Introduction

Total knee arthroplasty (TKA) improves mobility and quality of life and eliminates knee pain in patients with knee arthritis; however, a major postoperative complication is venous thromboembolism (VTE) [1]. Without thromboprophylaxis, the incidence of deep venous thrombosis (DVT) and pulmonary embolism (PE) ranges up to 22% and 10%, respectively [1,2,3,4,5,6,7,8]. The risk of thrombotic events can be decreased though, using prophylactic agents, such as low molecular weight heparin (LMWH) and fondaparinux that have reduced the cumulative incidence of VTE to 2.3% at 3 months after TKA [1]. Recently, new oral anticoagulants (dabigatran, rivaroxaban, apixaban) have been introduced for thromboprophylaxis in orthopedics [1,2,3]. Their use is favorable because of their easy administration, predictable pharmacokinetics, no requirement for monitoring, better patients’ compliance, and lower food and drug interactions [1,2,3]. However, the use of rivaroxaban (RIV) has been associated with a higher risk of hemorrhagic complications compared to LMWH [1,2,3,4,5,6,7,8].

Combined intravenous and intra-articular administration of tranexamic acid (TXA) in TKA patients has been associated with significant decrease in total blood loss, lower transfusion requirements, decreased postoperative hemoglobin (Hb) value, and shorter hospital stay [9,10,11,12,13,14]. Its excellent clinical efficacy and safety in comparison with intravenous TXA administration alone has been also demonstrated [9,10,11,12,13,14]. Additionally, the use of TXA in TKA patients has not been associated with adverse reactions [9,10,11,12,13, 15, 16]. However, to the best of our knowledge, the combined effect of TXA with thromboprophylaxis agents in TKA has not been specifically studied. Therefore, we performed this study to evaluate the efficacy of the combined intravenous and intra-articular administration of TXA to control the collateral effects and complications of RIV after TKA and to compare thromboprophylaxis schemes with and without TXA, RIV and LMWH.

Materials and methods

We prospectively studied a consecutive cohort of 158 patients with knee arthritis that underwent TKA at the authors’ institutions from January 2014 to January 2018. Inclusion criterion was primary varus gonarthrosis. Exclusion criteria were previous surgery at the knee, thrombocytopenia, anemia (Hb < 10 g/dl), warfarin therapy, coagulopathy, previous VTE and significant comorbidities such as ischemic heart disease, cerebrovascular accident, liver cirrhosis and renal disease. The mean follow-up was 2 years (range 7 months to 4 years); no patient was lost to follow-up. All patients gave written informed consent for their data to be included in this study. This study was approved by the Institutional Review Board/Ethics Committee of the authors’ institution.

At admission, the patients were randomly allocated into 3 study groups using a sealed envelope method. Group A patients (n = 46) were administered 10 mg/kg TXA intravenously (iv) before the inflation of the tourniquet and 1000 mg TXA intra-articularly (ia) in a 20 ml solution through the suction drainage after the completion of TKA and wound closure; postoperatively, these patients were administered 10 mg RIV daily for 25 days (a total of 30 days from the operation). Group B patients (n = 58) were administered TXA as in group A; postoperatively, these patients were administered LMWH subcutaneously (sc) daily as in group A. Group C patients (n = 54) were administered natural saline (NS) instead of TXA as in group A; postoperatively, these patients were administered RIV as in group A. Groups were similar with respect to the examined variables, without any potentially confounding variables between groups at baseline (Table 1).

Table 1 Details of the patients included in this series at baseline
Full size table

All patients underwent a unilateral, cemented TKA using the same implants (EVOLUTION® Medial-Pivot Knee System, Microport Orthopedics Inc, Arlington, TN, USA). An intramedullary alignment rod was used for the femoral and an extramedullary guiding system for the tibial preparation in all cases. The tourniquet was inflated before the incision and was not released before skin closure and application of compressive dressings. A single intra-articular drain was used that was clamped for the first 3 h postoperatively; thereafter, the drain was left open and removed 48 h after surgery, no matter what the drain output was. Postoperative rehabilitation included mobilization with a walker, muscle strengthening and range of motion exercises as tolerated by the first postoperative day. The patients were discharged from the hospital when they were able to walk independently with a walker (mean 5 days; range 4–7 days postoperatively).

At baseline and within hospitalization (days 1–4), we measured hemoglobin (Hb) and hematocrit (Hct) values, prothrombin time (PT), activated partial thromboplastin time (APTT), platelet count, and total suction drain blood volume. We calculated total Hb loss by the difference between Hb at baseline and Hb at postoperative day 4, because at that time we consider blood volume to be normalized. The cutoff point for allogenic blood transfusion was set at Hb value < 8 g/dl. After discharged, clinical evaluation was done at 1, 3 and 6 months, and yearly thereafter. Clinical follow-up included wound healing and hemorrhagic complications including ecchymosis or hematoma formation around the knee, cerebrovascular events, gastrointestinal or urinary hemorrhage, and clinical signs of DVT and PE [17]. Statistical analysis was performed using IBM SPSS Software, version 20 (IBM Corp., NY, USA). The Kruskal–Wallis test was used to compare categorical variables, and the one-way ANOVA test to compare continuous variables between the three study groups. A p value < 0.05 was considered indicative for statistical significance.

