Calcified Tissue International

, Volume 83, Issue 3, pp 167–175

Short-Term Effects of High-Dose Zoledronic Acid Treatment on Bone Mineralization Density Distribution After Orthotopic Liver Transplantation

Authors

    • Ludwig Boltzmann Institute of Osteology at Hanusch Hospital of WGKK and AUVA Trauma Center Meidling, 4th Medical Department at Hanusch HospitalUKH Meidling
  • M. Bodingbauer
    • Division of Transplantation, Department of SurgeryMedical University of Vienna
  • P. Roschger
    • Ludwig Boltzmann Institute of Osteology at Hanusch Hospital of WGKK and AUVA Trauma Center Meidling, 4th Medical Department at Hanusch HospitalUKH Meidling
  • T. Wekerle
    • Division of Transplantation, Department of SurgeryMedical University of Vienna
  • B. Pakrah
    • Department of NeurosurgeryKA Rudolfstiftung
  • M. Haas
    • Department of Internal Medicine III, Division of NephrologyMedical University of Vienna
  • A. Kainz
    • Department of Internal Medicine III, Division of NephrologyMedical University of Vienna
    • KH Elisabethinen
  • R. Oberbauer
    • Department of Internal Medicine III, Division of NephrologyMedical University of Vienna
    • KH Elisabethinen
  • F. Mühlbacher
    • Division of Transplantation, Department of SurgeryMedical University of Vienna
  • K. Klaushofer
    • Ludwig Boltzmann Institute of Osteology at Hanusch Hospital of WGKK and AUVA Trauma Center Meidling, 4th Medical Department at Hanusch HospitalUKH Meidling
Article

DOI: 10.1007/s00223-008-9161-2

Cite this article as:
Misof, B.M., Bodingbauer, M., Roschger, P. et al. Calcif Tissue Int (2008) 83: 167. doi:10.1007/s00223-008-9161-2

Abstract

Patients with “hepatic” bone disease exhibit increased fracture incidence. The effects on bone material properties, their changes due to orthotopic liver transplantation (OLT), as well as zolendronate (ZOL) treatment have not yet been investigated. We studied bone mineralization density distribution (BMDD) in paired transiliacal biopsies (at and 6 months after OLT) from patients (control CON n = 18, treatment group ZOL n = 21, the latter treated with i.v. ZOL at doses of 4 mg/month) for how bone at the material level was affected by the “hepatic” disease in general, as well as by OLT and ZOL in particular. (1) BMDD parameters at baseline reflected disturbed bone matrix mineralization in “hepatic” bone disease combined with low turnover. Trabecular bone displayed a decrease in mean and most frequent calcium concentration (CaMEAN −2.9% and CaPEAK −2.8%, respectively; both P < 0.001), increased heterogeneity of mineralization (CaWIDTH +12.2%, P = 0.01), and increased percentage of bone areas with low mineralization (CaLOW +32.4%, P = 0.02) compared to normal; however, there were no differences compared to cortical bone. (2) Six months after OLT, ZOL-treated trabecular bone displayed reduced CaLOW (−32.0%, P = 0.047), cortical bone increased CaMEAN (+4.2%, P = 0.009), increased CaPEAK (+3.3%, P = 0.040), and decreased CaLOW (−55.7, P = 0.038) compared to CON and increased CaMEAN compared to baseline (+1.9, P = 0.032) without any signs of hyper- or defective mineralization. These changes as consequence of the antiresorptive action of ZOL visible already after 6 months result in beneficial effects on bone matrix mineralization, likely contributing to the significant decrease in fracture incidence observed in these patients 2 years post transplantation.

Keywords

Liver transplantationHigh-dose zoledronic acidBone mineralization density distributionQuantitative backscattered electron imaging

Patients with “hepatic” bone disease who are referred for orthotopic liver transplantation (OLT) are characterized by high prevalence to fractures [1] and reduced T scores [25]. Altered biochemical blood serum markers (e.g., low osteocalcin and vitamin D levels) have also been reported for these patients [1, 3, 6, 7]. Histomorphometric analyses of transiliacal biopsies from patients with “hepatic” bone disease gave evidence for reduced trabecular bone volume [4, 8, 9] and decreased bone turnover [8, 10, 11]. Information on possible alterations of the bone material itself, such as of the mineralization pattern, is completely lacking. However, all changes in the properties of bone caused by the liver transplantation and treatment have to be considered to occur on this background of “hepatic” bone disease.

