Bone mineral density in systemic lupus erythematosus and its relation to age at disease onset, plasmatic estradiol and immunosuppressive therapy

Joint Bone Spine 70 (2003) 40–45 www.elsevier.com/locate/bonsoi

Original article

Bone mineral density in systemic lupus erythematosus and its relation to age at disease onset, plasmatic estradiol and immunosuppressive therapy > Ibsen Bellini Coimbra, Lilian Tereza Lavras Costallat * Rheumatology Unit, Department of Internal Medicine, Faculty of Medical Sciences, State University of Campinas, Brazil Received 1 June 2001; accepted 4 April 2002

Abstract Objective. – The aim of this paper was to evaluate bone mineral density (BMD) in patients with systemic lupus erythematosus (SLE), to determine the role of corticosteroids and cytotoxic drugs and to assess estrogen effect on BMD in SLE. Patients and methods. – BMD (DEXA) at lumbar vertebrae (L2–L4) and at femoral neck was performed in 60 pre-menopausal SLE patients and in 64 controls. Estradiol level was measured in all the individuals. Age, age at disease onset, body mass index (BMI), time of disease, disease activity (SLEDAI), prednisone dose at the evaluation, total cumulative and cumulative prednisone dose in the last year and cytotoxic drugs were assessed. Results. – The mean plasmatic estradiol was 175.9 pg/ml in patients and 149.9 in controls. BMD was inferior in patients than that in controls (P < 0.0001). The mean current, cumulative and previous year prednisone doses were, respectively, 19.17 mg/d, 28.78 g and 5.33 g. There was no association between corticosteroids or cytotoxic drug used and low bone mass. The serum concentration of estradiol did not influence the bone mass. The BMI and age at disease onset exhibited an influence on BMD at L2. Conclusions. – BMD was significantly lower in SLE patients but not related to CS (Corticosteroids)or other drugs; the estradiol in these patients had no effect on BMD. Low BMI interacting with early onset of disease might influence the probability of loss of bone mass. © 2002 Éditions scientifiques et médicales Elsevier SAS. All rights reserved. Keywords: Systemic lupus erythematosus; Bone mineral density; Bone loss; Estradiol; Corticosteroids; Cytotoxic drugs

1. Introduction Systemic lupus erythematosus (SLE) is a chronic, multisystem disease, which affects mainly women. The survival of SLE patients has increased dramatically in the last decade, from less than 50% after 4 years in 1984 to more than 80% after 10 years nowadays [1–2]. We ought to be aware of the long-term survivors’ morbidity as it happened [3–11]. Osteoporosis contributes to the morbidity of long-term survivors. The incidence of osteoporosis in SLE patients remains unclear, in spite of the fact that several authors were interested in it, mainly in relation to the risk of glucocorticosteroid induced osteoporosis, a well known osteopenia related risk > This paper was supported in part by the Fundo Remanescente of the Brazilian Society of Rheumatology. * Corresponding author. Department of Medical Clinic, FCM UNICAMP Cidade Universitaria Zeferino Vaz CEP 13083-970 Campinas, SP, Brazil. E-mail address: [email protected] (L.T.L. Costallat).

© 2002 Éditions scientifiques et médicales Elsevier SAS. All rights reserved. PII: S 1 2 9 7 - 3 1 9 X ( 0 2 ) 0 0 0 0 9 - X

factor [3]. A meta-analysis of the bone mineral density (BMD) of 61 SLE patients who never received corticosteroid treatment revealed a slight loss of bone mineral content in the spine, hip and forearm, suggesting that osteopenia might be disease related [12]. Other studies have also suggested an association between low BMD and corticosteroid usage [13–15]. Despite the extensive research on osteoporosis and SLE, many questions remain to be answered, including the factors involved in the development of osteoporosis. High doses of corticosteroid have been reported to be a cause by some [13–15], whereas others [3,5–11] have found no such relationship. Indeed, osteopenia or osteoporosis has been observed in SLE patients who have not taken corticosteroids, suggesting that other factors are involved in the mechanisms of bone loss in these individuals [12]. Elements of the lupus inflammatory process itself have been suggested to play a role in osteoporosis [8]. In addition, high levels of estrogens in women with SLE may offer some protection against os-

I.B. Coimbra, L.T.L. Costallat / Joint Bone Spine 70 (2003) 40–45

teoporosis [5,8]. In this study, we evaluated the BMD in pre-menopausal women, to determine the role of corticosteroids and cytotoxic drugs on BMD and to assess the effect of plasmatic estradiol on BMD in SLE patients.

