GA Utrecht and 10Department of Rheumatology, Hospital Hilversum, PO Box. 10016, 1201 DA Hilversum, The Netherlands. Abstract. Objective. This study was ...
Rheumatology 2001;40:1375–1383
Prevalence of vertebral deformities and symptomatic vertebral fractures in corticosteroid treated patients with rheumatoid arthritis R. N. J. de Nijs, J. W. G. Jacobs, J. W. J. Bijlsma, W. F. Lems1, R. F. J. M. Laan2, H. H. M. Houben3, E. J. ter Borg4, A. M. Huisman5, G. A. W. Bruyn6, P. L. M. van Oijen7, A. A. A. Westgeest8, A. Algra9 and D. M. Hofman10 on behalf of the Osteoporosis Working Group of the Dutch Society for Rheumatology Department of Rheumatology and Clinical Immunology, F02.127, University Medical Center, PO Box 85500, 3508 GA Utrecht, 1 Department of Rheumatology, Free University Hospital, PO Box 7057, 1007 MB Amsterdam, 2Department of Rheumatology, University Medical Center, PO Box 9101, 6500 HB Nijmegen, 3Department of Rheumatology, Atrium Medical Center, PO Box 4446, 6401 CX Heerlen, 4Department of Rheumatology, Sint Antonius Hospital, PO Box 2500, 3430 EM Nieuwegein, 5 Department of Rheumatology, Sint Franciscus Gasthuis, Kleiweg 500, 3045 PM Rotterdam, 6Department of Rheumatology, Hospital Medisch Centrum Leeuwarden, Henri Dunantweg 2, 8934 AD, Leeuwarden, 7Department of Rheumatology, Hospital Bosch Medicentrum, Nieuwstraat 34, 5211 NL’s-Hertogenbosch, 8Department of Rheumatology, Hospital Diaconessenhuis, PO Box 90052, 5600 PD Eindhoven, 9Julius Center for General Practice and Patient Oriented Research, University Medical Center, PO Box 85500, 3508 GA Utrecht and 10Department of Rheumatology, Hospital Hilversum, PO Box 10016, 1201 DA Hilversum, The Netherlands Abstract Objective. This study was designed to determine whether the prevalence of vertebral deformities in patients with rheumatoid arthritis (RA) treated with corticosteroids (Cs) is higher than in RA patients not receiving Cs therapy. Patients and methods. This multicentre cross-sectional study included 205 patients with RA who were receiving Cs orally on a daily basis and 205 patients with RA who did not receive Cs, matched for sex and age. Vertebral deformities were scored according to the Kleerekoper method. Results. Vertebral deformities were found in 52 (25%) patients on Cs and in 26 (13%) patients not on Cs. Sixteen (8%) patients in the group on Cs had experienced clinical manifestations of an acute vertebral fracture in the past vs only three patients (1.5%) among those not on Cs. The use of Cs tended to increase the risk of developing a vertebral deformity wadjusted odds ratio (OR) 1.56, 95% confidence interval (CI) 0.81–2.99x and symptomatic vertebral fracture (adjusted OR 1.42, 95% CI 0.24–8.32). Each 1-mg increase in the current daily Cs dose increased the risk of
Submitted 1 November 2000; revised version accepted 4 June 2001. Correspondence to: R. N. J. de Nijs, Department of Rheumatology and Clinical Immunology, F02.127, University Medical Center, PO Box 85500, 3508 GA Utrecht, The Netherlands.
1375
ß 2001 British Society for Rheumatology
1376
R. N. J. de Nijs et al.
a vertebral deformity (adjusted OR 1.05, 95% CI 0.98–1.13) and of a symptomatic vertebral fracture (adjusted OR 1.05, 95% CI 0.89–1.24). Conclusion. There is a higher prevalence of vertebral deformities and clinical manifestations of vertebral fractures in patients on Cs than in those not on Cs. Our data indicate that the use of Cs and each 1-mg increase in the current daily Cs dose may increase the risk of development of a vertebral deformity and symptomatic vertebral fracture in patients with RA. KEY WORDS: Corticosteroids, Prevalence, Rheumatoid arthritis, Symptomatic vertebral fractures, Vertebral deformities.
