CALCIUM INTAKE MAY be one of the many factors that

Transcription

1 X/05/$15.00/0 The Journal of Clinical Endocrinology & Metabolism 90(2): Printed in U.S.A. Copyright 2005 by The Endocrine Society doi: /jc Lack of Effect of Calcium Intake on the 25-Hydroxyvitamin D Response to Oral Vitamin D 3 Rula Goussous, Lingyi Song, Gerard E. Dallal, and Bess Dawson-Hughes Division of Endocrinology, Tufts-New England Medical Center (R.G.); and Bone Metabolism Laboratory (L.S., B.D.-H.), and Biostatistics Department (G.E.D.), Jean Mayer U.S. Department of Agriculture Human Nutrition Research Center on Aging at Tufts University, Boston, Massachusetts This study was conducted to examine the effect of calcium intake on the rise in serum 25-hydroxyvitamin D [25(OH)D] levels in response to supplemental vitamin D 3. Fifty-two healthy older men and women were randomly assigned to take calcium (500 mg twice daily with meals) or placebo tablets for 90 d between October 1 and the end of March. All participants were placed on 800 IU/d (20 g/d) vitamin D 3. Serum 25(OH)D measurements were made at baseline and on d 30, 60, and 90. The mean baseline 25(OH)D values were ng/ml ( nmol/liter) in the calcium group and ng/ml ( nmol/liter) in the control group (P 0.808). CALCIUM INTAKE MAY be one of the many factors that affect 25-hydroxyvitamin D [25(OH)D] levels in the blood. Bell found that 2000 mg/d calcium inhibited the rise in 25(OH)D in response to 100,000 IU vitamin D 3 given daily for 4 d (1). In contrast, Berlin and Bjorkhem (2) found that an increase in calcium intake caused an increase in serum 25(OH)D levels in rats on constant vitamin D intakes. Consistent with this latter finding, 2000 mg/d (50 mmol/d) supplemental calcium increased serum 25(OH)D levels significantly over a 6- to 7-wk period in 28 healthy young men studied during the winter months in Sweden (3). There are several potential mechanisms by which calcium may promote an enhanced 25(OH)D response to oral vitamin D. Increased calcium intake causes a subtle increase in the circulating ionized calcium concentration and a decline in the serum PTH level. Both the higher ionized calcium and the lower PTH concentrations suppress the production of 1,25- dihydroxyvitamin D [1,25(OH) 2 D] (4, 5). The lower 1,25(OH) 2 D levels may increase 25(OH)D levels by increasing the production of 25(OH)D as a result of release of negative feedback inhibition of 1,25(OH) 2 D on 25-hydroxylase (2, 6, 7). Alternatively, lower 1,25(OH) 2 D levels may increase 25(OH)D levels because of a modestly decreased utilization of 25(OH)D as substrate or by delaying the metabolic clearance of 25(OH)D in the liver (8, 9). In support of the latter possibility, Clements et al. (10) examined the impact of increased PTH [and 1,25(OH) 2 D] levels on 25(OH)D metabolism in seven patients with primary hyperparathyroidism before and after surgical removal of their First Published Online November 23, 2004 Abbreviations: 1,25(OH) 2 D, 1,25-Dihydroxyvitamin D; 25(OH)D, 25- hydroxyvitamin D. JCEM is published monthly by The Endocrine Society ( endo-society.org), the foremost professional society serving the endocrine community. 707 The difference in pattern of change in 25(OH)D was not statistically significant (group by time interaction, P 0.651); the calcium group increased ng/ml ( nmol/liter; P < 0.001), and the control group increased ng/ml ( nmol/liter; P < 0.001). The 95% confidence interval for difference in mean increase, calcium vs. control, was ng/ml ( 9.6, 8.7) nmol/liter. In older men and women, the level of calcium intake, within the range of mg/d, does not have an important effect on the rise in serum 25(OH)D that occurs in response to 800 IU (20 g)/d vitamin D 3. (J Clin Endocrinol Metab 90: , 2005) parathyroid adenomas (10). In these patients the hepatic clearance of 25(OH)D was 1.4 times faster before parathyroid surgery than afterward. If calcium intake does modulate serum 25(OH)D levels, then it should be considered when determining the vitamin D intake requirement in clinical studies