You are here

Article Text

Article menu

Download PDFPDF

Clinical and epidemiological research
Extended report
Assessment of hand bone loss in rheumatoid arthritis by high-resolution peripheral quantitative CT
  1. Anne Fouque-Aubert1,
  2. Stéphanie Boutroy1,
  3. Hubert Marotte2,
  4. Nicolas Vilayphiou1,
  5. Justine Bacchetta1,
  6. Pierre Miossec2,
  7. Pierre Dominique Delmas1,
  8. Roland D Chapurlat1
  1. 1Inserm U831, Université de Lyon, Service de Rhumatologie, Hôpital Edouard Herriot, Lyon, France
  2. 2Université de Lyon, Service de Rhumatologie, Hôpital Edouard Herriot, Lyon, France
  1. Correspondence to Dr Anne Fouque-Aubert, Inserm U831, Université de Lyon, Service de Rhumatologie, Hôpital Edouard Herriot, 5 Place d'Arsonval, 69437 Lyon, France; afamedezine{at}hotmail.com

Abstract

Objectives A new high-resolution peripheral quantitative CT (HR-pQCT) system allows for in vivo assessment of bone microarchitecture and volumetric bone mineral density (vBMD) with an 82 µm isotropic resolution. With this device, the microarchitecture impairment was evaluated in patients with rheumatoid arthritis (RA) in comparison with healthy controls and measured the erosion volume at metacarpal heads (MCPs).

Methods In this cross-sectional study, the reproducibility was first assessed by 3 HR-pQCT exams with repositioning in 14 patients with late RA and 14 healthy subjects. Then, HR-pQCT parameters were measured in a group of 93 patients with RA and 31 healthy controls. Two RA subgroups were distinguished: early RA (disease duration ≤2 years) (n=36) and late RA (n=57) and compared them to healthy controls.

Results The precision of the HR-pQCT volumetric measurements as assessed with coefficient of variation ranged from 0.7% to 1.8% in patients with late RA and from 0.6% to 1.4% in healthy subjects at MCPs. Total and trabecular vBMD and trabecular thickness were significantly decreased in patients with RA compared to healthy subjects and were significantly correlated to disease activity. The erosion volume was highly correlated to a semiquantitative assessment using the Outcome Measures in Rheumatoid Arthritis Clinical Trials (OMERACT) scoring system applied to the HR-pQCT slices.

Conclusions This study demonstrated the good reproducibility of the HR-pQCT volumetric measurements at MCPs and confirmed the involvement of trabecular compartment in periarticular osteopoenia. Thus, HR-pQCT appears interesting to simultaneously assess differences in bone volumetric density, microarchitecture and erosions.

https://doi.org/10.1136/ard.2009.114512

Statistics from Altmetric.com

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    Introduction

    In rheumatoid arthritis (RA), bone loss occurs as periarticular osteopoenia around inflamed joints and generalised osteoporosis.1 Periarticular bone loss is related to the production of several proinflammatory cytokines such as tumour necrosis factor α (TNFα), interleukin 6 and interleukin 1β by the RA synovium. These cytokines induce the expression of receptor activator of nuclear factor-κB ligand (RANKL) on the surface of activated T cells, rheumatoid synovial cells and osteoblastic cells. RANKL promotes osteoclastic differentiation and activation leading to periarticular osteoporosis and marginal erosions.2,–,4 In affected joints, hand and particularly metacarpal heads (MCPs) are among the earliest sites to be affected by RA.5 Several techniques including dual x-ray absorptiometry (DXA), digital x-ray radiogrammetry or quantitative ultrasound technique have been developed for measuring hand bone mineral density (BMD). However, so far no gold standard has been defined for periarticular osteoporosis assessment of the hand. Moreover, it remains to be clarified whether the trabecular or cortical bone compartment is preferentially involved.

