Abstract
Kidney transplantation is the renal replacement therapy of choice for patients with end stage renal disease. Advances in technology, surgical techniques and pharmacotherapy have improved renal allograft survival. Increasingly, we are seeing long term side effects related to renal transplantation, bone disease being a major one amongst them. Renal transplant patients have a higher risk of fragility fractures even when compared to those who remain on dialysis. This is likely to be related to pre-existing underlying bone disease and the emergence of new metabolic bone problems post-transplant. Conditions such as persistent hyperparathyroidism and the use of certain immunosuppressive agents have a deleterious effect on the post renal transplant bone.
Remarkable advances in the field of metabolic bone research have been made in the last decade and newer imaging techniques, biomarkers and therapeutic options are now available for osteoporosis in the general population. Interest is being focused on attempting to extrapolate these new discoveries to the management of bone disease post renal transplant. This review will briefly describe the metabolic bone changes that occur after transplantation and will provide an update on the currently available investigative options and therapeutic strategies for the management of post renal transplant bone disease.
Introduction
End stage renal disease (ESRD) is associated with multiple disorders of bone metabolism that are collectively referred to as chronic kidney disease-mineral and bone disorders (CKD-MBD). Abnormal bone and mineral metabolism thus often antedates transplantation and bone disease post-transplantation is in large part due to pre-existing bone damage acquired during renal insufficiency and dialysis. Renal transplantation is the treatment of choice for ESRD. The restoration of vitamin D synthesis, clearance of phosphate and reduction of parathyroid hormone (PTH) levels are all beneficial to the bone after transplantation. However, restoration of glomerular function does not always reverse bone disease. New bone disorders may emerge due to post-transplantation factors such as immunosuppressant use and persistent hyperparathyroidism. 1 The overall risk of fracture after renal transplantation is 4.6 (95% confidence interval (CI) 3.29–6.31) to 4.8 (95% CI 3.6–6.4)-fold higher than in healthy subjects,2,3 and hip fracture risk is 34% (aRR 1.34, 95% CI 1.12–1.61) higher during the first 3 years post-transplant compared to dialysis patients. 4
Changes in bone health after kidney transplantation
Bone health abnormalities in renal transplant patients may differ from that in other solid organ transplant recipients because risk factors that are unique and renally related, such as hyperparathyroidism and adynamic bone disease (ABD), exist in the former.
Osteoporosis
According to the World Health Organization, a T score of −1.0 to −2.5 on a dual energy absorptiometry (DXA) scan defines osteopenia and a T score ⩽ −2.5 defines osteoporosis. Though this has been expanded to include older men and other ethnicities, strictly speaking, this definition is only applicable to postmenopausal white women. Its applicability in younger men, pre-menopausal women and secondary osteoporosis remains uncertain. To date, there is still insufficient data to extrapolate this to the CKD and the renal transplant population. International Society of Clinical Densitometry guidelines recommend that in premenopausal women and men younger than 50 years, a Z score of −2.0 or lower be considered ‘below the expected range for age’ and a Z score above −2.0 as ‘within the expected range for age’. 5 Amidst these uncertainties, it may be more appropriate to classify patients with CKD and those post renal transplantation into those with ‘low bone density’ and those with ‘normal bone density’. Low bone mineral density (BMD) as defined above is common in the renal transplant population of Singapore with 45.6% of patients presenting to the renal bone disease clinic of a large tertiary public hospital having it at the time of tranplantation. 6 Tertiary hyperparathyroidism, age and post-menopausal status correlated with low BMD in this study. 6
Bone strength reflects the integration of two features: bone density and bone quality. Bone quality which is not measured by DXA scanning, is frequently affected in CKD and thus in CKD and renal transplantation, the National Institute of Health definition of osteoporosis as a skeletal disorder characterized by compromised bone strength predisposing to an increased risk of fracture may be of more relevance. In sharp contrast to the non-CKD patient with osteopenia/osteoporosis, in the CKD-MBD patient there is also the potential for low BMD to coexist with a range of functional abnormalities, from the high-turnover bone disease seen in uncontrolled hyperparathyroidism to the profound reduction of bone remodeling activity seen in ABD.
