Odds ratios for hip- and lower forearm fracture using peripheral bone densitometry; a case–control study of postmenopausal women


Henrik L. Jørgensen Department of Clinical Biochemistry, Hvidovre Hospital, Kettegaard Allé 30, 2650 Hvidovre, Denmark


Background: Dual-energy X-ray absorptiometry (DXA) measured at the lumbar spine and particularly at the hip remain the gold-standard for diagnosing osteoporosis. However, devices for assessing the peripheral skeleton present several advantages in terms of lower price and portability. A major concern when using peripheral densitometry is the poor correlation with the central measurements. The main aim of this study is, therefore, to assess the possibility of expressing ultrasound measurements at the heel and bone mineral density (BMD) measured at the distal forearm as fracture odds ratios rather than an absolute measure of bone mass. Methods: A total of 76 women with lower forearm fracture, 47 women with hip fracture and 231 age-matched women (controls) were included. All had broadband ultrasound attenuation (BUA) and speed of sound (SOS) measured at the heel using the DTU-one ultrasound scanner as well as BMD measured by dual X-ray absorptiometry on the DTX-200 at the distal forearm. Results:  BUA, SOS and BMD at the distal forearm were all significantly lower in fracture patients compared with their respective control groups. The odds ratio for lower forearm fracture was 3·1 (95% CI: 1·8; 5·2) for heel-BUA (T-score cut-off: –2·3), 4·1 (2·3; 7·4) for heel-SOS (–2·1) and 2·2 (1·3; 3·7) for lower forearm BMD (–2·7). The odds ratio for hip fracture was 3·4 (1·5–7·7) for heel-BUA (–2·7), 3·6 (1·6; 8·1) for heel-SOS (–2·6) and 3·2 (1·4; 7·4) for lower forearm BMD (–2·9). Conclusion:  Peripheral densitometry can discriminate between hip- and lower forearm fracture patients and age-matched controls. Significantly elevated odds ratios for incurring these fractures can be calculated using device- and site specific t-score cut-off values. The results from this case–control study need to be confirmed by prospective cohort studies.