SEARCH

SEARCH BY CITATION

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Patients and methods
  5. Study design and patients
  6. Methods
  7. Statistical analysis
  8. Results
  9. Demographic and clinical baseline data
  10. Bone mineral density
  11. Thoracic and lumbar fractures of the spine
  12. Correlations between demographic and clinical data, bone mineral density and spinal fractures
  13. Discussion
  14. Acknowledgements
  15. References
  16. Appendix

Background : A high prevalence of osteoporosis has been noted in Crohn's disease, but data about fractures are scarce.

Methods : The relationship between low bone mineral density and the prevalence of vertebral fractures was studied in 271 patients with ileo-caecal Crohn's disease in a large European/Israeli study. One hundred and eighty-one currently steroid-free patients with active Crohn's disease (98 completely steroid-naive) and 90 steroid-dependent patients with inactive or quiescent Crohn's disease were investigated by dual X-ray absorptiometry scan of the lumbar spine, a standardized posterior/anterior and lateral X-ray of the thoracic and lumbar spine, and an assessment of potential risk factors for osteoporosis.

Results : Thirty-nine asymptomatic fractures were seen in 25 of 179 steroid-free patients (14.0%; 27 wedge, 12 concavity), and 17 fractures were seen in 13 of 89 steroid-dependent patients (14.6%; 14 wedge, three concavity). The prevalence of fractures in steroid-naive patients was 12.4%. The average bone mineral density, expressed as the T-score, of patients with fractures was not significantly different from that of those without fractures (−0.759 vs. −0.837; P=0.73); 55% of patients with fractures had a normal T-score. The bone mineral density was negatively correlated with lifetime steroids, but not with previous bowel resection or current disease activity. The fracture rate was not correlated with the bone mineral density (P=0.73) or lifetime steroid dose (P=0.83); in women, but not in men, the fracture rate was correlated with age (P=0.009).

Conclusions : The lack of correlation between the prevalence of fractures on the one hand and the bone mineral density and lifetime steroid dose on the other necessitates new hypotheses for the pathogenesis of the former.


Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Patients and methods
  5. Study design and patients
  6. Methods
  7. Statistical analysis
  8. Results
  9. Demographic and clinical baseline data
  10. Bone mineral density
  11. Thoracic and lumbar fractures of the spine
  12. Correlations between demographic and clinical data, bone mineral density and spinal fractures
  13. Discussion
  14. Acknowledgements
  15. References
  16. Appendix

In recent years, a multitude of articles have been published indicating that osteopenia and its clinically relevant manifestation osteoporosis are frequent in inflammatory bowel disease,1–7 more so in Crohn's disease than in ulcerative colitis.8, 9 In the literature, the reported prevalence of osteopenia/osteoporosis in inflammatory bowel disease varies widely between 7% and 70%. So far, only a few studies have assessed thenatural history over longer periods of time.10–15 There have been manydiscordant results regarding therisk factors and pathogenesis of osteoporosis in inflammatory bowel disease, but most studies agree on the important negative effect of steroid treatment on bone mineral density.2, 8, 9, 16–19 The variation between study results can probably be explained by differences in the patient population, disease duration, cumulative disease activity, previous medical and surgical treatment, dietary status and smoking habits. Surprisingly, in inflammatory bowel disease, there have only been a few studies of the direct relationship between bone mineral density and fractures.1, 9

The present study describes the degree of lumbar spine osteopenia and osteoporosis and the prevalence and localization of radiologically established vertebral fractures in a large and well-characterized multinational group of patients with ileo-caecal Crohn's disease, with either active disease, presently in need of oral steroid medication, or quiescent disease during protracted steroid medication. The bipolar study population not only reflects frequent management situations in Crohn's disease but, by its diversity, may also render insights into the pathogenesis of osteoporosis in this particular disorder. The data published here represent the baseline of the international MATRIX study, in which the therapeutic effect of prednisolone and budesonide (controlled ileal release) will be examined in relation to a decrease in bone mineral density and development of vertebral fractures.

Study design and patients

  1. Top of page
  2. Summary
  3. Introduction
  4. Patients and methods
  5. Study design and patients
  6. Methods
  7. Statistical analysis
  8. Results
  9. Demographic and clinical baseline data
  10. Bone mineral density
  11. Thoracic and lumbar fractures of the spine
  12. Correlations between demographic and clinical data, bone mineral density and spinal fractures
  13. Discussion
  14. Acknowledgements
  15. References
  16. Appendix

The MATRIX study started in 1996 as a multicentre European/Israeli study. Between July 1996 and July 1999, 278 patients were recruited in 34 academic and non-academic centres in nine countries (see Appendix) and 271 patients participated in the study. The inclusion and exclusion criteria are described in Table 1.

Table 1.  Inclusion and exclusion criteria for the MATRIX study
  1. CDAI, Crohn's disease activity index.