    Bone erosions are also specific to RA, reflecting disease severity, and can be used to monitor disease progression.6 7 Most of the erosions occurs within 2 years of onset of the disease, but may progress even over the following 10 years.8 Presence of erosions at diagnosis is one of the poorest prognosis factors9 and one of the diagnostic arguments. To date, conventional radiographs have been considered as the gold standard to evaluate bone erosions.10 Despite a high specificity for detection of hand or feet erosions, radiographs lack sensitivity in early disease,11 probably because erosions are detectable only if they are tangential to the x-ray beam. Contrast is also poor for small erosions. Thus, other techniques such as MRI or ultrasonography have been recently developed to quantify bone erosions more accurately.12 13

    A new high-resolution peripheral quantitative CT system (HR-pQCT) has been used to assess in vivo trabecular architecture and volumetric BMD (vBMD) at the radius and tibia.14 15 In postmenopausal osteoporotic women, the HR-pQCT showed a reduction of volumetric density and some architectural alterations of trabecular and cortical bone which are associated with vertebral and non-vertebral fractures, independently of areal BMD assessed by DXA.14 15

    The aim of our study was to investigate hand bone damage in RA by HR-pQCT, including for the first time the evaluation of microarchitecture impairment and the quantification of bone erosions, at the second and third MCPs (MCP2 and MCP3). Specifically, we determined the reproducibility of this technique and we compared density and architectural parameters from healthy controls and patients with RA. We also compared these parameters in patients with early and late RA. Moreover, we measured the volume of bone erosions in patients with early and late RA and compared it to a semiquantitative assessment using the Outcome Measures in Rheumatoid Arthritis Clinical Trials (OMERACT) scoring system applied to the HR-pQCT slices.

    Patients and methods

    Patients

    To assess the short-term reproducibility, 14 patients with late RA (aged 27–63 years) and 14 healthy subjects (aged 20–48 years) underwent 3 separate HR-pQCT scans of MCP2 and MCP3. Two measurements were repeated at one time point with repositioning and a third measurement was performed 1 week later.16

    To determine metacarpal bone damage in RA, we compared three groups of subjects: patients with early RA, patients with late RA and healthy subjects. Diagnosis of RA was based on the American College of Rheumatology criteria17 and expert opinion. All patients with RA (n=93) were recruited from the rheumatology department of the Edouard Herriot University Hospital in Lyon (France). The early RA group (n=36, mean age=43.6±11.9 years, mean disease duration=1.0±0.5 years) included patients with disease duration shorter than 2 years and with no erosion on hand radiographs. The late RA group (n=57, mean age=49.7±8.7 years, mean disease duration=8.9±5.5 years) was defined as patients with a disease duration superior to 3 years. Exclusion criteria were the absence of informed consent, pregnancy or breastfeeding, diseases of bone metabolism, current glucocorticoid treatment and functional score of Steinbrocker over 3.18 The healthy controls (n=43, mean age=47.0±8.6 years) were recruited from word of mouth recommendation, most of them being staff members in our hospital. The protocol was approved by an independent ethics committee and each subject gave written informed consent before participation.

    Clinical assessment

    Prior to undergoing evaluation by HR-pQCT, all patients were assessed for disease activity by a trained rheumatologist using the Disease Activity Score 28 (DAS 28).19 The latest erythrocyte sedimentation rate (ESR) and C reactive protein (CRP) were collected. Rheumatoid factor (RF) and anti-cyclic citrullinated peptide antibody (anti-CCP Ab) at diagnosis were also collected.20 Radiographic hand bone damage was assessed by the Sharp score modified by van der Heijde on hands only by the same reader using the latest conventional radiograph available for the patient, provided that this radiograph had been performed within the past 6 months. Treatment data including the disease-modifying antirheumatic drugs (DMARDs) and the use of TNF blockers were collected (table 1).

    View this table:
    Table 1

    Patients and healthy controls characteristics

    HR-QCT measurements

    Metacarpal bone damage was assessed at MCP2 and MCP3 using a HR-pQCT system (XtremeCT; Scanco Medical AG, Bassersdorf, Switzerland). This system enables the simultaneous acquisition of a stack of 110 parallel CT slices with an isotropic resolution (voxel size) of 82 µm.14 The effective dose was approximately 3 µSv per measurement with a measurement time of 2.8 min.

