• bone histomorphometry;
  • metabolic bone disease;
  • bone loss;
  • hierarchical cluster analysis


  1. Top of page
  2. Abstract

Thirty-three patients with cholestatic liver disease underwent histomorphometric assessment of paired bone biopsy specimens at time of orthotopic liver transplantation (OLT) and 4 months thereafter. At 4 months after OLT, bone metabolism improved, with bone formation increasing to normal and no change in bone resorption. Early post-transplant bone loss may be attributed to an additional insult to bone formation early after transplantation.

Introduction: Patients with advanced liver disease, especially chronic cholestasis, often have osteopenia, which worsens early after orthotopic liver transplantation (OLT) before starting to recover. The changes in bone metabolism leading to this rapid loss of bone after OLT, and to its recovery, are poorly defined.

Materials and Methods: In thirty-three patients with advanced chronic cholestatic liver disease, tetracycline-labeled bone biopsy specimens were analyzed prospectively at time of OLT and at 4 months after OLT, as part of a randomized trial to study the efficacy of calcitonin on post-transplant bone loss. Hierarchical cluster analysis of histomorphometric parameters was performed in an attempt to establish the functional grouping of individual histomorphometric parameters before and after OLT.

Results and Conclusions: Results showed that from the time of OLT to 4 months after OLT, bone mineral density of the lumbar spine and histomorphometric parameters of bone volume decreased, consistent with early post-transplant bone loss. Histomorphometric resorption parameters were increased before OLT, with no change after OLT. Histomorphometric formation parameters increased from low values before OLT to normal values at 4 months after OLT, with the exception of mean wall thickness values, which further decreased after OLT, suggesting an additional insult to bone formation during the study period. Histomorphometric changes after OLT were similar in female and male patients, pre- and postmenopausal women, and in patients treated and not treated with calcitonin. Hierarchical cluster analysis suggested that before OLT, bone resorption was functioning independently of bone formation, but that by 4 months after OLT, their coupled relationship had improved. Therefore, despite post-transplant bone loss, by 4 months after OLT, bone metabolism had improved, with increased bone formation and more coupled bone balance, as suggested by hierarchical cluster analysis.


  1. Top of page
  2. Abstract

OSTEOPENIA IS A MAJOR complication of advanced chronic liver disease, especially in patients with chronic cholestatic liver disease (CCLD).(1–4) After orthotopic liver transplantation (OLT), bone mineral density (BMD) of lumbar spine (BMD-LS) decreases further during the first 4 months, leading to post-transplant fractures in approximately 20–40% of CCLD patients(5–8); this has been assumed to be related, in some way, to skeletal effects of high-dose immunosuppressive medications. After this early period of bone loss and with continuing normal allograft function, patients begin to gain bone mass during the subsequent post-transplant years.

Measurements of BMD identify loss or gain of bone density after OLT but fail to show disturbances of bone resorption and formation leading to loss or gain of bone mass. Histomorphometric analysis of bone biopsy specimens provides this essential information. Although there have been conflicting histomorphometric data from patients with cholestatic osteopenia,(9–15) we have recently found that both increased resorption and decreased formation contribute to pretransplant bone loss.(16) The changes in bone metabolism that lead to the additional insult to bone mass after OLT, as well as its eventual recovery, are poorly understood. Calcitonin is a bone antiresorptive agent, which may be of benefit in preventing bone loss after OLT; its effects on bone resorption and formation after OLT have not been studied.

To study changes in bone metabolism after OLT and the effect of calcitonin, 33 patients with CCLD underwent histomorphometric analysis of paired iliac crest bone biopsy specimens, taken at the time of OLT and at 4 months post-transplant. Hierarchical cluster analysis of histomorphometric parameters was performed in an attempt to establish the functional grouping of individual histomorphometric parameters before and after OLT.


  1. Top of page
  2. Abstract

Patient population

Sixty-three consecutive adult patients who fulfilled the following criteria were enrolled in a randomized controlled trial to test the efficacy of salmon calcitonin therapy (100 iu subcutaneously each day for the first 6 postoperative months) to prevent post-transplant bone loss(17): (1) advanced primary biliary cirrhosis (PBC) or primary sclerosing cholangitis (PSC), activated for liver transplantation; (2) absence of diseases other than PBC or PSC that affect bone metabolism; (3) no medications affecting bone metabolism in the 12 months preceding transplantation (corticosteroids, hormone replacement therapy, anticonvulsants, bisphosphonates, sodium fluoride); (4) normal creatinine clearance and normal thyroid function; (5) willingness to participate in the study, including consent to bone biopsy at the time of liver transplantation; and (6) completion of liver transplantation between April 1990 and July 1995. The diagnoses of PBC and PSC were made according to well-established criteria.(18–20)

