Impact of calcium and vitamin D insufficiencies on serum parathyroid hormone and bone mineral density: Analysis of the fourth and fifth Korea National Health and Nutrition Examination Survey (KNHANES IV-3, 2009 and KNHANES V-1, 2010)
Jean Mayer–U.S. Department of Agriculture Human Nutrition Research Center on Aging, Tufts University, Boston, MA, USA
Ajou University School of Medicine, Suwon, Republic of Korea
Serum 25-hydroxyvitamin D [25(OH)D] concentration may or may not reflect total body stores of vitamin D, but it is widely used as a reflection of total vitamin D exposure from food, supplement, and skin synthetic sources. Parathyroid hormone (PTH) is an important determinant of bone resorption and mass.1, 2 PTH increases in response to low calcium intake3 and low serum 25(OH)D concentrations,4 and the resulting secondary hyperparathyroidism causes bone resorption and bone loss.5 Large cross-sectional studies have examined the relative importance of calcium intake and serum 25(OH)D concentrations, to serum PTH levels6 and femoral neck bone mineral density (BMD).7 In healthy Icelandic adults, serum 25(OH)D concentration was inversely associated with serum PTH concentration independent of calcium intake.6 In contrast, calcium intake was inversely associated with serum PTH in subjects who had serum 25(OH)D concentrations below 25 nmol/L, but not in those with higher serum 25(OH)D concentrations.6 In National Health and Nutrition Examination Survey (NHANES) III, the BMD of men and women increased with increasing serum 25(OH)D concentration independent of calcium intake.7 In the women, calcium intake was significantly and positively associated with femoral neck BMD in those with serum 25(OH)D concentrations below 50 nmol/L, but not in those with higher 25(OH)D concentrations.7 In men, there was no association of calcium intake with femoral neck BMD at any serum 25(OH)D concentrations.7 The mean calcium intake in the Icelandic study population was about 1200 mg/d6 and that in the NHANES III cohort was over 700 mg/d for the women (age 50 years and older) and over 800 mg/d for the men.7 These intakes are considerably higher than calcium intakes in many parts of the world, including Asian countries such as China,8 India,9 Japan,10 and Korea.11 Whether the impact of calcium, shown above to be restricted to those with low serum 25(OH)D concentrations, would be similarly restricted in populations with lower mean calcium intakes is unknown. It is important to understand the contributions of calcium intake and 25(OH)D levels to bone mass in populations with widespread deficiencies in those nutrients in order to target interventions most effectively. In this investigation, we examine the relative importance of calcium intake and serum 25(OH)D concentration to serum PTH concentrations and BMD in a population with low calcium intake using recent data from Korea National Health and Nutrition Examination Survey (KNHANES).
Materials and Methods
KNHANES, conducted periodically by the Korea Centers for Disease Control and Prevention since 1998, provide comprehensive information on health status, health behavior, nutritional status, and sociodemographics in 600 national districts in Korea. Data from the fourth (IV-3, 2009) and fifth (V-1, 2010) KNHANES data samplings containing serum PTH concentrations and BMD data were used in this cross-sectional analysis. From an initial total of 19,491 men and women, 6926 subjects (3016 men and 3910 women) aged 50 years and older with PTH and BMD data were evaluated. Of the evaluated subjects, 1373 were excluded for missing data and an additional 891 subjects were excluded for specific illnesses, medications that affect calcium metabolism, vitamin or mineral supplement use, too-high serum PTH concentration, and overreporting daily calcium or calorie intake. A final 2095 men and 2567 women were used in this analysis as shown in Fig. 1. All participants provided written informed consent.
Biochemistry, BMD, and dietary assessment
Blood samples, after an 8-hour fast, were collected year-round. They were immediately processed, refrigerated, and transported in cold storage to the central testing institute (NeoDin Medical Institute, Seoul, South Korea), where they were analyzed within 24 hours. Serum 25(OH)D concentration was measured with a radioimmunoassay (RIA) kit (DiaSorin Inc., Stillwater, MN, USA) using a γ-counter (1470Wizard; PerkinElmer, Turku, Finland). The interassay coefficients of variation (CV) were 2.8% to 6.2% for 2009 samples and 1.9% to 6.1% for 2010 samples. Serum 25(OH)D was measured in the same institute conducting quality control every other week throughout the analysis period to minimize the analytical variation. Serum intact PTH was analyzed using a chemiluminescence assay (DiaSorin). The CV of serum PTH determined biweekly was 6.19% ± 0.4% during Jan 2009 to Sept 2010. Due to a reagent change, there was a larger variation on PTH analysis during Oct–Dec 2011 (13.95% ± 3.22%) when 14.7% (n = 687) of the current study samples were analyzed. The glomerular filtration rate (GFR, mL/min/1.73 m2) was estimated by the equation of (140–age) × (body weight)/(serum creatinine × 72) with a factor of 0.85 applied for women.
