Possible Toxic Effect of Tranexamic Acid and Aminocaproic Acid on Articular Cartilage: An In Vitro and In Vivo Analysis

Abstract

Objective

To compare the dose-dependent chondrotoxicity of tranexamic acid (TXA) and aminocaproic acid (ACA) in vitro and in vivo.

Design

In vitro, human chondrocytes were exposed to TXA or ACA (0-50 mg/ml), and cytotoxicity was assessed. In vivo, a rat model of monoiodoacetate-induced osteoarthritis was used to evaluate cartilage damage following intra-articular injections of TXA or ACA.

Results

In vitro, both TXA and ACA reduced chondrocyte viability in a dose-dependent manner, with significant cytotoxicity observed at concentrations ≥20 mg/ml. TXA was more toxic than ACA at these higher concentrations. Apoptosis increased markedly at 30 mg/ml for both agents. In the rat osteoarthritis model, joints treated with TXA or ACA showed greater cartilage erosion and matrix loss compared to controls, with TXA causing more severe damage.

Conclusion

Both TXA and ACA are chondrotoxic in a dose-dependent manner, with TXA demonstrating greater potency. Lower concentrations (≤10 mg/ml) are recommended for topical use in cartilage-preserving surgery to minimize potential damage.

Introduction

Limiting perioperative blood loss is a critical objective in invasive orthopedic surgeries, such as hip and knee surgery. ¹ The use of antifibrinolytic medications, particularly tranexamic acid (TXA) and aminocaproic acid (ACA), is increasingly common in these procedures. ² These synthetic lysine analogues competitively inhibit the binding sites of plasmin and plasminogen activator, preventing the conversion of plasminogen to plasmin—a fibrinolytic protease responsible for dissolving fibrin clots. ³ Consequently, TXA and ACA effectively reduce blood loss by promoting clot stability.

While both TXA and ACA can be administered orally, intravenously (IV), intra-articularly, or peri-articularly, the optimal route and dosage for most orthopedic surgeries remain subjects of ongoing debate. ⁴ IV administration, though widely regarded as safe, carries potential risks, including thrombosis and postoperative seizures. ⁵ In contrast, topical administration has gained popularity, especially in total knee arthroplasty and total hip arthroplasty, due to its ability to deliver high drug concentrations directly to the bleeding site with minimal systemic effects.^6,7 However, the application of these agents is not limited to joint replacement. Their use in joint-preserving procedures—such as arthroscopic repairs, ligament reconstructions, and osteotomies—raises critical questions about their potential toxic effects on the remaining viable articular cartilage.^{8 -10}

Despite the increasing use of topical TXA and ACA, there is a significant gap in understanding their effects on human chondrocytes (HCHs)—cells crucial to the integrity of articular cartilage.

Several studies have found that TXA can damage HCHs in vitro. TXA was shown to be toxic to HCHs, with damage occurring as early as 2.5 min after exposure. ¹¹ The toxic threshold appears to be above a concentration of 25 mg/ml. At concentrations of 50 mg/ml and 100 mg/ml, TXA significantly increased apoptosis in cartilage and subchondral bone. ¹² A study on bovine cartilage, however, explants and murine chondrocytes found no toxic potential at concentrations up to 25 mg/ml. ¹¹ This conflicting evidence highlights a critical need to clarify the precise dose-dependent effects of TXA on HCHs.

In stark contrast to the growing body of evidence regarding TXA’s chondrotoxic effects, studies examining ACA’s impact on HCHs are conspicuously absent from the literature. While both agents are commonly used antifibrinolytics in orthopedic surgery, the cellular response to ACA exposure remains largely uncharacterized. This knowledge gap is particularly concerning given that clinicians may use these agents interchangeably without evidence-based guidance regarding their comparative safety profiles for articular cartilage.

This study aims to evaluate and compare the impact of TXA and ACA on HCHs at various concentrations that are relevant to clinical practice. We hypothesize that TXA and ACA exert dose-dependent toxic effects on chondrocytes, which has critical implications for their safe use in cartilage-preserving surgeries.

Materials and Methods

This study comprised both in vitro experiments using HCHs and in vivo experiments using a rat model of osteoarthritis. The effects of TXA and ACA were evaluated in both settings.

HCHs Culture

Normal HCHs were purchased from PromoCell (Heidelberg, Germany). HCHs were cultured in T75 flasks using HCH growth medium supplemented with penicillin-streptomycin. Cells were maintained at 37 °C in 95% air and 5% CO₂, sub-cultured every 3 days, with medium replacement every 2 days. Passage 4 (P-4) HCHs were used for all experiments.

