concentrations, which were similar in the three age groups of rats. The absence of an effect of aging on the hepatotoxic effects of diquat indicates that redox cycling capacity is not limiting for the development of liver damage. Other effects of diquat were influenced by aging, but their relevancy to the hepatotoxicity is uncertain.

Total bone mineral content and body composition by x-ray densitometry in newborns.

Measurement of total body bone mineral content and body composition has not previously been convenient in the newborn. X-ray densitometry, otherwise known as dual-energy x-ray absorptiometry (DEXA), has been used in adults to assess with accuracy and precision the total body mineral content, lean mass, and fat. Body composition measurements were determined in vivo by DEXA in term newborns, and they were compared with values reported by chemical analysis of the cadaver, with skin-fold thickness measurements, and with bone mineral content measured by single photon absorptiometry. The intermeasurement coefficient of variation was < or = 2.5% for bone mineral content, and < or = 1% for fat and lean mass. The values for bone mineral content, fat, and lean mass fall within the ranges expected based on the reported values measured by chemical analysis of the infant cadaver. Triceps and quadriceps skin-fold measurements were best correlated with total body fat as measured by DEXA. The bone mineral content of the distal third radius site as measured by single photon absorptiometry in newborns showed significant correlation with total body bone mineral content. DEXA provides a reproducible and convenient method for the determination of body composition in newborns.

Bone mineral metabolism in full-term infants fed human milk, cow milk-based, and soy-based formulas.

OBJECTIVE: To study the hypothesis that ingestion of a modified soy-based formula with an improved mineral suspension system may result in bone mineral content similar to that observed in infants fed human milk or cow milk-based formulas. DESIGN: Prospective, self-selected group of infants fed human milk randomized between the two formula-fed groups. SETTING: University-based hospital nursery and follow-up. PARTICIPANTS: Fifty-six normal, healthy, full-term infants, free of major malformations or disorders, including 17 infants fed human milk, 19 infants fed a cow milk-based formula, and 20 infants fed a soy protein formula were followed up to 6 months' postnatal age. The soy-based formula studied was modified to improve the suspendability of the minerals. INTERVENTIONS: Infants were fed human milk or the study formula for the first 4 months, at which time beikost was permitted. Infants fed human milk received vitamin supplementation to provide 400 IU of vitamin D per day. MEASUREMENTS: Anthropometric variables, serum calcium, magnesium, phosphorus, alkaline phosphatase, and parathyroid hormone levels were measured at enrollment, and at 8, 16, and 24 to 26 weeks' postnatal age. Bone mineral content at the distal third radius site was measured with single photon absorptiometry at these times. Growth in the infants did not differ significantly among the groups. There was no significant difference in serum calcium, magnesium, alkaline phosphatase, or parathyroid hormone concentrations among the infants during the study. Serum phosphorus was significantly lower at 8 weeks in the group fed human milk than in that fed the cow milk-based formula. Bone mineral content at 16 and 24 to 26 weeks was higher in the group fed the soy-based formula than in that fed human milk, and bone width was also higher at 16 weeks in the infants fed the soy-based formula. CONCLUSIONS: Improving the suspendability of the mineral system in the soy formula results in bone mineralization in infants fed the soy-based formula similar to that measured in infants fed human milk and cow milk-based formula. We suggest that the suspendability of the minerals used is an important variable in the interpretation of the effect of feedings on the bone mineral status of infants.

Understanding the different phases of fetal growth.

Fetal growth was studied in 78 newborns who had serial scans in pregnancy. Weight at birth correlated with growth in the first 2 trimesters but better with the growth between 28 and 32 weeks. The correlations for weight at growth cessation were better than those for weight at birth. Crown-heel length at cessation did not correlate with fetal growth in the first 2 trimesters. Maternal weight increase correlated with fetal weight and crown-heel length but not with head circumference at cessation. Maternal prepregnancy weight correlated with fetal growth between 28 and 32 weeks, but not with fetal growth in the second trimester. A negative correlation was found between estimated duration of growth cessation and relative head circumference at birth.

Bone mineral content of healthy, full-term neonates. Effect of race, gender, and maternal cigarette smoking.

