| 2008 JASN IMPACT FACTOR 7.505 | HOME AUTHOR INFO EDITORIAL BOARD SUBSCRIBE FEEDBACK ALERTS HELP | |||
| CURRENT ISSUE | ARCHIVES | JASN Express | ONLINE SUBMISSION | |
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
CLINICAL RESEARCH |






Departments of * Nephrourology and || Radiology, Great Ormond Street Hospital for Children, London and
Vascular Physiology Unit and
Centre for Paediatric Epidemiology and Biostatistics, University College London Institute of Child Health, London, and
Department of Chemical Pathology, West Park Hospital, Epsom, United Kingdom
Correspondence: Dr. Rukshana Shroff, Research Registrar, Nephrourology Unit, Institute of Child Health, 30 Guilford Street, London WC1N 1EH, UK. Phone: 0207-4059200, ext. 0065; Fax: 0207-8138532; E-mail: shrofr{at}gosh.nhs.uk
Received for publication September 10, 2007. Accepted for publication January 3, 2008.
| Abstract |
|---|
|
|
|---|
3 mo and in 40 age-matched control subjects. All dialysis patients were prescribed daily oral 1-
hydroxyvitamin D3. 92% of patients were deficient in 25-hydroxyvitamin D [25(OH)D]. 1,25-dihydroxyvitamin D [1,25(OH)2D] levels were low in 36% and high in 11% of patients. There was a weak correlation between 1
-hydroxyvitamin D3 dosage and 1,25(OH)2D levels. Both carotid intima-media thickness and calcification scores showed a U-shaped distribution across 1,25(OH)2D levels: patients with both low and high 1,25(OH)2D had significantly greater carotid intima-media thickness (P < 0.0001) and calcification (P = 0.0002) than those with normal levels. Low 1,25(OH)2D levels were associated with higher high-sensitivity C-reactive protein (P < 0.0001). Calcification was most frequently observed in patients with the lowest 1,25(OH)2D and the highest high-sensitivity C-reactive protein. In contrast, 25(OH)D levels did not correlate with any vascular measure. In conclusion, both low and high 1,25(OH)2D levels are associated with adverse morphologic changes in large arteries, and this may be mediated through the effects of 1,25(OH)2D on calcium-phosphate homeostasis and inflammation. For optimization of strategies to protect the vasculature of dialysis patients, careful monitoring of 1,25(OH)2D levels may be required. | Introduction |
|---|
|
|
|---|
Because the failing kidney cannot convert the "nutritional" 25-hydroxyvitamin D3 [25(OH)D] to the biologically active or "hormonal" form 1,25-dihydroxyvitamin D3 [1,25(OH)2D],4 1,25(OH)2D deficiency is widely prevalent in patients with CKD.9,10 Thus, vitamin D analogues are routinely used, but both vitamin D deficiency and supplementation with vitamin D analogues have been implicated as potential risk factors in the development and progression of vascular disease.6,7,11–16
Vitamin D analogues increase calcium (Ca) and also phosphate (PO4) absorption and suppress parathyroid hormone (PTH) secretion via the vitamin D receptor on the parathyroid gland.4,17,18 Inadequate 1,25(OH)2D levels would be expected, therefore, to result in low plasma Ca and hyperparathyroidism, whereas high 1,25(OH)2D levels cause hypercalcemia, hyperphosphatemia, and oversuppression of PTH. In turn, high bone turnover (as a result of hyperparathyroidism) results in an efflux of Ca and PO4 from the bones into the soft tissues,4 whereas low bone turnover (as a result of oversuppression of PTH) results in an inability of the bone to buffer fluxes in serum Ca and PO4.13,17,18 Thus, it is known that optimal 1,25(OH)2D levels are required to manipulate PTH so as to achieve effective parathyroid gland suppression and allow a normal bone turnover, but the effects of 1,25(OH)2D on the vasculature are poorly understood.
