Repair of osteopenia in children with juvenile rheumatoid arthritis.

To test the hypothesis that the presence of osteopenia in juvenile rheumatoid arthritis is directly correlated with clinical disease activity and therefore reversible, we prospectively studied cortical bone mineral density (BMD) serially in 27 children. Twenty-four (89%) had BMD > or = 2 SD below age-related normal values (disease duration 49.3 +/- 7.7 months) at the beginning of the study. Of 27 children who had clinical disease improvement measured by a disease activity score during our study period, 17 (63%) had significant improvement or significant normalization, or both, of their BMD (0.34 +/- 0.13 gm/cm2 at initiation and 0.41 +/- 0.17 gm/cm2 at completion, p < 0.05; disease activity score of 3.4 +/- 0.2 at initiation and 1.4 +/- 0.2 at completion, p < 0.005). The increase in BMD was associated with a similar directional change in serum osteocalcin concentrations (4.6 +/- 1 ng/ml at initiation vs 9.1 +/- 1.1 ng/ml). The 10 patients whose disease became or remained active had a decreased or unchanged low serum osteocalcin level and BMD (BMD 0.37 +/- 0.17 gm/cm2 at initiation and 0.37 +/- 0.16 gm/cm2 at completion; disease activity score of 3.1 +/- 0.3 at initiation and 3.4 +/- 0.2 at study completion). We conclude that children with JRA who have improvement in their disease activity have an improvement in BMD heralded by an increase in serum osteocalcin values.