Rheumatology 1999;38:1239–1244 Osteopenia in young hypogonadal women with systemic lupus erythematosus receiving chronic steroid therapy: a randomized controlled trial comparing calcitriol and hormonal replacement therapy

Objective. To evaluate the efficacy of calcitriol and hormonal replacement therapy (HRT) in the treatment of steroid-induced osteoporosis in hypogonadal women. Methods. We studied 28 young patients (aged 37± 6 yr) with systemic lupus erythematosus (SLE) on chronic steroid therapy for 130± 22 months and requiring more than 10 mg/day prednisone. They were amenorrhoeic for more than 2 yr with proven ovarian failure. All had osteopenia with a T score at L2–4 of less than −1. They were randomized to receive HRT (conjugated oestrogen 0.625 mg daily from day 1 to day 21 plus medroxyprogesterone acetate 5 mg daily days 10–21) or calcitriol 0.5 mg daily. All received calcium carbonate 1 g/day. Results. There were no differences in the baseline demographic, bone mineral density (BMD) and biochemical data between the two groups. Lumbar spine BMD increased by 2.0± 0.4% after 2 yr with HRT (P< 0.05), but reduced by 1.74± 0.4% (P< 0.05) with calcitriol treatment. No change was seen at the distal one-third radius with HRT treatment but significant bone loss (2.3± 1.4%, P< 0.02) was observed with calcitriol therapy. BMD at the hip did not change in both groups. Comparing both treatment groups, significant differences in the BMD at the spine (P< 0.03) and radius (P< 0.05) were seen at the end of 2 yr. The changes in urinary n-telopeptide excretion but not serum osteocalcin at 6 months and 12 months were inversely correlated with the changes in lumbar spine BMD at 24 months. HRT did not cause an adverse effect on SLE disease activity. Conclusion. HRT but not calcitriol could prevent bone loss in young hypogonadal women on chronic steroid therapy. K : Osteopenia, Young hypogonadal women, Steroid, SLE, Calcitriol, Hormonal replacement therapy. Glucocorticoids are widely used in the treatment of formation and increase in bone resorption [6 ]. The mechanism of bone loss is believed to act through patients with autoimmune and chronic inflammatory inhibition of osteoblastogenesis and promotion of diseases due to their potent immunomodulatory action. apoptosis of osteoblasts and osteocytes [7]. In recent However, one of the major side-effects which limits the years, a number of studies have demonstrated that the use of this group of agents is the complication on bone adverse effect of glucocorticoid on bone loss can be loss and bone fractures, especially vertebral fractures. prevented [8]. Active agents include vitamin D and The pathogenesis of steroid-induced osteoporosis is its derivatives, calcitonin and bisphosphonates [8]. complex [1]. These include the reduction in intestinal Initiation of these agents within 4 weeks of steroid calcium absorption [2], increase in renal calcium excretherapy is effective in preventing bone loss which is tion [3], secondary hyperparathyroidism [4], suppresusually most marked within the first year of therapy [9]. sion of gonadotropin release [5], suppression of bone It has been shown that the effect of these agents is similar in both preand post-menopausal women. However, in patients who already have bone loss due Submitted 6 January 1999; revised version accepted 13 July 1999. to chronic steroid therapy, the effect of these agents is Correspondence to: A. Kung, Department of Medicine, University of Hong Kong, Queen Mary Hospital, Pokfulam Road, Hong Kong. less well established. In patients who are hypogonadal, © 1999 British Society for Rheumatology 1239 A. W. C. Kung et al. 1240 the general consensus is to give hormonal replacement Bone densitometry therapy (HRT) [10, 11]. However, there are only a few BMD, expressed as an areal density, was measured at studies documenting the effect of HRT in hypogonadal the lumbar spine (L2–4), femoral neck, trochanter, women on chronic steroid therapy [12, 13]. Although Ward’s triangle, and distal radius using dual-energy the use of HRT may be effective in preventing bone X-ray absorptiometry (DEXA; Hologic QDR 2000 plus, loss, HRT is known to cause flare up of certain autoHologic, Waltham, MA, USA). The in vivo precision of immune diseases and may be contraindicated, for the machine obtained from five post-menopausal women example, in diseases such as systemic lupus erythmeasured four times over 2 weeks for the lumbar spine, ematosus (SLE) [14]. femoral neck, Ward’s triangle, and distal one-third The use of non-hormonal agents such as