J Clin Endocrin Metab. First published ahead of print April 30, 2013 as doi:10.1210/jc.2013-1294

ORIGINAL E n d o c r i n e

ARTICLE R e s e a r c h

Relation Between Circulating Levels of 25(OH) Vitamin D and Parathyroid Hormone in Chronic Kidney Disease: Quest for a Threshold Marie Metzger, Pascal Houillier, Cédric Gauci, Jean Philippe Haymann, Martin Flamant, Eric Thervet, Jean-Jacques Boffa, François Vrtovsnik, Marc Froissart, Bénédicte Stengel, and Pablo Ureña-Torres on behalf of the NephroTest Study Group Institut National de la Santé et de la Recherche Médicale (INSERM) (M.M., C.G., M.Fr., B.S.), CESP Centre for Research in Epidemiology and Population Health, U1018, Diabetes, Obesity, and Chronic Kidney Disease Epidemiology Team, 94807 Villejuif, France; Université Paris Sud 11 (M.M., C.G., B.S.), Unité Mixte de Recherche (UMR) S 1018, F-94807, Villejuif, France; Departments of Physiology and Nephrology (C.G., P.H., E.T.), Hôpital Européen G Pompidou, Assistance Publique-Hôpitaux de Paris, France; Université Paris Descartes (P.H., E.T., M.Fr., P.U.-T.), Sorbonne Paris Cité, France; INSERM UMR S 872 (P.H.), Paris, France; Departments of Physiology (J.P.H., J.-J.B.) and Nephrology, Hôpital Tenon, Assistance Publique-Hôpitaux de Paris, Paris, France; Université Pierre et Marie Curie (J.P.H., J.-J.B.), Paris, France; INSERM UMR S 702 (J.P.H., J.-J.B.), Paris, France; Departments of Physiology (M.Fl., F.V.) and Nephrology, Hopital Bichat, Assistance Publique-Hôpitaux de Paris, Paris, France; Université Paris Diderot (M.Fl., F.V.), Paris, France; INSERM UMR S 699 (M.Fl., F.V.), Paris, France; INSERM UMR S 775 (M.Fr.), Paris, France; and Nephology Dialysis (P.U.-T.), Clinique du Landy, St Ouen, France

Context: Vitamin D deficiency is common in patients with chronic kidney disease (CKD). Current guidelines recommend treatment strategies in these patients similar to those for the general population, but the vitamin D nutritional status sufficient to prevent PTH levels from increasing in CKD is unknown. Objective, Main Outcome Measure: To study the relation between circulating PTH and 25(OH)D levels and to search for a 25(OH)D threshold associated with a significant PTH increase. Design, Setting, and Patients: In the hospital-referred NephroTest cohort study, we measured 25(OH)D, PTH, and glomerular filtration rate (mGFR) by 51Cr-EDTA renal clearance in 929 adult patients with nondialysis CKD stages 1 to 5 and no vitamin D supplementation. Patients’ mean age was 60.1 ⫾ 14.7 years; 71% were men, and 9% were black. Their mean mGFR was 37.8 mL/min/1.73 m2. Results: We found a 25(OH)D threshold of 8 ng/mL with an upper limit of 20 ng/mL (95% confidence interval) by linear piecewise regression modeling of log-PTH for 25(OH)D adjusted for mGFR, age, race, and ionized calcium level. The smoothed curve confirmed that PTH concentration rose steeply when circulating 25(OH)D levels fell to less than 20 ng/mL. Conclusions: Spontaneous 25(OH)D levels greater than 20 ng/mL seem sufficient to control serum PTH in CKD patients. This result reinforces guidelines to supplement vitamin D only if less than 30 ng/mL. (J Clin Endocrinol Metab 98: 0000 – 0000, 2013)

N

ative vitamin D, cholecalciferol (25[OH]D3) and ergocalciferol (25[OH]D2) are prehormones that play an essential role in mineral and bone homeostasis. If adequate vitamin D3 is not provided, then the adminis-

tration of less potent vitamin D2 is recommended. Vitamin D stimulates intestinal absorption and kidney reabsorption of calcium and phosphate, mainly through its dominant active metabolite 1,25(OH)2D3. In the para-

ISSN Print 0021-972X ISSN Online 1945-7197 Printed in U.S.A. Copyright © 2013 by The Endocrine Society Received February 1, 2013. Accepted April 9, 2013.

