The comparative efficacies of intravenous administration and intra-articular injection of tranexamic acid during anterior cruciate ligament reconstruction for reducing postoperative hemarthrosis: a prospective randomized study

With the popularization of sports and the frequent occurrence of accidents, the incidence of anterior cruciate ligament (ACL) injury increases year by year [1]. Arthroscopically assisted ACL reconstruction (ACLR) is a common, reproducible and minimally invasive procedure. However, after ACLR, hemarthrosis and its associated pain afflict patients and can disrupt rehabilitation programs [2]. Postoperative hemarthrosis has a 3 to 10% incidence after ACLR [2]. Hemarthrosis creates many adverse effects, including increased susceptibility to infection, potential toxic effects to the cartilage, possible subsequent synovitis and arthrofibrosis, the onset of fever episodes, delayed rehabilitation, and prolonged hospital stay [2]. These aspects negatively affect the outcomes of patients and result in increased costs [3]. Minimizing postoperative intra-articular hemarthrosis may be potentially advantageous for recovery in ACLR patients.

Tranexamic acid (TXA), as an antifibrinolytic agent derived from lysine, binds to lysine receptor sites on plasminogen and prevents the formation of plasmin, thus inhibiting the breakdown of fibrin clots and reducing active bleeding [4]. TXA has a great role in reducing postoperative blood loss, the need for transfusion and hospital-related costs after knee surgery without increasing the surgical risks [5‐8]. There are two major administration routes in the use of TXA: intravenous (IV) administration and intra-articular (IA) injection. Both IV administration and IA injection of TXA have been previously reported to effectively reduce hemarthrosis during ACLR. Some studies reported that IV administration of TXA could reduce the amount of drainage blood and postoperative hemarthrosis, reduce fever episodes and the need for aspiration of the knee, and improve knee range of motion (ROM) and early-phase outcomes without side effects after ACL reconstruction [9, 10]. Research on intra-articular injection of TXA during ACLR reported that it could significantly reduce postoperative intra-articular bleeding, pain and the grade of hemarthrosis, and no systemic side effects or need for aspiration was noted during the follow-up period [11]. However, the comparative efficacies of IV and IA TXA for reducing postoperative hemarthrosis during ACLR remain elusive.

This study aimed to compare the efficacies of intravenous administration and intra-articular injection of TXA during ACLR for improving early-phase outcomes. The hypotheses were that both IV administration and IA injection of TXA could reduce postoperative hemarthrosis, the amount of drainage blood, and pain, and that their performances might be similar.

In this study, 120 patients were included and completed the follow-up, and 40 patients were included in each group. Table 1 shows the comparison of preoperative and intraoperative data in the IV group, IA group and placebo group. According to the demographic data, there were no significant differences in age, sex or BMI among the different groups (p = 0.272 for age; p = 0.781 for sex; p = 0.353 for BMI). When comparing the time from injury to surgery and the operation time, no statistically significant differences were found among the three groups (p = 0.590 for time from injury to surgery; p = 0.100 for operation time). The proportions of patients who received general anesthesia and spinal/epidural blocks in the different groups were similar (p = 0.590). According to the meniscus treatment, the surgical methods were ACLR alone, ACLR + partial meniscectomy and ACLR + meniscus repair. We found that there was no significant difference in the proportions of patients who underwent ACLR alone, ACLR + partial meniscectomy and ACLR + meniscus repair in the IV group, IA group and placebo group (p = 0.854). In addition, no significant difference was found among the three groups in terms of tunnel diameter (p = 0.281).

Table 2 shows the comparison of postoperative outcomes among the different groups. The drain output volumes of the IV group, IA group and placebo group were 78.5 ± 38.4 mL, 63.3 ± 31.8 mL and 120.1 ± 57.4 mL, respectively. Statistical analysis showed that the drain output volumes of the IV group and IA group were significantly higher than that of the placebo group (p = 0.000). Although the drain output volume of the IV group was slightly higher than that of the IA group, no significant difference was found between them (p = 0.123).

The VAS scores and patellar circumferences of each group decreased gradually from 1 to 4 weeks postoperatively (Table 2). The VAS scores of the IV group and IA group were significantly lower than those of the placebo group 1 week and 2 weeks postoperatively (at 1 week: p = 0.000; at 2 weeks: IV vs. placebo p = 0.001, IA vs. placebo p = 0.000), but there was no significant difference in VAS scores between the IV group and IA group (at 1 week: p = 0.195; at 2 weeks: p = 0.680). The patellar circumferences of the IV group and IA group were significantly larger than those of the placebo group 1 week postoperatively (p = 0.000), but no significant difference was found between the IV group and IA group (p = 0.354); the comparison of the patellar circumference between different groups at 2 weeks after surgery had similar results. The hemarthrosis grades of the IV group and IA group at weeks 1 and 2 were significantly lower than that of the placebo group (at 1 week: p = 0.000; at 2 weeks: IV vs. placebo p = 0.015, IA vs. placebo p = 0.024), but the hemarthrosis grades of the IV group and IA group at weeks 1 and 2 were similar (at 1 week: p = 0.936; at 2 weeks: p = 0.932).

Four weeks postoperatively, the VAS scores, patellar circumferences and hemarthrosis grades of the IV group, IA group and placebo group exhibited no significant differences (p = 0.768 for VAS scores; p = 0.660 for patellar circumferences; p = 0.455 for hemarthrosis grades; Table 2). There were no infections in either group, and no patient developed DVT by postoperative week 4.