2. Patients and methods 2.1. Patients Sixty outpatient SLE patients, all female, unselected for any special features, diagnosed according to the ACR classification criteria [16], were analyzed in a cross-sectional study. The subjects were more than 16 years of age and were attended in the Rheumatology Unit, at the university hospital of UNICAMP. All the subjects gave their informed consent and the study was approved by the faculty’s Medical Ethics Committee. Patients underwent a standard interview, physical examination and a review of their medical records. Information about their age, age at disease onset, duration of disease, weight, height, body mass index (BMI), dietary calcium intake, and history of smoking were obtained, by the same investigator. The chronic use of medication was recorded, particularly cyclophosphamide, azathioprine and the total cumulative dose of corticosteroids, mean dose of corticosteroid per year of disease, current corticosteroid dose, and the total dose of corticosteroids in the year before evaluation. All patients were successfully followed up. Disease activity was measured using the SLEDAI [17]. 2.2. Exclusion criteria Irregularity of menses (non-cyclic unpredictable menses associated with prolonged periods of amenorrhoea) or serum foliculi stimulating hormone (FSH) levels above 14 mIU/ml (laboratory menopause), use of anticonvulsants, diuretics, barbiturics, hormone replacement therapy, sodium fluoride, calcium supplements, vitamin D supplement, calcitonin, oral anticoagulants and alcohol abuse. Patients suffering from conditions that related to secondary osteoporosis such as renal disease, intestinal disabsorptive disease, hepatic disease, thyroid or parathyroid diseases, prolonged immobilization (more than 6 weeks), or those suffering from other metabolic, congenital bone or other diseases associated with an acute phase response (infection, connective tissue disease other than SLE and malignant disease) were also excluded. 2.3. Estradiol (E2) Evaluation of plasmatic estradiol levels was carried out by radioimmunoassay (ICN-Biomedicals, Inc.) and the values were expressed in pg/ml. All the samples were always obtained in the follicular phase for all the patients. Patients with E2 levels <40 pg/ml were considered as group 1, those with E2 ranging between 40 and 125 pg/ml, group 2, those with E2 between 125 and 200 pg/ml, group 3 and those with E2 >200 pg/ml, group 4.

41

2.4. Bone mineral density Evaluation of BMD in g/cm2 of the left femoral neck and lumbar spine was carried out by dual energy X-ray absorptionmetry (DEXA) using a Lunar DPX densitometer with Lunar software version 3.6 Y (Lunar Radiation Corp., Madison, WI, USA). The physical and mathematical principles involved have been described [18]. The lumbar spine was measured from L1 to L4 and the mean lumbar BMD (L2–L4) was calculated. The results were expressed in g/cm2 and additionally in terms of Z-score and T-score. Z-score compared the patient’s observed BMD with an expected value for a female of the same age. T-score compared the patient’s observed BMD with an expected value for a young adult female aged 25–29 years (t-score). A T-score lower than –2.5 SD below the young adult mean was defined as osteoporosis according to the Word Health Association (WHO) [19]. The patients with standardized BMD measurements between 1 and 2.5 SD below the reference levels were considered as having osteopenia. 2.5. Control group Sixty-four healthy women followed up in the Gynecology Unit, Clinical Hospital of State University of Campinas, had agreed to participate as normal volunteers in this study. Inclusion criteria were age, ranging between 20 and 45 years, regular menstrual cycles, absence of any bone mass metabolism modifying the situation as described above for the patients. The exclusion criteria were the same as those in the patient group. Plasmatic estradiol levels and BMD by DEXA were checked for all. 2.6. Statistical analysis All data were expressed as mean ± standard error of the mean (SEM). Initial comparisons between groups were performed using multi-way chi-squared or Fisher’s exact probability test. Student’s t-test for comparison of normal distribution quantitative data was used. Wilcoxon’s test was performed for comparison of abnormal distribution quantitative data. Statistical significance was defined as P ≤ 0.05. Logistic regression analyses, univariate and multiple, using the stepwise variables (isolated or grouped in according to the clinical importance) selection were performed to evaluate the probable low bone mass associated variables either in femur or lumbar spine. The analysis of data was performed using the statistical software package SAS (Statistical analysis system) version 6.12. 3. Results 3.1. Characterization of the groups One hundred and twenty-four eligible women were included in the study, 60 SLE patients and 64 healthy ones as

