HM Medical Clinic

 

Icc-ibd.stijlloos.nl

GASTROENTEROLOGY 2011;140:116 –123 Histomorphometric Analysis Reveals Reduced Bone Mass and Bone
Formation in Patients With Quiescent Crohn's Disease
ANGELA E. OOSTLANDER,* NATHALIE BRAVENBOER,*,‡ EVELIEN SOHL,* PAULIEN J. HOLZMANN,*CHRISTIEN J. VAN DER WOUDE,§ GERARD DIJKSTRA,储 PIETER C. F. STOKKERS,¶ BAS OLDENBURG,#J. COEN NETELENBOS,* DANIEL W. HOMMES,** AD A. VAN BODEGRAVEN,‡‡ and PAUL LIPS* on behalf of the Dutch Initiative on Crohn and Colitis (ICC) *Department of Endocrinology, VU University Medical Center, Research Institute MOVE, Amsterdam; Departments of ‡Clinical Chemistry and ‡‡Gastroenterology andHepatology, VU University Medical Center, Amsterdam; §Department of Gastroenterology and Hepatology, Erasmus Medical Center, Rotterdam; 储Department ofGastroenterology and Hepatology, University Medical Center Groningen, University of Groningen, Groningen; ¶Department of Gastroenterology and Hepatology,Academic Medical Center, Amsterdam; #Department of Gastroenterology and Hepatology, University Medical Center Utrecht, Utrecht; and **Department ofGastroenterology and Hepatology, Leiden University Medical Center, Leiden, The Netherlands See related article, on page
Patients with inflammatory bowel disease (IBD), in particular patients with Crohn's disease (CD), are at 30 in CGH; see editorial on page 22.
increased risk for bone loss. The prevalence of osteopeniaand osteoporosis in patients with IBD varies betweenstudies and is reported to be up to 42% and 16%, respec- BACKGROUND & AIMS: Crohn's disease (CD) is asso-
The pathogenesis of bone loss in patients with ciated with an increased prevalence of osteoporosis, but the IBD is complex, multifactorial, and only partly under- pathogenesis of this bone loss is only partly understood. We stood. General risk factors such as malnutrition, malab- assessed bone structure and remodeling at the tissue level in sorption, calcium and vitamin D deficiency, and cortico- patients with quiescent CD. We also investigated the roles steroid treatment are known to be However, of osteocyte density and apoptosis in CD-associated bone these factors cannot completely explain the bone loss.
loss. METHODS: The study included 23 patients with
Recently, the inflammatory process itself has been sug- quiescent CD; this was a subgroup of patients from a large gested to play a pivotal role in IBD-associated bone loss.
randomized, double-blind, placebo-controlled, multicenter For example, proinflammatory cytokines that contribute trial. We obtained transiliac bone biopsy samples and per- to the intestinal immune response in IBD, such as tumor formed histomorphometric analysis. Results were com- necrosis factor ␣ and factors belonging to the receptor pared with data from age- and sex-matched healthy individ- activator for nuclear factor ␬B ligand (RANKL)/osteopro- uals (controls). RESULTS: Trabecular bone volume was
tegerin system, are known to enhance bone decreased among patients with CD compared with controls Bone remodeling involves a balanced process of bone (18.90% vs 25.49%; P ⬍ .001). The low bone volume was resorption by osteoclasts and subsequent bone formation characterized by decreased trabecular thickness (120.61 vs by osteoblasts. An imbalance in this process can be as- 151.42 ␮m; P ⬍ .01). Bone formation and resorption were sessed by measurement of biochemical markers of bone reduced, as indicated by a decreased mineral apposition rate turnover. In patients with IBD, bone resorption is in- (0.671 vs 0.746 ␮m/day; P ⬍ .01) and a low osteoclast creased, as indicated by elevated levels of the markers number and surface area compared with controls and pub- deoxypyridinoline and cross-linked N-telopeptide of type lished values, respectively. In trabecular bone of patients 1 Bone formation in patients with IBD is less with CD, osteocyte density and apoptosis were normal. The well characterized. Increased levels of (bone-specific) al- percentage of empty lacunae among patients was higher kaline decreased levels of than that of published values in controls. CONCLU-
and normal levels of bone formation markers have been SIONS: In adult patients with quiescent CD, bone his-
reported in patients with Therefore, based on tomorphometric analysis revealed a reduction in bone
biochemical markers of bone turnover, the bone remod- mass that was characterized by trabecular thinning. The
CD-associated bone loss was caused by reduced bone
formation, possibly as a consequence of decreased os-
Abbreviations used in this paper: 25(OH)D, 25-hydroxyvitamin D3; ALP, alkaline phosphatase; CD, Crohn's disease; CDAI, Crohn's Dis- teocyte viability in the patients' past.