The first period after transplantation is characterized by rapid bone loss [2, 7, 12] (up to 24% in the spine during the first few months after OLT [11]) and is caused by several factors, including immunosuppressive agents and immobilization before and after transplantation. Fractures occur very early and affect both patients with low or normal bone mineral density (BMD) [13]. Bisphosphonates are commonly used for the treatment of postmenopausal osteoporosis [14], but their antiresorptive effect, which is associated with decreased bone turnover and reduced fracture incidence, was also proven for the prevention and management of transplantation osteoporosis [13]. Alendronate and zoledronic acid (ZOL) were successful in preventing bone loss after OLT [5, 15], and in a recent clinical study of prophylactic high-dose ZOL therapy after OLT we could prove the efficacy of this bisphosphonate in preventing also fractures in liver recipients [16].

The aim of the present study was to gain a better understanding of the bone material properties in “hepatic” bone disease, their deviations from normal, and their possible changes caused by OLT and by preventive high-dose ZOL therapy. For this purpose, we measured the bone mineralization density distribution (BMDD) of paired transiliacal biopsies in a subgroup of patients from our previous clinical trial [16]. Bone matrix mineralization as characterized by BMDD can be principally obtained from quantitative microradiography [17], quantitative micro-computed tomography (CT) [18] or (as done in this work) quantitative backscattered electron imaging (qBEI) [19, 20]. BMDD is a key determinant of the stiffness and toughness of bone material [21, 22] and is consequently essential for the mechanical integrity of the bone. For healthy adult trabecular bone, BMDD has a specific shape, which is independent of several biological factors, such as gender, age, ethnicity, and skeletal site [23]. Specific deviations from normal BMDD occur in several bone diseases but also after therapies, which makes BMDD a sensitive tool for differential diagnosis or the evaluation of a therapy [20, 24]. The changes in BMDD after bisphosphonate treatment in postmenopausal osteoporosis have been studied intensively and are well characterized. Alendronate and risedronate made the bone mineralization more homogeneous and caused a slight shift to higher mineralization densities [17, 18, 25, 26]. Apart from a study of kidney transplantation patients treated with ZOL [27], information about the effects of bisphosphonates on the bone material after transplantation is rare. In the present study, for the first time, the transiliacal BMDD in “hepatic” bone disease before and after OLT, with and without ZOL treatment, was evaluated together with histomorphometry and serum biochemical markers of bone turnover of the same patients.

Materials and Methods

Patients

We studied transiliacal bone biopsies and serum biochemical markers of a subgroup of 39 patients from the previously published randomized controlled open label trial [16]: 21 were high-dose ZOL–treated (ZOL group, mean age 51 ± 8 years, 5 female/16 male) and 18 were in the control group (CON group, mean age 54 ± 6 years, 4 female/14 male). A detailed description of patients’ characteristics can be found in Bodingbauer et al. [16]. At baseline, age, gender, femoral and lumber spine T and Z scores, body mass index, Child-Turcotte-Pugh, model of end-stage liver disease index, and indications for OLT were similar for CON and ZOL patients. The ZOL group received ZOL (six infusions at 4 mg monthly after OLT), calcium carbonate (1,000 mg/day orally), and vitamin D3 (800 IE/day orally) during the first 12 months after OLT. The CON group received calcium and vitamin D at the same doses and duration as the ZOL group.