Almost all intervention trials on bone health post-transplant have used changes in BMD as an outcome measurement. Kidney Disease: Improving Global Outcomes (KDIGO) guidelines on post-transplant bone health suggest that patients who undergo kidney transplantation should undergo assessment of BMD using DXA within 3 months post-transplant. 7 This recommendation is supported by data that suggest that BMD is able to stratify fracture risk in renal transplant patients. 8 Studies also suggest that an initial sharp decline in BMD of up to 9% occurs during the initial 18 months post-transplant. 9 This has been corroborated in our population also. 6 BMD has been shown to improve progressively with time and approach normal values 10 years post transplantation. 10
Hyperparathyroidism
Persistent hyperparathyroidism has been associated with nephrocalcinosis, poorer graft function, extra-skeletal calcification,11,12 as well as an increased risk of fracture. 13 Elevated PTH stimulates osteoclast proliferation, increasing bone resorption and calcium release from the bone. It also increases 1-alpha hydroxylation of 25-hydroxyvitamin D (25(OH)D) which increases calcium absorption from the gut, increases re-absorption of calcium from the tubule and together with elevated FGF-23 level lead to phosphaturia, resulting in the clinical triad of raised PTH, hypercalcemia and hypophosphatemia. It has a reported incidence of 50% at 12 months post-transplant. 14 Serum PTH levels have been found to be an independent predictor of BMD at the time of transplant. 15 Some studies have also found an association between baseline PTH levels and post-transplant bone loss.9,16,17 This suggests that though successful renal transplantation theoretically should address the etiopathogenesis of bone disease arising from preceding renal failure, in reality, hyperparathyroidism, when too advanced, may result in consequences that persist well into the year after the transplant. 18 This also highlights the importance of trying to optimize the bone metabolic milieu even before transplantation.
Hypercalcemia is usually the main concern in persistent hyperparathyroidism with its prevalence shown to vary between 5–66%;14,19,20 52% and 15% of all patients post-transplant were found to be hypercalcemic at 3 and 24 months, respectively, after transplantation in one study. 14
Hypophosphatemia that may be due to elevated FGF-23 and PTH levels is common post-transplant. The prevalence of hypophosphatemia immediately post-transplant may be as high as 90% and has been shown to range between 21–39% in the first year post-transplant. The prevalence drops to 10% after 4 years post kidney transplant.19,21,22
Adynamic bone disease (ABD)
One of the main concerns of using anti-resorptive therapy in CKD and dialysis populations is the possible worsening of ABD. This may be an issue in the post renal transplant population also. Multiple studies suggest that low bone turnover disease is common and is a major problem post renal transplant.23,24 In a prospective review of 20 patients who had bone histomorphometry done prior and 6 month post-transplant, 60% had ABD. PTH and bone specific alkaline phosphatase was found to be significantly lower in the group with ABD compared to those without. Bone histomorphometry showed a decrease in osteoid and osteoblast surface and prolongation of mineralization lag time. 23 Another study of 20 patients who underwent bone biopsy at days 21–160 after transplantation concluded that impaired osteoblastogenesis and early osteoblast apoptosis may play important roles in the pathogenesis of post-transplant osteoporosis. 24 In Singapore, transplant patients have a much longer dialysis duration (96 months) than that mentioned in published literature and a significant number of our patients have secondary hyperparathyroidism (65.4%) and tertiary hyperparathyroidism (12.8%). 6 Thus the above data on the prevalence of ABD may be an over-estimate when pertaining to our local population. However, without currently available reliable methods to assess bone microarchitecture and bone turnover, iatrogenic causation of ABD with the use of anti-resorptive therapy remain a possibility.