ease Activity Index; IBD, inflammatory bowel disease; RANKL, receptor Keywords: Bone Histomorphometry; Osteocyte Apoptosis; activator for nuclear factor kappa B ligand; TNF, tumor necrosis factor.
2011 by the AGA Institute Inflammatory Bowel Disease; Osteoporosis.
BONE HISTOMORPHOMETRY IN QUIESCENT CD eling balance in patients with IBD is characterized by a patients with CD and is believed to progress to (presenile) net increase of bone resorption.
osteoporosis. Hence, as such it is the most relevant pop- At the tissue and cellular level, low bone mass is caused ulation to study. Of the 23 patients included, 12 were by a disturbed regulation of bone remodeling as women, of whom one was postmenopausal. Patients with Histomorphometric analysis of bone biopsy specimens is current or recent bisphosphonate or corticosteroid treat- a powerful tool to show disturbances in bone remodel- ment (⬍1 year and ⬍3 months, respectively, before in- ing. In patients with IBD, there are only few data con- clusion) were excluded, as well as patients with known cerning bone histomorphometry. Croucher et al showed metabolic bone diseases and serum levels of 25-hy- reduced bone formation and a mild mineralization defect droxyvitamin D3 less than 25 nmol/L. The study was in patients with IBD who had Ward et al approved by the institutional review board at each par- observed both a suppressed bone formation and bone ticipating medical center. All patients gave written in- resorption in children with newly diagnosed formed consent.
whereas Hessov et al reported normal bone remodelingand mineralization in young patients with The regulation of bone remodeling is believed to be Biochemical measurements in serum included cal- supervised by osteocytes, the third cell type in bone in cium (colorimetric assay; Roche Diagnostics, Mannheim, addition to osteoblasts and osteoclasts. As a conse- Germany), 25-hydroxyvitamin D3 (radioimmunoassay; quence, a lack of functionally active osteocytes might DiaSorin, Stillwater, MN), C-reactive protein (immuno- underlie bone pathology. In fact, the number of osteo- turbidimetric assay; Roche Diagnostics), and total alka- cytes appears to decline with and is lower in line phosphatase (International Federation of Clinical postmenopausal women, especially in those with a Chemistry method; Roche Diagnostics). Intra-assay and vertebral fracture, when compared with premenopausal inter-assay coefficients of variation of the assays were as Moreover, the percentage of apoptotic osteo- follows: calcium, 0.9% and 1.3%; 25-hydroxyvitamin D3, cytes in postmenopausal women has been shown to be 8% and 10%; C-reactive protein, 1.2% and 1.4%; and alka- associated with the level of bone Informa- line phosphatase, 0.7% and 2.4%. Measurements were tion on osteocyte density and osteocyte function in bone performed by the laboratories of Clinical Chemistry and of patients with IBD is lacking thus far.
Endocrinology of the VU University Medical Center.
To ascertain a better understanding of bone biology in patients with IBD, the aims of this study were to (1) assess bone structure and remodeling at the tissue level in Patients received 2 doses of tetracycline (250 mg patients with quiescent CD in comparison with healthy 4 times daily) for 2 days separated by an interval of 10 controls and (2) investigate whether osteocyte density days. Three to 7 days after the last dose, transiliac bone and osteocyte apoptosis are related to bone structure and biopsy specimens were taken using Bordier's trephine (ID, 8 mm). The bone biopsy specimens were fixed in cold4% phosphate-buffered formaldehyde, dehydrated in gradedethanol, and embedded in 80% methylmethacrylate (BDH Patients and Methods
Chemicals, Poole, England) supplemented with 20% dibuth- ylphtalate (Merck, Darmstadt, Germany), 8 g/L lucidol CH- Twenty-three patients with quiescent CD partici- 50L (Akzo Nobel, Deventer, The Netherlands), and 22 pated in this study. These patients were a subgroup ␮L/10 mL N,N dimethyl-p-toluidine (Merck). Five-micro- included in a large randomized, double-blind, placebo- meter-thick undecalcified sections were cut with a Polycut controlled, multicenter trial on the effect of risedronate 2500 S microtome (Reichert–Jung, Nussloch, Germany).