and in clinical practice. We conducted this study to test the impact of calcium intake, in the amounts usually recommended, on the rise in serum 25(OH)D that occurs in response to a daily oral dose of 800 IU (20 g) vitamin D 3. We also describe 25(OH)D levels achieved as a result of taking oral vitamin D supplements over a 3-month period in the winter. Subjects Subjects and Methods The subjects were healthy, ambulatory men and postmenopausal women, aged 50 yr and older. They were recruited through direct mailings and local newspapers advertisements. Telephone prescreening was used to identify subjects with usual calcium intakes of 600 mg/d (15 mmol/d) or less and to determine general eligibility. Exclusion criteria included any history or disorder known to alter calcium or vitamin D metabolism; therapy with a bisphosphonate, selective estrogen receptor modulators, estrogen, testosterone, glucocorticoids, anticonvulsants, or thiazide diuretics; travel to latitudes less than 35 N, or use of tanning salons 3 months before enrollment or during the study. Screening evaluation included blood and urine tests. Subjects were excluded if their screening 25(OH)D levels were outside the range of ng/ml ( nmol/liter) or their 24-h urinary calcium levels were above 300 mg/d (7.5 mmol/d). Ninety-one subjects were screened; 55 were found to be eligible and were enrolled. A couple withdrew from the study because the spouse was diagnosed with a serious medical condition, leaving 53 subjects who completed the study. One subject was excluded from the analysis because of an unexplained large increase in 25(OH)D level between screening and enrollment from 25 to 39 ng/ml (62.5 to 97.5 nmol/liter). Eight other subjects with 25(OH)D levels in the required range at screening, but with levels up to 31.2 ng/ml (78 nmol/ liter) at enrollment, were included in the analysis. All subjects were studied between October 1, 2003, and March 31, The investigation

2 708 J Clin Endocrinol Metab, February 2005, 90(2): Goussous et al. Calcium and Vitamin D TABLE 1. Clinical characteristics of the 52 study subjects Characteristic Calcium group Control group N Age (yr) Female (%) Caucasian (%) a Height (cm) Weight (kg) Smokers (%) Calcium intake (mg/d) Vitamin D intake (IU/d) No group differences were statistically significant. To convert values for calcium to mmol/d, divide by 40; to convert values for vitamin Dto g/d, divide by 40. a Other participants were: four blacks, three Hispanics, and one eastern Indian. review board at Tufts University approved the study, and written informed consent was obtained from each subject. Experimental methods Of the subjects enrolled, 44% reported current use of a multivitamin (frequency not specified). All subjects were asked to stop taking their own calcium and vitamin D supplements from 1 wk before the screening visit to the end of the study. Subjects were randomly assigned to high or low calcium intake groups throughout the 90-d study. Subjects in the high intake group took 1000 mg (25 mmol) calcium/d (500 mg chewable calcium carbonate twice a day with meals) on d 1 90; subjects in the low intake group took matching placebo supplements twice daily with meals on these days. All participants were placed on 800 IU (20 g) vitamin D 3 daily on d The calcium and placebo tablets were provided by GlaxoSmithKline (Pittsburgh, PA). The vitamin tablets were purchased from a local pharmacy (Nature s Bounty brand, Nature s Bounty, Inc., Bohemia, NY). Fasting blood was drawn at baseline and on d 30, 60, and 90, and 24-h urine collections were returned on d 1 and 90. Compliance was assessed at all visits by pill count and diary checks as recorded by the subjects. Dietary assessments Dietary intake of calcium and vitamin D over the preceding 3 months was assessed at baseline and on d 90 with use of the Fred Hutchinson Food Frequency Questionnaire (11). The questionnaires were selfadministered on site and reviewed for completeness. Biochemical measurements Serum 25(OH)D and 1,25(OH) 2 D levels were measured with RIA