The Efficacy of Combined Use of Rivaroxaban and Tranexamic Acid on Blood Conservation in Minimally Invasive Total Knee Arthroplasty a Double-Blind Randomized, Controlled Trial

The Efficacy of Combined Use of Rivaroxaban and Tranexamic Acid on Blood Conservation in Minimally Invasive Total Knee Arthroplasty a Double-Blind Randomized, Controlled Trial

The Journal of Arthroplasty xxx (2016) 1e6 Contents lists available at ScienceDirect The Journal of Arthroplasty journal homepage: www.arthroplastyj...

255KB Sizes 4 Downloads 42 Views

The Journal of Arthroplasty xxx (2016) 1e6

Contents lists available at ScienceDirect

The Journal of Arthroplasty journal homepage: www.arthroplastyjournal.org

Original Article

The Efficacy of Combined Use of Rivaroxaban and Tranexamic Acid on Blood Conservation in Minimally Invasive Total Knee Arthroplasty a Double-Blind Randomized, Controlled Trial Jun-Wen Wang, MD a, *, Bradley Chen, MD, ScD b, Po-Chun Lin, MD a, Shih-Hsiang Yen, MD a, Chung-Cheng Huang, MD c, Feng-Chih Kuo, MD a a b c

Department of Orthopaedic Surgery, Kaohsiung Chang Gung Memorial Hospital, College of Medicine, Chang Gung University, Kaohsiung, Taiwan Institute of Public Health, National Yangming University, Taipei, Taiwan Department of Radiology, Kaohsiung Chang Gung Memorial Hospital, College of Medicine, Chang Gung University, Kaohsiung, Taiwan

a r t i c l e i n f o

a b s t r a c t

Article history: Received 9 June 2016 Received in revised form 17 August 2016 Accepted 18 August 2016 Available online xxx

Background: Tranexamic acid (TXA) was reportedly to decrease postoperative blood loss after standard total knee arthroplasty (TKA). However, the blood-conservation effect of TXA in minimally invasive TKA, in particular, receiving a direct oral anticoagulant was unclear. The aim of the study was to investigate the efficacy of combined use of TXA and rivaroxaban on postoperative blood loss in primary minimally invasive TKA. Methods: In a prospective, randomized, controlled trial, 198 patients were assigned to placebo (98 patients, normal saline injection) and study group (100 patients, 1g TXA intraoperative injection) during primary unilateral minimally invasive TKA. All patients received rivaroxaban 10 mg each day for 14 doses postoperatively. Total blood loss was calculated from the maximum hemoglobin drop after surgery plus amount of transfusion. The transfusion rate and wound complications were recorded in all patients. Deep-vein thrombosis was detected by ascending venography of the leg 15 days postoperatively. Results: The mean total blood loss was lower in the study group (1020 mL [95% confidence interval, 960-1080 mL]) compared with placebo (1202 mL [95% confidence interval, 1137-1268 mL]) (P < .001). The transfusion rate was lower in the study group compared with placebo (1% vs 8.2%, P ¼ .018). Postoperative wound hematoma and ecchymosis were higher in placebo than the study group (P ¼ .003). There was no symptomatic deep-vein thrombosis or pulmonary embolism in either group. Conclusion: Systemic administration of TXA can effectively reduce the postoperative blood loss which results in lower rate of transfusion requirement and wound hematoma in minimally invasive TKA patients when rivaroxaban is used for thromboprophylaxis. Rivaroxaban has a high rate of bleeding complications when used alone in TKA patients. © 2016 Elsevier Inc. All rights reserved.

Keywords: rivaroxaban tranexamic acid venous thromboembolism total knee arthroplasty total blood loss wound complications

Total knee arthroplasty (TKA) is the most common and effective procedure for end-stage arthritis of the knee in terms of pain relief and functional recovery [1,2]. However, this procedure is associated

No author associated with this paper has disclosed any potential or pertinent conflicts which may be perceived to have impending conflict with this work. For full disclosure statements refer to http://dx.doi.org/10.1016/j.arth.2016.08.020. Financial support: This study was supported in part, by Ministry of Science and Technology, grant #NMRPG8136181-2. * Reprint requests: Jun-Wen Wang, MD, Department of Orthopaedic Surgery, Kaohsiung Chang Gung Memorial Hospital, Chang Gung University, College of Medicine, No.123, Da Pi Rd., Niao Sung Dist., Kaohsiung City, Taiwan, ROC. http://dx.doi.org/10.1016/j.arth.2016.08.020 0883-5403/© 2016 Elsevier Inc. All rights reserved.

