Pubertal Girls Only Partially Adapt to Low Dietary Calcium Intakes


  • Steven A Abrams MD,

    Corresponding author
    1. U.S. Department of Agriculture/Agricultural Research Service, Children's Nutrition Research Center, Department of Pediatrics, Baylor College of Medicine, Texas Children's Hospital, Houston, Texas, USA
    • USDA/ARS Children's Nutrition Research Center, 1100 Bates Street, Houston, TX 77030, USA
    Search for more papers by this author
  • Ian J Griffin,

    1. U.S. Department of Agriculture/Agricultural Research Service, Children's Nutrition Research Center, Department of Pediatrics, Baylor College of Medicine, Texas Children's Hospital, Houston, Texas, USA
    Search for more papers by this author
  • Penni D Hicks,

    1. U.S. Department of Agriculture/Agricultural Research Service, Children's Nutrition Research Center, Department of Pediatrics, Baylor College of Medicine, Texas Children's Hospital, Houston, Texas, USA
    Search for more papers by this author
  • Sheila K Gunn

    1. Section of Endocrinology, Department of Pediatrics, Baylor College of Medicine, Texas Children's Hospital, Houston, Texas, USA
    Search for more papers by this author

  • The authors have no conflict of interest.


We evaluated the effects of low calcium in the diets of young adolescent girls. We measured calcium absorption and excretion using stable isotopes. We found partial adaptation to low intakes but a persistent large deficit relative to recommended intakes. Low calcium intakes pose a substantial risk of inadequate calcium retention.

Introduction: A substantial number of adolescent girls in the United States have habitual calcium intakes <500 mg/day (about 40% of the current recommended intake). The ability to adapt to these very low intakes by increasing calcium absorption and decreasing calcium excretion is not known. We sought to determine the effects of recommended (REC-Ca) versus very low (LO-Ca) calcium intakes on calcium absorption and excretion in white and black girls.

Materials and Methods: Pubertal, but premenarcheal girls, were adapted to low or recommended calcium intakes for at least 2 weeks before each study. Calcium absorption (n = 51) and endogenous fecal calcium excretion (n = 36 of the 51) were determined by dual-tracer stable isotope studies. Subjects were then switched to the other diet for at least 6 weeks, and the study was repeated.

Results: Calcium intake was 389 ± 10 mg/day on LO-Ca and 1259 ± 35 mg/day on REC-Ca diets. Fractional absorption increased from 44.9 ± 1.9% on REC-Ca to 63.4 ± 1.7% on LO-Ca (p < 0.01), but the net calcium absorption remained less than one-half the value on LO-Ca as on REC-Ca. Despite decreases in both endogenous fecal calcium excretion and urinary calcium excretion, net calcium balance was much lower on LO-Ca compared with REC-Ca1 (131 ± 14 versus 349 ± 32 mg/day, respectively; p < 0.001). We found significantly lower urinary calcium excretion but not calcium absorption in black girls compared with white girls.

Conclusions: Very low calcium intakes are only partially adapted to by increased absorption and decreased excretion. Very low calcium intakes place both white and black pubertal girls at substantial risk for inadequate calcium retention.


There is now ample evidence that optimizing whole body calcium retention and bone mineral accretion in young adolescents in the United States requires calcium intakes >1000 mg/day.(1–4) Although several studies have emphasized evaluation of the high end of the calcium intake-retention curve to determine dietary requirements, virtually no attention has been paid to very low calcium intake levels and their consequences.(2,5) This may be a significant problem in that 10-20% of adolescent girls in the United States have calcium intakes of <500 mg/day, which is far below the recommended intake of 1300 mg/day.(2)

Minimal available data in adolescents suggests that calcium retention is less than 100 mg/day when calcium intake is below about 500 mg/day.(6) This is far below the 300 mg/day calcium retention expected during peak rates of growth when calcium intakes is at recommended levels.(2–6) Furthermore, the calcium retention data for intakes below 500 mg/day are based on only a handful of subjects who were studied using mass balance techniques of questionable accuracy performed many decades ago.(2,4,6) More recently, we estimated calcium retention to be <100 mg/day from absorption studies using tracers in adolescents who were provided diets with a calcium content of ∼300 mg/day for 2 weeks.