42

I.B. Coimbra, L.T.L. Costallat / Joint Bone Spine 70 (2003) 40–45

the control group. Among the patients, 51 (85%) were white, four (6.7%) black, one (1.7%) Asian and four mestizo (6.7%). In the control group, 49 (76.6%) were white, five (7.8%) black and three mestizo. There were no statistically significant differences in relation to race between the groups (P = 0.308). The mean age in the control group was 31.1 ± 7.2 years, and in the patient group, the mean was 32.8 ± 8.6 years. There were no statistically significant differences in age between the groups (P = 0.2304). The groups were also weight and height matched. The BMI was calculated using the relation BMI = weight/square height. The mean BMI was 25.1 ± 4.8 kg/m2 among patients and 25.2 ± 4.6 kg/m2 among the healthy women (P = 0.9208). The patient age at disease onset ranged between 12 and 52 years (mean age was 26.9 ± 9.4 years). The time of disease ranged from 12 up to 240 months, with a mean of 78.1 ± 52.3 months. The main clinical manifestation among the patients during the disease progression is shown in Table 1. 3.2. Corticosteroids and other cytotoxic drugs In relation to corticosteroid use, the current dose (on the day of BMD examination), the cumulative dose in the year prior to evaluation and the total cumulative dose were calculated. Among the 60 SLE patients, 59 used corticosteroids, 35 of whom concomitantly used them with antimalarials, nine with cyclophosphamide and five with azathioprine. Only one patient did not take corticosteroids. The mean doses of corticosteroids are shown in Table 2. 3.3. Smoking habits and disease activity Five patients were smokers (8.3%). Twenty-five had active disease by the SLEDAI. Table 1 Clinical features of the patients during the disease Fever, weight loss Arthritis Malar or discoid rash Alopecia Photosensitivity Oral ulcers Nephritis Serositis Seizures Psychosis Raynaud’s phenomenon Thromboembolic events Hematologic alterations

34 53 52 29 33 08 24 20 03 03 24 02 26

3.4. Plasmatic estradiol levels Fifty-four SLE patients were included in this evaluation and the mean of plasmatic estradiol was 175 ± 168 pg/ml (ranging from 10 up to 805 pg/ml). In the control groups, the mean was 149 ± 88 pg/ml (ranged from 32 up to 380). There was no statistical difference between the groups (P = 0.3133), even when the analysis was performed in each of the four groups according to its levels. 3.5. BMD measurements and T-score The mean of BMD in g/cm2 as well as the T-score by region (lumbar, spine and femur) were obtained from the patients and the control group. The data can be seen in Tables 3 and 4, respectively. The mean BMD was significantly inferior among the SLE women when compared to the healthy ones in all the analyzed regions. The means of Z-score were compared in both the groups and it was observed that they were significantly lower in SLE patients than in the healthy women (P = 0.0001) in all the analyzed regions. 3.6. Influence of the studied variables on the occurrence of low bone mass (T-score L1) The influence of some variables was analyzed by logistic regression, considering as a response the presence or absence of low bone mass. The following variables were included in the analysis: BMD, age, age at disease onset, time of disease, corticosteroid use (current, in the past few years and the total cumulative dose), activity of disease, plasmatic levels of estradiol, smoking history, use of cyclophosphamide and azathioprine. Some variables, even those not exhibiting an influence on low bone mass in linear analysis, but with a well-recognized clinical role in the bone mass were analyzed Table 3 BMD (g/cm2) in femur and lumbar spine

