Despite advances in cartilage repair strategies, treatment of focal chondral lesions remains an important challenge to prevent osteoarthritis. Articular cartilage is organized into several layers and lack of zonal organization of current grafts is held responsible for insufficient biomechanical and biochemical quality of repair-tissue. The aim was to develop a zonal approach for cartilage regeneration to determine whether the outcome can be improved compared to a non-zonal strategy. Hydrogel-filled polycaprolactone (PCL)-constructs with a chondrocyte-seeded upper-layer deemed to induce hyaline cartilage and a mesenchymal stromal cell (MSC)-containing bottom-layer deemed to induce calcified cartilage were compared to chondrocyte-based non-zonal grafts in a minipig model. Grafts showed comparable hardness at implantation and did not cause visible signs of inflammation. After 6 months, X-ray microtomography (µCT)-analysis revealed significant bone-loss in both treatment groups compared to empty controls. PCL-enforcement and some hydrogel-remnants were retained in all defects, but most implants were pressed into the subchondral bone. Despite important heterogeneities, both treatments reached a significantly lower modified O’Driscoll-score compared to empty controls. Thus, PCL may have induced bone-erosion during joint loading and misplacement of grafts in vivo precluding adequate permanent orientation of zones compared to surrounding native cartilage.

Reinforced hydrogels represent a promising strategy for tissue engineering of articular cartilage. They can recreate mechanical and biological characteristics of native articular cartilage and promote cartilage regeneration in combination with mesenchymal stromal cells. One of the limitations of in vivo models for testing the outcome of tissue engineering approaches is implant fixation. The high mechanical stress within the knee joint, as well as the concave and convex cartilage surfaces, makes fixation of reinforced hydrogel challenging. Methods. Different fixation methods for full-thickness chondral defects in minipigs such as fibrin glue, BioGlue®, covering, and direct suturing of nonenforced and enforced constructs were compared. Because of insufficient fixation in chondral defects, superficial osteochondral defects in the femoral trochlea, as well as the femoral condyle, were examined using press-fit fixation. Two different hydrogels (starPEG and PAGE) were compared by 3D-micro-CT (μCT) analysis as well as histological analysis. Results. Our results showed fixation of below 50% for all methods in chondral defects. A superficial osteochondral defect of 1 mm depth was necessary for long-term fixation of a polycaprolactone (PCL)-reinforced hydrogel construct. Press-fit fixation seems to be adapted for a reliable fixation of 95% without confounding effects of glue or suture material. Despite the good integration of our constructs, especially in the starPEG group, visible bone lysis was detected in micro-CT analysis. There was no significant difference between the two hydrogels (starPEG and PAGE) and empty control defects regarding regeneration tissue and cell integration. However, in the starPEG group, more cell-containing hydrogel fragments were found within the defect area. Conclusion. Press-fit fixation in a superficial osteochondral defect in the medial trochlear groove of adult minipigs is a promising fixation method for reinforced hydrogels. To avoid bone lysis, future approaches should focus on multilayered constructs recreating the zonal cartilage as well as the calcified cartilage and the subchondral bone plate.