ab21505

Risk assessment for osteoporosis by quantitative ultrasound of the heel in ankylosing spondylitis

T.L.Th.A. Jansen, M.H.M. Aarts¹, S. Zanen¹, G.A.W. Bruyn

Medisch Centrum Leeuwarden, Department of Rheumatology, Leeuwarden; ¹Isala klinieken, Department of Rheumatology, Zwolle, The Netherlands.

ABSTRACT
Objective
The aim of this cross-sectional cohort study is to assess the potential of quantitative ultrasound (QUS) of the calcaneus in pre-screening for vertebral/non-vertebral fractures, and in discriminating osteoporotic from normal bone density in patients with ankylosing spondylitis (AS); a second objective is to determine the prevalence of osteoporosis using dual-energy X-ray absorptiometry (DEXA) in this patient group.

Results
Included are 50 consecutive AS patients with no history of osteoporosis: mean (SD) age 52 (12) yrs, range 26-75 yr; female/male ratio 15/35. The mean (SD) DEXA T score in the lumbar spine (AP view) was -0.82 (1.73), mean (SD) DEXA T score in femoral neck -1.46 (1.12). The mean (SD) calcaneal QUS T score was -0.73 (0.95). In our population of AS patients the prevalence of femoral neck osteoporosis according to the WHO definition (DEXA T< -2.5) was 20%.
Osteoporosis criteria were met at the femoral neck in 10 (20%) patients, and 7 of them (70%) were correctly diagnosed using QUS, with T < -1.0 as cut-off value; normal bone density at the femoral neck was found in 15 AS patients (30%), yet in 2 of them the calcaneal QUS T was < -1.0. In AS the 20% pre-test probability of having femoral neck osteoporosis increased using calcaneal QUS, with a cut-off level T< -1.0 (70% sensitivity, 68% specificity), and then rose to 35% as the predictive value of a positive test, yielding a net result of QUS testing of +15%. The predictive value of a negative QUS test result was 90%, which makes QUS applicable to exclude severe osteoporosis. Vertebral and/or non-vertebral fractures occurred in 12 out of 50 AS patients (24%); 5 of them (10%) were associated with osteoporosis as defined by WHO criteria measured via DEXA.

Conclusion
The performance of QUS is similar to DEXA in finding patients with osteoporosis-associated fractures: the sensitivity of QUS T < -1.0 in finding the fracture is 80%, and the sensitivity of femoral neck DEXA T< -2.5 in finding fractured patients is 60%. We conclude that both osteoporosis and fractures are common sequelae in AS. Calcaneal QUS offers a promising approach to screen for osteoporosis, and may be applied to exclude osteoporosis-associated high fracture risk in AS.

Key words
Osteoporosis, osteopenia, fracture, ankylosing spondylitis (AS), quantitative ultrasound (QUS), DEXA.

Please address correspondence and reprint requests to: Tim L.Th.A. Jansen, Rheumatologist, Medisch Centrum Leeuwarden, Department Rheumatology, POB 888, 8901 BR Leeuwarden, The Netherlands.
E-mail: T.Jansen@znb.nl