calcitonin, radius was 1.2, 1.5, 2.2 and 1.4%, respectively. The bone bisphosphonates and vitamin D derivatives has been scans were performed by an independent person who reported to give variable benefits in established steroidwas unaware of the treatment the patients were receiving. induced bone loss [10, 11]. Calcitonin injection has been Results are presented as mean± .. shown to be effective for primary and secondary prevention of steroid-induced bone loss, but it does not signiLaboratory tests ficantly restore bone mass in the presence of steroids Fasting blood was obtained for laboratory tests. Serum [15]. The use of calcitonin is also constrained by its calcium, phosphorus, albumin, creatinine, total alkaline cost, side-effects, resistance to treatment with the develphosphatase were measured using a Hitachi 747 random opment of antibodies and a lack of efficacy to prevent access analyser (Boehringer, Mannheim, Germany). fractures [16, 17]. Bisphosphonates are effective in Serum intact osteocalcin, a marker of bone formation, increasing lumbar spine and femoral neck bone mineral was measured by enzyme-linked immunosorbent assay density (BMD) in steroid-treated patients when com(ELISA) method using commercial kits (Novocalcin, pared with controls [18, 19]. However, in view of the Metra Biosystems, Inc., CA, USA). The laboratory long half-life of bisphosphonates, the use of these agents intraand interassay coefficients of variation were 8.8 in young patients with immature skeletons is still of and 10.1%. Subjects were asked to bring a 24-h urine concern. Calcitriol has been shown to be effective in collection for calcium and creatinine measurement. A primary prevention of bone loss with corticosteroids 2-h fasting morning urine sample was also collected for [20], but whether it is effective in chronic steroid therapy n-telopeptide (NTx) measurement. Urine NTx, a marker is uncertain. The aim of our project was to evaluate the for bone resorption, was determined by ELISA effectiveness of HRT vs calcitriol in treating young (Osteomark, Ostex, Seattle, WA, USA). The intraand hypogonadal women with SLE on chronic steroid interassay coefficients of variation were 8.7 and 10.9%, therapy. respectively. All samples for individual subjects were measured in a single assay. Patients and methods Statistical analysis Patients Baseline values were compared using two-sample Twenty-eight women aged 37± 6 yr with SLE [21] on Student’s t-test. Results are reported as mean± .. The chronic steroid therapy were studied. These patients longitudinal data were analysed as the percentage change were put on steroid therapy at the diagnosis of the from baseline and the two groups were compared using disease. All of them were taking at least 10 mg of the ANOVA method. prednisone daily and were on steroid treatment for a mean period of 130± 22 (range 30–240) months. They Results were amenorrhoeic for at least 2 yr and proven to have ovarian failure with elevated luteinizing hormone/ BMD data follicular stimulating hormone. All had osteopenia with Table 1 shows the baseline characteristics of the two the T score at L2–4 of less than −1 according to local groups of patients. No differences were observed in their southern Chinese peak young mean values. All of them demographic, biochemical as well as BMD data. There had stable disease and were considered suitable for HRT was no correlation between the BMD values and the treatment. total steroid dose used and the duration of steroid Methods therapy in these patients. The serial changes of the various BMD results are The patients were randomly assigned to receive either HRT (conjugated oestrogen premarin 0.625 mg daily shown in Fig. 1. Treatment with HRT resulted in a 2.0± 0.4% increase in the lumbar spine BMD from day 1 to day 21 plus medroxyprogesterone acetate 5 mg daily from day 10 to day 21 of the 28 day (P< 0.05). There was no significant change in the BMD at both the hip and the forearm. Women treated with cycle (Wyeth, USA) or calcitriol (Rocaltrol, Roche Laboratories, USA) 0.5 mg daily. All patients received calcitriol continued to demonstrate bone loss. At the lumbar spine, there was a 1.6± 0.4% reduction at calcium carbonate 1 g daily. Biochemical bone markers and BMDs were determined at baseline and every 18 months (P< 0.05 vs baseline) and a 1.74± 0.4% reduction at 24 months (P< 0.05 vs baseline). Similarly, 6 months for 2 yr. Calcitriol and HRT in hypogonadal women on steroid therapy 1241 T 1. Baseline characteristics of the patients (mean± ..) Hormonal replacement therapy Calcitriol