Dietary calcium and vitamin D intake in elderly women: effect on serum parathyroid hormone and vitamin D metabolites 1-3

Transcription

1 Dietary calcium and vitamin D intake in elderly women: effect on serum parathyroid hormone and vitamin D metabolites 1-3 H Karimi Kinyamu, J Christopher Gallagher, Karen A Rafferty, and Kurt E Balhorn ABSTRACT In this study, the effect of dietary calcium and vitamin D on serum parathyroid hormone and vitamin D metabolites was measured in 376 free-living women aged y. Mean calcium intake in both groups was close to the recommended dietary allowance of 800 mg/d. Mean vitamin D intake in the 245 women not taking vitamin D supplements was 3.53 g/d (141 IU/d), which is below the recommended dietary allowance of 5 g/d (200 IU/d). To test the hypothesis that vitamin D is more important than calcium in reducing serum parathyroid hormone, the source of dietary calcium intake was subdivided into milk, which is fortified with vitamin D, and nonmilk sources. The serum parathyroid hormone concentration was inversely correlated with calcium intake derived from milk (r = 0.20, P < 0.01) but not from nonmilk sources (r = 0.06). Furthermore, serum calcidiol correlated with milk calcium intake (r = 0.35, P < 0.001) but not with nonmilk calcium intake (r = 0.10). Multivariate analysis showed a significant effect of season on serum calcidiol but not on serum parathyroid hormone. Serum parathyroid hormone was inversely correlated with serum calcidiol (r = 0.33, P < 0.001) and the regression predicted that mean serum parathyroid hormone would be reduced in the elderly to concentrations considered normal in the young when serum calcidiol is 122 nmol/l (49 ng/ml); this would require a much higher recommended dietary allowance for vitamin D than 5 g/d (200 IU/d). Am J Clin Nutr 1998;67: KEY WORDS Calcium intake, vitamin D intake, milk intake, vitamin D metabolites, vitamin D deficiency, serum parathyroid hormone, serum calcidiol, serum calcitriol, calcium absorption, elderly women INTRODUCTION Nutrition plays a role in the etiology and pathogenesis of senile osteoporosis. Two of the most important nutrients for bone health are calcium and vitamin D. For elderly women aged > 65 y, the recommended dietary allowance (RDA) for calcium is 800 mg/d and for vitamin D is 5 g/d (200 IU/d) (1). According to many nutritional surveys, a high proportion of the elderly living in North America consume less than the RDA for both calcium and vitamin D (2 5). In addition, although calcium intake may be normal, calcium absorption is less efficient in the elderly, thus limiting the amount of calcium absorbed from the diet (6). Adequate stores of vitamin D are essential for optimal calcium absorption. Elderly subjects are at risk of vitamin D deficiency because of insufficient dietary vitamin D intake (7, 8), inadequate sunlight exposure (9), and impaired renal synthesis of calcitriol (10 12). The best clinical measure of vitamin D status is the serum calcidiol concentration. In European countries, serum calcidiol concentrations < 30 nmol/l (12 ng/ml) have been defined as deficient (13) and serum calcidiol concentrations of 17.5 nmol/l (7 ng/ml) are commonly found (13 16). In North America, the prevalence of vitamin D deficiency is rarely reported in elderly ambulatory women (17, 18), and higher serum calcidiol concentrations (between 25 and 125 nmol/l, or 10 and 50 ng/ml) are usually found (17 20). Many studies have focused on the detrimental effects of calcium deficiency on bone. However, severe vitamin D deficiency causes malabsorption of calcium and osteomalacia and increases the risk of fractures. In contrast, a relative vitamin D deficiency or a low concentration of serum calcidiol is often associated with secondary hyperparathyroidism, which contributes to age-related bone loss. No studies have investigated the effect of the source of dietary vitamin D on serum calcidiol and serum parathyroid hormone (PTH) in freeliving elderly women. The aim of the study was to investigate whether it is the calcium or the vitamin D content of milk that is associated with the lower serum PTH concentrations in elderly women than in young women. To separate the effect of calcium from that of vitamin D, and because milk is the predominant source of dietary vitamin D, calcium intake was divided into milk and nonmilk sources. SUBJECTS AND METHODS Subjects The cross-sectional data presented in this paper were derived from baseline information collected at the Omaha site on women aged y who entered a multicenter osteoporosis trial (STOP- IT: Sites Testing Osteoporosis Prevention/or Intervention). All 1 From the Bone Metabolism Unit, Creighton University School of Medicine, Omaha. 