in patients with quiescent CD who had osteopenia (N ⫽ Goldner's trichrome stain was performed to distinguish 131, Crohn and Bone Study). The current study describes between calcified bone and noncalcified matrix (osteoid).
data obtained from patients at baseline. Patients were Tartrate-resistant acid phosphatase stain was used to visu- diagnosed with CD using clinical, endoscopic, histologic, alize osteoclasts. Unstained sections were used for fluores- and radiologic criteria according to Pa- cence microscopy to measure tetracycline labels.
tients had to be in remission (Crohn's Disease ActivityIndex [CDAI] ⬍150). Although 2 patients had a CDAI ⬎150, these patients had absence of symptoms of active Apoptotic cells were visualized by immunohisto- disease and no signs of active inflammation on sigmoid- chemical detection of activated caspase-3. Immunohisto- oscopy and as such were defined to be in clinical remis- chemistry was performed on two 5-␮m sections of each sion. Another inclusion criterion was a lumbar spine biopsy specimen, which were obtained with an interval of and/or total hip bone mineral density with a T-score of 30 ␮m. Sections were cut and transferred to poly-L- ⫺1 to ⫺2.5 SD (osteopenia). Osteopenia is the most lysine– coated slides. After deplastification and rehydra- prevalent state of bone loss in usually relatively young tion, sections were decalcified for 10 minutes with 1% GASTROENTEROLOGY Vol. 140, No. 1 acetic acid. Antigen retrieval was performed by a lacunae per bone area, total number of lacunae per bone 30-minute incubation with 0.5% saponin (Sigma, St area, percentage of positive osteocytes per total number Louis, MO) in phosphate-buffered saline (PBS) and a of osteocytes, and percentage of empty lacunae per total 10-minute incubation with 3.5 ␮g/mL deoxyribonuclease number of lacunae.
II (Sigma) in 25 mmol/L Tris plus 10 mmol/L magne- To compare the data of this CD population with sium sulfate. Sections were incubated with 3% hydrogen healthy controls, we used healthy control data from 17 peroxide in methanol to block endogenous peroxidase men and 26 women (aged 20 – 60 kindly provided and with 5% normal goat serum in PBS plus 0.05% by Prof Dr J. E. Compston (Cambridge, England). From this Tween20 to block nonspecific binding sites. Incubation population, sections of 10 age- and sex-matched healthy with primary antibody was performed overnight at 4°C controls were measured to verify comparability of the his- with 1/75 rabbit anti-cleaved caspase-3 antibody (Cell tomorphometric indices. The systemic measurement differ- Signaling Technology, Beverly, MA) in PBS plus 0.05% ence between our measurements and those from England Tween20. Sections were then incubated for 1 hour with was as follows: bone volume, 0.93; trabecular thickness, 1/100 biotin-labeled goat anti-rabbit immunoglobulin G 0.88; trabecular number, 0.99; osteoid volume, 0.93; and (Vector Laboratories, Burlingame, CA) in PBS plus 0.05% osteoid thickness, 0.98.
Tween20. Subsequently, the sections were incubated for30 minutes with the ABC kit (Vector Laboratories) and developed for 10 minutes with 3,3=-diaminobenzidine Results are expressed as mean ⫾ SD. Histomor- with nickel enhancement. Finally, sections were counter- phometric indices were compared between patients with stained with 0.025% toluidine blue in water, dehydrated, CD and healthy controls, as well as between men and and sealed in DePeX mounting medium (BDH Chemi- women using a 2-tailed Student t test for independent cals, VWR International, Poole, England).
samples. Correlations between histomorphometric pa-rameters and biochemical indices were calculated using Pearson's coefficient of correlation. All statistical analyses Histomorphometry was performed mainly on tra- were performed using SPSS software (version 15.0; SPSS becular bone. Static histomorphometry was performed Inc, Chicago, IL). A P value of ⬍.05 was considered automatically using NIS-Elements AR 2.10 (Nikon ¨sseldorf, Germany) at 100⫻ magnification.
