Aims. The intra-articular administration of tranexamic acid (TXA) has been shown to be effective in reducing blood loss in unicompartmental knee arthroplasty and anterior cruciate reconstruction. The effects on human articular cartilage, however, remains unknown. Our aim, in this study, was to investigate any detrimental effect of TXA on chondrocytes, and to establish if there was a safe dose for its use in clinical practice. The hypothesis was that TXA would cause a dose-dependent damage to human articular cartilage. . Materials and Methods. The cellular morphology, adhesion, metabolic activity, and viability of human chondrocytes when increasing the concentration (0 mg/ml to 40 mg/ml) and length of exposure to TXA (0 to 12 hours) were analyzed in a 2D model. This was then repeated, excluding cellular adhesion, in a 3D model and confirmed in viable samples of articular cartilage. Results. Increasing concentrations above 20 mg/ml resulted in atypical morphology, reduced cellular adhesion and metabolic activity associated with increased chondrocyte death. However, the cell matrix was not affected by the concentration of TXA or the length of exposure, and offered cellular protection for concentrations below 20 mg/ml. Conclusion. These results show that when in vitro chondrocytes are exposed to higher concentrations of TXA, such as that expected following recommended intra-articular administration, cytotoxicity is observed. This effect is dose-dependent, such that a tissue concentration of 10 mg/ml to 20 mg/ml could be expected to be safe. Cite this article: Bone Joint J 2018;100-B:404–12

Ovine articular chondrocytes were isolated from cartilage biopsy and culture expanded in vitro. Approximately 30 million cells per ml of cultured chondrocytes were incorporated with autologous plasma-derived fibrin to form a three-dimensional construct. Full-thickness punch hole defects were created in the lateral and medial femoral condyles. The defects were implanted with either an autologous ‘chondrocyte-fibrin’ construct (ACFC), autologous chondrocytes (ACI) or fibrin blanks (AF) as controls. Animals were killed after 12 weeks. The gross appearance of the treated defects was inspected and photographed. The repaired tissues were studied histologically and by scanning electron microscopy analysis.

All defects were assessed using the International Cartilage Repair Society (ICRS) classification. Those treated with ACFC, ACI and AF exhibited median scores which correspond to a nearly-normal appearance. On the basis of the modified O’Driscoll histological scoring scale, ACFC implantation significantly enhanced cartilage repair compared to ACI and AF. Using scanning electron microscopy, ACFC and ACI showed characteristic organisation of chondrocytes and matrices, which were relatively similar to the surrounding adjacent cartilage.

Implantation of ACFC resulted in superior hyaline-like cartilage regeneration when compared with ACI. If this result is applicable to humans, a better outcome would be obtained than by using conventional ACI.