Abbreviations: CI, confidence interval; CKD, chronic kidney disease; F, Fisher statistics; mGFR, measured glomerular filtration rate.

doi: 10.1210/jc.2013-1294

J Clin Endocrinol Metab

Copyright (C) 2013 by The Endocrine Society

jcem.endojournals.org

1

2

Metzger et al

25(OH) Vitamin D/PTH Threshold in CKD

thyroid gland, vitamin D suppresses PTH production (1). Consequently, low circulating vitamin D levels invariably result in elevated serum PTH concentrations in healthy individuals (2, 3) as well as in patients with chronic kidney disease (CKD) (4 – 6). Moreover, these low vitamin D levels are associated with a variety of nonskeletal alterations, including cardiovascular disease, diabetes mellitus, multiple sclerosis, cancer, and immune system dysfunction (1). In the general population, 25(OH)D status has been defined by either a single criterion or a combination of criteria or of changes in several of them: serum PTH concentration (2), circulating 1,25(OH)2D3 levels (7), intestinal calcium absorption (8), muscle strength, and bone mineral density (9). The definition of deficiency as ⬍10 ng/mL (25 nmol/L) has been widely recognized because of its association with muscle weakness, bone pain, fractures, and high PTH (1). The definitions of normality and insufficiency, however, remain controversial, with values proposed ranging between 10 and 40 ng/mL (25– 80 nmol/L). Despite the lack of scientific evidence, the Kidney Disease: Improving Global Outcomes recommended in 2009 that circulating 25(OH)D levels be monitored in patients with CKD stages 3–5D and treatment strategies adopted for vitamin D deficiency and insufficiency similar to those for the general population (10). For these reasons, and because of the ongoing debate about sufficient vitamin D nutritional status in CKD, we studied the relations between circulating 25(OH)D and PTH levels in patients with nondialysis CKD. We searched for a threshold 25(OH)D concentration below which PTH levels would significantly increase, with or without significant changes in blood and urinary levels of calcium and phosphate.

Materials and Methods Patients and study design The NephroTest study is a prospective hospital-based cohort, enrolling adult patients with CKD stages 1 to 5, not pregnant, not on dialysis or living with a kidney transplant, and referred to any of 3 physiology departments for extensive workups (11). All patients signed an informed consent at inclusion. Of 1294 patients included between January 2000 and December 2009, we analyzed baseline data from the 929 not receiving vitamin D supplementation (80% of the participants) and for whom PTH and 25(OH)D measurements and treatment information were available. All patients signed informed consent.

Measures As previously reported, 5-hour in-person visits for a complete nephrological workup provided a large quantity of clinical and laboratory data (12). We measured glomerular filtration rates (mGFR) by 51Cr-EDTA renal clearance. 25(OH) vitamin D concentrations were measured in plasma by a radioimmunologic

J Clin Endocrinol Metab

method (DiaSorin) that recognized both 25(OH) vitamin D2 and D3 with similar affinity. Normal values range from 10 to 40 pg/mL. We measured serum PTH concentrations using 3 secondgeneration 2-site radio-immunometric assays (Allegro-Intact PTH, then with the Allegro-calibrated Intact PTH Advantage assay; Nichols Institute Diagnostics, San Clemente, California [normal values 10 –58 pg/mL], and since January 2004 with the Elecsys chemoluminescent assay from Roche [normal values 10 – 65 pg/mL]) that yielded quite similar results (13). Phosphate concentrations were measured by colorimetry in plasma (phosphomolybdate assay) and ionized calcium by a specific electrode in serum.

