T2 Mapping versus Diffusion-weighted Imaging for Assessment of Cartilage Repair Tissue After Matrix-associated Autologous Chondrocyte Transplantation at 3.0 Tesla

Cartilage defects are common injuries in orthopaedics and trauma surgery, and isolated cartilage defects have been found in 37% of patients in a large series of arthroscopies.¹ Since articular cartilage has a limited capacity for spontaneous repair, in the last decade several surgical cartilage repair techniques have been developed:

• Microfracture has been shown to be an efficient one-step procedure, but produces mainly fibrous repair tissue with an incomplete filling of the defect and limited load-bearing capacity.^2–4
• In osteochondral autologous transplantation (OAT), osteochondral plugs are taken from non- or less frequent weight-bearing areas of the femoral condyles using a cylindrical cutting device and are implanted as a mosaic to fill the defect(s). OAT is limited with respect to the size of the defect that can be filled (the maximum size is 4cm2) because only a limited number of grafts are available, in a large defect the fixation of the grafts becomes instable and may result in uneven surfaces and impairment occurs due to the mechanical forces used during implantation, which may injure the cartilage layer of the osteochondral plug. ⁵
• Autologous chondrocyte implantation (ACI) requires the excision of a periosteal flap to keep the injected cultured autologous cell suspension in situ. ACI has been applied to 30,000 patients worldwide;6 however, cartilage overgrowth and delamination or fibrous degeneration of the newly formed tissue have been observed in 2.4–20% of cases.^7,8 As a consequence, there is substantial interest in improving ACI.
• New ACI techniques are often referred to as scaffold-guided or matrix-associated ACI (MACI), since biomaterials based on collagen,^9–11 hyaluronan^12–15 or polylactides16 are used as scaffolds for cell growth. MACI is less invasive and can be performed arthroscopically in central–anterior defects of the femoral condyle. Moreover, it can treat defects up to 10cm2.^17–20 An additional advantage may be more efficient redifferentiation of chondrocytes and hence the formation of hyaline-like repair tissue.²¹

In our study, Hyalograft C™ was used as a cell carrier. Matrix-associated autologous chrondrocyte transplantation (MACT) was performed as a two-stage procedure. Initially, approximately 150–200mg of healthy cartilage was arthroscopically obtained from the intercondylar notch; the cells were extracted from the cartilage matrix, grown in culture and then transferred onto a hyaluronic acid sponge called Hyaff 11™ (Fida Advanced Biopolymers, Abano Terme, Italy). In the second stage, the chondral defect was exposed using mini-arthrotomy. The cartilage defect was debrided to the subchondral bone and all unstable cartilage was removed. After determination of the defect size, the cell matrix transplants were fitted into the defect and the edges were fixed with fibrin glue.

With these developments in cartilage repair surgery, magnetic resonance imaging (MRI) of the articular cartilage has become increasingly important in pre-operative evaluation to select patients suitable for cartilage repair surgery. These new surgical techniques require reliable, non-invasive methods to monitor cartilage repair tissue over time to detect complications and variations of the normal maturation process at an early stage. As serial cartilage biopsy, which can be used for assessing cartilage repair, is not feasible in many patients because of its invasive nature and associated complications, the role of MRI is becoming even more important. ^22–27 Due to the increased use of clinical 3.0 Tesla (T)
scanners, superior gradient strengths and the use of dedicated coils, MRI of cartilage has significantly improved because of the higher signal-tonoise ratio (SNR) and shorter scan times possible with 3.0T imaging. ^28,29

Beyond the assessment of morphological characteristics of cartilage repair tissue, it is necessary to monitor the maturation process of the cartilage implant with regard to structural and biochemical development. Various MR sequences can visualise the ultrastructural components and biochemical composition of articular cartilage. Biochemical imaging at 3.0T benefits most from the higher SNR and its potential for higher-resolution scanning. ³⁰

One technique is T1 mapping with delayed gadolinium-enhanced MRI of cartilage (dGEMRIC), which is sensitive to the distribution and content of proteoglycans. ^31–33 Another technique is T2 mapping, which provides direct information about the water content, concentration and orientation of collagen, and therefore reflects structural variations within the cartilage. ^34–36 Both dGEMRIC and T2 mapping are now commonly used. Diffusion-weighted imaging (DWI) is another promising new technique for the assessment of cartilage. ^37–39 DWI is based on Brownian molecular motion, which is influenced by intra- and extra-cellular barriers. ^40,41 Measuring the molecular movement in articular cartilage reflects the biochemical structure and architecture of the tissue. ^42,43 Conventional DWI, based on spin-echo imaging, is relatively insensitive to susceptibility artefacts, but requires a rather long acquisition time, which limits its clinical feasibility. ⁴⁴ Echo-planar imaging (EPI)-based diffusion sequences, the gold standard for neuroradiological applications, are prone to susceptibility artefacts and have limitations with respect to contrast, as long echo times are needed. ⁴⁵ This renders these sequences impractical for low-T2 tissues such as muscle and cartilage.46 Alternatively, DWI can be based on steady-state free precession (SSFP) sequences, which can accomplish diffusion weighting in relatively short echo times. This is achieved by the application of a mono-polar diffusion-sensitising gradient, which, under steady-state conditions, leads to a diffusion weighting of consecutive echoes (spin echoes and stimulated echoes). For the assessment of diffusion-weighted images in our study, a 3D steadystate diffusion technique called reversed fast imaging with steady-state precession (PSIF) was used. ^47,48