Patients with rheumatoid arthritis (RA) are at risk of generalized osteoporosis. This is due to a combination of factors, in particular disease activity, immobility and medication. Corticosteroids (Cs) are often prescribed for patients with severe RA who respond poorly to antirheumatic drugs. Cs are known to uncouple bone formation and resorption, thus inducing osteoporosis w1, 2x. Cs decrease the intestinal absorption and increase the renal excretion of calcium w3, 4x. This may lead to secondary hyperparathyroidism w5x. In addition, Cs inhibit osteoblast proliferation and matrix attachment w6x. The effects of Cs occur mainly in regions of the skeleton with a high content of trabecular bone, in particular the ribs and spine, and seem to depend on the duration and dose of therapy w7, 8x. Prednisone doses exceeding 7.5 mg daily result in increased risks of fractures and bone loss w9, 10x, but there are also negative effects of lower prednisone doses on bone mineral density w11x. Disease activity in patients with active RA is associated with increased generalized loss of bone mass, probably because of increased release of bone-resorbing cytokines, such as interleukins 1 and 6, tumour necrosis factor a and interferon-c w12, 13x. The incidence of vertebral deformities in women with RA was over twice that in healthy controls w14x. In patients with early RA without Cs therapy, bone loss at the hips exceeded that in the control group w15x. The occurrence of osteoporosis is correlated with both disease activity and immobilization in patients with RA w16, 17x. A peripheral osteoporotic fracture is nearly always accompanied by symptoms and is often associated with trauma, whereas vertebral osteoporotic fractures often do not give rise to symptoms w18x and the association with trauma is less evident. The prevalence of vertebral deformities reported in the literature varies from 3 w19x to 27% w20x for post-menopausal women aged 65 yr and older. This wide range in prevalence is probably due to differences between centres in both the scoring method and the definition of vertebral deformities w21x. This is illustrated by a study of the prevalence of vertebral deformities in patients with RA treated with Cs, which varied from 35 to 79% depending upon the scoring method. The incidence of clinical manifestations of vertebral fractures was also higher in Cs-treated RA patients than in a group of patients with RA who were not taking Cs w22x. The influence of Cs on bone mineral density and vertebral deformities has been the subject of numerous
studies. Most of these studies, however, had a relatively small patient population and the patients had heterogeneous diseases. We report the results of a large crosssectional study of a group of RA patients. We investigated whether the prevalence of vertebral deformities and symptomatic vertebral fractures was higher in RA patients treated with Cs than in RA patients who did not receive Cs.
