The Relationship Between Serum 25(OH)D and Parathyroid Hormone Levels

CLINICAL RESEARCH STUDY The Relationship Between Serum 25(OH)D and Parathyroid Hormone Levels Walid Saliba, MD, MPH, a Ofra Barnett, PhD, a Hedy S. Rennert, MPH, a Idit Lavi, MA, a Gad Rennert, MD, PhD a,b a Department of Community Medicine and Epidemiology, Carmel Medical Center, Clalit Health Services, and Bruce Rappaport Faculty of Medicine, Technion-Israel Institute of Technology, Haifa, Israel; b Department of Epidemiology and Disease Prevention, Office of the Chief Physician, Clalit Health Services Headquarters, Tel Aviv, Israel. ABSTRACT OBJECTIVE: Low 25(OH)D levels are associated with increased parathyroid hormone levels leading to progressive bone loss. The serum levels of 25(OH)D sufficient to keep the parathyroid hormone level at a range that will prevent bone loss are still unclear. The current study was aimed at evaluating the relationship between 25(OH)D levels and concomitant parathyroid hormone levels. METHODS: The computerized laboratory database of Clalit Health Services, a not-for-profit health maintenance organization covering more than half of the Israeli population, was searched for all 25(OH)D and parathyroid hormone tests performed in 2009. Concomitant tests of parathyroid hormone and 25(OH)D were identified in 19,172 people. RESULTS: Serum parathyroid hormone levels were inversely correlated with 25(OH)D levels (r 0.176, P.001); 25(OH)D levels less than 50 nmol/l were associated with a steep increase in parathyroid hormone levels and hyperparathyroidism, which decreased with increasing 25(OH)D levels and reached a plateau at 25(OH)D levels of 75 to 85 nmol/l. The quadratic fit with plateau model showed that parathyroid hormone stabilizes at 25(OH)D level of 78.9 nmol/l. However, after excluding 5449 people with hypercalcemia or renal failure, the parathyroid hormone plateau was attained at a significantly lower 25(OH)D cut point of 46.2 nmol/l. CONCLUSION: Our data suggest that a 25(OH)D threshold of 50 nmol/l is sufficient for parathyroid hormone suppression and prevention of secondary hyperparathyroidism in persons with normal renal function. 25(OH)D levels greater than 75 nmol/l do not seem to be associated with additional change in parathyroid hormone levels. 2011 Elsevier Inc. All rights reserved. The American Journal of Medicine (2011) 124, 1165-1170 SEE RELATED EDITORIAL p. 1095 Vitamin D insufficiency is currently recognized as a worldwide epidemic 1,2 and has been found to be associated with increased morbidity. 3,4 It is widely accepted that vitamin D status is best assessed by serum 25(OH)D. 3,5,6 Because levels of 25(OH)D less than 25 nmol/l are associated with increased risk of rickets and osteomalacia, 5-7 there is a Funding: None. Conflict of Interest: None of the authors have any conflicts of interest associated with the work presented in this manuscript. Authorship: All authors had access to the data and played a role in writing this manuscript. Requests for reprints should be addressed to Walid Saliba, MD, MPH, Department of Community Medicine and Epidemiology, Carmel Medical Center, 7 Michal St, Haifa 34362, Israel. E-mail address: saliba_wa@clalit.org.il general agreement that this threshold defines vitamin D deficiency. 3,6 However, there is still lack of consensus on the definition of vitamin D insufficiency and on the physiologic range for its normal function. Because vitamin D controls dietary calcium absorption, even mild insufficiency of vitamin D is compensated by an increase in serum parathyroid hormone, which in turn is associated with increased bone turnover that leads to osteopenia, osteoporosis, and increased risk of fracture. 3-5 Many studies, therefore, use the relationship between serum parathyroid hormone and 25(OH)D to define the normal range of serum 25(OH)D, 4,5,7-11 and different cut points of serum 25(OH)D (37.5 nmol/l, 12 50 nmol/l, 2,10,11,13 or 75 nmol/l 2,3,5,8 ) are currently in use for insufficiency definition: According to the recently released report on Dietary 0002-9343/$ -see front matter 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.amjmed.2011.07.009

