Background/Purpose: Thoracic and/or thoraco-abdomino-pelvic (TAP) Computed Tomography (CT) may be performed during the follow up of patients with ankylosing spondylitis (AS) and are able to assess vertebral fracture and recently to assess bone fragility thought the scanographic bone attenuation coefficient of the first lumbar vertebra (SBAC-L1).   

The objective is to evaluate the correlation between the risk of bone fragility measured by the SBAC-L1 and spine structural severity assessed by the S (Modified stoke ankylosing spondylitis spinal score) in patients with AS.

Methods: This monocentric retrospective study included patients with spondyloarthritis followed from 2009 to 2017 at the University hospital of Nancy. Patients should fulfill New-York or ASAS 2009 criteria with radiographic sacroiliitis and a thoracic or TAP CT scan and radiographies (spine, pelvis) performed with a delay which did not exceed 2 years. Osteoporotic risk factors, Dual Energy X-ray Absorptiometry (DXA) measurements and clinical characteristics were collected. The mSASSS was performed by two readers with adjudications in case of discordance. The definition of structural spine involvement is retained for a mSASSS ≥ 2. Vertebral fractures were studied on sagittal spine radiographies according to Genant classification. The SBAC-L1 was measured in Hounsfield Units (HU) on axial L1 section in trabecular bone on CT. Intra- and inter-reader reliabilities for mSASSS and vertebral fractures were calculated. A SBAC L1 ≤ 145 HU (fracture threshold) defines patients at risk of VF.

Results: A total of 73 AS patients were included (age median: 60 (53-68.5) years, 8 women (11%)), disease duration median: 24 years (12-34)). Sixty patients (82.2%) have a mSASSS ≥ 2, with a mean score of 20.7 (±21.2). Presence of partial ankylosis is observed in 37 patients (50.7%). Fifty-three patients (72.6%) presented at least one clinical risk factor of osteoporosis. Assessment of osteoporosis was explored by DXA of the spine in only 16 patients (21.9%) and 3 (18.8%) of them presented a T-score ≤ -2.5 SD. T-score of AS patients with mSASSS + tended to be higher in comparison to AS patients with mSASSS – (p=0.051). Thirteen VFs were detected in 9 patients (12.3%) of the total population, 8 from them have a mSASSS ≥ 2 and 5 have de SBAC-L1 ≤ 145 HU. The mean SBAC-L1 was 141.1 HU (±45) in the whole population, 138.1 HU for mSASSS + and 154.8 for mSASSS – respectively. Forty-two patients (57.5%) presented a SBAC-L1 ≤ 145 HU: 60% in mSASSS + and 46.2% in mSASSS-. Patients with bone bridge have a lower SBAC-L1 (123.96 ± 41.1 HU) than patients mSASSS + without vertebral ankylosis (160.4 ± 41.9 HU) (p=0.02). The number of AS patients under the fracture threshold was higher in mSASSS+ in comparison to mSASSS – (73% vs 41.9%, p=0.006). The reproducibilities are very good (over 0.8) for mSASSS, poor to moderate for VF (0.3 to 0.6) and good to excellent for radiographic sacroiliitis.

Conclusion: There is no relation between the risk of bone fragility measured by the SBAC-L1 and spine structural severity assessed by the mSASSS in patients with AS, however patients with bone bridge have significantly lower SBAC-L1 and more of them are under the fracture threshold compared to patients without bone.

« Back to 2019 ACR/ARP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/relationship-between-structural-spine-involvement-and-the-scanographic-bone-attenuation-coefficient-of-l1-in-a-population-of-73-patients-with-ankylosing-spondylitis/