56.6% 88.3% 86.5% 48.3% 55% 13.3% 40% 33.3% 5% 5% 40% 3.3% 43.3%

Femur neck Ward Trochanter L2–L4 N

Patient M (g/cm2)

Controls M (g/cm2)

P*

0.950 ± 0.139 0.855 ± 0.160 0.760 ± 0.132 1.098 ± 0.178 60

1.036 ± 0.115 0.967 ± 0.132 0.842 ± 0.1 10 1.209 ± 0.137 64

0.0003 0.0001 0.0003 0.0002

* Student’s t-test. Table 4 Mean of T-score in femur and lumbar spine Patients

Table 2 SLE patients-doses of corticosteroids

Current (mg/d) Last year (g) Cumulative (g)

N

Mean

SD

Min

Max

58 57 59

19.83 5.61 29.27

18.35 3.93 17.60

5 0.60 1.20

80.00 15.80 76.70

Local Femur neck Ward’s triangle Trochanter L2–L4 N

Mean –0.204 –0.301 –0.211 –0.826 60

* Student’s t-test.

Controls SD 1.155 1.240 1.217 1.478

Mean 0.472 0.473 0.478 0.146 64

DP 0.930 0.972 0.987 1.123

P* 0.0005 0.0002 0.0008 0.0001

I.B. Coimbra, L.T.L. Costallat / Joint Bone Spine 70 (2003) 40–45

in groups, such as the estrogen levels (E2). Some variable interactions were also analyzed including interaction between BMI and age at disease onset (IT BMI ADO), BMI and time of disease (IT BMI TD), BMI and E2 levels (IT BMI E2), age at disease onset with E2 levels and BMI (IT ADO E2 BMI), age at disease onset and E2 (IT ADO E2); the interaction of the previous year’s cumulative dose of corticosteroids with BMI (IT LYC BMI), as well as with E2 levels (IT LYC E2); the interaction of the total cumulative dose of corticosteroids and BMI was also analyzed. 3.7. Lumbar spine L2–L4 region: In the densitometric study of this segment, 31 (51.7%) showed low bone mass, 12 had osteoporosis. The mean BMD in the group patients with low bone mass was 0.974 ± 0.128 g/cm2 and in those without it was 1.232 ± 0.117 g/cm2. In the analysis of the independent variables, as well as the logistic regression, none exhibited any significant influence on the low bone mass in this region. The final results of the logistic regression can be observed in Table 5. 3.8. Proximal femur Femur neck: The number of SLE patients with low bone mass in this region was 17 (28.4%), four of whom had osteoporosis. The mean BMD in the group with and without low bone mass were, respectively, 0.804 ± 0.105 g/cm2 and 1.007 ± 0.104 g/cm2. None of the analyzed variables exhibited an influence on bone mass in this region. Ward’s Triangle: Twenty-one (35%) SLE women showed low bone mass in this region, six of whom had osteoporosis. The means of BMD in the groups with and without low bone mass were, respectively, 0.688 ± 0.098 g/cm2 and 0.944 ± 0.105 g/cm2. None of the variables exhibited any influence on bone mass. Trochanter: Low bone mass was observed in this region in 18 (30%) SLE women, six of whom had osteoporosis. The mean of BMD for this region was 0.621 ± 0.085 g/cm2 in those with loss and 0.820 ± 0.099 g/cm2 in those without loss. None of the variables exhibited an influence on bone mass, except the activity of the disease whose significance level (P) was 5%. 4. Discussion The real incidence as well as the etiologic mechanism of osteoporosis among SLE patients remains unknown. It is Table 5 Final result of logistic regression for multiple analysis region L2–L4 Parameter

Coefficient

SE**

P

General mean IT BMI ADO* E2***

1.0320 –0.0019 0.0007

0.7698 0.0010 0.0017

NS NS NS

* IT BMI ADO: interaction between BMI and age at disease onset. ** SE: Standard error. *** E2: Estradiol.