Statistical analysis Clinical and biological data were expressed as percentages, means (⫾SD), or median (interquartile range), as appropriate. Associations of circulating 25(OH)D with serum PTH, phosphate, ionized calcium, and albumin-corrected calcium, as well as with urinary phosphate, 24-hour urinary calcium, and fasting urinary calcium, were tested with Spearman rank correlation coefficient. The Cochran-Armitage test for trend was used to investigate the relations between abnormal or extreme values of some of these parameters and 25(OH)D levels: prevalence of low-serum PTH (⬍20 pg/mL), hypercalcemia (defined by serum ionized calcium ⬎1.35 mmol/L or by serum albumin-corrected calcium ⬎2.65 mmol/L), and high urinary calcium (defined by fasting urinary calcium ⬎90th percentile or by 24-h urinary calcium ⬎90th) are reported by 25(OH)D classes (⬍10, 10 –20, 20 –30, 30 – 40, ⬎40 ng/mL). The relation between log-transformed PTH and 25(OH)D concentrations was modeled with piecewise linear regression: ln(PTH) ⫽ b0 ⫹ b1 䡠 min(25OHD-S,0) ⫹ b2 䡠 max(25OHD-S,0) ⫹ ⌺bixi, ⫹ e, with S the 25(OH)D threshold, bi the regression coefficient for covariates xi, and e the residual error. The regression function is continuous at the breakpoint but can be expressed before and after the breakpoint as follows: ln(PTH) ⫽ b0 ⫹ b1(25(OH)D-S)⫹ ⌺bixi, ⫹ e when 25(OH)D ⬍ S, and ln(PTH) ⫽ b0 ⫹ b2(25(OH)D-S) ⫹ ⌺bixi,, ⫹ e, when 25(OH)D ⱖ S. To determine whether the relation between log-transformed PTH and 25(OH)D was linear or whether there was a threshold of 25(OH)D associated with a significant log-PTH increase, we compared models with and without thresholds to estimate logPTH values. In practice, Fisher statistics (F) were computed to compare models with threshold values of [i] and the linear regression model, for all possible i values of 25(OH)D. The best 25(OH)D threshold for predicting log-PTH corresponds to the maximum F (Figure A in Supplemental data, published on The Endocrine Society’s Journals Online web site at http://jcem. endojournals.org). The 95% confidence interval (CI) of the 25(OH)D threshold was estimated with the bootstrap method and 1000 resamples. Models were run with and without adjustment for mGFR, serum ionized calcium, age, black race, serum albumin, winter season, and 24-hour urinary phosphate. Outlying observations were excluded (12 observations with PTH greater than 400 pg/mL and 1 with 25[OH]D ⫽ 80 ng/mL). Local polynomial regressions of PTH for 25(OH)D were also performed, and smoothed curves were traced through the scatter plot (lowess function in R). Statistical analyses were performed with SAS 9.2 (SAS Institute, Cary, North Carolina) and R (2.13; The R Foundation for Statistical Computing).

doi: 10.1210/jc.2013-1294

jcem.endojournals.org

Table 1. Patients’ characteristics (n ⫽ 929) Mean (SD), Median (IQR), or % 60.4 ⫾ 14.5 71 8.9 26.7 ⫾ 5.0 30 18 139 ⫾ 21/76 ⫾ 11

Age, y Men, % African origin, % Body mass index, kg/m2 Diabetes mellitus, % Cardiovascular disease, % Systolic/diastolic blood pressure, mm Hg Total cholesterol, mg/dL Serum albumin, g/dL Urinary albumin-to-creatinine ratio, mg/g 25(OH) vitamin D, ng/mL 1,25(OH)2D3, pg/mL mGFR, mL/min/1.73 m2 CKD stages, mGFR in mL/ min/1.73 m2, % 1–2 (ⱖ60) 3a (45–59) 3b (30 – 44) 4 (15–29) 5 (⬍15)

193 ⫾ 48 3.96 ⫾ 0.50 87 (19 – 480) 17.6 (11.0 –26.8) 24.0 (16.7–33.8) 37.8 (27.2–51.3) 15 19 31 27 8

Abbreviation: IQR, interquartile range.

Results Patient characteristics Patients were mainly men. Fewer than 10% came from sub-Saharan Africa or the French West Indies (Table 1). About 20% were obese (body mass index ⱖ30 kg/m2), and one-third had diabetes mellitus. Half had moderate CKD (stages 3a and 3b) and 35% had advanced CKD (stages 4 and 5). The prevalence of vitamin D deficiency (⬍10 ng/mL) and insufficiency (10 – 40 ng/mL) were dramati-

3

cally high, 19.4% and 72.3%, respectively; only 8.3% of the patients had a circulating 25(OH)D level greater than 40 ng/mL. Relation between circulating 25(OH)D and PTH concentrations Serum PTH levels increased significantly as circulating 25(OH)D concentrations declined (Table 2). Piecewise linear regression modeling of log PTH for 25(OH)D showed that a threshold model predicted PTH values better than a linear regression model, with an estimated threshold for 25(OH)D of 9.1 ng/mL, 95% CI [5.2–25.3], P ⬍ .001 (Figure 1). Adjusting for ionized calcium, mGFR, African origin, age, and center significantly improved the overall model performance (R2 ⫽ 0.39 versus 0.10 for the model without adjustment) and changed the 25(OH)D threshold only slightly, to 8.0 ng/mL, 95% CI [2.4 –20.0], P ⬍ .005 (Supplemental Data, Figure B). Estimated slopes were significantly negative both above and below this threshold, but the increase in log-PTH with decreasing 25(OH)D was stronger below it (Table 3). There was no significant interaction between 25(OH)D and any of the above adjustment variables in relation with log-PTH. Further adjustment for season, gender, serum albumin, and 24-hour urinary phosphate did not improve model prediction (data not shown). The smoothed curve between PTH and 25(OH)D also showed a nonlinear relation with an inflection point around 20 ng/mL of 25(OH)D (Figure 2). The prevalence of low PTH values (⬍20 ng/mL) increased slightly from 2.5% to 6.6% as 25(OH)D values increased from ⬍10 ng/mL to ⬎40 ng/mL (Table 4).