In the first week after successful liver transplantation, patients were randomized (1) to receive 100 MRC units of salmon calcitonin subcutaneously once daily for 6 months starting on the seventh postoperative day or (2) to receive no therapy. Randomization was by dynamic allocation, stratifying for gender (female versus male), diagnosis (PBC versus PSC), BMD level above or below the fracture threshold (0.98 g/cm2), and menopausal status. Patients underwent a protocolized immunosuppressive regimen with either triple therapy with prednisone, azathioprine, and cyclosporine, or dual therapy with prednisone and tacrolimus. Twenty-three patients were treated with triple therapy of cyclosporine, prednisone, and azathioprine. Cyclosporine was given to achieve the following trough levels: 250–350 ng/ml in the first week, 200–300 ng/ml from 2 weeks to 4 months, and 100–200 ng/ml from 4 to 12 months. Prednisone/solumedrol was given as follows: 1 g solumedrol at time of surgery; 100 mg BID on first 2 days, with taper to 15 mg BID by day 25; 15 mg BID on days 25–60; 20 mg/day from 61 days to 4 months; and 10 mg/day from 6 months onward. Azathioprine was given at 2 mg/kg/day. Ten patients received tacrolimus, with trough levels of 10–15 ng/ml for the first month and 5–10 ng/ml thereafter. Prednisone was given as follows: 1 g solumedrol at time of surgery; 25 mg QID on the first day, tapered to 15 mg/day by day 15; and 5 mg/day by 4 months. Standard therapy for acute cellular rejection was three intravenous doses of 1 g of methylprednisone. The study was approved by the Institutional Review Board of the Mayo Clinic. The BMD and fracture results of the randomized treatment trial with calcitonin have been previously reported(17) to show that calcitonin is ineffective in preventing post-transplant bone loss and fractures. As part of this randomized treatment trial, all patients were asked to consent to two bone biopsy specimens, the first at the time of OLT (n = 50) and the second at 4 months after OLT (n = 35); paired bone biopsy specimens were obtained in 33 patients, who form the study population for histomorphometric analysis. All patients had extensive clinical, biochemical, and radiologic examination before OLT, and at 4 and 12 months post-transplant.

Assessment of BMD and fractures

BMD was determined by DXA of the L1-L4 lumbar spine region, using a Hologic QDR 1000 densitometer (CV 2.2%) before OLT and at 4 and 12 months after OLT. Bone mass was corrected for bone size to calculate BMD (g/cm2). BMD measurements were compared with age- and sex-matched reference populations (Z-scores) and to young adult sex-matched reference populations at peak bone mass (T-scores). In patients with lumbar compression fractures, measurements were determined only on intact vertebrae. Large-volume paracentesis was performed, as necessary, for moderate/severe ascites before measurements of BMD to minimize the effects of ascites on BMD measurements.

Protocol-based chest X-rays and standard radiographs of the thoracolumbar spine at a tube distance of 120 cm were obtained to determine fractures before OLT and at 4 and 12 months after OLT. Additional radiographs were taken as clinically indicated at the sites of bone pain, and if X-rays were negative, bone scans were performed to evaluate possible fractures.

Tetracycline labeling

Study patients received tetracycline labeling before bone biopsy specimens to allow assessment of dynamic bone parameters. Labeling was done with cycles of oxytetracycline (course A) and demeclocycline (course B). At the time of enrollment, oxytetracycline, 250 mg QID, was given for 3 days, followed by 14 days off label, and then followed by 3 days of 250 mg QID. If more than 6 weeks had elapsed from the last dose of oxytetracycline and the patient had not yet undergone OLT, then course B was started (demeclocycline 150 mg QID for 3 days, 14 days off, then 150 mg QID for 3 days). Courses A and B were repeated at 6-week intervals until OLT. If a patient was not able to receive oral medication during initial labeling, doxycycline 100 mg was given intravenously every 12 h for 2 days. If a patient was called for OLT before the second label was administered (i.e., during the 14 days off-drug), doxycycline 100 mg was given intravenously every 6 h until the patient underwent OLT. Before the 4-month post-transplant bone biopsy, oxytetracycline and demeclocycline labeling was repeated as above, with bone biopsy specimens taken on days 23–27 of the labeling schedule. Because of uncertainty of timing of OLT and clinical problems related to liver disease, double tetracycline-labeling could be accurately done in 13 pretransplant and 23 post-transplant patients, resulting in 7 patients with paired dynamic data. All patients had assessment of all static histomorphometric parameters.

Bone biopsy specimens

The initial bone biopsy specimens were performed at time of OLT by the transplant surgeon, at the standard iliac crest bone biopsy site, using a 7.5-mm trephine.(21) Glucocorticoids were not administered until after the bone biopsy. Four-month bone biopsy specimens were done as an outpatient procedure from the contralateral iliac crest under local anesthetic. After the bone biopsy was taken, bone tissue was placed into 70% ethanol, dehydrated in 95% ethanol for 1 day, and dehydrated in 100% ethanol for 5 days, before immersion for 4 days in polymethyl methacrylate and embedding by controlled temperature polymerization. Four pairs of consecutive 5-mm sections were obtained at 100-mm intervals. Sections were stained with Goldner-Masson-Trichrome, Hematoxylin-Eosin, and Toluidine blue. Unstained sections were analyzed for fluorescent microscopy. Quantification of bone histomorphometric parameters was carried out by Bioquant System IV image analysis, using a Zeiss microscope and digitizing tablet (R and M Biometrics, Nashville, TN, USA). Bone biopsy specimens were read stepwise from corner to corner, and fields with more than 30% distortion were discarded. Primary and derived data were generated by the Bioquant IV software in accordance with standardized nomenclature and formulae,(22) for comparison with histomorphometric data for normal adult male and female references.