43

generally accepted that low BMD increases the risk of fractures and it is a common problem among patients with several rheumatic diseases. Multiple etiologic mechanisms are involved in the bone mass reduction in these patients. They can include some intrinsic factors of the inflammatory process, other disease related peculiarities and broken calcium imbalance. On the other hand, the several utilized drug related toxicity, which in same cases could alter the bone metabolism, is another possible factor associated with the low bone mass as in rheumatoid arthritis (RA). The pathogenic mechanisms in RA are related to age [20], patient functional status [21], menopause, time and disease activity [22], the dose and duration of treatment with corticosteroids [23], as well as the cytokines’ effect on bone [24], that can also be involved in SLE verified low bone mass. With the aim of analyzing the influence of some factors on the bone mineral density, we compared the SLE patients with and without low bone mass and also with an age and race matched control group of healthy women. Our findings showed that SLE patients have significantly less BMD than the normal women, and are in concordance with previous findings [6–11,25]. In this study, our goal was to study some variables on BMD reduction in SLE patients such as the cumulative corticosteroid doses, including the cumulative dose within the last year before evaluation, and the dose at the time of densitometry, as well as the plasmatic estradiol levels. For this purpose, the SLE patients were divided into two groups. In the first group, patients who showed a BMD reduction at any site were included. In the second, those who had normal BMD were included. The influence of these variables was analyzed in each vertebra at lumbar spine, including the total segment L2–L4, and in the three regions of proximal femur as well. In lumbar spine and femur, the relations among the low bone mass and some variables, such as age at disease onset, age at the time of evaluation, bone mass index, time of disease, corticosteroid utilized doses, plasmatic estradiol levels, use of cytotoxic drugs, smoking history, as well as the interaction among two or more variables, were not statistically different between patients with and without low BMD. Analysis of data showed that the interaction between low BMI and low age at disease onset might influence low bone mass at L2, but not in another region, except at L2–L4, where the same was observed, although at a non-significant level (P = 0.06). These findings have not been reported in other series even once. None of them analyzed the relation between low bone mass and bone mass index. The correlation between low BMD and anthropometric index (BMI) is well known [26–28], although the influence of the interaction of age and BMI has not yet been studied. Other analyzed variables, including smoking history, did not show any association with osteopenia in this series. The small number of patients with smoking history among them could in part explain this phenomenon. In relation to clinical manifestation, there were no statistical differences among the two groups, including renal dis-