2 Supported by NIH grants UO1-AG10373 and RO1-AG Address reprint requests to JC Gallagher, Bone Metabolism Unit, Creighton University School of Medicine, 601 North 30th Street, Room 6718, Omaha, NE jcg@creighton.edu. Received April 11, Accepted for publication August 6, Am J Clin Nutr 1998;67: Printed in USA American Society for Clinical Nutrition

2 DIETARY CALCIUM AND VITAMIN D IN ELDERLY WOMEN 343 women were volunteers who responded to advertisements in local newspapers or to mass mailing of letters inviting them to participate in a 3-y study. All women were recruited and enrolled in the study between November 1992 and February A larger proportion of the women (53%) were recruited in the winter than in the summer (30%) months. The subjects included 472 white women, 11 black women, 4 Hispanic women, 1 Asian woman, and 1 woman of mixed race. Subjects were excluded if they were taking medications or had diseases thought to influence calcium or phosphorus metabolism. Four hundred eighty-nine women were enrolled into the study; however, 99 women were excluded because they were taking diuretics at the time of the baseline tests, 13 because they did not complete their 7-d food diaries, and 1 because she had a serum creatinine concentration > 0.12 mmol/l (1.4 mg/dl). Of the 376 remaining subjects, 131 had a recent history of taking vitamin D supplements and 245 had no history of taking any medications known to affect calcium metabolism. All women were free-living and had a normal score for activities of daily living. The protocol was approved by the Creighton University Institutional Review Board. Dietary intake Dietary intake data were collected by using 7-d food diaries. Participants were instructed carefully by a dietitian to complete a 7-d food diary and a nutrient supplement record. Plastic food models (NASCO, Fort Artinson, WI) were used to help participants better estimate the quantities consumed. Average daily calcium and vitamin D intakes were calculated by using the FOOD PROCESSOR II PLUS nutrition and diet analysis system (version 5.1; Esha Research, Salem, OR). Calcium absorption test Calcium absorption was measured in a fasting state after oral administration of Bq (5 Ci) 45 Ca (Amersham, Arlington Heights, IL) in 100 mg CaCl 2 carrier given in a total of 250 ml distilled water (6). A blood sample was collected 2 and 3 h after the oral dose. 45 Ca activity was counted in 2 ml serum with a 1900 CA Tricarb Liquid Scintillation Analyzer (Packard Instrument, Meriden, CT). A parallel standard taken from the patient s dose before ingestion was counted at the same time. Calcium absorption was expressed as a percentage of the actual dose per liter of blood (%AD/L) and corrected for body mass index (BMI; in kg/m 2 ). Biochemical analyses Fasting blood and spot urine samples were collected before the calcium absorption test. Blood specimens were allowed to clot and were then centrifuged at 4 (C for 15 min at 2056 g to separate serum. All samples were stored frozen at 70 (C until analyzed. All serum and urine measurements were performed in fresh samples. Serum ionized calcium and serum and urine creatinine were analyzed by using automated procedures (Chemistry Analyzer; Nova Nucleus, Waltham, MA). Serum calcidiol was measured with a competitive binding assay (20) after extraction and purification of serum on Sep-Pak cartridges (Waters Associates, Milford, MA) (21). The limit for detection for the assay is 12.5 nmol/l (5 ng/ml) and our interassay variation was 5%. Serum calcitriol was measured with a nonequilibrium radioreceptor assay (Incstar Corp, Stillwater, MN) by using calf thymus receptor after extraction and purification of the serum on a nonpolar C 18 OH octadecylsilanol silica cartridge (22, 23). The limit of detection for the assay is 12 pmol/l (5 pg/ml) and our interassay variation was 10%. Serum intact PTH was measured with the Allegro immunoradiometric assay (Nichols Institute, San Juan Capistrano, CA) (24). The limit of detection for the assay is 1 ng/l (1 pg/ml) and our