Effects of development on techniques for calcium stable isotope studies in children

We evaluated the effects of age and pubertal status on the tracer excess measured in urine samples collected during the 120 h after the oral and intravenous administration of calcium stable isotopes to 90 children (16 preterm infants and 74 females aged 4.9-16.8 years). Analysis of subjects shows differences in distribution volume and fractional absorption of calcium based on developmental status, with four separate groups being identified. These are: (i) premature infants; (ii) prepubertal girls (Tanner stage 1); (iii) early pubertal girls (Tanner states 2-3); and (iv) late pubertal girls (Tanner stages 4-5). Identification of groups in this fashion allows for adequate dosing in calcium stable isotope studies, with the minimum dose being given to achieve an increase in urinary isotopic content consistent with the precision of the analytical equipment.

Determination of fractional absorption of dietary calcium in humans

Four dual-isotopic label methods for determining true fractional absorption of dietary calcium were compared in 23 subjects. The ratio of the integrals of oral label in a 24-h pooled urine to intravenous label in the same urine is called alpha 24h and was taken as the standard against which the others were compared. alpha Spot is the ratio of the fraction of oral label to the fraction of intravenous label in a single urine specimen; alpha Lag is the ratio of the level of oral label in blood 4 h after administration to the level of intravenous label in blood 2 h after administration. alpha Dec is obtained by deconvoluting response to the intravenous label from the response to the oral tracer. Results were as follows: alpha 24h = 0.273 +/- 0.124, alpha Dec = 0.300 +/- 0.101 (n = 14), alpha Spot = 0.359 +/- 0.179, and alpha Lag = 0.271 +/- 0.103. The Bland-Altman approach for comparison of methods was used to show that results for alpha Spot and alpha Lag can be expected, with a 95% confidence limit, to differ from the value of alpha 24h by 60 and 69%, respectively. The results for alpha Dec were shown to be not only indistinguishable from alpha 24h but identical from a theoretical perspective.

Extraction of magnesium from biological fluids using 8-hydroxyquinoline and cation-exchange chromatography for isotopic enrichment analysis using thermal ionization mass spectrometry

The use of 8-hydroxyquinoline to precipitate magnesium was evaluated as a method for preparing biological samples for isotopic enrichment analysis using thermal ionization mass spectrometry (TIMS). Standard curves in matrices of water, serum, and urine were prepared using varying amounts of 25Mg. The 25Mg/24Mg isotope ratio was measured by TIMS using a silica gel/phosphoric acid technique. Although the total Mg recovered by precipitation from the matrices varied considerably and was dependent on matrix, recovery was sufficient for isotopic enrichment analysis. Urine samples required cation-exchange chromatography (Bio-Rex AG 50W-X8 filter membrane, Bio-Rad Laboratories) prior to precipitation to remove contaminants which interfered with the thermal ionization process. The observed versus expected 25Mg/24Mg enrichments were evaluated using linear regression analysis: water, y = 0.016 + 1.022x; serum, y = 0.5 + 1.097x; urine, y = -0.004 + 0.943x. This method has proven useful for the isolation of magnesium from the biofluids tested.

Calcium kinetics in lactating women with low and high calcium intakes

Absorption of calcium and its mobilization from bone during lactation are important for delivery of calcium to breast-feeding infants; whether calcium intake offsets bone resorption is not known. We hypothesized that calcium absorption is increased in lactation and greater in women on low calcium diets, resulting in similar rates of bone resorption and accretion. Calcium absorption and kinetic indexes were calculated by using two stable isotopic tracers in 8 women; 6 were studied both during lactation and nonlactation. Women consumed low calcium diets, with half receiving supplemental calcium. Intestinal absorption was related to serum 1,25-dihydroxyvitamin D and did not increase during lactation. Despite decreased urinary calcium excretion during lactation, especially in women with low calcium intake, net balance tended to be lower during lactation. Mean residence time decreased and bone resorption exceeded accretion in almost all lactating women. Calcium need for milk production appears to be met by decreased urinary excretion and increased bone resorption, and not by increased intestinal absorption.