with substantial perioperative blood loss. An allogeneic blood transfusion rate as high as 69% was reported in patients receiving total hip and knee arthroplasty when the preoperative hemoglobin (Hb) level was <13 g/dL [3]. Both surgery and the application of a pneumatic tourniquet are considered to enhance coagulability and local fibrinolytic activity in the limb [4-6]. Tranexamic acid (TXA) is a synthetic amino acid derivative which inhibits the conversion of plasminogen to plasmin through the reversible blockade of fibrinolysin-binding sites on the plasminogen molecule [7-9]. Several clinical trials [10-13] and meta-analyses [14-17] have demonstrated that intravenous TXA reduced blood loss and the need for transfusion in primary total hip and knee arthroplasty.

2

J.-W. Wang et al. / The Journal of Arthroplasty xxx (2016) 1e6

Rivaroxaban is one of the first licensed oral factor Xa inhibitors for thromboprophylaxis in total hip and knee arthroplasty surgeries. The advantages of rivaroxaban, which include oral administration, no need to monitor blood levels, and no dosing adjustments, make this drug convenient for the short hospital stays required in contemporary TKA [18]. Its efficacy in preventing venous thromboembolism (VTE) after TKA has been shown to be superior to enoxaparin given subcutaneously in RECORD (Regulation of Coagulation in Orthopaedic surgery to prevent Deep-vein thrombosis and pulmonary embolism) 3 and 4 studies [19,20]. Despite its clinical efficacy in VTE prophylaxis, orthopedic surgeons are still skeptical regarding the routine use of rivaroxaban in knee and hip surgery and, in particular, the increased risk of bleeding complications [21]. A higher reoperation rate from wound complications within 30 days of hip and knee arthroplasty in the rivaroxaban group compared with the tinzaparin group (2.94% vs 1.8%) was also recently reported [22]. Similar concerns have been raised by other authors [21,23-26]. However, none of these studies used TXA as bleeding prophylaxis after hip and knee arthroplasty surgery. Owing to its antifibrinolytic effects, there is an increased risk of VTE after TXA use [27-29]. Furthermore, most of the studies reporting the blood-conservation effect of TXA in total knee and hip arthroplasty used lowemolecular weight heparin for thromboprophylaxis [10-12,30]. Minimally invasive techniques for TKA are increasing recently in orthopedic service. The advantages of minimally invasive TKA are less wound pain [31,32], faster rehabilitation [32,33], shorter hospital stay [33], and possible less blood loss [31] compared with conventional TKA. However, it is unclear that whether there is a blood-conservation effect of TXA in patients who undergo minimally invasive TKA and receive modern oral anticoagulants for thromboprophylaxis. The aim of this study was to conduct a prospective, randomized, double-blind study to assess the blood-conservation effect of TXA and wound hemotoma related to postoperative blood loss when rivaroxaban is used for thromboprophylaxis in minimally invasive TKA patients. Methods Patients were included in the study if they were 18 years of age, or older, and were scheduled for unilateral primary TKA. Patients were excluded from the study if they had a coagulopathy, severe renal impairment (creatinine clearance, <30 mL/min), concomitant use of protease inhibitors of human immunodeficiency virus, or fibrinolytic agents that contraindicated the use of rivaroxaban, prior surgery on the affected knee, a history of thromboembolic disease requiring life-long anticoagulant therapy or antiplatelet drugs that could not be stopped before operation, previous allergic history to TXA, or contrast medium for radiographic examination or a preoperative Hb level less than 10 g/dL. The present study was registered in the public ClinicalTrials.gov registry (NCT02458729) and was approved by the institutional review board of our institution. Written informed consent was obtained from all participating patients before randomization. Between August 2013 and April 2015, 250 patients scheduled for primary unilateral TKA were assessed for eligibility. Fifty patients were excluded because of ineligibility. The remaining 200 patients were randomly assigned, by means of a computer-generated randomization method, to either placebo (100 patients) or the study group (100 patients, 1 g TXA intravenous injection intraoperatively) (Table 1). Two patients in placebo group refused to participate after allocation. Thus, a total of 198 opaque, sealed envelopes which contained appropriate study medications and placebo were randomly numbered from 1 to 198 by means of a computer-generated method at the Institutional Clinical Trial

