Kortekaas Marion C, Kwok Wing-Yee, Reijnierse Monique, Wolterbeek Ron, Bøyesen Pernille, van der Heijde Desiree, Bloem Johannes L, Kloppenburg Margreet
From the Department of Rheumatology, Department of Radiology, and Department of Medical Statistics, Leiden University Medical Centre, Leiden, the Netherlands; Department of Rheumatology Diakonhjemmet Hospital, Oslo, Norway.M.C. Kortekaas, MD; W.Y. Kwok, MD, PhD, Department of Rheumatology, Leiden University Medical Centre; M. Reijnierse, MD, PhD, Department of Radiology, Leiden University Medical Center; R. Wolterbeek, PhD, Department of Medical Statistics, Leiden University Medical Center; P. Bøyesen, MD, PhD, Department of Rheumatology, Diakonhjemmet Hospital; D. van der Heijde, MD, PhD, Department of Rheumatology, Leiden University Medical Centre, and Department of Rheumatology, Diakonhjemmet Hospital; J.L. Bloem, MD, PhD, Department of Radiology, Leiden University Medical Center; M. Kloppenburg, MD, PhD, Department of Rheumatology, Leiden University Medical Centre.
J Rheumatol. 2015 Jul;42(7):1224-30. doi: 10.3899/jrheum.140338. Epub 2015 May 15.
To investigate criterion validity and intraobserver reliability of magnetic resonance imaging (MRI) in hand osteoarthritis (HOA).
In 16 patients with HOA (median age 57 yrs, 62% women, 13 with erosive OA), 3 Tesla MRI scans with gadolinium-chelate administration of right second to fifth distal interphalangeal/proximal interphalangeal joints were scored according to the Oslo HOA scoring method for synovial thickening, bone marrow lesions (BML), osteophytes, joint space narrowing (JSN), and erosions (grade 0-3). Ultrasound (US) was scored for synovial thickening and osteophytes, radiographs for osteophytes and JSN (Osteoarthritis Research Society International score), and anatomical phases (Verbruggen-Veys score). Pain was assessed during physical examination. Correlations of MRI with US and radiographic features were assessed with generalizability theory. With generalized estimating equations analyses, MRI features were associated with pain, adjusting for confounding.
Forty-three percent, 27%, 77%, and 61% of joints had synovial thickening (moderate/severe), BML, osteophytes, and erosions on MRI, respectively. Intraobserver reliability, assessed in 6 patients, was good (ICC 0.77-1.00). Correlations between osteophytes, JSN, and erosions on radiographs and MRI were moderate, substantial, and fair (ICC 0.53, 0.68, and 0.32, respectively); MRI showed more lesions than radiography. Correlation between synovial thickening and osteophytes on MRI and US was moderate (ICC 0.43 and 0.49, respectively). MRI was more sensitive for synovial thickening, US for osteophytes. Pain was associated with moderate/severe synovial thickening (adjusted OR 2.4, 95% CI 1.06-5.5), collateral ligaments (4.2, 2.2-8.3), BML (3.5, 1.6-7.7), erosions (4.5, 1.7-12.2), and osteophytes (2.4, 1.1-5.2).
MRI is a reliable and valid method to assess inflammatory and structural features in HOA. It gives additional information over radiographs and US.
探讨磁共振成像(MRI)对手部骨关节炎(HOA)的标准效度和观察者内信度。
对16例HOA患者(中位年龄57岁,62%为女性,13例为侵蚀性骨关节炎),采用3特斯拉MRI扫描,静脉注射钆螯合物,对右手第二至第五远端指间关节/近端指间关节,按照奥斯陆手部骨关节炎评分方法对滑膜增厚、骨髓病变(BML)、骨赘、关节间隙狭窄(JSN)和侵蚀(0 - 3级)进行评分。超声(US)对滑膜增厚和骨赘进行评分,X线片对骨赘和JSN(国际骨关节炎研究学会评分)以及解剖分期(Verbruggen - Veys评分)进行评分。在体格检查时评估疼痛情况。采用可推广性理论评估MRI与US及X线特征的相关性。通过广义估计方程分析,对混杂因素进行校正后,评估MRI特征与疼痛的相关性。
MRI检查显示,分别有43%、27%、77%和61%的关节存在滑膜增厚(中度/重度)、BML、骨赘和侵蚀。在6例患者中评估的观察者内信度良好(组内相关系数ICC为0.77 - 1.00)。X线片和MRI上骨赘、JSN和侵蚀之间的相关性分别为中度、高度和一般(ICC分别为0.53、0.68和0.32);MRI显示的病变比X线片更多。MRI和US上滑膜增厚与骨赘之间的相关性为中度(ICC分别为0.43和0.49)。MRI对滑膜增厚更敏感,US对骨赘更敏感。疼痛与中度/重度滑膜增厚(校正比值比OR 2.4,95%置信区间CI 1.06 - 5.5)、侧副韧带(4.2,2.2 - 8.3)、BML(3.5,1.6 - 7.7)、侵蚀(4.5,1.7 - 12.2)和骨赘(2.4,1.1 - 5.2)相关。
MRI是评估HOA炎症和结构特征的可靠且有效的方法。它比X线片和US能提供更多信息。