1166 The American Journal of Medicine, Vol 124, No 12, December 2011 Reference Intakes for vitamin D and calcium by the Institute of Medicine (IOM), persons are at risk of deficiency at serum 25(OH)D levels less than 30 nmol/l. Some, but not all, persons are potentially at risk for inadequacy at serum 25(OH)D levels between 30 and 50 nmol/l, and practically all persons are sufficient with serum 25(OH)D levels of at least 50 nmol/l. 14 The lack of a well-established consensus regarding thresholds to be used has clinical implications, and some thresholds may lead to under- or overtreatment. This study assesses the optimal range of serum 25(OH)D by testing the dynamic relationship between 25(OH)D and parathyroid hormone levels. MATERIALS AND METHODS CLINICAL SIGNIFICANCE Study Population and Data Collection Clalit Health Services (CHS) is a not-for-profit health maintenance organization covering more than half of the Israeli population. Data on laboratory tests and demographic variables are available from a computerized database. The study population includes all CHS members for whom 25(OH)D and parathyroid hormone test results performed at the same time in 2009 were available. When more than 1 pair of results was available for the same person, the most recent tests were selected. Paired test results were included only if concomitant serum creatinine and calcium levels also were available for the case, within 8 weeks from the index result. Clinical Definitions Primary hyperparathyroidism was defined as concomitant serum parathyroid hormone 65 pg/ml and serum calcium 10.5 mg/dl. Secondary hyperparathyroidism was defined as concomitant serum parathyroid hormone 65 pg/ml and serum calcium 10.5 mg/dl. Renal failure was defined as serum creatinine 1.5 mg/dl and hypercalcemia as serum calcium 10.5 mg/dl. 25(OH)D and Parathyroid Hormone Assay 25(OH)D and parathyroid hormone were tested in different laboratories; however, more than 90% of the blood tests were performed in 4 central laboratories using the same assay for both tests. 25(OH)D was measured using the LIAISON 25-OH Vitamin D TOTAL assay (DiaSorin USA, Stillwater, Minn), a competitive 2-step chemiluminescence assay with a measurement range of 4.0 to 150 ng/ml (10-375 nmol/l), analytic sensitivity 1.0 ng/ml (2.5 nmol/l), and functional sensitivity 4.0 ng/ml (10 nmol/l). The intra-assay precision is up to 5%, and the inter-assay precision is up to 15%. The specificity is 104% for 25-OH A 25(OH)D threshold of 50 nmol/l is sufficient for parathyroid hormone suppression and prevention of secondary hyperparathyroidism. 25(OH)D levels greater than 75 nmol/l are not associated with additional change in parathyroid hormone levels. By using the threshold of 50 nmol/l, a smaller proportion of the population will be defined as vitamin D insufficient. vitamin D2 and 100% for 25-OH vitamin D3. Parathyroid hormone was measured using the IMMULITE200 intact parathyroid hormone (Siemens USA, New York, NY), a 2-site chemiluminescent enzyme-labeled immunometric assay with a measurement range of 1 to 2500 pg/ml and functional sensitivity of 1 pg/ml. The intra-assay precision is up to 5.7%, and the inter-assay precision is up to 9%. Statistical Analysis Continuous data are presented as mean standard deviation, along with median and interquartile range. Spearman s correlation coefficient was used to test the relation between continuous variables, because some variables, including parathyroid hormone, presented a relatively right-skewed distribution. Continuous variables were compared between 2 groups by the unpaired Student t test or the Mann-Whitney test as appropriate. Logistic regression was used for testing an association between a continuous variable and a prevalence indicator. The mean serum parathyroid hormone is graphically presented in 12 categories (almost equally spaced) of serum 25(OH)D ( 12, 12-25, 25-37, 37-50, 50-62, 62-75, 75-85, 85-95, 95-105, 105-115, 115-125, and 125 nmol/l). A quadratic model with plateau was fitted to model the relationship between serum parathyroid hormone and serum 25(OH)D levels (exact continuous values) to objectively identify the 25(OH)D level where the parathyroid hormone reaches a plateau. This model assumes that for 25(OH)D level less than certain concentration (X 0 ) the equation relating parathyroid hormone and 25(OH)D is quadratic, and for 25(OH)D levels greater than X 0 the equation is constant (at a value P). The model equations thus are parathyroid hormone a b[25(oh)d] c[25(oh)d] 2 for 25(OH)D X 0, and parathyroid hormone P for 25(OH)D X 0. This model was estimated using nonlinear least squares model, as performed by the SAS NLIN procedure (SAS Institute Inc, Cary, NC). The joint point (X 0 ) need not be known for that application. All other statistical analyses were performed using SPSS 17.0 (SPSS Inc, Chicago, Ill). A P value of less than.05 was considered statistically significant. RESULTS Subject Characteristics Complete data were available for 24,109 pairs of serum parathyroid hormone and 25(OH)D test results in 20,045 CHS members who had test results in 2009. Included in the analysis are 19,172 members who fulfilled the inclusion