In addition to its endocrine effects, vitamin D has important autocrine/paracrine roles: The vitamin D receptor is ubiquitous19 and, importantly, is present in the vascular smooth muscle cells (VSMC),20 cardiomyocytes,9,21 and cells of the monocyte/macrophage lineage.22 1,25(OH)2D has important anti-inflammatory and immunomodulatory effects.15,23,24 Because inflammation plays a key role in vascular damage,25,26 vitamin D may have antiatherogenic effects. Thus, although vitamin D exerts potentially deleterious procalcific effects on the vasculature, its anti-inflammatory properties may confer a significant cardioprotective benefit. In this study, we hypothesize that 1,25(OH)2D levels in the normal range are associated with less vascular damage and calcification in children on dialysis.
| RESULTS |
|---|
|
|
|---|
|
|
|
1,25(OH)2D showed a linear correlation with both the Ca-PO4 product at the time of the study (P = 0.02, r = 0.28) as well as the mean time-averaged Ca-PO4 product (P = 0.03, r = 0.22; Table 2). Patients with high 1,25(OH)2D had more hypercalcemic episodes (15 versus 6%; P = 0.02). As classically observed, the mean time-averaged intact PTH (iPTH) level inversely correlated with 1,25(OH)2D [P = 0.02, r = 0.36 for mean time-averaged iPTH versus 1,25(OH)2D levels and P = 0.003, r = 0.41 for iPTH levels at the point of study versus 1,25(OH)2D levels]. No correlation was found between 25(OH)D or 1,25(OH)2D levels and the serum Ca, PO4, or alkaline phosphatase measured at any time point. Neither 25(OH)D nor 1,25(OH)2D showed any correlation with the patients' age, gender, race, dialysis vintage, BP, or body mass index. There was no seasonal variation in the levels.
Vitamin D Levels and Correlations with Vascular Measures
Both low and high levels of 1,25(OH)2D were associated with abnormal vascular measures: Patients with 1,25(OH)2D in the normal range had carotid intima-media thickness (cIMT) levels comparable to those in the control subjects (Table 3), but those with 1,25(OH)2D <40 or >150 pmol/L had significantly higher cIMT (P < 0.0001 for quadratic term; Figure 1A). The cardiac calcification score showed a similar relationship to 1,25(OH)2D: Calcification was seen in eight (36%) of 22 patients with 1,25(OH)2D levels <40 pmol/L, five (45%) of 11 patients with levels >150 pmol/L, and only one (3.6%) of 28 patient with levels in the normal range (P = 0.0002 for quadratic term; Figure 1B). No association was found between 1,25(OH)2D and pulse wave velocity (PWV) or between 25(OH)D and any of the vascular measures.
|
|
|
|
|
| DISCUSSION |
|---|
|
|
|---|
A total of 92% of our patients had 25(OH)D deficiency, and studies have shown that the majority of patients with CKD have low 25(OH)D levels9,27; however, unlike other studies in which the patients were naive to vitamin D sterols, all of our patients were prescribed daily alphacalcidol. The "nutritional" form, 25(OH)D, is unaffected by alphacalcidol treatment, so we also measured 1,25(OH)2D, the "hormonal," or "active," form of vitamin D produced by 25-hydroxylation of alphacalcidol in the liver,17,18 and found that 36% of the children had low 1,25(OH)2D. Reduced levels of 1,25(OH)2D despite treatment may be due to a loss of albumin-bound vitamin D in the peritoneum, as suspected in our patients, or nonconcordance with treatment. None of our patients had malabsorption syndromes or liver disorders to suggest poor absorption or reduced enzymatic conversion of alphacalcidol. Despite the strong inverse correlation between inflammation and vitamin D and the higher inflammatory status of HD patients, we found lower vitamin D levels in the PD cohort, suggesting that a loss of albumin-bound vitamin D did indeed play a major role. Earlier work from our group showed that in children on dialysis, the dosage of alphacalcidol influenced cIMT and calcification,3 but the lack of a consistent dosage-response relationship as shown in this study suggests that adjustments in the alphacalcidol dosage alone may not be sufficient for optimal management.
The effects of high vitamin D dosages on vascular calcification are widely recognized in clinical studies3,16,28–30 and in vitro work,20,31–33 and this study confirms an increased cIMT and calcification in patients with high 1,25(OH)2D levels. Vitamin D analogues may induce calcification by a number of mechanisms that include enhancing the gastrointestinal absorption of Ca and PO4, oversuppression of PTH leading to adynamic bone disease,13,17,18 and a direct effect on VSMC.20,31–33 An early study by Milliner et al.34 showed that, at autopsy, 60% of children with CKD had soft tissue calcification and 36% had systemic calcinosis, and use of any vitamin D analogue showed the strongest independent association with calcinosis. In addition, both 25(OH)D and 1,25(OH)2D can have a direct effect on the VSMC: 1,25(OH)2D upregulates the vitamin D receptor and induces cellular calcium uptake,31 decreases VSMC proliferation,20 and induces VSMC migration and osteoblastic conversion of the VSMC.32,33 Recent observational studies by Litwin et al.,29 Civilibal et al.,28 and Mitsnefes et al.30 and our own work3 have shown that a high vitamin D dosage adversely affects cIMT3,29,30 and calcification.3,28 Newer vitamin D analogues, such as paricalcitriol and doxercalciferol, are shown to be less calcemic35,36 but have only a marginal survival advantage over calcitriol,11,12 and their effects on the vasculature have not as yet been studied.