In 238 healthy, full-term infants, we evaluated the effects of race, gender, and maternal cigarette smoking on the newborn radial bone mineral content. Sixty-one percent of the infants were male and 39% were female; 161 of 238 mothers were nonsmokers and 77 mothers were cigarette smokers. Bone mineral content was significantly correlated with birth weight, length, and head circumference. There were no significant race- and gender-related differences in neonatal bone mineral content. There was no significant difference in the bone mineral content, bone width, and bone density of the newborn at the left distal-third radius site among the infants of smokers compared with those of nonsmokers, and there was no correlation between neonatal bone mineral content and the number of cigarettes reportedly smoked per day by the mother.

Effect of intravenous calcium infusions on serum chemistries in neonates.

In 36 newborn infants admitted to the Children's Hospital of Oklahoma, we studied the hypotheses that intravenous bolus infusions of Ca as calcium gluconate over 10 min may (a) result in acute metabolic acidosis, (b) increase serum osmolality, (c) increase serum free bilirubin, and (d) decrease serum phosphorus concentrations. All infants received 18 mg/kg of elemental calcium, as either a 5 or 10% solution of calcium gluconate. Blood ionized calcium (iCa) rose significantly with i.v. bolus Ca infusion (p less than 0.005) and blood pH and serum P declined significantly (p less than 0.05) with i.v. bolus Ca infusion in infants. Serum free bilirubin was not significantly altered. Serum osmolality rose significantly from baseline with bolus infusion of Ca as a 10% calcium gluconate solution and did not change significantly with bolus infusion of a diluted 5% solution. In neonates, intravenous bolus calcium infusion (a) decreases blood pH, (b) infusion of 10% but not 5% calcium gluconate increases serum osmolality, (c) serum free bilirubin concentrations were not altered, and (d) serum phosphorus concentrations were decreased.

Hypercalcitoninemia and hypocalcemia in acutely ill children: studies in serum calcium, blood ionized calcium, and calcium-regulating hormones.

We studied the hypotheses that serum calcium and blood ionized calcium would be low in acutely ill children and would rise with clinical improvement. In 15 children admitted to the pediatric intensive care unit, the blood ionized calcium level was 4.45 +/- 0.06 mg/dl (1.11 +/- 0.015 mmol/L) on entry versus 5.17 +/- 0.03 mg/dl (1.29 +/- 0.01 mmol/L) in control subjects (p less than 0.005), rose significantly on days 2 and 3, and was 5.12 +/- 0.04 mg/dl (1.28 +/- 0.01 mmol/L) at discharge (p less than 0.005). Changes in serum calcium level were similar, whereas serum magnesium and phosphorus levels were normal and did not change. Basal serum parathyroid hormone concentrations were elevated, rose further during the study, and were normal at discharge. Serum parathyroid hormone levels correlated inversely with blood ionized calcium levels, indicating that compensatory hyperparathyroidism occurs with low blood ionized calcium concentrations. Basal serum calcitonin values were evaluated on entry and decreased with clinical improvement. Serum calcitonin levels correlated significantly with low blood ionized calcium levels, indicating that hypercalcitoninemia may play a role in the pathogenesis of hypocalcemia in these children. Urine calcium excretion was not increased in the four children studied. We speculate that with clinical improvement, a rise in serum parathyroid hormone levels and a decline in serum calcitonin levels may help restore normocalcemia in these acutely ill children.

Effect of mineral supplementation of human milk on bone mineral content and trace element metabolism.

We studied the effect of feeding mineral fortified human milk to preterm infants (birth weight less than or equal to 1500 gm). Serum concentrations of calcium, magnesium, phosphorus, zinc, cooper, alkaline phosphatase, and parathyroid hormone were determined, and bone mineral content was measured, in infants fed unfortified human milk (group 1), fortified human milk (group 1), fortified human milk (group 2), and a "humanized," mineral-enriched premature infant formula (group 3). Serum calcium, magnesium, phosphorus, zinc, copper, and parathyroid hormone concentrations did not differ significantly among the groups studied. Serum alkaline phosphatase concentrations increased significantly only in the infants fed unfortified human milk, and bone mineral content in this group was significantly lower than in formula-fed infants.