Immunosuppression

All patients received induction treatment with rabbit antithymocyte globulin (Thymoglobulin) and either cyclosporin A or tacrolimus as maintenance immunosuppression. All patients received corticosteroids, which were gradually tapered and discontinued 3 months after OLT in the majority of cases (for details, see [16]). The paired biopsies were obtained at (baseline) and 6 months after OLT. The study was approved by the local institutional review board, and all patients provided written informed consent.

qBEI

Full details, microscope settings, and the precision of the qBEI method have been published elsewhere [20, 24]. In this work, qBEI was used to measure the BMDD of trabecular and cortical areas separately from transiliacal bone biopsies. Prior to analysis, undecalcified iliac bone samples were embedded in polymethylmethacrylate. The surface of the planoparallel block samples was carbon-coated for qBEI analysis in a digital scanning electron microscope (DSM 962; Zeiss, Oberkochen, Germany). Trabecular and cortical bone tissue areas were recorded by images of the same size (2 × 2.5 mm) at ×50 nominal magnification (corresponding to a resolution of 4 μm per pixel). In principle, qBEI measures the intensity of the signal of the backscattered electrons, which is mainly dependent on the local calcium concentration of the sample. The heterogeneous bone mineralization pattern is reflected by the different gray levels of the BE images (bright areas correspond to relatively high, dark areas to low calcium concentrations; see Fig. 1). Gray-level histograms of these BE images were further transformed to BMDDs as described previously [20]. Four parameters characterized the BMDD [23]: CaMEAN, the weighted mean Ca concentration of the bone area; CaPEAK, the peak position of the histogram, which indicates the most frequent (typical) calcium concentration of the studied bone area; CaWIDTH, the full width at half-maximum of the distribution, describing the variation in mineralization density; and CaLOW, the percentage of bone areas undergoing primary mineralization revealing a calcium concentration less than the 5th percentile of the normal reference BMDD (<17.68 wt.% Ca). All qBEI measurements were done blinded for treatment.
https://static-content.springer.com/image/art%3A10.1007%2Fs00223-008-9161-2/MediaObjects/223_2008_9161_Fig1_HTML.gif
Fig. 1

(a) Backscattered electron image of the cross section of a transiliacal bone biopsy from a 53-year-old male patient at baseline containing cortical and trabecular compartments. Bright areas show mineralized bone matrix (the brighter, the higher is the local Ca concentration), whereas embedding material appears black in this image. (b) Corresponding trabecular and cortical BMDD from the biopsy. Normal reference BMDD and 95% confidence interval for healthy adult individuals [23] is indicated by the light gray area

Bone Histomorphometry

Two-dimensional histomorphometric analysis was done on one 3-μm-thick Goldner’s trichrome–stained section of each specimen (studying a bone area of about 30 mm2 of each sample). These sections were analyzed for structural parameters such as trabecular bone volume (BV/TV), trabecular thickness (Tb.Th), and trabecular number (Tb.N). Mean cortical bone volume per cortical tissue volume (CV/TV = 100% − cortical porosity) and cortical thickness (Ct.Th) were determined from both cortices. Additionally, static parameters of bone formation and resorption, such as osteoid thickness (O.Th), osteoid volume per bone volume (OV/BV), osteoid surface per bone surface (OS/BS), and eroded surface per bone surface (ES/BS), were analyzed. All parameters were evaluated according to the guidelines of the American Society for Bone and Mineral Research nomenclature committee [28]. All histomorphometric measurements were done blinded for treatment.

Biochemical Markers

Blood samples for serum parameters as well as biochemical markers of bone turnover were taken before OLT and 6 months after OLT. Osteocalcin (OC) and type I collagen peptides CrossLaps (CTX) were measured by electrochemiluminescence (Elecsys; Roche, Reinach, Switzerland). 25-Hydroxyvitamin D (25[OH]D) was measured by radioimmunoassay (DiaSorin, Saluggia, Italy). All tests were carried out according to the manufacturer’s instructions in a routine laboratory. Generally, all values obtained before OLT were considered as baseline data. Further data were used for analysis if they were measured during a time period of 6 ± 1 months after OLT. If a parameter was measured more than once for a patient during the time interval, the mean value was used for further data analysis.