    During the examination, the right hand of the patient was immobilised in an anatomically ‘thermoformed’ hand splint and the arm was immobilised in a carbon cast. A single examination was performed for measurements of MCP2 and MCP3. An anteroposterior scout view was used to define the region of interest (ROI), which was chosen to start at the top of the MCP2, as shown in figure 1. The ROI was approximately 9 mm long in the axial direction, so that only a part of the MCP3 was captured during the scout. Consequently, when the effective ROI for MCP2 was about the complete 110 slices, we chose an effective ROI for MCP3 analysis of 73 slices corresponding to approximately 6 mm in the axial direction.

    Figure 1
    Figure 1

    Representative images of the second metacarpal head. A. Scout view demonstrating the measurement site (between dotted lines). B–E. Reconstruction of 20 high-resolution peripheral quantitative CT slices of (B) healthy control, (C) early rheumatoid arthritis (RA), (D) late RA, (E) early RA with an erosion (↓).

    vBMD and microarchitecture

    Two different density parameters were obtained: volumetric total (Dtot) and trabecular (Dtrab) BMDs in mg hydroxyapatite (HA)/cm3.The microarchitecture parameters obtained were: the cortical thickness (CTh, µm), the trabecular number (TbN, mm−1), thickness (TbTh, µm) and separation (TbSp, µm) and the distribution of trabecular separation (TbSpSD, µm) reflecting the heterogeneity of the trabecular network.21 TbN was measured whereas TbTh and TbSp were derived from Dtrab and TbN using standard methods from histomorphometry.22

    Bone erosion

    Erosions were defined as sharply marginated bone lesions with juxta-articular localisation with a cortical break seen in at least two adjacent slices. First, we determined the volume of erosions (using the entire volume of interest of MCP2 and MCP3) by manually defining the ROI of the erosion (V1) or excluding it (V2). The volume of erosions was then calculated as: Verosion = V1−V2; V = A× n × 0.082, where A is the mean area of slice (mm2), n is the total number of slice and 0.082 is the height of slice.14 Second, we performed a semiquantitative assessment of the erosion according to the definitions and principles of the OMERACT RA MRI scoring method developed for MRI.23 24 Briefly, the MCP2 and MCP3 were assigned a score by the percentage of bone volume involved (score 0–10, by 10% volume increments). Finally, the localisation and number of erosions of MCP2 and MCP3 detected with HR-pQCT were compared to standard radiographs.

    Statistical analysis

    For each subject in both reproducibility studies (for healthy subjects and patients with late RA), a coefficient of variation (CV) was calculated as the SD of the three repeated measurements divided by the subject mean. Further, the short-term precision errors were then calculated as the root mean square average of the precision errors for each of the subjects.

    vBMD and microarchitecture measurements were compared between patients with RA and healthy subjects with the Student t test or the Wilcoxon signed rank test, depending on the distribution of variables. Then, the differences between patients with early RA, patients with late RA and healthy subjects were assessed by analysis of variance or Kruskal–Wallis tests, followed, when significant, by a Student t test or a Wilcoxon signed rank test for pairwise comparisons. Moreover, these tests were age adjusted. For patients with RA, the relationships between the HR-pQCT parameters and the clinical evaluation were studied using Pearson or Spearman correlation (for age, disease duration, DAS 28, ESR, CRP and modified Sharp score). The association between HR-pQCT parameters and the presence of anti-CCP Ab or RF and treatments, was examined by the Student t test or the Wilcoxon signed rank test.

    For the bone erosions, we studied the correlation between the erosion volume by HR-pQCT and the semiquantitative assessment of bone erosion by OMERACT scoring system, using Spearman correlation.

    The statistical significance level was set at 0.05 and all statistical analyses were performed on SPSS software V.16.0 (SPSS, Chicago, Illinois, USA).

    Results

    Precision of HR-pQCT

    The CVs of volumetric density (total, trabecular and cortical) measurements ranged from 0.7% to 1.8% in patients with late RA and from 0.6% to 1.4% in healthy subjects. In comparison, the reproducibility of structural parameters (TbN, thickness, separation and distribution and CTh) was slightly poorer, with CVs ranging from 3.3% to 12.5% in patients with late RA and from 2.9% to 6.3% in healthy subjects (table 2).

