Persistent secondary hyperparathyroidism and vertebral fractures in kidney transplantation: Role of calcium-sensing receptor polymorphisms and vitamin D deficiency

Authors


Abstract

Bone morbidity remains a major problem even after successful renal transplantation. We investigated the role of calcium-sensing receptor (CaSR) polymorphisms and 25-hydroxyvitamin D levels on the persistence of secondary hyperparathyroidism (SHPT) and their relationships with vertebral fractures (VFx) in 125 renal allograft recipients transplanted 44 ± 23 months before. All patients underwent evaluation of the main biochemical parameters of calcium metabolism as well as vertebral and femoral bone density. In 87 patients, CaSR polymorphisms (A986S, R990G, and Q1011E) also were assessed. X-ray images of the lateral spine were obtained in 102 subjects to perform vertebral morphometry. High parathyroid hormone (PTH) and 25-hydroxyvitamin D lower than 80 nmol/L were found in 54% and 97% of patients, respectively, with 40% of these showing vitamin D levels lower than 30 nmol/L. VFx were detected in 57% of the subjects. After multiple adjustments, 25-hydroxyvitamin D, age, and hemodialysis duration, but not CaSR polymorphisms, were found to be significant predictors of high PTH, whereas age and time since transplant were positively related with lower 25-hydroxyvitamin D values. PTH and time since transplant were significantly associated with VFx. Patients with two or more VFx showed serum PTH levels 50% higher than patients without fractures. We therefore conclude that persistent SHPT is a very common feature after renal transplantation and that, unlike CaSR polymorphisms, low 25-hydroxyvitamin D is involved in its pathogenesis. High PTH levels, in turn, are associated with an increased VFx risk, which confirms the need for strategies aimed at lowering serum PTH in this setting as well. © 2010 American Society for Bone and Mineral Research.

Introduction

Renal transplantation is a well-established procedure that effectively corrects most of the abnormalities induced by renal insufficiency. However, because of several factors, the alterations in bone metabolism may persist after transplantation in a lot of cases.1 Indeed, a significant number of these patients present with reduced bone mineral density (BMD) and increased risk of fractures even long after transplantation.2–5 The negative effect of immunosuppressive therapy on bone6, 7 is considered one of the major causes of bone morbidity. Another important risk factor for bone morbidity is represented by secondary hyperparathyroidism (SHPT) that tends to persist in up to 50% of patients after successful transplantation.8–11 The restoration of normal renal function leads to a marked decrease in parathyroid hormone (PTH) levels soon after transplantation.9, 12 However, long-term elevation of PTH levels has been found even in transplanted patients with good renal function,11, 13 which indicates that the persistence of SHPT after transplantation is not exclusively dependent on the reduction in glomerular filtration rate. Although the mechanisms responsible for this phenomenon are not yet fully understood, several factors have been identified. They include the long-term persistence of parathyroid gland enlargement, the long period of time required for the involution of parathyroid gland hyperplasia, and an alteration in calcium set point.9 A number of genetic factors may play a role, too. The polymorphism of the vitamin D receptor (VDR), which seems responsible for a higher incidence of sporadic primary hyperparathyroidism and secondary hyperparathyroidism in patients with chronic renal insufficiency,9, 14 also can predict higher PTH levels in kidney transplant patients.9, 10

Calcium-sensing receptor (CaSR) polymorphisms have been demonstrated to be involved in the pathogenesis and severity of sporadic primary hyperparathyroidism,15, 16 and the reduction in CaSR expression in parathyroid adenomas may have a causative effect.17, 18CaSR expression has been found to be normal in kidney transplant patients with SHPT because of diffuse parathyroid hyperplasia, whereas its expression is downregulated in subjects with nodular hyperplastic glands, thus suggesting that the presence of nodular hyperplasia may be one of the most striking causes of persistent SHPT.19 However, it is currently not known whether different genotypes of this receptor may be related to increased prevalence of persistent SHPT in renal allograft recipients.

Low plasma levels of 25-hydroxyvitamin D [25(OH)D] have been reported to be very common after renal transplantation.20–25 The data available support the view that vitamin D insufficiency/deficiency may have an important effect on the persistence and severity of SHPT in this condition, thus indirectly contributing to skeletal morbidity even in these patients.20, 23–25 However, on the basis of statistical calculations, some authors tended to minimize the possibility of reverting SHPT through an increase in serum 25(OH)D concentrations.22

This study aims at evaluating the relationships among parathyroid hormone levels, CaSR polymorphisms, and vitamin D status and their role in bone mineral density and vertebral fracture risk in renal transplant recipients.

