Rheumatol Int (2008) 29:47–51 DOI 10.1007/s00296-008-0638-0
O R I G I N A L A R T I CL E
Associations with subregional BMD-measurements in patients with rheumatoid arthritis H. Franck · J. Gottwalt
Received: 22 January 2008 / Accepted: 15 June 2008 / Published online: 3 July 2008 © Springer-Verlag 2008
Abstract Patients with rheumatoid arthritis (RA) have bone loss to various degrees at diVerent skeletal sites. The subregional bone mineral density (BMD) of the hand and the correlation of BMD to other regional bone losses, parameters of inXammation or bone resorption was evaluated in 421 patients with RA and controls. RA patients had signiWcantly (P < 0.01) lower BMD values in the carpus (0.405 § 0.004 g/cm2), metacarpal joint II (0.318 § 0.036 g/cm2) and metacarpal joint III (0.326 § 0.022 g/cm2) compared to controls. There was no diVerence in bone density at the lumbar spine or hip. SigniWcant (P < 0.001) correlations were found between BMD total of the hand, its subregions, the forearm and hip. Parameters of inXammation correlated signiWcantly (P < 0.001) with pyridinolines (r = 0.378), desoxypyridinolines (r = 0.183), forearm (r = ¡10, P < 0.05), MCP II (r = ¡0.190, P < 0.001), MCP III (r = 0.204, P < 0.001) and carpus (r = 0.191, P < 0.001). Keywords Subregional BMD · Rheumatoid arthritis · InXammation · Glucocorticoids · C-reactive protein
Introduction Patients with rheumatoid arthritis (RA) have bone loss to various degrees at diVerent skeletal sites. They possess several
H. Franck (&) Schwerpunktspraxis für Rheumatologie/Endokrinologie, Godesbergerallee 90, 52175 Bonn, Germany e-mail:
[email protected] J. Gottwalt Rheumazentrum Oberammergau, Hubertusstr. 40, 82487 Oberammergau, Germany
risk factors for bone loss including inXammatory activity and oral steroids to control inXammation. One of the predominant radiological changes in RA is a loss of peripheral bone. Consequently, juxtaarticular bone loss is also part of the deWnition of RA (American College of Rheumatology, ACR) criteria for classiWcation of RA [1]. It precedes the development of erosions. In most cases, the amount of disease related bone losses in hands of patients with RA are estimated by the evaluation of X-ray. However, osteopenia can not be identiWed radiologically until a considerable part of bone mineral has been lost. A quantiWcation for juxtaarticular bone loss should be a sensitive index for joint damage [2]. Furthermore, this local osteopenia could be faster or earlier than a generalized bone loss. Several studies have been undertaken to evaluate bone loss in patients with RA at various sites by diVerent methods [2, 3]. Bättcher and colleagues reported a signiWcant decrease in bone digital loss measured by digital X-ray radiogrammetry in patients with RA [4]. Some studies have used dual energy X-ray absorptiometry (DXA) for assessment of peripheral bone loss in small populations of RA [2, 5]. Few studies used DXA for detailed measurement around the Wnger joints in small groups [6]. The aim of the study was to evaluate whether subregional BMD bone mineral area density (BMD) of the hand is signiWcant diVerent in patients with RA and controls. Furthermore, if BMD correlates to other regional bone losses, parameters of inXammation or bone resorption.