Vascular calcification
Vascular calcification is common and appears to progress at a rapid rate in the dialysis population. It has been associated with increased cardiovascular and all-cause mortality. 25 Vascular calcification is a complex process. Deranged mineral bone metabolism and the uremic milieu seem to be a major contributor to its development in CKD. The understanding of the pathophysiology of vascular calcification has changed from one that is of passive calcification due to high calcium-phosphate product to one which is complex, involving active transformation of vascular smooth muscle cells into distinct ‘osteoblast-like’ cells with a secretary phenotype. A complex combination of factors such as chronic inflammation, oxidative stress, dyslipidemia, accumulation of advance glycation end-products, hyperglycemia and hyperphosphatemia, have been shown to trigger the process of vascular calcification. 26 The lack of calcification inhibitors such as matrix GLA protein, fetuin-A and pyrophosphate in the CKD milieu may further accelerate vascular calcification in these patients. 27
Coronary artery calcification (CAC) detected on computed tomography (CT) coronary angiogram was a strong predictor of cardiovascular mortality post renal transplant in a prospective study. 28 Studies on progression of CAC post kidney transplant have been contradictory. Previous preliminary reports suggest kidney transplant slows down the progression of vascular calcification. 29 A recently published review summarizing 13 published studies on vascular calcification (CAC and abdominal aorta) in kidney transplant recipients concluded that CAC slows down but does not halt after transplant. 30
Vitamin D deficiency
Serum 25(OH)D levels have been found to be significantly lower in renal transplant recipients compared with controls.31,32 This has been noted in our population also. 6 Secondary hyperparathyroidism due to vitamin D insufficiency has been associated with bone loss in the general osteoporotic population, 33 but this association is not clear in renal transplant recipients. A positive correlation between 25(OH)D levels and BMD was found in a Turkish, 34 but not in a white population of renal transplant patients. 32 No correlation between 25(OH)D levels and low baseline BMD or subsequent bone loss was found in the study conducted amongst Singaporean post renal transplant patients. 6 Ethnic differences on the effects of vitamin D deficiency on bone mass have to be thus considered. The complicated pathophysiology of CKD-MBD with altered bone turnover and hyperparathyroidism may overshadow the consequences of vitamin D insufficiency on bone mass. Vitamin D receptor (VDR) polymorphisms have also been recognized to influence PTH levels and BMD in kidney transplant recipients. Conflicting reports exist, with some suggesting that patients with the bb genotype of the VDR have decreased bone loss,35,36 and others identifying the BB genotype with lower PTH levels and protection from bone loss.36,37 The mechanism by which VDR gene polymorphisms affect parathyroid gland function is not yet clearly defined.
1,25-Dihydroxyvitamin D (1,25(OH)D) levels are not measured routinely and its status post renal transplantation is less well defined. A recently published prospective study shows a progressive increase in 1,25(OH)D level post kidney transplant and postulated that the rise was due to progressive decline in FGF-23. FGF-23 normally inhibits 1-alpha hydroxylation of 25(OH)D. The persistent high PTH then accelerates the conversion of 25-(OH) D to its active form. 38
Immunosuppressants and bone health
Glucocorticoids
Histomorphometric and several cross-sectional studies suggest a causal relationship between bone loss and glucocorticoid use in renal transplant recipients.9,39,40 The pathogenesis of steroid induced bone loss is multifactorial. The process of uncoupling of BMU seen in osteoporosis is magnified in transplant patients due to the effect of glucocorticoids. Glucocorticoids decrease the replication of cells of the osteoblast lineage and impair osteoblast differentiation and maturation. In mature osteoblasts, glucocorticoids are also pro-apoptotic. In osteoclasts, glucocorticoids increase the expression of RANK Ligand and decrease expression of its soluble decoy receptor, osteoprotegerin, leading to increase in osteoclast activity and at the same time a decrease in apoptosis of mature osteoclasts. Glucocorticoids also inhibit calcium reabsorption from the gastrointestinal tract as well as the renal tubules leading to PTH stimulation They also inhibit the release of gonadotropins, and as a result estrogen and testosterone production.