Statistical Analysis

Statistical analysis was done using SigmaStat for Windows, version 2.03 (SPSS, Inc., Chicago, IL). Comparisons between BMDD parameters and normal healthy values, between trabecular and cortical BMDD, and between the treatment groups CON and ZOL at baseline and 6 months after OLT were done by t-tests or Mann–Whitney rank sum tests (if data were not normally distributed). For comparisons to baseline, paired t-tests or Wilcoxon signed rank tests (when appropriate) were used. Testing eventual differences between the three groups of liver recipients (A, B, and C according to the Child-Turcotte-Pugh classification) was done by one-way analysis of variance or Kruskal-Wallis one-way analysis of variance on ranks when appropriate. Non-normally distributed data are presented by median (interquartile range) in the tables. P < 0.05 was considered statistically significant.

Results

Patients at Baseline (Bone Matrix Mineralization in “Hepatic” Bone Disease)

Bone matrix mineralization was studied in the trabecular and cortical compartments of the iliac crest biopsy from liver recipients. An example of a biopsy at baseline (before OLT) from a male patient aged 53 years is shown in Fig. 1a. Generally, OLT patients (independent of Child-Turcotte-Pugh classification) revealed a shift of trabecular and cortical BMDD to lower mineralization densities together with a broadening of the peak compared to normal (Fig. 1b). For trabecular bone, comparison with normal BMDD values (previously published in [23]) resulted in CaMEAN −2.9% (P < 0.001), CaPEAK −2.8% (< 0.001), CaWIDTH +12.2% (P = 0.01), and CaLOW +32.4% (P = 0.02) (Table 1). The BMDD of the cortex was similar to that of trabecular bone for our patients with “hepatic” bone disease (Fig. 1b). Considering the classification of patients according to Child-Turcotte-Pugh, no difference could be found between the three classes A, B, and C for any of the BMDD parameters in trabecular or cortical bone.
Table 1

BMDD parameters of trabecular and cortical bone from patients with “hepatic” bone disease at baseline compared to normal reference data for trabecular bone from a previous study [23]

 

Trabecular bone

Cortex

Normal (n = 52) from [23]

Patients at baseline (n = 39)

Difference from normal (%)

P

Patients at baseline (n = 39)

CaMEAN (wt.%)

22.20 (0.45)

21.55 (0.80)

−2.9

<0.001

21.76 (21.22–22.28)

CaPEAK (wt.%)

22.96 (22.70–23.14)

22.30 (0.70)

−2.8

<0.001

22.36 (22.01–22.88)

CaWIDTH (wt.%)

3.29 (3.12–3.47)

3.76 (0.71)

+12.2

0.01

3.64 (3.29–4.29)

CaLOW (%)

4.52 (3.87–5.79)

6.52 (3.79)

+32.4

0.02

5.24 (3.27–7.08)

Data shown are mean (SD) or median (interquartile range). P values show difference from reference values of trabecular BMDD parameters

Static histomorphometry, although showing a large variation, revealed clear evidence of decreased bone formation. As shown in Table 2, median values of static parameters of bone formation OV/BV and OS/BS were reduced compared to the normal range (deviations >1 SD from normal [29]). Mean values for the structural parameters BV/TV, Tb.Th, Tb.N, O.Th, and ES/BS were within the normal range. In contrast, mean Ct.Th and CV/TV were both decreased for CON and in the lowest normal range for ZOL at baseline.
Table 2

Histomorphometric parameters at baseline in comparison to normal and changes after OLT and treatment

 

Trabecular bone

Baseline

6 months after OLT

Normal rangea

CON (n = 18)

ZOL (n = 21)

CON vs. ZOL

CON (n = 18)

ZOL (n = 21)

CON vs. ZOL

BV/TV (%)

20.0 (6.8)

20.3 (8.8)

ns

23.2 (9.0)

19.7 (5.0)

ns

20.6 (5.2)

Tb.Th (μm)

125 (35)

122 (27)

ns

133 (30)

125 (25)

ns

137 (27)

Tb.N (mm−1)

1.61 (0.39)

1.66 (0.61)

ns

1.75 (0.58)

1.58 (0.29)

ns

1.6 (0.4)

OV/BV (%)