Table 2. Parathyroid Hormone (PTH), Serum and Urinary Phosphate, and Calcium Levels According to Circulating 25(OH) Vitamin D Levels 25(OH) Vitamin D, ng/mL

N PTH, pg/mL Phosphate Serum, mg/dL Urinary, mg/24 h Calcium Serum ionized, mg/dL Serum total,b mg/dL 24-h urinary, mg/24 h Fasting urinary CaU, mg/mmol creatinine

Overall

50

P Value

929 65 (40 –109)

180 99 (57–161)

341 66 (43–114)

227 58 (40 –90)

104 42 (30 – 84)

50 46 (32–75)

27 40 (25– 66)

⬍.0001

3.35 ⫾ 0.69 646 (495– 845)

3.39 ⫾ 0.73 623 (436 – 805)

3.36 ⫾ 0.67 614 (501– 834)

3.34 ⫾ 0.73 696 (520 – 886)

3.32 ⫾ 0.61 667 (498 – 832)

3.25 ⫾ 0.61 677 (492– 816)

3.25 ⫾ 0.83 809 (521– 872)

.2 .004

4.84 ⫾ 0.26

4.76 ⫾ 0.25

4.84 ⫾ 0.26

4.90 ⫾ 0.27

4.88 ⫾ 0.26

4.88 ⫾ 0.20

4.83 ⫾ 0.24

⬍.0001

9.03 ⫾ 0.52

9.00 ⫾ 0.53

9.04 ⫾ 0.51

9.03 ⫾ 0.53

9.03 ⫾ 0.49

8.96 ⫾ 0.42

9.14 ⫾ 0.75

.99

49 (26 –92)

38 (19 – 68)

44 (23– 88)

57 (30 –106)

58 (29 –98)

54 (32–99)

57 (35–145)

⬍.0001

2.4 (1.0 –5.6)

1.6 (0.8 – 4.0)

2.3 (0.9 –5.3)

2.8 (1.2– 6.0)

3.0 (1.2– 6.8)

2.4 (1.2–5.8)

1.7 (0.8 – 8.0)

⬍.0001

Values are mean (SD) or median (interquartile range). a

P for trend using Spearman rank correlation to test statistical significance between quantitative variables.

b

Albumin-corrected calcium.

4

Metzger et al

25(OH) Vitamin D/PTH Threshold in CKD

Figure 1. Piecewise linear regression model, log-PTH according to 25(OH)D level. Best threshold at 9.1 ng/mL (95%CI 5.2–25.3) of 25(OH)D.

Blood and urinary calcium and phosphate according to 25(OH)D values Serum ionized calcium concentration, 24-hour urinary calcium excretion, fasting urinary calcium/creatinine ratio, and 24-hour urinary phosphate excretion all increased significantly as 25(OH)D concentration increased (Table 2). Hypercalcemia (ionized calcium ⬎1.35 mmol/L), howTable 3. Piecewise Linear Regression Results, Log(PTH) According to 25(OH) Vitamin D Level

Below threshold, 25(OH)D ⬍8 ng/mL Intercept Slope Above threshold, 25(OH)D ⱖ8 ng/mL Intercept Slope Adjustment variables African vs non-African origin Age, y Serum ionized calcium, mmol/L CKD stages, mGFR in mL/min/1.73 m2 5 (⬍15) 4 (15–29) 3b (30 – 44) 3a (45–59) 1–2 (ⱖ60) Center 1 vs 2

␤

Standard Error

P

6.68 ⫺.07

0.44 0.02

⬍.0001 ⬍.001

6.20 ⫺.01

0.44 0.001

⬍.0001 ⬍.0001

.18

0.07

.006

.003 ⫺1.57

0.001 0.33

.05 ⬍.0001

.73 .39 Ref ⫺.27 ⫺.59 ⫺.15

0.09 0.05

⬍.0001 ⬍.0001

0.05 0.06 0.04

⬍.0001 ⬍.0001 ⬍.001

J Clin Endocrinol Metab

Figure 2. Nonparametric regression (LOESS).

ever, was very uncommon, seen in fewer than 1% of patients, and did not vary significantly with circulating 25(OH)D levels (Table 4). The prevalence of 24-hour urinary calcium values above the 90th percentile increased significantly in patients with the highest 25(OH)D values, but no significant variation was observed for fasting urinary calcium. We found no association between plasma phosphate and 25(OH)D levels.