Inclusion criteria
 Age between 20 and 70 years
 Confirmed diagnosis of Crohn's disease (by combination of X-ray, endoscopy, histology and scintigraphy)
 Disease confined to the distal ileum, the ileo-caecal region and/or the ascending colon
 For the steroid-free group, patients not having received steroids during 6 months before the study and having active disease  (CDAI > 150)
 For the steroid-dependent group, patients having received 7–20 mg/day of steroids for at least 4 out of 6 months immediately prior to  inclusion in the study and having quiescent disease (CDAI ≤ 200)
 Signed informed consent
Exclusion criteria
 Previous gastric surgery (except for closure of a perforation or selective vagotomy)
 Ileostomy, colostomy or pouch
 Small bowel resection exceeding 100 cm
 Any resection distal to the mid-transverse colon
 Disease proximal to the ileum
 Active Crohn's disease in the rectum (verified by rectoscopy)
 Complicated Crohn's disease (abscess, obstruction, perforation, active fistulas)
 Clinically relevant renal, hepatic, cardiovascular or psychiatric disease
 Uncontrolled diabetes mellitus
 Active peptic disease
 Rheumatoid arthritis
 Ankylosing spondylitis
 Primary sclerosing cholangitis
 Hyperparathyroidism
 Treatment with calcitonin and/or bisphosphonates within the last 6 months
 Treatment with fluoride, androgens, anabolic steroids, active metabolites of vitamin D within the last 6 months

Demographic and clinical data.  The demographic and clinical data recorded for all patients included the gender, age at the time of investigation, age at the diagnosis of Crohn's disease, extent of Crohn's disease (including localization and extent of previous operations), current medication for Crohn's disease and for non-related diseases, cumulative steroid dose, current steroid dose in the steroid-dependent group, current Crohn's disease activity index,20 body mass index and a physical activity index.21 A smoking history (non-smokers, ex-smokers and smokers) and menstrual history (pre- and postmenopausal female, hormone replacement therapy) were also obtained.

Radiography of the thoracic and lumbar spine.  Lateral and posterior/anterior radiographs of the thoracic and lumbar spine were taken at the patient's clinical centre following a study-specific procedure manual applying the European Vertebral Osteoporosis Study recommendations.22, 23 Radiographs were sent with a coded identification to two expert radiologists (L.L., D.F.) who assessed the findings unaware of relevant clinical or laboratory data. In the radiological assessments, vertebrae with an osteoporotic deformation and a height reduction of more than 20% were considered to be fractured using a morphometric evaluation (six-point measurement) following the algorithm of Felsenberg et al. (Figure 1a,b).22 Osteoporotic vertebral fractures were subdivided into wedge, concavity, biconcavity and crush fractures. Vertebral fractures or deformities with an aetiology other than osteoporosis (for example, degenerative deformities or traumatic fractures) were not included in the assessment.

image

Figure 1. (A) Vertebra with fracture of inferior endplate (six-point measurement). (B) Two vertebrae with wedge fracture.

Download figure to PowerPoint

Bone mineral density measurement.  The bone mineral density of the lumbar spine (posterior/anterior), the left femoral neck and the total body was measured using dual X-ray absorptiometry [Hologic QDR 1000 W, 1500, 2000, 2000 plus, 4000 (Hologic Inc., Bedford, MA, USA); or Lunar DPX, DPX/L or DPXplus (Lunar Inc., Madison, WI, USA)]. All scans were performed according to the manufacturers' instructions. A study-specific quality assurance and procedures manual was distributed by a central quality assurance programme provider (Synarc/MDM Hologic, Maynard, MA, USA). All dual X-ray absorptiometry scans were stored on floppy disks and sent to Synarc/MDM Hologic for centralized review and data collection.

The dual X-ray absorptiometry results were expressed as the bone mineral density in absolute values (g/cm2) and as gender-controlled T-scores using the World Health Organization criteria definition [‘normal’, less than 1 standard deviation (s.d.) below the mean for the reference population; ‘osteopenia’, between 1 s.d. and 2.5 s.d. below the mean; ‘osteoporosis’, more than 2.5 s.d. below the mean].24 The mean bone mineral density from at least three evaluable vertebrae (four when available) from L1 to L4 was used. For the purpose of this study, only the lumbar spine results are referred to, in order to enable correlation with lumbar X-ray findings.

Statistical analysis

  1. Top of page
  2. Summary
  3. Introduction
  4. Patients and methods
  5. Study design and patients
  6. Methods
  7. Statistical analysis
  8. Results
  9. Demographic and clinical baseline data
  10. Bone mineral density
  11. Thoracic and lumbar fractures of the spine
  12. Correlations between demographic and clinical data, bone mineral density and spinal fractures
  13. Discussion
  14. Acknowledgements
  15. References
  16. Appendix

The bone mineral density was calculated by dividing the total bone mineral content by the total area. Fractured vertebrae as seen on the X-ray were excluded (one vertebra for each of 15 patients). For the sake of comparability with Hologic data, Lunar data were transformed using the formula: new bone mineraldensityvalue = (0.906 × Lunar value) − 0.025. T-scores were calculated from bone mineral density values using formulae supplied by Synarc/MDM Hologic.