0.3 (0.4–0.6)

1.2 (0.5–1.4)

ns

1.1b (0.5–7.0)

0.2 (0.1–1.1)

P = 0.021

3.0 (1.6)

O.Th (μm)

10.3 (4.9–12.3)

10.0 (5.7)

ns

10.2 (6.7–15.7)

7.4 (4.2)

ns

8.7 (2.0)

OS/BS (%)

1.8 (0.3–6.2)

8.9 (5.1–14.0)

ns

9.4b (5.0–28.6)

3.6 (1.4–10.1)

P = 0.047

17.1 (6.1)

ES/BS (%)

2.8 (2.2)

2.9 (1.7–4.6)

ns

2.9 (2.3)

0.9 (0.3–4.4)

ns

3.6 (1.0)

 

Cortical bone

Baseline

6 months after OLT

Normal rangea

CON (n = 7c)

ZOL (n = 7)

CON vs. ZOL

CON (n = 7)

ZOL (n = 7)

CON vs. ZOL

Ct.Th. (mm)

0.737 (0.275)

1.006 (0.467)

ns

0.950 (0.608)

1.132 (0.457)

ns

1.303 (0.348)

CV/TV (%)

90.1 (4.8)

93.5 (2.1)

ns

90.5 (2.6)

90.7 (5.3)

ns

94.4 (1.0)

Data are mean (SD) or median (interquartile range)

aNormal range values are data from healthy men aged 51–60 years from Tovey and Stamp [29]

bSignificantly different from baseline (paired t-test or Wilcoxon signed rank test)

cFor the evaluation of mean Ct.Th and CV/TV, only the biopsies containing both cortices before and after OLT were used

Biochemical markers of bone turnover at baseline are summarized in Table 3. The bone formation marker OC was in the lower normal range, and the bone resorption marker CTX was slightly elevated. Patients before OLT had inadequate serum levels of 25(OH)D.
Table 3

Serum levels of biochemical markers of bone turnover at baseline and 6 months after OLT for controls and ZOL-treated patients

 

Baseline

6 months after OLT

Normal range

CON

ZOL

CON vs. ZOL

CON

ZOL

CON vs. ZOL

OC (ng/mL)

15.00 (11.73–27.31) n = 15

14.70 (11.70–29.00) n = 14

ns

87.98** (53.80–117.65) n = 12

25.23* (19.50–43.20) n = 14

<0.001

14–34

CTX (ng/mL)

5.08 (3.36) n = 15

4.47 (2.76) n = 15

ns

3.10 (1.68–9.03) n = 9

0.20* (0.10–2.23) n = 13

0.01

0.9–3.6

25(OH)D (pg/mL)

23.75 (10.96) n = 15

29.48 (17.26) n = 15

ns

78.99*** (28.56) n = 12

79.04** (23.50) n = 14

ns

22–92

Data shown are mean (SD) or median (interquartile range). n indicates the number of patients within the group (note that the number of patients studied 6 months after OLT is different from the number at baseline for most parameters). Comparison between CON and ZOL at baseline or 6 months after OLT is indicated by the corresponding P value. Intraindividual changes are indicated by * < 0.05, ** < 0.01, and *** P < 0.001 for the comparison to baseline (paired t-tests or Wilcoxon signed rank test, corresponding mean or median values not shown)

Effect of OLT, Ca/Vitamin D Supplementation, and ZOL

A significant increase of the BMDD parameter cortical CaMEAN 6 months after OLT compared to baseline was observed for ZOL-treated patients (+1.9%, P = 0.032, paired t-test). Comparison of the groups CON and ZOL 6 months after OLT revealed higher CaMEAN and CaPEAK of the cortex in the ZOL group (+4.2%, P = 0.009 and +3.3%, P = 0.040, respectively). Additionally, reduced CaLOW in trabecular (−32.0%, P = 0.047) and cortical (−36.7%, = 0.046) bone was observed for ZOL compared to CON (see Table 4).
Table 4

BMDD parameters of CON and ZOL patients before and after OLT

 