Materials and methods A total of 421 patients with RA (357 women: mean age 58.40 § 12.87 years and 64 men: mean age 56.11 § 12.80
123
48
years) fulWlling the ACR criteria for classiWcation for RA [1] were recruited from the center of Rheumatology and included in the study. All patients gave informed consent and were examined by rheumatologists. The duration of disease related symptoms were 4.5 § 1.6 years for men and 4.8. § 1.4 years for women. 221 women (61%) and 45 men (69%) were treated with glucocorticoids. Patients with RA were treated with disease modiWying drugs mainly with methotrexate, sulfasalazin or chloroquine. Patients received no drugs aVecting signiWcantly bone metabolism except for glucocorticoids. A total of 98 healthy volunteers served as controls. 67 women (mean age 57.8 § 8.0 years) and 31 men (mean age 56.2 § 10.0 years) were free of clinical signs and symptoms of inXammatory joint diseases or bone diseases and were not taking drugs known to aVect bone metabolism. BMD of the lumbar vertebrae and diVerent sites of the right femoral region (femoral neck, Ward’s triangle, intertrochanter, total), the forearm and hand was measured by dual energy X-ray absorptiometry (DXA) using the Hologic QDR 4500 device. Scans were analyzed by the same operator using the software recommended by the manufacturers. Forearm measurements were performed on all subjects in sitting position. Short-term precision errors of BMD of both forearms were assessed in 12 normal volunteers. These measurements were also performed to assess mid-term precision errors in 85 healthy volunteers once at baseline and after 4 weeks. Every subject was tested approximately at the same time of each day. Each individual was scanned twice on the same day with repositioning between the scans. Every person was positioned according to the manufacturers suggestions. All scan analyses were performed by the same technician. The regions of interest (ROI) were placed as recommended by the manufacturers. The subregional BMD of the left hand was measured using the modiWed technique according to Peel et al. [2] and Harrison et al. [7]. Furthermore, measurement of the global BMD of the hand, the metacarpophalangeal joints (MCP) II, the MCP III and the carpus were performed as well. The coeYcient of variation (CV) was determined according to Slosman et al. [8]. Biochemical markers of bone turnover (Pyridinolin and Deoxypyridinolines) were quantiWed as reported [9]. Grip strength was determined using a vigorimeter.
Results In vitro precision was 0.9%, in vivo precision errors were less than 1.5% at the lumbar spine and less than 1.0% at the hip [10, 11]. Short-term precision errors of BMD total of the right and left forearms and the parts of the ultradistal radius were
123
Rheumatol Int (2008) 29:47–51
well below 1.2%, Mid-term precision errors of the left and right forearm were all below 1.5% and precision errors of BMD total were 1%. The precision errors of the subregional scans were 0.9–1.4% for short-term and 1.5–2.3% for mid-term. Patients with RA were associated with signiWcantly (P < 0.01) lower BMD values in the carpus (0.405 § 0.004 g/ cm2), metacarpal joint II (0.318 § 0.036 g/cm2) and metacarpal joint III (0.326 § 0.022 g/cm2) compared to controls (carpus: 0.424 § 0.005 g/cm2, MCP II (0.347 § 0.036 g/ cm2) and MCP III (0.340 § 0.037 g/cm2). A gender speciWc veriWcation of lower bone mass in RA was performed as well. Women with RA had signiWcantly (P < 0.01) lower BMD values in the carpus (0.401 § 0.001 g/cm2), MCP II (0.316 § 0.032 g/cm2) and MCP III (0.325 § 0.023 g/cm2) than controls (carpus: 0.421 § 0.021 g/cm2; MCP II: 0.344 § 0.037 g/cm2 and MCP III: 0.340 § 0.034 g/cm2). Similiar results were obtained for men with corresponding higher values. Men with RA also had signiWcantly (P < 0.01) lower BMD values in the carpus (0.427 § 0.01 g/cm2), MCP II (0.330 § 0.062 g/cm2) and MCP III (0.325 § 0.023 g/cm2) compared to controls (carpus: 0.490 § 0.022 g/cm2; MCP II: 0.402 § 0.047 g/cm2 and MCP III: 0.380 § 0.036 g/cm2). There were no diVerences in bone density at the lumbar spine (women: 0.921 § 0.157 g/cm2, men: 0.941 § 0.144 g/cm2) and hip (women: 0.811 § 0.140 g/cm2, men: 0.895 § 0.143 g/cm2) in patients with RA compared to controls (lumbar spine women: 0.930 § 0.146 g/cm2; men: 0.960 § 0.146 g/cm2; hip women 0.820 § 0.122 g/cm2; men: 0.899 § 0.144 g/cm2). However, we could demonstrate a close