Saliba et al Sufficient 25(OH)D Levels 1167 Table 1 Demographic Data and Clinical Characteristics of the Study Population, Clalit Health Services 2009 All (N 19,172) Gender Female 72.0% 79.2% Age Mean SD 63.7 15.5 62.0 15.4 Median (IQ range) 65.0 (56-75) 63 (54-73) Ethnicity Jews 92.9% 92.8% Arabs 7.1% 7.2% Hypercalcemia 14.1% Impaired renal function 15.8% Primary hyperparathyroidism 8.7% Secondary hyperparathyroidism 31.7% 29.7% SD standard deviation; IQ interquartile. *Creatinine 1.5 mg/dl. Calcium 10.5 mg/dl. Parathyroid hormone 65 pg/ml and calcium 10.5 mg/dl. Parathyroid hormone 65 pg/ml and calcium 10.5 mg/dl. Subjects with Normal Renal Function* and Normal Calcium Levels (N 13,723) criteria. Demographic and clinical characteristics are presented in Table 1. Seventy-two percent are female, and the overall mean age was 63.7 15.5 years. Laboratory tests were equally distributed throughout the different seasons of the year (data not shown). Serum 25(OH)D Levels Serum 25(OH)D ranged between 10 and 375 nmol/l (16 people had levels 200 nmol/l) with a mean of 62.3 26.8 (Table 2). No difference in mean serum 25(OH)D was detected between those with primary hyperparathyroidism compared with those with secondary hyperparathyroidism: 57.3 25.3 and 56.4 27.5 nmol/l, respectively (P 0.240). Mean serum 25(OH)D level was lower in people with renal failure (creatinine 1.5 mg/dl) compared with those with normal renal function: 57.0 27.5 nmol/l and 63.3 26.6 nmol/l, respectively (P.001). Association Between Parathyroid Hormone and 25(OH)D Levels Serum parathyroid hormone levels were weakly and inversely correlated with serum 25(OH)D level (Spearman r 0.176; P.001). The low strength of the correlation may be the result of the nonlinear nature of the association. Serum parathyroid hormone levels increased steeply at 25(OH)D levels less than 50 nmol/l, increased mildly at 25(OH)D level between 50 and 75 nmol/l, and reached a plateau at 25(OH)D levels greater than 75 nmol/l. A segmented model including quadratic equation followed by a plateau was fitted for the whole study group and estimated that serum parathyroid hormone began to increase at serum 25(OH)D less than 78.9 nmol/l (X 0 78.9). This point corresponded to a serum parathyroid hormone level of 62.5 pg/ml (P 62.5) in all cases tested. The prevalence of hyperparathyroidism (defined as serum parathyroid hormone 65 pg/ml) increased with decreasing serum 25(OH)D (P.001) (data not shown). Association of Parathyroid Hormone with Creatinine and Calcium A significant positive correlation was found between serum parathyroid hormone and serum creatinine levels (Spearman r 0.27; P.001). This correlation was stronger when serum creatinine was greater than 1.5 mg/dl (Spearman r 0.44; P.001). Overall, the correlation of serum parathyroid hormone and serum calcium was significant but weak (Spearman r 0.03; P.001) and negative in those with serum calcium 10.5 mg/dl (r 0.112; P.001). However, for those with hypercalcemia (calcium 10.5 mg/dl) the correlation turned positive and stronger (r 0.284; P.001). Because impaired renal function and increased serum calcium levels were found to be associated with increased serum parathyroid hormone levels, the association between serum parathyroid hormone and 25(OH)D levels was studied separately in people with normal renal function and normal serum calcium (n 13,723). The demographic, clinical, and laboratory characteristics of this subgroup are presented in Tables 1 and 2. The association between serum parathyroid hormone levels by 25(OH)D levels for this subgroup is described in Figure. When the segmented model was fitted to this pop-