Table 1 Patients Included and Excluded for Analyses. Enrollment of Patients

Placebo (%)

Study Group (%)

Randomized Patients underwent surgery and took study medication Patients in primary and secondary safety outcome analysis Refused to participate after randomization Patients in primary safety outcome analysis Patients in radiographic evaluation for deep-vein thrombosis Refused venographic examination Inadequate renal function (GFR <60 mL/min/1.73m2) Failure of venography

100 (100) 100 (100)

100 (100) 100 (100)

98 (98)

100 (100)

2 (2)

0 (0)

98 (98)

100 (100)

66 (66)

67 (67)

21 (21) 7 (7)

21 (21) 6 (6)

4 (4)

6 (6)

GFR, glomerular filtration rate.

Pharmacy. The 198 envelopes included 98 envelopes containing 2 ampoules of 20 mL of normal saline (placebo group) and 100 envelopes containing one ampoule of 20 mL of normal saline and four ampoules of 250 mg/5 mL TXA (China Chemical and Pharmaceutical Co, Taiwan) (study group). All patients underwent unilateral primary TKA using the minimidvastus technique according to Haas et al [31] under general anesthesia. They described: the skin incision was made along the medial border of mid-to-distal tibial tubercle, and the vastus medialis oblique muscle was split approximately 2-cm inline with its fibers from the superior medial pole of the patella [32]. A pneumatic tourniquet was inflated to a pressure of 300 mmHg before the incision and released at the end of surgery after skin closure. All TKAs were cemented using the same prosthesis (NexGen, Legacy, Posterior-Stabilized Prosthesis; Zimmer, Warsaw, IN). Before wound closure, 2 drain tubes were placed into the knee joint and connected to the vacuum bag. Then, the envelop was sequentially opened by circulating nurse not involved in this study before each surgery. The attending anesthetists and the nurse in charge, who were not involved in the study, administered the study medications according to the groupings. Thus, patients in placebo received 20-mL saline intravenously, 5 minutes before deflation of the tourniquet, and patients in the study group received 1 g TXA by slow intravenous infusion 5 minutes before deflation of the tourniquet. Patients, surgeons, research assistant participating in clinical evaluation were blinded as randomization until data analysis was completed. The length of wound in full extension of the knee was recorded at the end of operation in all patients. All patients received 10 mg of oral rivaroxaban (Xarelto, Bayer Shering Pharma AG, Wuppertal, Germany) once daily starting from postoperative day (POD) 1 to POD 14 for VTE prophylaxis. No other anticoagulants or antiembolic devices were used in all patients. Preoperative data, including Hb level, hematocrit, prothrombin time, activated partial thromboplastin time, D-dimer levels, and platelet count were collected. The demographics of the patients, including age, gender, body mass index, American Society of Anesthesiologists grade [34], and blood indices were similar between the 2 groups (Table 2). Safety Outcomes Postoperative Hb levels were recorded on the first, second, and fourth days postoperatively. The drop of Hb level on POD1, POD2, and POD4 compared with the preoperative Hb level of all patients were recorded. It was assumed that blood volume was normalized

J.-W. Wang et al. / The Journal of Arthroplasty xxx (2016) 1e6 Table 2 Details of the Patients. Characteristics

Placebo (N ¼ 98)

Age (y) BMI (kg/m2) Women (%) Preoperative Hb (g/dL) Preoperative Hct (%) PT INR aPTT INR Platelet count (1000/uL) ASAdno./total no. (%) I II III

69.60 28.08 72 13.15

(7.78;68.0-71.2) (4.01;27.3-28.9) (77.6) (1.15;12.9-13.4)

Study Group (N ¼ 100) 68.19 28.16 74 13.28

P Value

(6.62;66.9-69.5) (4.19;27.3-29.0) (74) (1.22;13.0-13.5)

.171 .903 .477 .443

39.94 (3.17;39.3-40.6)