The finding of glucocorticoid-induced bone loss has not been replicated in all studies. Comparisons of patients who were treated with early steroid withdrawal with those who were maintained on low doses of steroids have also demonstrated only a modest benefit that seems to be restricted to the lumbar spine.41,42 Bone loss at 3 months post transplantation, has been found to be similar irrespective of whether patients were withdrawn from prednisolone early or maintained on it. In both groups, significant recovery in BMD was found to occur at 12 months after transplantation. 42
Calcineurin inhibitors
Cyclosporine has a controversial role in bone health. It had been shown in vitro to inhibit bone resorption. 43 However in in-vivo rodent models, it appears to have the opposite effect and causes severe bone loss. 44 In humans, cyclosporine in a steroid-free regimen does not appear to be associated with a decrease in BMD. 45 A recent study looking at the effects of cyclosporine on BMD in patients with nephrotic syndrome who were treated with prednisolone also came to a similar conclusion, 46 suggesting that, clinically, cyclosporine may not have a deleterious effect on bone health, despite animal models showing otherwise. There is minimal data available on the skeletal effects of Tacrolimus, another calcineurin inhibitor in renal transplant. A cross-sectional study in 72 kidney transplant patients showed that patients on higher tacrolimus dose had more bone loss compared with those on a lower dose. 47
Mammalian target of rapamycin inhibitors (mTORi)
Mammalian target of rapamycin inhibitors (mTORi) exhibit anti-proliferative and antiangiogenic activities. The two available mTORi are sirolimus and everolimus. Sirolimus has been shown to markedly decrease bone longitudinal growth rate in growing mice, 48 and may also cause a state of resistance to endogenous insulin like growth factor-1 (IGF-1). 49 IGF-1 action is needed for osteocyte stimulation and osteoblast survival and differentiation. Data from phase 2 clinical trials in renal transplant recipients receiving triple therapy with either cyclosporine or sirolimus suggest that sirolimus elevates bone turnover markers. Compared to the group treated with cyclosporine, the increase in bone turnover markers was less however. 50 A more recent study published showed osteoclast suppression with sirolimus both in-vitro and in-vivo. 51 This suggests that sirolimus may not be as detrimental to the bone as initially seen in mice. Everolimus has also been studied in mouse and human bone cells and was shown to inhibit osteoclast activity as well as osteoclast differentiation. In ovariectomized rat models, it has been shown to prevent up to 60% of cancellous bone loss. 52 In the BOLERO 2 study of breast cancer patients treated with everolimus, bone turnover was suppressed and the accelerated bone resorption that is seen in breast cancer with significant skeletal involvement and also with the use of aromatase inhibition was reversed with the use of everolimus. 53 There is however at present no clinical study exploring the bone effects of everolimus in renal transplant patients.
Bone loss has not been shown with the antimetabolites azathioprine and mycophenolate mofetil in rat models though human studies are lacking.54,55
Management of post renal transplant bone disease
Assessment of bone mineral density and quality
BMD estimation using DXA forms the cornerstone currently for assessment of postmenopausal osteoporosis. Due to the aforementioned inherent problems with DXA in the CKD and post-transplant population, KDIGO guidelines do not recommend it currently in patient with CKD stages 3–5. However, a KDIGO working group on CKD-MBD has recently concluded that a lower femoral neck BMD is associated with increased fracture risk in CKD. 56 This was based on data from two prospective studies.57,58 The earlier guidelines are being reviewed currently and an updated version incorporating the new data can be expected soon. KDIGO however currently recommends that patients with stage 1–3 CKD without biochemical evidence of CKD-MBD should be managed the same as the general population. Most transplant patients belong to this category and DXA may thus be used in this group of patients. DXA does have additional limitations. It is unable to provide information regarding bone quality or bone turnover and measurement accuracy may be compromised by extra-skeletal calcifications. If used, the results of DXA scanning should thus be interpreted together with clinical history, bone turnover markers and bone biopsy if available.
Bone histomorphometry with double tetracycline is the gold standard to assess bone quality (bone turnover, bone volume and mineralization). However, it is invasive and interpretation of the result requires expertise. It is not routinely done in the CKD population and data on renal transplant population are lacking. Efforts have been ongoing to find suitable alternatives to bone biopsy. The use of bone turnover markers (BTMs) is an attractive option. The combination of a low serum bone specific alkaline phosphatase (BSAP) and a low serum PTH is suggestive of low turnover and an elevated serum BSAP alone or in combination with an increased serum PTH appears to be highly sensitive and specific for high turnover in CKD and hemodialysis patients.59,60 The main limitation of using BTMs is that some of the markers such as osteocalcin, procollagen type 1 N terminal (P1NP) and collagen type 1 cross-linked C-telopeptide (CTX) are renally cleared and the lack of studies correlating them with turnover on histomorphometry in CKD.