kits from Diasorin (Stillwater, MN), and serum PTH was determined by TABLE 2. Biochemical measures in the calcium and control subjects immunoradiometric assay (Nichols Institute Diagnostics, San Juan Capistrano, CA). The coefficients of variation of these assays ranged from %. Serum calcium and phosphorous and urinary creatinine were measured with a Cobas Mira chemistry Analyzer (Roche, Indianapolis, IN), and 24-h urinary calcium was determined with a Nucleus Chemistry Analyzer (Nova Biochemical, Waltham, MA) with a coefficient of variation below 3%. Statistical analysis Subjects were randomized in blocks of 10. Three subjects missed one visit each, and their previous 25(OH)D values were carried forward. Baseline characteristics of the two groups were compared using 2 tests (categorical variables) and two-sample t tests (continuous variables). Variables with more than two time points [serum 25(OH)D and PTH] were examined using repeated measures ANOVA, and variables with only two time points were examined using two-sample t tests. The main statistical analyses were performed with SPSS 11.5 for Windows (version , SPSS, Inc., Chicago, IL). All results are the mean sd unless otherwise stated. All P values were two-sided, and P 0.05 was considered to indicate statistical significance. Results The baseline clinical characteristics of the 52 subjects in the high and low calcium intake groups are shown in Table 1. The groups were similar in age, ethnicity, sex distribution, weight, and dietary calcium and vitamin D intakes. Compliance with vitamin D pills was % in the calcium group (n 22; data missing in one subject) and % in the control group (P 0.83). Compliance with the calcium (n 22; data missing in one subject) and placebo pills was similar in the two groups ( % and %, respectively; P 0.533). The mean baseline 25(OH)D values were ng/ml ( nmol/liter) in the calcium group and ng/ml ( nmol/liter) in the control group (P 0.808; Table 2). There was no statistically demonstrable difference between the two groups in the pattern of change in 25(OH)D (group by time interaction, P 0.651), as shown in Fig. 1. Both groups increased significantly over time (P 0.001). The mean change in 25(OH)D from d 1 90 was ng/ml ( nmol/liter) in the calcium group and ng/ml ( nmol/liter) in the control group [P 0.923; 95% confidence interval for difference in mean increase, calcium-control: 3.8, 3.5 ng/ml ( 9.6, 8.7 nmol/ liter)]. In all subjects, baseline 25(OH)D levels and the change Calcium group (n 23) Serum Day 1 Day 30 Day 60 Day 90 Change (day 90 day 1) 25(OH)D (nmol/liter) PTH (pmol/liter) ,25(OH) 2 D (pmol/liter) c c Calcium (mmol/liter) c c Phosphorus (mmol/liter) c Urinary Ca/Cr ratio (mmol/mol) e c To convert values for serum 25(OH)D to ng/ml, divide by 2.5; serum PTH to pg/ml, divide by 0.106; 1,25(OH) 2 D to pg/ml, divide by 2.4; serum calcium to mg/dl, divide by 0.25; serum phosphorus to mg/dl, multiply by ; 24-h urinary calcium:creatinine (Ca/Cr) ratio to mg/g, divide by a Confidence interval for difference (calcium-control) in mean change (day 90 day 1). b n 28 subjects. c n 22 subjects. d Differs from calcium group on day 1 at P e Differs from calcium group at P

3 Goussous et al. Calcium and Vitamin D J Clin Endocrinol Metab, February 2005, 90(2): FIG. 1. Mean serum 25(OH)D levels ( SEM) in the calcium (dashed line) and control (solid line) groups on d 1, 30, 60, and 90 in the 52 subjects. The pattern of change in the two groups did not differ significantly (by ANOVA, P 0.700). The 25(OH)D levels increased significantly over time in both groups (P 0.001). To convert values for serum 25(OH)D to nanograms per milliliter, divide by 2.5. in 25(OH)D levels (d 1 90) were inversely correlated (r 0.565; P 0.001). Mean serum PTH values were similar in the two groups at baseline (Table 2), and the patterns of change in PTH during the study did not differ significantly in the two groups (group by time interaction, P 0.871). Baseline serum 25(OH)D and PTH levels were significantly correlated (r 0.317; P 0.023). The 24-h