Dynamic histomorphometry was performed semiauto- matically using OsteoMeasure (Osteometrics, Atlanta, GA). All measurements were performed according to theAmerican Society of Bone and Mineral Research nomen- Patient characteristics are listed in Be- Mean cortical thickness was assessed manually cause several parameters showed relatively high variation by measuring both cortices, each at 4 equidistant places.
in the total population, data are presented for men and Trabecular bone volume, trabecular bone surface, and women separately. The quiescent state of disease in our trabecular thickness were measured and used to calculate CD population was ascertained by a mean CDAI of trabecular number and trabecular separation. Besides 93.8 ⫾ 71.7 and a mean C-reactive protein level of 7.3 ⫾ these structural indices, parameters associated with bone 8.4 mg/L. Disease activity was similar in men and women formation were measured, including osteoid volume, os- with CD. Disease duration tended to be longer in male teoid surface, and osteoid thickness. Furthermore, the patients with CD (P ⫽ .067). Levels of biochemical pa- distance between double tetracycline labels and the la- rameters were all within the reference values of our lab- beled (mineralizing) surface was measured under UV oratory. However, alkaline phosphatase levels in men light and used to calculate mineral apposition rate, bone with CD were at the upper limit of the reference range formation rate, adjusted apposition rate, and mineraliza- and were higher than alkaline phosphatase levels in tion lag time. Bone resorption was assessed as osteoclast women with CD (86 vs 60 IU/L; P ⫽ .003).
number (measured manually using an integrated eye-piece, Zeiss II; Zeiss, Oberkochen, Germany) and osteo- clast surface. All measurements were performed by one Histomorphometric data concerning bone struc- ture are summarized in Trabecular bone volume In addition, the total number of osteocytes, the num- and trabecular thickness were reduced in patients with ber of cleaved caspase-3–positive osteocytes, and the CD when compared with healthy controls (P ⬍ .001 and number of empty lacunae in the entire trabecular bone P ⫽ .006, respectively; and B). Trabecular area were counted at 200⫻ magnification. Duplicate sec- number was unaffected. Comparison of bone structure tions were analyzed by the same investigator. From these parameters between men and women showed no differ- data, the following parameters were calculated: total ences within the control population. However, in patients number of osteocytes per bone area, number of empty with CD, trabecular bone volume tended to be lower in BONE HISTOMORPHOMETRY IN QUIESCENT CD Table 1. Patient Characteristics Body mass index (kg/m2) 82 ⫾ 61 (0–194) 105 ⫾ 81 (6–322) C-reactive protein level (mg/L)a 6.6 ⫾ 6.2 (1–17) 7.9 ⫾ 10.3 (1–38) Location of disease Large intestine/small intestine/both Bowel resection, yes/no (n) Disease duration (y) Age at diagnosis (y) Mesalamine derivatives 25-Hydroxyvitamin D3 (nmol/L)a Alkaline phosphatase (U/L)a 86 ⫾ 25b,c Bone mineral density lumbar spine (g/cm2) Bone mineral density total hip (g/cm2) NOTE. Data represent mean ⫾ SD; in the case of CDAI and C-reactive protein, ranges are shown in parentheses.
aReference ranges obtained from the Department of Clinical Chemistry of the VU University Medical Center: C-reactive protein, ⬍8 mg/L;25-hydroxyvitamin D3, 25–150 nmol/L; calcium, 2.20 –2.60 mmol/L; alkaline phosphatase, ⬍120 IU/L.
bN ⫽ 10; alkaline phosphatase value of one patient with a liver function test abnormality was excluded.
cSignificantly different from women with CD (P ⬍ .01).
men compared with women (P ⫽ .056). Trabecular num- in patients with CD when compared with healthy con- ber was lower in men than in women with CD (P ⫽ .037), trols (P ⫽ .007; ). None of the patients had and trabecular separation was higher in men (P ⫽ .049).
signs of osteomalacia. Osteoid volume, mineralizing sur- In data of male and female patients with CD are face, and bone formation rate were comparable between summarized, including variables that were only available healthy controls and patients with CD , D, and in patients with CD.