    View this table:
    Table 2

    Reproducibility of density and structural measurements at MCP2 and MCP3

    Evaluation of bone density and microarchitecture

    At MCP2, four patients were not analysed (one patient with early RA and three patients with late RA) because their hand did not fit into the carbon cast excluding their MCP2 of the ROI. At MCP3, we evaluated only patients with at least 73 slices available, including 26 patients with early RA, 36 patients with late RA and 30 healthy controls.

    Comparison of patients with RA and healthy subjects

    We first compared the whole RA group to the healthy controls. As shown in table 3, patients with RA had significantly lower total vBMD at MCP2 (−6.2%, p<0.05) than healthy controls. They also had lower trabecular vBMD (−7.8% and −5.3%, p<0.05 at MCP2 and MCP3, respectively) and TbTh (−6.5% and −8.4%, p<0.05 at MCP2 and MCP3, respectively). Results were similar when adjusted for age.

    View this table:
    Table 3

    Comparison of the HR-pQCT parameters in patients with RA and healthy subjects

    Comparison of the patients with early RA, patients with late RA and healthy controls

    We observed a significant difference between controls, patients with early and late RA for trabecular vBMD at MCP2 and for TbTh at both MCPs (table 4). Trabecular vBMD was lower in late RA and in early RA compared to controls, although the difference did not reach statistical significance between early RA and controls (−8.7%, p<0.05 and −6.1%, p=0.057, respectively). Moreover, TbTh was reduced at MCP2 and MCP3 in patients with late RA (respectively −6.5% and −7.3%, p<0.05) and at MCP3 in patients with early RA (−8.4%, p<0.05) compared to healthy controls. For these parameters, patients with early and late RA did not differ significantly (data not shown). Moreover, the results were similar with age-adjusted comparison test.

    View this table:
    Table 4

    Comparison of the HR-pQCT parameters between the three groups: patients with early RA, patients with late RA and healthy subjects

    Association between density, microarchitecture measured at MCP2 and MCP3 and clinical data

    Despite the restricted analysis of 73 slices at MCP3, each parameter at MCP2 was highly correlated to its homologous at MCP3 (r=0.80–0.93) (data not shown).

    As shown in table 5, total vBMD, trabecular vBMD and TbTh were negatively correlated with the disease activity characterised by DAS 28, ESR and CRP. There was no significant correlation between HR-pQCT parameters, age, disease duration and Sharp score (except the correlation of TbSpSD with Sharp score at MCP2).

    View this table:
    Table 5

    Correlations of HR-pQCT parameters with clinical data

    We found no significant difference in HR-pQCT parameters between patients with or without anti-CCP Ab and across various treatments such as DMARDs and TNF blockers (data not shown).

    We did not use cortical measurements because cortical bone is particularly thin at MCPs leading to a partial effect volume, which has already been reported in a previous study at the distal radius.14

    Evaluation of bone erosions

    At MCP2, 7 patients with early RA and 18 patients with late RA exhibited bone erosions and at MCP3, erosions were detected in 6 patients with early RA and 15 patients with late RA. For the patients with early RA, the mean erosion volume was 5.3±3.0 mm3 at MCP2 and 4.4±3.7 mm3 at MCP3, ranging from 0.8 to 9.3 mm3. For patients with late RA, the mean erosion volume was 23.4±36.2 mm3 at MCP2 and 21.5±34.3 mm3 at MCP3, ranging from 0.7 to 148.7 mm3.

    Then, we assessed bone erosions using the semiquantitative OMERACT grading score, which was applied to the HR-pQCT slices. The OMERACT score ranged from 1 to 2 for patients with early RA and from 1 to 6 for patients with late RA.

    The erosion volume was strongly correlated to the assessment of bone erosion by the OMERACT score at MCP2 (r=0.80, p<0.05) and MCP3 (r=0.83, p<0.05).

    Finally, in patients with late RA, we compared the bone erosion observed with the HR-pQCT with those seen on hand radiographs (which were available for 14/18 patients with MCP2 HR-pQCT erosions and 12/15 patients with MCP3 HR-pQCT erosions). At MCP2 and MCP3, erosions were radiologically detected, respectively, in 5/14 patients and 3/12 patients. Consequently, HR-pQCT detected infraradiological erosions in 9/14 patients at MCP2 and 9/12 patients at MCP3. In patients with early RA, the lack of erosion on radiographs was an inclusion criteria.