relationship between BMD total of the hand, its subregions, the forearm and hip of patients with RA (Table 1). To further characterize the nature of bone loss, additional analyses of inXammation, treatment with glucocorticoids and grip strength in relation to bone density were performed. Treatment with glucocorticoids had a signiWcant (P < 0.05) eVect on bone density in most areas in patients with RA versus no treatment with glucocorticoids (Table 2). In female RA patients treated with corticosteroids BMD was signiWcantly (P < 0.05) reduced in the hip (0.787 § 0.142 vs. 0.848 § 0.129 g/cm2), the lumbar spine (0.895 § 0.149 vs. 0.963 § 0.162 g/cm2), BMD total of the hand (0.327 § 0.05 vs. 0.340 § 0.05 g/cm2) MCP II (0.306 § 0.060 vs. 0.323 § 0.058 g/cm2) and MCP III (0.296 § 0.063 vs. 0.317 § 0.064 g/cm2). The BMD total of the forearm (0.502 § 0.093 vs. 0.524 § 0.082 g/cm2) was just not signiWcant in the steroid treated group. Table 3. Systemic inXammation had an signiWcant impact on the BMD of our patients. Patients with inXammatory activity (elevated C-reactive protein, CRP > 5 mg/l) had signiW-
Rheumatol Int (2008) 29:47–51
49
Table 1 Correlation coeYcients between BMD of subregional areas of the hand with BMD at diVerents sites (forearm, lumbar spine and hip) BMD
Lumbar spine (total)
Hip (total)
Forearm (total)
UD forearm
MCP II
MCP III
Hand (global)
Lumbar spine (total)
1.0 **
0.585**
0.544**
0.535**
0.393**
0.421**
0.454**
Hip (total)
0.585**
1.0**
0.623**
0.608**
0.522**
0.566**
0.,491**
Forearm (total)
0.544**
0.623**
1.0**
0.836**
0.663**
0.691**
0.577**
UD Forearm
0.535**
0.608**
0.836**
1.0**
0.625**
0.666**
0.632**
MCP II
0.393**
0.522**
0.663**
0.625**
1.0**
0.844**
0.630**
MCP III
0.421**
0.566**
0.691**
0.666**
0.844**
1.0**
0.799**
Hand (global)
0.454**
0.491**
0.577**
0.632**
0.630**
0.799**
1.0**
* P < 0.05, **P < 0.001 MCP metacarpophalangeal joints Table 2 BMD(g/cm2) § SD (standard deviation) in patients with and without treatment of glucocorticoids Areas of BMD
BMD of patients on glucocorticoids
BMD of patients without glucocorticoids
BMD total of the hip
0.786 § 0.142*
0.848 § 0.128
Lumbar spine
0.895 § 0.149*
0.963 § 0.161
Mid forearm
0.510 § 0.100*
0.535 § 0.090
Ultradistal forearm
0.374 § 0.078*
0.393 § 0.072
MCP II
0.301 § 0.060*
0,323 § 0.05
MCP III
0.296 § 0.060*
0.317 § 0.06
(r = 0.30, P < 0.001), MCP II (r = 0.27, P < 0.001), MCP III (r = 0.333, P < 0.005), BMD total of the forearm (r = 0.28, P < 0.005), forearm distal 1/3 (r = 0.29, P < 0.005), and BMD total of the hip (r = 0.25, P = 0.005). No signiWcant correlation was found between grip strength and BMD total of the lumbar spine. Equivalent correlations were detected between pain (visual analogue scale) and BMD of the sub regional areas, forearm and hip. Grip strength also correlated signiWcantly with CRP (r = ¡0.224, P < 0.005) and VAS (r = ¡0.318, P < 0.001).
* P < 0.05, MCP metacarpophalangeal joints
cantly lower BMD total of the hip (0.753 § 0.144 vs. 0.825 § 0.138 g/cm2), BMD of the ultra distal forearm (0.366 § 0.090 vs. 0.390 § 0.070 g/cm2), global hand (0.319 § 0.060 vs. 0.340 § 0.05 g/cm2), metacarpal joint II (0.286 § 0.050 vs. 0.319 § 0.050 g/cm2) and metacarpal joint III (0.274 § 0.05 vs. 0.313 § 0.050 g/cm2). No diVerences in BMD total of lumbar spine were detected. Parameters of inXammations (CRP) correlated signiWcantly (P < 0.001) with bone resorption markers such as pyridinolines (r = 0.378), deoxypyridinolines (r = 0.183) and bone density, BMD total of the hip (r = ¡0.142), forearm (r = ¡0.10, P < 0.05), MCP II (r = ¡0.190, P < 0.001), MCP III (r = 0.204, P < 0.001) and the carpus (r = 0.0191, P < 0.001). Grip strength also had an inXuence on BMD especially the appendicular bone. RA patients showed signiWcant correlations between grip strength and BMD total of the hand
Discussion BMD This study has demonstrated a signiWcant reduction in appendicular bone density in patients with RA using DXA measurement tools. The measurement of bone loss in patients with RA at diVerent sites is an important diagnostic feature for detecting osteoporosis. This is especially true for peripheral bone loss, which is related to speciWc variables of the disease. A lot of diVerent tools have been tested to determine hand bone mass in RA (radiogrammmetry, quantitative ultrasound etc.). However, DXA is characterized by good precision [2] allowing to assess BMD in diVerent study populations.