Calcium and protein kinetics in prepubertal boys. Positive effects of testosterone

We investigated the effects of 4-6-wk administration of testosterone on calcium and protein metabolism in six healthy prepubertal short boys (mean age +/- SE = 12.9 +/- 0.6 yr). At baseline, subjects received a 4-h infusion of L-[1-13C]leucine and L-[2-15N]glutamine, and were given 42Ca intravenously, and 44Ca PO. Testosterone enanthate (approximately 3 mg/kg) was given I.M. 2 wk apart (two doses n = 5, three doses n = 1), and the study was repeated 4-5 d after the last injection. After testosterone therapy, there were significant increases in serum testosterone and mean peak and total growth hormone concentrations. Net calcium absorption (Va) and retention (Vbal) also increased (Va 13.3 +/- 2.3 vs 21.5 +/- 2.3; mg.kg-1.d-1, Vbal 8.0 +/- 2.1 vs 16.6 +/- 2.5, mg.kg-1.d-1, P < .05 both), as well as Ca's net forward flow into bone and total exchangeable pool (16 and 20%, respectively). The rate of appearance of leucine (an indicator of proteolysis) increased by 17.6 +/- 5.9%, P = 0.036. Leucine oxidation decreased by 48.6 +/- 8.0%, P = 0.004; thus, nonoxidative leucine disappearance, which estimates protein synthesis, increased significantly by 34.4 +/- 7.7%, P = 0.009. Glutamine's rate of appearance also increased (+32%), mostly through enhanced glutamine de novo synthesis (+42%). In conclusion, short term testosterone administration significantly increases calcium's retention and net forward flow into bone in prepubertal humans, as well as whole body estimates of protein and calcium anabolism. These effects may represent a pure androgen effect, an amplification of growth hormone's action or some combination of these factors.

Pathophysiology of hypocalciuria in preeclampsia: measurement of intestinal calcium absorption

OBJECTIVE

To analyze calcium absorption using stable isotopes in patients with preeclampsia and in normotensive controls.

METHODS

Fifteen pregnant subjects were studied: eight with preeclampsia (hypertension and proteinuria) and seven normotensive controls. All patients were ingesting their normal diet. The subjects received two stable calcium isotopic tracers. An oral tracer (44Ca, 0.0124 mmol/kg) was given with milk, while an intravenous tracer (42Ca, 0.00249 mmol/kg) was infused over 7-10 minutes. Calcium concentration was determined by atomic absorption spectrophotometry, and isotope ratios by thermal ionization mass spectrometry from pooled 24-hour urine samples.

RESULTS

No difference was noted in fractional intestinal absorption between preeclamptic subjects (0.282 +/- 0.051) and normotensive controls (0.306 +/- 0.079) (P = .49). However, the fraction of dietary calcium appearing in the urine differed significantly (0.06 for preeclamptic subjects and 0.087 for normotensive controls; P = .008).

CONCLUSIONS

Despite the indirect evidence of others, calcium absorption does not appear to be impaired in patients with preeclampsia. The retention site of the unexcreted calcium is unidentified.

Hypercalciuria in premature infants receiving high mineral-containing diets

We evaluated urinary calcium excretion in 21 premature infants fed either a formula containing high concentrations of calcium (Ca) and phosphorus (P) or human milk fortified with a commercially available human milk fortifier. Dual-tracer stable-isotope Ca absorption studies were performed on all infants. Urinary Ca excretion was not significantly related to Ca or P intake or true Ca absorption. The recovery of the orally administered tracer in the urine was used to evaluate the source of calciuria in study subjects. In almost all subjects, tissue-derived (Vbu), rather than diet-derived (Vou), Ca was the principal source of urinary Ca. Hypercalciuric subjects demonstrated greater Vbu and Vou than nonhypercalciuric subjects. Our data demonstrate that moderate hypercalciuria is common in premature infants whose diets are high in mineral content and that hypercalciuria is not related to inadequate mineral intake or Ca absorption but is related, instead, to losses of both tissue and dietary Ca.