Discussion The main result of this study is that the negative relation between serum PTH and 25(OH)D concentrations in CKD patients is nonlinear. Serum PTH concentrations rise steeply when circulating 25(OH)D levels fall below the threshold of 8 ng/mL, and the increase in PTH is significantly attenuated when 25(OH)D values exceed 20 ng/mL. We also show that the prevalence of hypercalcemia and markedly suppressed PTH is low when circulating 25(OH)D concentrations are within normal limits. The major strength of the NephroTest study is its reliance on a large number of well-phenotyped patients. Here, this enables an accurate examination of the shape of the relation between serum PTH and 25(OH)D levels in CKD. The assessment of serum ionized calcium as well as of urinary calcium and phosphate excretion also allowed us to see whether normal or relatively high circulating 25(OH)D or low PTH levels were associated with any potentially harmful biochemical alterations, such as hypercalcemia, hypercalciuria, or hyperphosphaturia.

doi: 10.1210/jc.2013-1294

jcem.endojournals.org

5

Table 4. Proportion (%, n) of Low Serum PTH (⬍20 pg/mL), Hypercalcemia (serum-albumin corrected ⬎2.65 mmol/L or serum ionized ⬎1.35 mmol/L), and High Urinary Calcium (24-h and fasting) According to Circulating 25(OH) Vitamin D Levels 25(OH) Vitamin D (ng/mL)

No. of patients Low serum PTH (⬍20 pg/mL) Calcium Serum ionized (⬎1.35 mmol/L) Serum albumin-corrected (⬎2.65 mmol/L) Fasting urinary (⬎90th percentile) 24-h urinary (⬎90th percentile) a

Overall

40

P for Trend

842a 4.2 (35)

159 2.5 (4)

303 4.0 (12)

209 3.8 (8)

95 6.3 (6)

76 6.6 (5)

.01

0.5 (4)

0 (0)

0 (0)

1.4 (3)

0 (0)

1.3 (1)

.1

2.4 (20)

1.3 (2)

1.7 (5)

4.8 (10)

3.2 (3)

0 (0)

.6

–

8.2 (13)

8.3 (25)

11.5 (24)

12.6 (12)

11.8 (9)

.4

–

5.7 (9)

7.6 (23)

11.0 (23)

15.8 (15)

15.8 (12)

.0007

Reduced sample size because of missing data for fasting and 24-h urinary calcium.

In healthy subjects from the general population, serum PTH is inversely correlated with 25(OH)D (14 –16). In addition, more than 70 studies have reported circulating 25(OH)D thresholds that fluctuate between 10 and 50 ng/mL (25–125 nmol/L) without any clearcut threshold (17). In most of these studies, serum PTH concentrations begin to decrease as circulating 25(OH)D levels rise to 15 to 20 ng/mL (37.5 to 50 nmol/L) and are maximally suppressed at 30 to 40 ng/mL (75 to 100 nmol/L) (3, 14, 18 –21). Some studies also suggest that the relation between 25(OH)D and PTH may vary with age and ethnicity (22, 23). Another study, using median values of 25(OH)D from a very large dataset, intimated that there is no threshold, because the circulating 25(OH)D-dependent PTH levels revealed no threshold above which increasing 25(OH)D failed to suppress PTH further (22). The negative correlation between circulating 25(OH)D and PTH in CKD populations is also well known (24 –26). Unfortunately, none of these studies assessed whether a 25(OH)D threshold could be established below which PTH would rise markedly. They all suggested that PTH increases significantly when 25(OH)D values fall to less than 30 ng/mL. Here, the piecewise linear regression model provided the best fit after adjusting for ionized calcium, mGFR, ethnicity, and age, and estimated that there was a 25(OH)D threshold at 8 ng/mL and an upper limit at 20 ng/mL (95%CI). Local polynomial regression between PTH and 25(OH)D showed an inflection point around 20 ng/mL of 25(OH)D. What this means is that when 25(OH)D levels exceed 20 ng/mL, serum PTH is likely to be within normal limits, and if it does rise, the reason is less likely to be related to the circulating 25(OH)D concentration. The principal clinical implication of these findings is that circulating 25(OH)D levels less than 20 ng/mL should