F). However, analysis of the individual data revealed a Histomorphometric data concerning bone remodeling cluster of 9 patients with CD who had a very low min- are shown in Mineral apposition rate was lower eralizing surface (3.78 ⫾ 1.80) as well as bone formation Table 2. Bone Structure and Remodeling Parameters in Healthy Controls and Patients With CD Abbreviation (unit) Controls (n ⫽ 43) Patients with CD (n ⫽ 23) 18.90 ⫾ 5.28a Trabecular thickness Tb.Th (␮m) 120.61 ⫾ 22.16b Trabecular number Tb.N (/mm) Osteoid thickness O.Th (␮m) Mineralizing surface Mineral apposition rate MAR (␮m/day) 0.671 ⫾ 0.098b Bone formation rate Adjusted apposition rate Aj.Ar (␮m/day) Mineralization lag time NOTE. Data represent mean ⫾ SD.
aSignificantly different from the control group (P ⬍ .001).
bSignificantly different from the control group (P ⬍ .01).

GASTROENTEROLOGY Vol. 140, No. 1 Figure 1. Histomorphometric measurements on bone structure and remodeling in healthy controls and patients with CD. (A) Bone volume (BV/TV)
is decreased in patients with CD. (B) Trabecular thickness is decreased in patients with CD. (C) Osteoid volume (OV/BV) is unaffected in
patients with CD. (D) Mineralizing surface (MS/BS) is unaffected in patients with CD; however, a subset of 9 patients with CD had a very low
mineralizing surface. (E) Mineral apposition rate (MAR) is decreased in patients with CD. (F) Bone formation rate (BFR) is unaffected in patients with
CD; however, a subset of 9 patients with CD had a very low bone formation rate.
rate (0.022 ⫾ 0.011) in comparison with both the total patients had a low osteoid volume (0.42 ⫾ 0.34), a low CD population and healthy controls. Moreover, the 9 number of osteoclasts and osteoclast surface (0.19 ⫾ 0.22 patients in this subgroup (4 men and 5 women; age, 43 ⫾ and 1.36 ⫾ 0.80, respectively), a low mineral apposition 13 years) turned out to have a low bone volume (15.47 ⫾ rate (0.59 ⫾ 0.08), and a low adjusted apposition rate 2.90), which was accompanied by a low total hip bone (0.29 ⫾ 0.16). Comparison of bone remodeling parame- mineral density (T-score, ⫺1.38 ⫾ 0.55). Trabecular thick- ters between men and women showed no statistically ness and trabecular number were low as well (104.61 ⫾ significant differences in both healthy controls and pa- 16.05 and 1.68 ⫾ 0.33, respectively). Furthermore, these tients with CD. Examination of the relationship between Table 3. Bone Structure and Remodeling Parameters in Male and Female Patients With CD Abbreviation (unit) Cortical thickness Ct.Th (␮m) 16.79 ⫾ 4.58a Trabecular thickness Tb.Th (␮m) Trabecular number Tb.N (/mm) 1.63 ⫾ 0.30b Trabecular separation Tb.Sp (␮m) 522.9 ⫾ 116.6b Osteoid thickness O.Th (␮m) Osteoclast number N.Oc/TA (/mm2) Osteoclast surface Oc.S/BS (%) Mineralizing surface Mineral apposition rate MAR (␮m/day) Bone formation rate Adjusted apposition rate Aj.Ar (␮m/day) Mineralization lag time NOTE. Data represent mean ⫾ SD.
aTendency towards significantly different from women with CD (P ⫽ .056).
bSignificantly different from women with CD (P ⬍ .05).

BONE HISTOMORPHOMETRY IN QUIESCENT CD Figure 2. Osteocyte apoptosis in patients with CD. (A) Immunohistochemical staining of a bone biopsy section for cleaved caspase-3. The black
arrowhead indicates an apoptotic osteocyte, and white arrowheads indicate empty lacunae. (B) Positive correlation between disease activity (CDAI)
and the percentage of apoptotic osteocytes.
any of the histomorphometric parameters and disease controls. Bone remodeling parameters did not differ be- activity did not reveal statistically significant correlations.
tween sexes.