Cell Treatment and Viability Assay

Cell viability was assessed using a CCK-8 solution (Dojindo, Kumamoto, Japan) following the manufacturer’s instructions. In summary, HCHs were counted and plated at a density of 2.5 × 10³ cells/well in 96-well plates. After one day, the cells were treated with TXA or ACA at concentrations of 0, 10, 20, 30, and 50 mg/ml. The 0 mg/ml group served as the negative control, and the concentration range was selected to reflect doses used in clinical topical applications. ⁶ Following 48 h of treatment, 10 ml of CCK-8 was added, and the plates were incubated at 37 °C for 2 h. The absorbance of each well was then measured at 450 nm. Each experiment was conducted independently with 6 replicates.

Live/Dead Fluorescence Assay

To optimize the concentrations of TXA and ACA in our study, we employed the LIVE/DEAD cell assay following the manufacturer’s guidelines (Live/Dead Viability/Cytotoxicity Kit, Invitrogen, Carlsbad, CA). HCH cells were seeded at a density of 5 × 10⁴ cells/ml in Falcon 24-well plates (Corning NY, USA; Cat. Number 353047) and allowed to adhere for 24 h before exposure to TXA or ACA at 0, 10, and 30 mg/ml. After a 48-h incubation, cell viability was assessed using 4 μM EthD-1 and 2 μM calcein-AM. Fluorescence imaging was performed via confocal microscopy, utilizing Alexa Fluor (488 nm) and Texas Red (543 nm-594 nm) lasers with a 10x objective, enabling precise visualization of live and dead cell populations.

Apoptosis Assay

Flow cytometry was used to determine various types of cell death including early and late stage apoptosis as well as necrosis by quantifying cell surface annexin V-FITC (Sigma, Saint-Louis, MO, USA) staining as we previously described ¹³ Briefly, cells were seeded in 96-well plates at a density of 5 × 10⁴ cells/ml per well and incubated for 24 h. HCHs were treated with the different concentrations of TXA and ACA (0, 10, and 30 mg/ml). After 48 h, HCHs were harvested by centrifugation (HeraeusMultifuge X3R, Thermo Fisher Scientific, Villebon-sur-Yvette, France) at 1800 rpm for 5 min then the supernatant was removed. The cells were washed with 200 µl of 1x binding buffer. After removal of the supernatant, apoptosis or necrosis in the cultured cells was evaluated after performing Annexin V-FITC/propidium iodide (PI) double staining, utilizing the Annexin-V FITC Apoptosis Detection Kit I (BD Biosciences, San Jose, CA) in accordance with the manufacturer’s guidelines.

Gene Expression Analysis by Quantitative Reverse Transcription PCR

To examine the molecular effects of TXA and ACA on HCHs differentiation, the expression levels of key cartilage-related genes were analyzed 72 h after treatment. Total RNA was extracted using Trizol reagent (Ambion, Waltham, MA). cDNA was synthesized using the Primescript II 1st strand cDNA synthesis kit (BIOLINE, London, UK). Real-time PCR was performed using Power SYBR Green PCR master mix (Applied Biosystems Inc., USA) with gene-specific primers for COMP, Aggrecan, Osteonectin, Col2A1, and Runx2. The 2-∆∆Cq method was used for the relative quantification of gene expression. The mRNA levels of the target genes were normalized to those of GAPDH. The sequences of the specific primers were as follows: cartilage oligometric matrix protein (COMP), forward 5′- AAGAACGACGACCAAAAGGAC-3′, reverse 5′-CATCCCCTATACCATCGCCA-3′; Osteonectin, forward 5′-CCCATTGGCGAGTTTGAGAAG-3′, reverse 5′- CAAGGCCCGATGTAGTCCA-3′; Col2A1, forward 5′-TGGACGCCATGAAGGTTTTCT-3′, reverse 5′-TGGGAGCCAGATTGTCATCTC-3′; Runx2, forward 5′- TCAACGATCTGAGATTTGTGGG-3′, reverse 5′-GGGGAGGATTTGTGAAGACGG-3′, GAPDH, forward 5′-GGAGCGAGATCCCTCCAAAAT-3′, reverse 5′- GGCTGTTGTCATACTTCTCATGG-3′.