interassay variation was 3.5%. Bone markers Serum osteocalcin was measured by radioimmunoassay (Incstar Corp). The limit for detection for the assay is 0.78 g/l (0.78 ng/ml) and our interassay variation was 5%. Urine collagen crosslinks were measured by enzyme-linked immunosorbent assay (Osteomark International, Seattle) as N-telopeptides, which is a specific marker for bone type 1 collagen. Statistical analysis Data were analyzed with the SPSS statistical package for Windows (SPSS Inc, Chicago). The 376 women were divided into two groups: 245 women who did not take any vitamin D supplements and 131 women who were taking vitamin D supplements regularly. Differences between the descriptive and biochemical measurements in the two groups were tested by using Student s t test. Simple linear regression methods and Pearson correlation coefficients were used to examine the association between calcium absorption and the following: calcium intake, vitamin D intake, milk consumption, vitamin D metabolites, and PTH. The effect of milk consumption on serum vitamin D metabolites and serum PTH was examined further by testing the effect of nonmilk calcium on these variables. Nonmilk calcium intake was calculated by subtracting milk calcium from dietary calcium intake. Multivariate regression analyses were used to determine the most important predictors of serum calcidiol and serum PTH concentrations. Because there is seasonal variation in serum vitamin D metabolite and serum PTH concentrations, general factorial analysis of variance (ANOVA) was used to test the interaction effect between season and milk calcium intake or season and nonmilk calcium intake (comparison of season slopes) on serum PTH and serum calcidiol. For this purpose, two distinct seasons were defined. The summer season was defined as the months of June to October and the winter season was defined as the months of December to April. May and November were regarded as transition months and the results for 38 women in these 2 mo were not included in the seasonal analyses. RESULTS Biochemical and dietary intake characteristics of the study population Baseline biochemical and dietary intake characteristics of the 376 women, who were divided into two groups (245 not taking vitamin D supplements and 131 taking vitamin D supplements), are shown in Table 1. There were no significant differences between the two groups in age or BMI. In addition, the mean dietary calcium intake of the two groups was not significantly different: 666 and 660 mg/d, respectively. The group not taking vitamin D supplements, however, had a mean total calcium intake of 704 mg/d compared with an intake of 818 mg/d in the group taking vitamin D supplements (P < 0.05). Calcium from

3 344 KINYAMU ET AL TABLE 1 Mean differences in descriptive and biochemical data between groups taking and not taking vitamin D supplements 1 No vitamin D Vitamin D Variable (n = 245) (n = 131) Age (y) 71 ± 3 71 ± 4 BMI (kg/m 2 ) 27 ± 4 26 ± 4 Total calcium intake (mg/d) 704 ± ± 366 Dietary calcium intake (mg/d) 666 ± ± 265 Milk calcium intake (mg/d) 235 ± ± 181 Nonmilk calcium intake (mg/d) 431 ± ± 160 Total vitamin D intake (IU/d) 141 ± ± 80 ( g/d) 3.53 ± ± 2.0 Dietary vitamin D intake (IU/d) 141 ± ± 80 ( g/d) 3.53 ± ± 2.0 Milk vitamin D intake (IU/d) 78 ± ± 60 ( g/d) 1.96 ± ± 1.5 Serum calcidiol (nmol/l) 73.6 ± ± 28.2 Serum calcitriol (pmol/l) 83.3 ± ± 18.5 Serum PTH (ng/l) 36.7 ± ± 13.2 Serum ionized calcium (mmol/l) 1.24 ± ± 0.04 Serum creatinine (mmol/l) ± ± Serum osteocalcin ( g/l) 3.87 ± ± h Urine calcium excretion (mmol/d) 3.14 ± ± h Urine N-telopeptides (nmol BCE/mmol creatine) 51.9 ± ± 21.9 Calcium absorption (%AD/L, BMI corrected) ± ± x ± SD. PTH, parathyroid hormone; BCE, bone collagen equivalents; AD, actual dose. 2 Significantly different from vitamin D, P < 0.05 (Student s t test). 