To investigate the correlation between X-ray and preoperative magnetic resonance imaging (MRI) cartilage grading in patients with degenerative knee osteoarthritis (OA) after arthroscopic surgery.

A total of 91 cases of elder patients with knee OA underwent arthroscopic surgery were retrospectively reviewed, including 25 males and 66 females. The patients aged from 60 to 85 years who met the diagnostic criteria for knee osteoarthritis with complete preoperative imaging data (X-ray and MRI), Kellgren-Lawrence (K-L) grade zero to Ⅲ, received arthroscopy alone, and could be followed up after surgery were selected. The patients who underwent arthroscopy combined with other treatment, image data incomplete, and could not be followed-up were excluded. According to the postoperative curative effect duration, the patients were divided into the effective group and the invalid group. All the patients were followed-up for 24 months. The preoperative and postoperative knee injury& osteoarthritis score (KOOS) were evaluated. Knee images of preoperative X-ray and MRI classification were retrospectively analysed, including K-L grades of X-ray, the whole-organ magnetic resonance imaging score(WORMS)of MRI. The correlation of preoperative imaging classification (K-L grades and WORMS) and the duration of curative effect were analysed. Student t test, covariance analysis and Spearman correlation analysis were used for differences analysis in the groups and between the groups; correlation was analyzed by multiple linear regression analysis.

Among 91 patients with 91 knee joints, the average age was (69.3±6.3) years, the effective group were 32 patients of K-L grade Ⅱ, one patient with WORMS grade 2.5, 15 patients were WORMS grade three and 16 patients were WORMS grade four. The invalid group were 56 patients, among which 31 patients were K-L grade Ⅱ, 28 patients with K-L grade Ⅲ, 10 patients were WORMS grade four, 29 patients were WORMS grade five, and 20 patients were WORMS grade six. The postoperative KOOS scores were higher than those before surgery, but the postoperative KOOS score of the effective group was significantly higher than that of the ineffective group. The univariate covariance analysis on postoperative KOOS score showed that the postoperative scores change of the two groups had statistically significant differences(F=134.813, 56.785, 138.996, 28.502, 58.799, all P<0.001), the postoperative score of the effective group was significantly higher than the invalid group. The multiple regression analysis showed that MRI classification had statistically significant effects on the postoperative treatment duration(t=-12.526, P<0.001). The MRI grading had a higher correlation with postoperative efficacy duration (r=-0.927, P<0.001). The survival curve showed that the postoperative efficacy time became shorter with the increase of MRI grading(χ²=111.0, P<0.001).

Combined with WORMS classification of MRI, it is helpful to identify early knee osteoarthritis of elderly patients. Arthroscopic palliation plus bone marrow stimulation can effectively relieve symptoms, which is worthy of clinical application.