HCHs Pellet Culture and Staining

The formation of cartilage-like tissue in vitro was investigated in high-density pellet cultures. ¹⁴ Briefly, 5 × 10⁵ cells in 0.5 ml of DMEM containing penicillin-streptomycin, ITS⁺³, 0.1 mM ascorbic acid-2- phosphate, 0.4 mM L-proline, 100 nM dexamethasone (Sigma-Aldrich, Buchs, Switzerland) with or without TXA and ACA were centrifuged at 250 g for 5 min in 15-ml polypropylene tubes. The cell pellets were cultured for 3 weeks and the medium was changed twice a week. Then, pellet cultures were fixed in 4% paraformaldehyde for 4 h at room temperature, dehydrated, and embedded in paraffin. The pellet cultures were cut into 5-μm sections and stained with standard hematoxylin and eosin (H&E) and Alcian blue.

Western Blot

After 72 h of 30 mg/ml TXA or ACA treatment on HCHs, the proteins were extracted using PRO-PREP™ protein extraction solution (iNtRON Biotechnology Inc., Seongnam, Republic of Korea), denatured protein samples (40 μg per lane) were separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) using 10% and 12% gels and then transferred to polyvinylidene difluoride membranes.

The membranes were blocked with 5% skim milk in Tris-buffered saline with 0.1% Tween-20 (TBST) at room temperature for 2 h and incubated with specific primary antibodies (diluted 1:1000 in 5% skim milk) at 4 °C overnight.

After washing three times in TBST, the membranes were incubated with secondary antibodies (diluted at 1:2000) for 2 h at room temperature.

The specific primary antibodies were as follows: anti-proliferating cell nuclear antigen (anti-PCNA; Santa Cruz Biotechnology, sc-55550, 1:1000), anti-XIAP (Santa Cruz, sc-55550, 1:1000), anti-Bax (Santa Cruz, sc-23959, 1:1000), anti-Bcl-2 (Santa Cruz, sc-7382, 1:1000), and anti-β-actin (Santa Cruz, sc-47778, 1:1000), anti-cleaved PARP (Cell Signaling Technology, #9541, 1:1000), and anti-cleaved caspase-3 (Cell Signaling, #9664, 1:1000). The membranes were then washed three times with TBST (10 min each) and incubated with HRP-conjugated anti-mouse IgG or anti-rabbit IgG secondary antibodies (ACE BioLabs, #A1012S and #A1013S, 1:2000) in TBST at room temperature for 2 h, followed by detection using a chemiluminescent substrate (Bio-Rad Laboratories, Hercules, CA).

All blots were visualized using enhanced chemiluminescence (ECL) substrate (Bio-Rad Laboratories).

Animals

Twenty-four male Sprague Dawley rats (220-250 g, 7 weeks old) were sourced from. Animal care and experimental procedures were approved by the Institutional Animal Care and Use Committee of CHA University (IACUC No. 200010), and this study was conducted according to AVMA guidelines. The rats were housed in pairs in standard polycarbonate cages with corncob bedding, under a controlled 12-h light/dark cycle at a constant temperature (22 ± 2 °C) and humidity (50 ± 10%). They had ad libitum access to standard laboratory chow and filtered water. Environmental enrichment was provided in the form of nesting material.

Sample size determination

The sample size of n = 6 animals per group was determined using the “resource equation” method, suitable for exploratory studies where prior data on effect size is limited. ¹⁵ The equation is E = (Total number of animals) − (Total number of groups). For this study, E = (24 − 4) = 20. An E-value between 10 and 20 is considered optimal, indicating the sample size is adequate to detect a biological effect without being excessive. ¹⁵

Experimental unit and blinding

The experimental unit was the individual animal. To minimize assessment bias, the pathologist performing the macroscopic and histological scoring was blinded to the treatment group of each specimen.

Osteoarthritis Induction and Treatment

Osteoarthritis was induced by intra-articular injection of monoiodoacetate (MIA, 0.3 mg) into the left knee. Animals were randomly allocated to the four treatment groups using a random number generator. To ensure isotonic conditions and prevent osmolarity-related cytotoxicity, TXA and ACA were dissolved in phosphate-buffered saline (PBS), which mimics physiological fluids and prevents osmotic activity gradients across the chondrocyte membrane.^{16 -18} The control groups received the same vehicle (PBS). A total standardized volume of 1 ml of TXA or ACA solution (30 mg/ml), or PBS for controls, was administered intra-articularly 1 week after MIA injection. Rats were divided into four groups (n = 6 per group): PBS control, PBS + MIA, MIA + TXA, and MIA + ACA. The right knee served as an untreated control. No animals were excluded from the study.