There is no doubt that adolescents may adapt to very low calcium intakes by increasing fractional absorption and decreasing both urinary and endogenous fecal calcium excretion. However, the relative contributions and total savings to the “calcium economy” of these adaptations are unknown. Potential ethnic differences in this response are not known but may be important because low calcium intakes may be more common, especially among blacks and Asians.(2)

We therefore sought to evaluate the effects in pubertal girls of very low calcium intake on calcium absorption, excretion, and net balance using stable isotope techniques. We compared these values on calcium intakes at the current dietary requirements and a value of approximately one-third of that value in a crossover design. Comparisons were made between groups of black and white girls.



Girls were eligible for the study if they were otherwise healthy and were determined, by physical examination performed by a pediatric endocrinologist (SKG), to be Tanner stage 2 (breast), but had not reached menarche. Girls whose weight was greater than the 97th percentile for age were excluded. Girls were recruited for the study by community advertising.

We screened 60 girls for the study (30 whites and 30 blacks). Nine girls withdrew from the study after completing only one of the two planned studies because of school or personal commitments. Therefore, we ultimately were able to enroll and complete both calcium absorption studies on high and low intake in 51 of the girls. Of these 51, 15 girls failed to complete both sets of in-patient fecal collections; therefore, 36 girls completed both balance studies.

Written informed consent was obtained from a parent or legal guardian for each subject; written assent was obtained from all of the study subjects. The Institutional Review Board of Texas Children's Hospital/Baylor College of Medicine approved this protocol.

Clinical protocol

This study was a crossover study evaluating the effects of high and low calcium intakes on calcium absorption and net calcium retention (balance). Each of the 51 girls completed stable isotope balance studies on a low (LO-Ca) and a recommended (REC-Ca) calcium intake. The choice of which was done first was determined from their usual dietary intake, which was based on a complete 24-h dietary recall aided for accuracy with a list of commonly eaten foods containing 50 mg or more of calcium. This was to ensure that the first study most accurately reflected their usual diet before an intervention. This was felt to be especially important for those girls who had habitually low intakes. One-half of the girls (26) had habitual very low calcium intakes (<500 mg/day) and were therefore started as LO-Ca for the first study. One-half (25) had higher intakes (>500 mg/day), and therefore, their first study was at the REC-Ca intakes. Most of the girls in the normal intake group had habitual intakes >800 mg/day. Girls whose habitual intakes were low maintained their usual diet for 2 weeks before the study. A complete 3-day home weighed diet (2 weekdays and 1 weekend day) was performed before admission to verify usual nutrient intakes. Girls whose intakes were normal and who were to be started on the REC-Ca diet were counseled about ensuring that their diet contained ∼1300 mg/day of calcium for at least 2 weeks before the study.

After completing the first metabolic study, girls whose habitual intake was low were counseled to increase their dietary calcium intake. Because of their very low intakes and avoidance of dairy products, they required daily calcium supplements to achieve recommended calcium intakes. This was provided to them as calcium citrate malate (600 mg/day). Compliance was verified based by dietary recall and by return of the empty supplement container and a supplement record sheet. Some variability in time between the two studies was needed to ensure subjects did not need to miss school for the 6-day inpatient studies. In general, it was ∼6 weeks between the two studies.

In girls whose habitual intake was normal (those whose first study was REC-Ca), it was felt to be unethical to lower their calcium intake for a prolonged period of time. Therefore, they were only asked to do so for 2 weeks before the LO-Ca study by avoiding virtually all dairy products and calcium-fortified foods. During the first week of the adaptation period, a 24-h dietary recall was conducted to assure compliance to the diet. They were admitted for 48 h before the study and placed on a low-calcium diet to help ensure compliance.