Pubertal changes in calcium kinetics in girls assessed using 42Ca

The rate of calcium deposition in bone (Vo+) and the size of the exchangeable calcium pool in bone (EP) were measured in 50 girls aged 4.9 to 16.7 y using i.v. administered 42Ca. Both Vo+ and EP reached a maximum during early puberty and decreased in late puberty and more than 2 y postmenarche. In contrast, the initial mass of calcium distribution did not decrease in late puberty. The ratio of calcium deposition in bone to the exchangeable pool size, i.e. ko+ = Vo+/EP, was more closely correlated with length of time since menarche (r = 0.83, p < 0.001) than with age (r = 0.65, p < 0.001), weight (r = 0.51, p < 0.001), or body mass index (r = 0.28, p = 0.05). The lower ko+ in late puberty indicates that as bone matures the fraction of the calcium pool that enters bone decreases.

Mineral balance and bone turnover in adolescents with anorexia nervosa

We evaluated seven female adolescents with anorexia nervosa to determine whether calcium metabolism was affected by their disorder. We measured calcium absorption, urinary calcium excretion, and calcium kinetics, using a dual-tracer, stable-isotope technique during the first weeks of an inpatient nutritional rehabilitation program. Results were compared with those from a control group of seven healthy adolescent girls of similar ages. The percentage of absorption of calcium was lower in subjects with anorexia nervosa than in control subjects (16.2% +/- 6.3% vs 24.6% +/- 7.2%; p < 0.05). Urinary calcium excretion was greater in subjects with anorexia nervosa than in control subjects (6.4 +/- 2.5 vs 1.6 +/- 0.7 mg.kg-1 x day-1; p < 0.01) and was associated with bone resorption rather than calcium hyper-absorption. Calcium kinetic studies demonstrated a decreased rate of bone formation and an increased rate of bone resorption. These results suggest marked abnormalities in mineral metabolism in patients with anorexia nervosa. From these results, we hypothesize that improvement in bone mineralization during recovery from anorexia nervosa will require resolution of hormonal abnormalities, including hypercortisolism, in addition to increased calcium intake.

Calcium absorption and metabolism in children with juvenile rheumatoid arthritis assessed using stable isotopes

OBJECTIVE

To assess calcium intake, absorption, urinary excretion and the fraction of urinary calcium originating from bone and diet in patients with juvenile rheumatoid arthritis (JRA).

METHODS

A dual tracer stable isotope technique was used to study 6 girls and 3 boys with JRA.

RESULTS

Fractional absorption in the 6 girls, ages 4-9, with JRA was significantly lower than that in 10 similar, healthy girls (22.6 +/- 4.7% vs 30.4 +/- 8.4%, p = 0.033). Urinary calcium excretion tended to be higher in the girls with JRA than in controls, (2.9 +/- 1.5 vs 1.6 +/- 1.7, p = 0.15). The urinary calcium in patients with JRA was derived principally from bone, and there was no increase in diet derived urinary calcium. One of the boys with new onset JRA was markedly hypercalciuric and in negative calcium balance (-222 mg/day).

CONCLUSION

Our data show that hypercalciuria in patients with JRA results from bone resorption, not hyperabsorption of dietary calcium and suggest that increases in calcium intake may benefit children with JRA.