All bone biopsy specimens were read nonblinded by the three trained technicians in the Mayo Bone Histomorphometric Laboratory. To reduce intraobserver variation, a mean of four readings was used for all measured histomorphometric parameters. To reduce interobserver variation, the three technicians responsible for quantifying the bone biopsy specimens were required to measure bone histomorphometric parameters on a reference bone biopsy within 1 SD of the mean for these parameters each month to ascertain quality control. The study bone biopsy specimens were measured by the same three technicians as the normal control bone biopsy specimens, using the same methods. Bone histomorphometric parameters were expressed as Z-scores (sex-adjusted histomorphometric values), using normal female and male histomorphometric reference values of the Mayo Clinic Bone Histomorphometry Laboratory (see Table 1).

Table Table 1. Age and Sex Distribution of Reference and Study Populations for Bone Histomorphometry
Thumbnail image of

Static histomorphometric parameters

The following static parameters were analyzed.

1. Cancellous bone volume (BV/TV: measured as percentage of the total medullary bone volume from an unstained slide).

2. Cancellous bone architectural parameters: trabecular thickness (Tb.Th), trabecular number (Tb.N), and trabecular separation (Tb.Sp) are derived data from fractional cancellous bone volume and bone perimeter. Trabecular number and separation serve as indirect measures of bone resorption activity, because they reflect the effects of increased bone resorption activity on trabecular bone structure; patients with increased bone resorption activity have more widely separated, less numerous, and disconnected trabeculae.(23)

3. Osteoid parameters: osteoid thickness (O.Th) was measured by dividing each seam into four equal measurements in 50 fields or more and expressed as mean osteoid thickness of osteoid in micrometers. Osteoid volume (OV/BV) is expressed as percentage of total bone volume and osteoid surface (OS/BS) as percentage of cancellous surfaces covered with osteoid.

4. Number of osteoclasts per 100 mm of trabecular surface length (N.Oc): osteoclasts were identified as large amorphous cells that interface with a bone-resorbing surface within a resorption pit, display characteristically dense and somewhat granular cytoplasm (Goldner stain), and contain one or more irregularly shaped nuclei with prominent nucleoli.

5. Eroded surface (ES/BS): identified as a scalloped surface eroded to a depth of one lamella or more and expressed as a percentage of cancellous surface showing resorption cavities.

6. Osteoblast-osteoid interface (N.Ob): percentage of osteoid surface covered by osteoblasts, defined as cuboidal pyronine-staining cells.

7. Cortical thickness (Ct.Th): mean thickness of cortical seams in micrometers (average of 12 measurements).

8. Mean wall thickness (W.Th): mean thickness of the total bone structural unit in micrometers, measured as the distance between the cement line and quiescent, mineralized bone surface.

Dynamic histomorphometric parameters

The tetracycline double- and single-label lengths were measured using the unstained slide, which was scanned until 50 measurements were obtained. Mineral apposition rate (MAR) was calculated as the mean of four equally spaced interlabel thickness measurements (obtained from all available double labels on cancellous surfaces), divided by the time of the labeling periods (μm/day). The following dynamic parameters were derived using standardized formulae based on previously described variables.

1. Bone formation rate per unit bone surface (BFR/BS): amount of new bone mineralized per micrometer of cancellous bone surface area per day (mm3/mm2/year).

2. Bone formation rate per total bone volume (BFR/BV): amount of newly mineralized bone per total volume of cancellous bone (mm3/mm2/year).

3. Adjusted rate of bone apposition (Aj.AR): the product of mineralization rate and mineralizing surface divided by the osteoid surface (mm3/mm2/year).

4. Mineralization lag time (Mlt): the average lag time in days between apposition of osteoid and its mineralization.

5. Activation frequency (Ac.f): the activation frequency, or the rate at which new remodeling cycles are initiated, was calculated by determining the reciprocal of the total period (which is calculated by summing the duration of the formation period, quiescent period, and erosion period; days−1).

Histomorphometric adult female and male reference populations

Table 1 summarizes the age and sex distribution of the bone histomorphometric reference and study populations. The normal adult female and male bone histomorphometric reference parameters were established by analyzing iliac bone biopsy specimens from healthy volunteers with (1) no prior history of medical disease or drug therapy known to affect bone metabolism; (2) no evidence of vertebral fractures as assessed by lumbar and thoracic spine X-rays and no history of any hip or distal forearm (Colles') fractures; (3) lumbar spine BMD within the age- and sex-adjusted normal range; and (4) no laboratory abnormalities affecting bone metabolism. Normal adult reference and study bone biopsy specimens were analyzed by the same Mayo Clinic Bone Histomorphometry Laboratory technicians, using the same quantification and analysis procedures. Characteristics of the normal adult reference populations have been previously published.(16,24)