    Discussion

    In this study, we evaluated for the first time the ability of HR-pQCT to assess volumetric BMD, microarchitecture of the MCP and the volume of metacarpal bone erosions in patients with RA.

    We defined a ROI on the MCPs and determined the reproducibility of the technique, which was quite similar at both MCPs with CVs less than 5.5% (except for TbSpSD). As expected, CVs were higher in patients with late RA than in healthy subjects, essentially for trabecular parameters. In fact, hand positioning and scan analysis were more difficult in patients with RA than in healthy subjects due to articular deformation. However, the reproducibility at the MCPs is comparable to the reproducibility obtained at the radius for which CVs are ranging from 1.5% to 4.5%.14 and the reproducibility of trabecular architecture measurements is also similar to, or better than, in previous studies using a lower resolution in vivo HR-pQCT system21 25 or high-resolution MRI with CVs ranging from 2% to 9%.26 27

    Total vBMD was significantly reduced in patients with RA compared to controls (−6.2%, p<0.05) at MCP2. This result is consistent with previous studies using DXA,28 dual x-ray radiogrammetry29 or quantitative phalangeal ultrasound30 31 showing a reduction of BMD either in periarticular regions30 or in the whole hand.32 33 But so far, there has not been any gold standard for the assessment of bone density or microarchitecture on the hand and no study previously compared these different techniques.

    The reduction of total vBMD can be related to the involvement of the trabecular compartment. Because of the thinness of cortical bone at MCPs leading to partial volume effect, cortical parameters were not rightfully interpretable. The reduction of TbTh appears even in early stages of the disease at MCP2 and MCP3 (−8.4% and −7.3%, respectively, p<0.05), suggesting that metacarpal bone loss may be induced by increased proinflammatory cytokines activating osteoclasts rather than mechanically by synovial proliferation.

    Then, we investigated the relationships between vBMD and microarchitecture impairment and several clinical parameters. We found a significant negative but moderate correlation of total vBMD, trabecular vBMD and TbTh with disease activity assessed by DAS 28 score, ESR and CRP. These results are in agreement with most of the DXA studies that described a significant correlation between hand bone loss and disease activity evaluated by CRP or ESR.33 34

    As HR-pQCT parameters were highly correlated at MCP2 and MCP3, the analysis of both MCPs provided similar information, despite the restricted analysis to 73 slices at MCP3. Consequently, we can consider that we can rightfully assess only the MCP3 when MCP2 cannot be analysed.

    To our knowledge, our study is the first to evaluate the quantitative measurement of erosion volumes in patients with RA by HR-pQCT. MRI has been previously used to measure erosion volumes in patients with RA at MCP and wrist joints showing a strong positive correlation between the total erosion volume and the total erosion score (r=0.93–0.94).35 We have demonstrated a good correlation between bone erosion volume and the OMERACT score (r=0.80 at MCP2 and r=0.83 at MCP3, p<0.05). HR-pQCT may represent an alternative tool to assess hand bone damage in RA, with shorter imaging time than MRI, but requires further investigations. We also need prospective studies to investigate the sensitivity to detect change of this technique, specifically the appearance of bone erosions. If the sensitivity to change is confirmed in prospective studies, HR-pQCT could become a useful tool in longitudinal observational studies and in clinical trials to detect early response to treatment.

    Our study has several limitations. The cortical bone is too thin to be correctly assessed at MCPs with the current evaluation method. The measurements essentially consist of trabecular bone. Our study has a cross-sectional design and the sample size is limited. Moreover, although the OMERACT erosion score has been used in CT,36 this score was applied for the first time to the HR-pQCT scan and has never been validated in HR-pQCT analysis. Therefore, our results need to be replicated.

    In conclusion, HR-pQCT at the MCPs seems to be a reproducible technique to assess vBMD and microarchitectorial parameters. Periarticular bone loss in RA may stem from predominantly trabecular bone loss. Volumetric BMD, trabecular thinning and erosions are detectable early in the course of RA using HR-pQCT. Therefore, the use of this technique might be interesting in clinical studies in RA.