Despite the clearly deleterious calciotropic effects of vitamin D, in this study, low levels of 1,25(OH)2D were also associated with increased cIMT and calcification. Low 1,25(OH)2D results in an increase in PTH levels that can promote soft tissue calcification through its effect on calcium absorption and an efflux of Ca and PO4 from a high-turnover bone state.13,17,18 Our findings are supported by two studies of adult HD patients that demonstrated a approximately 20% survival advantage of any vitamin D formulation over no vitamin D treatment.11,12 In a small cohort of Japanese HD patients, Shoji et al.37 showed that the use of 1-
hydroxyvitamin D3 was associated with a 70% lower risk for death from cardiovascular disease as compared with a group not on any vitamin D supplements. London et al.38 recently showed that in a cohort of adult dialysis patients who all were naive to vitamin D analogues, low 25(OH)D and 1,25(OH)2D were associated with greater vessel stiffness and reduced brachial artery distensibility. In an earlier study, we showed that children who were on dialysis and had PTH levels more than two-fold the upper limit of normal (ULN) had higher cIMT, PWV, and calcification than those with PTH levels less than two-fold the ULN3; however, as seen in this study, 1,25(OH)2D levels were associated with cIMT and calcification independent of PTH, suggesting that the biologic consequences of 1,25(OH)2D extend beyond the regulation of Ca-PO4 homeostasis alone.
Anti-inflammatory actions of vitamin D22,26 may also contribute to its effects on the vasculature but have not been explored in clinical studies. Interestingly, we found a strong inverse correlation between hs-CRP and 1,25(OH)2D levels that was associated with vascular calcification but no association between 25(OH)D and hs-CRP. 1-
Hydroxylase is expressed on many cell types such as macrophages, endothelial cells, and dendritic cells, where it acts in an autocrine/paracrine manner independent of the PTH–bone axis and is unaffected by renal failure but may be regulated by immune stimuli. In vitro studies as well as studies in other inflammatory disease states, such as rheumatoid arthritis,39 have shown that vitamin D can influence various aspects of inflammation,40 including inhibition of antigen-presenting cell maturation, downregulation of NF-
B, and modulation of cytokine production to create an anti-inflammatory environment22,41 (increased IL-10 and decreased IL-6, IL-12, and TNF-
), but this is the first clinical study that has found an association between 1,25(OH)2D levels and inflammation; however, it is also possible that inflammation leads to low vitamin D levels. In a cross-sectional study, it would be impossible to discern a cause–effect relationship between inflammation-malnutrition and vitamin D levels, but given the results of in vitro studies that have shown a causal effect of vitamin D on inflammation, it is likely that low vitamin D levels contribute to the proinflammatory milieu in dialysis patients. Studies of patients without renal failure have shown that vitamin D supplementation can suppress serum TNF-
and increase IL-10 levels42: TNF-
promotes atherosclerosis and IL-10 has antiatherogenic properties. Vitamin D also has direct cardioprotective effects such as an antiproliferative effect on cardiomyocytes21 and negative endocrine regulation of the renin-angiotensin system.43
In this cross-sectional study, an association between vitamin D levels and vascular phenotype does not necessarily indicate a cause–effect relationship; however, this study serves as a starting point to stimulate cell biology work and generate hypotheses for randomized, controlled studies of the effects of vitamin D analogues on cardiovascular health. We were unable to find a correlation between hs-CRP levels and cIMT, suggesting that either the study population was too small to demonstrate an effect or inflammation has a greater influence on calcification than on the mechanisms involved in vessel thickening. Unlike the study by London et al.,38 we did not find a correlation between vitamin D levels and vessel stiffness. The greater plasticity of children's vessels and their shorter dialysis vintage may allow for compensatory mechanisms that can maintain normal vessel function in the face of early structural damage to the vessel. Vessels from children provide an ideal model to study uremic influences on the arterial wall, because they do not have the confounding proatherosclerotic risk factors that are prevalent in the adult CKD population.
