Article Text

Effect of hormone replacement therapy on bone density in a patient with severe osteoporosis caused by anorexia nervosa
  1. Department of Rheumatology, St Helier Hospital, Carshalton, Surrey SM5 1AA

    Statistics from

    Osteoporosis is a recognised complication of anorexia nervosa, in part because of oestrogen deficiency.1 Data on the effect of oestrogen supplementation in osteoporotic patients with anorexia nervosa are limited. The effect of oestrogen supplementation on bone density in a patient with severe osteoporosis because of anorexia nervosa is reported.

    The patient presented at the age of 26 years with a history of intermittent back pain and kyphosis. Her menarche was at the age of 13 years when she weighed 53 kg. At that time she developed a restrictive eating habit but continued to menstruate until the age of 19 years when she became amenorrhoeic and weighed 32 kg. Although she received psychological support her weight only improved temporarily. She remained underweight at 41 kg for the two years preceding presentation. The patient admitted to drinking minimal amounts of milk, although did have yoghurts regularly. The only exercise she did was walking as part of daily activities. There was no history of binge eating or vomiting and she denied laxative, diuretic, alcohol or drug abuse.

    On examination her height was 1.57 m, weight 41 kg (body mass index 17). Her sclera were normal and joint hypermobility and excessive skin laxity were not found. She had a kyphosis with spinal tenderness on percussion and reduced spinal movements. General examination was normal. Investigations showed normal full blood count, biochemistry including serum calcium, phosphate, alkaline phosphatase, and thyroid function. Serum oestradiol was 55 pmol/l (range 120 to 570), luteinising hormone 0.5 IU/l (range 1 to 10), follicle stimulating hormone 4.6 IU/l (range 1 to 8), and testosterone 0.4 nmol/l. Radiographs showed multiple vertebral fractures of her thoracic spine but her lumbar spine was normal. Biochemical markers of bone turnover, parathyroid hormone, and vitamin D levels were not measured. Bone mineral density (BMD) was measured using a Lunar DPX densitometer and showed severe osteoporosis with a lumbar spine (L2 to L4) BMD of 0.36 g/cm2 (T score −7.0, Z score −6.2) and femoral neck BMD of 0.51 g/cm2 (T score −3.9, Z score −3.38).

    She underwent psychiatric assessment but declined help. After some discussion, with an emphasis on the severity of her osteporosis, she agreed to take hormone replacement therapy (HRT) and calcium. She was treated with oestradiol valerate 2 mg and norethisterone acetate 1 mg (Kliofem) with calcium 500mg (Calcichew) daily. Over the following year her lumbar spine BMD increased from 0.36 g/cm2 to 0.51 g/cm2 (a gain of 38% ), but her femoral neck BMD remained unchanged (fig 1). The DXA scan lumbar spine images and L2-L4 area remained similar in all three scans suggesting that the increase in lumbar spine BMD was not resulting from incident lumbar vertebral fracture. After one year she discontinued treatment because she felt bloated. Her lumbar spine BMD has declined since this point. During the period of time that she was taking HRT and calcium no other drugs were given and the patient made no changes in lifestyle. Her weight increased by 2 kg and her height remained stable.

    Figure 1

    Change in bone mineral density with hormone replacement therapy.

    The patient presented in the case report is interesting for a number of reasons including the large increase in lumbar spine BMD because of the effect of HRT. Caution has to be used in the way this increase in lumbar spine BMD is interpreted as the extreme severity of the patients osteoporosis emphasises the percentage increase because of a denominator effect. However, in terms of outcome, her BMD has increased by over 2 standard deviations over a short time course, representing a potential fracture reduction of about fourfold (though this inference is based on data from postmenopausal women).

    Oestrogen replacement to increase BMD seems reasonable in amenorrhoeic women with osteoporosis secondary to anorexia nervosa. However, an open randomised controlled study of comparing no treatment with either HRT (Premarin 0.625 mg and Provera 5 mg) or the combined oral contraceptive pill (ethinyl estradiol 35 μg) concluded that oestrogen supplementation did not prevent progressive bone loss.2 A subgroup of six patients with initial body weights of less than 70% of ideal did however show a small increase in BMD (4.4%). The lack of treatment response when compared with the patient reported could reflect differences in the type, dose, and frequency of delivery of the oestrogen or alternatively the type and dose of progestogen. For example, norethisterone is known to have oestrogenic and androgenic properties in addition to being progestogenic.

    The magnitude of the increase in lumbar spine BMD suggests that there is more than filling in of the remodelling space, and possibly an anabolic effect of oestrogen is being seen, as suggested in recent studies of postmenopausal women.3 4 The lack of increase in femoral BMD is worthy of comment. It is well recognised that different parts of the skeleton and different bone envelopes differ in their response to hormones such as oestrogen. Thus the femoral neck, which is predominantly cortical bone, is likely to be affected differently by oestrogen compared with the spine, which is predominantly trabecular bone. In addition increased skeletal loading may be required to facilitate the maximum effect of oestrogen at certain skeletal sites such as the hip.

    Although restoration of normal body weight and menses may partially restore BMD in anorexic patients with osteoporosis, the findings in the patient reported suggest that oestrogen supplementation should not be neglected. Further studies are required to determine the most effective HRT regimen, combined with appropriate changes in exercise and weight gain in this group of patients.


    View Abstract

    Request Permissions

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