Presented in abstract form at the American Society of Bone and Mineral Research meeting, Cincinatti, Ohio, U.S.A., 1997, and at the National Osteoporosis Society Meeting, Bath, United Kingdom, 1998.
Cross-Sectional Analysis of Renal Transplantation Osteoporosis†
Version of Record online: 1 NOV 1999
Copyright © 1999 ASBMR
Journal of Bone and Mineral Research
Volume 14, Issue 11, pages 1943–1951, November 1999
How to Cite
Parker, C. R., Freemont, A. J., Blackwell, P. J., Grainge, M. J. and Hosking, D. J. (1999), Cross-Sectional Analysis of Renal Transplantation Osteoporosis. J Bone Miner Res, 14: 1943–1951. doi: 10.1359/jbmr.19184.108.40.2063
- Issue online: 2 DEC 2009
- Version of Record online: 1 NOV 1999
- Manuscript Accepted: 23 JUN 1999
- Manuscript Revised: 12 MAR 1999
- Manuscript Received: 2 DEC 1998
We report a cross-sectional study of 54 adult female renal transplant recipients. We measured bone mineral density (BMD) of the lumbar spine, femoral neck, total hip, and mid- and total radius, and 38 patients underwent transiliac crest bone biopsy. Osteopenia was widespread with 31/54 (57%) of patients osteoporotic at one or more sites. Seventeen out of 54 (32%) of the patients had a prevalent low-trauma fracture. There was a clear trend in BMD reduction across spine, hip and midradius, with the predominantly cortical midradial site showing the greatest loss. We found no relationship between BMD and body mass index, parathyroid hormone (PTH), dose of immunosuppressant, years since transplantation, age at menopause, or years since menopause. Histologically, abnormal biopsies could be classified into three categories: hyperparathyroid (n = 20), adynamic (n = 14), and osteomalacic (n = 2). Mean PTH was lower (p = NS) and mean cumulative prednisolone dose was higher (p = 0.04) in the adynamic group compared with the hyperparathyroid group, but because of overlap between groups neither was an effective discriminator of histology. We suggest that bone biopsy is indicated in these patients to direct appropriate treatment. At the cellular level, there were significant negative correlations between osteoclast function (eroded surface, r = 0.47, p = 0.003) and osteoblast numbers (osteoblast surface, r = –0.40, p = 0.01) and cumulative exposure to prednisolone. We postulate that suppression of osteoblast function by prednisolone with unopposed bone resorption may result in relative hypercalcaemia and low PTH. This progressive reduction in bone turnover may promote or prolong the adynamic state.
Renal transplantation is the treatment of choice in end-stage renal failure. Over recent years, modification of post-transplantation immunosuppressant regimens, in particular the introduction of cyclosporine, has prolonged the lifespan of transplanted organs.
Renal transplant recipients have many risk factors for osteoporosis prior to transplantation, including low dietary calcium intake, reduced exercise, heparin therapy, amenorrhea, low body weight, and premature menopause. Although transplantation may correct many of the biochemical imbalances associated with chronic renal failure, such as reduced renal function and secondary hyperparathyroidism, immunosuppressants, essential for preventing transplant rejection, offer a continuing insult to bone.
There is a growing body of evidence that patients have diminished bone mineral density (BMD) at transplantation,(1,2) although perhaps surprisingly this is not a universal finding.(3) Most studies address the 12–18 month post-transplantation period, and suggest that bone loss is especially rapid during the first few months post-transplantation, when doses of all immunosuppressants are maximal to combat rejection. Rates of bone loss vary considerably, but figures approximating 5% at the lumbar spine during the first 6 months following transplantation have been reported.(2,3) There is no consensus at present as to which risk factors are most strongly associated with reduced BMD, although prednisolone is a likely candidate.(2,4) Opinion remains divided as to the effect cyclosporine has on the skeleton,(5,6) but this agent is probably catabolic to bone.