1168 The American Journal of Medicine, Vol 124, No 12, December 2011 Table 2 Laboratory Data of the Study Population, Clalit Health Services 2009 All (N 19,172) Calcium (mg/dl) Mean SD 9.8 0.65 9.7 0.43 Median (IQ range) 9.7 (9.4-10.1) 9.7 (9.4-10) Creatinine (mg/dl) Mean SD 1.20 1.2 0.84 0.21 Median (IQ range) 0.85 (0.71-1.1) 0.79 (0.68-0.94) PTH (pg/ml) All Mean SD 77.7 95.5 57.0 38.0 Median (IQ range) 56.1 (37.3-86.0) 49.6 (34.5-70.0) Winter and Spring Mean SD 75.9 89.3 56.9 38.3 Median (IQ range) 55.8 (37.4-85.0) 49.6 (34.8-69.9) Summer and Autumn Mean SD 79.4 101.1 57.1 37.6 Median (IQ range) 56.6 (37.3-87.0) 49.6 (34.3-70.3) 25(OH)D (nmol/l) All Mean SD 62.3 26.8 63.5 26.6 Median (IQ range) 61.3 (43.5-78.8) 62.7 (45.2-79.8) Winter and Spring Mean SD 57.8 26.8 58.8 26.6 Median (IQ range) 56.5 (38.2-74.6) 57.7 (39.2-75.7) Summer and Autumn Mean SD 66.6 26.1 68.4 25.8 Median (IQ range 65.5 (49.2-82.3) 67.3 (51.6-83.6) SD standard deviation; IQ interquartile; PTH parathyroid hormone. *Creatinine 1.5 mg/dl. Calcium 10.5 mg/dl. No significant difference between seasons for both groups (P.2). Significant difference between seasons for both groups (P.001). Subjects with Normal Renal Function* and Normal Calcium Levels (N 13,723) ulation, the parathyroid hormone plateau was reached at serum parathyroid hormone levels of 53.8 pg/ml (P 53.8), which corresponded with the lower serum 25(OH)D level of 46.2 nmol/l (X 0 46.2). Figure Mean serum parathyroid hormone (pg/ml) by serum 25(OH)D subgroups, CHS 2009. Subjects with normal serum calcium and normal renal function (N 13,723). DISCUSSION Our study shows that the optimal serum 25(OH)D level is approximately 50 nmol/l, as measured by a competitive 2-step chemiluminescence assay, in people with normal kidney function and normal serum calcium levels, if the criteria of plateau serum parathyroid hormone level is accepted as a measure of 25(OH)D adequacy. Parathyroid hormone stabilized at this threshold, and serum 25(OH)D levels greater than 46.2 nmol/l were not associated with continued decrease in parathyroid hormone levels in this group of patients. The 25(OH)D measurement has not been standardized. 15 Binkley et al 16 found that 25(OH)D assays used by different laboratories yielded markedly different results; thus, it should be emphasized that in the present study 25(OH)D was measured by a competitive 2-step chemiluminescence assay. The significant correlation between serum 25(OH)D and serum parathyroid hormone levels is in line with findings in other studies. 4,17 It is suggested that the optimal range of 25(OH)D for skeletal health is one that reduces parathyroid

Saliba et al Sufficient 25(OH)D Levels 1169 hormone levels to a minimum. 3 A wide range of normal serum 25(OH)D levels are recommended in the literature, with most estimates clustered from 70 to 80 nmol/l and 40 to 50 nmol/l. 17 Our study identified 2 different 25(OH)D thresholds: 46.2 nmol/l for people with normal creatinine/ calcium and 78.9 nmol/l when no restriction is made. This finding can explain the discrepancy of results reported in the literature. Both renal failure and hypercalcemia were associated with increased parathyroid hormone levels irrespective of vitamin D status. Thus, we suggest the threshold 46.2 nmol/l to define vitamin D sufficiency for parathyroid hormone stabilization. The overall threshold of 78.9 nmol/l in our population is compatible with the findings reported by Holick et al 5 and Chapuy et al. 8 Holick et al 5 used the same model that we used and reported that parathyroid hormone plateaued at 25(OH)D 29.8 ng/ml (equivalent to 74.5 nmol/l). By using an exponential model, Chapuy et al 8 found that serum parathyroid hormone held a stable plateau as long as serum 25(OH)D levels were higher than 78 nmol/l. However, they found no correlation between serum parathyroid hormone and serum creatinine, and concluded that the increase in parathyroid hormone was not secondary to impaired renal function. Okazaki et al 4 recently suggested a 25(OH)D threshold of 28 ng/ml (70 nmol/l). Heaney et al 18 found that intestinal calcium absorption was higher at serum 25(OH)D levels averaging 86.5 nmol/l than at levels averaging 50 nmol/l and concluded that normal serum 25(OH)D levels ranged from 80 to 90 nmol/l. Nevertheless, Hansen et al 19 showed that patients with osteoporosis and vitamin D insufficiency had only 3% increase in fractional calcium absorption after treatment with vitamin D; according to previous studies, this