Also, no correlations were found between other biochem- The observed low bone mass in patients with quiescent ical parameters and any of the histomorphometric indi- CD is in agreement with previous studies on bone histo- morphometric analysis in patients with Weshowed that the mild bone mass deficit in patients with quiescent CD is related to thinning of trabeculae. Only Staining for cleaved caspase-3 clearly visualized one other study published details on bone microarchi- apoptotic osteocytes, as depicted in . In this tecture in patients with IBD, reporting a mild cortical figure, empty lacunae can be identified as well. In trans- bone deficit but a normal amount and structure of tra- iliac bone biopsy specimens of patients with CD, the total becular bone in children with newly diagnosed In lacunar density was 293 ⫾ 87 per mm2 bone consisting of our study, cortical thickness in bone of patients with CD 265 ⫾ 76 osteocytes per mm2 bone and 28 ⫾ 11 empty was comparable to values in healthy controls (J. Comp- lacunae per mm2 bone. The percentage of empty lacunae ston, personal communication). This discrepancy might over the total number of lacunae was 9.27% ⫾ 3.43%. The be due to a different response of the developing skeleton number of positive osteocytes over the total number of to IBD than the mature skeleton. In addition, patients osteocytes was 3.01% ⫾ 2.23%. None of the parameters in the study of Ward et al were included at diagnosis, differed were statistically significant between men and before treatment had been instituted, which might be a women with CD.
cause of the difference as well.
Examination of the relationship between bone remod- The relatively small reduction in bone formation we eling parameters and osteocyte density as well as osteo- observed in patients with quiescent CD was also reported cyte apoptosis did not reveal statistically significant cor- by Croucher et It must be mentioned that, in con- relations. Examination of the relationship between trast to our population, the majority of their patients disease activity and osteocyte density as well as osteocyte received prednisolone therapy at time of the bone biopsy.
apoptosis showed a positive correlation of CDAI with the This is likely to have been a contributory factor to the percentage of caspase-3–positive osteocytes (r ⫽ 0.496, bone loss and might explain the stronger reduction in P ⫽ .022; see ).
bone formation they observed in comparison with ourfindings.
Interestingly, in our study population, a subset of In this study, bone structure and remodeling in patients had histomorphometric characteristics resem- transiliac bone biopsy specimens from patients with qui- bling osteoporosis rather than osteopenia. In these pa- escent CD were investigated. In short, our results indicate tients, all bone formation indices were low and compa- that trabecular bone volume was lower in patients with rable to the extent of reduction of bone formation seen CD than in healthy age- and sex-matched controls. The by Croucher et al. Because recent corticosteroid treat- reduction of bone mass was more pronounced in men ment was an exclusion criterion in this study, current than in women and was characterized by a reduction in corticosteroid therapy cannot be the cause of the reduced trabecular thickness as well as trabecular number. Besides bone formation in our patients. Sex, age, current medi- structural differences, we observed a change in bone cation use, and number of patients with small bowel remodeling as indicated by a reduced mineral apposition resection(s) in the past were similar in the subset and rate in this CD population when compared with healthy total population. Unfortunately, prospective data on life- GASTROENTEROLOGY Vol. 140, No. 1 time cumulative dose of glucocorticoids and the history osteocyte density in bone of patients with CD is within of number as well as severity of relapses are not available the range observed in healthy Literature on in the current study. Both aspects could well be related to osteocyte apoptosis in healthy controls is limited, be- a more severely affected bone phenotype. However, a cause the assessment of apoptotic osteocytes in bone retrospective estimate showed no statistically significant specimens is relatively new in bone morphometry. None- correlation between cumulative glucocorticoid dose and theless, the percentage of apoptotic osteocytes in our CD any of the bone formation parameters.
population is comparable to the percentage observed in Glucocorticoids inhibit differentiation and activity of hip fracture controls by Sambrook et However, the osteoblasts and its precursor cells and, moreover, induce percentage of empty lacunae is relatively high when com- osteoblast An advantage of this study is that pared with the controls in that study and resembles more only patients without recent corticosteroid treatment the percentage of empty lacunae observed in their pred- were included. Inherent to this aspect, patients in this nisone-treated patients with rheumatoid arthritis. This study were in a quiescent state of disease, which is a finding may be explained by an increased osteocyte apo- poorly investigated but interesting group of patients be- ptosis in the past, because empty lacunae mark sites cause there are indications that the cumulative effects of where osteocytes have died previously. This hypothesis is even small excesses in cytokine levels over many months corroborated by the positive correlation we observed be- can be clinically relevant. One of our findings supporting tween disease activity (CDAI) and the percentage of ap- this theory is that bone of male patients with CD, who optotic osteocytes. In that case, osteocyte apoptosis tended toward longer disease duration, was more severely might have been increased during an active period of affected than bone of female patients. Men with CD in disease in patients with CD. However, C-reactive protein this cohort had lower bone volume and both less and level did not correlate with the percentage osteocyte ap- thinner trabeculae than women with CD. The latter as- optosis. Because CDAI is inaccurate in the low range, the pect is in contrast to observations in healthy controls correlation of CDAI with osteocyte apoptosis might be describing no difference in bone structure between men due to chance. Therefore, the biological relevance of this and Moreover, in postmenopausal popula- finding remains to be elucidated.