Statistical analysis

All parameters (biochemical, clinical, histomorphometric) were reported as mean ± SE. Raw bone histomorphometric values were converted to Z-scores by taking the difference between the mean study histomorphometric measurements from the mean values of sex-matched normals and dividing this result by the SD of the sex-matched normals. Paired t-tests of the biochemical and histomorphometric parameters were used to assess the within patient changes after transplantation, and independent t-tests were used to identify differences between female and male patients, PBC and PSC patients, and pre- and postmenopausal women. Associations between BMD and histomorphometric parameters were assessed using the Pearson correlation coefficient. Data analyses were performed using the SAS data analysis system (SAS Institute, Cary, NC, USA).(25)

Hierarchical cluster analysis is a well-established, multivariate statistical method(26,27) used to organize a large number of related parameters into a single system (a hierarchical tree or dendrogram). The positioning of individual parameters within the tree is such that each parameter is closest to the other parameters with which it shares the most functional similarity and furthest away from those most dissimilar. Individual histomorphometric parameters were first converted to Z-scores, and then cluster analyses were performed using the Splus statistical program (Insightful Corp., Seattle, WA, USA) to identify the degree of similarity among all the histomorphometric parameters within patients. These parameters were then organized into discrete functional groups (clusters) with maximum intracluster similarity. The degree of similarity among the different clusters was assessed, and the process proceeded sequentially until all variables were “clustered” and formed a single hierarchical tree (dendrogram), the branch length of the dendrogram corresponding to dissimilarity between/among clusters.


  1. Top of page
  2. Abstract

Demographics, BMD, and fracture assessment of total study population

Baseline demographic data of the study patients (11 PBC, 22 PSC; 12 males, 21 females; 6 postmenopausal) and post-transplant immunosuppression and hospitalization days are shown in Table 2. Pretransplant BMD of the lumbar spine (BMD-LS) was low, with a mean BMD-LS T-score of −2.1 ± 0.23. Thirty-nine percent (13) of the patients had osteoporosis (T-scores < −2.5), and 36% (12) had osteopenia (T-scores between −1 and −2.5). Six patients (18%) had fractures before OLT, one patient had a single rib fracture, two patients had vertebral fractures only (two thoracic and two lumbar), and three patients had both rib and vertebral fractures (five rib fractures, three lumbar, and four thoracic fractures).

Table Table 2. Clinical Data for 33 Patients With CCLD Before and After OLT
Thumbnail image of

Mean BMD T-score decreased to −2.5 ± 0.2 at 4 months post-transplant (p < 0.001) before partially recovering to −2.3 ± 0.2 at 1 year (p < 0.01). This is similar to the changes in absolute BMD values as shown in Fig. 1. Twelve patients (36%) sustained fractures during the first post-transplant year, 2 patients had rib fractures only (both multiple), 6 patients had vertebral fractures only (3 lumbar and 8 thoracic fractures), and 4 patients had both rib fractures (all multiple) and vertebral fractures (12 thoracic and 6 lumbar). One patient sustained a femoral neck fracture during the first year.

thumbnail image

Figure FIG. 1.. Changes in BMD of the lumbar spine after OLT in 33 patients with CCLD.

Download figure to PowerPoint

Bone histomorphometric parameters of total population

Cancellous bone volume was decreased at the time of OLT; cortical thickness was normal, but showed a significant decrease after OLT (Fig. 2). Both direct resorption parameters (eroded surface and osteoclasts per surface length) and indirect resorption parameters (trabecular number and separation) showed increased bone resorption before OLT, with no significant change after OLT (Table 3). The static bone formation parameters (number of osteoblasts and osteoid parameters) as well as the dynamic bone parameters (bone formation rates and adjusted appositional rate) increased significantly after OLT from low to normal values, with the exception of mean wall thickness, which further decreased after OLT. Activation frequency and mineralization rate also increased from low to normal values by 4 months post-transplant; mineralization lag time remained unchanged.

Table Table 3. Bone Histomorphometric Parameters Before and After Liver Transplantation in 33 Patients With CCLD
Thumbnail image of
thumbnail image

Figure FIG. 2.. Changes in bone histomorphometric parameters after OLT in 33 patients with CCLD.

Download figure to PowerPoint

There were no histomorphometric signs of osteomalacia before or after OLT. With the exception of mineralization lag time (p < 0.05), which increased in male patients and decreased in female patients after OLT, there were no significant differences in histomorphometric changes after OLT between patients with PBC and those with PSC, between female and male patients, and between pre- and postmenopausal women (data not shown).

Effect of calcitonin therapy on bone histomorphometric parameters

Fourteen patients (9 females, 5 males) were randomized to undergo calcitonin therapy, and 19 patients (12 females, 7 males) served as control patients. With regard to any identified pre- or post-transplant variables that could have influenced the effect of calcitonin, no significant differences were found between calcitonin-treated and control patients in any clinical, biochemical, radiological, or BMD parameter (Tables 1 and 4). Histomorphometric comparison of pretransplant bone biopsy specimens of the two study populations showed that there were no differences in pretransplant bone histomorphometric findings. There were no differences in histomorphometric parameters at 4 months post-transplant between patients treated and not treated with calcitonin, and histomorphometric changes after OLT were identical in calcitonin patients and controls (Table 4).

