Background Bone loss in patients (pts) with early axial spondyloarthritis is insufficiently studied.
Objectives To assess 1-year bone mineral density (BMD) changes at the femoral neck (FN) and lumbar spine (LS) in pts with early axial spondyloarthritis (axSpA).
Methods 47 pts (20 men and 27 women) with axSpA and <5 years duration of inflammatory back pain (ASAS 2009 criteria) of the Russian CORSAR cohort were observed. Pts included had baseline and 1-year-visit BMD measurements. Pts matched the ASAS criteria and were observed according to ASAS guidelines. At baseline, the median age was 27.0 (24.0–31.0) years, disease duration 22.0 (10.0–34.0) months. 90% of pts were HLA-B27 positive. 46% of pts alongside inflammatory back pain also had peripheral arthritis. Median BASDAI was 3.5 (2.1–5.2), ASDASCRP 2.3 (1.4–3.2), ESR 9.0 (5.0–24.5) mm/h, CRP 3.9 (1.0–19.7) mg/L. Femoral neck (FN) and lumbar spine (LS) (L2–4) BMD were measured in all pts using dual-energy x-ray absorptiometry (DXA). Low BMD was defined as Z score ≤-2 SD (at least in one investigated site). MRI of sacroiliac joints (SIJs), LS and hip joints (HJs) (in case of hip joint involvement) were performed using Signa Ovation 0.35 T scanner (matrix 288x192).
Results In all 47 pts, baseline median Z-score was -0.6 (-1.3 – 0.2) SD for the FN and -0.7 (-1.8 – 0.3) SD for the LS. Z-score for the FN was significantly lower in women: -1.0 (-1.5 – -0.3) SD and 0.0 (-0.75 – 0.4) SD, respectively, p=0.0065. At baseline examination low BMD at least in one of the sites was evidenced in 8 (17%) pts (4 men and 4 women). While all 8 pts had low LS BMD, 3 of them had also low FN BMD. At baseline examination of the 47 pts association between lumbar spine MRI-spondylitis and low BMD was revealed. MRI-spondylitis was evidenced in 8 (17%) pts: 3 (62.5%) of them had low FN and/or LS BMD, while 5 (62.5%) pts of this group had normal BMD values. In contrast, within the group of pts without the LS MRI inflammation (n=39) nobody had low BMD (p=0.0006). Initially, there was found a relationship between high ASDAS activity and low BMD: 11.5% of pts with high ASDASCRP values had low BMD. In contrast, within the group of pts with low or moderate ASDASCRP values nobody had low BMD (p=0.054).
Among the 8 pts with low baseline BMD, 1-year BMD normalized in 3 pts (all women and all initially had low LS BMD). In one patient (male) with normal baseline BMD values low LS BMD at the 1-year visit was observed. Therefore, at 1-year visit fewer pts had low BMD than at baseline examination (n=8, 17% vs n=6, 12.8%, respectively). In males the lowering of LS BMD at 1-year visit was evidenced with much higher confidence than in females (5 among 20 pts, 25%, vs 1 among 27 pts, 3.8%, respectively; p=0.043).
Neither at baseline examination nor at 1-year visit any correlation was detected between low BMD and disease duration, high disease activity according to BASDAI, raised CRP or ESR levels, the presence of MRI-sacroiliitis or MRI-coxitis.
Conclusions In the Russian cohort of axSpA pts there was found an association between active MRI-spondylitis of the LS and BMD reduction in LS and/or FN. The findings corroborate the hypothesis that low BMD in pts with early axSpA is a result of inflammation. Males more often than females had lowered LS BMD after 1 year of observation.
Disclosure of Interest None declared