44

I.B. Coimbra, L.T.L. Costallat / Joint Bone Spine 70 (2003) 40–45

ease, in agreement with a previous report [4]. It could be in part explained by the patient profiles. Only 10 of the 60 studied patients presented with renal disease, none of whom with renal failure, that can develop a secondary hyperparathyreoidism related osteopenia. The disease activity measured by the SLEDAI, did not show any influence on the probability of low bone mass as observed by others [8,25], except in trochanter, where the disease activity was the only one that influenced the probability of low bone mass. Therefore the statistical significance was not well demonstrated (P = 0.0525). This could suggest an association between low bone mass in this region and disease activity, indicating that those patients tended to show low bone mass. However, further studies are required to better explain this association, probably with a greater number of patients. There are evidences of interleukine-1 (IL-1) and tumor necrosis factor alpha (TNF-a) increased production induced by estrogen deficiency [29]. Those increases could enhance the bone resorption. There was no association among low bone mass with either the total cumulative dose of corticosteroids, nor with the last year cumulative dose and the dose at the moment of evaluation, in none of the studied regions. Those findings are similar to some previously reported [4–5,7–8,11,30,31] but not with Petri’s [15] and others [32–34], that observed an association between the cumulative dose of corticosteroids and low BMD on spine. In spite of the statistical model utilized was the same utilized as that in this series, the ethnic characteristics of both groups were different. AfroAmericans were more frequent in that series and Caucasians were in ours. Low levels of serum complement (fraction C4), were the activity index used in that series, while SLEDAI was in ours. Those differences between these series may explain the conflicting findings. Another study also found an association between the corticosteroids dose and L2 low BMD, but not in other regions of the lumbar spine [14], and two others with high dose of corticosteroids and low bone mass [13,25]. In the first study [14] the authors also observed a significant influence of high dose of corticosteroids on femur neck low BMD in the first 12 month of treatment, explained by just a statistical influence, once clinically it was established they could not verify that influence. Yet another study found an association between corticosteroid use and low bone mass in femur [25]. However, many factors may bias their results. The participants in that study had a higher fracture rate than those who declined to be studied, suggesting that BMD may have been lower among patients. This was not verified among our patients. Besides, in this study the patients observed may have had more severe disease, as evidenced by a greater percentage of non-zero SLICC (Systemic Lupus International Collaborative Clinics) scores. Also, among their patients, they could observe 37 post menopause patients, 20 of whom were on hormone replacement therapy, six were taking vitamin D and 27 were on oral calcium supplementation. This is substantially different from our study, where all se-

lected patients were pre-menopause and the use of hormone replacement, vitamin D and calcium supplementation was considered as exclusion criteria. These findings support that other factors, which are not well known yet, may be implicated in the low bone mass in some SLE patients under corticosteroid regimen. The influence of cytotoxic drugs on bone mass was also verified since some cytotoxic agents, such as cyclophosphamide, can lead to an ovarian failure that can cause an estrogen deficiency. However, this was not observed in this study, and no influence of therapeutic regimen on BMD was verified. The association of estrogen levels and BMD was also analyzed, through the plasmatic estradiol levels measured in the folicullar phase. A chronic hyperestrogenism was observed in SLE women patients as well in men [35]. A protection against low bone mass in those patients as previously suggested could be expected [5]. In our study, the plasmatic estradiol levels were analyzed by logistic regression, univariate and multiple analyses, in patients who showed low bone mass compared to those who did not. No statistical differences were found between them. Also, no influence of these levels interacting with other variables was verified on bone mass. Those findings can suggest that in those patients, the plasmatic estradiol levels did not play a protective role against either osteopenia or osteoporosis, as previously suggested [5,9,36]. In summary, in this study, low bone mass was more frequent among SLE women patients than in healthy women. These findings confirm earlier studies as discussed before, including the osteopenia among the clinical manifestations that can occur in SLE patients. We also verified that as the younger the patient at disease onset and smaller the bone mass index, the bigger is the probability to show osteopenia in L2. There was no association between low bone mass and corticosteroid doses, neither with doses nor with other studied variables. Therefore, in younger patients with low bone mass index, under corticosteroid regimen, a prophylactic therapeutic against osteopenia may be useful, although further studies must be performed on this field. The use of cyclophosphamide and azathioprine did not seem to influence the low bone mass in these patients. In relation to the association between BMD, we verified that the plasmatic estradiol did not influence the presence of low bone mass. Other mechanisms may indirectly be involved, such as diminished physical activity, often associated with the arthritis episodes, other systemic complains, as well as the long periods of hospitalization that can occur with SLE patients. Also, the deliberate avoidance of sunlight exposure always recommended to SLE patients, can result in abnormalities of vitamin D hydroxylation and diminished calcium intestinal absorption. In addition, individuals with SLE may not attain an optimum peak bone mass at skeletal maturity, since their disease often begins in early adult life. As the SLE population grows older, and the BMD measurements become more available, further studies are re-

I.B. Coimbra, L.T.L. Costallat / Joint Bone Spine 70 (2003) 40–45

quired for better knowledge of all the mechanisms involved in the low bone mass observed in some SLE patients and also in its prevention.