small increase was not associated with decreased risk of hip fracture or increase bone mass. 19 Some studies have suggested higher thresholds. Dawson-Hughes et al 9 found that the change point estimate of 25(OH)D above which the association between parathyroid hormone and 25(OH)D is flat was 110 nmol/l. In line with our suggested threshold of 25(OH)D at approximately 50 nmol/l as measured by a competitive 2-step chemiluminescence assay, Malabanan et al 10 found that supplementation with oral vitamin D 2 and oral calcium decreased serum parathyroid hormone only in subjects with basal serum 25(OH)D levels of less than 50 nmol/l, and concluded that this threshold was the minimum required to optimize serum parathyroid hormone and prevent secondary hyperparathyroidism. The same threshold was suggested by Lips et al 11 from the reanalysis of the MORE trial. Recently, this threshold was defined as the minimum required to support and maintain normal bone and mineral health by the 14th Vitamin D Workshop held in Brugge Belgium in October 2009. 20 Moreover, the recently released IOM report concluded that a serum 25(OH)D level of at least 50 nmol/l is practically sufficient for everyone. 14 Other studies found lower thresholds. Thomas et al 21 found that parathyroid hormone increased steeply as serum 25(OH)D declined to less than 15 ng/ml (37.5 nmol/l), and that for levels greater than 37.5 nmol/l the slope was not significantly different from zero (P.13). It is possible that the wide range of previously suggested thresholds may reflect different demographic 22 and clinical characteristics of the studied populations, as well as the statistical models used and the assay used to measure 25(OH)D levels. The strength of our study is in the large number of parathyroid hormone/25(oh)d paired test results that were available from a large general population with demographic heterogeneity. Also, the studied tests were equally distributed among the seasons of the year. Another strength of the study is that we were able to evaluate the influence of serum calcium levels and renal function status on the parathyroid hormone/25(oh)d relationship. Yet, the current study also has several limitations: data on calcium intake, body mass index, and serum magnesium, all known to influence serum parathyroid hormone levels, were not available. Our study used the total serum calcium level and not the more accurate ionized calcium or calcium corrected to serum albumin. Renal function assessment was based on creatinine levels rather than by the more accurate glomerular filtration rate. The individual timing of blood collection for parathyroid hormone testing was not available; however, the instructions state that the blood sample should be collected in the morning after an overnight fast. Another limitation of our study is that we were not able to identify and exclude from the analysis patients taking medication that may affect 25(OH)D levels, such as anticonvulsants and glucocorticoids. Normal serum parathyroid hormone levels are influenced by serum 25(OH)D levels. 15 Some data suggest lower thresholds of serum parathyroid hormone levels in vitamin D-replete subjects. 23 In this study, we used a serum parathyroid hormone level greater than 65 pg/ml to define hyperparathyroidism in all patients irrespective of their individual serum 25(OH)D levels. However, we still do not have optimal reference intervals for parathyroid hormone values based on coexisting 25(OH)D levels. 15 Moreover, a large proportion of the population included in our study were vitamin D depleted and are expected to have higher parathyroid hormone levels. It should be emphasized that we did not have data on bone density and therefore cannot state if the limit increase in parathyroid hormone also was associated with a limit in influence on bone density. Vitamin D deficiency, usually based on high thresholds of 75 nmol/l, is an increasingly described phenomenon affecting the majority of Western populations, and drastic corrective measures are often recommended. 2,3,5,8 Our finding, that 25(OH)D levels greater than 46.2 nmol/l seem to be sufficient, is in line with those of the recent IOM report, which also suggested that even lower levels may be adequate for some. In light of these findings, it seems that a smaller proportion of the population will be defined as vitamin D insufficient.