All diets during the first metabolic study period were customized to reflect the subject's usual Ca intake. All diets for the subsequent metabolic study reflected prescribed diets that did not reflect usual intakes (from low Ca intake to high Ca intake; from high Ca intake to low Ca intake). A weighed dietary record of all food and drink consumed during the in-patient period was recorded. Nutrient calculations were performed using the Nutrition Data System for Research (NDS-R, version 4.02; University of Minnesota).

Calcium absorption study

Some (n = 15) of the 51 study subjects were unable to stay for both 6-day periods as in-patients and thus could not have two measurements of endogenous fecal excretion performed. Thus, we performed 36 completed sets of two full 6-day balances sets for assessment of all values including endogenous fecal calcium excretion. However, all 51 girls had paired absorption measurements.

Stable isotope studies were conducted as previously described.(7) Each subject was given a breakfast containing ∼350 mg of calcium while consuming the REC-Ca diet. Toward the end of breakfast, the subjects were given46Ca (0.4 μg/kg), which had been premixed (and allowed to equilibrate in the refrigerator for 24 h) with 120 ml of milk. The calcium in the milk was part of the 350 mg in the daily calculations. Subjects consuming the LO-Ca diet were given a breakfast containing ∼150 mg of calcium. With this diet,46Ca (0.4 μg/kg) was premixed with 40 ml of milk. The calcium in this milk was also counted as part of the total 150 mg given with breakfast. One hour after breakfast,42Ca (5 mg) was infused over 10 minutes through a heparin lock catheter. Beginning with breakfast, a complete 5- to 6-day urine and fecal collection was obtained.

Analytical methods

Urine samples were prepared for mass spectrometric analysis, as previously described, using an oxalate precipitation technique.(7) Samples were analyzed for isotopic enrichment with a Finnigan MAT 261 (Bremen, Germany) magnetic sector thermal ionization mass spectrometer. Each sample was analyzed for the ratio of42Ca/43Ca and46Ca/43Ca, with correction for fractionation to the reference44Ca/43Ca ratio. Accuracy and precision of this technique for natural abundance samples compared with standard data are 0.15% or better, depending on the ratio being measured.

Calculations and statistics

Endogenous fecal calcium (EFC) is that portion of the total intestinal calcium (TIC) entering the gut from endogenous sources that does not get absorbed.(7–10) EFC excretion was calculated as the ratio of urine versus fecal recovery of the intravenously administered isotopes.

Calcium balance was calculated as the difference between total absorbed calcium from the dual-tracer method and the sum of the 5-day urinary and endogenous fecal calcium excretion values.

Comparisons of values between study time points were made using ANOVA. Differences in anthropometric and calcium metabolic parameters were evaluated for the complete data sets (Statview 4.5 for Macintosh; Statview Corp., Berkeley, CA, USA). All data are presented as the mean ± SE.


We measured calcium absorption in 51 girls whose anthropometric characteristics at each study time were as shown in Table 1. Calcium intake for the whole group of 51 girls averaged 389 ± 10 mg/day on LO-Ca and 1259 ± 35 mg/day on REC-Ca. Absorption fraction was 63.4 ± 1.7% on LO-Ca versus 44.9 ± 1.9% on REC-Ca. This led to 320 mg lower total calcium absorption on LO-Ca compared with REC-Ca diets (p < 0.0001 for each). Of the 51 girls, 28 were white and 23 were black. Of the 28 whites, 14 had low baseline intakes and 14 had normal baseline intakes. Of the 23 blacks, 12 had normal baseline intake and 11 had low baseline intakes.

Table Table 1.. Anthropometric Characteristics of the Study Subjects (n = 51)
original image

Complete absorption and excretion data for the 36 girls in whom two sets of fecal collections were made are shown in Table 2. Calcium intake was essentially the same for this subgroup of 36 girls as for the entire group of 51 girls (386 ± 14 mg/day on LO-Ca and 1222 ± 42 mg/day on REC-Ca). Calcium intakes and fractional absorptions for the 15 girls who did not perform the fecal collections were similar to those of the 36 who completed them.