Patients and methods Patients Patients were recruited by members of the Osteoporosis Working Group of the Dutch Society for Rheumatology at 10 hospitals in The Netherlands, university and nonuniversity centres. Two hundred and five patients with RA w1987 American Rheumatism Association (ARA) criteriax w35x had been using Cs orally on a daily basis for at least 1 month. Another 205 patients with RA who were currently not receiving Cs, were matched oneto-one for sex and age (the greatest difference was 5 yr) to form a comparison group. Informed consent was obtained from all patients. The ethics committees of the participating hospitals approved of the study. Clinical evaluation In this cross-sectional study, the following data were obtained from the medical records of the patients: duration of RA; cumulative oral prednisone dose (or prednisone equivalent dose); duration of Cs treatment; other medications that might affect bone (use of calcium, vitamin D, fluoride, bisphosphonates, oestrogens, thiazide diuretics and methotrexate at the time of inclusion); and laboratory data wmean and cumulative (since the diagnosis of RA) erythrocyte sedimentation rate (ESR) and rheumatoid factor test statusx. Data on the patients were obtained from a form filled in by a rheumatologist at each of the participating centres, and included age, sex, height, weight, ARA functional class (Steinbrocker) w23x, daily dietary calcium intake and current daily oral prednisone dose or prednisone equivalent dose. The Dutch version of the Health Assessment Questionnaire (HAQ) (Vragenlijst Dagelijks Functioneren) w24x was filled in by the patients. The range of the HAQ is 0–3; 0 represents optimal and 3 poor functional capacity. Symptomatic vertebral fractures were identified from reports in the medical records of vertebral deformities
Vertebral deformities in corticosteroid-treated patients with RA
after a period of back pain that led to prescription of analgesics or bed rest w25x. Vertebral deformities Lateral X-rays of the thoracic and lumbar spine were made of all patients of both RA groups, as part of this study. All X-rays were scored twice, once by a regular observer and once by one of three alternating observers. The X-rays were scored blind following the method of Kleerekoper et al. w26x. This method consists of nakedeye inspection of vertebrae Th4–L5, comparing each vertebra with the vertebrae below and above it. If an abnormally shaped vertebra was found, the anterior, middle and posterior heights of this vertebra were measured with a ruler. Measurement of the height of an abnormally shaped vertebra was repeated if the interobserver difference in the measurement exceeded 15%. If the difference in the measurement again exceeded 15%, the height was discarded. This occurred in 0.4% of the measurements, which implies that 0.4% of the vertebrae could not be scored according to the Kleerekoper method. The mean of the heights measured by the regular observer and one other observer was used. The scoring system was as follows: 0 = normal shape and dimensions; 1 = only endplate deformity, middle height 15 (n = 37)
22 (30)
14 (24)
6 (17)
10 (27)
5 (7)
4 (7)
3 (8)
4 (11)
There were no statistically significant differences between groups.
OR of prevalence of vertebral deformities in RA patients receiving Cs is influenced by the concomitant medication (bisphosphonates, vitamin D and calcium) given to these patients to treat or prevent osteoporosis, which might influence the prevalence of vertebral deformities, we also performed multivariate logistic analysis with and without this medication to see if there was any change in the adjusted OR. When the influence of current daily dose of Cs on the prevalence of vertebral deformities in the total group was analysed, we found a crude OR of 1.10 (95% CI 1.04–1.16). The adjusted OR with concomitant medication taken into account wfunctional capacity (HAQ) and calcium supplementation (yesuno) were included in the multivariate modelx was 1.05 (95% CI 0.98–1.13). The adjusted OR without calcium supplementation (yesuno) was 1.08 (95% CI 1.02–1.15). This means that, with an increase in the current prednisone (or equivalent) dose of 1 mguday in RA patients, the risk of a vertebral deformity will increase by a factor of 1.08. We would like to stress, however, that even after correction by means of logistic regression analysis, there may be residual confounding due to unknown risk factors. Dykman et al. w7x showed that long-term therapy with various prednisone regimens resulted in glucocorticoidinduced osteopenia and fractures. This effect was cumulative and occurred in all patient groups. Julian et al. w27x, however, could not demonstrate a correlation between the degree of bone loss and the cumulative dose of glucocorticoids. Lems et al. w25x could not find a correlation between cumulative prednisone dose and the prevalence of vertebral deformities. In our study, the OR of cumulative Cs dose vs the prevalence of vertebral deformities was 1.00, demonstrating that there was no effect of the cumulative Cs dose on the prevalence of vertebral deformities. When we divided the RA patients on Cs into subgroups according to cumulative Cs dose, there were no statistically significant differences in the prevalence of vertebral deformities between the subgroups. We have no plausible explanation for the fact that, in our study, the prevalence of vertebral deformities was not related to the cumulative Cs dose. There are, however, differences between our study and the study of Dykman. We had the larger patient population, matched for age and sex, which consisted only of patients with RA; furthermore, we used a validated method to score vertebral deformities.