Calcium kinetics in the hyperprostaglandin E syndrome

Metabolic investigations, including the use of stable isotopes of calcium, were used to study calcium kinetics in three children with the hyperprostaglandin E syndrome. The studies were performed both during indomethacin treatment and in the absence of therapy. Off therapy, each child had hypercalciuria (mean urinary calcium excretion 0.478 mM/kg/d), hyperprostaglandinuria, and elevated serum calcitriol concentration. All had diminished bone density and were euparathyroid. Indomethacin treatment was associated with a marked reduction in serum calcitriol concentration, as well as decreased prostaglandin E excretion. Mean urinary calcium excretion fell to 0.135 mM/kg/d. The stable isotope studies defined two components to the hypercalciuria of this disease: an indomethacin-sensitive dietary contribution and a relatively indomethacin-resistant bone resorptive element. Bone densitometry confirmed the presence of the resorptive element by demonstrating skeletal demineralization.

Bone mineralization outcomes in human milk-fed preterm infants

We evaluated bone mineralization by single photon absorptiometry at 2 y in a cohort of preterm infants studied since birth. Infants were fed human milk fortified with Ca [to achieve 80 mg/dL (19.96 mmol/L)] and P [40 mg/dL (12.91 mmol/L)] from wk 2 through 8 after birth. After hospital discharge, infants were divided into two groups (HM and F) determined by the timing of the introduction of cow milk-based formula. Mid-radius bone mineral content (BMC) was assessed in 10 infants who were breast-fed (HM) for a minimum of 2 mo after hospital discharge and 11 who were bottle-fed (F). The mean duration of human milk-feeding differed by design between HM and F groups (31 +/- 15 versus 11 +/- 3 wk, respectively). Although we had observed previously that group F had significantly greater BMC values at 16, 25, and 52 wk compared with values in group HM, we found similarities in BMC values (180 +/- 30 mg/cm) between groups at 2 y. The 2-y cohort comprised healthy infants and the groups had similar birth weights, lengths of gestation, and values for weight (10.8 +/- 1.1 kg), length (82 +/- 2 cm), and bone width (7.8 +/- 1.1 mm). Follow-up outcomes at 2 y in preterm infants fed fortified human milk in hospital suggest that if they continue to receive human milk after hospital discharge, radius BMC will "catch-up" to that of similar infants given formula in the posthospitalization period.

Developmental changes in calcium kinetics in children assessed using stable isotopes

Total exchangeable calcium pool size (TEP) and bone calcium accretion rate (Vo+) were measured using stable isotopes in healthy children and young adults. 42Ca or 46Ca was given intravenously to 10 children aged 10 months to 14 years and 3 women aged 23-33 years. Calcium kinetic parameters were determined using a two- or three-exponential curve of the resultant serum and urine tracer excesses. These data were compared with previously reported (radiotracer) kinetic studies of 21 children and 5 adults without known bone disease. Current results are comparable to those previously obtained, and the data from all studies were analyzed together. Total Vo+ was significantly greater in children aged 3-16 years than in adults (2.8 +/- 1.6 versus 0.7 +/- 0.2 g/day, p less than 0.01). Both TEP and Vo+ were significantly correlated to age independently of variations in body weight (p less than 0.01 for each). The ratio ko+ = Vo+/TEP was greater in children than adults (0.36 +/- 0.15 versus 0.12 +/- 0.03 day-1, p less than 0.001). These data demonstrate increased bone flow of calcium associated with increases in exchangeable calcium pools in children compared to adults. Vo+ and TEP may be maximum in early adolescence, associated with peak rates of net calcium accretion. The use of stable isotopes permits the safe evaluation of calcium kinetics in patients of all ages.

Calcium and phosphorus deficiencies affect mineral distribution in neonatal miniature piglets

Current efforts to monitor the mineral status of preterm infants fed human milk may not provide sufficient information on the distribution of body minerals. To investigate the body distribution of calcium and phosphorus during various degrees of mineral deficiency, neonatal miniature piglets were raised for 2 wk on diets differing only in calcium and phosphorus. Groups A, B, and C were fed 100%, 60%, and 20%, respectively, of the recommended amounts of calcium and phosphorus that, when adjusted for rates of growth, approximated the range of dietary intakes of preterm infants. Group C manifested biochemical and body-composition evidence of mineral deficiency when compared with group A: lower serum phosphorus; higher serum alkaline phosphatase activity; less fat-free tissue, calcium, and phosphorus in tibiae, vertebrae, and whole carcasses. Neonatal miniature piglets are useful for studying mineral deposition during mineral deficiency in preterm infants.