Materials and Methods

Patients

We selected 125 Caucasian subjects [body mass index (BMI) 24 ± 4 kg/m2, 87 males aged 50 ± 11 years, and 38 females aged 52 ± 11 years] from a larger cohort of patients who had undergone kidney transplantation (KTx) at our university hospital. Inclusion criteria were age 25 to 65 years, time since transplant 1 to 120 months, and serum creatinine ≤227 µmol/L. Patients were excluded if they had serum creatinine levels greater than 228 µmol/L, if they had a history of diabetes mellitus before transplantation, or if they had been treated with calcium supplements, vitamin D (including calcitriol), estrogens, or antiresorptive drugs after the graft. The causes of end-stage renal failure were chronic glomerulonephritis (45.6%), polycystic kidney disease (17.6%), unknown (12%), reflux nephropathy (9.6%), hypertensive nephropathy (8.8%), toxic nephropathy (3.2%), obstructive nephropathy (1.6%), and nephrolithiasis (1.6%). Of the patients, 88 had undergone hemodialysis (mean duration 33 ± 51 months), 28 had been on peritoneal dialysis (mean duration 8 ± 22 months), and 9 had undergone both before kidney transplantation. The subjects were on treatment with different combinations of oral immunosuppressive therapies. Forty-four patients were on a combination of cyclosporine A (CsA), mycophenolate mofetil, and methyl prednisolone; 26 patients were on a combination of tacrolimus, mycophenolate mofetil, and methyl prednisolone; and 9 patients were on a combination of CsA and methyl prednisolone. The remaining 46 subjects were being treated with several combinations of the above-mentioned drugs and/or rapamicine and azathioprine. The cumulative intake of the immunosuppressive drugs was calculated for each patient. The mean cumulative intakes were cyclosporine A (CsA) 171 ± 225 g, methyl prednisolone 8 ± 7 g, tacrolimus 3 ± 21 g, mycophenolate mofetil 881 ± 1039 mgr, and rapamicine 0.16 ± 0.75 g.

All patients gave their written fully informed consent, and the study was approved by the Institute's Ethical Committee. The study was carried out between February 1 and April 30, 2007.

Methods

Biochemical assay

Fasting blood samples were obtained from all patients. Total unadjusted serum calcium, phosphate, and creatinine were analyzed by Automatic Analyzer (Technicon Instruments Corp., Tarrytown, NY, USA). Bone-specific alkaline phosphatase (BAP) was determined by monoclonal immunoenzimatic assay (OCTEIA, Ostase BAP, IDS, Ltd., Bolden, Tyne & Wear, UK; EOS, Cervarese S. Croce, Padova, Italy). The reference range is 2.7 to 22.4 µg/L. Quantitative determination of PTH was obtained by using a direct two-site sandwich type chemiluminescent immunoassay (N-tactTM PTH DiaSorin S.p.A., Saluggia, Vercelli, Italy). The reference range for this method is 10 to 65 pg/mL. However, in accordance with Messa and colleagues,9 we considered PTH values above 80 pg/mL to be consistent with hyperparathyroidism in this specific population. Quantitative determination of 25(OH)D [LIAISON 25(OH)D total Assay, DiaSorin, Saluggia, Vercelli, Italy] was performed by a direct competitive chemiluminescence immunoassay. Patients with 25(OH)D serum levels of less than 30 nmol/L were defined as vitamin D–deficient. Patients with 25(OH)D serum levels between 31 and 80 nmol/L were defined as vitamin D–insufficient.