Incapacitance Test

The primary outcome measure in vivo was the histological assessment of cartilage integrity (Mankin score), with weight-bearing serving as a secondary functional outcome. Weight distribution between hind limbs was measured at weeks 2 and 5 using an incapacitance meter (WPI, Sarasota, FL). Weight-bearing was calculated as % weight on left limb = 5 L/(L + R) × 100%, where L and R represent the weight (grams) on the left and right hind limbs, respectively.

Cartilage Scoring

Following euthanasia at 5 weeks, the both knees were harvested by cutting the femur and tibia/fibula 2 cm above and below the joint line, respectively. After removal of soft tissue, the specimens were photographed using a stereo microscope (Nikon SMZ745). The cartilage degeneration and bone destruction of the femoral and tibial condyles were evaluated using a macroscopic scoring system on a scale of 0-5 points (Table 1). ¹⁹ Next, joints were fixed in a 10% neutral buffered formalin solution for 14 days at room temperature. Knees were then decalcified, dehydrated in a graded series of alcohols, embedded in paraffin wax, and stored at room temperature until sectioning.

Table 1.

Macroscopic Cartilage and Bone Scoring (0-5) for Rat Arthritis Induced by Monoiodoacetic Acid (MIA).

Points	Findings
0	Intact articular surface
1	≤10 punctate depressions per condyle
2	>10 punctate depressions per condyle
3	Erosion (≤50% of joint surface)
4	Erosion (>50% of joint surface)
5	Bone destruction

Histologic Analysis

The paraffin-embedded sections of knee joint were prepared, dewaxed, and rehydrated. Serial sagittal plane sections of 5-μm sections were cut and stained with standard H&E and Alcian blue. Histological sections were quantified using a modified Mankin histology scoring system (Table 2). ²⁰

Table 2.

Modified Mankin’s Histological Scores.

Subscore	Details
Cartilage structure
Normal	0
Surface irregularities	1
Pannus and surface irregularities	2
Clefts to transitional zone	3
Clefts to radial zone	4
Clefts to calcified zone	5
Complete disorganization	6
Cartilage cells
Normal	0
Pyknosis, lipid degeneration, hypercellularity	1
Clusters	2
Hypocellularity	3
Safranin O, thionine, Alcian blue
Normal	0
Slight reduction	1
Moderate reduction	2
Severe reduction	3
No staining	4
Tidemark integrity
Intact	0
Destroyed	1

Statistical Analysis

Data are presented as mean ± standard deviation (SD). Data were analyzed using GraphPad Prism (version 8). All data were assessed for normality using the Shapiro-Wilk test. As data were not normally distributed, the Mann-Whitney U test was used for comparisons between two groups, and the Kruskal-Wallis test with Bonferroni post hoc analysis was used for multiple group comparisons. In vitro experiments were performed at least in triplicate. A P-value of less than 0.05 was considered statistically significant. Exact P-values are reported where possible.

Results

HCH viability was significantly reduced after treatment with 20 mg/ml of TXA or ACA, with TXA exhibiting greater toxicity than ACA. However, cell viability remained unaffected at 10 mg/ml concentrations of both TXA and ACA (Fig. 1). As shown in Figure 1 , cell viability was not significantly affected at 10 mg/ml for either agent (TXA: 87.67 ± 5.30%, P = 0.056; ACA: 88.53 ± 6.19%, P = 0.111). However, at concentrations of 20 mg/ml and higher, both TXA and ACA significantly reduced chondrocyte viability in a dose-dependent manner. Viability was reduced to 74.07 ± 1.96% (P = 0.008) with 20 mg/ml TXA, 64.23 ± 2.56% (P = 0.008) with 30 mg/ml TXA, and 52.56 ± 0.81% (P = 0.008) with 50 mg/ml TXA. For ACA, viability was reduced to 77.59 ± 4.80% (P = 0.016) at 20 mg/ml, 62.12 ± 1.07% (P = 0.008) at 30 mg/ml, and 53.64 ± 2.24% (P = 0.008) at 50 mg/ml.

Figure 1.

1. Human chondrocyte viability assessed by Cell Counting Kit-8 assay. (A) shows human chondrocyte viability when treated with TXA; (B) shows human chondrocyte viability when treated with ACA. Cell viability was significantly reduced at concentrations of 20 mg/ml and higher for both agents. Data are presented as mean ± SD (n = 6 replicates). Statistical significance was determined using the Kruskal-Wallis test with Bonferroni post hoc analysis. TXA = tranexamic acid; ACA = aminocaproic acid. *P < 0.05, **P < 0.01 versus control group.

