Background: The Knee Inflammation MRI Scoring System (KIMRISS) is a semi-quantitative grading tool (range 0-500) used to measure features of inflammation in knee osteoarthritis (OA), including compartment-specific BML scores.

Objectives: To investigate; 1) Inter-rater reliability of KIMRISS BML change scores including expert readers and a trainee reader, 2) Validity of KIMRISS BML scores by comparing medial and lateral femoral compartments in surgical and contralateral limbs before and 1 year after medial opening wedge high tibial osteotomy (HTO).

Methods: We included 27 patients with varus alignment and medial compartment knee OA [preoperative Kellgren-Lawrence grades ≤3, age = 53.1 ± 5.9, 22 males (81%), body mass index = 29.9 ± 4.3] undergoing unilateral HTO. Sagittal 2D turbo spin-echo sequences were acquired at 3-Tesla in both knees pre and 1-year post HTO after surgical removal of hardware. Femoral BML scores in the medial and lateral compartments were assessed on images blinded to limb and time by masking surgical evidence. Three reviewers independently graded the same 108 images (27 patients, 2 limbs, 2 time points). We evaluated reliability pre-to-post HTO BML change scores in the medial and lateral femur by calculating intraclass correlation coefficients (ICC) and Bland-Altman plots with 80% Limits of Agreement (LoA) for each rater pair.

Results: Means and standard deviations are shown in Figure 1 . Cumulative percent distribution plots of changes after surgery show both increases and decreases in the medial and lateral femur ( Figure 2 ). When combining all raters, the ICC (95% confidence interval (CI)) for the change in BML score for the surgical knee was 0.81(0.64, 0.91) in the medial femur and 0.73 (0.49, 0.87) in the lateral femur. 80% LOA for change in BML score in the surgical knee was -3.56 to 5.44. ICCs and LoAs for rater pairs are in Table 1 . The surgical limb medial femur BML score was significantly greater than all other scores pre and post HTO ( Figure 1 ). There were no significant changes after HTO. The mean change (95% CI) in medial femur BML score of the surgical knee was 1.08 (-1.21, 3.38). Using the mean scores of all raters, we compared BML scores in surgical and contralateral limbs before and after HTO using a 2-way repeated measures (limb by time) ANOVA, seperately for the medial and lateral compartments.

Fig 1.

Means and standard deviations for KIMRISS femoral BML scores in the medial and lateral compartment for the surgical and contralateral knee.

Fig 2.

Probability plots for the 12-month change after surgery in femoral BML score in the medial (a) and lateral (b) compartments of the surgical knee. Change scores from each of the three raters is shown, as well as the average of all four raters.

Conclusion: The KIMRISS can reliably detect differences between femoral BML scores in symptomatic and contralateral limbs, supporting the inter-rater reliability, feasibility and validity of compartment-specific BML scores.

Disclosure of Interests: Jenna Schulz: None declared, Trevor Birmingham: None declared, Walter P Maksymowych Grant/research support from: Received research and/or educational grants from Abbvie, Novartis, Pfizer, UCB, Consultant of: WPM is Chief Medical Officer of CARE Arthritis Limited, has received consultant/participated in advisory boards for Abbvie, Boehringer, Celgene, Eli-Lilly, Galapagos, Gilead, Janssen, Novartis, Pfizer, UCB, Speakers bureau: Received speaker fees from Abbvie, Janssen, Novartis, Pfizer, UCB., Robert G Lambert: None declared, Stephany Pritchett: None declared, Frank Beier: None declared, J. Robert Giffin: None declared, Thomas Appleton Grant/research support from: AbbVie and Pfizer