Clinical Radiologists, MRI Researcher, Osteoarthritis Clinical Trial Researcher

Plain radiographs are insensitive to early osteoarthritic (OA) change and at present any intervention that may halt progression of OA needs to be applied in the very early stages before tissue loss. Early OA is characterized by loss of proteoglycans[1], which are the fixed charge elements in the cartilage extracellular matrix. Delayed Gadolinium Enhanced MRI of Cartilage (dGEMRIC) is a contrast MRI technique that allows the detection of loss of charge in cartilage. It has been validated extensively in vitro and in vivo[2]. dGEMRIC has been shown to be a good predictor of outcome after pelvic osteotomy for acetabular dysplasia[3]. It has shown to correlate with clinical symptoms in both FAI[4] and dysplastic hips[5] and also correlate with the severity of acetabular dysplasia[6].

Pelvic osteotomy for acetabular dysplasia corrects the mechanical abnormality in the hip and has been shown to improve clinical symptoms. In the dysplastic hip the femoral coverage is insufficient, which will lead to increased mechanical stress on the articular cartilage. Initially, the articular cartilage will respond to the increased mechanical load by increasing charge density. Eventually, the tissue will degenerate leading to OA if the increase mechanical stress is beyond the capacity of the tissue to adapt. Although clinical symptoms improve after osteotomy, we do not know what happens to the articular cartilage at the molecular level. dGEMRIC technique is ideal to address this question. Similar unanswered questions remain when cam decompression is performed in femoroacetabular impingement (FAI).

We performed a prospective cohort study to evaluate the dGEMRIC index of articular hip cartilage and the morphologic changes in the labrum, cartilage, and bone preoperatively and at twelve and twenty-four months following periacetabular osteotomy for correction of developmental dysplasia of the hip. The hypothesis was that the dGEMRIC index (likely reflecting the proteoglycan content of cartilage) is modulated by alteration of the hip joint biomechanics.

Thirty-seven patients (thirty-seven hips) with no or minimal osteoarthritis were treated with periacetabular osteotomy for symptomatic acetabular dysplasia. All patients had preoperative and one-year follow-up dGEMRIC scans. Twenty-eight of the thirty-seven also had two-year scans. The changes in dGEMRIC findings and hip morphology between the preoperative visit and the examinations at one and two years following the periacetabular osteotomy were assessed. MRI was performed with use of a single 1.5-T system (MAGNETOM Avanto; Siemens Healthcare, Erlangen, Germany) and a flexible surface coil. A standardized protocol was used for dGEMRIC[7]. A three-dimensional isotropic dual-flip-angle technique was used for T1 mapping with voxel size of 0.83mm. For morphologic evaluation, an isotropic true-FISP (fast imaging with steady-state precession) sequence was used with an isotropic voxel size of 0.63 mm. The isotropic data sets of the T1-mapping and true-FISP sequences were used for radial reconstruction and evaluation. To account for the change in acetabular position after the osteotomy and to avoid partial volume artifacts, we generated separate radial reformats for the acetabulum and the femur using osseous anatomic landmarks. The region-of-interest evaluations were performed manually by a trained reader on anonymized T1 data sets. A musculoskeletal radiologist performed morphologic grading of the labrum, cartilage, and bone on the true-FISP reformats.

The mean preoperative dGEMRIC index (and standard deviation) was 561.6 ± 117.6 ms; this decreased to 515.2 ± 118.4 ms at one year after periacetabular osteotomy but subsequently recovered to 529.2 ± 99.1 ms at two years postoperatively. The decrease in the dGEMRIC index of the acetabular cartilage after surgery to normal levels appears to be most pronounced at the superior aspect of the acetabulum, where the decrease in mechanical loading after periacetabular osteotomy would be most pronounced. There was no evidence of OA progression on radiographs or on morphologic MRI. All domains of the Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) demonstrated significant improvement from the preoperative to the postoperative visits (all p < 0.001).

Cartilage is a biological tissue that responds to mechanical loading. Initially, with increased mechanical load the articular cartilage density would increase, which will manifest as an increase in T1 value on dGEMRIC scans. Periacetabular osteotomy for developmental dysplasia of the hip appears to alter the mechanical loading of articular cartilage in the hip, which in turn alters the cartilage matrix composition, as demonstrated by dGEMRIC.