3 One subject with an unusually high N-telopeptide value of 777 nmol BCE/mmol creatinine was excluded from the mean. milk accounted for 30% of dietary calcium. Average milk consumption was 240 ml/d (approximately three-fourths of an 8- oz glass) and 10% of the elderly did not consume any milk. Mean daily dietary vitamin D intakes were not significantly different between the two groups: 3.53 g/d (141 IU/d) in those not taking vitamin D supplements and 3.4 g/d (136 IU/d) in those taking vitamin D. In the 131 subjects taking a multivitamin containing 10 g vitamin D/d (400 IU/d), the estimated mean vitamin D intake was 13.4 g/d (536 IU/d). The estimated sources of vitamin D were as follows: 51% milk, 18% fish, 9% fortified cereals, 8% meats, 7% eggs, and 7% others. Association among calcium intake, milk consumption, vitamin D intake, parathyroid hormone, and vitamin D metabolites To compare the effects of dietary calcium or vitamin D on serum PTH, only data in the 245 women not taking supplementary vitamin D were analyzed. The serum PTH concentration was inversely correlated with dietary calcium intake (r = 0.19, P < 0.01) in all women. When the dietary calcium intake was divided into milk and nonmilk sources of calcium, the serum PTH concentration was inversely correlated with milk calcium intake but not with nonmilk calcium intake (Figure 1). Exclusion of the person with a high milk intake did not change the regression lines. The relation between serum PTH and calcium intake is shown separately for winter and summer months. Results were available for 76 women in the summer months, 131 women in the winter months, and 38 in the transitional spring and autumn months. Although milk calcium appeared to have a greater effect on serum PTH in the summer months than in the winter months, the ANOVA showed only a significant effect of milk calcium on serum PTH (P < 0.001), no effect of season (P > 0.753), and no significant interaction effects of milk calcium and season (P > 0.142). There was no significant effect of nonmilk calcium or season on serum PTH (Figure 1). The serum calcidiol concentration was significantly correlated with dietary calcium intake (r = 0.33, P < 0.001) in all women; however, when women were separated into groups on the basis of milk and nonmilk sources of calcium, the serum calcidiol concentration was significantly correlated only with milk calcium intake and not with nonmilk calcium intake (r = 0.10) (Figure 2). There was an independent and significant effect of milk calcium and season (P < 0.001) on serum calcidiol but no interaction effect of milk calcium and season (P > 0.404) as shown by ANOVA. There was a significant (P < ) difference of 16.3 nmol/l (6.5 ng/ml) in serum calcidiol between the intercepts of summer and winter months in the milk calcium source and a difference (P < 0.001) of 14.3 nmol/l (5.7 ng/ml) between the intercepts for the summer and winter months in the nonmilk calcium source. The serum calcidiol concentration was also significantly correlated with dietary vitamin D intake and the serum PTH concentration was inversely correlated with dietary vitamin D intake (Figure 3). Serum PTH concentration was inversely correlated with serum calcidiol (Figure 4). Because there were no significant differences between the regression slopes of serum PTH on serum calcidiol in women taking or not taking vitamin D supplements, data were pooled for the 376 women.

4 DIETARY CALCIUM AND VITAMIN D IN ELDERLY WOMEN 345 FIGURE 1. Correlation between serum parathyroid hormone (PTH) and dietary calcium intake derived from milk or nonmilk sources in 245 women. Serum PTH was inversely correlated with milk calcium intake (r = 0.20, P < 0.01). There was no significant difference between samples collected in the summer (r = 0.34, P < 0.01) and those collected in the winter (r = 0.18, P < 0.05). There was no correlation between serum PTH and nonmilk calcium intake (r = 0.06) and no seasonal difference. Multiple regression analysis The main determinants of serum calcidiol and serum PTH were examined by multiple regression analysis in those taking and not taking vitamin D supplements (Table 2). For serum calcidiol, the variables entered in the model as independent predictors were age, calcium intake, milk calcium intake, nonmilk calcium intake, dietary vitamin D, season, and body weight. In the 245 women not taking vitamin D supplements, milk calcium intake was the main determinant of serum calcidiol but in the 131 women taking vitamin D supplements, none of the variables were found to be determinants of serum calcidiol. For serum PTH, the variables entered in the model as independent predictors were dietary calcium, milk calcium intake, nonmilk calcium intake, body weight, season, serum calcidiol, serum calcitriol, serum creatinine, and serum ionized calcium. In women not taking vitamin D supplements the main determinant of serum PTH concentration