The other crucial feature in renal transplant bone disease relates to the pathological processes underlying osteoporosis. Age-related changes may be superimposed on a background of secondary hyperparathyroidism, osteomalacia, adynamic bone disease, or mixed bone disease (hyperparathyroid bone disease and osteomalacia in combination). At present the only means of correctly identifying pathology in an individual remains bone biopsy.(7) Intuitively, any treatment intervention to preserve BMD in transplant patients should be directed at the underlying cause, thus identifying a noninvasive method of predicting histology for an individual is an attractive proposition.
This cross-sectional study was designed to establish the prevalence of osteoporosis and osteoporotic fractures in the Nottingham female renal transplant population, characterize any BMD reduction by qualitative and quantitative analysis of bone histology, and identify any determinants of low BMD and bone histology.
MATERIALS AND METHODS
We invited female renal transplant recipients with serum creatinine values <300 μmol/l who were under regular follow-up by, and living within a 20 mile radius of the Mid-Trent Renal Transplant Center (City Hospital, Nottingham, U.K.), to attend for assessment of osteoporotic risk. Fifty-four women (72% of those approached) attended for initial evaluation between February 1996 and September 1996.
Osteoporotic risk was assessed by questionnaire, biochemical profile, plain radiographs, and dual-energy X-ray absorptiometry (DXA) scanning at lumbar spine, femoral neck, and distal radius. To demonstrate that our selected population was representative of the local female renal transplant population, we surveyed patients excluded from the study on geographic grounds by postal questionnaire. The response rate was 61%, and we include descriptive data on 19 patients.
In our study population, we report complete scan data from 51 patients. Three patients did not attend for radial scans. With the exception of one Asian woman, all subjects were Caucasian. Twenty-nine women were postmenopausal, and of these six were on hormone replacement therapy (HRT) at the time of the study. Because of the effects of HRT on BMD and fracture risk, these women are considered separately in some analyses. Fifteen women were taking calcium and/or vitamin D supplements as prophylaxis against osteoporosis. None of the women had ever been treated with calcitriol or bisphosphonates.
The cause of end-stage renal failure was glomerulonephritis in 15 patients, reflux nephritis in 14, polycystic kidneys in 5, hypertensive nephropathy in 4, diabetic nephropathy in 3, obstructive uropathy in 2, tuberculosis in 1, and septicemia in 1. In 9 cases the cause was unknown. Fifty-one patients had been on long-term dialysis pretransplantation; 3 had been transplanted without dialysis. Seven patients were diabetic.
Informed consent was obtained from all patients and local ethical committee approval was granted.
Comparison group for BMD measurement
To establish whether there was a significant reduction in BMD in our study population, we compared absolute BMD (g/cm2) at the lumbar spine, femoral neck, total hip, and total radius with BMD measured at the same sites and on the same densitometer in a control group of 572 early postmenopausal women from the Nottingham area (median age 53 years, range 45–61 years), who made up the initial screening population for a multicenter cohort study (Early Postmenopausal Intervention Cohort, EPIC).(8) We controlled for age, years since menopause, and body mass index (BMI). We made a further comparison of BMD between the two populations after excluding those women who had ever used HRT, leaving residual populations of 343 in the control group and 48 in the study population.
All subjects had received immunosuppression post-transplantation with prednisolone, azathioprine, or cyclosporine, either alone or in combination. Cumulative doses were calculated from hospital and computerized records and for prednisolone included any pulsed doses of intravenous methylprednisolone given during transplant rejection episodes. For the purpose of this study, methylprednisolone was regarded as being equipotent with oral prednisolone. Mean dose of each drug was calculated as the cumulative drug dose divided by time elapsed since first transplantation and current drug dose as the cumulative drug dose over the 6 months preceding screening with all values expressed in grams.