Means were compared using Student's t-test, and the chi-squared test was used for the distribution of T scores. Correlations were assessed using Spearman rank correlations. Analyses involving several independent variables to explain one dependent variable were performed as analysis of covariance and multiple regression with backward selection of variables.

Demographic and clinical baseline data

  1. Top of page
  2. Summary
  3. Introduction
  4. Patients and methods
  5. Study design and patients
  6. Methods
  7. Statistical analysis
  8. Results
  9. Demographic and clinical baseline data
  10. Bone mineral density
  11. Thoracic and lumbar fractures of the spine
  12. Correlations between demographic and clinical data, bone mineral density and spinal fractures
  13. Discussion
  14. Acknowledgements
  15. References
  16. Appendix

As defined by the study entry criteria, the subgroups of steroid-free and steroid-dependent patients were significantly different in their clinical presentation (Table 2). The steroid-free group consisted of 98 steroid-naive patients (i.e. they had never used corticosteroid treatment for their bowel disease) and 83 patients with previous steroid use. The steroid-free group had a higher disease activity, a shorter duration of disease in the subgroup of steroid-naive patients and a shorter time since the last exacerbation. The groups were similar with regard to gender distribution, smoking habits and, in female patients, hormone replacement therapy. The age was not significantly different between the groups. Five patients aged 17–19 years were erroneously included, as was one patient with four fractures. They are all included in the analyses in this paper.

Table 2.  Patient and disease characteristics in 271 patients with Crohn's disease, either steroid-free or steroid-dependent
 Steroid-free patientsSteroid-dependent patientsP value
Steroid-naivePrevious steroidsOngoing steroids
Number988390 
Male/female47/5138/4547/43N.S.
Mean age (range) (years)36.436.638.7N.S.
Median duration of disease (years)3.39.57.3< 0.001
Time since last exacerbation (months)30.235.337.90.0048
Patients with previous resection (%)2447280.0028
Body mass index (kg/m2)22.222.223.80.0099
Crohn's disease activity index234240133 
Median lifetime steroid dose (mg)036008480 
Postmenopausal women (yes/no)10/418/3710/33N.S.
Hormone replacement therapy4/463/413/40N.S.
Active smokers46 (47%)43 (52%)43 (48%) 
Non-smokers37 (38%)29 (35%)30 (33%)N.S.
Ex-smokers15 (15%)11 (13%)17 (19%) 
Physical activity index14.514.113.5N.S.

Bone mineral density

  1. Top of page
  2. Summary
  3. Introduction
  4. Patients and methods
  5. Study design and patients
  6. Methods
  7. Statistical analysis
  8. Results
  9. Demographic and clinical baseline data
  10. Bone mineral density
  11. Thoracic and lumbar fractures of the spine
  12. Correlations between demographic and clinical data, bone mineral density and spinal fractures
  13. Discussion
  14. Acknowledgements
  15. References
  16. Appendix

The bone mineral density, mean T scores and prevalence of osteopenia and osteoporosis in the three patient subgroups are shown in Table 3. In terms of theabsolute values, the bone mineral density was significantly different in the three groups, with the lowest values in the steroid-dependent group. A subgroup difference was also apparent in the distribution of patients with normal bone mineral density, osteopenia and osteoporosis: nearly two-thirds of steroid-free patients had a normal bone mineral density, compared with less than half of the steroid-dependent patients. Applying a chi-squared test, the difference in the distribution of T-scores was significant (P < 0.001).

Table 3.  Prevalence of vertebral fractures, bone mineral density and T scores in 179 steroid-free and 90 steroid-dependent patients with Crohn's disease
  Steroid-free patients (n = 179)Steroid-dependent patients (n = 90)
 Steroid-naive (n = 97)Previous steroids (n = 82)Ongoing steroids (n = 90)P value
  • *

    Definition of osteopenia.

  • Definition of osteoporosis.

Bone mineral density (g/cm2)1.020.970.94< 0.001
T score−0.47− 0.87− 1.17< 0.001
T score values, n (%)
 T score > −164 (66)46 (56)38 (42) 
 T score = −1 to −2.5*30 (31)27 (33)38 (42)0.006
 T score < −2.53 (3)9 (11)14 (16) 
No. of patients with fractures (%)12/97 (12.4)13/82 (15.9)13/89 (14.6)0.79

Thoracic and lumbar fractures of the spine

  1. Top of page
  2. Summary
  3. Introduction
  4. Patients and methods
  5. Study design and patients
  6. Methods
  7. Statistical analysis
  8. Results
  9. Demographic and clinical baseline data
  10. Bone mineral density
  11. Thoracic and lumbar fractures of the spine
  12. Correlations between demographic and clinical data, bone mineral density and spinal fractures
  13. Discussion
  14. Acknowledgements
  15. References
  16. Appendix

Fifty-six fractures were discovered in 38 of 268 (14.2%) evaluable patients. Forty-one of the fractures were wedge fractures and 15 were concavity fractures. There was no significant difference with regard to prevalence and type between the groups (steroid-free group: 27 wedge, 12 concavity; steroid-dependent group: 14 wedge, three concavity). Biconcavity and crush fractures were not found in the baseline radiographs. All fractures were asymptomatic. Twelve (of 97) steroid-naive patients had vertebral fractures. These 12 steroid-naive patients with fractures were not significantly different from the remaining steroid-free patients with respect to any of the demographic, clinical or laboratory data.