Trabecular bone

Baseline

6 months after OLT

CON (n = 18)

ZOL (n = 21)

CON vs. ZOL

CON (n = 18)

ZOL (n = 21)

CON vs. ZOL

CaMEAN (wt.%)

21.66 (0.77)

21.46 (0.83)

ns

21.50†† (20.32–22.25)

21.77†† (21.57–22.02)

ns

CaPEAK (wt.%)

22.38 (0.62)

22.23 (0.78)

ns

22.36†† (21.32–23.05)

22.53††† (22.18–22.79)

ns

CaWIDTH (wt.%)

3.72 (0.77)

3.80 (0.66)

ns

4.02 (0.89)

3.78 (0.65)

ns

CaLOW (%)

5.17 (3.85–6.81)

5.39 (4.66–7.87)

ns

7.00†† (4.24–12.11)

4.76 (3.88–6.36)

P = 0.047

 

Cortex

Baseline

6 months after OLT

CON (n = 18)

ZOL (n = 21)

CON vs. ZOL

CON (n = 18)

ZOL (n = 21)

CON vs. ZOL

CaMEAN (wt.%)

21.81 (0.84)

21.67 (0.66)

ns

21.19††† (1.18)

22.09* (0.86)

P = 0.009

CaPEAK (wt.%)

22.47 (0.77)

22.31 (0.60)

ns

21.94†† (1.29)

22.67 (0.83)

0.040

CaWIDTH (wt.%)

3.68 (0.66)

3.80 (0.60)

ns

3.95†† (0.78)

3.71 (0.56)

ns

CaLOW (%)

5.44 (3.47)

5.67 (2.31)

ns

6.08 (3.81–13.31)

3.85 (2.88–6.30)

0.046

Data shown are mean (SD) or median (25–75%). Differences between CON and ZOL were tested by t-tests or Mann–Whitney rank sum tests

P = 0.032 for comparison with baseline (paired t-test);  < 0.05, †† < 0.01, and ††† < 0.001 for comparison 6 months after OLT vs. normal trabecular reference data (which are shown in Table 1)

Comparison with normal reference BMDD data [23] showed that the increase in mineralization density due to ZOL did not cause a shift of BMDD toward higher mineralization than normal (CaMEAN and CaPEAK values did not exceed those of normal bone): Trabecular CaMEAN and CaPEAK after ZOL treatment were still lower compared to normal. In the cortex, CaPEAK was lower and CaMEAN was within the normal range (Table 4). In both trabecular and cortical bone, the percentage of bone area undergoing primary mineralization (CaLOW) was similar to normal 6 months after OLT and ZOL treatment.

Changes of histomorphometric parameters compared to baseline were found only for the CON group. Histomorphometric parameters of bone formation were changed due to OLT and Ca/vitamin D supplementation, revealing increases in OV/BV of +267% (< 0.05) and in OS/BS of +422% (< 0.05) in the CON group compared to baseline. Comparison of CON vs. ZOL 6 months after OLT showed lower OV/BV (−81%, P = 0.021) and OS/BS (−62%, P = 0.047) for ZOL (see Table 2). Neither trabecular nor cortical bone structural parameters (BV/TV, Tb.Th, Tb.N, CV/TV, Ct.Th) were significantly changed due to OLT and Ca/vitamin D supplementation alone or OLT and ZOL treatment.

Changes in bone turnover markers could be observed 6 months after OLT and Ca/vitamin D supplementation and ZOL treatment (see Table 3). Compared to baseline, OC was increased in CON (+487%, < 0.01) and ZOL (+72%, < 0.05). Elevated OC levels were far above normal for CON and within the normal range for ZOL. Compared to baseline, CTX was decreased in ZOL (−96%, < 0.05). 25(OH)D was increased compared to baseline in both CON and ZOL (+233%, < 0.001 and +168%, < 0.01). Comparison of the two treatment groups 6 months after OLT showed significant differences for OC, which was higher in CON compared to ZOL (+249%, < 0.001), and CTX, which was reduced in ZOL (−94%, P = 0.01).