be avoided in CKD patients to prevent the stimulation of PTH secretion. They also suggest that the level for starting supplementation with native vitamin D in CKD patients recommended by K-DOQI and Kidney Disease: Improving Global Outcomes (30 ng/mL) is probably adequate for avoiding a further decrease if the main objective is PTH control (10, 27). Our analysis does not, however, allow us to infer the existence of a similar 25(OH)D threshold for the association of vitamin D with mortality and CKD progression. The 3 main previous studies in CKD patients (28 –30) found an excess of risk for mortality and progression toward end-stage renal disease when 25(OH)D values were lower than 10 –15 ng/mL compared with values greater than 10 –15 ng/mL, with no evidence of any threshold. Similarly, we cannot know whether the same vitamin D threshold applies to other extramineral action of vitamin D, including protection against cardiovascular diseases, type 2 diabetes mellitus, immunological disorders, and malignancy (1, 31). It is generally accepted that intestinal calcium absorption is reduced in CKD subjects and that increasing circulating 25(OH)D levels significantly increase this absorption (32). Unfortunately, intestinal calcium absorption was not measured in this study, and thus, we do not know whether the vitamin D threshold of 20 ng/mL observed for PTH might also apply to the degree of intestinal calcium absorption. We observed that circulating 25(OH)D was positively correlated with ionized serum calcium levels (12) and with 24-hour urinary calcium excretion, findings that illustrate the dependence of intestinal calcium absorption on 25(OH)D. Moreover, controversy subsists about the potential harmful effects of excessive natural vitamin D treatment or high circulating 25(OH)D concentrations: reports are contradictory (33–39). Here we observed that serum ion-

6

Metzger et al

25(OH) Vitamin D/PTH Threshold in CKD

ized calcium, 24-hour urinary calcium excretion, fasting urinary calcium/creatinine ratio, and 24-hour urinary phosphate excretion each increased significantly when 25(OH)D exceeded 20 ng/mL (Table 2). However, the values of most of these parameters were still within normal ranges, and hypercalcemia remained very uncommon (⬍1% of patients). Likewise, the prevalence of extremely low PTH values increased only slightly, from 2.5% to 6.6%, as 25(OH)D values rose from ⬍10 ng/mL to ⬎40 ng/mL. Thus, taken together, these results demonstrate the safety and low risk of changes in mineral levels with circulating 25(OH)D levels up to 40 ng/mL in this particular CKD population. Our study has some limitations. Circulating 25(OH)D concentrations were measured by a standard commercial assay (DiaSorin) and not by a highly specific method, such as HPLC or HPLC-tandem mass spectrometric analysis, which recognizes 25(OH)D2 and D3 separately. The same is true for the PTH assays: the 2 second-generation assays we used interfere with some of the N-truncated PTH fragments (40). Finally, it should be recognized that the relation between 25(OH)D and PTH in a given subject is the result of a complex interplay of factors, including age, sex, ethnicity, genetics, renal function, nutritional calcium, phosphate, and magnesium intake, and the variability of PTH and 25(OH)D measurements. This makes it impossible to select a single threshold. In conclusion, given the high prevalence of vitamin D deficiency and insufficiency in this particular population of CKD patients and the availability of inexpensive and easy interventions to boost circulating concentrations of 25(OH)D, we should systematically supplement with vitamin D before 25(OH)D values drop to less than 20 ng/mL to reduce the risk of secondary hyperparathyroidism and thus reinforce the current guidelines that call for supplementation less than 30 ng/mL.

Acknowledgments Collaborators in the NephroTest Study Group are Emmanuel Letavernier, Pierre Ronco, Hafedh Fessi (Hôpital Tenon); Eric Daugas, Caroline du Halgouet (Hôpital Bichat); Renaud de La Faille, Christian d’Auzac, Gerard Maruani, Marion Vallet, Laurence Nicolet-Barousse, Mélanie Roland, and Christian Jacquot (Hôpital Européen G. Pompidou). We thank Jo Ann Cahn for revising the English of this manuscript. Address all correspondence and requests for reprints to: Marie Metzger, Inserm, CESP U1018, Team 10, 16 Avenue Paul Vaillant Couturier, 94807 Villejuif, France. E-mail: [email protected]. This work was supported by the NephroTest CKD cohort study grants from Inserm GIS-IReSP AO 8113LS TGIR (B.S.);