1170 The American Journal of Medicine, Vol 124, No 12, December 2011 CONCLUSIONS 25(OH)D levels greater than 46.2 nmol/l seem to be sufficient for suppression of parathyroid hormone and prevention of secondary hyperparathyroidism. 25(OH)D levels greater than 75 nmol/l do not seem to be associated with additional change in parathyroid hormone levels. More studies are needed to identify the optimal level of 25(OH)D in correlation with clinical outcomes. References 1. Holick MF. Vitamin D deficiency. N Engl J Med. 2007;357:266-281. 2. Mithal A, Wahl DA, Bonjour JP, et al; IOF Committee of Scientific Advisors (CSA) Nutrition Working Group. Global vitamin D status and determinants of hypovitaminosis D. Osteoporos Int. 2009;20: 1807-1820. 3. Holick MF. High prevalence of vitamin D inadequacy and implications for health. Mayo Clin Proc. 2006;81:353-373. 4. Okazaki R, Sugimoto T, Kaji H, et al. Vitamin D insufficiency defined by serum 25-hydroxyvitamin D and parathyroid hormone before and after oral vitamin D(3) load in Japanese subjects. J Bone Miner Metab. 2011;29:103-110. Epub 2010 Jun 22. 5. Holick MF, Siris ES, Binkley N, et al. Prevalence of vitamin D inadequacy among postmenopausal North American women receiving osteoporosis therapy. J Clin Endocrinol Metab. 2005;90:3215-3224. 6. Pearce SH, Cheetham TD. Diagnosis and management of vitamin D deficiency. BMJ. 2010;340:142-147. 7. Carnevale V, Nieddu L, Romagnoli E, et al. Regulation of PTH secretion by 25-hydroxyvitamin D and ionized calcium depends on vitamin D status: a study in a large cohort of healthy subjects. Bone. 2010:626-630. 8. Chapuy MC, Preziosi P, Maamer M, et al. Prevalence of vitamin D insufficiency in an adult normal population. Osteoporos Int. 1997;7: 439-443. 9. Dawson-Hughes B, Harris SS, Dallal GE. Plasma calcidiol, season, and serum parathyroid hormone concentrations in healthy elderly men and women. Am J Clin Nutr. 1997;65:67-71. 10. Malabanan A, Veronikis IE, Holick MF. Redefining vitamin D insufficiency. Lancet. 1998;351:805-806. 11. Lips P, Duong T, Oleksik A, et al. A global study of vitamin D status and parathyroid function in postmenopausal women with osteoporosis: baseline data from the multiple outcomes of raloxifene evaluation clinical trial. J Clin Endocrinol Metab. 2001;86:1212-1221. 12. Nesby-O Dell S, Scanlon KS, Cogswell ME, et al. Hypovitaminosis D prevalence and determinants among African American and white women of reproductive age: Third National Health and Nutrition Examination Survey, 1988-1994. Am J Clin Nutr. 2002;76:187-192. 13. Yetley EA. Assessing the vitamin D status of the US population. Am J Clin Nutr. 2008;88:558S-564S. 14. National Academy of Science, Institute of Medicine (2010). Dietary references intake for calcium and vitamin D. Available at: http:// www.nap.edu/catalog.php?record_id 13050. Accessed January 8, 2011. 15. Eastell R, Arnold A, Brandi ML, et al. Diagnosis of asymptomatic primary hyperparathyroidism: proceedings of the third international workshop. J Clin Endocrinol Metab. 2009;94:340-350. 16. Binkley N, Krueger D, Cowgill CS, et al. Assay variation confounds the diagnosis of hypovitaminosis D: a call for standardization. J Clin Endocrinol Metab. 2004;89:3152-3157. 17. Aloia JF, Talwar SA, Pollack S, et al. Optimal vitamin D status and serum parathyroid hormone concentrations in African American women. Am J Clin Nutr. 2006;84:602-609. 18. Heaney RP, Dowell MS, Hale CA, Bendich A. Calcium absorption varies within the reference range for serum 25-hydroxyvitamin D. J Am Coll Nutr. 2003;22:142-146. 19. Hansen KE, Jones AN, Lindstrom MJ, et al. Vitamin D insufficiency: disease or no disease? J Bone Miner Res. 2008;23:1052-1060. 20. Henry HL, Bouillon R, Norman AW, et al. 14th Vitamin D Workshop consensus on vitamin D nutritional guidelines. J Steroid Biochem Mol Biol. 2010;121:4-6. 21. Thomas MK, Lloyd-Jones DM, Thadhani RI, et al. Hypovitaminosis D in medical inpatients. N Engl J Med. 1998;338:777-783. 22. Saliba W, Rennert HS, Kershenbaum A, Rennert G. Serum 25(OH)D concentrations in sunny Israel. Osteoporos Int. 2011 DOI: 10.1007/ s00198-011-1597-y. 23. Souberbielle JC, Lawson-Body E, Hammadi B, et al. The use in clinical practice of parathyroid hormone normative values established in vitamin D-sufficient subjects. J Clin Endocrinol Metab. 2003;88: 3501-3504.