Intact parathyroid hormone (PTH (1–84), 12–72 ng/l) was measured by a solid-phase, two-site chemiluminescent enzyme immunometric assay on a DPC-Immulite automated immunoassay analyzer (Diagnostics Product Corporation, Los Angeles, CA, U.S.A.). Routine biochemical measurements including serum creatinine, phosphate, and corrected serum calcium were measured using an Olympus AU 800 automated multichannel analyzer (Olympus Optical Co., U.K. Ltd., Hampshire, U.K.). Samples were taken in the nonfasting state. Biochemical markers of bone turnover were also measured but will not be discussed further in this paper.
Plain radiographs of the thoracolumbar spine (anteroposterior and lateral views) were analyzed for prevalent vertebral fractures. Vertebral fractures were classified as anterior, central, or posterior and graded as follows according to the percentage reduction in vertebral height: grade 1, 0–25%; grade 2, 25–50%; grade 3, 50–75%; grade 4, 75–100%.(9)
Plain pelvic radiographs (anteroposterior, to include both hip joints) were also performed and examined together with the spinal films to identify any artefactual abnormalities that could influence BMD as measured by DXA.
BMD measured in grams per square centimeter was determined using DXA with the Hologic QDR 2000 densitometer (Hologic, Waltham, MA, U.S.A.), for the lumbar spine (L1–L4 in anteroposterior direction), total hip, femoral neck, and total and midshaft radius. BMD was also expressed in standard deviation units as T scores (comparison with young adult mean) when defining osteoporosis with reference to WHO criteria,(10) and otherwise as Z scores (comparison with age-matched mean). Precision errors, established with a local normal population (female age range 45–59: spine and hip sites n = 410, radial sites n = 277) were as follows: L1–L4, 1.48%; total hip, 0.99%; femoral neck, 1.77%; midradius, 1.07%; ultradistal radius, 1.82%; and distal third radius, 1.17%. Spinal phantoms were scanned daily with a precision error of 1.5%.
Bilateral calcaneal ultrasound measurements were made using the Lunar Achilles Plus (Lunar Corp., Madison, WI, U.S.A.) for speed of sound (SOS in m/s), broadband ultrasound attenuation (BUA in dB/MHz), and stiffness. Stiffness (%) is a combination of BUA and SOS measurements and was expressed as T and Z scores (Lunar reference ranges) in the same way as detailed above for the BMD measurements. Daily phantoms were scanned with a precision error (stiffness) using manufacturer's data of 1.5%.
Patients were screened in two stages. The initial group of 26 patients (age range 45–61 years) was selected as an age-matched population with our control group. We performed bone biopsies on all those with low BMD (Z score <1.5 at spine or hip, n = 13 plus n = 2 patients with Z scores >1.5 but prevalent low-trauma fractures). Because all of these initial biopsies demonstrated significant abnormalities, we went on to biopsy all 23 patients subsequently screened.
Transiliac crest bone biopsies were taken according to standard technique(11) from the side contralateral to the functioning renal transplant. Thirty-eight women underwent biopsy. Apart from five samples all were bicortical.
Thirty-three patients received adequate double tetracycline labeling. Beginning ∼1 month prior to biopsy women received two time-spaced cyclical doses of oral demeclocycline hydrochloride (150 mg twice daily) following a 2 days on, 10 days off, 2 days on, 4 days free schedule.
Biopsy specimens were fixed in absolute alcohol and processed in the dark at 4°C in LR white resin monomer (London Resin Co., Berkshire, U.K.). Resin polymerization was carried out at 60°C under 2 atm of nitrogen in a pressure vessel. Seven-micrometer sections of undecalcified bone were stained with von Kossa's stain, toluidine blue, and solachrome azurine, and 15 μm unstained sections were cut for examination of tetracycline fluorescence in ultraviolet light. Sections underwent histomorphometric analysis using a VIDS II image analyzer.(12) Reference ranges based on a local (Manchester, U.K.) “normal” population were divided into groups according to decade of age. Results were expressed as absolute values and as Z score comparisons with age-matched means.
