Tissue engineering has been proposed to alleviate injury to the knee meniscus, which leads to loss of function and damage to the surrounding articular cartilage. Relatively few studies have been performed to tissue engineer the meniscus;​ however, much guidance can be found by studying related tissues such as articular cartilage. One technique that has proved beneficial for producing tissue engineered articular cartilage constructs is mechanical stimulation. This thesis describes the investigation of scaffold choices and the development of culture techniques for producing meniscal constructs. In addition, a direct compression stimulator was developed, validated, and applied to the tissue engineered meniscal constructs.

Poly-glycolic acid (PGA) and agarose were examined for use as scaffolds. It was found that agarose did not support fibrochondrocyte growth, and that while PGA supported cell production and proliferation, constructs were not mechanically robust after 7 wks in culture. A direct compression bioreactor was developed and validated using articular cartilage and meniscal explants. An attempt to produce better constructs than previously achieved in vitro in terms of mechanical properties, matrix production and organization, was successful using the self-assembly (SA) method. The SA method developed for the meniscus employs a ring-shaped agarose mold seeded with articular chondrocytes (AC) and meniscal fibrochondrocytes (MFC). It was found that a spectrum of biomechanical and biochemical properties could be achieved based on the seeding ratio of ACs:​MFCs. It was determined that a 50:​50 ratio of AC:​MFC produced a construct that best replicated the cross-section of the native meniscus. A direct comparison of SA constructs to constructs formed with PGA was made via an investigation into the effect of dynamic direct compression stimulation. It was found that, after 4 wks in culture, PGA constructs lacked sufficient mechanical integrity to undergo loading. Wk 8 SA constructs were 3.4 times stronger and stiffer in the circumferential direction than in the radial direction. In addition, SA constructs had 3 times more GAG and 2 times more collagen than PGA constructs. The application of dynamic stimulation did not further increase mechanical properties or matrix production in SA constructs, but merits further study examining different loading regimens.