Dual tracer stable isotopic assessment of calcium absorption and endogenous fecal excretion in low birth weight infants

Using a dual tracer (44Ca orally and 46Ca i.v.) stable isotope technique, true dietary Ca absorption, endogenous fecal Ca excretion, and net Ca retention were measured in 12 low birth weight (1426 +/- 260 g) infants fed a high Ca-containing formula. Endogenous fecal Ca excretion averaged 7.2 +/- 4.1% of intake, and exceeded 10% of intake in three infants. Net Ca retention, 103 +/- 38 mg/kg/d, was consistent with previous studies of Ca retention obtained using mass balance techniques and correlated closely (r = 0.98, p less than 0.001) with true Ca absorption but not with endogenous fecal excretion (r = -0.40, p = 0.19). Although endogenous fecal excretion may represent a significant source of Ca loss for some low birth weight infants, these data suggest that net Ca retention in low birth weight infants fed a high Ca-containing formula is primarily determined by the total dietary Ca absorbed.

Stable isotopic measurement of endogenous fecal calcium excretion in children

Using a stable isotopic technique in which 42Ca was administered via a bolus injection, we measured endogenous fecal calcium excretion, Vf, in five healthy children, aged 3-14 years. The Vf averaged 1.4 +/- 0.4 mg/kg/day, and was lower than urinary Ca excretion (Vu) in four of the five children. These results for Vf are consistent with previously reported results for Vf in healthy adults and much lower than those reported in premature infants. These results may be useful in understanding developmental changes in Ca metabolism and in interpreting dual tracer Ca isotope studies in children.

Unequal distribution of a stable isotopic calcium tracer between casein and whey fractions of infant formulas, human milk and cow's milk

Measurement of calcium absorption with tracers assumes a complete equilibration of tracer with milk calcium. In this study, the equilibration of tracer between the micellar casein and soluble fractions (primarily whey) of infant formulas, human milk and cow's milk was measured in vitro using milk samples enriched with 42Ca and analyzed by thermal ionization mass spectrometry. Incomplete equilibration of tracer occurred with the micellar casein fraction of all milks. The least equilibration with micellar casein was found with premature infant formula, for which the ratio of slopes of the equilibration lines (whey/casein) was 8.5/1. These differences may be due to Ca-casein binding in cow's milk-based formulas. The effects of the lack of tracer equilibration in vivo cannot be determined. However, unequal bioavailability of casein- vs. whey-bound Ca may exist.

Recent studies of human calcium metabolism using stable isotopic tracers

Stable isotopes of calcium are used safely as tracers for calcium in human populations ranging in age from infants to postmenopausal women. Thermal ionization mass spectrometry is used to measure calcium isotope ratios with relative accuracies of about 1% for natural abundance ratios at precisions of about 1% relative to the mean. Perturbations of natural abundance ratios are determined for the calcium in blood, urine, and feces with a limit of detection of about 2 delta % excess. The mathematical rationale for clinical studies of fractional absorption of dietary calcium and the kinetics of calcium's internal distribution are presented.