Table Table 4. Biochemical, BMD and Histomorphometric Parameters in Calcitonin-Treated Patients and Controls
Thumbnail image of

Hierarchical cluster analyses of bone histomorphometric parameters

The outcome of hierarchical cluster analyses using all histomorphometric parameters is shown in the dendrograms, in which the parameters having the most functional similarity are joined by the shortest branches into clusters; there is increasing dissimilarity between the clusters and between parameters as branches go to the right and join at the right side of the dendrogram the top of the dendrogram (Fig. 3A). Hierarchical cluster analysis of bone histomorphometric parameters at the time of OLT showed that the resorption parameters, osteoclast number, eroded surface, and trabecular separation organized into one cluster. This cluster had the greatest dissimilarity from the remaining bone metabolism markers. After OLT, the dendrogram (Fig. 3B) showed a different pattern of similarity of the various bone histomorphometric markers; bone resorption markers had more similarity with the remaining markers and were part of a more integrated system of bone histomorphometric parameters.

thumbnail image

Figure FIG. 3.. Dendrograms by hierarchical cluster analysis of bone histomorphometric variables at time of (A) OLT and (B) at 4 months post-transplant.

Download figure to PowerPoint

Correlations of BMD changes with histomorphometric changes

Bone loss, which occurred within the first 4 months after OLT, correlated with the decrease in trabecular thickness after OLT (r = 0.47, p < 0.01). The bone gain thereafter (4–12 months) correlated positively with the increase in osteoid markers from OLT to 4 months post-OLT (osteoid volume [r = 0.35, p = 0.05] and osteoid thickness [r = 0.53, p < 0.01]). No other correlations were observed between BMD and histomorphometric markers.


  1. Top of page
  2. Abstract

Histomorphometric analysis of paired iliac crest bone biopsy specimens in 33 cholestatic patients provides static and dynamic bone parameters by which to assess the changes in bone volume, resorption, and formation after OLT. The severity of disturbances in bone metabolism in our study population was reflected by low pretransplant BMD of the study patients, with 39% of patients meeting criteria for osteoporosis (T scores < −2.5) at the time of OLT. In addition, histomorphometric analysis of iliac crest bone biopsy specimens immediately pretransplant showed a profound negative bone balance with decreased bone formation (decreased osteoblast number, osteoid markers, and bone formation rates) and increased bone resorption (increased osteoclast number, eroded surface areas, and trabecular separation; decreased trabecular number). After OLT, BMD decreased, leading to lower BMD values at 4 months post-transplant, after which an increase of BMD occurred; this is consistent with previous publications describing changes in cholestatic patients after OLT.(1–8)

Despite the decrease in BMD at 4 months post-transplant, the iliac crest bone biopsy specimens taken at that time suggested histomorphometric improvement. All static and dynamic parameters of bone formation had significantly increased from below normal values at the time of OLT to values within the normal range by 4 months after OLT, with the exception of mean wall thickness. At the same time, 4-month measurements of direct (eroded surface, osteoclast number) and indirect (trabecular separation, trabecular number) parameters of bone resorption showed increased bone resorption, similar to the values at time of OLT. Activation frequency increased from low to normal values after OLT. Although activation frequency may directly reflect the increases in bone formation, it probably also indicates a return toward normalization of bone turnover after OLT. The histomorphometric findings of our study suggest that compensatory mechanisms resulting in increased bone formation are active by 4 months, although postoperative bone loss has not yet been reversed.

A few studies have investigated histomorphometric changes after OLT in smaller patient populations and with variable etiology of underlying chronic liver disease.(28–30) These studies also found evidence of significant increases in bone formation after OLT, and no significant change in bone resorption parameters. It has been difficult to know if the results of these previous studies could be extrapolated to a purely cholestatic population, which has never been extensively studied, but it would seem that the results of the previous studies are consistent with our findings in cholestatic patients.

Despite apparently improved bone formation parameters at 4 months after OLT, bone loss between OLT and 4 months was reflected by reductions in histomorphometric bone volume and densitometry measurements over this time period. Interestingly, mean wall thickness decreased significantly after OLT to even lower values at 4 months (mean Z-scores decreased from −1.9 to −2.5). The change in mean wall thickness between OLT and 4 months later reflects changes in mean thickness of completed bone remodeling periods preceding the 4-month bone biopsy. In normal bone, the bone formation period lasts for about 3–4 months(31); however, this time is considerably shortened by high-dose glucocorticoids to about 1–2 months.(32,33) It is possible that some bone formation periods have started before OLT and finished after OLT, leading to an underestimation of the reported change in mean wall thickness. Despite this, mean wall thickness decreased by >10% during this time period, indicating an additional insult to bone formation during the study period. This insult to bone formation and mean wall thickness values is consistent with the known effect of high dose glucocorticoids on bone formation and suggests the important negative effects of glucocorticoids after OLT. Because, by 4 months, bone formation had improved back to normal (as shown by analysis of the remaining bone formation parameters), the additional insult to bone formation had occurred early after OLT.

