Articular cartilage repair is assumed to improve by covering the cartilage lesion with a biomaterial scaffold tailored to the specific requirements of the weight-bearing joint surface. We have tested the feasibility of a novel composite collagen-polylactide scaffold rhCo-PLA in cartilage repair. To confirm these results and further challenge the scaffold, we tested it in a large porcine cartilage defect. A critical-sized full-thickness chondral defect was made in the medial femoral condyle of 18 domestic pigs. This technically widest possible defect size of 11×17 mm was determined in a pilot test. Five weeks later, the defect was either treated with the novel rhCo-PLA scaffold or left untreated to heal spontaneously. After four months, the medial condyles were evaluated macroscopically using Goebel's score, in which the worst possible result receives a total of 20 points and imaged with µCT to evaluate subchondral bone. Macroscopic score and subchondral bone microstructure were similar in both study groups. The total Goebel score was higher in spontaneous group (9.75±3.9 for spontaneous and 9.1±3.7 for rhCo-PLA, respectively) but differences between individual animals were large. Subchondral bone volume fraction was 48.2±3.6% for rhCo-PLA and 44.2±3.4% for spontaneous. Trabecular thickness was greater in operated joints (207.9±18.8 µm for spontaneous and 242.9±32.9 µm for rhCo-PLA) than in contralateral non-operated joints (193.3±15.1 µm and 213.4±33.2 µm, respectively). These preliminary data demonstrate that individual differences in the macroscopic appearance were large but there were no significant differences between the two study groups in the score or subchondral bone structure.

Use of scaffolds for articular cartilage repair (ACR) has increased over the last years with many biomaterials options suggested for this purpose. It is known that scaffolds for ACR have to be optimally biodegradable with simultaneous promotion of chondrogenesis, favouring hyaline cartilage formation under rather complex biomechanical and physiological conditions. Whereas improvement of the scaffolds by their conditioning with stem cells or adult chondrocytes can be employed in bioreactors, “ideal” scaffolds should be capable of performing such functions directly after implantation. It was previously considered that scaffold structure and composition would be the best if it mimics the structure of native cartilage. However, in this case no clear reparative stimuli are being imposed on the scaffold area, which would drive chondrocytes activity in a desired way.

In this work, we studied new xeno-free, recombinant human type III collagen-laden polylactide (PLA) mesh scaffolds, which have been designed, produced, and biomechanically optimized in vitro and in vivo validated in a porcine and equine model. The scaffolds were additionally assessed for relative performance simulated synovial fluids for both human conditions and veterinary cases.

It was experimentally shown that success of the scaffolds in ACR eventually require lower stiffness than surrounding cartilage yet matching the strain compliance, different in static and dynamic conditions. This ensures an optimal combination of load transfer and oscillatory nutrients supply to the cells, which otherwise is difficult to rely on just with a passive diffusion in avascular cartilage conditions. The results encourage further development of such scaffold structures targeted on their best clinical performance rather than trying to imitate the respective original tissue.

The authors would like to thank Finnish Agency for Innovation (Tekes) for providing financial support to this project. A.Z. also acknowledges Teknos Foundation (Finland) for the scholarship.

For a meaningful evaluation of biomaterials, in vitroenvironments that mimic the physiological properties of the in vivoenvironment are desirable with relevant control of key factors. For faster screening and reduction of time and costs, combination and control of different critical parameters are needed.

Commercial Hyalograft® and ChondroGide® scaffolds were compared to a new experimental recombinant human collagen-PLA (rhCo-PLA) [1] and pure PLA scaffolds under BEST protocol [2] in pseudostatic (creep), dynamic (frequency scans, strain sweeps), and combined conditions (simulated operative periods) relevant for orthopaedic applications. Temperatures 25–37°C, dry and fully immersed wet (water, 0.9% NaCl) conditions were analysed and aggregate, complex dynamic moduli and loss factor were obtained. Additionally a method was developed for estimation of the swelling pressure under variable compression. ChondroGide and rhCo-PLA were compared in vivoin earlier experiments [1].

All scaffold materials have a non-linear and non-uniform behaviour when immersed in a fluid, accompanied by rapid change in starting porosity (down for Hyalograft® and ChondroGide®, up for PLA), but nearly stable for rhCo-PLA. Too hydrophilic materials exhibited partial non-wetting (dry spots) under a slight compression eventually by closure of the specimen rim due to elastocapillary effect, where as hydrophobic (PLA) shown substantial expansion. The swelling pressure of PLA was measured of ∼1 kPa (water, 25°C). Observed creeping cannot be reliably fit with simple viscoelastic models, but can be approximated with biphasic theory with variable complex moduli and permittivity values. No significant differences were observed in creep for 1 h and 5 h runs, showing that a shorter time is sufficient to catch the main effects in these biomaterials. No substantial differences were observed between water and NaCl solution at 37°C, except for ChondroGide® which swells in NaCl more than in water. Besides of some differences in swelling, no significant differences observed between 25 and 37°C tests for creep. For dynamic conditions all materials undergo densification and “stiffening” (50% and more) upon cyclic strain deformation, with the effect being higher at 37°C than at 25°C. rhCo-PLA scaffolds exhibit relatively stable modulus in water and loss factor with physiologically-compatible behaviour (∼0.1 with a minimum values range around 1 Hz) at frequency scans (0.01–20 Hz). On the contrary, ChondroGide® has the highest loss factor (up to 0.6–0.7).

Water at 25°C seems to be sufficient to rapidly test these kinds of materials for biomechanical screening, unless additions or specific effects are of interest. The applied deformation level is more important to predict materials properties in dynamic conditions than experiment time. This means that better in vitrodata can be obtained in shorter runs. The animal studies have also exhibit rhCo-PLA producing better quality (ICRS median score 12.5 vs. 8.5 for ChondroGide®).