Bone volume (BV/TV, %), osteoid surface (OS, %), OS/BV; osteoid thickness (O.Th, μm); osteoid volume (OV, %); osteoblast surface (Ob.S, %), Ob.S/BS; mineralization front (%), mineralizing surfaces as a proportion of the osteoid surfaces, MS/OS[sLS + dLS]; eroded surfaces (ES, %), ES/BS; osteoclast surface (Oc.S, %), Oc.S/BS; osteoclast numbers (N.Oc, ×10−2/mm2), N.Oc/TVt; cortical thickness (Ct.Th, μm); trabecular thickness (Tb.Th, μm); cortical volume (CV, %), CV/TVc. Presence or absence of paratrabecular marrow fibrosis and aluminum staining were also noted.
Bone formation rate (BFR, %/year, volume referent), BRF/BVt [dLS + ½sLS].
PTH results were positively skewed and therefore natural log transformed for statistical analysis. Univariate linear regression analysis (Pearson correlation) was used to identify factors associated with both bone loss and histomorphometric variables and between biochemical parameters and histomorphometry. t-tests (two-tailed) were used to assess significance. ξ2-tests assessed differences in dichotomous variables at baseline. Multiple linear regression was used to compare our study population with the EPIC population. Multivariate analysis was used to assess the relative strength of the associations between cumulative prednisolone and azathioprine and histomorphometric parameters. Results are expressed as means (± SEM) with a p value of < 0.05 regarded as significant.
Table 1 contains descriptive characteristics, immunosuppressant dosages, and biochemical parameters of all 54 female renal transplant patients in whom BMD was assessed.
We compared our study population with female transplant recipients excluded from the study on geographic grounds (n = 19) and data for this group is also included in Table 1. We found no significant differences between the two populations in age, years since first transplant, age at menopause, years since menopause, or dialysis duration.
Only two biopsies were classified as normal. The remainder could be clearly segregated into three groups (hyperparathyroid, n = 20; osteomalacia, n = 2; adynamic n = 14, Table 2). Hyperparathyroid bone disease was defined in the presence of two out of three of the following: increased ES (>2 SD above the age- and gender-matched mean), elevated Oc.S (>2 SD above the age- and gender-matched mean) and increased N.Oc (>20 × 10−2/mm2). In 13 of the 20 hyperparathyroid patients, paratrabecular fibrosis was also present. In those with adynamic bone disease Ob.S, N.Oc, and O.Th were within or below the normal range of age- and gender-matched controls. In addition, a mineralization defect (mineralizing surface as a proportion of OS) and BFR >2 SD below the mean of age- and gender-matched controls were present in adynamic(13) and osteomalacic subjects, with the latter showing elevated OV.
Bone mineral density
Comparison of renal transplant recipients with a control population:
Results from multiple linear regression analysis comparing the BMD values for the study and control groups after controlling for age, years since menopause, and BMI are shown in Table 3. BMD values at lumbar spine, femoral neck, total hip, and total radius were significantly lower in the transplant group. When women who had ever used HRT were excluded from both groups, the reduction in BMD in the transplant population remained highly significant (p < 0.001 at all sites except lumbar spine p = 0.017).
Densitometry and ultrasound:
Mean BMD measurements (g/cm2) at the lumbar spine, femoral neck, total hip, and midradius for the original 54 patients, together with heel ultrasound (stiffness) results are given in Table 4. Results are also expressed as T scores.
The majority of patients had some degree of reduction of BMD irrespective of underlying histology. Thirty-one out of 54 (57%) were osteoporotic by World Health Organization criteria at one or more sites (Figs. 1A and 1B). There was a clear pattern of bone loss between sites, with the greatest reduction at the midradius, followed by femoral neck and total hip, then lumbar spine.
BMD measurements and histology:
There was a trend at all sites toward greater bone loss in the hyperparathyroid group compared with the adynamic group (Table 4). These differences reached significance at the femoral neck (p = 0.05) and total hip (p = 0.02).
We identified five instances of dystrophic calcification (three cases of vascular calcification affecting the lumbar spine and two of calcific periarthritis at the hip) with the potential to “falsely” elevate BMD as measured by DXA, rendering these sites less reliable indicators of BMD in affected individuals.