J Clin Endocrinol Metab

French Ministry of Health AOM 09114 and AOM 10245 (M.Fr.); Inserm AO 8022LS (B.S.); Agence de la Biomédecine R0 8156LL (B.S.); AURA, and Roche 2009-152-447G (M.Fr.). Disclosure Summary: P.U.-T. has received honoraria, research funds, and consulting fees from Abbott, Amgen, Novartis/ Genzyme, Reata, Shire, and Fresenius. P.H. has received honoraria and/or research funds from Amgen and Otsuka Ltd. M.Fr. has received honoraria and/or research funds from Affymax, Genzyme, Hoffmann-La-Roche, Novartis, Sandoz, and Vifor. M.Fr. has been employed by Amgen since January 1, 2011, but was a full-time academic associate professor during the time of study conception and data collection. B.S. has received honoraria from Amgen and Gambro.

References 1. Holick MF. Vitamin D deficiency. N Engl J Med. 2007;357:266 – 281. 2. Thomas MK, Lloyd-Jones DM, Thadhani RI, et al. Hypovitaminosis D in medical inpatients. N Engl J Med. 1998;338:777–783. 3. Chapuy MC, Preziosi P, Maamer M, et al. Prevalence of vitamin D insufficiency in an adult normal population. Osteoporos Int. 1997; 7:439 – 443. 4. Dusso AS, Brown AJ, Slatopolsky E. Vitamin D. Am J Physiol Renal Physiol. 2005;289:F8 –F28. 5. Holick MF. Vitamin D for health and in chronic kidney disease. Semin Dial. 2005;18:266 –275. 6. Ritter CS, Armbrecht HJ, Slatopolsky E, Brown AJ. 25-Hydroxyvitamin D(3) suppresses PTH synthesis and secretion by bovine parathyroid cells. Kidney Int. 2006;70:654 – 659. 7. Holick MF. High prevalence of vitamin D inadequacy and implications for health. Mayo Clin Proc. 2006;81:353–373. 8. 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. 9. Bischoff-Ferrari HA, Dietrich T, Orav EJ, Dawson-Hughes B. Positive association between 25-hydroxy vitamin D levels and bone mineral density: a population-based study of younger and older adults. Am J Med. 2004;116:634 – 639. 10. KDIGO. KDIGO clinical practice guideline for the diagnosis, evaluation, prevention, and treatment of Chronic Kidney Disease-Mineral and Bone Disorder (CKD-MBD). Kidney Int Suppl. 2009;S1– S130. 11. Moranne O, Froissart M, Rossert J, et al for the NephroTest Study Group. Timing of onset of CKD-related metabolic complications. J Am Soc Nephrol. 2009;20:164 –171. 12. Urena-Torres P, Metzger M, Haymann JP, et al for the NephroTest Study Group. Association of kidney function, vitamin D deficiency, and circulating markers of mineral and bone disorders in CKD. Am J Kidney Dis. 2011;58:544 –553. 13. Souberbielle JC, Boutten A, Carlier MC, et al. Inter-method variability in PTH measurement: implication for the care of CKD patients. Kidney Int. 2006;70:345–350. 14. Krall EA, Sahyoun N, Tannenbaum S, Dallal GE, Dawson-Hughes B. Effect of vitamin D intake on seasonal variations in parathyroid hormone secretion in postmenopausal women. N Engl J Med. 1989; 321:1777–1783. 15. Kruse K, Bartels H, Kracht U. Parathyroid function in different stages of vitamin D deficiency rickets. Eur J Pediatr. 1984;141:158 – 162. 16. 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;67:644 – 650. 17. Sai AJ, Walters RW, Fang X, Gallagher JC. Relationship between

doi: 10.1210/jc.2013-1294

18.

19.

20.

21. 22.

23. 24.

25.

26.

27.

28.