The Live/Dead assay confirmed these findings (Fig. 2). As shown in Figure 2A , TXA treatment led to a significant decrease in cell viability in a dose-dependent manner. At 10 mg/ml, TXA-treated cells exhibited a reduction in viability compared with the control (P = 0.001). At the higher concentration of 30 mg/ml, TXA-induced cell viability to drop below 40% (P < 0.001). ACA showed a milder impact, with a significant decrease in viability only observed at 30 mg/ml (P < 0.001). Fluorescence microscopy images (Fig. 2B) visually confirmed these results, demonstrating that control cells remained largely viable, whereas cells treated with 30 mg/ml TXA showed a marked increase in dead cells. The 10 mg/ml TXA treatment also resulted in decreased cell viability, though to a lesser extent than the higher dose. Consistent with the viability assay, ACA treatment exhibited a milder impact, with only a slight reduction in live cells at both concentrations.

Figure 2.

Live/dead viability/cytotoxicity assay results. (A) Cell viability decreased significantly with 10 mg/ml TXA (P = 0.001) and 30 mg/ml TXA (P < 0.001), as well as 30 mg/ml ACA (P < 0.001). (B) Fluorescence microscopy images visually confirm the quantitative data, showing a marked increase in dead (red) cells with 30 mg/ml TXA treatment compared to controls and ACA. Data are presented as mean ± SD (n = 6 replicates). Statistical significance was determined using the Kruskal-Wallis test with Bonferroni post hoc analysis.

To assess the apoptotic effects of TXA and ACA on chondrocytes, Annexin-V and PI staining were performed, followed by flow cytometry analysis. As shown in Figure 3 , TXA treatment led to a marked increase in apoptosis, particularly at the higher concentration of 30 mg/ml. In comparison, ACA treatment induced a lower level of apoptosis than TXA at the same concentrations. Although 30 mg/ml ACA led to an increase in apoptotic cells relative to the control, the proportion of apoptotic cells was considerably lower than that observed with 30 mg/ml TXA.

Figure 3.

Flow cytometry analysis following annexin-V and propidium iodide staining after 48 h. Chondrocyte apoptosis increased without a corresponding increase in necrosis rate after treatment with 30 mg/ml of TXA or ACA.

Gene Expression Analysis

COMP (cartilage oligomeric matrix protein) and COL2a1 (collagen type II alpha 1), both crucial for cartilage extracellular matrix formation, showed no significant differences between the control and treated groups. Aggrecan and osteonectin, essential for cartilage structure and function, were significantly downregulated in TXA- and ACA-treated cells compared with the control group (P < 0.001), indicating impaired cartilage matrix synthesis. In contrast, RUNX2, a key transcription factor involved in chondrocyte hypertrophy and cartilage degeneration, was significantly upregulated in both TXA- and ACA-treated groups (P = 0.009 and P = 0.04, respectively). This suggests that TXA and ACA may induce a shift toward hypertrophic or osteogenic differentiation, which is often associated with cartilage degradation and osteoarthritis progression ( Fig. 4 ).

Figure 4.

Quantitative real-time PCR analysis of gene expression. COMP, aggrecan, and osteonectin gene expressions were significantly downregulated after 30 mg/ml TXA and ACA treatment, while RUNX2 expression was significantly upregulated. Data are presented as mean ± SD (n = 3 independent experiments). Statistical significance was determined using the Kruskal-Wallis test with Bonferroni post hoc analysis. **P < 0.01, ***P < 0.001.

To assess the impact of TXA and ACA on the chondrogenic potential of HCHs, histological staining was performed on high-density pellet cultures maintained for 3 weeks ( Fig. 5 ). The control group exhibited minimal matrix deposition, as expected. In contrast, the chondrogenic differentiation media (CM)-only group displayed robust chondrogenic differentiation, characterized by well-formed pellet morphology and strong Alcian blue staining, indicative of glycosaminoglycan (GAG) production. The CM + TXA group showed markedly reduced chondrogenic differentiation, as evidenced by disrupted pellet morphology and decreased Alcian blue staining. This suggests that TXA, particularly at the tested concentration, significantly impairs the ability of HCHs to undergo chondrogenic differentiation. The CM + ACA group also demonstrated a reduction in Alcian blue staining compared with the CM-only group, though the effect was less severe than that observed with TXA. This indicates that while ACA may slightly impact chondrogenic differentiation, its effect is not as detrimental as TXA.

Figure 5.

Histological sections of HCH pellets cultured for 3 weeks, stained with H&E and Alcian blue (scale bar: 200 μm). Groups treated with chondrogenic medium (CM) containing TXA or ACA showed reduced chondrogenic differentiation.