Discussion

Beneficial effects of ZOL on fracture incidence 2 years after OLT have been previously described [16]. However, changes in BMD in our analyses of the data of this clinical trial were less conclusive and could not explain the decrease in fracture risk. In this work, we analyzed bone matrix mineralization changes in relation to histomorphometric parameters and biochemical markers in a subgroup of patients of the previous study, providing new information beyond BMD in “hepatic” bone disease as well as its changes due to transplantation, calcium/vitamin D supplementation, and bisphosphonate therapy 6 months after OLT.

“Hepatic” Bone Disease—Baseline Characteristics

Measurements of BMDD revealed that the underlying liver diseases were associated with a shift toward the lower mineralization of trabecular and cortical bone matrix compared to healthy adults described previously [23]. These deviations from normal were independent of the classification of the liver diseases according to Child-Turcotte-Pugh. Potential differences were possibly too small to be detected by the relatively low number of patients in each classification group.

It is interesting that reduced mineralization densities were found in “hepatic” bone disease despite low bone formation, as shown by the histomorphometric observations in biopsies of our patients and as reported by others [8, 10, 11]. It has to be emphasized that, generally, the shape and position of the BMDD peak are dependent on two different processes: first, on kinetics of mineralization and, second, on bone turnover [24]. There is a strong negative linear correlation between the BMDD peak position and histomorphometric bone turnover parameters, as has been shown in cases of osteoporotic patients and patients with primary hyperparathyroidism [3032]. This means that normally average bone matrix mineral content is decreased when turnover parameters are elevated and vice versa. In consequence, the data of our patients indicate that the mineralization kinetics and/or matrix composition have to be altered by the “hepatic” bone disease to explain the decreased mineralization at low turnover status. A comparable situation where low bone turnover is accompanied by decreased mineralization, normal or low BMD, and fragility fractures was found in idiopathic osteoporosis of premenopausal women [33, 34]. Inadequate levels of vitamin D, as have been found in our and other patients referred for OLT [3, 15], could play a crucial role in the etiology of this shift toward lower mineralization in “hepatic” bone disease.

In previous histomorphometric studies of patients referred for OLT, lower trabecular bone volume had been reported [4, 7, 35]. However, in the majority of our patients, trabecular bone volume was not reduced but cortical width and volume were both lower than normal. The latter and the decrease in bone matrix mineralization are consistent with the reduced BMD in our patients and might contribute essentially to increased bone fragility in “hepatic” bone disease already before OLT.

Biochemical parameters of our patients confirmed previous observations. Low vitamin D and low OC levels were also found by others [1, 3, 6]. The reduction of OC was mirrored by reduced osteoid volume and surface in our patients.

Effect of OLT, Ca/Vitamin D Supplementation, and High-Dose ZOL

Already 6 months after liver transplantation and ZOL treatment, a significant shift of BMDD toward higher mineralization, although still lower than normal, could be observed in cortical bone compared to baseline values and compared to controls despite the short time period with respect to the bone remodeling process. CaLOW revealed significantly reduced values in trabecular bone and in the cortex due to ZOL treatment, and cortical CaMEAN and CaPEAK were higher in the ZOL group compared to CON. This is what one would expect, when bone remodeling is reduced: The overall average tissue age should be increased (increase in CaMEAN and CaPEAK) because of a lower percentage of new forming bone sites (reduced CaLOW). These changes in BMDD due to ZOL treatment are in line with the histomorphometrically measured lower osteoid volume and surface. The reduction of bone turnover by ZOL treatment is also reflected by the decreased CTX levels compared to CON. OC levels of the ZOL group were (although significantly increased vs. baseline values) still within the normal range 6 months after OLT. This is in agreement with histomorphometric observations by Vedi et al. [36], who reported trends toward increased bone turnover in pamidronate-treated patients after OLT, indicating that the bisphosphonates clearly reduced, but did not reverse, the increase of bone turnover after OLT. What has also been observed after bisphosphonate treatment in several other studies of women with postmenopausal osteoporosis was a clear narrowing (reduced CaWIDTH) of BMDD [25, 26]. Such a narrowing after short-term reduction of bone remodeling was also proven by computed modeling of BMDD [37]. The lack of this characteristic effect of reduction of CaWIDTH in the ZOL group, which was typically measured after 2 or 3 years of antiresorptive treatment, is likely due to the rather short treatment period (6 months) and the low bone turnover situation before OLT compared to patients with postmenopausal osteoporosis.

