原文

译文

FRI0502

MRI ACTIVITY IN THE SACROILIAC JOINTS LEADS TO STRUCTURAL DAMAGE ON MRI IN PATIENTS WITH EARLY INFLAMMATORY LOW BACK PAIN: A LONGITUDINAL ANALYSIS IN THE ESPAC STUDY.

M. van Onna ^1,*, A.-G. Jurik ², D. van der Heijde ³, A. van Tubergen ¹, L. Heuft-Dorenbosch ⁴, R. Landewé ¹

¹Rheumatology, UNIVERISITY HOSPITAL MAASTRICHT, Maastricht, Netherlands, ²Radiology, Aarhus University Hospital, Aarhus, Denmark, ³Rheumatology, Leiden University Medical Center, Leiden, ⁴Rheumatology, Laurentius Hospital Roermond, Roermond, Netherlands

Background: There is limited knowledge about the development and rate of progression of MRI structural changes of the sacroiliac (SI) joints, and about the relationship between inflammation and structural changes especially in patients with early spondylarthritis (SpA).

Objectives: To assess whether MRI inflammation leads to (progression of) structural changes on both MRI and conventional radiographs (CRs) of the SI joints in patients with early inflammatory back pain (IBP).

Methods: A 2-year follow-up study was conducted in patients (38% male; mean (SD) age 34.9 (10.3) years) with IBP of less than 2 years duration who were included in the Early Spondyloarthritis Cohort (ESpAC). MRI (T1 pre- and post gadolinium (Gd) and Short Tau Inversion Recovery (STIR)) and CRs of the SI joints were obtained at baseline, 1 and 2 years follow-up. The MRIs were scored by one experienced reader for both active and structural lesions. MRIs were scored as a series per patient with blinding for time order. A MRI was considered positive if at least one lesion was present on 2 successive slices or in case of more than one lesion in 1 slice (in analogy to the ASAS/OMERACT consensus for MRI active lesions). CRs were scored according to the modified New York Criteria (mNYC) by one (different) observer. A GEE model was developed that correlated MRI structural changes or CR damage to MRI activity 1 year earlier (autoregressive time-lag model). This was done on a per SI-joint level, using both the dichotomous and continuous MRI scores as dependent variable.

Results: Of the 68 patients enrolled at baseline, 24 patients (35%) had signs of MRI activity (of which 66% were HLA-B27 positive) and 17 patients (25%) (of which 71% were HLA-B27 positive) had structural changes at baseline. Fourteen patients (of which 86% were HLA-B27 positive) fulfilled the mNYC at baseline. MRI erosions increased significantly during follow-up (p = 0.046) whilst the overall CR damage score did not change during follow-up. MRI activity was associated with an increase in MRI structural changes 1 year later, both for erosion and fatty degeneration score (table 1). The GEE analysis did not show a similar relationship between MRI activity and CR progression score.

Conclusions: In patients presenting with IBP of recent onset, MRI activity is correlated with progression of structural changes on MRI one year later. This may imply that MRI inflammation precedes erosions and fatty degeneration in the SI-joints in a causal manner.

早期IBP病人骶髂关节MRI炎症与1年后MRI结构破坏之间的关系

van Onna M, et al. EULAR 2011. Present No: FRI0502.

背景：目前对骶髂关节(SIJ)MRI结构破坏的演变和进展速率，以及炎症与结构改变之间的关系，尤其是在早期脊柱关节炎(SpA)病人中，知之甚少。

目的：在早期炎性背痛(IBP)患者中评估MRI炎症是否会导致(或演变为)MRI与传统放射学(CR)可见的结构损害。

方法：病程小于2年的IBP病人均来自早期SpA队列(ESpAC)，将接受2年的随访。其中男性38%，平均年龄34.9岁(SD: 10.3)。在基线、随访1年和2年时进行传统X线与MRI检查，所用方法包括增强前后的T1以及短时反转回复(STIR)。由一名资深读片师对MRI成像中的活动性和结构性损害进行评分。阅片时遮蔽病人信息和摄片时期。当连续2个扫描层面均发现至少1个病灶或同一层面至少有2个病灶，则判定为阳性MRI发现（与ASAS/OMERACT有关MRI活动性损害的定义相似）。CR评分遵循纽约修订版分类标准，由另一位读片师进行操作。我们开发了一个GEE模型，将MRI或CR结构变化与1年前MRI活动性病变进行关联分析(自回归时间滞后模型)。相关分析是在每个SIJ水平进行，采用二分式和连续式MRI评分作为因变量。

结果：入组68例病人。基线时，24例病人(35%)有MRI活动性征像(HLA-B27阳性者占66%)，17例病人(25%)有结构性损伤(HLA-B27阳性者占71%)。基线时有14例病人符合纽约改良版分类标准。与整个随访过程中CR评分无变化相比，MRI侵蚀评分上升显著(p=0.046)。基线时MRI活动度与随访1年后MRI结构损伤增加有关联性，侵蚀和脂肪沉积均如此(表1)。GEE分析没有发现MRI活动度与放射学进展之间存在关联。

结论：新发病IBP病人中MRI活动度与一年后MRI结构损伤进展之间存在关联。这可能意味着MRI炎症出现于SIJ侵蚀病变和脂肪变性之前，并可能与后两者之间存在因果关系。

Table 1: Autoregressive GEE showing the correlation between MRI activity and the progression of MRI structural changes during follow-up in 68 patients included in ESpAC.

表1. 针对EspAC队列中68病人进行随访随访并用自回归GEE分析MRI活动度与MRI结构性损害在随访1年之后的进展之间存在相关性