Paratrabecular fibrosis (marrow fibrosis) was graded histologically as grade 0 (absent), grade 1 (minimal), and grade 2 (moderate). None of our subjects exhibited grade 3 (severe) or grade 4 (replacement of bone by fibrous tissue) marrow fibrosis. It has been suggested that marrow fibrosis can cause an artefactual elevation of BMD,(14) but we observed no association with BMD in our patients.
Detectable aluminum was present in three biopsies and in all cases was found deep within bone rather than at actively remodeling sites. Two patients were classified as adynamic, one as hyperparathyroid.
There were significant associations between histomorphometric determinants of bone mass and both BMD and ultrasound parameters (Table 5). Total hip BMD (cortical and trabecular bone) correlated with both cortical and trabecular histomorphometry, while lumbar spine (trabecular bone) only correlated with trabecular markers. Midradius (cortical bone) correlated with Ct.Th, but also with Tb.Th. BUA and SOS correlated better with trabecular histomorphometry than cortical parameters.
We found no relationship between age-matched BMD or heel ultrasound (stiffness) and BMI, PTH, dose of immunosuppressant, years since transplantation, age at menopause, or years since menopause.
Seventeen subjects in the group as a whole (32%) had experienced a total of 25 low-trauma fractures (5 with more than one fracture) at the following sites: thoracic spine (n = 2), hip (n = 3), forearm (n = 8), tibia (n = 2), other (n = 10: metatarsal, metacarpal, rib, humerus). Nineteen of the fractures were sustained post-transplantation and six pre-transplantation. Fourteen of the patients with fractures had a bone biopsy. Individuals with spine, hip, and radial fractures did not always have a low BMD at the site of fracture but the fracture population as a group had a lower mean BMD at any measured site compared with those without a fracture (Table 6). The differences between means reached statistical significance at the midradius (p = 0.005) and for heel SOS (p = 0.02). Hyperparathyroid patients were more likely to fracture (9/20, 45%) than the adynamic patients (4/14, 29%, p = NS for between-group difference); one of the osteomalacic patients had a fracture while the remaining three fractures occurred in those who did not have a biopsy.
Determinants of histology
Serum PTH was higher in the hyperparathyroid subjects (mean 107.7 ± 23.0 ng/l) compared with the adynamic patients (63.2 ± 15.2 ng/l; Fig. 2), but this difference failed to reach statistical significance (p = 0.15) and there was considerable overlap in the lower range.
There was no correlation between PTH and Ob.S as an index of osteoblast number (r = 0.15, p = NS) nor between PTH and ES as an index of osteoclast function (r = 0.26, p = NS).
Serum calcium and phosphate
Mean serum calcium and phosphate levels were slightly higher in those with hyperparathyroid bone disease (calcium 2.51 ± 0.03 mmol/l, phosphate 0.92 ± 0.05 mmol/l) than those with the adynamic bone lesion (calcium 2.46 ± 0.05 mmol/l, phosphate 0.81 ± 0.05 mmol/l). The differences between means did not reach statistical significance.
Although it might be expected that a transplant recipient's daily prednisolone dosage would gradually decrease with years since transplantation, this was not true of our population. The cumulative dose continued to show a strong linear relationship with time (r = 0.91, p < 0.001) even in those patients who were transplanted more than 15 years previously (Fig. 3).
Because cumulative glucocorticoid exposure has been implicated in post-transplantation bone loss, we examined the relationship between cumulative prednisolone dose and histomorphometric cell surface parameters and found the strongest associations between Ob.S (r = –0.40, p = 0.01) and ES (r = –0.47, p = 0.003; Figs. 4A and 4B).
We also examined the relationship between these histologic parameters of bone turnover and current dose, mean dose, and pulsed dose of prednisolone, but found no association.
Ob.S and ES correlated less well with time since transplantation (r = –0.18, p = NS and r = –0.41, p = 0.01) than with cumulative steroid dose, thereby suggesting that the latter was the major determinant.