Table Table 2.. Calculated Calcium Balance Values in Girls During Early Puberty
original image

There were no differences in fractional absorption of calcium whether the LO-Ca or REC-Ca diet was given first, and therefore, there was no difference in the results based on the girls' usual intakes as low or normal for age. Therefore, results were not segregated based on prestudy intake. This is consistent with the expectation that 2 weeks is adequate for adaptation to low and higher calcium intakes.(3)

We found no significant difference in results for black compared with white girls except for urinary calcium, which was significantly lower in black compared with white girls (Table 3). Fractional absorption of calcium was virtually identical (differences of <2%) on both LO-Ca and REC-Ca intakes. Total calcium absorption was similar on the LO-Ca diet (249 ± 12 versus 243 ± 14 mg/day) for white and black girls. For REC-Ca intakes, total calcium absorption was 587 ± 35 mg/day for whites and 540 ± 47 mg/day for blacks (p = 0.41).

Table Table 3.. Calcium Absorption and Urinary Calcium Excretion (mg/day) on LO-Ca and REC-Ca Intakes: Ethnic Effects
original image

Adaptation to low calcium intake/absorption also occurred through changes in endogenous fecal excretion. An inverse relationship between EFC excretion (Vendo) and fractional absorption (α) was found (r = −0.30; Fig. 1). There was no significant correlation between α at the two studies or between Vendo at the two studies. Vendo was similar between ethnic groups, with a non-physiologically or statistically significant difference of <5 mg/day between black and white girls at both the high and low intakes. Overall, the mean calculated net calcium retention was 48 mg lower on the LO-Ca diet (p = 0.09) and 42 mg lower on the REC-Ca diet (p = 0.51) in white girls.

Figure FIG. 1.

Linear relationship between percentage absorption of calcium and daily endogenous fecal calcium excretion: y = −57.4x + 101.5 (r = −0.30).

In considering the net differences between calcium balance on low and high intake, one approach is to evaluate how adaptation was distributed. In this case, one can calculate that by increasing fractional absorption by 20%, there was a net “adaptation” of 20% of the LO-Ca calcium intake of 390 mg/day, which equals 78 mg/day. Savings in excretion from combined decreased urinary and endogenous fecal excretion were 67 mg/day (Table 2). Therefore, adaptation at LO-Ca was approximately equally distributed between increased absorption and decreased excretion.


There are virtually no data evaluating the effects on calcium absorption and excretion in adolescents of calcium intakes <500 mg/day, although a substantial proportion of adolescents have intakes in this range.(1–5) We found that on diets containing these very low calcium intakes, fractional absorption, as expected, increased substantially. It is likely, however, that, even at very low intakes, it is difficult to regularly exceed 60-70% fractional absorption from dietary sources, even in adolescents.(2–4) Some of this increased absorption is probably caused by the small calcium load used in the low calcium study rather than any changes in vitamin D-mediated calcium absorption.(11) However, because matching calcium loads would not accurately reflect the actual meal differences in calcium in this type of study, these two effects cannot readily be distinguished. On a practical level, there clearly is a maximum adaptation to small load sizes and lowered overall calcium intakes, even at these low intake levels.

In considering the overall adaptation to differing diets, an important outcome is the net calcium balance, also referred to as the net calcium retention. This value represents the net difference in total absorbed calcium with adjustment for differences in urinary calcium excretion. The adaptation of both increased calcium absorption and decreased calcium excretion was not even close to allowing for the level of calcium retention shown by others using longitudinal BMC measurements to occur during pubertal growth in girls.(12,13) A deficit of 220 mg/day of total retained calcium from diets containing very low calcium intakes, if maintained over a prolonged period, would be considerable in terms of total skeletal calcium. It would represent an annual deficit of 80 g/year in calcium accretion or about 5-8% of total skeletal calcium accreted. The strength of a cross-over study such as this is that genetic factors are less of a factor in interpreting the findings. It is clear that the girls, when receiving more calcium, were capable of retaining more calcium. Although the low intake tested here was well below population mean intakes, it represents an intake typical of at least 10% of American adolescent girls, a substantial group to develop this large a bone calcium deficit. Because calcium intakes of adolescents in Asia and Africa are commonly much lower than in the United States, it would be important if these findings were to be replicated in those populations.