urinary calcium/creatinine ratio was higher in the calcium group than in the control group at baseline and increased more in the calcium group than in the control group over 90 d ( vs mmol/mol, respectively; P 0.035; 95% confidence interval for difference in mean increase, calcium vs. control: 7, 192; Table 2). There was no significant group difference in the mean baseline values or in changes during the study in serum 1,25(OH) 2 D, calcium, or phosphorous levels. In all subjects, serum TABLE 2. Continued FIG. 2. Serum 25(OH)D values of the individual subjects on d 1, 30, 60, and 90 in response to supplementation with 800 IU vitamin D 3 daily for 3 months. To convert values for serum 25(OH)D to nanograms per milliliter, divide by (OH)D and 1,25(OH) 2 D were significantly correlated at baseline (r 0.362; P 0.010; n 50), but changes in these two parameters were not significantly correlated. The 25(OH)D values of the individual subjects on d 1, 30, 60, and 90 are shown in Fig. 2. By d 90, 44% of the subjects had reached a 25(OH)D level of 28 ng/ml (70 nmol/liter), and only 21% of the subjects had reached 32 ng/ml (80 nmol/liter). Discussion This study demonstrates that in healthy older men and women, calcium intake in amounts commonly consumed ( mg/d) does not have a clinically important effect on the rise in serum 25(OH)D levels that occurs in response to oral supplementation with 800 IU/d (20 g/d) vitamin D 3. This finding contradicts that of Berlin and Bjorkhem (3), who reported that men had a large 8.4 ng/ml (21 nmol/liter) increase in 25(OH)D in response to calcium supplementa- Control group (n 29) Day 1 Day 30 Day 60 Day 90 Change (day 90 day 1) 95% CI a ( 9.6, 8.7) b b b c ( 0.8, 0.4) b c ( 42.3, 5.0) b c ( 0.04, 0.04) b c ( 0.04, 0.09) d b (7, 192)

4 710 J Clin Endocrinol Metab, February 2005, 90(2): Goussous et al. Calcium and Vitamin D tion. Several design differences in the two studies may explain the discordant results. In the Berlin study, the subjects were younger (mean age, 29 vs. 62 yr), the calcium supplement dose was higher [2000 mg (50 mmol) vs mg (25 mmol)/d], and the starting levels of 25(OH)D were higher [29.2 and 26.8 ng/ml vs and 19.6 ng/ml (73 and 67 nmol/liter vs and 49.1 nmol/liter in the calcium and control groups, respectively] than those in our study. It is possible that these differences allowed for greater absorption of calcium and greater suppression of PTH and 1,25(OH) 2 D levels in the Berlin study. Although the Berlin study did not find that the calcium suppressed serum PTH, calcium did lower the mean serum 1,25(OH) 2 D concentration. As indicated previously, the reduction in serum 1,25(OH) 2 D could have increased the production and/or decreased the metabolism of 25(OH)D (2, 6, 7). Another design difference that could account for the different findings is that the Berlin study did not actually employ vitamin D supplementation and did not estimate the dietary vitamin D intake of the study subjects. Although the subjects in that study were randomized to the calcium or control group, the study was small (14 subjects/group), so there is no certainty that the vitamin D intakes of the two groups were similar during the study. There is also no certainty that dietary calcium intakes were balanced in the two groups, because dietary calcium intake was not estimated, and urinary calcium excretion at baseline was about 50% greater in the control than in the calcium supplement group [ ( sem) vs mg/24 h ( vs mmol/24 h)]. Finally, the Berlin study was carried out for 6 7 wk, a period that is only half that needed to reach a steady state 25(OH)D level after changing vitamin D intake (12) and presumably also the rate of 25(OH)D metabolism. Both studies were conducted at high latitudes (Sweden and Boston) in the winter, when the contribution of sunshine to vitamin D synthesis is minimal (13). Our results also differ from those of Bell et al. (1), who found that 2,000 mg/d calcium blunted the rise in 25(OH)D in response to 100,000 IU (2,500 g) vitamin D daily for 4 d. In that