Clinical Endocrinology (1997) 47, 367–370 Bone mineral density in Turner’s syndrome—a longitudinal study N. J. Shaw*, V. K. Rehan†, S. Husain†, T. Marshall‡ and C. S. Smith† *Department of Growth and Endocrinology, Birmingham Children’s Hospital, Birmingham, †Department of Child Health, Royal Liverpool Children’s Hospital, Liverpool and ‡Department of Public Health and Epidemiology, University of Birmingham, Birmingham, UK (Received 30 January 1997; returned for revision 20 February 1997; finally revised 24 March 1997; accepted 5 June 1997) Summary OBJECTIVE Osteoporosis is a recognized problem in adult women with Turner’s syndrome, the aetiology of which is unclear. The aim of this study was to examine bone mineralization longitudinally in a group of girls with Turner’s syndrome and to study the effect of different treatments on bone mineral density. DESIGN A prospective observational study. PATIENTS Eighteen girls with Turner’s syndrome aged 4–17 years attending a paediatric endocrine clinic. MEASUREMENTS Bone mineral density of the lumbar spine was assessed using dual energy X-ray absorp- tiometry at several time points over a 2 . 5-year period. RESULTS Only one girl had evidence of a significant reduction in bone density when comparisons were made with control data related to body weight and pubertal status. No advantage was found for any form of treatment in optimizing bone mineralization. CONCLUSIONS As there is little evidence of reduced bone mineral density in girls with Turner’s syndrome there is no justification for an early introduction of oestrogen replacement during the prepubertal years. Since the early descriptions of Turner’s syndrome osteoporosis has been listed as an associated feature (Finby & Archibald, 1963). Two main explanations have been proposed for this. The first is that it is due to oestrogen deficiency consequent on ovarian failure comparable to post-menopausal osteoporosis. The second is that it is due to an intrinsic bone defect similar to the other bony abnormalities seen in the condition, which has been likened to a skeletal dysplasia. There have been several studies of bone mineral density in Turner’s syndrome using a variety of techniques which have produced conflicting results. These studies have usually been cross-sectional in nature with little longitudinal follow-up. The aim of this study was to examine bone mineralization in a group of girls with Turner’s syndrome in a longitudinal manner and to assess the effects of different treatments they were receiving. Methods The patients studied were a group of 18 girls aged 4–17 years with a karyotype consistent with Turner’s syndrome who were attending the regional paediatric endocrine clinic at the Royal Liverpool Children’s Hospital in 1991 when the study commenced. At that time 5 of them were receiving treatment with growth hormone alone, 7 were receiving treatment with growth hormone and oestrogen, 2 were on oestrogen replace- ment alone and 4 girls were receiving no treatment. In this clinic growth hormone was introduced when growth failure became apparent and continued until growth was completed and oestrogen replacement was commenced around the age of 12 years, using ethinyl oestradiol in a low dose of 2 mg per day, which was progressively increased over 2 years and then changed to a combined oestrogen/progestogen preparation containing 20 mg of ethinyl oestradiol. None of these girls received treatment with low dose oestrogen prior to the age of 11 years and none were treated with the anabolic steroid oxandralone. Each girl had measurements of bone mineral density (gm/ cm 2 ) of the lumbar spine (L 2 –L 4 ) performed by dual energy X- ray absorptiometry (DXA) using a Hologic QDR 1000 scanner. This has a radiation dose of 4 mRem and coefficient of variation of 2%. All the girls had a scan at baseline, 6 months, 1 year and then between 2 and 2 . 5 years 14 girls had a further scan. Four girls were not scanned at this point as they had either transferred to an adult Turner’s clinic or had moved from the area. Two sources of reference data for bone mineral density were used in this study. The first, which allows comparison with age and gender-related controls, is from a study of 75 schoolgirls aged 5–13 years performed in Oswestry in the United Kingdom (Davie & Haddaway, 1994) using a Hologic QDR 1000 scanner. The second source of reference data, which allows comparison with girls of similar body weight and pubertal status, is a study of 134 American girls aged 1–19 years 367  1997 Blackwell Science Ltd Correspondence: Dr N. J. Shaw, Department of Growth and Endocrinology, Birmingham Childrens Hospital, Ladywood Middleway, Birmingham, B16 8ET, UK. Fax: 0121-456-4697.