Effects of Tranexamic Acid Cytotoxicity on In Vitro Chondrocytes
Use of topical tranexamic acid (TXA) in orthopedic surgery has been expanding over the past decade, with increasing evidence confirming reductions in perioperative blood loss and transfusion requirements, but there is minimal evidence regarding effects of TXA on native cartilage.
We conducted a study to understand the in vitro effects of TXA on bovine cartilage and murine chondrocytes and ultimately to expand the clinical application of topical TXA to include scenarios with retained native cartilage, such as hemiarthroplasty. Bovine cartilage explants were exposed to TXA at a concentration of 100 mg/mL, and glycosaminoglycan (GAG) release and cell viability were measured at 8, 24, and 48 hours. Monolayer murine chondrocytes were exposed to TXA 25, 50, and 100 mg/mL, and viability was measured at 8, 24, and 48 hours.
GAG released from bovine explants was significantly higher in the samples exposed to TXA 100 mg/mL at all time points. Cell viability was significantly decreased in the explants exposed to TXA 24 and 48 hours after initial incubation. Bovine chondrocyte viability was not affected by TXA 25 mg/mL. Murine chondrocyte viability was similar between the TXA 25 mg/mL and control samples at all time points. The TXA 50 mg/mL sample dropped from 66.51% viability at 8 hours to 6.81% viability at 24 hours and complete cell death by 48 hours. The TXA 100 mg/mL samples had no observable viable cells at 8, 24, and 48 hours.
Our data indicated that TXA 100 mg/mL damaged and was cytotoxic to bovine explanted cartilage and was cytotoxic to murine chondrocytes. Murine and bovine chondrocyte viability were not affected by TXA 25 mg/mL.
Cell viability was notably higher in the control groups 24 and 48 hours after initial incubation (Figure 2), with a visually observable (Figure 3) but variable and nonsignificant (P = .33) difference in viability at 8 hours, control (mean, 63.87%; SD, 13.63%) versus TXA (mean, 46.08%; SD, 22.51%). As incubation time increased from 8 hours, there were significant decreases in cell viability at 24 hours (mean, 39.28%; SD, 4.12%; P = .024) and 48 hours (mean, 21.98%; SD, 2.15%; P = .0005) relative to controls.
After results of exposing murine cells to TXA at different concentrations were obtained, bovine explants were exposed to TXA 25 mg/mL, and viability was recorded 24 and 48 hours after exposure (Figure 4). There was no significant difference in viability between samples.
Cell viability was similar between the TXA 25 mg/mL and control samples at all time points (Figure 5). The TXA 50 mg/mL sample dropped from 66.51% viability at 8 hours to 6.81% viability at 24 hours and complete cell death by 48 hours. The TXA 100 mg/mL samples had no observable viable cells at 8, 24, and 48 hours (Figure 6, confirmed with light microscopy). The TXA 0 mg/mL and 25 mg/mL samples remained largely unchanged: 78.28% and 92.99% viable at 8 hours, 97.29% and 90.22% viable at 24 hours, and 91.62% and 91.35% viable at 48 hours, respectively. See Figures 4 and 5 for viability at all time points and concentrations. Statistical analyses were not performed on these data because the zero values obtained for all samples incubated in TXA 100 mg/mL and the 48-hour TXA 50 mg/mL samples prevented accurate estimation of P values and thus meaningful comparisons of the treatment groups.
Discussion
The results of this study showed that TXA is cytotoxic to both bovine and murine chondrocytes at a concentration of 100 mg/mL. There is a time-dependent increase in GAG release as well as a decrease in chondrocyte viability in intact bovine cartilage. These data suggest that topical or intra-articular administration of TXA at this concentration in the setting of native cartilage may have unintended, detrimental effects.
Murine chondrocyte monolayer cultures exposed to TXA at lower concentrations did not exhibit a concentration-dependent curve with respect to viability. Chondrocytes exposed to TXA 25 mg/mL had no reduction in viability relative to control samples. When the concentration was doubled to 50 mg/mL, however, viability was reduced to 6.81% by 24 hours (Figure 5). These data suggest that, between 25 mg/mL and 50 mg/mL, there is a concentration at which TXA becomes cytotoxic to murine chondrocytes. It should be cautioned that, though TXA was cytotoxic to chondrocytes in this study, the effects are still unknown and indeed may be similar to effects on other types of cells that are present in a replaced joint—such as synovial cells, inflammatory cells, and osteoblasts.
The unaffected viability of murine chondrocytes with TXA 25 mg/mL indicated that this may be a cutoff concentration for safety in the presence of cartilage. To confirm these results, we exposed the bovine explants to TXA 25 mg/mL as well. Consistent with the prior study, chondrocyte viability was unaffected at 48 hours. Some clinical studies have effectively used topical TXA at this concentration, or at a lower concentration, to reduce blood loss in TJA,25 which suggests that 25 mg/mL may be a safe yet effective dose for clinical use of topical TXA.
As the methods used in this study did not distinguish between late-apoptotic and necrotic cell death, we could not determine which mechanism of death led to the viability loss observed. If apoptosis is occurring, how TXA initiates this sequence is unclear. There have been no studies directly linking TXA to apoptotic events, though some studies have indicated that TXA interacts with several molecules other than plasminogen, including GABA (γ-aminobutyric acid) receptors, glycine receptors, and tachykinin neurokinin 1 receptors.26-28 According to these studies, these interactions may be responsible for seizure activity and increased emesis caused by TXA use. In addition, TXA-containing compounds, such as trans-aminomethylcyclohexanecarbonyl-l-(O-picolyl)tyrosine-octylamide, have been shown to induce apoptosis.29
It appears that the extracellular matrix (ECM) of native cartilage explants has a protective effect on chondrocytes. With exposure to TXA 100 mg/mL, the explants retained 52% viability at 24 hours, whereas the monolayer cultures were nonviable at that point. The weak negative charge of the molecule may retard its penetration into the ECM, though there was an inconsistent presentation of cell death at explant superficial zones in treated samples (Figure 3). Consistent surface layer cell death would be expected if slowed penetration were the only protective mechanism. It is possible that the ECM acts as a buffer or solvent, effectively reducing the concentration of TXA directly interacting with the chondrocytes. Further exploration is needed to elucidate the significance of the ECM in protecting chondrocytes from TXA.