40.26 (3.46;39.6-40.9)

.503

0.98 (0.06;0.97-0.99) 0.98 (0.06;0.97-0.99) 225 (51;214-235)

0.98 (0.05;0.97-0.99) 0.98 (0.06;0.96-0.99) 232 (54;222-243)

.783 .166 .280 .908

2/98 (2) 60/98 (61) 36/98 (37)

2/100 (2) 63/100 (63) 35/100 (35)

3

recorded. The extent of hematoma or ecchymosis around the operated knee was recorded. The degree of ecchymosis or wound hematoma appearing 1 day after the last dose of rivaroxaban was classified as follows: (1) low grade: minimal ecchymosis around the knee, no hematoma and (2) high grade: obvious hematoma formation around knee (with or without blister formation), major ecchymosis in the knee with or without extension to the ipsilateral thigh or leg. Deep-vein thrombosis (DVT) was assessed on the second day after the last dose of rivaroxaban, or earlier, if symptomatic, by bilateral ascending venography of the legs using the Rabinov and Paulin technique [38]. All cases of suspected DVTs were verified by venography. In cases of suspected pulmonary embolism, computed tomographic angiography of the chest was performed. All the radiographic images were interpreted by an independent radiologist. Statistical Analysis

All data were described as mean (standard deviation and range). BMI, body mass index; Hb, hemoglobin; Hct, hematocrit; INR, international normalized ratio; PT, prothrombin time; aPTT, activated partial thromboplastin time.

on the fourth POD. Total Hb loss was calculated by subtracting the lowest Hb level after operation from the preoperative Hb level. Total blood loss was calculated according to the method of Nadler et al [35], using the maximum postoperative decreased in Hb level adjusted for weight and height of the patients. The formula of total blood loss is as follows: total blood volume X (change in Hb level/ mean Hb) þ volume transfused. Hidden blood loss was calculated by subtracting the amount of drainage for the total blood loss. The drain tubes were removed 36-48 hours after operation. The total amount of the drainage was recorded. The trigger for allogenic red blood cell transfusion was set at a Hb level of 7 g/dL in healthy patients between 7 and 8 gL/dL in patients with symptoms signs of acute anemia including tachycardia, short of breath, and dizziness or between 8 g/dL and 9 g/dL in patients with cardiovascular dis te  Française dAnesth sie et de Re animation ease, according to Socie e guidelines [36]. The amount blood transfusion of all patients was recorded. The length of hospital stay was recorded in all patients. VTE Evaluation and Wound Complications All patients were followed up postoperatively at 2, 6 weeks, and 3 months and were screened for clinical symptoms and signs for deepvein thrombosis such as thigh and calf swelling >3cm compared with the contralateral leg, together with calf tenderness, according to the criteria devised by Well et al [37] at each visit. The wound condition including hematoma formation, bleeding, drainage, or infection were recorded. Any wound complications requiring return to surgery because of hematoma and infection within 30 days of the procedure and 30-day readmission to the orthopedic service were

Based on the mean difference in blood loss and standard deviation obtained from our previous study, a mean difference of blood loss 200 mL with a standard deviation of 300 mL was assumed between groups. To obtain a statistical power of 0.90 and an alpha error of 0.05, 67 patients were required in each group. Expecting 10% loss of follow-up and 20% exclusion due to venographic analysis, 100 patients in each group were enrolled; the total case number was 200 patients. The Mann-Whitney U test was used to determine the differences between the 2 groups in the distribution of demographics and preoperative clinical data. Descriptive data (including gender and American Society of Anesthesiologists level) between the 2 groups were compared using chi-square test or Fisher's exact test. An intention-to-treat analysis was used in patients who undergone planned surgery, had taken study drugs, and had undergone venographic examination to assess for thromboembolism. The independent t test was used to determine the difference in the total blood loss and hidden blood loss between placebo group and the study group. The Mann-Whitney U test was used to determine the differences in the length of hospital stay, postoperative Hb drop, and blood loss in drainage between the 2 groups. Descriptive data including safety outcomes, wound complications, and blood transfusion rate between the placebo and the study group were compared using chi-square test or Fisher's exact test. All statistical comparisons were made using the Statistical Package for Social Sciences (SPSS) (version 18; SPSS Inc, Chicago, IL). Results Safety Outcomes There was a significant drop of Hb level postoperatively in placebo compared with the study group on POD 1 (2.29 ± 0.83 g/dL vs