In this study, we found both intravenous administration and intra-articular injection can reduce intra-articular hemarthrosis, joint pain and swelling during ACL reconstruction, and no significant difference in the efficacies of reducing hemarthrosis, joint pain and swelling was found between intravenous administration and intra-articular injection.

Intravenous administration of TXA can reduce the amount of drainage blood and postoperative hemarthrosis after ACLR without side effects [9, 10, 15] (Table 3). In this study, it was found that intravenous administration of TXA could reduce postoperative drainage volume and hemarthrosis following ACLR and alleviate pain and knee swelling, similar to the results of the previous study [9, 10, 15]. TXA may penetrate large joints efficiently after IV administration [9, 16]. TXA was shown to exert its beneficial effects not only by reducing blood loss but also through its anti-inflammatory effects, which might improve analgesia after surgery [17]. This might explain the significantly lowered VAS score in patients who received IV-administered TXA. However, we found that IV administration of TXA did not significantly improve the Lysholm score at 4 weeks. Hetsroni [15] reported that compared to no use of TXA, IV administration of TXA did not improve the Lysholm score at 3 months in patients undergoing ACLR, but Karaaslan et al. [10] found a significant difference in the Lysholm score between IV administration and no use of TXA 2 and 4 weeks postoperatively. The differences in the comparisons of Lysholm scores might be attributed to differences in TXA dose and frequency in different studies. Because the mean duration of the effect of TXA is approximately 3 h, some studies suggested that a second dose should be administered to extend the effect over the first 6 h, when most bleeding occurs [18, 19]. Our study and Hetsroni’s study [15] only gave intravenously administered TXA before tourniquet release, which might not be sufficient to maximize the positive physiological effects of TXA; however, 10 mg/kg of intravenous TXA was given before surgery and an intravenous infusion of 10 mg/kg/h was continued for 3 h after the operation in Karaaslan’s study [10]. The regimen for intravenous administration of TXA requires further investigation.

In this study, we found that intra-articular injection of TXA could also significantly reduce drainage blood volume and postoperative hemarthrosis and relieve pain and knee swelling. Intra-articular injection of TXA can be quickly absorbed locally; the physiological half-life of TXA in the joint cavity is approximately 3 h [20], thus local hemostasis can be achieved. To date, there has been only one study on intra-articular injection of TXA to reduce intra-articular hemarthrosis after ACLR [11], and the results of that study were similar to ours. The confusion of intra-articular injection of TXA after ACLR was that it was unclear whether TXA could harm the knee environment, particularly by potentially causing apoptosis of chondrocytes and tenocytes. Some studies have shown that a higher concentration of TXA might have a detrimental effect on animal or human chondrocytes in vitro [21, 22]. Parker et al. [23] showed that TXA (with concentrations up to 40 mg/mL) had no effect on the glycosaminoglycan content of human articular chondrocyte hydrogels after 6 h of exposure. Ambra et al. [24] found doses of TXA for clinical topical use did not demonstrate any cytotoxic effects on cartilage explants in a Yucatan mini pig model. It is worth noting that a post-arthroscopic knee is filled with some irrigation fluid and hemarthrosis, which might further lower the true concentration of TXA. The optimal dosage of intra-articular TXA still needs to be clarified in the future.

To the best of our knowledge, no study has compared the efficacies of intravenous administration and intra-articular injection of TXA in reducing postoperative hemarthrosis after ACLR. We found that there were no significant differences between intravenous administration and intra-articular injection of TXA in reducing postoperative blood drainage and hemarthrosis, alleviating pain and knee swelling, and improving knee function. Similarly, the comparative efficacies of IV administration and IA injection for reducing blood loss during total knee arthroplasty (TKA) are still controversial. Some studies reported that compared to IV administration, IA injection of TXA seemed to be more effective in terms of reducing blood loss and transfusion frequency [21, 22, 25]. In addition, some researchers concluded that there was no statistically significant difference in blood loss and transfusion rates between IA and IV use in TKA [26, 27]. Most likely, either IA administration or IV injection had its own advantage. For example, Sarzaeem et al. [28] reported that IA injection of TXA was more effective at decreasing postoperative drainage after TKA, but IV administration of TXA seemed to reduce the number of transfused units and the magnitude of the drop in hemoglobin more effectively. The argument for intravenous administration of TXA was that only a small portion of intravenously administered TXA reached the target surgical site, whereas intra-articular application of TXA could supply a higher concentration. The argument for intra-articular administration of high dose of TXA is the problem of potential toxicity to chondrocytes. Doses of TXA for clinical topical use did not demonstrate any cytotoxic effects on cartilage in vitro and in vivo [23, 24]. Therefore, it may be the surgeon’s preference as to what route of administration to use.

There were several limitations in this study. First, the follow-up period (4 weeks) was not long enough and there were not enough observation times points for the Lysholm scores. The first 3 months postoperatively is a significant period for preventing a delay in the rehabilitation program, so the exact long-term effect of TXA application on outcomes after ACLR could not be effectively analyzed in this study. Second, we only administered TXA intravenously 10 min before tourniquet release, which might not be sufficient to maximize the positive physiological effects of TXA. The regimen of intravenous administration of TXA requires further investigation.

Both intravenous and intra-articular applications of TXA were effective and efficient for reducing postoperative blood drainage, intra-articular hemarthrosis, joint pain and swelling after arthroscopic ACL reconstruction. The efficacies of intravenous administration and intra-articular injection of TXA during ACL reconstruction for reducing postoperative hemarthrosis were similar.