The BMD (g/cm2) was measured at the lumbar spine (L1–L4) and femoral neck by dual-energy X-ray absorptiometry (DXA, DISCOVERY-W fan-beam densitometer; Hologic Inc., Marlborough, MA, USA) with CVs of 1.9% and 2.5%, respectively.
Nutrient intakes, including total calorie and calcium intakes, were assessed with a 24-hour dietary recall questionnaire administered by a trained dietician. The results were calculated using the Food Composition Table developed by the National Rural Resources Development Institute (7th revision).12 Contents of dietary supplements were not documented in KNHANES.
Physical activity was assessed by a questionnaire and categorized as “yes” or “no” with “yes” meaning >30 minutes of moderate physical activity three or more times in the last week in which the subject was tired compared to ordinary levels. Current smokers were defined as those who had smoked more than five packs of cigarettes during their life and were currently smoking; ex-smokers were smokers who had smoked in the past but had quit; nonsmokers had no history of smoking. Regular alcohol drinkers were those who currently drank alcohol more than one time per month and nondrinkers were all others. Occupations were classified into two groups: indoor workers and outdoor workers. The indoor workers included managers, professionals and related workers, clerical office workers, service workers, and sales workers. Unemployed subjects were classified as indoor workers considering their limited outdoor activities. The outdoor workers included skilled agricultural, forestry, and fishery workers. Women were classified into menopausal status as well as hormone replacement therapy (HRT). The estrogen (ER) (+) women were defined as premenopausal and postmenopausal women taking HRT.
The complex sample analysis was used for the KNHANES data for weighting all values following the guidance of statistics from the Korea Centers for Disease Control and Prevention. Before any adjustment was made for the weighted data, simple correlations were run to evaluate the associations of dietary calcium intake and serum 25(OH)D concentration with serum PTH and BMD. Serum 25(OH)D concentrations were then stratified as <50 nmol/L, 50 to 75 nmol/L, and >75 nmol/L. The general characteristics of the study population, such as age, body mass index, daily dietary calcium intake, and GFR, were compared across the 25(OH)D groups after weighting all values without any adjustment using a general linear model. The differences of physical activity, smoking status, and occupation across the 25(OH)D groups were analyzed by χ2 test in complex sample analysis. To determine the association of serum 25(OH)D concentration with serum PTH concentration and BMD, serum 25(OH)D concentration was entered as fixed factor, and serum PTH concentration or BMD as dependent factors for the analysis of covariance (ANCOVA) test before and after adjustment for covariates, which included the continuous variables: age, gender, body mass index (BMI), GFR, and dietary calcium intake, and the categorical variables: gender (male/female), smoking (never, past, current), occupation (indoor/outdoor), physical activity (yes/no), and season (Jan–Mar, Apr–June, July–Sept, Oct–Dec). To determine the association of dietary calcium intake with the serum PTH concentration and BMD (lumbar spine and femoral neck), daily dietary calcium intakes were divided into quintiles and the differences of serum PTH concentration or BMD across quintiles by the serum 25(OH)D categories were determined by the ANCOVA test after adjusting for age, gender, BMI, GFR, smoking, occupation, physical activity and season.
The differences of dietary calcium intake, serum 25(OH)D and PTH concentrations, and BMD by age (50–59, 60–69, ≥70 years old) and gender were determined by the ANCOVA test after adjusting for BMI, GFR, smoking, occupation, physical activity, season for men and women, and for ER (+/–) for women. The p for trend was used to assess the significance of serum 25(OH)D concentration and calcium intake on the serum PTH concentration or BMD within the age and gender groups. Data were analyzed using SPSS 18.0 (SPSS Inc., Chicago, IL, USA) to account for the complex sampling design.
Initial analyses indicated that dietary calcium intake was significantly correlated with serum PTH concentration (r = –0.082, p < 0.001) and with femoral neck (r = 0.254, p < 0.001) and lumbar spine (r = 0.184, p < 0.001) BMD. Similarly, serum 25(OH)D concentrations were significantly correlated with serum PTH (r = –0.196, p < 0.001), femoral neck (r = 0.121, p < 0.001), and lumbar spine BMD (r = 0.074, p < 0.001).
To explore these associations further, the subjects were divided into three serum 25(OH)D groups: lowest <50 nmol/L, middle 50 to 75 nmol/L, and highest >75 nmol/L. Subject characteristics of these groups are shown in Table 1. Most subjects (58.7%, n = 2739) were in the lowest 25(OH)D group (<50 nmol/L) followed by 32.5% subjects (n = 1517) in the middle group, and 8.8% (n = 406) in the highest group. Mean dietary intake of calcium was significantly higher in the middle 25(OH)D group than in the other groups. Subjects in the lowest 25(OH)D group had significantly lower GFR (p for trend <0.003) than the other groups. Similar numbers of subjects were measured in each season. Physical activity was low (12%–18%) in all groups and only 11% to 12% of subjects were outdoor workers.