Adynamic patients as a group had received a significantly higher cumulative prednisolone dose compared with the hyperparathyroid patients (mean 43.6 ± 5.3 g and 23.6 ± 4.0 g, respectively, p = 0.04).
The correlations between cumulative azathioprine with Ob.S (r = –0.21, p = NS) and ES (r = –0.38, p = 0.02) were weaker than those for cumulative prednisolone. Cumulative azathioprine (doses of azathioprine < 2 g excluded) correlated well with cumulative prednisolone (r = 0.78, p < 0.001). Multivariate analysis suggested that the relationship between azathioprine and Ob.S and ES reflected the strong association between cumulative doses of prednisolone and azathioprine rather than an independent effect of azathioprine.
We found no association between cumulative dose of cyclosporine and any histologic parameter.
Although renal transplantation aims to restore normal biochemical balance, even when good renal function results, this is not necessarily accompanied by normalization of other variables such as PTH. Even when PTH returns to the normal range there may still exist abnormalities of parathyroid function as assessed by calcium infusion testing.(15) To minimize any effects in our study of ongoing renal impairment on mineral balance, we selected patients who, with only two exceptions, had serum creatinine values < 200 μmol/l (mean 129.0 ± 7.2 μmol/l).
There are currently ∼19 female and 38 male patients undergoing renal transplantation per annum at the City Hospital, Nottingham (average over last 3 years). We restricted our recruitment to the female renal transplant population in order to make a direct comparison between the prevalence of osteoporosis in the age 45–61 years subset and a population of nearly 600 “well” Nottingham women of the same age range known to us through the EPIC study.(8) The significant reduction in BMD at all sites in our transplant population emphatically confirms earlier work indicating that osteoporosis is a substantial problem in the renal transplant population. Of our population, 57% were osteoporotic at one or more scanned sites and 32% had already sustained low-trauma fractures. Ramsey-Goldman et al.(16) recently reported incident fracture rates in a U.S. cohort of female renal transplant recipients of 13% (all ages) and 45% (postmenopausal). These figures reflected rates 18 times (age 25–44 years) and 34 times (age 45–64 years) those of age- and gender-matched noninstitutionalized controls.
In the normal course of their post-transplant care, ranging in duration from 1–22 years, 12 of 17 of our fracture patients had received no prophylaxis against, or treatment for, bone loss either in the form of calcium and/or vitamin D supplementation or as HRT.
Our findings conflict with previous observations(1,17) because we show a clear stepwise decrease in BMD across sites as the proportion of cortical bone increases. These findings are supported by our observational data on fracture prevalence, which indicate a greater prevalence of radial than spinal fractures and are shared by Ramsey-Goldman et al.(16) who also documented a preponderance of limb fractures (n = 28) compared with axial fractures (n = 5). Because most of our patients, irrespective of underlying pathology, have a degree of elevation of PTH, this may represent the mechanism that underlies the preferential cortical bone loss.
We found significant associations between BMD and ultrasound parameters and several of the histologic indices of BMD, suggesting that scan measurements are valid surrogates for BMD in the renal transplant population. However, we failed to identify any variable that could be used to identify those with low BMD without densitometry.
Equally, although patients in the adynamic group have lower mean PTH values, and exposure to higher cumulative prednisolone doses than the hyperparathyroid group, because of the degree of overlap neither factor can be used to predict histologic subtype.
Histologically, our patients could be segregated into three subtypes, namely hyperparathyroid (n = 20), adynamic (n = 14), and osteomalacic (n = 2). Osteomalacia has previously been linked with aluminum deposition in dialysis patients,(18) but aluminum was absent in both our patients, with no other clinical features distinguishing them from other subjects.