In contrast to these changes in the ZOL group, the CON group reflected the tendency toward higher amounts of low mineralized bone and higher bone turnover. The numerical increase in CaLOW in trabecular and cortical bone (although not statistically significant) is in agreement with our histomorphometric findings of increased bone formation parameters, which have also been reported by others [4, 7, 35, 36, 38]. These findings are in line with our data on dramatically elevated OC levels in the control group after OLT, which confirm previously published observations [7]. However, CTX levels remained at the slightly elevated level and did not further rise in our CON patients. These findings of increased bone formation but no further increase in bone resorption have also been reported by others [7, 11, 35, 38]. Somewhat in contrast, Crosbie et al. [39] observed an increase in bone resorption immediately after OLT and Crawford et al. [15] found time-dependent increases (although not statistically significant) of deoxypyridinoline in placebo-treated liver recipients that might be an indication for transient increases also in bone resorption shortly after OLT, which were diminished at 6 months after OLT.

In both of our groups, vitamin D levels were significantly increased compared to baseline, most likely reflecting the compliant vitamin D supplementation and/or recovery of liver function and 25(OH)D hydroxylation in all patients.

The question of whether the reduction of bone turnover by bisphosphonates leads to “frozen bone” and microdamage accumulation is not fully clarified yet and is controversially discussed even in the normal treatment regime [40]. In animal studies, high-dose (supertherapeutic) bisphosphonate treatment was reported not to change the nature of the bone mineral itself [41] but to decrease mechanical toughness [42]. However, beneficial effects of high-dose ZOL on fracture rates, mortality, as well as BMD have been found in our patients [16] and at much lower ZOL doses for postmenopausal women [43, 44]. Osteonecrosis of the jaw, which is a known side effect of high-dose bisphosphonate therapy, was observed in one of our patients after the 12-month surveillance period and successfully treated. No cardiovascular events, including atrial fibrillation, have been observed in our patients. Considering bone matrix mineralization, high-dose ZOL treatment did not cause a shift to higher mineral densities than normal CaMEAN and CaPEAK did not exceed normal values, and CaLOW was reduced to normal after bisphosphonate therapy. Additionally, no signs of mineralization defects, which had been also discussed [41], could be detected in the biopsies from the patients. This is supported by observations from the zoledronic phase III studies in postmenopausal osteoporosis [45], where no evidence of marrow fibrosis, woven bone, or osteomalacia was found.

In conclusion, our patients with “hepatic” bone disease revealed abnormally low bone matrix mineralization together with decreased cortical bone volume, low bone formation, and slightly elevated bone resorption parameters, which all can contribute to bone fragility and increased fracture risk. Interestingly, OLT caused a significant increase in bone formation parameters without further increase in bone resorption markers, which remained 6 months after OLT at the preoperative level. ZOL treatment, however, caused a significant reduction of bone turnover compared to Ca/vitamin D-treated patients. This was accompanied by some restoration of mineralization already visible at 6 months after OLT, which might indicate improvement of the bone material properties, resulting in a beneficial effect of ZOL treatment on fracture risk observed 2 years after OLT.

Acknowledgments

The authors thank G. Dinst, D. Gabriel, P. Messmer, and S. Thon for careful sample preparation and qBEI measurements and histomorphometric analyses at the bone material laboratory of the Ludwig Boltzmann Institute of Osteology, Vienna, Austria. This study was supported by the AUVA (Austrian Social Insurance for Occupational Risk), the WGKK (Social Health Insurance Vienna), the Austrian Science Fund (FWF P-18325), and the Austrian Academy of Science (OELZELT EST370/04).

Copyright information

© Springer Science+Business Media, LLC 2008