Table 1. Subject Characteristics Categorized by Serum 25-Hydroxyvitamin D Concentrations
Values are expressed as mean (SEM) after weighting of all variables without adjustment except where indicated.
BMI = body mass index; HRT = hormone replacement therapy; GFR = glomerular filtration rate.
p for trend analyzed by general linear model in complex data analysis.
To convert serum 25-hydroxyvitamin to ng/mL, divide by 2.496.
p < 0.05 in the comparison of <50 nmol/L versus >75 nmol/L and marked on the higher mean value.
p < 0.05 in the comparison of 50–75 nmol/L versus >75 nmol/L and marked on the higher mean value.
p < 0.05 in the comparison of <50 nmol/L versus 50–75 nmol/L and marked on the higher mean value.
GFR (mL/min/1.73 m2) was estimated by the equation of [(140–age) × (body weight)]/(serum creatinine × 72) and a factor 0.85 was applied for women.
Physical activity is defined as leisure (not work related) exercise over 30 minutes of moderate intensity more than 5 days per week.
Outdoor workers: skilled agricultural, forestry, and fishery workers. Indoor workers: managers, professionals and related workers, clerical office workers, service workers, sales workers, and unemployed subjects.
Mean serum PTH concentration in each 25(OH)D group is shown in Table 2. Mean PTH concentrations decreased across the 25(OH)D groups (p for trend <0.001); subjects in the lowest serum 25(OH)D group had significantly higher PTH than those in the middle and highest groups. Adjustment for potential confounders, calcium intake, age, gender, BMI, GFR, smoking, occupation, season, and physical activity, did not substantially alter the mean PTH concentrations (Table 2). The adjusted mean serum PTH concentrations declined significantly (p for trend <0.001), by about 15% across the 25(OH)D groups.
Table 2. Serum PTH Concentrations and BMDs Categorized by Serum 25(OH)D Concentrations
Mean femoral neck BMD increased across the 25(OH)D groups (p for trend <0.001). Subjects in the lowest 25(OH)D group had significantly lower femoral neck BMD than those in the middle and highest groups, as presented in Table 2. Adjustment for potential confounders, calcium intake, age, gender, BMI, GFR, smoking, occupation, season, and physical activity did not substantially alter the mean BMD. The adjusted mean femoral neck BMD was 2% lower in the lowest serum 25(OH)D group than the middle 25(OH)D group. Although lumbar spine BMD increased across the 25(OH)D groups (p for trend = 0.002), the significant difference was disappeared after adjustment for potential confounders.
The impact of dietary calcium on PTH was examined by serum 25(OH)D category. To accomplish this, subjects were further divided into quintiles of calcium intake. The PTH levels, adjusted for age, gender, BMI, GFR, smoking, occupation, season, physical activity, are presented by quintile of calcium intake within each 25(OH)D group in Fig. 2. Adjusted mean serum PTH concentrations declined with increasing quintile of calcium intake in both the lowest (p for trend = 0.032) and highest (p for trend = 0.023) 25(OH)D groups.
The impact of dietary calcium on BMD was also examined by serum 25(OH)D category after adjustment for age, gender, BMI, GFR, smoking, occupation, season, and physical activity as shown in Fig. 2. The adjusted mean femoral neck BMD increased significantly across the calcium quintiles in each of the three 25(OH)D groups. Calcium intake was positively associated with the adjusted mean BMD of lumbar spine in the lower two 25(OH)D groups.
Dietary calcium intakes, serum 25(OH)D and PTH concentrations, and BMD are shown by sex and age (50–59, 60–69, and ≥70 years old) in Table 3. These values were adjusted for BMI, GFR, smoking, occupation, physical activity, season, and, in the case of women, ER (+/–). The men had higher dietary calcium intake, serum 25(OH)D concentration, and femur neck and lumbar spine BMD (p < 0.001) and lower serum PTH concentration (p < 0.001) than the women. Mean dietary calcium intakes were significantly higher in the 50- to 59-year-old age group (p for trend <0.001) than in older age groups. In the men and women, there were no significant differences in serum 25(OH)D concentrations across the age groups. In both the men and women, the serum PTH concentrations were significantly lower in the 50 to 59-year-old age group (p for trend <0.001) than in the 60 to 69 and ≥70-year-old age groups. In both genders, femur neck and lumbar spine BMD decreased significantly with age.
