OP0222 STUDY OF VERTEBRAL FRACTURE PREVALENCE AND SCANOGRAPHIC BONE ATTENUATION COEFFICIENT (SBAC-L1) IN PATIENTS WITH RHEUMATOID ARTHRITIS AND ANKYLOSING SPONDYLITIS VS. CONTROLS

Background Osteoporosis is a common disease whose prognosis can be seriously impacted by the development of fractures that lead to functional limitations and may even have life-threatening sequelae. This disease is often under-screened, especially in at-risk populations that require multidisciplinary care such as patients with rheumatoid arthritis (RA) and ankylosing spondylitis (AS) [1,2,3,4,5,6]. Moreover, a recent study showed that the scanographic bone attenuation coefficient of the first lumbar vertebra (SBAC-L1) with a threshold at 145 Hounsfield Units (HU) identified 96.6% of patients in the general population with a vertebral fracture (VF), whereas DEXA (with a T-score ≤ -2,5) identified only 39% of these patients [7]. Objectives The aim of the study was to identify the prevalence of vertebral fractures (VFs) and to measure the scanographic bone attenuation coefficient of the first lumbar vertebra (SBAC-L1) based on CT-scan examinations of patients with rheumatoid arthritis (RA), patients with ankylosing spondylitis (AS) and in a control group. Methods This monocentric and retrospective study included patients who were evaluated between 2009 and 2017 with a diagnosis of RA based on the ACR/EULAR criteria, those with a diagnosis of AS based on the New-York criteria, and a RA-matched control group. All of the patients received a CT-scan. The osteoporosis risk factors, data from dual energy X-ray absorptiometry (DEXA) and clinical characteristics were collected. VFs were determined via CT-scans according to the Genant classification, and the SBAC-L1 was measured in Hounsfield units (HU). SBAC-L1 ≤ 145 HU (fracture threshold) defined patients at risk of VFs. Results A total of 244 patients were included (105 RA, 83 AS, 56 controls). The AS group was younger and primarily consisted of males. Of the 4,365 vertebrae studied, 66 osteoporotic VFs were found in 36 patients: 18 (17.1%) patients with RA, 13 (15.7%) patients with AS and 5 (8.9%) controls. The SBAC-L1 was 142.2 (±48.4) HU for the RA group and 142.8 (±48.2) for the AS group, both of which were significantly lower than that of the control group (161.8 (±42.7) HU). Of the 36 patients with VFs and rheumatism, 28% had a T-score ≤ -2.5 SD, and 71.4% had a SBAC-L1 ≤ 145 HU. A T-score ≤ -2.5 SD and a SBAC-L1 ≤ 145 HU were associated with the presence of a VF (OR = 2.35 [CI95%: 1.12-4.92] and 2.06 [CI95%: 1.04-4.10]), respectively. Conclusion The SBAC-L1 was significantly lower in the RA and AS groups than in the control group. Furthermore, SBAC-L1 ≤ 145 HU was associated with a higher risk of VFs, with an odds ratio similar to that of a DEXA. References [1] Toledano E. Reumatol Clin. 2012 Nov-Dec;8(6):334-41. [2] Dougado M. Ann Rheum Dis. 2014;73:62-8. [3] McKeown E. Rheumatol Oxf Engl. 2012;51:1662-9. [4] Van der Weijden MA. Clin Rheumatol. 2012 Nov;31(11):1529-35. [5] Ghozlani I. Bone. 2009 May;44(5):772-6. [6] Ardizzone M. Rev Med Interne. 2006 May;27(5):392-9. [7] Pickhardt PJ. Ann Intern Med. 2013;158(8):588–95. Disclosure of Interests None declared