Table 3 Correlation coeYcients between bone resorptions, inXammatory parameters and BMD at diVerents sites CRP
ESR
BMD hip (total)
BMD UD forearm
BMD MCP II
BMD MCP III
BMD hand (global)
Desoxypyridinolines
0.183**
0.288**
¡0.133*
¡0.172**
¡0.182**
¡0.168**
¡0.217**
Pyridinolines
0.375**
0.353**
¡0.238**
¡0.202**
¡0.236**
¡0.212**
¡0.194**
CRP
1.0**
0.622**
¡0.142**
¡0.10*
¡0.187**
¡0.204**
¡0.194**
ESR
0.622**
1.0**
¡0.097
¡0.126*
¡0.234**
¡0.237**
¡0.218**
* P < 0.05, **P < 0.001 MCP metacarpophalangeal joints
123
50
Our precision errors measuring bone density in the hand were in the same range as reported before [6]. It allowed longitudinal studies as well, which have already been performed in smaller patient groups. Consequently, subregional measurements of the hand in patients with RA is a valuable tool assessing lower regional bone mass. This is important since bone density in RA is correlated with radiological changes [3]. In early arthritis, some authors [12] did not Wnd any diVerences in BMD of lumbar spine or hip but signiWcant lower values in BMD of the whole hand using Dual Energy Absorptiometry (DXA). Haugeberg et al. [12] could report that hand BMD is a sensitive tool measuring bone damage in RA. Our results demonstrated that the standard measurement of BMD (in the lumbar spine and hip) does not detect and reXect bone loss to such an extent. This has been conWrmed by several authors [13, 14]. Therefore, only weak correlations between BMD total of the spine and hand were found. In our large patient group, we could not conWrm a signiWcant decrease in BMD of the total hip as reported by Haugeberg et al. [13], and no loss could be shown in the lumbar spine as has been reported in some cases [12, 13]. Shibuya et al. [14] also showed a signiWcantly lower BMD in patients with RA in all sites except for the lumbar vertebra. However, in this study the peripheral BMD was measured by peripheral quantitative computed tomography. In contrast, a change of BMD of the lumbar spine could be observed in patients receiving anti-TNF-alpha antibody in a prospective open label pilot study [15] but an increase was not conWrmed in the Best study [16]. Other studies also showed asteopenia in the calcaneus [13] or the distal forearm in patients with RA [17]. Iwamoto et al. [17] reported a lower BMD of the distal forearm in a large patient group with RA showing correlations with ESR. However, other bone measurements of the hand or lumbar spine were not performed, since this DXA machine by osteometer was especially designed for forearm measurements.
Rheumatol Int (2008) 29:47–51
Similiarly, Sivas et al. [22] presented signiWcant correlations of CRP and BMD of the forearm with lower BMD of the right forearm and the right hip of the patient group. Haugeberg et al. [13] reported that CRP and rheumatoid factor were signiWcantly correlated with changes in BMD hand in early arthritis. Steroids The eVects of glucocorticoids on bone density are discussed controversially. A deleterious eVect of glucocorticoids could not be proven in a small study group [23] measuring appendicular bone mass with peripheral computed tomography. In contrast, Kelly et al. [24] showed that peripheral bone loss (forearm) depends on the use of oral steroids. We could demonstrate in our large patient group, that BMD was lower in almost all measured sites in women treated with glucocorticoids. Grip strength was signiWcantly correlated with both BMD of the metacarpals and the forearm implying that MCP joints bone density is related to hand function and also to the degree of disease activity. Towheed et al. [25] could proof a signiWcant correlation between grip strength and peripheral bone density as well.