In men, the fracture rate tended to be generally higher than that in women (Table 4), but was similar in all age groups; in women, a significant increase in fracture rate was observed with increasing age (P = 0.0051; linear regression analysis) with a sharp rise in the postmenopausal period. With regard to the localization of fractures, prevalence peaks were found in the middle thoracic spine and thoracic–lumbar junction for wedge fractures, and in the thoracic–lumbar junction and lumbar spine for concavity fractures (Figure 2).

Table 4.  Number (and proportion) of patients with vertebral fractures divided by sex and age group
 TotalAge (years)
17–3031–4041–5051–6061–69
  1. Three patients (all women) had no evaluable X-rays.

Patients26810963463416
Patients with fractures 38 (14%) 12 (11%) 8 (13%) 7 (15%) 6 (18%) 5 (31%)
Women136 49362716 8
Women with fractures 17 (13%)  3 (6%) 4 (11%) 4 (15%) 3 (19%) 3 (38%)
Men132 60271918 8
Men with fractures 21 (16%)  9 (15%) 4 (15%) 3 (16%) 3 (17%) 2 (25%)
image

Figure 2. Prevalence of vertebral fractures in 268 patients with Crohn's disease by type and localization.

Download figure to PowerPoint

Correlations between demographic and clinical data, bone mineral density and spinal fractures

  1. Top of page
  2. Summary
  3. Introduction
  4. Patients and methods
  5. Study design and patients
  6. Methods
  7. Statistical analysis
  8. Results
  9. Demographic and clinical baseline data
  10. Bone mineral density
  11. Thoracic and lumbar fractures of the spine
  12. Correlations between demographic and clinical data, bone mineral density and spinal fractures
  13. Discussion
  14. Acknowledgements
  15. References
  16. Appendix

In the study population as a whole, the bone mineral density and T score were negatively correlated with the lifetime steroid dose (P < 0.001 in both cases), but not with previous bowel resections or current disease activity. Men had significantly lower T scores than women (−1.06 vs. − 0.60; P = 0.0022).

In univariate analyses, the vertebral fracture rate was not significantly influenced by smoking, bowel resections or current disease activity. In women, there was a positive correlation with age (P = 0.009): women above the age of 50 years tended to have more fractures than women below that age (P = 0.059); however, the fracture rate of women above 50 years was not significantly higher than that of men above 50 years of age.

The fracture rate was not correlated with the bone mineral density (P = 0.73) or T-score (P = 0.82) in the patient group as a whole, not even in the subgroup of postmenopausal women (P = 0.83 for T score). The mean lumbar T score of the 38 patients with fractures was − 0.759 ± 1.339 and that of the remaining 230 patients without fractures was − 0.837 ± 1.253 (P = 0.73). Fifty-five per cent of the patients with fractures had a normal T score.

In a multiple regression analysis, none of the following factors was found to independently determine the prevalence of vertebral fractures: gender; age; age at disease onset; previous resection; lifetime steroid dose; current Crohn's disease activity index; steroid-free or steroid-dependent group.

Discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Patients and methods
  5. Study design and patients
  6. Methods
  7. Statistical analysis
  8. Results
  9. Demographic and clinical baseline data
  10. Bone mineral density
  11. Thoracic and lumbar fractures of the spine
  12. Correlations between demographic and clinical data, bone mineral density and spinal fractures
  13. Discussion
  14. Acknowledgements
  15. References
  16. Appendix

This study of 271 patients, either currently steroid-free and with active Crohn's disease, or currently steroid-treated and with inactive or quiescent Crohn's disease, represents, to our knowledge, the only major patient population in which the prevalence of osteopenia and osteoporosis has been investigated simultaneously with a radiological assessment of vertebral fractures. Overall, a reduced bone mineral density (expressed as a T-score < −1) was seen in about 64% of patients and vertebral fractures in 14%. The bone mineral density was significantly lower in the steroid-dependent group than in the steroid-free group and, consequently, a negative correlation between the lifetime steroid dose and the bone mineral density was found. Surprisingly, the fracture rate was similar in both subgroups and was not correlated with the bone mineral density of the lumbar spine. These observations render the pathogenesis and pathophysiology of this important clinical complication of inflammatory bowel disease uncertain.