Urinary calcium excretion and EFC excretion were similar to each other at both intake levels. An inverse relationship between fractional absorption and endogenous excretion was also reported in adults by Heaney et al.(10) The small difference between the total intestinal calcium between recommended and low intakes suggests that “adaptation” to low intake did not lead to a marked decrease of the total secreted calcium in the intestine. Instead, the greater calcium absorption fraction at low intakes led to increased absorption of the secreted calcium and lower net endogenous secretory calcium losses. It is possible, but unlikely, that adaptation to low intakes might be increased after longer periods of time; however, there is no evidence for this in adolescents.

We found the expected lower level of urinary calcium at both intake levels in black compared with white girls. This has been seen in previous studies of adolescent girls.(14–16) However, unlike our earlier cross-sectional study(14) and that of Bryant et al.,(15) we did not find a greater calcium absorption at either intake level among white and black girls. The explanation for these variations between study results is not obvious, but our current study results are similar to the findings of Bell et al.(16) However, the age of our subjects was ∼2 years younger than those in the study of Bryant et al.(15) Furthermore, in our earlier study, we identified a much greater racial difference in calcium absorption (but not urinary calcium excretion) after menarche than before menarche. In fact, we did not find a significant effect for total calcium absorption in the premenarcheal girls in the earlier study,(14) but did find a significant benefit for urinary calcium in the premenarcheal girls.

Taken together, these findings suggest that a savings to the calcium economy from decreased urinary calcium occurs at a relatively early age in black girls. This difference continues during adulthood and is likely related to different renal handling of calcium.(17) There may be a relatively greater increase in calcium absorption in blacks at the peak of the pubertal growth spurt that disappears later in adolescence and adulthood.(17,18) Regulation of endogenous fecal excretion of calcium is related to gastrointestinal factors and does not depend on ethnicity.(15)

Overall, our findings indicate that the failure to fully adapt to low calcium intakes was present for both black and white girls, although black girls adapted more by decreasing urinary calcium more than white girls at low calcium intakes. We have not evaluated pubertal Hispanic or Asian girls to determine if these findings would include them as well.

We conclude that a substantial proportion of American girls are at risk for low total calcium retention based on their low intakes. Adaptive mechanisms cannot provide the majority of the lost calcium caused by these very low intakes.


The authors thank the nursing staff of the Metabolic Research Unit of the Children's Nutrition Research Center and the General Clinical Research Center of Texas Children's Hospital for caring for the study subjects: Lily Liang and Becky Gorham for technical and dietary support; Dorothy Powledge for study recruitment; and Dr Christopher Branner, Courtney Edwards, Dr Dalia Galicia, Yana Krisman, Deborah Melendez, Elizabeth Napoli, Holly Olvey, Lisa Turner, and Tracey Lipsey for subject management and laboratory assistance. This work is a publication of the U.S. Department of Agriculture (USDA)/Agricultural Research Service (ARS) Children's Nutrition Research Center, Department of Pediatrics, Baylor College of Medicine, and Texas Children's Hospital, Houston, TX. This project has been funded in part with federal funds from the USDA/ARS under Cooperative Agreement 58-6250-6-001, the National Institutes of Health, NCRR General Clinical Research for Children Grant RR00188, and National Institutes of Health HD36591. Contents of this publication do not necessarily reflect the views or policies of the USDA, nor does mention of trade names, commercial products, or organizations imply endorsement by the U.S. Government.