Table 3. Dietary Calcium Intakes, Serum 25(OH)D and PTH Concentrations, and BMDs by Age and Sex
This study examined calcium intake and serum concentrations of 25(OH)D in relation to serum PTH concentration and BMD in a population with a high prevalence of low calcium intake. By nature of its design, this cross-sectional study cannot establish causality. However, it does confirm that lower 25(OH)D concentrations are associated with higher serum PTH and lower BMD levels,6, 7 It differs, however, from previous reports of the associations of calcium intake with PTH and BMD. The important link of lower calcium intakes to higher serum PTH levels and lower BMD has been limited previously to subjects with serum 25(OH)D levels below 25 nmol/L6 and 50 nmol/L,7 respectively. In contrast, our study identified significant inverse associations between calcium intake and serum PTH levels not only at lower (< 50 nmol/L) but also at higher (> 75 nmol/L) serum 25(OH)D concentrations. Furthermore, significant positive associations between calcium intakes and femoral neck BMD were found in the lowest, the middle, and the highest 25(OH)D groups in this study. Unlike previous studies conducted in Iceland6 and the United States7 in subjects with relatively higher calcium intakes, 1200 mg/d and 749 mg/d, respectively, the current study population in Korea had average calcium intake of 485 mg/d. These findings emphasize that low calcium intake is detrimental to calcium metabolism and bone mass and that it cannot be compensated with higher 25(OH)D levels alone.
Our findings indicate that subjects in the top quintile of calcium intake (adjusted mean 951 mg/d; range, 668–1986 mg/d), had significantly lower PTH levels and higher BMD of the femoral neck and spine than those in the adjacent 4th quintile (adjusted mean, 557 mg/d; range, 467–667mg/d). This indicates that 80% of the study population would benefit from increasing the calcium intake. It also suggests that the current Korean recommended daily allowance (RDA) (700 mg, ≥50 years old),13 is not adequate to suppress PTH and maximize BMD in older adults. There is evidence from randomized, controlled trials that increasing calcium intake with supplements14, 15 and with milk powder16 lowers PTH levels and reduces rates of bone loss17 in healthy older men and women, with the greatest benefit seen in those with low (<400 mg/d) usual calcium intakes.18
In this study, 59% of the population had 25(OH)D levels below 50 nmol/L. Factors contributing to the low serum 25(OH)D levels may include limited skin synthesis of vitamin D, the high latitude (37 degrees North) and cultural preferences, low intakes of vitamin D–rich foods,19 and limited availability of vitamin D–fortified foods.20 It is true that serum 25(OH)D is inactive and it may not represent actual total body store of 25(OH)D or 1,25-dihydroxyvitamin D [1,25(OH)2D], the main regulatory hormone21; however, 25(OH)D is the substrate for 1,25(OH)2D, presenting in circulation at ∼1000 times higher concentration than 1,25(OH)2D.22 We found that the PTH level was lowest and BMD was highest in the participants with 25(OH)D levels ≥ 50 nmol/L. Considering that a vitamin D dosage of 800 IU/d resulted in increases of serum 25(OH)D levels to greater than 50 nmol/L in 97.5% of postmenopausal women as shown in a recent study,23 the majority of the current study population seems to benefit from a higher vitamin D intake. The current adequate intake (AI) for vitamin D in Korea is 400 IU per day (≥50 years old).13 Increasing vitamin D intake to levels above the current Korean AI is needed to optimize calcium metabolism and BMD.
In conclusion, calcium intake is a significant determinant of serum PTH and BMD in this population with a high prevalence of low calcium intakes. This cross-sectional study suggests that a calcium intake of at least 668 mg/d is needed to achieve suppressed PTH levels and higher BMD. In contrast to previous findings in populations with higher calcium intakes, the very low calcium intakes seen here were detrimental at high as well as low 25(OH)D levels, indicating that calcium insufficiency cannot be compensated for by higher serum 25(OH)D levels. This study confirms that a 25(OH)D level of 50 nmol/L or higher is associated with greater suppression of serum PTH and higher BMD. A majority of participants had levels below this threshold. Substantial increases in both calcium and vitamin D intake are recommended for improving calcium metabolism and increasing BMD in this at-risk population.
All authors state that they have no conflicts of interest.
This research was supported in part by the U.S. Department of Agriculture, under agreement number 58-1950-7-707. Any opinions, findings, conclusion, or recommendations expressed in this publication are those of the authors and do not necessarily reflect the view of the U.S. Department of Agriculture. We thank all the members of the Korea Institute for Health and Social Affairs who conducted the national survey.
Authors' roles: Study concept and design: NJ, BD, and KY. Acquisition of data: NJ, KO, and YK. Data analysis: NJ, KO, and KY. Data interpretation: NJ, BD, and KY. Drafting manuscript: NJ and KY. Revising manuscript content: BD and KY. Approving final version of manuscript: NJ, BD, YK, KO, and KY. NJ takes responsibility for the integrity of the data analysis.