Conclusion In this study we highlighted important changes in appendicular bone mass and its relationship to inXammatory activity, grip strength and treatment with glucocorticoids in patients with RA. Hand and subregional bone density measured by DXA is an accurate and reproducible tool of juxtaarticular bone loss. It detects bone loss in patients with RA even if standard DXA measurement of the spine and hip are normal [26, 27], especially in conditions of increased inXammatory activities or treatment with glucocorticoid. Last but not least peripheral DXA measurement should be accessible to patients, since it is associated with little exposure of radiation.
CRP InXammatory activity has an important inXuence on bone turnover or bone loss in RA [18–20]. Momohara et al. [21] showed a signiWcant correlation with bone resorption markers and ESR or CRP in patients with RA. However, it did not correlate with BMD in particular, and peripheral BMD of the hand was not measured. In contrast, we could demonstrate a lower BMD in patients with higher inXammatory activity and signifcant correlations of CRP with both BMD of hip, forearm and MCP joints and bone resorption markers.
123
References 1. Arnett FC, Edworthy SM, Bloch DA, McShane DJ, Fries JF, Cooper NS et al (1988) The American Rheumatism Association 1987 revised criteria for the classiWcation of rheumatoid arthritis. Arthritis Rheum 31:315–324. doi:10.1002/art.1780310302 2. Peel NF, Spittlehouse AJ, Bax DE, Eastell R (1994) Bone mineral density of the hand in rheumatoid arthritis. Arthritis Rheum 37:983–991. doi:10.1002/art.1780370702 3. Berglin E, Lorentzon R, Nordmark L, Nilsson-Sojka B, Rantapää Dahlqvist S (2003) Predictors of radiological progression and changes in hand bone density in early rheumatoid arthritis. Rheu-
Rheumatol Int (2008) 29:47–51
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
matology (Oxford) 42:268–275. doi:10.1093/rheumatology/ keg077 Böttcher J, Pfeil A, Heinrich B, Lehmann G, Petrovitch A, Hansch A et al (2005) Digital radiogrammetry as a new diagnostic tool for estimation of disease-related osteoporosis in rheumatoid arthritis compared with pQCT. Rheumatol Int 25:457–464. doi:10.1007/ s00296-004-0560-z Deodhar AA, Brabyn J, Jones PW, Davis MJ, Woolf AD (1994) 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 53:685–690 Castañeda S, González-Alvaro I, Rodríguez-Salvanés F, Quintana ML, LaVon A, García-Vadillo JA (2007) Reproducibility of metacarpophalangeal bone mass measurements obtained bydual-energy X-ray absorptiometry in healthy volunteers and patients with early arthritis. J Clin Densitom 10:298–305. doi:10.1016/ j.jocd.2007.04.003 Harrison BJ, Hutchinson CE, Adams J, Bruce IN, Herrick AL (2002) Assessing periarticular bone mineral density in patients with early psoriatic arthritis or rheumatoid arthritis. Ann Rheum Dis 61:1007–1011. doi:10.1136/ard.61.11.1007 Slosman DO, Rissoli R, Donath A, Bonjour JP (1990) Vertebral bone mineral density measured laterally by dual-energy X-ray absorptiometry. Osteoporos Int 1:23–29. doi:10.1007/BF01880412 Franck H, Meurer T, Hofbauer LC (2004) Evaluation of bone mineral density, hormones, biochemical markers of bone metabolism, and osteoprotegerin serum levels in patients with ankylosing spondylitis. J Rheumatol 31:2236–2241 Franck H, Munz M, Scherrer M (1995) Evaluation of dual-energy X-ray absorptiometry bone mineral measurement - comparison of a single-beam and fan-beam design: The eVect of osteophytic calciWcation on spine bone mineral density. Calcif Tissue Int 56:192– 195. doi:10.1007/BF00298608 Franck H, Munz M, Scherrer M (1997) Bone mineral density of opposing hips using dual energy X-ray absorptiometry in singlebeam and fan-beam design. Calcif Tissue Int 61:445–447. doi:10.1007/s002239900365 Haugeberg G, Green MJ, Conaghan