    Acknowledgments

    The CeNGEPS (Centre National de Gestion des Essais de Produits de Santé) is acknowledged for financial support of AF-A's research fellowship.

    References

    1. ↵
      1. Rehman Q,
      2. Lane NE
      . Bone loss. Therapeutic approaches for preventing bone loss in inflammatory arthritis. Arthritis Res 2001;3:221–7.
      OpenUrlCrossRefPubMedWeb of Science
    2. ↵
      1. Hofbauer LC,
      2. Khosla S,
      3. Dunstan CR,
      4. et al
      . The roles of osteoprotegerin and osteoprotegerin ligand in the paracrine regulation of bone resorption. J Bone Miner Res 2000;15:2–12.
      OpenUrlCrossRefPubMedWeb of Science
    3. ↵
      1. Hofbauer LC,
      2. Lacey DL,
      3. Dunstan CR,
      4. et al
      . Interleukin-1beta and tumor necrosis factor-alpha, but not interleukin-6, stimulate osteoprotegerin ligand gene expression in human osteoblastic cells. Bone 1999;25:255–9.
      OpenUrlCrossRefPubMedWeb of Science
    4. ↵
      1. Fouque-Aubert A,
      2. Chapurlat R
      . Influence of RANKL inhibition on immune system in the treatment of bone diseases. Joint Bone Spine 2008;75:5–10.
      OpenUrlCrossRefPubMedWeb of Science
    5. ↵
      1. Fleming A,
      2. Crown JM,
      3. Corbett M
      . Early rheumatoid disease. I. Onset. Ann Rheum Dis 1976;35:357–60.
      OpenUrlAbstract/FREE Full Text
    6. ↵
      1. Solau-Gervais E,
      2. Legrand JL,
      3. Cortet B,
      4. et al
      . Magnetic resonance imaging of the hand for the diagnosis of rheumatoid arthritis in the absence of anti-cyclic citrullinated peptide antibodies: a prospective study. J Rheumatol 2006;33:1760–5.
      OpenUrlAbstract/FREE Full Text
    7. ↵
      1. van der Heijde DM
      . Radiographic imaging: the ‘gold standard’ for assessment of disease progression in rheumatoid arthritis. Rheumatology (Oxford) 2000;39(Suppl 1):9–16.
    8. ↵
      1. Plant MJ,
      2. Jones PW,
      3. Saklatvala J,
      4. et al
      . Patterns of radiological progression in early rheumatoid arthritis: results of an 8 year prospective study. J Rheumatol 1998;25:417–26.
      OpenUrlPubMedWeb of Science
    9. ↵
      1. Dixey J,
      2. Solymossy C,
      3. Young A
      . Is it possible to predict radiological damage in early rheumatoid arthritis (RA)? A report on the occurrence, progression, and prognostic factors of radiological erosions over the first 3 years in 866 patients from the Early RA Study (ERAS). J Rheumatol Suppl 2004;69:48–54.
      OpenUrlAbstract/FREE Full Text
    10. ↵
      1. Turner RA,
      2. Flint KP,
      3. Semble EL,
      4. et al
      . Clinical evaluation of radiographic progression in rheumatoid arthritis. Clin Exp Rheumatol 1990;8:583–6.
      OpenUrlPubMedWeb of Science
    11. ↵
      1. Brower AC
      . Use of the radiograph to measure the course of rheumatoid arthritis. The gold standard versus fool's gold. Arthritis Rheum 1990;33:316–24.
      OpenUrlPubMedWeb of Science
    12. ↵
      1. Wakefield RJ,
      2. Gibbon WW,
      3. Conaghan PG,
      4. et al
      . The value of sonography in the detection of bone erosions in patients with rheumatoid arthritis: a comparison with conventional radiography. Arthritis Rheum 2000;43:2762–70.
      OpenUrlCrossRefPubMedWeb of Science
    13. ↵
      1. Foley-Nolan D,
      2. Stack JP,
      3. Ryan M,
      4. et al
      . Magnetic resonance imaging in the assessment of rheumatoid arthritis--a comparison with plain film radiographs. Br J Rheumatol 1991;30:101–6.