tions, bone loss due to the loss of whole trabeculae has A drawback of bone histomorphometry is a consider- been reported mainly in A higher bone turn- able measurement variation. Differences in staining tech- over in male patients with CD in this cohort, as indicated niques, methods to measure the specimen, as well as by higher levels of alkaline phosphatase than in female intersection and interobserver variance can be substan- patients, might partially explain this finding. However, tial. The number of patients studied is relatively small, direct differences in bone remodeling indices could not but it still is the largest series in well-characterized pa- be detected between men and women with CD, which tients with IBD. Another limitation is that patients with suggests the existence of differences between male and IBD reflect a heterogeneous population. In this study, female patients in the past.
only patients with CD were included. In addition, these Bone resorption data were not available in the healthy patients were in a quiescent state of disease and had not control population, but when compared with the litera- used glucocorticoids for at least 3 months before inclu- ture, osteoclast number and osteoclast surface seemed to sion. These restrictions to patient inclusion should have be low in our patients with Our data are in reduced the impact of heterogeneity in this study. How- concordance with previous studies addressing bone re- ever, we still might not have had enough statistical power sorption in patients with IBD, which showed a slight to detect certain differences and correlations.
decrease in bone resorption indices in both children and In conclusion, in this study, for the first time a cohort adults with These findings raise the question of adult patients with quiescent CD was evaluated using whether treatment with antiresorptive agents is beneficial bone histomorphometry. In these patients, bone mass is to all patients with CD with osteoporosis or osteopenia.
reduced as characterized by trabecular thinning. Further- However, additional data on eroded surface, resorption more, our results show that CD-associated bone loss is cavities, and serum levels of bone resorption markers are caused by a reduced bone formation at the tissue level, necessary to make firm conclusions on bone resorption possibly as a consequence of decreased osteocyte viability in this study population.
in the past.
Osteocyte apoptosis interrupts signaling between os- teocytes and the effector cells of bone: osteoblasts andosteoclasts. Therefore, a lack of osteocytes or inefficacy of their function can be related to a defective bone structure 1. Bernstein CN, Leslie WD. Review article: osteoporosis and inflam- and remodeling. As far as we know, osteocyte density and matory bowel disease. Aliment Pharmacol Ther 2004;19:941–952.
apoptosis in transiliac bone biopsy specimens from pa- 2. Vestergaard P. Bone loss associated with gastrointestinal dis- tients with CD have not been studied before. From com- ease: prevalence and pathogenesis. Eur J Gastroenterol Hepatol parison of our findings with the literature, it appears that 2003;15:851– 856.
BONE HISTOMORPHOMETRY IN QUIESCENT CD 3. Bernstein CN, Leslie WD. The pathophysiology of bone disease in 18. van Essen HW, Holzmann PJ, Blankenstein MA, et al. Effect of gastrointestinal disease. Eur J Gastroenterol Hepatol 2003;15: raloxifene treatment on osteocyte apoptosis in postmenopausal women. Calcif Tissue Int 2007;81:183–190.
4. Moschen AR, Kaser A, Enrich B, et al. The RANKL/OPG system is 19. Lennard-Jones JE. Classification of inflammatory bowel disease.
activated in inflammatory bowel disease and relates to the state Scand J Gastroenterol Suppl 1989;170:2– 6.
of bone loss. Gut 2005;54:479 – 487.