Vertebral deformities are frequently asymptomatic and sometimes occur during normal physical activities. A vertebral deformity detected on an X-ray might be the result of a symptomatic fracture or an asymptomatic deformity that occurred earlier. Symptomatic vertebral fractures are defined here as acute vertebral deformities leading to the prescription of a therapy. Lems et al. w25x found that 13% of the patients treated with Cs had a symptomatic vertebral fracture, and the occurrence of vertebral deformities in their study was 35–79% depending upon the method used to score the vertebral deformities. In their study, only 2% of the RA patients not taking Cs had a manifest vertebral fracture, whereas the prevalence of vertebral deformities in this group varied from 25 to 67% w22x. According to our data, 8% of the RA patients in the group taking Cs suffered from a manifest vertebral fracture vs 1.5% of the group of RA patients not receiving Cs. Some degree of surveillance bias cannot be excluded, however. RA patients on Cs who have back pain may have more X-ray investigations of the spine. To identify symptomatic vertebral fractures, we used reports in the medical records on vertebral deformities after a period of back pain that led to the prescription of analgesics or bed rest. This method may result in underestimation of the number of symptomatic vertebral fractures because some patients with an episode of back pain may not be treated by their rheumatologist but by their general practitioner, and this information may not be passed on to their rheumatologist. The prevalence of vertebral deformities in the RA group taking Cs was 25% vs 13 for the RA group not on Cs. A discrepancy between the prevalence of vertebral deformities and the prevalence of symptomatic vertebral fractures has also been reported by Cooper et al. w28x. Kleerekoper hypothesized that vertebral fractures lead to vertebral deformities but that not all vertebral deformities are the consequence of a vertebral fracture w18x. Because of the use of the Kleerekoper method, the prevalence of vertebral deformities might have been overestimated slightly in our study as this semiquantitative method does not take into account the normal shape of the vertebrae. The thoracic and lumbar vertebrae are to some degree wedge-shaped and inversely wedge-shaped respectively. Another reason for overestimation of the prevalence of vertebral deformities by the Kleerekoper method is the low threshold (height reduction of 15% or more) for diagnosing
5.93 5.49 5.43 5.42 5.05 5.02 4.74 4.29 4.10 3.11 1.28
(1.69–20.81) (1.57–19.24) (1.55–18.98) (1.53–19.15) (1.44–17.77) (1.43–17.66) (1.34–16.76) (1.19–15.40) (1.15–14.57) (0.83–11.75) (0.30–5.54)
1.05 (0.89–1.24) With Vitamin D supplement YuNa Calcium supplement YuNa
1.19 (1.04–1.35) Without Vitamin D supplement YuNa Calcium supplement YuNa
Height Sex Weight Age Functional capacity (HAQ) Vitamin D supplement YuNa Calcium supplement YuNa
1.16 (1.05–1.28) 1.20 1.19 1.18 1.18 1.14 1.14 1.05
(1.07–1.34) (1.07–1.32) (1.06–1.32) (1.06–1.32) (1.02–1.28) (1.02–1.28) (0.92–1.18)
Total group of patients with RA
1.17 (0.97–1.42) With Bisphosphonate use YuNa
1.17 (0.97–1.42) Without Bisphosphonate use YuNa
Height Bisphosphonate use YuNa Weight Age Sex
1.13 1.12 1.11 1.10 1.09
Patients with RA on Cs 1.07 (0.92–1.24)
Current daily Cs dose
(0.96–1.34) (0.95–1.32) (0.94–1.30) (0.93–1.30) (0.93–1.29)
Multivariate logistic analysis shows the adjusted OR with and without concomitant medication prescribed for the treatment or prevention of osteoporosis. Interpretation of one-variable adjusted ORs, e.g. the crude OR in the column ‘Cs use YuN’, decreased from 5.70 (95% CI 1.63–19.87) to 4.29 (95% CI 1.19–15.40) after non-comparabilities between the Cs groups (YuN) with regard to the use of vitamin D supplement had been taken into account. a Concomitant (anti-osteoporotic) medication. PuN, positiveunegative; YuN, yesuno. CI, confidence interval.