In conclusion, we have shown, for the first time, that both low and high levels of 1,25(OH)2D are associated with adverse morphologic changes in the large arteries and that the vascular damage may be determined by the effects of vitamin D on Ca-PO4 homeostasis and inflammation. Given the narrow therapeutic window for vitamin D analogues on vascular health, optimal vascular protective strategies in dialysis patients may require careful monitoring of not only the vitamin D dosage but also 1,25(OH)2D levels.
| CONCISE METHODS |
|---|
|
|
|---|
3 mo were recruited. Levels of 25(OH)D, 1,25(OH)2D and hs-CRP were measured and related to cIMT, PWV, and the presence of coronary and valvular calcification on multislice computed tomography (CT) scan. Patients were compared with healthy age- and gender-matched school children who underwent vascular scans in our department as part of a parallel study investigating normal levels of cIMT and PWV in healthy children. From this cohort, we included 40 consecutive children aged 5 to 18 yr with no known medical illnesses or family history of heart disease or diabetes to serve as control subjects in our study, in a 1.5:1 patient:control subject ratio. Of note, the majority of children in the control group were overweight, although no intentional selection bias was involved. Informed written consent was obtained from all parents or caregivers and from children when appropriate. The study was approved by the local research ethics committee.
In our center PD is offered to all children as the initial dialysis modality unless specific contraindications such as recent intra-abdominal surgery, peritoneal infections, or difficult family circumstances preclude it. As a result, our group of patients on HD are often older and have a longer dialysis vintage. All patients in this study were prescribed daily oral 1-
hydroxycholecalciferol (alphacalcidol), titrating the dosage so as to keep the iPTH level less than two-fold the ULN. Hyperphosphatemia was managed by dietary PO4 restriction and calcium-based PO4 binders. Sevelamer, a newer non–calcium-based PO4 binder, was used only for patients with hypercalcemia and persistently high iPTH levels.
Data Collection
Serum Ca, PO4, and iPTH levels and the dosages of elemental calcium intake from phosphate binders and alphacalcidol were recorded at monthly intervals from the start of stage 4 CKD, and mean time-averaged values were calculated (the sum total of the monthly values for each variable was divided by the number of months of exposure to provide a mean time-averaged value). Given the variable amounts of time spent in stage 4 CKD, the biochemical values and medication dosages at the time of the study are expressed separately in Table 2. Because the iPTH assay has changed over time, iPTH is expressed as multiples of the ULN. The number of hypercalcemic episodes (defined as albumin-adjusted serum Ca levels >2.5 mmol/L and expressed as a percentage of the total number of albumin-adjusted serum Ca measurements in each patient) was calculated. In control subjects, a single blood test at the time of the scans was performed.
Vitamin D Levels and hs-CRP
All biochemical data were determined on blood samples taken before the vascular scans and before a midweek session of HD but at varying times throughout the year. There was an interval of 10 to 12 h between alphacalcidol intake and blood sampling. Serum 25(OH)D and 1,25(OH)2D levels were measured by a single technician using the enzyme immunoassay (IDS-OCTEIA 25-HydroxyvitaminD EIA kit; Immunodiagnostics Systems Ltd., Tyne and Wear, UK) and RIA (1,25 dihydroxyvitamin D 125I RIA kit; DiaSorin, Stillwater, MN), respectively. 1,25(OH)2D levels in 120 healthy normal children have been measured and validated in our laboratory (unpublished data), so we measured 25(OH)D levels only in the 40 control subjects. hs-CRP levels were determined by ELISA (EIA test kit; MP Biomedicals, Orangebury, NY) for all patients and control subjects.
Carotid Ultrasonography for cIMT
B-mode ultrasound of both common carotid arteries was performed using a 12-MHz linear array transducer (Vivid 7; GE Medical, Horton, Norway). Longitudinal two-dimensional images of the vessel 1 to 2 cm proximal to the carotid bulb were acquired on the R wave of the electrocardiogram (ECG), frozen in diastole, and analyzed off-line. The cIMT was calculated as the distance between the leading edge of the lumen-intima interface and the media-adventitia interface on the far wall of the artery.
Applanation Tonometry for PWV
Applanation tonometry in the carotid and femoral arteries was performed with a micromanometer (SPC-301; Millar Instruments, Houston, TX) and recorded directly onto a computer running proprietary software (SphygmoCor version 7.0; Scanmed, Gloucestershire, UK). An ECG gated signal and standard distance measurements were used. PWV was computed from contour analysis of 10 consecutive pressure waveforms recorded consecutively in the carotid and femoral arteries to give the aortic PWV. PWV was not measured in the control subjects because permission was not granted for femoral artery measurements.