Lowered serum Ca, blood ionized Ca, and unresponsive serum parathyroid hormone with oral glucose ingestion in infants of diabetic mothers.

Oral glucose ingestion may lower serum Ca in infants of diabetic mothers (IDMs). Six metabolically stable IDMs were studied following ingestion of 1.7 +/- 0.1 g/kg (mean +/- SE) of glucose over 20 min and serum Ca, Mg, P, blood iCa, serum PTH, and CT were measured at 0, 1/2, 1, and 2 h. Data obtained in IDMs were compared with previously reported findings in 10 normal neonates. In IDMs as in normal neonates, serum Ca, Mg, P declined significantly after oral glucose ingestion. Blood Ca2+ was significantly lower at +1/2 h in IDMs versus normal neonates, and by analysis of covariance, trends in blood Ca2+ were significantly different in IDMs versus normal neonates, (p less than 0.05). Serum PTH concentrations were unaltered in IDMs versus a significant rise in serum PTH noted in normal neonates. The difference between the two groups was significant statistically (p less than 0.05). Baseline serum CT was elevated in both groups and did not change. Thus, in IDMs responses to oral glucose ingestion differs from that seen in normal neonates as follows: blood Ca2+ is lowered in IDMs versus normal neonates, and serum parathyroid hormone (PTH) does not respond to a decline in blood Ca2+ in IDMs, whereas in normal neonates serum PTH rises and blood Ca2+ is maintained. We speculate that relative parathyroid gland unresponsiveness occurs in IDMs, which may result in lowered blood Ca2+ after oral glucose ingestion in these infants.

Calcium, phosphorus, magnesium, and calcitonin concentrations in the serum and cerebrospinal fluid of children.

Mineral metabolism in the cerebrospinal fluid (CSF) of children is poorly understood. Recent reports have suggested a neuroregulatory role for calcitonin. We examined the hypotheses that in children (1) CSF levels of calcium and phosphorus might be low, (2) CSF levels of magnesium might be higher than serum levels of magnesium, and (3) immunoreactive calcitonin might be present in the CSF. We examined serum and CSF samples of 45 children, aged 8 days to 16 years, undergoing spinal taps for suspected meningitis or as part of leukemia therapy. Both serum and CSF levels of calcium correlated with those of magnesium. There was no correlation for CSF levels vs serum levels of calcium, magnesium, or phosphorus. The CSF levels of calcium and phosphorus were lower than the serum levels of these elements, but the CSF levels of magnesium were higher than the serum levels of magnesium. Calcitonin was detected in the CSF of 8% of samples assayed (range, 14 to 175 ng/L [14 to 175 pg/mL]). Two of these five samples had bacteriologically proven meningitis, and two samples were from patients less than 2 months of age. The CSF levels of calcitonin did not correlate with the serum levels of calcitonin. Thus, in children CSF levels of calcium and phosphorus are low, CSF levels of magnesium are higher than the serum levels, and the level of immunoreactive calcitonin is usually not present in the CSF but possibly is elevated in meningitis and early infancy.

Pathogenesis of early neonatal hypocalcemia: studies of serum calcitonin, gastrin, and plasma glucagon.

In 64 maternal-infant pairs, we tested the hypotheses that serum calcitonin, serum gastrin, and plasma glucagon concentrations are elevated in infants at risk for early neonatal hypocalcemia, and that elevated serum gastrin and plasma glucagon result in elevated serum calcitonin and low serum calcium values in neonates. Serum Ca declined significantly in neonates at 24 hours of age, and was inversely correlated with serum calcitonin. Cord serum calcitonin, gastrin, and plasma glucagon concentrations rose significantly at 24 hours of age. Cord calcitonin was significantly higher in preterm compared with term infants, and there was no significant difference between asphyxiated and nonasphyxiated preterm neonates; in term neonates cord calcitonin concentration was inversely correlated with Apgar scores at 1 and 5 minutes. Cord calcitonin was not correlated with cord gastrin or glucagon. Cord and 24-hour gastrin and glucagon values were not related to prematurity; cord glucagon, but not gastrin, was related to birth asphyxia. We conclude that (1) serum calcitonin, gastrin, and plasma glucagon values rise postnatally; cord calcitonin is elevated in preterm and in asphyxiated term infants; serum calcitonin concentration does not correlate with the elevated serum gastrin and plasma glucagon values; and at 24 hours of age, decreased serum Ca is correlated with serum calcitonin, and hence calcitonin might play a role in the pathogenesis of early neonatal hypocalcemia.

