Zellisen and colleagues [1] concluded that a small difference in replacement glucocorticoid dose was probably responsible for the decreased bone mineral density (BMD) observed in 32% of men with treated Addison disease. The absolute glucocorticoid dose was not statistically higher in men with decreased compared with normal BMD (mean, 31.0 ± 6.3 mg/d compared with 28.4 ± 7.3 mg/d of hydrocortisone; P = 0.32); however, the weight-adjusted dose was reported to be significantly different (0.43 ± 0.08 mg/kg compared with 0.35 ± 0.10 mg/kd per day). Because the statistical difference between groups was marginal (P = 0.032) and because several variables were compared, it is important to know whether the authors calculated the P value using a Bonferroni or other statistical correction for multiple comparisons.

The mean absolute glucocorticoid doses differed by 9.2%, whereas the weight-adjusted doses differed by 22.9%; therefore, the mean body weight must have been lower in the men with decreased BMD than in those with normal densities. The body mass index tended to be lower in patients with decreased BMD (P = 0.17). Is the difference in weight between the two groups significant? Because body mass makes an important contribution to BMD [2, 3], were body mass or body mass index tested as variables in the stepwise linear regressions?

If glucocorticoids were responsible for low BMD, it is surprising that the duration of glucocorticoid therapy was not a more important factor. Among the 10 men with low BMD, the mean duration of replacement therapy was 11.0 =/- 10.1 years. Did some men have only short-duration replacement? And, if so, is it reasonable to attribute the low BMD to glucocorticoid treatment in those persons? Is it possible that inactivity or other factors relating directly to Addison disease could have contributed to low BMD?

Long-term, low-dose glucocorticoids are common treatments for inflammatory diseases such as rheumatoid arthritis, in which the inflammatory disease itself may contribute to osteopenia. Therefore, the lessons learned from patients with Addison disease, as studied by Zelissen and colleagues, may be important to many patients and physicians.