      OpenUrlAbstract/FREE Full Text
    14. ↵
      1. Boutroy S,
      2. Bouxsein ML,
      3. Munoz F,
      4. et al
      . In vivo assessment of trabecular bone microarchitecture by high-resolution peripheral quantitative computed tomography. J Clin Endocrinol Metab 2005;90:6508–15.
      OpenUrlCrossRefPubMedWeb of Science
    15. ↵
      1. Sornay-Rendu E,
      2. Boutroy S,
      3. Munoz F,
      4. et al
      . Alterations of cortical and trabecular architecture are associated with fractures in postmenopausal women, partially independent of decreased BMD measured by DXA: the OFELY study. J Bone Miner Res 2007;22:425–33.
      OpenUrlCrossRefPubMedWeb of Science
    16. ↵
      1. Glüer CC,
      2. Blake G,
      3. Lu Y,
      4. et al
      . Accurate assessment of precision errors: how to measure the reproducibility of bone densitometry techniques. Osteoporos Int 1995;5:262–70.
      OpenUrlCrossRefPubMedWeb of Science
    17. ↵
      1. Arnett FC,
      2. Edworthy SM,
      3. Bloch DA,
      4. et al
      . The American Rheumatism Association 1987 revised criteria for the classification of rheumatoid arthritis. Arthritis Rheum 1988;31:315–24.
      OpenUrlCrossRefPubMedWeb of Science
    18. ↵
      1. Steinbrocker O,
      2. Traeger CH,
      3. Batterman RC
      . Therapeutic criteria in rheumatoid arthritis. J Am Med Assoc 1949;140:659–62.
      OpenUrlCrossRefPubMedWeb of Science
    19. ↵
      1. Prevoo ML,
      2. van ‘t Hof MA,
      3. Kuper HH,
      4. et al
      . Modified disease activity scores that include twenty-eight-joint counts. Development and validation in a prospective longitudinal study of patients with rheumatoid arthritis. Arthritis Rheum 1995;38:44–8.
      OpenUrlCrossRefPubMedWeb of Science
    20. ↵
      1. Marotte H,
      2. Pallot-Prades B,
      3. Grange L,
      4. et al
      . The shared epitope is a marker of severity associated with selection for, but not with response to, infliximab in a large rheumatoid arthritis population. Ann Rheum Dis 2006;65:342–7.
      OpenUrlAbstract/FREE Full Text
    21. ↵
      1. Laib A,
      2. Newitt DC,
      3. Lu Y,
      4. et al
      . New model-independent measures of trabecular bone structure applied to in vivo high-resolution MR images. Osteoporos Int 2002;13:130–6.
      OpenUrlCrossRefPubMedWeb of Science
    22. ↵
      1. Parfitt AM,
      2. Mathews CH,
      3. Villanueva AR,
      4. et al
      . Relationships between surface, volume, and thickness of iliac trabecular bone in aging and in osteoporosis. Implications for the microanatomic and cellular mechanisms of bone loss. J Clin Invest 1983;72:1396–409.
      OpenUrlCrossRefPubMedWeb of Science
    23. ↵
      1. Conaghan P,
      2. Bird P,
      3. Ejbjerg B,
      4. et al
      . The EULAR-OMERACT rheumatoid arthritis MRI reference image atlas: the metacarpophalangeal joints. Ann Rheum Dis 2005;64(Suppl 1):i11–21.
      OpenUrlAbstract/FREE Full Text
    24. ↵
      1. Østergaard M,
      2. Edmonds J,
      3. McQueen F,
      4. et al
      . An introduction to the EULAR-OMERACT rheumatoid arthritis MRI reference image atlas. Ann Rheum Dis 2005;64(Suppl 1):i3–7.
      OpenUrlAbstract/FREE Full Text
    25. ↵
      1. Müller R,
      2. Hildebrand T,
      3. Häuselmann HJ,
      4. et al
      . In vivo reproducibility of three-dimensional structural properties of noninvasive bone biopsies using 3D-pQCT. J Bone Miner Res 1996;11:1745–50.
      OpenUrlPubMedWeb of Science
    26. ↵
      1. Gomberg BR,
      2. Wehrli FW,