Protein Expression Analysis

To further investigate the molecular mechanisms underlying TXA-induced apoptosis, the expression of key apoptotic markers was analyzed using Western blot ( Fig. 6 ). The results indicate that TXA treatment at 30 mg/ml significantly upregulated markers of apoptosis, including cleaved-PARP and cleaved-caspase-3, compared with the control and ACA-treated groups. The quantitative analysis confirms a notable increase in cleaved-caspase-3 expression, further supporting enhanced apoptotic activity. In contrast, the expression of the X-linked inhibitor of apoptosis protein (XIAP), a known suppressor of apoptosis, was markedly reduced following TXA treatment. A similar trend was observed for Bcl-2, an anti-apoptotic protein, which exhibited decreased expression in the TXA group. These results suggest that TXA enhances apoptosis by suppressing key survival pathways. The Bax/Bcl-2 ratio, a critical determinant of apoptotic signaling, was increased in TXA-treated cells, indicating a shift toward a pro-apoptotic state. While ACA treatment also led to an increase in Bax expression, the effect was more pronounced in the TXA group, reinforcing the notion that TXA promotes apoptotic signaling more strongly than ACA.

Figure 6.

Western blot analysis of cleaved-PARP, XIAP, Bcl-2, Bax, and cleaved-caspase-3 protein expression in chondrocytes. Protein quantifications were normalized to β-actin.

Animals

Incapacitance test

Figure 7 shows the effect of TXA or ACA on weight-bearing. At baseline, rats distributed roughly 50% of body weight on each hind limb to maintain a balanced posture. Injection of PBS did not alter this distribution. However, at 2 and 5 weeks after MIA injection, there was a slight reduction in weight-bearing on the ipsilateral (damaged) hind limb. Intra-articular injection of TXA or ACA similarly caused a weight-bearing imbalance, with more weight borne on the contralateral limb (interpreted as reduced load on the injured side), although this effect was not statistically significant.

Figure 7.

Weight-bearing percentage in rats with monosodium MIA-induced osteoarthritis at 2 and 5 weeks. Data are presented as mean ± SD (n = 6 animals per group). Statistical significance was determined using the Kruskal-Wallis test.

Macroscopic observation of tibial cartilage

To assess the effects of TXA and ACA on cartilage integrity in the MIA-induced osteoarthritis model, macroscopic evaluation of the femoral condyle and tibial plateau was performed (Fig. 8). In the control group, both femoral and tibial cartilage surfaces appeared smooth and intact, with no visible signs of degeneration (score = 0). In contrast, the MIA + PBS group exhibited surface irregularities, fibrillation, and focal erosions, resulting in elevated macroscopic damage scores (femur = 2.0 ± 0.8; tibia = 1.73 ± 0.5).

Figure 8.

Macroscopic features and cartilage scores of femoral and tibial condyles. (Note: Left knee images in control group were horizontally flipped for consistency.) The 0.3 mg MIA + 30 mg TXA or ACA groups showed more punctate depressions, cartilage erosion, and higher scores compared with control or PBS + 0.3 mg MIA. Femoral and tibial cartilage showed similar macroscopic changes. Data are presented as mean ± SD (n = 6 animals per group). Statistical significance was determined using the Kruskal-Wallis test with Bonferroni post hoc analysis. *P < 0.05, **P < 0.01.

Compared with MIA + PBS, both MIA + TXA and MIA + ACA groups showed further cartilage deterioration. The MIA + TXA group demonstrated pronounced surface erosion and fibrillation, with significantly increased macroscopic scores in both femoral (3.29 ± 0.6; P = 0.0113) and tibial regions (3.61 ± 0.6; P = 0.00017). Similarly, the MIA + ACA group exhibited extensive cartilage loss, particularly in the tibial plateau (femur = 3.50 ± 0.6; P = 0.0049; tibia = 2.70 ± 0.5; P = 0.0049).

Quantitative comparison ( Fig. 8 , right panel) confirmed that both TXA and ACA significantly aggravated macroscopic cartilage degeneration compared with MIA + PBS, with the tibial plateau showing the greatest degree of damage. These results indicate that intra-articular administration of TXA or ACA exacerbates cartilage destruction in the MIA-induced osteoarthritis model.