Adjusted OR (95% CI) (with concomitant medication)
Adjusted OR (95% CI) (without concomitant medication)
4.31 (1.13–16.47) Without Vitamin D supplement YuNa Bisphosphonate use YuNa Calcium supplement YuNa 1.42 (0.24–8.32) With Vitamin D supplement YuNa Bisphosphonate use YuNa Calcium supplement YuNa
Rheumatoid factor PuN Methotrexate use YuN Mean ESR Age Height Weight Steinbrocker functional class Vitamin D supplement YuNa Functional capacity (HAQ) Bisphosphonate use YuNa Calcium supplement YuNa
One-variable adjusted OR (95% CI)
Multivariate logistic analysis
5.70 (1.63–19.87)
Crude odds ratio (95% CI) Bivariate logistic analysis
Cs use (YuN) Total group of patients with RA
TABLE 4. Crude and adjusted ORs for the association between the prevalence of symptomatic vertebral fractures and the use of Cs in the total group of patients with RA (n = 410), and the current daily Cs dose (per 1-mg increase in prednisone or prednisone equivalent dose) in the total group and in the group of patients with RA who were on Cs (n = 205)
Vertebral deformities in corticosteroid-treated patients with RA 1381
1382
R. N. J. de Nijs et al.
vertebral deformities. Some investigators take a cut-off point of 20% for vertebral deformity w29x. However, the Kleerekoper method is one of the most feasible approaches in the clinical setting. It is easy to perform and complex calculations are not necessary. Experienced assessors are highly reliable in the identification of vertebral deformities w30x. In our study, we showed that the use of Cs in patients with RA increases the risk of development of a symptomatic vertebral fracture (crude OR 5.70, 95% CI 1.63–19.87). The multivariate adjusted OR was 1.42 (95% CI 0.24–8.32) with and 4.31 (95% CI 1.13–16.47) without concomitant medication wbisphosphonate use (yesuno), vitamin D and calcium supplementation (yesuno)x. This means that calcium supplementation, bisphosphonate use, vitamin D supplementation, functional capacity and Steinbrocker functional class had a considerable influence. The adjusted OR, with concomitant medication taken into account, of the influence of the current daily Cs dose on the development of symptomatic vertebral fractures was 1.05 (95% CI 0.89–1.24), demonstrating that an increase of 1 mg in the daily Cs dose (prednisone or equivalent) was associated with an increase of 1.05 in the risk of a symptomatic vertebral fracture and an increase of 1.19 after adjustment when the model multivariate did not include vitamin D and calcium supplementation (yesuno). The prevalence of Cs therapy in combination with preventive therapy of osteoporosis was assessed in two studies in the UK. The study by Walsh et al. w31x in a community (in Nottinghamshire) showed that only 14% of the patients on oral Cs received medication to prevent secondary osteoporosis. Only 6% of the Cs-treated patients studied by Peat et al. w32x received calcium supplementation. These figures do not meet the recommendations published by the American College of Rheumatology Task Force on Osteoporosis, which stress the need for preventive treatment of patients receiving Cs w33x. In our study, more RA patients on Cs received calcium, vitamin D and bisphosphonate medication than RA patients not using Cs. Despite our statistical adjustment, this treatment, although used only in a small number of patients, might have obscured the effect of Cs on the number of vertebral deformities, because this medication may have been prescribed more frequently to patients with a history of deformities. An important question is what kind of prevention should RA patients treated with Cs receive? There is a need for studies to determine which strategy should be used for the preventive treatment of patients receiving Cs w34x. In summary, we demonstrated in a large crosssectional study in patients with RA, designed to evaluate the final outcome measure of Cs on bone, namely fracture, that vertebral deformities and symptomatic vertebral fractures occurred more often in patients with RA who were treated with Cs than in patients with RA who were not receiving Cs (matched for age and sex). We performed multivariate logistic analyses to adjust for confounding and showed that the use of Cs in patients with RA, even after correction for
confounding, carries a substantial risk of the development of a vertebral deformity and symptomatic vertebral fracture.