study the calcium supplement increased serum Ca 2 and lowered serum 1,25(OH) 2 D levels, suggesting that calcium inhibited the production of 25(OH)D (1). The different calcium (and vitamin D) doses and durations in the two studies probably account for the different results. Our negative findings are not likely to have resulted from ineffective use of calcium or vitamin D supplements. Compliance with the calcium and vitamin D supplements, to the extent that self-reports and pill counts are reliable, was good at 98% and 97%, respectively. The calcium tablets were chewable, and so were known to disintegrate. The calcium supplements were taken with meals (14, 15), and the dosage was split at 500 mg, twice daily (16), to optimize absorption. The rise in urinary calcium excretion in the calcium group is additional evidence that the supplements were consumed. The vitamin D tablets were not assayed independently, but the increment in 25(OH)D levels in both groups is within the range predicted by Heaney s dose-response study in men [0.28 ng/ml (0.7 nmol/liter)/40 IU (1 g) input of vitamin D 3 ] (17), suggesting that the vitamin D tablets were consumed and that they contained the expected amount of vitamin D. Finally, our two study groups had similar baseline 25(OH)D levels because of the restricted range imposed as an entry criterion. This is important, because the starting level of 25(OH)D is a significant determinant of the change in 25(OH)D in response to vitamin supplementation. It is somewhat surprising that the supplements did not induce significant changes in serum PTH and perhaps in 1,25(OH) 2 D. However, changes in these hormones are variable. In our calcium and vitamin D supplement trial [700 IU (17.5 g)/d vitamin D 3 and 500 mg (12.5 mmol)/d calcium vs. placebo] in older men and women, for instance, there was a significant effect of supplements on serum PTH, but not on serum 1,25(OH) 2 D (18). Barger-Lux et al. (19) found that supplementation with 1,000 IU (25 g)/d vitamin D 3 did not significantly alter serum PTH or 1,25(OH) 2 D levels in men, but higher daily doses of vitamin D (5,000 and 10,000 IU or 125 and 250 g) did. In another study of men treated with 800 IU (20 g) vitamin D 3 daily for 2 months, we observed an increase in serum 1,25(OH) 2 D, but not in serum PTH (20). In neither of the latter two studies did subjects receive supplemental calcium, and their self-selected calcium intakes were not reported (19, 20). Although 25(OH)D levels in both the calcium and control groups increased significantly over time, relatively few subjects reached the level considered by many to be optimal for bone health (21), i.e. 32 ng/ml (80 nmol/liter), after taking 800 IU (20 g) vitamin D daily for 3 months. This dose is twice that recommended by the National Academy of Sciences for men and women the age of those in this study (22). In conclusion, in healthy older men and women, a calcium intake within the range usually consumed and recommended does not appear to have an important effect on the rise in serum 25(OH)D levels that occurs in response to daily doses of 800 IU (20 g) vitamin D 3. Moreover, only 21% of these subjects reached a 25(OH)D level of 32 ng/ml (80 nmol/liter) in the winter after 3 months of supplementation. Acknowledgments We are grateful to the staff of Metabolic Research Unit at Jean Mayer U.S. Department of Agriculture Human Nutrition Research Center on Aging at Tufts University for assistance in carrying out this study. We also appreciate the support of the Division of Endocrinology at Tufts- New England Medical Center, where Rula Goussous is a fellow. Finally, we thank GlaxoSmithKline for providing the calcium and placebo tablets used in this study. Received July 14, Accepted November 10, Address all correspondence and requests for reprints to: Dr. Bess Dawson-Hughes, Bone Metabolism Laboratory at the Jean Mayer, U.S. Department of Agriculture Human Nutrition Research Center on Aging at Tufts University, 711 Washington Street, Boston, Massachusetts bess.dawson-hughes@tufts.edu. This work was supported by the U.S. Department of Agriculture under Agreement Any opinions, findings, conclusions, or recommendations expressed in this publication are those of the authors and do not necessarily reflect the view of the NIH or the U.S. Department of Agriculture. References 1. Bell NH, Shaw S, Turner RT 1987 Evidence that calcium modulates circulating 25-hydroxyvitamin D in man. J Bone Miner Res 2: Berlin T, Bjorkhem I 1987 On the regulatory importance of 1,25-dihydroxyvi-