Histological observations of tibial cartilage

To further evaluate cartilage degeneration at the histological level, Safranin O staining was performed on femoral and tibial cartilage sections, and the Mankin score was used to quantify structural integrity ( Fig. 9 ). In the control group, cartilage displayed a well-preserved architecture with intense Safranin O staining, indicating abundant proteoglycan content and minimal degeneration (Mankin score 0-1). In contrast, the MIA + PBS group exhibited moderate cartilage degradation with reduced Safranin O staining and surface fibrillation, corresponding to higher Mankin scores (femur = 5.8 ± 2.0; tibia = 7.09 ± 1.5).

Figure 9.

Histological analysis of femoral and tibial condyles. Sagittal sections of medial tibial and femoral condyles stained with Safranin O. Mankin scores are indicated in each image. The 0.3 mg MIA + 30 mg TXA or ACA groups demonstrated moderate to severe loss of staining and higher scores compared with other groups. M: Mankin score. Data are presented as mean ± SD (n = 6 animals per group). Statistical significance was determined using the Kruskal-Wallis test with Bonferroni post hoc analysis. *P < 0.05, **P < 0.01.

The MIA + TXA and MIA + ACA groups demonstrated markedly aggravated histologic damage compared with MIA + PBS. The MIA + TXA group showed extensive surface erosion, deep fissuring, and near-complete loss of Safranin O staining, with significantly elevated Mankin scores in both femoral (11.44 ± 1.6; P = 0.0113) and tibial cartilage (11.73 ± 0.1; P = 0.00017). Similarly, the MIA + ACA group exhibited pronounced cartilage loss and proteoglycan depletion, with high Mankin scores in the femur (10.8 ± 1.0; P = 0.0049) and tibia (12.73 ± 1.8; P = 0.0049).

Quantitative analysis of Mankin scores ( Fig. 9B ) confirmed that both TXA and ACA significantly increased histologic cartilage degeneration compared with MIA + PBS, consistent with the macroscopic findings. These results indicate that intra-articular administration of TXA or ACA exacerbates structural and biochemical cartilage destruction in the MIA-induced osteoarthritis model.

Discussion

The findings of this study demonstrate concentration-dependent chondrotoxic effects of TXA and ACA, with TXA exhibiting significantly greater cytotoxicity than ACA at clinically relevant concentrations. These results align with the hypothesis that both agents may compromise chondrocyte viability at higher doses, though their mechanisms and critical thresholds differ substantially. In addition, using a rat model of MIA-induced osteoarthritis, we investigated the effect of TXA or ACA on cartilage in vivo. Based on the results, it was observed that topically administrated TXA or ACA had significant damage on cartilage tissue both macroscopically and histologically. Therefore, our data suggest that both TXA and ACA exert cytotoxic effects on cartilage.

Our selected in vitro concentrations of 10-50 mg/ml reflect those achieved in clinical settings. The local concentration of topically applied TXA can vary significantly based on the administered dose and carrier volume. For instance, a common intra-articular dose of 1.5 g of TXA in 50 ml of saline results in a concentration of 30 mg/ml, while higher concentrations, up to 100 mg/ml (e.g., 2 g in 20 ml), can be reached with different preparations.^21,22 Therefore, our findings of dose-dependent chondrotoxicity are directly relevant to established clinical protocols. Given that cartilage is avascular, direct topical application can lead to sustained high local concentrations, making in vitro toxicity data highly relevant for predicting clinical outcomes in joint-preserving surgeries.

The use of TXA or ACA to reduce operative and postoperative bleeding and the subsequent need for blood transfusion is well defined for total knee arthroplasty (TKA). ²³ A previous meta-analysis by Riaz et al. ²⁴ compared the effect of TXA and ACA on TKA, and showed that the efficacy was similar between the two. Moreover, they suggest that ACA may be an appealing alternative to TXA for TKA surgery since it is more cost-effective. Many studies have shown that topical application of TXA in TKA is effective and safe, therefore comparable to intravenous systemic application. Topical administration of ACA has shown satisfactory effects and safety in total joint arthroplasty, although it has been reported relatively rarely. ²⁵ However, it is debatable whether topical use of TXA and ACA is not harmful to the human cartilage. This topic is relevant in TKA because patella is preserved in many cases and there are studies where TXA was used in unicompartmental knee arthroplasty (UKA). ²⁶ Furthermore, TXA (and in some cases ACA) have been employed in a wide range of other orthopedic procedures beyond arthroplasty, including high tibial osteotomy, anterior cruciate ligament reconstruction, periacetabular osteotomy, hip hemiarthroplasty, and arthroscopic rotator cuff repair.^{8 -10,27,28} The broad use of these antifibrinolytics underscores the importance of understanding their potential chondrotoxic effects. Therefore, the possibility of TXA- or ACA-induced cartilage damage should be taken into account when applying these agents in clinical practice