This additional insult to bone formation may be the key component of early post-transplant bone loss, although it has always been assumed that early post-transplant bone loss is related to increased bone turnover. Whether bone resorption, which is increased at time of OLT and at 4 months, further increases early after OLT is not evident from our study findings. Because no change in trabecular structure (trabecular number and separation) occurred during the study period, one can speculate that no clinically important increase in osteoclastic activity had occurred. In addition, previous histomorphometric studies(28–30) do not support a further increase in bone resorption by 4 months after OLT. Nor have biochemical parameters of bone metabolism provided a clear and consistent answer. However, parathyroid hormone (PTH), which increases bone resorption, has been shown to increase within the first month after OLT,(34) but within normal ranges. Biochemical resorption markers have been studied by Crosbie et al.(35) in 12 patients after OLT, who showed an increase of bone resorption indices (free pyridinoline and deoxypyridinoline cross-links) by 2 months. Interestingly, the bone formation indices (osteocalcin and procollagen type 1 carboxy propeptide) first decreased early after OLT and then started to progressively increase after 2–3 months post-transplant. As in this study, this suggests an additional insult to bone formation early after OLT. Despite uncertainties about a temporary further increase in bone resorption early after OLT, the ongoing high bone resorption from before OLT to 4 months after OLT undoubtedly also promoted bone loss by contributing to the negative bone remodeling balance.

The normalization of bone formation parameters by 4 months after OLT most likely reflects the early recovery of bone metabolism, leading to increased BMD by 1 year. In favor of this are the correlations in our study of the increased osteoid parameters at 4 months with the gain in BMD thereafter. Improvement in bone metabolism at 4 months post-transplant was also suggested by hierarchical cluster analysis. Bone histomorphometric analysis provides information on individual bone turnover parameters compared with normal, but does not provide any information on the functional groupings of these parameters. Hierarchical cluster analysis is a statistical technique extensively used in gene studies to organize genes into functional populations.(36,37) It is also used in clinical studies(38,39) to identify clusters of variables with highest similarity and can be applied on histomorphometric parameters. Pretransplant cluster analysis of histomorphometric parameters in our study population showed that bone resorption parameters were organized together into one cluster, functioning separately from the bone formation and mineralization parameters. This “uncoupling” of bone resorption and formation, which has not been previously demonstrated in cholestatic patients, may explain the unusual finding of increased resorption and decreased formation in these patients.(16) Interestingly, after OLT, hierarchical cluster analysis of the histomorphometric parameters of bone metabolism showed improvements toward a more integrated and “coupled” balance. Therefore, both bone formation and functional status (coupling) of bone metabolism markers seem to have improved after OLT. The change in position of mean wall thickness in the dendrograms is also of interest. At 4 months, all resorption and formation parameters exhibit greater similarity than before OLT, with the exception of mean wall thickness, which is isolated from all other histomorphometric parameters as completely dissimilar. Mean wall thickness at 4 months still reflects the early post-transplant period when bone formation was inhibited by factors integral to the post-transplant course. On the other hand, the other markers of bone resorption and formation represent more closely the present state of bone metabolism at 4 months after OLT, with improved bone formation and more coupling of bone resorption and formation. The dissimilarity between mean wall thickness (reflecting bone loss) and the other bone formation markers (reflecting improvement) provides further support of an overall improvement in bone metabolism by 4 months post-transplant. Moreover, it further illustrates the usefulness of hierarchical cluster analyses in detecting functional (dis)similarities among individual histomorphometric bone markers.

The use of calcitonin, an inhibitor of bone resorption, was thought to be potentially beneficial in preventing osteopenia in cholestatic patients who had increased bone resorption at time of OLT. However, analysis showed that calcitonin had no effect on either direct (osteoclast number, eroded surface areas), or indirect (trabecular thickness, number, separation) parameters of bone resorption; both treated and untreated patients had increased bone resorption before and after OLT, without any change after OLT. In addition, no other histomorphometric effects of calcitonin were noted, including no effects on bone volume, formation, or mineralization indices. These findings are consistent with the lack of efficacy of calcitonin on BMD and fractures in the post-transplant period.(17) The small patient numbers and short duration of therapy may have obscured a small effect of calcitonin. It is likely that the relatively weak action of calcitonin is overwhelmed by other factors operative early in the post-transplant course. Moreover, because our study suggests that an additional insult to bone formation may be a key component of early post-transplant bone loss, the mild antiresorptive effect of calcitonin may be of little benefit in this clinical setting. With the dual insult of decreased bone formation and persistently increased bone resorption after OLT, prevention of early post-transplant bone loss may require combination therapy of a potent antiresorptive agent such as a biphosphonate with a bone forming agent such as human recombinant PTH.

In conclusion, bone histomorphometric analysis of paired iliac crest bone biopsy specimens from 33 cholestatic patients showed that bone formation parameters significantly increased from low to normal values at 4 months post-transplant, whereas bone resorption remained persistently increased with no change from baseline. The increase in bone formation by 4 months post-transplant most likely reflects an early recovery of bone remodeling balance, which was also supported by hierarchical cluster analysis showing a more coupled and integrated bone turnover status at 4 months post-transplant. Although histomorphometric mechanisms of post-transplant bone loss are not fully elucidated by this study, the decrease in mean wall thickness after OLT suggests that an additional insult to bone formation early after OLT may the key component of post-transplant bone loss. The changes in bone metabolism after OLT were similar in all study populations, indicating no effect of gender, disease, or calcitonin treatment. Correlations of histomorphometric changes with clinical, biochemical, and immunosuppressive variables during this critical time period may identify the main causes of early post-transplant bone loss and more effective prevention strategies.