Results

Hct and Hb values statistically significantly decreased in group C compared to groups A and B, without any difference between groups A and B (Fig. 1a, b, Table 2). Transfusions (2 units of packed red blood cells) were necessary in one patient of group B and two patients of group C. Suction drain blood volume output was significantly higher in group C compared to groups A and B, without any difference between groups A and B (Fig. 1c). The mean suction drain blood volume output was 440 ml, 511 ml and 763 ml for groups A, B and C, respectively.

Fig. 1
figure 1

a A graph shows Hct decrease was significantly higher in group C compared to groups A and B. b A graph shows that Hb decrease was significantly higher in group C compared to groups A and B. c A graph shows that suction drain blood volume output was significantly higher in group C compared to groups A and B

Full size image
Table 2 Statistical comparison (p values) between study groups
Full size table

By direct comparison of data, hemorrhagic complications were more common in group C. One patient of group A experienced knee swelling and hematoma formation that gradually subsided during the follow-up. Three patients of group C experienced knee swelling and hematoma formation, one patient experienced extended pre-patellar ecchymosis and two patients experienced urinary tract hemorrhage (macroscopic hematuria). No patient of group B experienced any hemorrhagic complications. No patient of any group experienced any clinical signs of DVT or PE that required further imaging investigation during the follow-up period.

Discussion

TKA may be associated with substantial blood loss and need for transfusion; however, blood loss and transfusion requirements have been associated with immunologic reactions, disease transmission, and increased risk of surgical complications, time for hospitalization and overall cost of treatment [9,10,11,12,13,14]. In this setting, the use of TXA is promising, as a safe and effective method to reduce hemorrhage [9,10,11,12,13, 15, 16, 18,19,20,21], especially in combined intravenous and intra-articular administration [9,10,11,12,13]. In the present study, we evaluated the effect of TXA associated with RIV or LMWH in TKA patients. Our results showed that combined intravenous and intra-articular administration of TXA is associated with lower blood loss and fewer transfusion requirements and hemorrhagic complications compared to placebo (saline) in TKA patients, without any difference in thromboprophylaxis with RIV and LMWH. These results support the role of TXA in TKA patients regardless of the thromboprophylactic agent administered postoperatively in these patients.

ΤΧΑ is an inhibitor of the conversion of plasminogen to plasmin through the reversible blockade of fibrinolysin-binding sites on the plasminogen. Previous studies have reported the efficacy of either intravenous or intra-articular ΤΧΑ in TKA patients including significantly reduced blood loss, transfusion requirements and hospital stay without any related complications [11, 22,23,24,25,26,27]. However, to the best of our knowledge, only one published study has reported on the efficacy of combined intravenous and intra-articular TXA administration in TKA patients [26]. All the above studies concluded that TXA is effective and safe and significantly reduces postoperative blood loss and transfusion requirements, as well as wound hematoma formation [22, 23, 28,29,30,31]. Less transfusion requirements, fewer laboratory examinations and shorter hospital stay are advantages associated with TXA administration that may lead to reduction in the overall cost of treatment [21, 32,33,34]. After TKA, a 3-h suction drain clamping can result in temporary hemostasis and may control hematoma formation without increasing thromboembolic events or the risk of hematoma formation and wound healing complications [35,36,37]. There is evidence that TXA administration alone could reduce blood loss and transfusion requirements [25], but it seems that TXA plus suction drain clamping, as in this series, can achieve better results in terms of hemostasis [23].

Using appropriate thromboprophylaxis in TKA patients has decreased the incidence of DVT to 1–3% and of PE to 0.2–1.1% [38]. Conventional anticoagulants, such as LMWH, warfarin, aspirin, as well as newer agents, such as fondaparinux, RIV, apixaban and dabigatran, are available for thomboprophylaxis [16, 39]. The most widely used agents and current preferences in most guidelines are LMWH; however, LMWH have been associated with limitations such as the injectable administration, risk of heparin-induced thrombocytopenia (approximately 0.2%), and inactivation of factors related to clotting [40]. Additionally, the incidence of significant hemorrhagic complications with the use of any type of thromboprophylaxis is reported from 1.8 to 5.1%, which may be underestimated in the well-controlled reports, and this incidence may be increased with the more effective agents [41, 42]. Newer oral anticoagulants have shown equal efficacy and safety compared to LMWH, improved patients’ quality of life and treatment compliance, and healthcare cost reduction [43,44,45,46]. RIV showed superior efficacy as compared to enoxaparin for the prevention of DVT after TKA, however at the cost of an increased risk of hemorrhagic and wound healing complications, even though these were not reported at a statistically significant difference [6, 8, 38, 47, 48]. RIV has also been associated with a higher rate of hemorrhagic complications when used without TXA after TKA [30, 49]. Similarly, dabigatran led to a significant increase in postoperative wound leakage (20%) that resulted in increased hospital stay [50]. To reduce the risk of oral anticoagulants-related complications switch-therapy modalities have been recommended; the patients can take advantage of the safety of LMWH in the first postoperative days, when it is most likely to present wound bleeding events and then switch to the more convenient oral anticoagulants during the outpatient follow-up period [17].

In conclusion, combined intravenous and intra-articular administration of TXA is safe and effective in TKA, with fewer hemorrhagic complications compared to placebo. Thromboprophylaxis with RIV and LMWH is similar.