Multislice-Spiral CT for Cardiac Imaging
Sixteen-slice spiral CT scans were performed (Somatom Sensation 16; Siemens Medical Solutions, Erlangen, Germany) using the standard Ca scoring protocol. Prospective ECG triggering was performed to obtain image acquisition in diastole, using 60% of the RR interval for image reconstruction. The Agatston score44 for each main epicardial coronary artery, cardiac valves, and aorta and an overall score for each patient were determined. In the control group, CT was not undertaken because of radiation concerns.
Statistical Analyses
Results are presented as means ± SD or median and range, depending on the distribution. Comparisons between groups were made using t test or the Mann-Whitney U test as appropriate, and correlations were tested using Pearson or Spearman correlation tests for parametric and nonparametric data, respectively. The two main outcome variables were cIMT and calcification score [transformed to log10(CS + 1) to adjust for skewness]. Factors affecting the two outcome variables were explored using multiple regression analysis. From univariate analyses, variables with P < 0.15 were entered into the stepwise multiple regression analyses. The nonlinear effect of vitamin D levels was fitted by including a quadratic term: The outcome variables were regressed on 1,25(OH)2D and the square of 1,25(OH)2D, and the significance of the squared (quadratic) term was noted. A third outcome measure, hs-CRP, was analyzed after log transformation to account for skewness, and its correlation with 1,25(OH)2D was strengthened when that, too, was log-transformed. The interaction between 1,25(OH)2D and hs-CRP was tested by two way ANOVA. P
0.05 was considered statistically significant. Statistical analyses were performed using SPSS version 12.0.1 (SPSS, Chicago, IL).
| DISCLOSURES |
|---|
|
|
|---|
| Acknowledgments |
|---|
We gratefully acknowledge Dr. Atul Singhal, Childhood Nutrition Research Centre, Institute of Child Health, London, for contributing data on the control subjects.
| Footnotes |
|---|
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
H. Eddington and J. G. Heaf Clinical management of disturbances of calcium and phosphate metabolism in dialysis patients NDT Plus, August 1, 2009; 2(4): 267 - 272. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. L. Melamed and R. Thadhani Low Calcidiol Levels and Coronary Artery Calcification: True, True, and Related? J. Am. Soc. Nephrol., August 1, 2009; 20(8): 1663 - 1665. [Full Text] [PDF] |
||||
![]() |
D. V. Barreto, F. C. Barreto, S. Liabeuf, M. Temmar, F. Boitte, G. Choukroun, A. Fournier, and Z. A. Massy Vitamin D Affects Survival Independently of Vascular Calcification in Chronic Kidney Disease Clin. J. Am. Soc. Nephrol., June 1, 2009; 4(6): 1128 - 1135. [Abstract] [Full Text] [PDF] |
||||
|
|
R. Shroff MONITORING CARDIOVASCULAR RISK FACTORS IN CHILDREN ON DIALYSIS Perit. Dial. Int., February 1, 2009; 29(Supplement_2): S173 - S175. [Abstract] [Full Text] [PDF] |
||||
![]() |
T. B. Drueke and E. Ritz Treatment of Secondary Hyperparathyroidism in CKD Patients with Cinacalcet and/or Vitamin D Derivatives Clin. J. Am. Soc. Nephrol., January 1, 2009; 4(1): 234 - 241. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. C. Shroff, R. McNair, N. Figg, J. N. Skepper, L. Schurgers, A. Gupta, M. Hiorns, A. E. Donald, J. Deanfield, L. Rees, et al. Dialysis Accelerates Medial Vascular Calcification in Part by Triggering Smooth Muscle Cell Apoptosis Circulation, October 21, 2008; 118(17): 1748 - 1757. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. C. Shroff, V. Shah, M. P. Hiorns, M. Schoppet, L. C. Hofbauer, G. Hawa, L. J. Schurgers, A. Singhal, I. Merryweather, P. Brogan, et al. The circulating calcification inhibitors, fetuin-A and osteoprotegerin, but not Matrix Gla protein, are associated with vascular stiffness and calcification in children on dialysis Nephrol. Dial. Transplant., October 1, 2008; 23(10): 3263 - 3271. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
HOME
CURRENT ISSUE
ARCHIVES
JASN Express
ONLINE SUBMISSION
AUTHOR INFO
EDITORIAL BOARD SUBSCRIBE FEEDBACK ALERTS HELP |