Early neonatal hypocalcemia in extremely preterm infants. High incidence, early onset, and refractoriness to supraphysiologic doses of calcitriol.

Nineteen preterm infants born at or before 32 weeks of gestation were studied to determine the dose of calcitriol that would be effective in the prophylaxis of early neonatal hypocalcemia (serum calcium level, less than 7.0 mg/dL [less than 1.75 mmol/L]). In these infants the course of early neonatal hypocalcemia was not modified by calcitriol administration. Serum 1,25-dihydroxyvitamin D level rose in response to intramuscular administration of calcitriol. The incidence of hypocalcemia in these infants was 37% by 12 hours, 83% by 24 hours, and 89% by 36 hours. Thus, in extremely preterm infants, the incidence of early neonatal hypocalcemia is higher and the onset earlier than in larger preterm infants; furthermore, in these infants the hypocalcemia is refractory even to high doses of calcitriol.

Decline in serum calcium, magnesium, and phosphorus values with oral glucose in normal neonates: studies of serum parathyroid hormone and calcitonin.

In 10 normal term infants aged 52 +/- 2.5 hours, serum calcium, magnesium, phosphorus, ionized calcium, parathyroid hormone, and calcitonin were studied at 0, 1/2, 1, and 2 hours after administration of 1.77 +/- 0.08 gm/kg glucose orally over 20 minutes. In response to glucose administration, serum glucose concentration rose and serum P, Ca, and Mg concentrations fell. Serum PTH concentration rose significantly, and blood ionized Ca and pH were unaltered. Serum calcitonin was elevated, as compared with adult values, and did not change. We suggest that in neonates, as in adults, oral ingestion of glucose lowers serum Ca, Mg, and P, and a compensatory rise in serum PTH concentration maintains blood ionized Ca concentration.

Effect of hypocalcemia on cardiac function in very-low-birth-weight preterm neonates: studies of blood ionized calcium, echocardiography, and cardiac effect of intravenous calcium therapy.

Traditionally, in infants, a serum calcium value less than 7.0 mg/dL is considered to impair cardiac function. In very-low-birth-weight infants, we studied the hypotheses that decline in serum calcium to 6.0 mg/dL (1) would not impair cardiac function and (2) ionized calcium would remain greater than 3.0 mg/dL. We also evaluated the effect of calcium infusion on cardiac function. We studied 15 normokalemic and normonatremic infants whose birth weights were 822 to 1,450 g and were less than 32 weeks' gestation. When serum calcium declined to less than 6.0 mg/dL, 18 mg/kg of calcium as 5% calcium gluconate was infused for 10 minutes. Serum total calcium concentration, blood ionized calcium concentration, ECG, and M-mode echocardiogram were obtained on entry into the study, when the infants were hypocalcemic, immediately after treatment with calcium, and eight hours after treatment. Ionized calcium values were calculated based on serum total calcium and serum protein, and corrected calcium values were calculated based on serum total calcium, serum albumin, and blood pH. In all infants, serum calcium value declined to less than 7.0 and in eight infants to less than 6.0 mg/dL. Assessment of heart rate, systolic blood pressure, ejection fraction, left ventricular systolic time interval, right ventricular systolic time interval, fiber shortening index, and left ventricular mean velocity of circumferential fiber shortening showed no significant alteration from baseline during hypocalcemia or in association with intravenous slow bolus infusion of 18 mg/kg of calcium.(ABSTRACT TRUNCATED AT 250 WORDS)

Postnatal changes in calcium-regulating hormones in very-low-birth-weight infants. Effect of early neonatal hypocalcemia and intravenous calcium infusion on serum parathyroid hormone and calcitonin homeostasis.