20. Parfitt AM, Drezner MK, Glorieux FH, et al. Bone histomorphom- 5. Robinson RJ, Iqbal SJ, Abrams K, et al. Increased bone resorption etry: standardization of nomenclature, symbols, and units. Re- in patients with Crohn's disease. Aliment Pharmacol Ther 1998; port of the ASBMR Histomorphometry Nomenclature Committee.
12:699 –705.
J Bone Miner Res 1987;2:595– 610.
6. Dresner-Pollak R, Karmeli F, Eliakim R, et al. Increased urinary 21. Vedi S, Compston JE, Webb A, et al. Histomorphometric analysis N-telopeptide cross-linked type 1 collagen predicts bone loss in of bone biopsies from the iliac crest of normal British subjects.
patients with inflammatory bowel disease. Am J Gastroenterol Metab Bone Dis Relat Res 1982;4:231–236.
2000;95:699 –704.
22. Dalle CL, Bertoldo F, Valenti MT, et al. Histomorphometric anal- 7. Gilman J, Shanahan F, Cashman KD. Altered levels of biochemi- ysis of glucocorticoid-induced osteoporosis. Micron 2005;36: cal indices of bone turnover and bone-related vitamins in patients with Crohn's disease and ulcerative colitis. Aliment Pharmacol 23. Kimmel DB, Recker RR, Gallagher JC, et al. A comparison of iliac bone histomorphometric data in post-menopausal osteoporotic 8. Abitbol V, Roux C, Chaussade S, et al. Metabolic bone assess- and normal subjects. Bone Miner 1990;11:217–235.
ment in patients with inflammatory bowel disease. Gastroenter- 24. Ott SM, Oleksik A, Lu Y, et al. Bone histomorphometric and ology 1995;108:417– 422.
biochemical marker results of a 2-year placebo-controlled trial of 9. Bischoff SC, Herrmann A, Goke M, et al. Altered bone metabolism raloxifene in postmenopausal women. J Bone Miner Res 2002; in inflammatory bowel disease. Am J Gastroenterol 1997;92: 25. Sambrook PN, Hughes DR, Nelson AE, et al. Osteocyte viability 10. Bjarnason I, Macpherson A, Mackintosh C, et al. Reduced bone with glucocorticoid treatment: relation to histomorphometry. Ann density in patients with inflammatory bowel disease. Gut 1997; Rheum Dis 2003;62:1215–1217.
40:228 –233.
11. E.F.Eriksen BLMK. The cellular basis of osteoporosis. Spine: state of the art reviews. 8th ed. 1994.
12. Croucher PI, Vedi S, Motley RJ, et al. Reduced bone formation in Received June 3, 2010. Accepted September 9, 2010.
patients with osteoporosis associated with inflammatory boweldisease. Osteoporos Int 1993;3:236 –241.
Reprint requests 13. Ward LM, Rauch F, Matzinger MA, et al. Iliac bone histomorphom- Address requests for reprints to: Nathalie Bravenboer, PhD, etry in children with newly diagnosed inflammatory bowel dis- Departments of Endocrinology and Clinical Chemistry, VU University ease. Osteoporos Int 2010;21:331–337.
Medical Center, PO Box 7057, 1007 MB Amsterdam, The 14. Hessov I, Mosekilde L, Melsen F, et al. Osteopenia with normal Netherlands. e-mail: fax: (31) 0 20 44 42 vitamin D metabolites after small-bowel resection for Crohn's disease. Scand J Gastroenterol 1984;19:691– 696.
15. Qiu S, Rao DS, Palnitkar S, et al. Age and distance from the surface but not menopause reduce osteocyte density in human The authors thank Prof Dr Juliet Compston and Linda Skingle for cancellous bone. Bone 2002;31:313–318.
supplying the control material and data.
16. Mullender MG, Tan SD, Vico L, et al. Differences in osteocyte density and bone histomorphometry between men and women Conflicts of interest and between healthy and osteoporotic subjects. Calcif Tissue Int The authors disclose no conflicts.
17. Qiu S, Rao DS, Palnitkar S, et al. Reduced iliac cancellous osteocyte density in patients with osteoporotic vertebral fracture.
The Initiative on Crohn and Colitis (ICC) Foundation received a J Bone Miner Res 2003;18:1657–1663.
research grant from Sanofi-Aventis.

Source: http://icc-ibd.stijlloos.nl/upload/files/publicaties/2011-Oostlander%20et%20al-Gastroenterology.pdf