Approximately one third of our patients were in the adynamic category, a similar proportion to that in the dialysis population where rates of 32%(19) and 49%(20) have been described. Published literature incorporating bone biopsy data is scarce in renal transplant patients. In contrast to our findings, Julian and colleagues(2) describe only hyperparathyroidism in their patients, although their population differed in several respects from ours (younger mean age, shorter pretransplantation dialysis period, and lower post-transplantation PTH level). In several small cross-sectional studies, other authors have found a preponderance of either adynamic bone disease(17) or mixed bone disease.(21)
Adynamic bone disease is a relatively recently recognized phenomenon and was first described in 1983.(22) Originally, it was found to be almost invariably due to aluminum deposition (two out of three patients in our study in whom aluminum was detected had adynamic bone disease) but more recently this view has altered as a form of adynamic bone disease is now known to occur with no associated aluminum intoxication.(13,23) Factors implicated in this nonaluminum adynamic state in dialysis patients include diabetes, increasing age, and CAPD.(23) In our patients, none of these factors seemed to be associated; the mean age of the adynamic group was slightly lower than the hyperparathyroid group (45 ± 2.97 years compared with 48.7 ± 2.39 years, p = NS) and only one patient with adynamic bone disease was diabetic (the remaining six diabetics belonged to the hyperparathyroid group). Although 20 of the biopsied patients had a history of CAPD treatment, only four of these were classified as adynamic. The two features that did seem to distinguish patients with the adynamic bone lesion from the hyperparathyroid group were a lower mean PTH (63.2 ± 15.2 ng/l compared with 107.7 ± 23.0 ng/l) and a higher mean cumulative prednisolone dose (43.6 ± 5.3 g compared with 23.6 ± 4.0 g), although there was considerable overlap. Although one might expect a higher PTH in the hyperparathyroid group by comparison with those with adynamic bone disease, it is harder to account for the discrepancy in cumulative steroid dose between the two groups. It is tempting to implicate cumulative glucocorticoid exposure in the pathogenesis of adynamic bone disease, but this causal relationship seems improbable because adynamic bone occurs with similar frequency in long-term dialysis patients who have had no exposure to significant doses of prednisolone.(20)
It has been suggested in the dialysis population that excessive PTH suppression results in nonaluminum-related adynamic bone. Several investigators have noted lower PTH levels in patients with adynamic bone disease compared with other forms of renal osteodystrophy.(19,24) It may also be that adynamic bone can spontaneously evolve into other forms of renal bone disease.(13) Fifty percent of children on dialysis with biopsy-proven hyperparathyroid bone disease treated with calcitriol went on to develop adynamic bone disease.(25) It follows that avoiding excess PTH suppression with calcium-binding agents may prevent this condition. Therapeutically, stimulating PTH may reverse the aplastic disorder. In one study, lowering dialysate calcium stimulated a 3-fold increase in PTH and restored bone formation to normal.(26) It would seem therefore that both excessive elevation and over-suppression of PTH is deleterious to bone.
At the cellular level, we find both osteoclast function and osteoblast numbers to be influenced by cumulative use of prednisolone. It is possible that in renal transplant recipients, suppression of osteoblast function through prolonged exposure to prednisolone with concurrent unopposed bone resorption may result in relative hypercalcemia and a decrease in PTH. This may either cause the adynamic state to persist in patients with this lesion at transplantation or promote it at some later interval.
In conclusion, we have shown that bone loss and associated low-trauma fractures in female renal transplant patients represent a considerable clinical problem. Studies examining therapeutic prevention and intervention directed at the underlying cause of low BMD are urgently required. Intuitively, because treatment should be influenced by the underlying histology for reasons of both safety and efficacy, and until such time as optimal therapy has been established, bone biopsy is indicated in all renal transplant patients with low BMD in whom treatment is being considered. To save time and to avoid additional sedation, this procedure could be performed at the time of transplantation itself.
Special thanks go to the Nottingham female renal transplant recipients who participated in the study. We thank Kay Dove, Sue Smith, Dane Blake, and Pat San for their help in coordinating study visits, Dr. Julia Fairburn (Department of Radiology, City Hospital, Nottingham) for assistance with X-ray analysis, and the Department of Bone Densitometry, City Hospital, Nottingham, for scan support.