was serum calcidiol and in women taking vitamin D supplements serum calcidiol and serum creatinine accounted for equal variation in serum PTH. Prevalence of vitamin D deficiency Vitamin D deficiency, defined as a serum calcidiol concentration < 30 nmol/l (12 ng/ml), was observed in 10 of 245 (4%) of the women not taking vitamin D supplements and in 1 of 131 (< 1%) of the women taking vitamin D supplements. Absolute secondary hyperparathyroidism, defined as a PTH concentration FIGURE 2. Correlation between serum calcidiol and calcium intake from milk or nonmilk sources in 245 women. Serum calcidiol was significantly correlated with milk calcium intake (r = 0.35, P < 0.001). There was an independent significant effect of season on serum calcidiol: winter (r = 0.37, P < ) and summer (r = 0.22, P < 0.05). There was no correlation between serum calcidiol and nonmilk calcium intake (r = 0.10). > 65 ng/l (65 pg/ml) was observed in 6 of 245 (2%) of the women not taking vitamin D supplements and in 5 of 131 (4%) of the women taking vitamin D supplements; however, compared with young women in our laboratory, mean serum PTH increased by 30%. Renal function In the 245 women not taking vitamin D supplements, the serum PTH concentration was not significantly correlated with the serum creatinine concentration (r = 0.11, P > 0.09), but was significantly correlated with serum calcitriol (r = 0.18, P < 0.01). Serum calcitriol was inversely correlated with serum creatinine (r = 0.17, P < 0.01). In the 131 women taking vitamin D supplements, the serum PTH concentration was significantly correlated with serum creatinine (r = 0.26, P < 0.01) but not with serum calcitriol (r = 0.08). Serum calcitriol was inversely correlated with serum creatinine (r = 0.17, P < 0.05). Because of the effect of serum creatinine on serum PTH, serum PTH was adjusted for serum creatinine and correlated with serum calcidiol by using partial correlation analysis. Unadjusted serum PTH was correlated with serum calcidiol also. The equation for adjusted serum PTH was y = 37.9 ng/l 0.18 nmol creatinine/l, compared with that for unadjusted serum PTH, which was y = 48.8 ng/l 0.17 nmol creatinine/l. The unadjusted and adjusted intercepts were significantly different (P < ) but the slopes were not. The adjusted intercepts and slopes were similar to those in 85 healthy

5 346 KINYAMU ET AL FIGURE 3. Correlation between serum calcidiol and vitamin D intake (r = 0.34, P < 0.001) and between serum parathyroid hormone (PTH) and vitamin D intake (r = 0.15, P < 0.05) in 245 women. Serum calcidiol = 61 nmol/l mg/d (vitamin D intake); serum PTH = 40.0 ng/l mg/d (vitamin D intake). young women studied in our laboratory (y = 38.6 ng/l 0.14 nmol/l, r = 0.34, P < 0.01). Calcium absorption In both groups there was a nonsignificant decline in calcium absorption (% AD/L, BMI corrected) within the small age range studied (65 77 y). Calcium absorption in women taking vitamin D supplements was not significantly different from that in women not taking vitamin D supplements (Figure 5). Serum calcitriol was significantly correlated with calcium absorption at 3 h (r = 0.24, P < ) in women not taking vitamin D supplements but not in women taking vitamin D supplements (r = 0.11). Serum calcidiol was not correlated with calcium absorption in either group. DISCUSSION The average mean calcium intakes of 704 mg/d in women not taking vitamin D supplements and 818 mg/d in women taking vitamin D supplements were close to the RDA of 800 mg/d. On average, milk consumption accounted for 30% of the total calcium intake and calcium supplements accounted for 10%. Dietary vitamin D intakes of 3.53 g/d (141 IU/d) and 3.4 g/d (136 IU/d) in the two groups were below the US RDA of 5 g/d (200 IU/d). However, as expected, vitamin D intake was greater than the RDA in women taking vitamin D supplements containing 10 g/d (400 IU/d). Milk consumption accounted for 51% of the vitamin D intake in women not taking vitamin D supplements. Several studies reported vitamin D intakes lower than the RDA in the elderly in the United States. In a healthy elderly population in New Mexico (8) the mean dietary vitamin D intake FIGURE 4. Correlation between serum parathyroid hormone (PTH) and serum calcidiol in 376 women:, women not taking vitamin D supplements (n = 245);, women taking vitamin D supplements (n = 131). Serum PTH was