Mylène Jansen

MRI cartilage thickness after KJD and HTO 279 14 For HTO patients, bi-plane medial-based opening-wedge osteotomy was performed. The method of Miniaci was used to preoperatively define the amount of correction needed and TomoFix medial high tibial plates and screws (DePuy Synthes, Switzerland) or Synthes locking compression plate (LCP) system (DePuy Synthes, Switzerland) were used for fixation. 26 After surgery, partial weight-bearing (maximum 20 kg) was allowed for 6 weeks, after which full weight-bearing was started gradually. Prophylactic anticoagulant was used for 6 weeks. At 18 months after surgery, the metal plate and screws were removed, to allow imaging at 2 years. Image acquisition and analysis 1.5T or 3T MRIs with 3D spoiled gradient recalled imaging sequence with fat suppression (SPGR-fs) were acquired at baseline (before treatment) and 2 years after treatment. While the MRI field strength differed as some patients were included in an extended imaging study, the protocols used were the same for both 1.5T and 3T MRI scans. To prevent bias, only patients who underwent MRI scans of sufficient quality to allow analysis and were scanned with the same hardware (1.5 or 3T) at both baseline and 2 years follow-up were included in the analyses. Reasons for insufficient quality to allow analysis were severe motion artifacts or insufficient positioning (e.g. relevant parts of the joint cut-off). There were no constraints regarding concomitant treatment during the 2 years of follow-up. Cartilage structure in the knee was measured using Chondrometrics Works 3.0 software. 27 The primary and secondary outcome parameters for the present study were the 2-year change in mean cartilage thickness over the total subchondral bone area (ThCtAB) of the MAC and the percentage of denuded subchondral bone area (dABp) of the MAC, respectively. 28 On an exploratory basis, longitudinal changes were determined for 16 femorotibial subregions: the central, internal, external, anterior, and posterior tibia and the central, internal, and external femur for both the MAC and the LAC. 27 Further, location-independent analysis was used to determine the total (summed) thinning and thickening scores across all subregions. 29 Standardized semi-flexed weight-bearing radiographs were performed at the same time points, according to the Buckland-Wright protocol, using an aluminum step wedge as a reference standard for image analysis. 30,31 Using knee images digital analysis (KIDA) software, the mean JSW of the MAC was determined. 32 Both MRI and radiograph analyses were performed by experienced observers blinded to the type of intervention and acquisition order. For the radiograph analyses, 1 observer analyzed all images. For the MRI analyses, 2 different observers analyzed the images, where each of the observers processed pairs of baseline and follow-up of each patient in the same session. Also, the number of patients from each treatment was equally divided between the 2 observers. The reproducibility of both types of analysis have been reported before in detail. 27,32–34