- 11994 Changes in bone mass early after kidney transplantation J Bone Miner Res 9:1–9., , , , ,
- 21991 Rapid loss of vertebral mineral density after renal transplantation N Engl J Med 325:544–550., , , , ,
- 31994 Loss of regional bone mineral density in the first 12 months following renal transplantation Nephron 66:52–57., , ,
- 41995 Bone loss after kidney transplantation: A longitudinal study in 115 graft recipients Nephrol Dial Transplant 10:2096–2100., , , , , ,
- 51994 Missing impact of cyclosporine on osteoporosis in renal transplant recipients Transplant Proc 26:2652–2653., , , , ,
- 61996 Post-transplantation bone disease: The role of immunosuppressive agents and the skeleton J Bone Miner Res 11:1–7.
- 71994 The role of bone biopsy in the management of patients with renal osteodystrophy J Am Soc Nephrol 4:1631–1642.,
- 81998 Cigarette smoking, alcohol, and caffeine consumption, and bone mineral density in post-menopausal women: The Nottingham EPIC Study Group Osteoporos Int 8:355–363., , , ,
- 91989 Epidemiology of vertebral fractures in women Am J Epidemiol 129:1000–1011., , , , ,
- 10WHO Study Group 1994 Assessment of fracture risk and its application to screening for postmenopausal osteoporosis. Report no. 843. WHO, Geneva, Switzerland.
- 111998 Bone histomorphometry. In: StevensonJC, LindsayR (eds.) Osteoporosis. Chapman & Hall Medical, London, U.K., pp. 85–115.
- 121994 Age related histomorphometric changes in bone in normal British men and women J Clin Pathol 47:529–534., , ,
- 131994 Histological, radiological, and biochemical features of the adynamic bone lesion in continuous ambulatory peritoneal dialysis patients Am J Nephrol 14:19–29., , , , ,
- 141997 Renal osteodystrophy Bailliere's Clin Endocrinol Metab 11:165–194.
- 151991 Parathyroid function in normocalcemic renal transplant recipients: Evaluation by calcium infusion J Clin Endocrinol Metab 72:350–355., , , ,
- 161999 Increased risk of fracture in patients receiving solid organ transplants J Bone Miner Res 14:456–463., , , , , , , ,
- 171998 What are the bone lesions in patients with more than four years of a functioning renal transplant? Nephrol Dial Transplant 13(Suppl):103–104., , , ,
- 181978 Osteomalacic dialysis osteodystrophy: Evidence for a water-borne aetiological agent, probably aluminium Lancet 1:841–845., , , ,
- 191994 Adynamic bone disease with negative aluminium staining in pre-dialysis patients: Prevalence and evolution after maintenance dialysis Nephrol Dial Transplant 9:517–523., , , , , , , , ,
- 201996 The aplastic form of renal osteodystrophy Nephrol Dial Transplant 11:29–31., , ,
- 211996 Adynamic bone lesion in renal transplant recipients with normal renal function Nephrol Dial Transplant 11(Suppl):58–64., , ,
- 221983 Uremic osteodystrophy: classification, cause, and treatment. In: FrameB, PottsJT (eds.) Clinical Disorder of Bone and Mineral Metabolism. Excerpta Medica, Amsterdam, The Netherlands, pp. 254–259., , , ,
- 231993 Aplastic osteodystrophy without aluminum: The role of “suppressed” parathyroid function Kidney Int 44:860–866., , , , , , , , ,
- 241993 Correlation of bone histology with parathyroid hormone, vitamin D3, and radiology in end-stage renal disease Kidney Int 44:1071–1077., , , , , ,
- 251994 Development of adynamic bone in patients with secondary hyperparathyroidism after intermittent calcitriol therapy Kidney Int 46:1160–1166., , , , , ,
- 261993 Aplastic osteodystrophy: Effect of lowering dialysate calcium J Am Soc Nephrol 4:697., , ,