  1. Top of page
  2. Abstract
  • 1
    Hay JE, Lindor KD, Wiesner RH, Dickson ER, Krom RAF, LaRusso NF 1991 The metabolic bone disease of primary sclerosing cholangitis. Hepatology 14:257261.
  • 2
    Diamond TH, Stiel D, Lunzer M, McDowall D, Eckstein RP, Posen S 1989 Hepatic osteodystrophy. Static and dynamic bone histomorphometry and serum bone Gla-protein in 80 patients with chronic liver disease. Gastroenterology 96:213221.
  • 3
    Monegal A, Navasa M, Guanabens N, Peris P, Pons F, Martinez de Osaba MJ, Ordi J, Rimola A, Rodes J, Munoz-Gomez J 1997 Osteoporosis and bone mineral metabolism disorders in cirrhotic patients referred for orthotopic liver transplantation. Calcif Tissue Int 60:148154.
  • 4
    Giannini S, Nobile M, Ciuffreda M, Iemmolo RM, Dalle Carbonare L, Minucuci N, Casagrande F, Destro C, Gerundo GE, Sartori L, Crepaldi G 2000 Long-term persistence of low bone density in orthotopic liver transplantation. Osteoporos Int 11:417424.
  • 5
    Porayko MK, Wiesner RH, Hay JE, Krom RA, Dickson ER, Beaver S, Schwerman L 1991 Bone disease in liver transplant recipients: Incidence, timing, and risk factors. Transplant Proc 23:14621465.
  • 6
    Meys E, Fontanges E, Fourcade N, Thomasson A, Pouyet M, Delmas PD 1994 Bone loss after orthotopic liver transplantation. Am J Med 97:445450.
  • 7
    Haagsma EB, Thijn CJ, Post JG, Slooff MJ, Gips CH 1988 Bone disease after orthotopic liver transplantation. J Hepatol 6:94100.
  • 8
    Guichelaar MMJ, Hay JE, Egan K, Therneau T 2000 Incidence and pretransplant risk factors for posttransplant fractures in patients with chronic cholestatic liver disease. J Hepatol 31:49.
  • 9
    Cuthbert JA, Pak CY, Zerwekh JE, Glass KD, Combes B 1984 Bone disease in primary biliary cirrhosis: Increased bone resorption and turnover in the absence of osteoporosis or osteomalacia. Hepatology 4:18.
  • 10
    Mitchison HC, Malcolm AJ, Bassendine MF, James OF 1988 Metabolic bone disease in primary biliary cirrhosis at presentation. Gastroenterology 94:463470.
  • 11
    Hodgson SF, Dickson ER, Wahner HW, Johnson KA, Mann KG, Riggs BL 1985 Bone loss and reduced osteoblast function in primary biliary cirrhosis. Ann Intern Med 103:855860.
  • 12
    Stellon AJ, Webb A, Compson J, Williams R 1987 Low bone turnover state in primary biliary cirrhosis. Hepatology 7:137142.
  • 13
    Hodgson SF, Dickson ER, Eastell R, Eriksen EF, Bryant SC, Riggs BL 1993 Rates of cancellous bone remodeling and turnover in osteopenia associated with primary biliary cirrhosis. Bone 14:819827.
  • 14
    Guanabens N, Pares A, Marinoso L, Brancos MA, Piera C, Serrano S, Rivera F, Rodes J 1990 Factors influencing the development of metabolic bone disease in primary biliary cirrhosis. Am J Gastroenterol 85:13451362.
  • 15
    Diamond T, Stiel D, Lunzer M, McDowall D, Eckstein RP, Posen S 1989 Hepatic osteodystrophy; static and dynamic bone histomorphometry and serum bone gla-protein in 80 patients with chronic liver disease. Gastroenterology 96:213221.
  • 16
    Guichelaar MMJ, Malinchoc M, Sibonga J, Clarke BL, Hay JE 2002 Bone metabolism in advanced cholestatic liver disease: Analysis by bone histomorphometry. Hepatology 36:895903.
  • 17
    Hay JE, Malinchoc M, Dickson ER 2001 A controlled trial of calcitonin therapy for the prevention of post-liver transplantation atraumatic fractures in patients with primary biliary cirrhosis and primary sclerosing cholangitis. J Hepatol 34:292298.
  • 18
    Wiesner RH, LaRusso NF 1980 Clinicopathologic features of the syndrome of primary sclerosing cholangitis. Gastroenterology 79:200206.
  • 19
    Dickson ER, Fleming CR, Ludwig J 1979 Primary biliary cirrhosis. Prog Liver Dis 6:487502.
  • 20
    Dickson ER, LaRusso NF, Wiesner RH 1984 Primary sclerosing cholangitis. Hepatology 4:33S35S.
  • 21
    Hodgson SF, Johnson KA, Muhs JM, Lufkin EG, McCarthy JT 1986 Outpatient percutaneous biopsy of the iliac crest: Methods, morbidity, and patients acceptance. Mayo Clin Proc 61:2833.
  • 22