The entry of patients into this study attempted to match, as closely as possible, the everyday clinical use of steroids in Crohn's disease: patients were either being treated with protracted low-dose steroids to suppress disease activity (refractory to treatment with mesalamine and/or azathioprine alone), but were not being considered for surgery (steroid-dependent patients), or were being considered for steroids because of recent exacerbation (steroid-free patients). Half of the latter group had never been treated with steroids, the other half only on previous occasions. The multicentre international participation in the study allowed the inclusion of a large population of Crohn's disease patients and the neutralization of possible local or regional management strategies. At the same time, the strict definition of inclusion criteria, the uniform collection of demographic and clinical data and the standardized methods used for laboratory tests and morphological investigations allowed valid comparisons to be made between such data and the prevalence of low bone mineral density and fractures. For the assessment of possible risk factors for osteopenia/osteoporosis and fractures, it was an advantage that a number of confounding clinical conditions were excluded at study entry.

The two subgroups showed distinct differences: the steroid-free patients were slightly younger, had a shorter history of Crohn's disease and a lower lifetime dose of steroids. In this group of patients with active disease at the time of inclusion, the proportion of patients with osteopenia/osteoporosis (38%) was comparable to the prevalence found in most cross-sectional studies. In the steroid-dependent subgroup, the combined prevalence of osteopenia and osteoporosis (58%), however, exceeded the reported prevalence of most comparable studies.1–4, 6, 7, 13, 17

The lifetime steroid dose emerged as the strongest causative factor for a decreased bone mineral density in the total study group, followed by male sex. Other factors reported in the literature as contributors to osteopenia and osteoporosis in Crohn's disease and/or ulcerative colitis were not identified in this large survey: age25; smoking habits;26 previous bowel resection;27 or current disease activity.17, 28 Clearly, in the steroid-dependent subgroup, the disease activity is suppressed due to steroid treatment, but not even in the steroid-free subgroup was a correlation found between disease activity and bone mineral density. However, it should be noted that the correlation between disease activity and bone mineral density at any given point may not be conclusive. Only a prospective cumulative registration of the Crohn's disease activity index over a longer period could possibly disclose whether increased systemic inflammatory activity, involving pro-inflammatory cytokines, is a determinant of osteoporosis.29–32

To our knowledge, this is the first study to use systematically both bone mineral density measurement by dual X-ray absorptiometry and radiological assessment of vertebral fractures in a large and well-defined group of Crohn's disease patients. In the study by Abitbol et al., only 34 of the 84 investigated patients had Crohn's disease.1 The proportion of fractures was not reported, nor was the previous steroid use. Jahnsen et al. found fractures of the arms, legs, spine and other locations in 16 (27%) of 60 patients with Crohn's disease;9 however, only symptomatic patients underwent X-ray investigation. The overall rates of spinal fracture of around 14% in the present study appear high in this generally young patient population. The prevalence of vertebral fractures increases up to 43% in female patients above 60 years; however, this group was small (seven patients). The present study suffers from an essential drawback: it cannot compare fracture rates with an age- and gender-matched population from a comparable geographical area. The best comparison for the present study stems from the European Vertebral Osteoporosis Study, in which community-based subjects above the age of 50 years were investigated, and a crude prevalence of osteopenic/osteoporotic deformities of about 10% was found in men and women in the age group 50–65 years.33 The radiological method recommended by the European Vertebral Osteoporosis Study was applied in the present study. Fractures were subdivided into wedge fractures and concavity fractures, the former being significantly more frequent than the latter.

In the present study, no comparison was performed between the bone mineral density of the hip and fractures in that area. Recently, a large, well-controlled, clinical–epidemiological study, based on hospital records, has shown that the relative risk of Crohn's disease patients developing symptomatic vertebral fractures is increased overall by about 50% compared to the background population.34 Yet another study reported an increased risk of low-energy fractures of various anatomical locations in Crohn's disease compared to controls (15.7% vs. 1.4%, respectively) based on a self-administered postal questionnaire and compared with the data of population-based controls.35

The present study describes a similar prevalence of vertebral fractures in male and female Crohn's disease patients. The questionnaire study by Vestergaard et al. reported a higher frequency in women (relative risk, 2.5), but no increased risk in male patients with Crohn's disease.35 The reason for this gender difference is unclear.36 Hypothetically, it could be caused by environmental factors, such as the higher physical activity of men during work and leisure activities. In the present study, in the postmenopausal period, women overtake men with regard to the prevalence of fractures, even though the figures are too small to draw any far-reaching conclusions. It is probable that, in women in this period of life, the effect of disease-induced bone damage combined with hormonally induced alterations is responsible for the disproportionately high fracture rate.37