Genotyping

Genomic DNA from 87 subjects was extracted from peripheral white blood cells by using standard proteinase K-SDS digestion and phenol-chloroform extraction. DNA samples were kept at −20°C until used. A fragment of exon 7 of the CaR gene containing three polymorphisms, A986S, R990G, and Q1011E, was amplified by polymerase chain reaction (PCR) with the following primers: forward: 5′-TCCCGCAACACCATCGAGGA and reverse: 5′- TCTTCCTCAGAGGAAAGGAG. At least two different PCR amplifications from genomic DNA were sequenced on double-strand with sense and antisense primers. Reverse primers were biotinylated. PCR was performed in a final volume of 50 µL containing 1 µg of DNA, 50 mM KCl, 10 mM Tris HCl, pH 8.3, 1 mM MgCl2, 0.2 mM dNTP, 0.5 U Taq polymerase (Cetus, Emeryville, CA, USA), and 10 pmol of each primer. Annealing temperature was 54°C. PCR product then was subjected routinely to 10% and 12% polyacrylamide gels containing 5% glycerol and visualized by silver staining. Contamination problems were ruled out by including PCR control samples with no DNA as template. Extractions of DNA and pre-PCR reactions were performed in different rooms with respect to post-PCR reactions.

Bone densitometry

Bone mineral density (BMD) was measured by dual-energy X-ray absorptiometry (DXA, Hologic QDR 4500 A, Hologic, Inc., Waltham, MA, USA) at the lumbar spine (L2–4) and at the hip. The results were expressed as bone mineral density (BMD, g/cm2) and T-score (number of SDs to measure the difference between the patient's BMD value and the normal young adults' BMD level). According to World Health Organization (WHO) recommendations, osteoporosis is defined as a T-score value ≤ 2.5 SD.

Vertebral morphometry

Conventional lateral and anteroposterior thoracolumbar radiographs were obtained in 102 patients. In order to determine the type of deformities, a visual assessment of spinal radiographs was performed by one of us (SG). The vertebrae were identified and evaluated by using dedicated software for quantitative morphometry (MorphoXpress, P&G Pharmaceuticals, Egham, UK). The characteristics and performances of MorphoXpress are reported in detail in a recent paper.26 In brief, the MorphoXpress operates as follows: Original lateral vertebral radiographs are digitised using a TWAIN scanner (UMAX Power Look 1000, Techville, Dallas, TX, USA). The analysis then is initialized by the manual targeting of the centers of the upper and lower vertebrae to be examined. After that, the software automatically determines the positions of landmarks for a standard six-point morphometric measurement. If necessary, the software allows the operator to move these points before they are confirmed as correct. The positions of the confirmed points then are used by the software to calculate anterior, middle, and posterior vertebral heights, which also may be used for the determination of deformity shape. The fractures we analyzed in our study were defined as mild, moderate, or severe on the basis of a height-ratio decrease of 20% to 25%, 25% to 40%, and more than 40%, respectively, in accordance with the graduation of Genant.27

Statistical analysis

Statistical analyses were performed using SPSS Version 15.0 (SPSS Inc., Chicago IL, USA). The results were expressed as mean ± SD. Univariate and multiple linear regression analyses were used to evaluate the relationships between the variables considered. Multiple logistic regression was used when appropriate. General linear model (GLM) analysis was used to compare means by adjusting for covariates.

Results

The main parameters considered in our study are reported in Table 1. Since there were no significant differences between groups as to dialysis status, the patients were analyzed as a homogeneous group. Serum calcium levels were higher than normal in 13 patients (10.4%). Mean phosphate levels tended to be low, with 39 patients (31.2%) below the normal limit. PTH levels decreased after the transplant but remained high in more than 50% of subjects. 25(OH)D was normal in only 4 patients, whereas 97% of subjects showed vitamin D levels lower than 80 nmol/L (see Table 1). In addition, approximately 50% of the study population was in the range of vitamin D deficiency. BAP was in the normal range, with only 10 patients showing high levels.

Table 1. Clinical, Biochemical, and Bone Parameters in Renal Transplant Patients According to Dialysis Typea
 ALL (125)HD (88)PD (28)HD + PD (9)p*
  • a

    HD, hemodialysis; PD, peritoneal dialysis; HD + PD, both.

  • *

    HD versus PD.