Acknowledgements We would like to thank Dr. Lucia H. Costa Paiva and Dr. Aarão Mendes Pinto from the Department of Tocogymnecology, Clinical Hospital, UNICAMP, for the control group in this study.

References [1] [2] [3]

[4]

[5]

[6] [7]

[8]

[9]

[10]

[11] [12]

[13]

[14]

[15] [16]

[17]

Bresnihan B. Outcome and survival in systemic lupus erythematosus. Ann Rheum Dis 1989;48:443–5. Mills JA. Systemic lupus erythematosus. New Engl J Med 1994;330: 1871–9. Dykman TR, Gluck OS, Murphy WA, Hahn BH. Evaluation of factors associated with glucocorticoid-induced osteopenia in patients with rheumatic diseases. Arthritis Rheum 1985;28:361–8. Ramsey-Goldman R, Schilling E, Daugherty C, Conte C, Rairie J, Manzi S. Fractures in patients with lupus. Arthritis Rheum 1996; 39(Suppl 9):S89. Dhillon VB, Davies MC, Hall ML, Round JM, Ell PJ, Jacobs HS, et al. Assessment of the effect of oral corticosteroids on bone mineral density in systemic lupus erythematosus: a preliminary study with dual energy X-ray absorptionmetry. Ann Rheum Dis 1990;49:624–6. Kalla AA, Van Wyk Kotze TJ, Meyers OL. Metacarpal bone mass in systemic lupus erythematosus. Clin Rheumatol 1992;11:475–82. Kalla AA, Fataar AA, Jessop SJ, Bewerunge L. Loss of trabecular bone mineral density in systemic lupus erythematosus. Arthritis Rheum 1993;36:1726–34. Formiga F, Moga I, Nolla JM, Pac M, Mitjavila F, Roig-Escofet D. Loss of bone mineral density in pre-menopausal women with systemic lupus erythematosus. Ann Rheum Dis 1995;54:274–6. Houssiau FA, Lefebvre C, Depresseaux G, Lambert M, Devogelaer JP, Nagant-de Deuxchans C. Trabecular and cortical bone loss in systemic lupus erythematosus (SLE). Br J Rheumatol 1996;35:244–7. Horslev-Petersen K. Influence of disease activity and drug treatment on bone mineral density (BMD) in patients with systemic lupus erythematosus. Rheumatology in Europe 1995;24(Suppl. 3):11. Sels F. Bone density in systemic lupus erythematosus. Rheumatology in Europe 1995;24(Suppl. 3):353. Sels F, Dequeker J, Verwilghen J, Mbuyi-Muamba JM. SLE and osteoporosis: dependence and/or independence on glucocorticoids. Lupus 1996;5:89–92. Rosin M, Yarboro C, Nuzzo V, Klippel J, Gourley M. Corticosteroid (CS) effects on bone mineral density (BMD) in patients with systemic lupus erythematosus (SLE). Arthritis Rheum 1996;39(Suppl 9):S293. Hearth-Holmes M, Nandy I, Elder C, Wolf R. Bone mineral density (BMD) in two groups of patients with systemiclupus erythematosus. Arthritis Rheum 1996;39(Suppl. 9):S292. Petri M. Osteoporosis in SLE: prednisone affects lumbar spine more than other areas. Arthritis Rheum 1996;39(Suppl. 9):S138. Tan EM. Systemic lupus erythematosus: immunologic aspects. In: McCarty DJ, editor. Arthritis andallied conditions. Philadelphia: Lea and Febiger; 1985. p. 1049–54. Bombardier C, Gladman DD, Urowitz MB, Caron D, Chang CH. The committee on prognosis studies in SLE. Derivation of the SLEDAI: a disease activity index for lupus patients. Athritis Rheum 1992;35: 630–40.