In this large, well-documented group of Crohn's disease patients, no correlation could be found between the degree of osteopenia and vertebral fractures. Obviously, such a correlation has been assumed to be self-evident for the special case of Crohn's disease, such that it has not been investigated systematically.38, 39 Looking for potential explanations for the lack of correlation in the present study, it could be hypothesized that Crohn's disease patients are more susceptible to fractures without any significant osteopenia in certain biological phases, such as infancy, adolescence, post-menopause or old age. The report of Semeao et al. on six children could be interpreted in this way,40 as could our findings in postmenopausal women. However, in male patients with fractures in the present study, no correlation with either disease onset or senescence could be demonstrated. Another possible explanation is that subclinical fractures occur during phases of active disease — whether treated with high doses of steroids or not, and whether with transient osteopenia or not — and the bone matrix recovers thereafter, obscuring the relationship with the fracture.15 As a third hypothesis, we suggest that fractures might occur as a result of completely different pathophysiological events changing the matrix structure in the trabecular bone of these patients. Such microarchitectural changes in the trabecular bone may weaken the strength of the vertebrae. Whilst the loss of bone mineral density in Crohn's disease is partially reversible,27 the structural changes of the trabecular bone, i.e. cortical thinning and disruption of the trabecular lattice,41 may persist. Non-infectious inflammatory events must also be considered. Recently, one case of osteonecrosis of the hip has been reported, in which granulomatous inflammation with multinucleated giant cells was present histologically;42 in a second case, osteonecrosis was found in a patient with long-standing Crohn's disease never treated with steroids.43 In this sense, vertebral fractures may represent a further extra-intestinal manifestation of inflammatory bowel disease, comparable to primary sclerosing cholangitis or pyoderma gangrenosum, disorders that mainly occur in patients with inflammatory bowel disease but progress surprisingly independent of the natural or treated course of the bowel disorder.

The demonstrated discordance between the low bone mineral density and vertebral fractures in Crohn's disease needs further investigation; it may be necessary to obtain bone biopsies in various phases of Crohn's disease activity and treatment to improve our understanding of the vertebral fractures so frequently found. It would be premature to conclude that, in Crohn's disease, osteopenia does not contribute to the increased fracture risk, as reported by Bernstein et al.34 On the contrary, the combination of early structural changes of the bone matrix, increasing steroid-, age- and/or gender-dependent osteopenia, and incidental low- or high-energy trauma may jointly result in asymptomatic and symptomatic fractures.

In summary, this study demonstrated a high prevalence of osteopenia and osteoporosis in patients with Crohn's disease, which was greater in steroid-dependent than in steroid-free patients. The lifetime steroid dose emerged as the factor most strongly correlated with a reduced bone mineral density, which may reflect a true causative relationship and/or an expression of a high cumulative disease activity. The high frequency of vertebral fractures in Crohn's disease and its lack of correlation with osteopenia and steroid use necessitates further clarification. In the meantime, a pragmatic therapeutic strategy should be chosen, acting on the possible causes of osteopenia (steroids, smoking, physical immobility, sex hormone and vitamin D deficiency) and controlling disease activity with non-steroidal anti-inflammatory and immunosuppressive drugs. If dual X-ray absorptiometry measurement reveals manifest osteoporosis, anti-osteoclastic treatment with bisphosphonates should be considered.44