Acknowledgements We wish to thank E. Buskens, clinical epidemiologist, for help with the statistical analysis. This work was supported by grant VRu713 from the Dutch League against Rheumatism (Het Nationaal Reumafonds).
References 1. Lukert BP, Raisz LG. Glucocorticoid-induced osteoporosis: pathogenesis and management. Ann Intern Med 1990;112:352–64. 2. Adachi JD, Bensen WJ, Hodsman AB. Corticosteroidinduced osteoporosis. Semin Arthritis Rheum 1993; 22:375–84. 3. Klein RG, Arnaud SB, Gallagher JC, DeLuca HF, Riggs BL. Intestinal calcium absorption in exogenous hypercortisonism. Role of 25-hydroxyvitamin D and corticosteroid dose. J Clin Invest 1977;60:253–9. 4. Hahn TJ, Halstead LR, Bran DT. Effects of short term glucocorticoid administration on intestinal calcium absorption and circulating vitamin D metabolite concentrations in man. J Clin Endocrinol Metab 1981;52:111–5. 5. Suzuki Y, Ichikawa Y, Saito E, Homma M. Importance of increased urinary calcium excretion in the development of secondary hyperparathyroidism of patients under glucocorticoid therapy. Metabolism 1983;32:151–6. 6. Gronowicz G, McCarthy MB. Glucocorticoids inhibit the attachment of osteoblasts to bone extracellular matrix proteins and decrease b-1 integrin levels. Endocrinology 1995;136:598–608. 7. Dykman TR, Gluck OS, Murphy WA, Hahn TJ, Hahn BH. Evaluation of factors associated with glucocorticoidinduced osteopenia in patients with rheumatic diseases. Arthritis Rheum 1985;28:361–8. 8. Reid IR, Heap SW. Determinants of vertebral mineral density in patients receiving longterm glucocorticoid therapy. Arch Intern Med 1990;150:2545–8. 9. Verstraeten A, Dequeker J. Vertebral and peripheral bone mineral content and fracture incidence in postmenopausal patients with rheumatoid arthritis: effect of low dose corticosteroids. Ann Rheum Dis 1986;45:852–7. 10. Michel BA, Bloch DA, Wolf F, Fries JF. Fractures in rheumatoid arthritis: an evaluation of associated risk factors. J Rheumatol 1993;20:1666–9. 11. Laan RF, van Riel PL, van Erning LJ, Lemmens JA, Ruijs SH, van de Putte LB. Vertebral osteoporosis in rheumatoid arthritis patients: effect of low dose prednisone therapy. Br J Rheumatol 1992;31:91–6. 12. Eastgate JA, Wood NC, Di Giovine FS, Symons JA, Grinlinton FM, Duff GW. Correlation of plasma interleukin 1 levels with disease activity in rheumatoid arthritis. Lancet 1988;ii:706–9. 13. Arend WP, Dayer J-M. Cytokines and cytokine inhibitors or antagonists in rheumatoid arthritis. Arthritis Rheum 1990;33:305–15. 14. Spector TD, Hall GM, McCloskey EV, Kanis JA. Risk of vertebral fracture in women with rheumatoid arthritis. Br Med J 1993;306:58.