One of the drawbacks of traditional DXA is that it cannot differentiate between cortical and trabecular bone. Newer imaging techniques such as quantitative CT (QCT) and peripheral quantitative CT (pQCT) allows three-dimensional imaging of the skeleton to provide volumetric bone mineral density thus allowing distinction between cortical and trabecular compartments. pQCT avoids large doses of radiation by focusing on tibia and distal radius. QCT measured at the spine has been correlated with trabecular bone volume from bone histomorphormetry. 61 A single center study of 72 dialysis patients showed that prevalent vertebral fractures were best predicted by lumbar spine cortical BMD. Every 1 mg increase of bone mineral content in cortical bone was associated with a 4% decrease of fracture risk. 62 A cross-sectional study in 52 hemodialysis patients using pQCT showed that decreases in distal radial cortical vBMD, cortical area, cortical thickness as well as decreases in torsional strength and bending strength were all significantly associated with odds of a fracture. 63 High resolution (HR) pQCT has a resolution of 100 μm compared with the 0.5 μm afforded by QCT. It has the added advantage of being able to evaluate cortical porosity. Cortical porosity and decrease in cortical thickness are associated with bone fragility. HR pQCT has been used to detect endocortical bone loss in renal transplant patients. 64 A less expensive and more readily available way to assess trabecular bone, the trabecular bone score (TBS), has been made available recently. It is a scoring system employing conventional DXA images. TBS is related to bone microarchitecture and provides information that is not captured from standard BMD measurements. 65 It has the potential to discern differences between DXA scans that show similar BMD measurements. 66 TBS has been shown to correlate with QCT and HR pQCT. 67 TBS is yet to be used in the transplant setting and whether it may add additional information remains unexplored. Despite these advances in imaging techniques, it has to be remembered that radiological assessment gives only a static assessment of bone microarchitecture and provides little information on bone turnover.
Prevention and treatment of bone loss post renal transplant
Preventive measures for osteoporosis in the general population also apply to transplant recipients. This includes smoking cessation, decreasing alcohol consumption, fall risk reduction and weight bearing exercises. No one treatment option will address all the problems in post-transplant bone disease. Treatment has to be individualized depending on patient profile, level of kidney function, pre-existing done disease, risk of fractures and risk of rejection.
Steroid sparing and steroid withdrawal
Steroid sparing and/or withdrawal protocols in renal transplant in an attempt to obviate the potential deleterious effects of steroids on bone health have attracted significant attention. However, as mentioned previously, the data regarding this modality is controversial. Though some beneficial effect on lumbar spine BMD has been reported,68,69 fracture data is limited to observational studies and this too has shown conflicting results. A US registry database of kidney transplant patients followed up over 3.9 years showed that steroid withdrawal was associated with a 31% fracture risk reduction. 68 However in another observational study that included kidney/liver and kidney/pancreas recipients, a high rate of fracture occurred despite minimal steroid exposure. 70
Vitamin D and its derivatives
Both native and activated vitamin D have been shown to improve BMD in transplant patients but fracture data are lacking. A Cochrane meta-analysis suggested that vitamin D supplementation in renal transplant patients improved BMD at the femoral neck with a trend towards higher BMD at the spine. 71 Low doses of vitamin D3 and calcium replacement for one year was shown to cause a reduction in lumbar spine, femoral neck, and femoral total bone loss in 52 renal transplant patients in a subsequent study. 72 A large randomized trial looking at the benefit of vitamin D supplementation post kidney transplant is ongoing. 73 Activated forms of vitamin D such as paracalcitol have also been shown to improve BMD in the transplant setting. The benefit is postulated to be related to the effective suppression of high turnover bone disease that characterizes secondary hyperparathyroidism. 74
Bisphosphonates
Bisphosphonates have been shown in multiple trials to improve BMD in the post-transplant setting. Bisphosphonates suppress osteoclast function and restore coupling of bone remodeling units that is perturbed in osteoporosis. A Cochrane meta-analysis in 2009 showed that bisphosphonates improve BMD at both lumbar spine and femoral neck without affecting serum creatinine level in renal transplant patients. 71 More trials have been published since, primarily looking at change in BMD. These have again reported positive outcomes.75–78 Fracture outcomes however remain elusive. A Cochrane summary of 6 trials was unable to demonstrate fracture reduction with the use of bisphosphonates in renal transplantation. 71
The lack of fracture reduction may be due to the trials being underpowered to detect clinical fractures. As mentioned previously, low bone turnover is prevalent in renal transplant patients. Hence, there is also a theoretical risk of worsening ABD with bisphosphonate therapy.