Table 3 Postoperative Data and Blood Loss Measurement. Event Wound length (cm) mean (SD; 95% CI) Drop of hemoglobin level (g/dL) mean (SD; 95% CI) POD1 POD2 POD4 Total drainage (mL) mean (SD; 95% CI) Blood transfusion no./total no. (%) Hidden blood loss (mL) mean (SD; 95% CI) Total blood loss (mL) mean (SD; 95% CI)

Placebo (N ¼ 98) 8.79 (1.02; 8.58-8.98) 2.29 3.60 3.98 991 8/98 212 1202

(0.83; 2.13-2.46) (0.95; 3.41-3.79) (1.15; 3.75-4.21) (391; 914-1068) (8.2) (421; 129-295) (327; 1137-1268)

All patients had intake of rivaroxaban 10-mg daily for 14 doses starting from POD 1. SD, standard deviation; CI, confidence interval; POD, postoperative day; OR, odds ratio.

Study Group (N ¼ 100) 8.64 (1.18; 8.41-8.87) 1.64 2.86 3.50 953 1/100 60 1020

(0.82; 1.48-1.80) (0.99; 2.66-3.06) (1.05; 3.29-3.71) (328; 888-1018) (1) (400; 19-139) (301; 960-1080)

P Value (OR; 95% CI) .342 <.001 <.001 .003 .464 .018 (8.8; 1.1-71.7) .010 <.001

4

J.-W. Wang et al. / The Journal of Arthroplasty xxx (2016) 1e6

Table 4 Complications related to postoperative bleedings. Complications

Placebo (N ¼ 98)

Study Group (N ¼ 100) No. With Events/Total No. (%)

P Value (OR; 95% CI)

Up to POD 15 All ecchymosis Wound hematoma Deep infection Return to OT because of wound complication 30-day readmission to orthopedic service

29/98 15/98 0/98 0/98 0/98

17/100 3/100 0/100 0/100 0/100

.036 (2.05; 1.04-4.04) .003 (5.84; 1.64-20.89) d d d

(30) (15) (0) (0)

(17) (3) (0) (0)

All patients had intake of rivaroxaban 10-mg daily for 14 doses starting from POD 1. POD, postoperative day; OT, operating theater; CI, confidence interval; OR, odds ratio.

1.64 ± 0.82 g/dL, P < .001), on POD2 (3.60 ± 0.95 d/dL vs 2.86 ± 0.99g/ dL, P < .001) and on POD4 (3.98 ± 1.15g/dL vs 3.50 ± 1.05g/dL, P ¼ .003) (Table 3). The mean wound length in full extension of the knee and total drainage amount was similar between the 2 groups. The mean total blood loss was 1202 ± 327 mL (95% confidence interval, 1137-1268 mL) in placebo which was higher than 1020 mL (95% confidence interval, 960-1080 mL) in the study group (P < .001). The mean hidden blood loss was significantly higher in placebo than in the study group (P ¼ .01). The rate of blood conservation by TXA injection was 15%. The transfusion requirement occurred in 8 of 98 (8.2%) patients in placebo which was higher than 1 in 100 (1%) patients in the TXA group (P ¼ .018) (Table 3). No patients in each group required return to the operating room or 30-day readmission to the orthopedic service because of wound complications related to postoperative bleeding (Table 4). Postoperative ecchymosis of the operated lower limb at the first clinic visit after surgery occurred in 29 of 98 (30%) patients in placebo which was higher than 17 of 100 (17%) patients in the study group (P ¼ .036). There was a significantly higher rate of wound hematoma in placebo (15%) compared with the study group (3%) (P ¼ .003). However, there were no differences in the length of hospital stay between the 2 groups (6.52 vs 6.40 days, P ¼ .34) (Table 5). All ecchymoses resolved spontaneously 1 week after the first clinic visit. There were no episodes of joint stiffness or any latent infections of the knee up to 3 months of follow-up in both groups. VTE Evaluation Sixty-six patients in placebo (66%) and 67 patients in the study group (67%) underwent ascending venography of both legs on POD15. The reasons for patients who did not have a VTE evaluation by venography were shown in Table 1. The results showed that the incidence of positive venography of the leg was similar between the 2 groups (15.2% vs 14.9%). All DVTs occurred in the leg veins. No patients in both groups had symptomatic DVT or pulmonary embolism up to 3 months follow-up (Table 5). Discussion This study shows that TXA administered systemically reduces postoperative Hb loss, total blood loss, hidden blood loss, and

