Purpose: To assess the influence of extent of disease on the skeletal status of men with ankylosing spondylitis (AS).
Patients and methods: Fourteen men with AS were studied at entry and again after 15 months. Bone mineral density (BMD) was assessed by single photon absorptiometry (SPA), dual energy x-ray absorptiometry (DXA), and quantitative computed tomography (QCT). Calciotropic hormones and bone turnover were also assessed, and biopsies of iliac crest and skin were taken after tetracycline double-labeling from 10 subjects. Clinical evaluation and Health Assessment Questionnaires were used to assess functional status.
Results: Of the 14 participants, 7 had sacroiliitis alone without radiologic evidence of spinal involvement (early disease) and 7 had sacroiliitis with extensive vertebral calcifications and immobilization (late disease). QCT baseline lumbar spine BMD was very low in both groups compared with normative standards (Z score = -3.08 +/- 1.83, P < 0.0001) and did not change significantly over 15 months. This low BMD was more marked in late disease than in early disease subjects (P < 0.01). DXA BMD at the lumbar spine was lower than predicted in early disease subjects (Z score = -1.08 +/- 0.67, P = 0.005) but not in the late disease group. DXA BMD was also low at the all three hip sites (Z score = -0.96 +/- 0.86, P < 0.01). Significant differences between late disease group and normative values were apparent at all hip sites. Values in early disease subjects, however, did not differ from age-predicted norms. Bone mineral status did not change significantly over the 15-month period of observation. Circulating parathyroid hormone (PTH) and vitamin D metabolites were normal in both groups as were creatinine clearance and urinary excretion of calcium and hydroxyproline. Osteocalcin levels were normal in all but the two youngest subjects in the early disease group. Histomorphometry of the iliac crest showed no consistent change in bone turnover. Bone volume and trabecular width were low in many cases. Cancellous bone volume correlated with lumbar spine BMD by QCT (r = 0.69, P = 0.026) but not with DXA. Although beneficial changes occurred in exercise tolerance and pain over time, anthropometric measurements did not improve.
Conclusion: BMD is low in both the axial and peripheral skeleton in men with AS and is independent of spinal immobilization. Anterioposterior lumbar spine DXA in late AS is less useful than QCT in determining the degree of osteopenia in late AS. Bone mineral deficits in AS do not reflect measurable metabolic derangement or hypogonadism. Although bone histomorphometry suggests both trabecular thinning and loss of structural elements as mechanisms involved in low bone volume, the exact cause of osteopenia in AS remains to be determined.