45

[18] Cullun ID, Ell PJ, Ryder JP. X-Ray dual-photon absorptionmetry: a new method for the measurement of bone density. Br J Radiology 1989;62:587–92. [19] World Health Organization (WHO). Assessment of fracture risk and its application to screening for postmenopausal osteoporosis. Geneva: WHO; 1994 130 p. [Reports series, 843]. [20] Magaro M, Tricerri A. Piane D.Generalized osteoporosis in nonsteroid treated rheumatoid arthritis. Rheumatol Int 1991;2:73–6. [21] Kroger H, Honkanen R, Saarikoski S, Alhava E. Decreased axial bone mineral density in perimenopausal women with rheumatoid arthritis–a population based study. Ann Rheum Dis 1994;53:18–23. [22] Gough AKS, Lilley J, Eyre S, Holder RL, Emery P. Generalized bone loss in patients with early rheumatoid arthritis. Lancet 1994;344: 23–7. [23] Laan RFJM, Van Riel PLCM, Van Erning LJTHO, Lemmens JAM, Ruijs SHJ, Van De Putte LBA. Vertebral osteoporosis in rheumatoid arthritis patients: effect of low dose prednisone therapy. Br J Rheumatol 1992;31:91–6. [24] Offe I, Epstein S. Osteoporosis associated with rheumatoid arthritis: Pathogenesis and management. Semin Arthritis Rheum 1991;20: 256–72. [25] Kipen Y, Buchbinder R, Forbes A, Strauss B, Littlejohn G, Morand E. Prevalence of reduced bone mineral density in systemic lupus erythematosus and the role of steroids. J Rheumatol 1997;24:1922–9. [26] Liel Y, Edward J, Shary J, Spicer KM, Gordon L, Bell NH. The effects of race and body habitus on bone mineral density of the radius, hip and spine in pre-menopausal women. J Clin Endocrinol Metab 1988;66: 1247–50. [27] Katzman DK, Bachrachl K, Carter DR, Marcus R. Clinical anthropometric correlates of bone mineral acquisition in healthy adolescent girls. J Clin Endocrinol Metab 1991;73:1332–9. [28] Heiss CJ, Sanborn CF, Nichols DL, Bonnick SL, Alford BB. Association of body fat distribution, circulating sexhormones and bone density in postmenopausal women. J Clin Endocrinol Metab 1995;80: 1591–6. [29] Pacifici R, Brown C, Puscheck E, Friedrich E, Slatopolsky V, Magio D, et al. Effect of surgical menopause and estrogen replacement on cytokine release from human blood mononuclear cells. Proc Natl Acad Sci USA 1991;88:5134–8. [30] Formiga F, Nolla JM, Mitjavila F, Bonnin R, Navarro MA, Moga I. Bone mineral density and hormonal status in men with systemic lupus erythematosus. Lupus 1996;5:623–6. [31] Lakshminarayanan S, Walsh S, Mohanraj M, Rothfield N. Factors associated with low bone mineral density in female patients with systemic lupus erythematosus. J Rheumatol 2001;28:102–8. [32] Jardinet D, Lefebvre C, Depresseux G, Lambert M, Devogelaer JP, Houssiau FA. Longitudinal analysis of bone mineral density in premenopausal female systemic lupus erythematosus patients: deleterious role of glucocorticoid therapy at the lumbar spine. Rheumatology (Oxford) 2000;39:389–92. [33] Sinigaglia L, Varenna M, Binelli L, Zucchi F, Ghiringhella D, Gallazzi M, et al. Determinants of bonemass in systemic lupus erythematosus: a cross sectional study on pre-menopausal women. J Rheumatol 1999;26:1280–4. [34] Pons F, Peris P, Guanabens N, Font J, Huguet M, Espinosa G, et al. The effect of systemic lupus erythematosus and long-term steroid therapy on bone mass in pre-menopausal women. Br J Rheumatol 1995;34:742–6. [35] Lahita RG. Sex steroids and SLE: metabolism of androgens and estrogens [Editorial]. Lupus 1992;1:125–7. [36] Redlich K, Ziegler S, Kiener HP, Spitzauer S, Stohlawetz P, Bernecker P, et al. Bone mineral density and biochemical parameters of bone metabolism in female patients with systemic lupus erythematosus. Ann Rheum Dis 2000;59:308–10.