Age (years)51 ± 1151 ± 1151 ± 1153 ± 10n.s.
Weight (kg)69 ± 1169 ± 1068 ± 1075 ± 17n.s.
Months since transplant44 ± 3350 ± 3428 ± 2527 ± 23.002
s-Creatinine (53–115 µmol/L)146 ± 32148 ± 31139 ± 31148 ± 36n.s.
s-Calcium (2.10–2.60 mmol/L)2.4 ± 0.22.4 ± 0.22.4 ± 0.22.4 ± 0.1n.s.
s-Phosphate (0.87–1.45 mmol/L)0.9 ± 0.20.9 ± 0.20.9 ± 0.30.8 ± 0.2n.s.
PTH (10–65 pg/mL)113 ± 83114 ± 88100 ± 67146 ± 69n.s.
PTH before renal transplantation (10–65 pg/mL)304 ± 316297 ± 293302 ± 391374 ± 242n.s.
25(OH)D (>80 nmol/L)40 ± 2241 ± 2435 ± 1738 ± 20n.s.
BAP (2.7–22.4 µg/L)11 ± 411 ± 412 ± 59 ± 3n.s.
PTH >65 pg/mL (%)68676472n.s.
PTH >80 pg/mL (%)54525062n.s.
Vitamin D insufficiency (%)57585460n.s.
Vitamin D deficiency (%)40374035n.s.
Lumbar spine T-score (SD)−1.4 ± 1.4−1.4 ± 1.5−1.5 ± 1.2−1.3 ± 1.9n.s.
Total hip T-score (SD)−1.3 ± 0.9−1.3 ± 0.9−1.3 ± 0.9−1.1 ± 1.2n.s.
Femoral neck T-score (SD)−1.7 ± 0.9−1.7 ± 0.9−1.6 ± 1.0−1.6 ± 1.1n.s.
Mean daily intake of CsA (mg)145 ± 179154 ± 192120 ± 148135 ± 136n.s.
Mean daily intake of Methyl prednisolone (mg)9 ± 138 ± 914 ± 2210 ± 5n.s.
Patients with vertebral fractures (%)57556057n.s.

The 87 patients who had undergone genotyping of CaSR did not differ from the whole study population with regard to age, sex distribution, duration of renal failure, type of dialysis (hemodialysis 73%, peritoneal dialysis 22%, both 5%), time since transplant, cumulative dosage of immunosuppressive drugs, PTH levels before and after transplant, serum creatinine level, and vitamin D level. The A986S polymorphism was observed most commonly (35%), whereas the other two CaSR polymorphisms, R990G and Q1011E, occurred in a minority of cases (11% and 1%, respectively). The allelic distribution of the three polymorphisms is reported in Table 2.

Table 2. PTH Levels (mean ± SD) by Genotypes and Haplotypes of Calcium-Sensing Receptor (CaSR) in 87 Renal Transplant Patients [General Linear Model with Renal Function, Months Since Transplant, PTH Before Renal Transplantation, Serum Calcium, Age, Hemodialysis Duration, Cumulative Dosage of Methyl Prednisolone, 25(OH)D as Covariates]
LocusbGenotypenaPTH (pg/mL)
A986SAA56124 ± 101
 AS29129 ± 59
 SS297
R990GRR77123 ± 90
 RG10144 ± 112
 GG0
Q1011EQQ86126 ± 91
 QE168
 EE0
AlleleHaplotypenaPTH (pg/mL)
  • a

    Number of patients.

  • b

    No evidence was found for Hardy-Weinberg disequilibrium at any of the three loci.

ARQX/X297 ± 45
 X/ARQ39129 ± 75
 ARQ/ARQ46126 ± 100
SRQX/X56127 ± 100
 X/SRQ29128 ± 70
 SRQ/SRQ297 ± 47
AGQX/X77123 ± 90
 X/AGQ10144 ± 110
AREX/X86126 ± 90
 X/ARE167

Lumbar spine bone density (T-score) was −1.4 ± 1.4 SD, with 26% of patients showing values of −2.5 SD or less. Total hip and femoral neck bone density values were −1.3 ± 0.9 SD and −1.7 ± 0.9 SD, respectively. Sixteen percent of the patients showed osteoporosis of the femoral neck. Fifty-eight patients (57%) had at least one fracture at vertebral morphometry. Two or more vertebral fractures were present in 32% of patients, and 17% showed three or more fractures. On the basis of the parameters of the Genant algorithm, we did not find any severe fractures, whereas 30% were moderate and 70% were mild.