5 Goussous et al. Calcium and Vitamin D J Clin Endocrinol Metab, February 2005, 90(2): tamin D 3 and dietary calcium on serum levels of 25-hydroxyvitamin D 3 in rats. Biochem Biophys Res Commun 144: Berlin T, Bjorkhem I 1988 Effect of calcium intake on serum levels of 25- hydroxyvitamin D 3. Eur J Clin Invest 18: LoCascio V, Adami S, Galvanini G, Ferrari M, Cominacini L, Tartarotti D 1985 Substrate-product relation of 1-hydroxylase activity in primary hyperparathyroidism. N Engl J Med 313: Eisman JA, Wark JD, Prince RL, Moseley JM 1979 Modulation of plasma 1,25-dihydroxyvitamin D in man by stimulation and suppression tests. Lancet 2: Lore F, Di Cairano G, Periti P, Caniggia A 1982 Effect of the administration of 1,25-dihydroxyvitamin D 3 on serum levels of 25-hydroxyvitamin D in postmenopausal osteoporosis. Calcif Tissue Int 34: Bell NH, Shaw S, Turner RT 1984 Evidence that 1,25-dihydroxyvitamin D 3 inhibits the hepatic production of 25-hydroxyvitamin D in man. J Clin Invest 74: Halloran BP, Bikle DD, Levens MJ, Castro ME, Globus RK, Holton E 1986 Chronic 1,25-dihydroxyvitamin D 3 administration in the rat reduces the serum concentration of 25-hydroxyvitamin D by increasing metabolic clearance rate. J Clin Invest 78: Clements MR, Johnson L, Fraser DR 1987 A new mechanism for induced vitamin D deficiency in calcium deprivation. Nature 325: Clements MR, Davies M, Fraser DR, Lumb GA, Mawer EB, Adams PH 1987 Metabolic inactivation of vitamin D is enhanced in primary hyperparathyroidism. Clin Sci 73: Block G, Woods M, Potosky A, Clifford C 1990 Validation of a self-administered diet history questionnaire using multiple diet records. J Clin Epidemiol 43: Vieth R, Chan PC, MacFarlane GD 2001 Efficacy and safety of vitamin D3 intake exceeding the lowest observed adverse effect level. Am J Clin Nutr 73: Webb AR, Kline L, Holick MF 1988 Influence of season and latitude on the cutaneous synthesis of vitamin D 3 : exposure to winter sunlight in Boston and Edmonton will not promote vitamin D 3 synthesis in human skin. J Clin Endocrinol Metab 67: Heaney RP, Smith KT, Recker RR, Hinders SM 1989 Meal effects on calcium absorption. Am J Clin Nutr 49: Recker RR 1985 Calcium absorption and achlorhydria. N Engl J Med 313: Harvey JA, Zobitz MM, Pak CY 1988 Dose dependency of calcium absorption: a comparison of calcium carbonate and calcium citrate. J Bone Miner Res 3: Heaney RP, Davies KM, Chen TC, Holick MF, Barger-Lux MJ 2003 Human serum 25-hydroxycholecalciferol response to extended oral dosing with cholecalciferol. Am J Clin Nutr 77: Dawson-Hughes Harris SS, Krall EA, Dallal GE 1997 Effect of calcium and vitamin D supplementation on bone density in men and women 65 years of age or older. N Engl J Med 337: Barger-Lux MJ Heaney RP, Dowell S, Holick MF 1998 Vitamin D and its major metabolites: serum levels after graded oral dosing in healthy men. Osteoporos Int 8: Harris SS, Dawson-Hughes B 2002 Plasma vitamin D and 25OHD responses of young and old men to supplementation with vitamin D 3. J Am Coll Nutr 21: Dawson-Hughes B, Heaney RP, Holick M, Lips P, Meunier P, Vieth RP 2004 Vitamin D round table. In: Burckhardt P, Dawson-Hughes B, Heaney RP, eds. Nutritional aspects of osteoporosis, 2nd ed. San Diego: Elsevier; Standing Committee on the Scientific Evaluation of Dietary Reference Intakes 1997 Dietary reference intakes: calcium, phosphorus, magnesium, vitamin D, and fluoride. Washington DC: National Academy Press; 432 JCEM is published monthly by The Endocrine Society ( the foremost professional society serving the endocrine community.