Ilhanli İlker, Güder Necip, Tosun Alptekin, Avci Esin, Çelik Canan
Department of Physical Medicine and Rehabilitation, Medical Faculty of Giresun University, Giresun, Turkey.
Department of Physical Medicine and Rehabilitation, Physical Medicine and Rehabilitation State Hospital, Giresun, Turkey.
Arch Rheumatol. 2017 Apr 17;32(4):339-346. doi: 10.5606/ArchRheumatol.2017.5940. eCollection 2017 Dec.
This study aims to assess the validity and reproducibility of computer-assisted joint space area measurement in knee roentgenograms of patients with knee osteoarthritis and compare it with a qualitative method in knee roentgenograms and quantitative and semi-quantitative methods in magnetic resonance imaging.
The study included 40 knees of 40 patients diagnosed as osteoarthritis (14 males, 26 females; mean age 57.4±5.9 years; range 47 to 67 years). Only the patients who wrote consents for publication of their radiologic data, and with knee roentgenograms and magnetic resonance images of the same knees were selected. Computer-assisted measurements were applied to joint spaces by two blinded physicians, for two times with an interval of one week. Data were evaluated for intraobserver and interobserver consistency. Also, data were compared with qualitative (Kellgren-Lawrence classification), quantitative (joint space width, cartilage thickness, meniscal thickness in magnetic resonance images) and semi-quantitative methods (whole-organ magnetic resonance imaging score).
Intraobserver consistency was evaluated for each physician, which revealed no differences. Interobserver consistency was evaluated by comparing the measurements of two blinded physicians and no differences were found (p>0.05). There was no significant correlation between the grade of Kellgren-Lawrence classification and other variables; such as grade of meniscus, meniscal thickness, cartilage thickness and computer- assisted joint space area measurements (p>0.05). While there was a positive correlation between computer-assisted joint space area measurement and other quantitative measurements, there was a negative correlation between computer-assisted joint space area measurement and whole-organ magnetic resonance imaging scores.
When compared with qualitative, quantitative, and semi-quantitative methods, computer-assisted joint space area measurement seems to be a useful, reproducible, and cost-effective quantitative method for evaluating knee osteoarthritis.
本研究旨在评估计算机辅助测量膝关节骨关节炎患者膝关节X线片关节间隙面积的有效性和可重复性，并将其与膝关节X线片的定性方法以及磁共振成像的定量和半定量方法进行比较。
本研究纳入40例被诊断为骨关节炎的患者的40个膝关节（男性14例，女性26例；平均年龄57.4±5.9岁；年龄范围47至67岁）。仅选择那些同意发表其放射学数据、且有同一膝关节的X线片和磁共振图像的患者。两名不知情的医生对关节间隙进行计算机辅助测量，测量两次，间隔一周。评估数据的观察者内和观察者间一致性。此外，将数据与定性（Kellgren-Lawrence分级）、定量（磁共振图像中的关节间隙宽度、软骨厚度、半月板厚度）和半定量方法（全器官磁共振成像评分）进行比较。
对每位医生评估观察者内一致性，结果显示无差异。通过比较两名不知情医生的测量结果评估观察者间一致性，未发现差异（p>0.05）。Kellgren-Lawrence分级与其他变量之间无显著相关性；如半月板分级、半月板厚度、软骨厚度和计算机辅助关节间隙面积测量（p>0.05）。虽然计算机辅助关节间隙面积测量与其他定量测量之间呈正相关，但计算机辅助关节间隙面积测量与全器官磁共振成像评分之间呈负相关。
与定性、定量和半定量方法相比，计算机辅助关节间隙面积测量似乎是一种用于评估膝关节骨关节炎的有用、可重复且具有成本效益的定量方法。