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Patients and methods
  5. Study design and patients
  6. Methods
  7. Statistical analysis
  8. Results
  9. Demographic and clinical baseline data
  10. Bone mineral density
  11. Thoracic and lumbar fractures of the spine
  12. Correlations between demographic and clinical data, bone mineral density and spinal fractures
  13. Discussion
  14. Acknowledgements
  15. References
  16. Appendix
  • 1
    Abitbol V, Roux C, Chaussade S, et al. Metabolic bone assessment in patients with inflammatory bowel disease. Gastroeneterology 1995; 108: 41722.
  • 2
    Bjarnason I, Macpherson A, Mackintosh C, Buxton-Thomas M, Forgacs I, Moniz C. Reduced bone density in patients with inflammatory bowel disease. Gut 1997; 40: 22833.
  • 3
    Compston JE, Judd D, Crawley EO, et al. Osteoporosis in patients with inflammatory bowel disease. Gut 1987; 28: 4105.
  • 4
    Pigot F, Roux C, Chaussade S, et al. Low bone mineral density in patients with inflammatory bowel disease. Dig Dis Sci 1992; 37: 1396403.
  • 5
    Robinson RJ, Al-Azzawi F, Iqbal SJ, et al. Osteoporosis and determinants of bone density in patients with Crohn's disease. Dig Dis Sci 1998; 43: 25006.
  • 6
    Schoon EJ, Van Nunen AB, Wouters RSME, Stockbrugger RW, Russel MGVM. Osteopenia and osteoporosis in Crohn's disease. Scand J Gastroenterol 2000; 35: 437.
  • 7
    Silvennoinen JA, Karttunnen TJ, Niemela SE, Manelius JJ, Lehtola JK. A controlled study of bone mineral density in patients with inflammatory bowel disease. Gut 1995; 37: 716.
  • 8
    Gosh S, Cowen S, Hannan WJ, Ferguson A. Low bone mineral density in Crohn's disease, but not in ulcerative colitis, at diagnosis. Gastroenterology 1994; 107: 10319.
  • 9
    Jahnsen J, Falch JA, Aadland E, Mowinckel P. Bone mineral density is reduced in patients with Crohn's disease but not in patients with ulcerative colitis: a population-based study. Gut 1997; 40: 3139.
  • 10
    Clements D, Motley RJ, Evans WD, et al. Longitudinal study of cortical bone loss in patients with inflammatory bowel disease. Scand J Gastroenterol 1992; 27: 105560.
  • 11
    Dresner-Pollak R, Karmeli F, Eliakim R, Ackerman Z, Rachmilewitz D. Increased urinary N-telopeptide cross-linked type 1 collagen predicts bone loss in patients with inflammatory bowel disease. Am J Gastroenterol 2000; 95: 699704.
    Direct Link:
  • 12
    Motley RJ, Crawley EO, Evans C, et al. Increased rate of spinal trabecular bone loss in patients with inflammatory bowel disease. Gut 1988; 29: 13326.
  • 13
    Motley RJ, Clements D, Evans WD, et al. A four-year longitudinal study of bone loss in patients with inflammatory bowel disease. Bone Mineral 1993; 23: 95104.
  • 14
    Roux C, Abitbol V, Chaussade S, et al. Bone loss in patients with inflammatory bowel disease: a prospective study. Osteoporos Int 1995; 5: 15660.
  • 15
    Schulte C, Dignass AU, Mann K, Goebell H. Bone loss in patients with inflammatory bowel disease is less than expected. Scand J Gastroenterol 1999; 34: 696702.
  • 16
    Bernstein CN, Seeger LL, Sayre JW, Anton PA, Artinian L, Shanahan F. Decreased bone density in inflammatory bowel disease is related to corticosteroid use and not disease diagnosis. J Bone Miner Res 1995; 10: 2505.
  • 17
    Bischoff SC, Herrmann A, Goke M, Manns MP, Von ZurMuhlen A, Brabant G. Altered bone metabolism in inflammatory bowel disease. Am J Gastroenterol 1997; 92: 115763.
  • 18
    Floren CH, Ahren B, Bengtsson M, Bartosik J, Obrant K. Bone mineral density in patients with Crohn's disease during long-term treatment with azathioprine. J Intern Med 1998; 243: 1236.DOI: 10.1046/j.1365-2796.1998.00246.x
  • 19
    Pollak RD, Karmeli F, Eliakim R, Ackerman Z, Tabb K, Rachmilewitz D. Femoral neck osteopenia in patients with inflammatory bowel disease. Am J Gastroenterol 1998; 93: 148390.
    Direct Link:
  • 20
    Best WR, Becktel JM, Singleton JW, Kern F. Development of a Crohn's disease activity index. National Cooperative Crohn's Disease Study. Gastroenterology 1976; 70: 43944.
  • 21
    Baecke JA, Burema J, Frijters JE. A short questionnaire for the measurement of habitual physical activity in epidemiological studies. Am J Clin Nutr 1982; 36: 93642.
  • 22
    Felsenberg D, Wieland E, Gowin W, et al. Morphometric analysis of spine X-rays for diagnosis of osteoporotic fractures. Med Klin 1998; 93: 2530.
  • 23
    Kiel D. Assessing vertebral fractures. National Osteoporosis Foundation Working Group on Vertebral Fractures. J Bone Miner Res 1995; 10: 51823.
  • 24
    Assessment of fracture risk and its application to screening for postmenopausal osteoporosis. Report of a WHO Study Group. World Health Organ Tech Rep Ser 1994; 843: 1129.
  • 25
    Andreassen H, Hylander E, Rix M. Gender, age, and body weight are the major predictive factors for bone mineral density in Crohn's disease. Am J Gastroenterol 1999; 94: 8248.
    Direct Link:
  • 26
    Silvennoinen JA, Lehtola JK, Niemela SE. Smoking is a risk factor for osteoporosis in women with inflammatory bowel disease. Scand J Gastroenterol 1996; 31: 36771.
  • 27
    Compston JE, Ayers AB, Horton LW, Tighe JR, Creamer B. Osteomalacia after small-intestinal resection. Lancet 1978; i: 912.
  • 28
    Hyams J, Wyzga N, Kreutzer DL, Justinich CL, Gronowicz GA. Alterations in bone metabolism in children with IBD: an in vitro study. J Pediatr Gastroenterol Nutr 1997; 24: 28995.
  • 29
    Fiocchi C. Production of inflammatory cytokines in the intestinal lamina propria. Immun Res 1991; 10: 23946.
  • 30
    Issenman RM. Bones and Crohn's disease: cytokines, a missing link? J Pediatr Gastroenterol Nutr 1997; 24: 3612.
  • 31
    Raisz LG. Local and systemic factors in the pathogenesis of osteoporosis. N Engl J Med 1988; 318: 81828.
  • 32
    Macdonald BR, Gowen M. Cytokines and bone. Br J Rheumatol 1992; 31: 14955.
  • 33
    Lunt M, Felsenberg D, Reeve J, et al. Bone density variation and its effects on risk of vertebral deformity in men and women studied in thirteen European centers: the EVOS Study. J Bone Miner Res 1997; 12: 188394.
  • 34
    Bernstein CN, Blanchard JF, Leslie W, Wajda A, Yu BN. The incidence of fracture among patients with inflammatory bowel disease. A population-based study. Ann Intern Med 2000; 133: 7959.
  • 35
    Vestergaard P, Krogh K, Rejnmark L, Laurberg S, Mosekilde L. Fracture risk is increased in Crohn's disease, but not in ulcerative colitis. Gut 2000; 46: 17681.DOI: 10.1136/gut.46.2.176
  • 36
    Selby PL, Davies M, Adams JE. Do men and women fracture bones at similar densities? Osteoporos Int 2000; 11: 1537.
  • 37
    Cummings SR, Browner WS, Bauer D, et al. Endogenous hormones and the risk of hip and vertebral fractures among older women. Study of Osteoporotic Fractures Research Group. N Engl J Med 1998; 339: 7338.
  • 38
    Monsen U. Inflammatory bowel disease. An epidemiological and genetic study. Acta Chir Scand Suppl 1990; 559: 142.
  • 39
    Angel V, Devesa JM, Martinez R, Vicente E, Nuno J. [Ileoanal anastomosis with a J reservoir. The evolution of the technique and the results.] Anastomosis ileoanal con reservorio en J. Evolucion de la tecnica y resultados. Rev Esp Enferm Dig 1993; 83: 105.
  • 40
    Semeao EJ, Stallings VA, Peck SN, et al. Vertebral compression fractures in pediatric patients with Crohn's disease. Gastroenterology 1997; 112: 17103.
  • 41
    Oleksik A, Ott SM, Vedi S, Bravenboer N, Compston J, Lips P. Bone structure in patients with low bone mineral density with or without vertebral fractures. J Bone Miner Res 2000; 15: 136875.
  • 42
    Freeman HJ, Qwen D, Millan M. Granulomatous osteonecrosis in Crohn's disease. Can J Gastroenterol 2000; 14: 9514.
  • 43
    Khan A, Illiffe G, Houston DS, Bernstein CN. Osteonecrosis in a patient with Crohn's disease unrelated to corticosteroid use. Can J Gastroenterol 2001; 15: 76568.
  • 44
    Haderslev KV, Tjellesen L, Sorensen HA, Staun M. Alendronate increases lumbar spine bone mineral density in patients with Crohn's disease. Gastroenterology 2000; 119: 63946.