Clinical and biochemical predictors of high PTH levels

Beside a univariate regression analysis, a multiple linear regression analysis model was constructed including PTH as a dependent variable and a number of predictive factors. Age, cumulative dosage of methyl prednisolone, and 25(OH)D level were considered significant predictors (Table 3). Approximately equivalent results were obtained when using the same variables in a multiple logistic regression in which PTH levels were subdivided in accordance with the upper normal limit of 65 pg/mL. Age [odds ratio (OR) 1.07, 95% confidence interval (CI) 1.01–1.14), cumulative dosage of methyl prednisolone (OR 1.18, 95% CI 1.01–1.40), and 25(OH)D level (OR 0.96, 95% CI 0.93–0.99) remained significant predictors. However, when the 80 pg/mL cutoff for the PTH upper normal limit was used, cumulative dosage of methyl prednisolone was no longer significant (Table 4). Even after adjusting for covariates (Fig. 1), patients with 25(OH)D ≤ 30 nmol/L showed mean PTH levels that were 30 pg/mL higher than those of subjects with 25(OH)D > 30 nmol/L.

Table 3. Clinical and Biochemical Predictors of PTH Levelsa
Variableβ Coefficientp
  • a

    Multiple linear regression analysis: r2 = 0.38, p < .001. Adjustment for renal function, hemodialysis duration, months since transplant, PTH level before renal transplantation, and serum calcium level.

Age0.233.036
Cumulative dosage of methyl prednisolone0.438.001
25(OH)D−0.302.003
Table 4. Clinical and Biochemical Variables Associated With PTH ≤ Versus >80 pg/mL (as Dependent Variable)a
VariableOdds ratio95% Confidence interval
  • a

    Multiple logistic regression. Adjustment for sex, renal function, months since transplant, PTH level before renal transplantation, serum calcium level, and cumulative dosage of methyl prednisolone.

Age1.101.03–1.18
25(OH)D0.960.93–0.99
Hemodialysis duration1.021.01–1.03
Figure 1.

PTH levels in renal transplant patients with 25(OH)D ≤ 30 versus > 30 nmol/L. Age, renal function, months since transplant, PTH level before renal transplantation, serum calcium level, cumulative dosage of methyl prednisolone, and hemodialysis duration were used as covariates in a general linear model (F = 4.80, p < .0001). ap < .05.

PTH levels did not differ as to calcium-sensing receptor polymorphisms when genotypes were considered or patients were grouped for different haplotypes, as analyzed by Scillitani and colleagues28 (see Table 2).

Predictors of vitamin D levels

In a multiple linear regression analysis model, considering serum vitamin D as a dependent variable, only age (β Coefficient −0.203, p = .016) and transplant duration (β Coefficient 0.369, p = .002) were significant predictors of vitamin D levels, after adjusting for hemodialysis duration, cumulative dosage of methyl prednisolone, cumulative dosage of CsA, and weight.

Predictors of bone turnover, bone density, and vertebral fractures

BAP serum levels increased with increasing PTH values and cumulative dosage of CsA and decreased with increasing cumulative dosage of methyl prednisolone (Table 5). None of the variables considered were associated with lumbar or femoral bone density.

Table 5. Clinical and Biochemical Variables Associated With Bone Alkaline Phosphatase (BAP) Serum Levelsa
Variableβ Coefficientp
  • a

    Multiple linear regression analysis: r2 = 0.23, p < .001. Adjustment for age, months since transplant, serum vitamin D level, and weight.

PTH0.36<.001
Cumulative dosage of cyclosporin A0.29.017
Cumulative dosage of methyl prednisolone−0.45.001

PTH levels and transplant duration were slight but significant risk factors for vertebral fractures, whereas cumulative dosages of CsA and mycophenolate mofetil were slightly protective (Table 6). By stratifying patients on the basis of the number of vertebral fractures, PTH levels were significantly more elevated in those with two or more fractures, being 50% higher even after multiple adjustments (Fig. 2).

Table 6. Clinical and Biochemical Variables Associated With Vertebral Fracturesa
VariableOdds ratio95% Confidence interval
  • a

    Multiple logistic regression analysis. Adjustment for age, sex, hemodialysis duration, PTH level before renal transplantation, cumulative dosage of methyl prednisolone, serum vitamin D, and lumbar and femoral bone density.

Months since transplant1.041.01–1.08
Cumulative dosage of cyclosporin A0.990.99–1.00
Cumulative dosage of mycophenolate mofetil0.990.99–1.00
PTH1.021.01–1.02
Figure 2.