Vertebral deformities in corticosteroid-treated patients with RA
15. Gough AK, Lilley J, Eyre S, Holder RL, Emery P. Generalised bone loss in patients with early rheumatoid arthritis. Lancet 1994;344:23–7. 16. Laan RF, Buijs WC, Verbeek AL et al. Bone mineral density in patients with recent onset rheumatoid arthritis: influence of disease activity and functional capacity. Ann Rheum Dis 1993;52:21–6. 17. Hooyman JR, Melton LJ, Nelson AM, O’Fallon WM, Riggs BL. Fractures after rheumatoid arthritis. Arthritis Rheum 1984;27:1353–61. 18. Kleerekoper M. Vertebral fracture or vertebral deformity. Calcif Tissue Int 1992;50:5–6. 19. Ha¨rma¨ M, Heliovaara M, Aromaa A, Knekt P. Thoracic spine compression fractures in Finland. Clin Orthop 1986;205:188–94. 20. Melton LJ III, Kan SH, Frye MA, Wahner HW, O’Fallon MO, Riggs BL. Epidemiology of vertebral fractures in women. Am J Epidem 1989;129: 1000–10. 21. O’Neill TW, Varlow J, Felsenbeng D et al. Variation in vertebral height ratios in population studies. European Vertebral Osteoporosis study. J Bone Miner Res 1994; 9:1895–907. 22. Lems WF, Jahangier ZN, Jacobs JWG, Bijlsma JWJ. Methods to score vertebral deformities in patients with rheumatoid arthritis. Br J Rheumatol 1997;36: 220–4. 23. Hochberg MC, Chang WR, Dwosh I, Lindsey S, Pincus T, Wolfe F. The American College of Rheumatology 1991 revised criteria for the classification of global functional status in rheumatoid arthritis. Arthritis Rheum 1992; 35:498–502. 24. Bijlsma JWJ, Oude Heuvel CHB, Zaalberg A. Development and validation of the Dutch questionnaire capacities of daily life (VDF) for patients with rheumatoid arthritis. J Rehabil Sci 1990;3:71–4. 25. Lems WF, Jahangier ZN, Jacobs JWG, Bijlsma JWJ. Vertebral fractures in patients with rheumatoid arthritis
26.
27.
28. 29. 30.
31.
32. 33.
34. 35.
1383
treated with corticosteroids. Clin Exp Rheumatol 1995;13:293–7. Kleerekoper M, Parfitt AM, Ellis BI. Measurement of vertebral fracture rates in osteoporosis. In: Christiansen C, Arnaud CD, Nordin BEC, Parfitt AM, Peck WA, Riggs BL, eds. Proceedings of the Copenhagen International Symposium on Osteoporosis 1984. Copenhagen: Aalberg Stiftsbogtrykkeri, 1984:103–9. Julian BA, Laskow DA, Dubovsky J, Dubovsky EV, Curtis JJ, Quarles LD. Rapid loss of vertebral mineral density after renal transplantation. N Engl J Med 1991; 325:544–50. Cooper C, O’Neill T, Silman A. The epidemiology of vertebral fractures. Bone 1993;14:S89–97. Genant HK, Wu CY, van Kuijk C, Nevitt MC. Vertbral fracture assessment using a semiquantitative technique. J Bone Miner Res 1993;8:1137–48. Wu CY, Li J, Jergas M, Genant HK. Comparison of semiquantitative and quantitative techniques for the assessment of prevalent and incident vertebral fractures. Osteoporosis Int 1995;5:354–70. Walsh LJ, Wong CA, Pringle M, Tattersfield AE. Use of oral corticosteroids in the community and the prevention of secondary osteoporosis: a cross sectional study. Br Med J 1996;313:344–6. Peat ID, Healy S, Reid DM, Ralston SH. Steroid induced osteoporosis: an opportunity for prevention. Ann Rheum Dis 1995;54:66–8. American College of Rheumatology Task Force on Osteoporosis Guidelines. Recommendations for the prevention and treatment of glucocorticoid-induced osteoporosis. Arthritis Rheum 1996;39:1791–801. Bijlsma JWJ. Prevention of glucocorticoid induced osteoporosis. Ann Rheum Dis 1997;56:507–9. Arnett FC, Edworthy SM, Bloch DA et al. The American Rheumatism Association 1987 revised criteria for the classification of rheumatoid arthritis. Arthritis Rheum 1998;31:315–24.