    Parfitt AM, Drezner MK, Glorieux FH, Kanis JA, Malluche H, Meunier PJ, Ott SM, Recker RR 1987 Bone histomorphometry: Standardization of nomenclature, symbols and units. Report of the ASBMR histomorphometry nomenclature committee. J Bone Miner Res 2:595609.
  • 23
    Parfitt AM, Mathews CHE, Villanueva AR, Kleerekoper M, Frame B, Rao DS 1983 Relationship between surface, volume and thickness of iliac trabecular bone in aging and in osteoporosis. Implications for the microanatomic and cellular mechanisms of bone loss. J Clin Invest 72:13961409.
  • 24
    Clarke BL, Ebeling PR, Jones JD, Wahner HW, O'Fallon WM, Riggs BL, Fitzpatrick LA 1996 Changes in quantitative bone histomorphometry in aging healthy men. J Clin Endocrinol Metab 81:22642270.
  • 25
    SAS Institute 1990 The FREQ procedure. In: SAS/STAT User's Guide, 4th ed., vol. 1. SAS Institute, Cary, NC, USA, pp. 852888.
  • 26
    Everitt BS, Landau S 1998 The use of multivariate statistical methods in psychiatry. Stat Meth Med Res 7:253277.
  • 27
    Massart DL, Kaufman L 1983 Hierarchical clustering methods. In: MassartDL, KaufmanL (eds.) The Interpretation of Analytical Chemical Data by the Use of Cluster Analysis. John Wiley & Son, New York, NY, USA, pp. 75101.
  • 28
    McDonald JA, Dunstan CR, Dilworth P, Sherbon K, Sheil AGR, Evans RA, McCaughan GW 1991 Bone loss after liver transplantation. Hepatology 14:613619.
  • 29
    Monegal A, Navasa M, Guanabens N, Peris P, Pons F, Martinez de Osaba MJ, Ordi J, Rimola A, Rodes J, Munoz-Gomez J 2001 Bone disease after liver transplantation: A long-term prospective study of changes, hormonal status and histomorphometric characteristics. Osteoporos Int 12:484492.
  • 30
    Vedi S, Greer S, Skingle SJ, Garrahan NJ, Ninkovic M, Alexander GA, Compston JE 1999 Mechanism of bone loss after liver transplantation: A histomorphometric analysis. J Bone Miner Res 14:281287.
  • 31
    Eriksen EF, Gundersen HJG, Melsen F, Mosekilde L 1984 Reconstruction of the formative site in iliac trabecular bone in 20 normal individuals employing a kinetic model for matrix and mineral apposition. Metab Bone Dis Relat Res 5:243252.
  • 32
    Dempster DW, Arlot MA, Meunier PJ 1983 Mean wall thickness and formation periods of trabecular bone packets in corticosteroid-induced osteoporosis. Calcif Tissue Int 35:410417.
  • 33
    Dalle Carbonare L, Arlto ME, Chavassieux PM, Roux JP, Portero N, Meunier PJ 2001 Comparison of trabecular bone structure and microarchitecture and remodeling in glucocorticoid-induced and postmenopausal osteoporosis. J Bone Miner Res 16:97103.
  • 34
    Compston JE, Greer S, Skingle SJ, Stirling DM, Price C, Friend PJ, Alexander G 1996 Early increase in plasma parathyroid hormone levels following liver transplantation. J Hepatol 25:715718.
  • 35
    Crosbie OM, Freaney R, McKenna MJ, Curry MP, Hegarty JE 1999 Predicting bone loss following orthotopic liver transplantation. Gut 44:430434.
  • 36
    Nielsen TO, West RB, Linn SC, Alter O, Knowling MA, O-Connell JX, Zhu S, Fero M, Sherlock G, Pollack JR, Brown PO, Botstein D, van de Rijn M 2002 Molecular characterisation of soft tissue tumours: A gene expression study. Lancet 359:13011307.
  • 37
    Rosenwald A, Wright G, Chan WC, Connors JM, Campo E, Fisher RI, Gascoyne RD, Muller-Hermelink HK, Smeland EB, Giltnane JM, Hurt EM, Zhao H, Averett L, Yang L, Wilson WH, Jaffe ES, Simon R, Klausner RD, Powell J, Duffey PL, Longo DL, Greiner TC, Weisenburger DD, Sanger WG, Dave BJ, Lynch JC, Vose J, Armitage JO, Montserrat E, Lopez-Guillermo A, Grogan TM, Miller TP, LeBlanc M, Ott G, Kvaloy S, Delabie J, Holte H, Krajci P, Stokke T, Staudt LM 2002 The use of molecular profiling to predict survival after chemotherapy for diffuse large-B-cell lymphoma. N Engl J Med 346:19371947.
  • 38
    Ozguler A, Guegen A, Leclerc A, Landre M-F, Piciotti M, Le Gall S, Morel-Fatio M, Boureau F 2002 Using the Dallas pain questionnaire to classify individuals with low back pain in a working population. Spine 27:17831789.
  • 39
    Ketelaars CAJ, Abu-Saad HH, Schlosser MAG, Mostert R, Wouters EFM 1997 Long-term outcome of pulmonary rehabilitation in patients with COPD. Chest 112:363369.