Antibody to receptor activator of nuclear factor KB ligand
Denosumab is a fully human monoclonal antibody to RANKL that inhibits osteoclast activity. It has been used successfully in treatment of osteoporosis. 79 Denosumab is hepatically cleared. This thus eliminates the concern of drug accumulation in renal insufficiency unlike as is the case with bisphosphonates. A re-analysis of the fracture reduction evaluation of denosumab in osteoporosis every 6 months (FREEDOM) study stratified by level of kidney function has shown that denosumab is both efficacious and safe in patients with up to stage 4 CKD. 80 However studies in dialysis patients have been limited due to severe hypocalcemic events reported with the use of denosumab.81,82 The possible explanation for hypocalcemia with denosumab may be akin to the phenomenon of hungry bone syndrome noted after parathyroidectomy. CKD patients are more dependent on hyperparathyroidism mediated bone turnover to maintain normal calcium level. With the use of denosumab, sudden shifts of calcium back into the bone can occur resulting in hypocalcemia. Trials evaluating the effect of denosumab on either BMD or fracture outcome in renal transplant patients are lacking.
Calcimimetics
Cinacalcet is currently the only commercially available calcimimetic. This calcium sensing receptor blocker successfully reduces PTH levels and has revolutionized treatment of secondary hyperparathyroidism. A small study in renal transplant patients has shown promising results in reducing serum calcium, PTH and improving serum phosphate. 83 A bone histomorphometry study in dialysis patients showed improved bone histology after 6–12 months of treatment with cinacalcet. 84
A more recent randomized control trial, designed to look at treatment of hypercalcemia in renal transplant patients with persistent hyperparathyroidism showed that 78.9% of patient treated with cinacalcet were able to achieve a serum calcium of <2.55 mmol/L compared with 3.5% of patients in the placebo group. The PTH level in the treatment arm was also significantly lower in the treatment group. Despite successful reduction in serum PTH and serum calcium, the secondary end point of change in BMD at the lumbar spine, femoral neck and distal 1/3 radius was not achieved. 80 Similarly, in the dialysis population, data from the EVOLVE study, suggest neither survival nor fracture reduction benefit in patients who were given cinacalcet.85,86 This has resulted in the European Renal Association-European Dialysis and Transplant Association (ERA-EDTA) recommending that cinacalcet not be used in CKD stage 5D if the specific aim is to improve survival. 87
Parathyroidectomy
Parathyroidectomy has been the treatment of choice for persistent hyperparathyroidism. In the transplant as well as in dialysis populations, parathyroidectomy improves bone mineral density as well as corrects the biochemical abnormalities prevalent in persistent hyperparathyroidism.88–91 There have been no studies on fracture reduction with parathyroidectomy in transplant patients. The concern of potentially worsening ABD with parathyroidectomy also exists. There also have been multiple case reports that suggest possible negative impact on renal allograft function post parathyroidectomy in the early post-transplant period.92–94
Teriparatide
Teriparatide is recombinant human PTH and is the first commercially available pure anabolic anti-osteoporosis agent. It has been shown to increase BMD and prevent fractures in postmenopausal women receiving glucocorticoids. 95 A small study on 26 post renal transplant patients treated with a 6 month course of teriparatide showed no benefit of using teriparatide compared to placebo. 96
Conclusion
There is no doubt that renal transplantation offers a second lease of life for patients with end stage renal disease. However, while overall long-term patient survival has improved, new worrisome health outcomes such as secondary osteoporosis and fragility fractures have emerged. Bone health post renal transplant is complicated. The progressive renal insufficiency pre-transplant, long years of dialysis, pre-existing renal bone disease, as well as potential co-existent traditional risk factors for osteoporosis all play a role in the pathogenesis of metabolic bone problems after transplantation. Trying to achieve optimal bone health while preserving the patient’s transplanted kidney is definitely a challenge. Close collaboration between osteoporosis experts and nephrologists is vital to achieve this aim. Most of the current treatment options in renal transplant patients have originated from the management of postmenopausal and senile osteoporosis. Areas of potential research include studies exploring the possible protective effect of VDR polymorphisms, the role of new biomarkers and bio-imaging modalities in the diagnosis and assessment of vascular calcification and bone loss and prospective studies on fracture risk prediction and reduction in renal transplantation.
Footnotes
Conflict of interest
The authors declare that there are no conflicts of interest.
Funding
This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.