PG, Quinn M, WakeWeld R, Proudman SM et al (2007) Hand bone densitometry: a more sensitive standard for the assessment of early bone damage in rheumatoid arthritis. Ann Rheum Dis 66:1513–1517. doi:10.1136/ard. 2006.067652 Haugeberg G, Ãrstavik RE, Uhlig T, Falch JA, Halse JI, Kvien TK (2003) Comparison of ultrasound and X-ray absorptiometry bone measurements in a case control study of female rheumatoid arthritis patients and randomly selected subjects in the population. Osteoporos Int 14:312–319. doi:10.1007/s00198-002-1365-0 Shibuya K, Hagino H, Morio Y, Teshima R (2002) Cross-sectional and longitudinal study of osteoporosis in patients with rheumatoid arthritis. Clin Rheumatol 21:150–158. doi:10.1007/s10067-0028274-7 Lange U, Teichmann J, Müller-Ladner U, Strunk J (2005) Increase in bone mineral density of patients with rheumatoid arthritis treat-
51
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
ed with anti-TNF-alpha antibody: a prospective open-label pilot study. Rheumatology (Oxford) 44:1546–1548. doi:10.1093/rheumatology/kei082 Güler-Yüksel M, Bijsterbosch J, Goekoop-Ruiterman YP, de Vries-Bouwstra JK, Hulsmans HM, de Beus WM, et al. (2007) Changes in bone mineral density in patients with recent onset, active rheumatoid arthritis. Ann Rheum Dis (Aug) 17 Iwamoto J, Takeda T, Ichimura S (2002) Forearm bone mineral density in postmenopausal women with rheumatoid arthritis. Calcif Tissue Int 70:1–8. doi:10.1007/s00223-001-1054-6 Franck H, Ittel TH, Tasch O, Herborn G, Rau R (1994) Osteocalcin in patients with rheumatoid arthritis. A one-year followup study. J Rheumatol 21:1256–1259 Molenaar ET, Lems WF, Dijkmans BA, de Koning MH, van de Stadt RJ, Voskuyl AE (2000) Levels of markers of bone resorption are moderately increased in patients with inactive rheumatoid arthritis. Rheumatology (Oxford) 39:742–744. doi:10.1093/rheumatology/39.7.742 Oelzner P, Franke S, Müller A, Hein G, Stein G (1999) Relationship between soluble markers of immune activation and bone turnover in post-menopausal women with rheumatoid arthritis. Rheumatology (Oxford) 38:841–847. doi:10.1093/rheumatology/ 38.9.841 Momohara S, Okamoto H, Yago T, Furuya T, Nanke Y, Kotake S et al (2005) The study of bone mineral density and bone turnover markers in postmenopausal women with active rheumatoid arthritis. Mod Rheumatol 15:410–414. doi:10.1007/s10165-005-0435-5 Sivas F, Barösa N, Onder M, Ozoran K (2006) The relation between joint erosion and generalized osteoporosis and disease activity in patients with rheumatoid arthritis. Rheumatol Int 26:896–899. doi:10.1007/s00296-006-0104-9 Martin JC, Munro R, Campbell MK, Reid DM (1997) EVects of disease and corticosteroids on appendicular bone mass in postmenopausal women with rheumatoid arthritis: comparison with axial measurements. Br J Rheumatol 36:43–49. doi:10.1093/rheumatology/36.1.43 Kelly C, Bartholomew P, Lapworth A, Basu A, Hamilton J, Heycock C (2002) Peripheral bone density in patients with rheumatoid arthritis and factors which inXuence it. Eur J Intern Med 13:423. doi:10.1016/S0953-6205(02)00129-2 Towheed TE, Brouillard D, Yendt E, Anastassiades T (1995) Osteoporosis in rheumatoid arthritis: Wndings in the metacarpal, spine, and hip and a study of the determinants of both localized and generalized osteopenia. J Rheumatol 22:440–443 Haugen IK, Slatkowsky-Christensen B, Orstavik R, Kvien TK (2007) Bone mineral density in patients with hand osteoarthritis compared to population controls and patients with rheumatoid arthritis. Ann Rheum Dis 66:1594–1598. doi:10.1136/ard.2006. 068940 Güler-Yüksel M, Bijsterbosch J, Goekoop-Ruiterman YP, de Vries-Bouwstra JK, Ronday HK, Peeters AJ et al (2007) Bone mineral density in patients with recently diagnosed, active rheumatoid Arthritis. Ann Rheum Dis 66:1508–1512. doi:10.1136/ard. 2007.070839
123