transfusion requirement in patients undergoing minimally invasive TKA with the use of rivaroxaban for thromboprophylaxis. This is the first randomized clinical trial of combined use of TXA and rivaroxaban in minimally invasive TKA. Previously, our clinical trials of TXA in minimally invasive TKA patients showed similar results; however, in those studies, lowemolecular weight heparin was used for thromboprophylaxis [13,39]. Previous reports have indicated that anticoagulants which reduce the incidence of DVT are associated with increased risk of postoperative bleeding [40,41]. Similarly, rivaroxaban, a direct factor Xa inhibitor, has shown a higher risk of postoperative bleeding (6.8% vs 2.2%, P ¼ .004) and relevant nonmajor bleeding (5.5% vs 2.0%, P ¼ .012) compared with enoxaparin after total hip or knee arthroplasty surgeries in one study [42], as well as meta-analyses of 4 clinical trials [43]. An increased postoperative infection rate from 1% to 2.5% and higher hospital stay due to wound complications in total hip and knee patients after introduction of rivaroxaban have been reported [21]. Brimmo et al [24] reported a higher early periprosthetic infection in the rivaroxaban group than in the control group in primary THA and TKA patients. Furthermore, potential long-term problems, such as joint stiffness and latent infection related to prolonged wound drainage, were associated with chemical thromboprophylaxis [44,45]. We demonstrated a 15% blood-conservation effect of intravenous TXA associated with rivaroxaban in primary TKA compared with placebo, which is similar to our previous study using lowe molecular weight heparin for thromboprophylaxis [39] but is low compared with earlier studies in conventional TKA [12]. As with previous report, the low blood-conservation effect may be related to the minimally invasive technique [39]. In our previous retrospective study comparing the efficacy and safety of rivaroxaban with enoxaparin in TKA patients using TXA for bleeding prophylaxis [46], the nonmajor bleeding and wound complications were not different between placebo and the TXA groups, which was similar to this randomized clinical trial. However, in this trial, the major bleeding events (blood transfusion 2 units of blood) were higher in placebo than the TXA group (P ¼ .01), which did not affect the clinical outcome except for a higher incidence of wound hematoma and ecchymosis. The ecchymosis spontaneously resolved 1 week later. No further treatment or operations were required in those patients. Recently, a large cohort of retrospective study

Table 5 Incidence of Venous Thromboembolism. Event

Placebo (N ¼ 98)

Study Group (N ¼ 100)

P Value (OR; 95% CI)

Up to POD 15 pulmonary embolism Proximal deep-vein thrombosis Distal deep-veina thrombosis Symptomatic venous thromboembolism of all patients

0/66 0/66 10/66 0/98

0/67 0/67 10/67 0/100

d d .97 (1; 0.39-2.63) d

(0) (0) (15.2) (0)

All patients had intake of rivaroxaban 10-mg daily for 14 doses starting from POD 1. POD, postoperative day; CI, confidence interval; OR, odds ratio. a The numbers of positive deep-vein thromboses were detected by ascending venography of the legs only.

(0) (0) (14.9) (0)