      3. Vasilic B,
      4. et al
      . Reproducibility and error sources of micro-MRI-based trabecular bone structural parameters of the distal radius and tibia. Bone 2004;35:266–76.
      OpenUrlCrossRefPubMedWeb of Science
    27. ↵
      1. Pothuaud L,
      2. Newitt DC,
      3. Lu Y,
      4. et al
      . In vivo application of 3D-line skeleton graph analysis (LSGA) technique with high-resolution magnetic resonance imaging of trabecular bone structure. Osteoporos Int 2004;15:411–19.
      OpenUrlCrossRefPubMedWeb of Science
    28. ↵
      1. Haugeberg G,
      2. Green MJ,
      3. Conaghan PG,
      4. et al
      . Hand bone densitometry: a more sensitive standard for the assessment of early bone damage in rheumatoid arthritis. Ann Rheum Dis 2007;66:1513–17.
      OpenUrlAbstract/FREE Full Text
    29. ↵
      1. Böttcher J,
      2. Pfeil A,
      3. Rosholm A,
      4. et al
      . Digital X-ray radiogrammetry combined with semiautomated analysis of joint space widths as a new diagnostic approach in rheumatoid arthritis: a cross-sectional and longitudinal study. Arthritis Rheum 2005;52:3850–9.
      OpenUrlCrossRefPubMedWeb of Science
    30. ↵
      1. Alenfeld FE,
      2. Diessel E,
      3. Brezger M,
      4. et al
      . Detailed analyses of periarticular osteoporosis in rheumatoid arthritis. Osteoporos Int 2000;11:400–7.
      OpenUrlCrossRefPubMedWeb of Science
    31. ↵
      1. Njeh CF,
      2. Boivin CM,
      3. Gough A,
      4. et al
      . Evaluation of finger ultrasound in the assessment of bone status with application of rheumatoid arthritis. Osteoporos Int 1999;9:82–90.
      OpenUrlCrossRefPubMedWeb of Science
    32. ↵
      1. Deodhar AA,
      2. Brabyn J,
      3. Jones PW,
      4. et al
      . Measurement of hand bone mineral content by dual energy x-ray absorptiometry: development of the method, and its application in normal volunteers and in patients with rheumatoid arthritis. Ann Rheum Dis 1994;53:685–90.
      OpenUrlAbstract/FREE Full Text
    33. ↵
      1. Peel NF,
      2. Spittlehouse AJ,
      3. Bax DE,
      4. et al
      . Bone mineral density of the hand in rheumatoid arthritis. Arthritis Rheum 1994;37:983–91.
      OpenUrlPubMedWeb of Science
    34. ↵
      1. Devlin J,
      2. Lilley J,
      3. Gough A,
      4. et al
      . Clinical associations of dual-energy X-ray absorptiometry measurement of hand bone mass in rheumatoid arthritis. Br J Rheumatol 1996;35:1256–62.
      OpenUrlAbstract/FREE Full Text
    35. ↵
      1. Bird P,
      2. Lassere M,
      3. Shnier R,
      4. et al
      . Computerized measurement of magnetic resonance imaging erosion volumes in patients with rheumatoid arthritis: a comparison with existing magnetic resonance imaging scoring systems and standard clinical outcome measures. Arthritis Rheum 2003;48:614–24.
      OpenUrlCrossRefPubMedWeb of Science
    36. ↵
      1. Møller Døhn U,
      2. Boonen A,
      3. Hetland ML,
      4. et al
      . Erosive progression is minimal, but erosion healing rare, in patients with rheumatoid arthritis treated with adalimumab. A 1 year investigator-initiated follow-up study using high-resolution computed tomography as the primary outcome measure. Ann Rheum Dis 2009;68:1585–90.
      OpenUrlAbstract/FREE Full Text

    Footnotes

    • Ethics approval This study was conducted with the approval of the CPP sudest II.

    • Provenance and peer review Not commissioned; externally peer reviewed.

    • Patient consent Obtained.

    • Competing interests None declared.