Measurement of hand bone mineral content by dual energy x-ray absorptiometry: development of the method, and its application in normal volunteers and in patients with rheumatoid arthritis.
OBJECTIVES--To develop a method of measuring hand bone mineral content (BMC) by dual energy x ray absorptiometry (DXA); to apply this method of measuring hand BMC to normal volunteers to ascertain causes of variability; and to measure hand BMC in patients with rheumatoid arthritis (RA) of varying duration and severity. METHODS--The x ray beam of the Hologic QDR 1000 dual energy x ray absorptiometer was hardened by introducing a perspex-aluminium plate and the analysis software altered to allow for the small tissue bulk of the hand compared with the torso. Ninety five volunteers (46 men age 24-81 and 49 women age 20-83) had scans of both hands. Eight volunteers were assessed repeatedly to establish reproducibility and effect of hand position. Fifty six patients (22 men, 34 women, age range 25-86 years) with RA of differing duration and severity, had hand BMC measurement by DXA. RESULTS--The precision of BMC measurement was 2.3% with no additional variation due to hand position. Hand dominance had no significant effect on BMC. In men, hand BMC correlated with height (r = 0.57, p < 0.0001), weight (r = 0.58, p < 0.0001), forearm span (r = 0.5, p = 0.0006) and hand volume (r = 0.66, p < 0.0001). In women hand BMC correlated with height (r = 0.66, p < 0.0001), weight (r = 0.4, p = 0.003), forearm span (r = 0.3, p = 0.03) and hand volume (r = 0.49, p = 0.0008). After correcting for all these variables, male volunteers had significantly higher hand BMC than female volunteers (p = 0.01) and patients with RA had lower hand BMC than normal volunteers (total hand BMC in male volunteers 90.9 gms, 95% CI 86.9-95, in male patients 81.7 gms, 95% CI 73.7-89.6, p < 0.004, total hand BMC in female volunteers 62.2 gms 95% CI 59.8-64.5, female patients 52.3 gms, 95% CI 48.1-56.5, p < 0.005). In patients with RA, the hand BMC showed an inverse correlation with age (r = -0.44, p = 0.01), disease duration (r = -0.62, p = 0.0003), Larsen's grades (r = -0.62, p = 0.0002) and modified Sharp's method score (r = -0.69, p < 0.0001) in female patients only. CONCLUSIONS--A new, sensitive and reproducible technique of measurement of hand bone mineral content by DXA, has been developed and this method has been applied to normal volunteers and patients with RA. Hand dominance had no significant effect on hand BMC. After correcting for physical size, men have higher hand BMC than women. Hand BMC inversely correlates in women patients with disease duration and other validated methods of assessing radiological outcome in RA. Longitudinal studies are needed to establish its role in monitoring disease progression.