Bone mineral content reflects total body calcium in neonatal miniature piglets

We measured bone mineral content (BMC) in 18 neonatal miniature piglets by single photon absorptiometry, total body calcium (TBC) by total body neutron activation analysis, growth, and serum indices of mineral status (calcium, phosphorus, alkaline phosphatase activity). Measurements were begun on day 6, when the piglets were weaned, and were continued to day 19. After weaning, the piglets were assigned randomly to receive one of three diets which differed only in their concentrations of calcium and phosphorus: 100% of the recommended level (diet A), 60% (diet B), and 20% (diet C). No differences were observed among groups during the 19-day study, either in weight gain (48 +/- 2 g/day) or increment in crown-rump length (2.4 +/- 0.2 cm/wk). BMC correlated significantly (p less than 0.001) with TBC at 6 (r = 0.83), 13 (r = 0.77), and 19 (r = 0.93) days. BMC correlated significantly (p less than 0.001) with the ash weight (r = 0.87) and calcium content (r = 0.90) of the corresponding tibial bone segment. Anthropometric parameters and serum indices of mineral status did not predict TBC as accurately as did BMC measurements. We observed a range in BMC measurements in this study that was similar to the range reported for infants in the 1st yr of life. The high correlation between BMC and TBC suggested that BMC is useful in the assessment of mineral status in infants.

Bioavailability of calcium and phosphorus in human milk fortifiers and formula for very low birth weight infants

Growth, 96-hour balance of nutrients (nitrogen, fat, calcium (Ca), phosphorus (P), and magnesium), metabolizable energy, and serum biochemical markers of mineral status (Ca and P concentrations and alkaline phosphatase activity) were measured in 22 very low birth weight infants to investigate the bioavailability of minerals from specialized formula and from human milk fortifiers. The intakes of Ca and P were similar between group FORM ("Preemie" SMA) and group CMF (1:1 wt/wt, human milk and Similac Natural Care or Similac Special Care). The intakes of nitrogen, energy, fat, and magnesium differed between groups. Group CMF had significantly greater fecal losses and significantly lower absorption and retention of Ca and P in comparison with those of group FORM. Retention of Ca and P in both groups, however, was greater than 25% below intrauterine estimates of accretion. Retention rates of Ca, P, and magnesium were not correlated with their respective intakes. Weight gain during the balance study and during the entire study interval was significantly less in group CMF. The ratio of Ca retention to either weight gained or nitrogen retained was lower in group CMF, which suggested that the low retention of Ca was related less to the slower rate of growth in these infants than to their greater fecal losses of Ca. Although the cause of the greater fecal losses of Ca and P in this group is unclear, the data suggest an insolubility of the mineral sources. Our results indicate that sole reliance on the absolute mineral concentrations of the milk selected for very low birth weight infants may be unrealistic; the bioavailability of Ca and P from particular mineral sources should be evaluated.

Mineral balance studies in very low birth weight infants fed human milk

Mineral homeostasis often is disrupted in the very low birth weight (VLBW) infant fed either human milk or commercial formula that contains insufficient quantities of available calcium (Ca) and phosphorus (P). Alterations in mineral homeostasis include abnormal patterns of serum (Ca and P concentrations and alkaline phosphatase activity) and urine (Ca and P) biochemical markers, low net Ca and P retentions in comparison with intrauterine estimates of mineral accretion, and decreased bone mineral content. A two-phase study was conducted in our laboratory to test for these alterations in mineral homeostasis. In phase 1, VLBW infants fed a preparation of fortified human milk (either human milk-derived fortifier I or II or cow milk-derived fortifier) or cow milk-based formula specially designed for VLBW infants were evaluated during their hospitalization. In phase 2, after hospitalization, these infants were evaluated during the first 6 months of life when fed either their mother's milk or routine formula exclusively. The bioavailability of Ca and P from the tested preparations varied widely. Although the fortification of human milk resulted in both an improved biochemical pattern and net retention of Ca and P, optimal intrauterine mineral accretion was not achieved in any group tested. Longitudinal assessments of bone mineralization, by single photon absorptiometry, demonstrated that human milk-fed former VLBW infants had reduced bone mineral content. These investigations suggest that former VLBW infants fed human milk exclusively may be at risk for Ca and P deficiencies.