J.-W. Wang et al. / The Journal of Arthroplasty xxx (2016) 1e6

comparing rivoraxaban and enoxaparin in THA and TKA patients showed no demonstrate differences in rate of infections, reoperation, transfusion, or major bleedings between the 2 groups [47]. However, a tendency to greater risk of minor bleeding and wound complications in patients with rivaroxaban than enoxaparin in hip and knee arthroplasty was reported [26]. Furthermore, a higher early periprosthetic joint infection in patient using rivoraxaban than other form of chemical prophylsxia (2.5% vs 0.2%, P < .015) has been reported [24]. There have been limited studies regarding the combination of TXA with rivaroxaban in total hip and knee arthroplasty. Recently,  et al [48] conducted a prospective, nonrandomized, none Clave double-blind study and reported the efficacy (and tolerance for) and safety of rivaroxaban thromboprophylaxis using a combination of TXA with rivaroxaban in primary total hip arthroplasty. Wood et al [49] reported a reduction in the transfusion requirement from 10.5% to 3% by adding TXA in total knee patients with rivaroxaban prophylaxis by a retrospective review of a prospective cohort. In our current prospective, randomized, controlled trial in minimally invasive TKA patients, TXA reduced the blood loss from 1202 mL in placebo to 1020 mL (15%) accompanied by a lower transfusion rate (8.2% vs 2%) (P ¼ .02). No symptomatic VTE was observed in either group. We recognized limitations of this study. First, our sample size was small. However, our power calculation in recruiting patients was based on expected total blood loss of previous studies not on the occurrence of thromboembolism. Second, although we commenced patients to perform ascending venography of the leg in the postoperative period, only two-third of our patients underwent this procedure because of personal reasons. Therefore, the data of incidence of positive venogram of either group were not used for drawing a conclusion. A larger sample size is needed for investigating the safety of this regimen in terms of thrombosis incidence. Nevertheless, in the current study, no symptomatic VTE has been observed up to 3 months after surgery. In conclusion, our randomized, controlled trial demonstrated the blood-conservation effect of TXA which resulted in a reduced rate of transfusion and wound hematoma when patients used rivaroxaban prophylaxis in minimally invasive TKA. No symptomatic VTE was observed in all patients. Rivaroxaban has a high rate of bleeding complications when used alone in TKA patients. References 1. Clutier JM. Results of total knee arthroplasty with a non-constrained prosthesis. J Bone Joint Surg Am 1983;65:906. 2. Ewald FC, Jacobs MA, Miegel RE, et al. Kinematic total knee replacement. J Bone Joint Surg Am 1984;66(7):1032. mez LA, et al. Preoperative hemoglobin levels and the 3. Salido JA, Marín LA, Go need for transfusion after prosthetic hip and knee surgery: analysis of predictive factors. J Bone Joint Surg Am 2002;84-A(2):216. 4. Fahmy NR, Patel DG. Hemostatic changes and postoperative deep-vein thrombosis associated with use of a pneumatic tourniquet. J Bone Joint Surg Am 1981;63(3):461. 5. Murphy WG, Davies MJ, Eduardo A. The haemostatic responses to surgery and trauma. Br J Anaesth 1993;70(2):205. €ja € J, Myllynen P, Myllyla € G, et al. Fibrinolysis after application of a pneu6. Peta matic tourniquet. Acta Chir Scand 1976;153(11-12):647. 7. Dunn CJ, Goa KL. Tranexamic acid: a review of its use in surgery and indications. Drug 1999;57:1005. 8. Mannucci PM. Hemostatic drugs. N Engl J Med 1988;339:245. 9. Prentice CR. Basis of antifibrinolytic therapy. J Clin Pathol Suppl (R Coll Pathol) 1980;14:35. 10. Benoni G, Fredin H. Fibrinolytic inhibition with tranexamic acid reduces blood loss and blood transfusion after knee arthroplasty: a prospective, randomised, double-blind study of 86 patients. J Bone Joint Surg Br 1996;78(3):434. €ck G, Axelsson K, Ryttberg L, et al. Tranexamic acid reduces blood loss in 11. Ekba total hip replacement surgery. Anesth Analg 2000;91(5):1124. 12. Hiippala S, Strid L, Wennerstrand M, et al. Tranexamic acid (Cyklokapron) reduces perioperative blood loss associated with total knee arthroplasty. Br J Anaesth 1995;74(5):534.