29.

vitamin D, parathyroid hormone, and bone health. J Clin Endocrinol Metab. 2011;96:E436 –E446. Aloia JF, Talwar SA, Pollack S, Feuerman M, Yeh JK. Optimal vitamin D status and serum parathyroid hormone concentrations in African American women. Am J Clin Nutr. 2006;84:602– 609. 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. Ginde AA, Wolfe P, Camargo CA Jr, Schwartz RS. Defining vitamin D status by secondary hyperparathyroidism in the U.S. population. J Endocrinol Invest. 2012;35:42– 48. Monk RD, Bushinsky DA. Making sense of the latest advice on vitamin D therapy. J Am Soc Nephrol. 2011;22:994 –998. Valcour A, Blocki F, Hawkins DM, Rao SD. Effects of age and serum 25-OH-Vitamin D on serum parathyroid hormone levels. J Clin Endocrinol Metab. 2012;97:3989 –3995. Aloia JF, Chen DG, Chen H. The 25(OH)D/PTH threshold in black women. J Clin Endocrinol Metab. 2010;95:5069 –5073. González EA, Sachdeva A, Oliver DA, Martin KJ. Vitamin D insufficiency and deficiency in chronic kidney disease. A single center observational study. Am J Nephrol. 2004;24:503–510. Ishimura E, Tsuchida T. [Vitamin D deficiency/insufficiency in patients with chronic kidney disease stage 3 and 4 — current concept and its therapeutic strategy]. Clin Calcium. 2004;14:764 –769. LaClair RE, Hellman RN, Karp SL, et al. Prevalence of calcidiol deficiency in CKD: a cross-sectional study across latitudes in the United States. Am J Kidney Dis. 2005;45:1026 –1033. Moe SM, Chertow GM, Coburn JW, et al. Achieving NKF-K/DOQI bone metabolism and disease treatment goals with cinacalcet HCl. Kidney Int. 2005;67:760 –771. Melamed ML, Astor B, Michos ED, Hostetter TH, Powe NR, Muntner P. 25-hydroxyvitamin D levels, race, and the progression of kidney disease. J Am Soc Nephrol. 2009;20:2631–2639. Pilz S, Henry RM, Snijder MB, et al. Vitamin D deficiency and myocardial structure and function in older men and women: the Hoorn Study. J Endocrinol Invest. 2010;33:612– 617.

jcem.endojournals.org

7

30. Ravani P, Malberti F, Tripepi G, et al. Vitamin D levels and patient outcome in chronic kidney disease. Kidney Int. 2009;75:88 –95. 31. Souberbielle JC, Body JJ, Lappe JM, et al. Vitamin D and musculoskeletal health, cardiovascular disease, autoimmunity and cancer: recommendations for clinical practice. Autoimmun Rev. 2010;9: 709 –715. 32. Rutherford WE, Blondin J, Hruska K, Kopelman R, Klahr S, Slatopolsky E. Effect of 25-hydroxycholecalciferol on calcium absorption in chronic renal disease. Kidney Int. 1975;8:320 –324. 33. Faupel-Badger JM, Diaw L, Albanes D, Virtamo J, Woodson K, Tangrea JA. Lack of association between serum levels of 25-hydroxyvitamin D and the subsequent risk of prostate cancer in Finnish men. Cancer Epidemiol Biomarkers Prev. 2007;16:2784 –2786. 34. Jia X, Aucott LS, McNeill G. Nutritional status and subsequent all-cause mortality in men and women aged 75 years or over living in the community. Br J Nutr. 2007;98:593–599. 35. Melamed ML, Michos ED, Post W, Astor B. 25-hydroxyvitamin D levels and the risk of mortality in the general population. Arch Intern Med. 2008;168:1629 –1637. 36. Stolzenberg-Solomon RZ, Jacobs EJ, et al. Circulating 25-hydroxyvitamin D and risk of pancreatic cancer: Cohort Consortium Vitamin D Pooling Project of Rarer Cancers. Am J Epidemiol. 2010; 172:81–93. 37. Tuohimaa P, Tenkanen L, Ahonen M, et al. Both high and low levels of blood vitamin D are associated with a higher prostate cancer risk: a longitudinal, nested case-control study in the Nordic countries. Int J Cancer. 2004;108:104 –108. 38. Vieth R. Vitamin D toxicity, policy, and science. J Bone Miner Res 22 Suppl. 2007;2:V64 –V68. 39. Visser M, Deeg DJ, Puts MT, Seidell JC, Lips P. Low serum concentrations of 25-hydroxyvitamin D in older persons and the risk of nursing home admission. Am J Clin Nutr. 2006;84:616 – 622; quiz 671– 612. 40. Brossard JH, Cloutier M, Roy L, Lepage R, Gascon-Barré M, D’Amour P. Accumulation of a non-(1– 84) molecular form of parathyroid hormone (PTH) detected by intact PTH assay in renal failure: importance in the interpretation of PTH values. J Clin Endocrinol Metab. 1996;81:3923–3929.