PTH levels in renal transplant patients without vertebral fractures (VFx = 0), with one vertebral fracture (VFx = 1), and with two or more vertebral fractures (VFx = 2+). Age, renal function, months since transplant, PTH level before renal transplantation, serum calcium level, 25(OH)D level, and hemodialysis duration were used as covariates in a general linear model (F = 3.64, p = .032). ap < .05 versus VFx = 0.

Discussion

Persistent secondary hyperparathyroidism is very common in the majority of renal transplant patients8–11 in both the short and long term after surgery, even in those with good renal function.11, 13 Several causes have been identified, and among these, a major role is played by the persistence of parathyroid gland hyperplasia, especially when the nodular pattern is present.9 The expression of both vitamin D and calcium-sensing receptors has been found to remain substantially downregulated in kidney transplant patients with nodular hyperplastic glands,19 thus confirming their striking importance in the persistence of SHPT after the graft. Some reports also indicate that vitamin D receptor polymorphisms may be associated with higher PTH levels after renal transplantation.9, 10 However, it is currently not known whether different genotypes of CaSR may be linked with an increased prevalence of persistent SHPT in renal allograft recipients. CaSR polymorphisms have been reported to play a role in the pathogenesis and severity of primary hyperparathyroidism,15, 16 thus indicating that this genetic influence may be important in the development and maintenance of the parathyroid gland hyperfunction. In this study, the assessment of CaSR polymorphisms was obtained in two-thirds of a large population of kidney transplant patients and then analyzed on both a genotype and aplotype basis. However, we were not able to find any associations between CaSR polymorphisms and PTH levels, not even after correcting for the most important variables affecting PTH levels in this specific setting. Moreover, the different CaSR polymorphisms were not significant predictors either of bone density or of vertebral fractures in our study population. We therefore conclude that CaSR polymorphisms cannot play any central role as a risk factor for persistent hyperparathyroidism in these patients. However, we cannot exclude that other, more relevant conditions liable to affect parathyroid gland function in this setting may have overwhelmed a contributing role of CaSR polymorphisms.

In the last few years, several authors have reported low 25(OH)D levels in up to 70% to 95% of kidney transplant patients.20–25 In this study, only 4 subjects had normal serum vitamin D, 97% had values below 80 nmol/L, whereas 40% were in the range of clear 25(OH)D deficiency. Our data are perfectly in keeping with those obtained from studies on populations of kidney transplant patients living in northern countries, such as Denmark,25 the United Kingdom,24 the United States,23 Canada,22 and Germany.21 Considering the importance of sun exposure for vitamin D status, these results could appear rather surprising in view of the stronger effect of the sun at our latitude. However, it is well known that transplanted patients are strongly recommended to avoid sun exposure because of their increased risk of developing skin cancer.29 We did not specifically evaluate any parameters related to the amount of sun exposure. However, it has already been demonstrated that avoiding sun exposure is a very common practice in a large majority of renal transplant recipients, which mostly explains low vitamin D levels.25 In addition, an increased conversion of calcidiol to calcitriol owing to high PTH levels cannot be invoked as a cause of the low vitamin D levels detected in these patients because 1,25-dihydroxyvitamin D [1,25(OH)2D] values have been reported to be low normal in kidney transplant patients.25, 30

The proportion and severity of hypovitaminosis D found in our study seem even more remarkable than those observed in the general elderly Italian population.31 Vitamin D status does not exclusively affect skeletal metabolism, but it is probably involved in the genesis of cardiovascular, oncologic, metabolic, and immunologic diseases,32 which are fairly common among kidney transplant patients. The efforts to achieve vitamin D repletion in these subjects therefore should be at least as important as those currently made for the elderly population. As already suggested,33 for instance, vitamin D status might reduce the risk of rejection in kidney transplant patients.