inversely correlated with serum calcidiol (r = 0.33, P < 0.001). Solid line (elderly women): serum PTH = 48.8 ng/l 0.17 nmol/l (serum calcidiol); dotted line (young women): serum PTH = 38.6 ng/l 0.14 nmol/l (serum calcidiol). was 2.2 g/d (88 IU/d) and in a group of healthy postmenopausal women in Boston (25) the mean vitamin D intake was 2.68 g/d (107 IU/d). The relation between the 7-d food diaries and serum PTH and serum calcidiol in this group of healthy elderly women revealed interesting correlations. To separate the effects of calcium from those of vitamin D supplements on serum PTH and serum calcidiol, the relations were examined only in the group not taking vitamin D supplements. The finding that serum PTH correlated significantly with calcium intake from milk, but not with other sources of dietary calcium, suggested that the effect on serum PTH was due to another factor besides calcium. A plausible explanation was that it was due to the presence of vitamin D in milk. In support of this explanation is the finding that serum calcidiol was significantly correlated with calcium intake from milk but not with calcium intake from nonmilk sources. Further support that indicates that vitamin D plays a more important role than serum calcium in suppressing serum PTH is the inverse correlation between serum PTH and serum calcidiol in both women taking and not taking vitamin D supplements. Thus, the vitamin D source, whether from milk, sunlight exposure, or vitamin D supplements, was important in increasing serum calcidiol and suppressing serum PTH. There has not been much awareness of the physiologic importance of milk as a source of vitamin D. A recent study in an elderly, Irish, institutionalized population (mean age: 85 y) showed that vitamin D fortified milk was effective in increasing serum calcidiol and correcting hypovitaminosis D; however, the effect on serum PTH was not reported (26). Other support for the role of serum calcidiol in the suppression of serum PTH has been shown in animal studies. In rats, pharmacologic doses (500 pmol) of calcidiol have been shown to decrease pre-pro PTH messenger RNA concentrations by 20% in vivo, and subsequently decrease PTH synthesis (27). In addition, a pharmacologic injection of ng calcidiol in the carotid artery completely suppressed

6 DIETARY CALCIUM AND VITAMIN D IN ELDERLY WOMEN 347 TABLE 2 Determinants of serum calcidiol and serum parathyroid hormone concentrations in two groups of elderly women taking and not taking vitamin D supplements No vitamin D Vitamin D (n = 245) (n = 131) Multiple r r 2 Multiple r r 2 Serum calcidiol Milk calcium Season Body weight Serum parathyroid hormone (PTH) Serum calcidiol Serum calcitriol Body weight Serum calcidiol Serum creatinine FIGURE 5. Calcium absorption in the women not taking vitamin D supplements ( ; solid line, n = 245) and in the women taking vitamin D supplements ( ; dotted line, n = 131). Calcium absorption was not significantly greater in the women taking vitamin D supplements. AD/L, actual dose per liter of blood. PTH secretion in dogs (28). However, calcitriol binds to the PTH receptor with greater affinity than does calcidiol and is regarded as the normal physiologic control for PTH secretion. One might argue that the effect of milk calcium on the suppression of serum PTH is mediated through increased calcium absorption because of a higher vitamin D intake, but the results show that calcium absorption in the group taking vitamin D supplements was not different from that in the group not taking supplements. The finding of an inadequate absorptive response to vitamin D in older people is supported by the results of another recent study in which oral administration of calcitriol (25 g/d, or 1000 IU/d) did not increase calcium absorption (29). As shown in this study, a vitamin D intake of 12.5 g/d (500 IU/d) normalized serum calcidiol to a concentration known not to be associated with osteomalacia (30), yet was not able to restore calcium absorption to normal. Thus, the major effect of vitamin D supplements in the elderly may be the prevention of osteomalacia by a direct mineralizing effect on bone rather than by an increase in calcium absorption. Another unexpected finding in this study was that serum calcidiol was in the normal range in 96% of these elderly women. Several studies have shown that vitamin D deficiency, defined