Appendix

  1. Top of page
  2. Summary
  3. Introduction
  4. Patients and methods
  5. Study design and patients
  6. Methods
  7. Statistical analysis
  8. Results
  9. Demographic and clinical baseline data
  10. Bone mineral density
  11. Thoracic and lumbar fractures of the spine
  12. Correlations between demographic and clinical data, bone mineral density and spinal fractures
  13. Discussion
  14. Acknowledgements
  15. References
  16. Appendix

The MATRIX study group consists of the following members [country: main investigator (city), number of patients contributed].

Italy: G. Bianchi Porro (Milan), n=22; P. Bianchi (Milan), n=2. Israel: S. Bar-Meir (Tel-Hashomer), n=1; Z. Halperin (Tel Aviv), n=10; Y. Niv (Petach Tikva), n=10; E. Goldin (Jerusalem), n=19. Spain: J.Panez-Diaz (Barcelona), n=18; J. Maria Bilbao (Malaga), n=3; R. Urribarena (Zaragoza), n=1; J.Hinojosa (Valencia), n=3; P. Nos (Valencia), n=2; A. Lanas (Zaragoza), n=5; A. Benages (Valencia), n=1; M. Roca Garcias (Barcelona), n=7; J. Martinez Salmeron (Motril), n=3; J. Garcia Paredes (Madrid), n=5; V Gonzales Lara (Madrid), n=1. Belgium: P.Rutgeerts (Leuven), n=11; M. de Vos (Ghent), n=3. Denmark: J. Dahlerup (Aarhus), n=8. The Netherlands: R. Stockbrügger (Maastricht), n=20; C.Lamers (Leiden), n=1. Norway: J. Jahnsen (Oslo), n=5; M.Vatn (Oslo), n=10; J. Florholmen (Tromsoe), n=17. Sweden: L. Loof (Uppsala), n=11; R. Lofberg (Huddinge), n=8; C. H. Floren (Malmö), n=3. UK: S.Gosh (Edinburgh), n=12; P. Mills (Glasgow), n=19; A. B. Hawthorne (Cardiff), n=13; J. Walters (London), n=7; V. Mani (Leigh); n=8; B. Norton (Derby), n=10.