In very-low-birth-weight (VLBW) infants, we studied the hypotheses that in early neonatal hypocalcemia the serum parathyroid hormone (PTH) concentration would rise; the serum calcitonin (CT) concentration would decline; and, in response to intravenous (IV) calcium (Ca) infusion, the serum PTH concentration would be lowered; and the serum CT concentration would rise. Fifteen infants appropriate for gestational age (age, less than 32 weeks; birth weight, less than 1,500 g) were enrolled in the study. In eight infants in whom the serum Ca level declined to less than 6.0 mg/dL, changes in serum magnesium, phosphorus, PTH, CT, and whole blood ionized calcium (iCa) were evaluated on entry into the study, when serum Ca declined to less than 6.0 mg/dL, immediately after infusion of 18 mg/kg of elemental calcium as calcium gluconate, and at eight hours post-Ca infusion (+ 8 hr). The serum Ca concentration declined from 7.9 +/- 0.6 baseline (mean +/- SE) to 5.2 +/- 0.2 mg/dL pre-Ca infusion and rose to 9.17 +/- 0.74 mg/dL post-Ca infusion and 7.1 +/- 0.5 mg/dL at +8 hr post-Ca infusion. Whole blood iCa declined from 4.82 +/- 0.24 to 3.72 +/- 0.19 mg/dL pre-Ca infusion, rose to 6.68 +/- 0.32 mg/dL post-Ca infusion, and was 4.12 +/- 0.21 mg/dL at + 8 hr post-Ca infusion. The serum P concentration did not change significantly. The serum PTH concentration rose from 116 +/- 17 to 204 +/- 34 pmole/L pre-Ca infusion, declined to 149 +/- 22 pmole/L post-Ca infusion, and was 187 +/- 28 pmole/L at + 8 hr post-Ca infusion. The serum CT concentration was elevated and did not change significantly. Thus, in infants less than 32 weeks' gestation, the serum PTH level rises in early neonatal hypocalcemia and is suppressed by IV Ca infusion; the serum CT level is markedly elevated and is not altered in early neonatal hypocalcemia and does not rise further in response to IV Ca infusion in VLBW infants. We suggest that hypercalcitoninemia occurs in VLBW infants and that serum CT concentrations are unresponsive to changes in serum Ca.

Late infantile tetany and secondary hyperparathyroidism in infants fed humanized cow milk formula. Longitudinal follow-up.

Five full-term infants with birth weights appropriate for gestational age presented with hypocalcemic tetany at 5 to 9 days of age. All infants had been fed Similac 20, a cow milk formula. Initial mean serum calcium (Ca), phosphorus (P), and magnesium (Mg) levels of the tetanic infants were 6.8, 9.5, and 1.6 mg/dL, respectively. The mean serum parathyroid hormone (PTH) level was elevated at 79 mu LEq/mL (adult normal values, less than or equal to 57 mu LEq/mL). Following restoration of normocalcemia with Ca supplements, feeding was reinstituted with Similac 20 in two infants and Similac PM 60/40 in three infants. Serum biochemical and hormonal values were compared with those of 18 exclusively breast-fed infants followed up from three weeks to six months and 14 Similac 20-fed full-term infants followed up from one week to six months. In tetanic infants, serum Ca concentrations became elevated (10.4 +/- 0.05 mg/dL; mean +/- SEM) by six weeks (vs 9.2 +/- 0.3 mg/dL in breast-fed infants) (P less than .001) and serum Mg concentrations (2.26 +/- 0.01 mg/dL) by four weeks (vs 1.92 +/- 0.07 mg/dL in breast-fed infants) (P less than .01). Mean serum P concentrations declined progressively. Mean serum PTH concentrations were elevated and ranged from 74 to 143 mu LEq/mL at two to 16 weeks (vs mean 28 to 35 mu LEq/mL in breast-fed infants (P less than .0001). In 14 formula-fed-nontetanic full-term infants, serum PTH concentrations were intermediate between formula-fed-tetanic and breast-fed infants, mean serum Ca concentrations ranged from 10.2 to 10.4 mg/dL, and mean serum P concentrations declined from 8.3 to 7.1 mg/dL. We speculate that acute hypocalcemic tetany in the study infants was induced by the relatively high P load in cow milk formulas (vs human milk); with the continued P load, secondary hyperparathyroidism continued, maintaining P, Ca, and Mg homeostasis.