Previous studies on cytotoxicity to cartilage were conducted mainly for TXA in vitro, while studies on ACA or those in vivo were lacking. The present study provides some insight into how ACA as well as TXA act in vivo. Parker et al. ²⁹ showed that increasing concentrations above 20 mg/ml resulted in atypical morphology, reduced cellular adhesion, and metabolic activity associated with increased chondrocyte death. Also, they recommended that tissue concentration of 10 to 20 mg/ml could be expected to be safe to topically apply. In contrast, Ambra et al. ²⁶ reported that topical use of TXA did not demonstrate any cytotoxic effects on chondrocyte. However, it is noteworthy that cartilage explants in a Yucatan mini pig model were used in their study, and relatively lower concentration of TXA (1, 2, and 4 mg/ml) was applied. It is notable that McLean et al. ³⁰ suggest that attention must be paid when exposing articular tissues to TXA at high levels or prolonged time. The method of intra-articular injection or infusion extended surgical soaking, or tissue saturation, especially in an enclosed joint space, needs to be examined in conjunction with the findings of our study that demonstrated increasing cell mortality.

In theory, topical administration of TXA or ACA may avoid the danger of potentially increased systemic side effects including increased coagulating risk, so it may be especially beneficial in some high-risk patients. Nevertheless, some studies suggest that topically used TXA in doses of 1.5 and 3 g can reach mean plasma values of 4.5 and 8.5 mg/l. ⁶ In that study, plasma concentrations 1 h after administration of 10 mg/kg TXA have yielded mean levels of 18 mg/l. Therefore, topical application of TXA, ACA may still have systemic effects and, possibly, may result in systemic side effects also.^6,31,32

It should be noted that, although TXA was cytotoxic to chondrocytes in this study, its effects on other cell types present in the joint (such as synovial cells, inflammatory cells, and osteoblasts) remain unknown and could be similar.^30,33

In the literature, effective dosing for topical TXA ranges from 250 mg to 3 g, so we commenced to study the effect on cell viability at 10 and 30 mg/ml, maintaining TXA in culture for 24 and 48 h. These concentrations correspond to those used clinically, supporting the relevance of our findings.

The selection of the MIA-induced osteoarthritis (OA) model was deliberate and offers a key advantage for this study’s translational relevance. ¹⁶ Clinically, topical antifibrinolytics are applied to joints with pre-existing degenerative changes or surgical trauma. By inducing cartilage damage with MIA prior to TXA or ACA administration, our model more accurately simulates this clinical context than would a model using healthy, intact cartilage. This approach allows for a direct assessment of drug-induced toxicity on vulnerable chondrocytes, which is the central question of our investigation. The MIA model is a well-validated and reproducible method for inducing OA-like pathology in rats, characterized by chondrocyte apoptosis and progressive cartilage degradation that mimics human OA.^17,18

A key strength of this study is that it compares the effects of TXA and ACA on cartilage and chondrocytes in both in vitro and in vivo settings. Previous investigations typically examined TXA’s effects on cartilage or chondrocytes in vitro only, so our combined approach may better reflect clinical conditions.

Nonetheless, our study has several limitations. First, conclusions drawn from animal studies may not fully translate to human tissues due to species-specific biological differences. For instance, rat cartilage is thinner and has higher cellular density compared with human cartilage, which may affect drug diffusion dynamics and metabolic responses. Second, the sample size was relatively small. Third, cartilage damage in vivo was assessed with only a single scoring method, which may not capture all facets of cartilage degeneration.

The MIA model, while advantageous for simulating pre-existing damage, induces OA through chemical chondrocyte necrosis, which differs from the slow, mechanical, and inflammatory cascade of idiopathic human OA, limiting the translatability of findings related to disease mechanisms. ¹⁷ Future studies should explore a wider range of concentrations to better define safety thresholds and evaluate effects on other intra-articular tissues.

Conclusion

We demonstrated that both TXA and ACA exhibit dose-dependent and time-dependent toxic effects on human chondrocytes. This chondrotoxicity was observed both in vitro and in vivo, with TXA showing greater detrimental effects than ACA. From a safety perspective, using lower concentrations (10 mg/ml) is recommended for topical application of TXA or ACA to minimize potential damage to chondrocytes and cartilage tissue. These results should be taken into consideration when determining the optimal dosage and administration method for TXA or ACA in procedures where cartilage preservation is crucial.