as a serum calcidiol concentration 30 nmol/l (12 ng/ml), is extremely common in elderly women in European countries (13 16), and in North America one study showed that 15% of the elderly had a vitamin D deficiency (8). In the present study, only 4% of those not taking vitamin D supplements had a serum calcidiol concentration < 30 nmol/l. However, the inverse correlation between serum PTH and serum calcidiol indicates that secondary hyperparathyroidism occurs at serum calcidiol concentrations not normally thought to be associated with vitamin D deficiency. From the regression analysis it can be estimated that serum PTH in the elderly would equal the mean PTH concentration of 28 ng/l (28 pg/ml) in young women at a serum calcidiol concentration of 122 nmol/l (49 ng/ml), whereas for young women the corresponding mean serum calcidiol concentration would be 75 nmol/l (30 ng/ml). Correlations between dietary vitamin D intake and serum calcidiol and between vitamin D intake and serum PTH show that elderly women would need to consume > 10 g vitamin D/d (400 IU/d) to reach a serum calcidiol concentration of 122 nmol/l (49 ng/ml) and a normal PTH concentration. Although serum PTH was within the normal range in 97% of this elderly population, serum PTH was 30% higher than in younger women (10). There are other factors that may cause an age-related increase in serum PTH, such as decreased absorption of calcium with age and decreased concentrations of serum calcitriol, both of which are found in women aged > 80 y (6, 18). A decline in renal function with age is normally associated with a decline in serum calcitriol and, as found in this study, serum calcitriol was inversely correlated with serum creatinine. Thus, decreased renal function with aging causes a decrease in calcitriol production by the kidney, resulting in malabsorption of calcium and secondary hyperparathyroidism. In these elderly women, serum calcitriol concentrations were similar to those of young women (10, 18), suggesting that serum PTH remains high in the elderly as a compensatory mechanism to maintain normal concentrations of serum calcitriol. Our results also showed an increase in serum PTH with declining renal function. Even after serum PTH was adjusted for serum creatinine, serum PTH remained correlated with serum calcidiol, and lower concentrations of serum calcidiol would be required to suppress secondary hyperparathyroidism. Thus, in the elderly, secondary hyperparathyroidism was due in part to vitamin D deficiency and to decreased renal function. In summary, this study showed milk to be an important nutritional source of vitamin D in the elderly, providing 50% of the dietary vitamin D intake. An increase in vitamin D intake, either by an increase in milk intake or from vitamin D supplements, should increase serum calcidiol and decrease serum PTH. The results from this study suggest that an adequate intake of vitamin D played a more significant role than did the calcium intake ( mg/d) in suppressing secondary hyperparathyroidism within the calcium intake range. Elderly women probably need to consume more than the current RDA of vitamin D (5 g/d, or 200 IU/d) to increase serum calcidiol and decrease serum PTH to normal concentrations. These conclusions are based on observational data. It is important that these findings be confirmed in a longitudinal, intervention study.

7 348 KINYAMU ET AL We thank the Research Coordinators Kimberly Petranick, Michelle Wilson, and Patty Fannon for their efforts in this study, and Kay Ryschon for the statistical analysis. REFERENCES 1. National Research Council. Recommended dietary allowances. 10th ed. Washington, DC: National Academy Press, Payette H, Gray-Donald K. Dietary intake and biochemical indices of nutritional status in an elderly population, with estimates of the precision of the 7-d food record. Am J Clin Nutr 1991;54: Delvin EE, Imbach A, Copti M. Vitamin D nutritional status and related biochemical indices in an autonomous elderly population. Am J Clin Nutr 1988;48: McGandy RB, Russell RM, Hartz SC, et al. Nutritional status survey of healthy non institutionalized elderly: energy and nutrient intake from three-day diet records and nutrient supplements. Nutr Res 1986;6: Garry PJ, Goodwin JS, Hunt WC, Hooper EM, Leonard AG. Nutritional status in a healthy elderly population: dietary and supplemental intakes. Am