Vitamin D deficiency also was found to be one of the major predictors of high PTH levels in our study. This is of importance because persistent secondary hyperparathyroidism is currently recognized as one of the main bone-damaging factors after kidney transplantation. Indeed, high PTH levels have been regarded as responsible for increased and unbalanced bone turnover after renal transplantation,13, 34, 35 thus leading to bone loss.10, 11, 13, 36 This study documents a relationship between the severity of persistent secondary hyperparathyroidism and the prevalence of vertebral fractures on a relatively large sample of patients included on the basis of homogeneous clinical and laboratory criteria. We also observed that patients with two or more fractures had PTH levels approximately 50% higher than subjects without fractures. The cross-sectional design of this study may limit the strength of this cause-effect relationship because some of these fractures may have occurred before the observation period. However, these data are in line with the increased risk of vertebral and nonvertebral fractures found in patients with primary hyperparathyroidism.37, 38 In addition, secondary persistent hyperparathyroidism has been found to be associated with increased bone turnover and with a progressive decline in bone density in longitudinal studies on renal allograft recipients as well.11, 36 A previous longitudinal observation on pancreas-kidney transplant patients also reported a link between high PTH levels and incident fractures39 Finally, this association may have been found infrequently in other studies on renal transplant patients owing to the relatively small number of patients in whom vertebral fractures were detected.3–5, 40 On the whole, persistently high PTH levels can be considered an important risk factor for fractures even in the long term after transplantation. Decreasing PTH levels therefore would be very beneficial in this condition as well. To this purpose, cinacalcet use would be a promising strategy even after transplantation because this treatment results in a substantial decrease in PTH and serum calcium and in an increase in bone density.41 However, calcimimetic drugs are indicated in case of hypercalcemia, which, in turn, can be present in a minority of these patients. Furthermore, these drugs cannot solve the problem of vitamin D insufficiency/deficiency, the correction of which may be helpful in reducing PTH levels, as evidenced by the data provided by this and other studies.21, 25 As far as this issue is concerned, it has been hypothesized that a full vitamin D replacement could not substantially decrease PTH levels in renal transplant allografts, given the multifactorial origin of the persistent secondary hyperparathyroidism in this setting.22 In contrast, a number of perspective studies found that cholecalciferol and calcium treatment may markedly reduce PTH levels after renal transplantation.42, 43 Even if PTH-lowering cholecalciferol effects need further evaluation, this therapy seems to be a promising option.

The prevalence of vertebral fractures reported in our study is higher than that of studies in which symptomatic fractures were confirmed by radiology,3, 39 but it is in keeping with the 32% to 50% observed by means of vertebral morphometry, even in the absence of any clinical signs.4, 5 Because of the cross-sectional design of the study, we cannot exclude that many of these events may have occurred several times before the observation period. However, a progressive rise in fracture risk was found in a 15-year follow-up study after renal transplantation.2 In their 4 year follow-up study on pancreas-kidney transplanted subjects, Smets and colleagues39 also demonstrated that all the fractures occurred more than 1 year after transplantation (range 1.3 to 4.0 years). Correspondingly, we found that time since transplant was a significant predictor of vertebral fracture risk. These data are not surprising because most risk factors for fractures tend to persist over the course of the clinical life of these patients.

Despite its bone-damaging effect, treatment with glucocorticoids is necessary for most patients in the long term after solid-organ transplantation as well. As remarked in preceding studies,2 no association was found between cumulative glucocorticoid intake and fractures in our analysis. However, the negative effects of these drugs on bone are well established.44 Accordingly, we observed that the cumulative dosage of methyl prednisolone was negatively associated with bone alkaline phosphatase, which indicates a negative effect of this drug on osteoblastic activity. A disproportionate increase in fracture risk for the relatively spared bone density status also may be considered as a possible expression of a glucorticoid-induced bone damage,3–5 even if we cannot exclude that the absence of correlation between bone density and fracture prevalence may be due to such factors as the cross-sectional design of the study and the alteration in bone quality. As far as immunosuppressive therapy is concerned, it is worth noting that CsA was associated with increased BAP and decreased vertebral fracture prevalence. Although our study was not specifically designed to address this issue, it is of interest that a bone-sparing effect of CsA has already been reported.45, 46 All these data suggest that the period of observation to detect bone morbidity should be extended in the long term after renal transplantation as well.

In conclusion, we confirm that persistent secondary hyperparathyroidism is fairly common in long-term renal transplant recipients, being substantially involved in the pathogenesis of vertebral fractures. While CaSR polymorphism does not affect PTH levels in this setting, a low vitamin D level is a well-established determinant of this parathyroid function disorder. Further studies are needed to explore the potential multilevel benefits of cholecalciferol supplementation.

Disclosures

All the authors state that they have no conflicts of interest.

Ancillary