J Clin Nutr 1982;36: Gallagher JC, Riggs BL, Eisman J, Hamstra A, Arnaud SB, DeLuca HF. Intestinal calcium absorption and serum vitamin D metabolites in normal subjects and osteoporotic patients. J Clin Invest 1979;64: Lukert BP, Carey M, McCarty B, et al. Influence of nutritional factors on calcium regulating hormones and bone loss. Calcif Tissue Int 1987;40: Omdahl JL, Garry PJ, Hunsaker LA, Hunt WC, Goodwin JS. Nutritional status in a healthy elderly population: vitamin D. Am J Clin Nutr 1982;36: MacLaughlin J, Holick MF. Aging decreases the capacity of human skin to produce vitamin D 3. J Clin Invest 1985;76: Kinyamu HK, Gallagher JC, Petranick KM, Ryschon KL. Effect of parathyroid hormone (hpth[1-34]) infusion on serum 1,25-dihydroxyvitamin D and parathyroid hormone in normal women. J Bone Miner Res 1996;11: Tsai KS, Heath H III, Kumar R, Riggs BL. Impaired vitamin D metabolism with aging in women. J Clin Invest 1984;73: Francis RM, Peacock M, Barkworth SA. Renal impairment and its effects on calcium metabolism in elderly women. Age Ageing 1984;13: Lips P, Wiersinga A, van Ginkel FC, et al. The effect of vitamin D supplementation on vitamin D status and parathyroid function in elderly subjects. J Clin Endocrinol Metab 1988;67: Bouillon RA, Auwerx JH, Lissens WD, Pelemans WK. Vitamin D status in the elderly: seasonal substrate deficiency causes 1,25-dihydroxycholecalciferol deficiency. Am J Clin Nutr 1989;45: McKenna MJ, Freaney R, Meade A, Muldowney FP. Hypovitaminosis D and elevated serum alkaline phosphatase in elderly Irish people. Am J Clin Nutr 1985;41: Toss G, Almqvist S, Larsson L, Zetterqvist H. Vitamin D deficiency in welfare institutions for the aged. Acta Med Scand 1980;208: Sherman SS, Hollis BW, Tobin JD. Vitamin D status and related parameters in a healthy population: the effects of age, sex and season. J Clin Endocrinol Metab 1992;71: Kinyamu HK, Gallagher JC, Balhorn KE, Petranick KM, Rafferty KA. Serum vitamin D metabolites and calcium absorption in normal young and elderly free-living women and in women living in nursing homes. Am J Clin Nutr 1997;65: Dawson-Hughes B, Dallal GE, Krall EA, Harris S, Sokoll LJ, Falconer G. Effect of vitamin D supplementation on wintertime and overall bone loss in healthy postmenopausal women. Ann Intern Med 1991;115: Haddad JG, Chyu KJ. Competitive protein-binding radioassay for 25-hydroxycholecalciferol. J Clin Endocrinol Metab 1971; 33: Reinhardt TA, Horst RL. Simplified assays for the determination of 25-OHD, 24,25-(OH) 2 D and 1,25-(OH) 2 D. In: Norman AW, Schaefer K, Grigoleit HG, Herrath DV, eds. Vitamin D, molecular, cellular and clinical endocrinology. New York: Walter de Gruyter & Co, 1988: Reinhardt TA, Horst RL, Orf JW, Hollis BW. A microassay for 1,25- dihydroxyvitamin D not requiring high performance liquid chromatography: application to clinical studies. J Clin Endocrinol Metab 1984;58: Hollis BW. Assay of circulating 1,25-dihydroxyvitamin D involving a novel single-cartridge extraction and purification procedure. Clin Chem 1986;32: Nussbaum SR, Zahradnik RJ, Lavigne JR, et al. Highly sensitive two-site immunoradiometric assay of parathyrin, and its clinical utility in evaluating patients with hypercalcemia. Clin Chem 1987;33: Dawson-Hughes B, Harris SS, Krall EA, Dallal GE, Falconer G, Green CL. Rates of bone loss in postmenopausal women randomly assigned to one of two dosages of vitamin D. Am J Clin Nutr 1995;61: Keane EM, Rochfort S, Cox J, McGovern D, et al. Vitamin-D-fortified liquid milk_a highly effective method of vitamin D administration for house-bound and institutionalised elderly. Gerontology 1992;38: Silver J, Naveh-Many T, Mayer H, Schmeizer HJ, Popovtzer M. Regulation by vitamin D metabolites of parathyroid hormone gene transcription in vivo in the rat. J Clin Invest 1986;78: Canterbury JM, Lerman S, Claflin AJ, Henry H. Inhibition of parathyroid hormone secretion by 25-hydroxycholecalciferol and 24,25-dihydroxycholecalciferol in the dog. J Clin Invest 1978;61: Francis RM, Boyle IT, Moniz C, et al. A comparison of the effects of alfacalcidiol treatment and vitamin D 2 supplementation on calcium absorption in elderly women with vertebral fractures. Osteoporosis Int 1996;6: Lips P, Netelenbos JC, Jongen MJM, et al. Histomorphometric profile and vitamin D status in patients with femoral neck fracture. Metab Bone Dis Relat Res 1982;4:85 93.