5

13. Lin PC, Hsu CH, Chen WS, et al. Does tranexamic acid save blood in minimally invasive total knee arthroplasty? Clin Orthop Relat Res 2011;469:1995. 14. Alshryda S, Sarda P, Sukeik M, et al. Tranexamic acid in total knee replacement: a systematic review and meta-analysis. J Bone Joint Surg Br 2011;93(12):1577. 15. Cid J, Lozano M. Tranexamic acid reduces allogeneic red cell transfusions in patients undergoing total knee arthroplasty: results of a meta-analysis of randomized control trials. Transfusion 2005;45:1302. 16. Ho KM, Insmail H. Use of intravenous tranexamic acid to reduce allogeneic blood transfusion in total hip and knee arthroplasty: a meta-analysis. Anaesth Intensive Care 2003;31:529. 17. Yang ZG, Chen WP, Wu LD. Effective and safety of tranexamic acid in reducing blood loss in total knee arthroplasty: a meta-analysis. J Bone Joint Surg Am 2012;94:1153. 18. Perzborn E, Strassburger J, Wilmen A, et al. In vitro and in vivo studies of the novel antithrombotic agent BAY 59-7939ean oral, direct factor Xa inhibitor. J Thromb Haemost 2005;3(3):514. 19. Lassen MR, Ageno W, Borris LC, et al. RECORD3 Investigators. Rivaroxaban versus enoxaparin for thromboprophylaxis after total knee arthroplasty. N Engl J Med 2008;358(26):2776. 20. Turpie AG, Lassen MR, Davidson BL, et al. RECORD4 Investigators. Rivaroxaban versus enoxaparin for thromboprophylaxis after total knee arthroplasty (RECORD4): a randomised trial. Lancet 2009;373(9676):1673. 21. Jameson SS, Rymaszewska M, Hui AC, et al. Wound complications following rivaroxaban administration: a multicenter comparison with low-molecularweight heparins for thromboprophylaxis in lower limb arthroplasty. J Bone Joint Surg Am 2012;94(17):1554. 22. Jensen CD, Steval A, Partington PF, et al. Return to theatre following total hip and knee replacement, before and after the introduction of rivaroxaban: a retrospective cohort study. J Bone Joint Surg Br 2011;93(1):91. 23. Lotke PA. Rivaroxaban for thromboprophylaxis. N Engl J Med 2008;359:2174. 24. Brimmo O, Glenn M, Klika AK, et al. Rivaroxaban use for thrombosis prophylaxis is associated with early periprosthetic joint infection. J Arthroplasty 2016;31: 1295. 25. Rath NK, Goodson MW, White SP, et al. The use of rivaroxaban for chemical thromboprephylaxis following total knee replacement. Knee 2013;20:397. 26. Sindali K, Rose B, Soueid H, et al. Elective hip and knee arthroplasty and the effect of rivaroxaban and enoxaparin thromboprophylaxis on wound healing. Eur J Orthop Surg Traumatol 2013;23:481. 27. Arnljots B, Wieslander JB, Dongan P, et al. Importance of fibrinolysis in limiting thrombus formation following severe microarterial trauma: an experimental study in the rabbit. Microsurgery 1991;12:332. 28. Astedt B, Liedholm P, Wingerup L. The effect of tranexamic acid on the fibrinolytic activity of vein walls. Ann Chir Gynaecol 1978;67:203. 29. Wong J, Abrishami A, El Beheiry H, et al. Topical application of tranexamic acid reduces postoperative blood loss in total knee arthroplasty: a randomized, controlled trial. J Bone Joint Surg Am 2010;92(15):2503.  G, Serra-Prat M, et al. Efficacy of aminocaproic, tranexamic 30. Camarasa MA, Olle acids in the control of bleeding during total knee replacement: a randomized clinical trial. Br J Anaesth 2006;96(5):576. 31. Tria Jr AJ. Advancement in minimally invasive total knee arthroplasty. Orthopedics 2003;26(8 Suppl):S859. 32. Haas SB, Cook S, Beksac B. Minimally invasive total knee replacement through a mini midvastus approach: a comparative study. Clin Orthop Relat Res 2004;428:68. 33. Bonutti PM, Mont MA, McMahon M, et al. Minimally invasive total knee arthroplasty. J Bone Joint surg Am 2004;86(Supple 2):26. 34. Owens WD, Felts JA, Spitznagel Jr EL. ASA physical status classification: a study of consistency of ratings. Anesthesiology 1978;49:239. 35. Nadler SB, Hidalgo JU, Bloch T. Prediction of blood volume in normal human adults. Surgery 1962;51:224. 36. Rosencher N, Kerkkamp HE, Macheras G, et al. Orthopedic Surgery Transfusion Hemoglobin European Overview (OSTHEO) study: blood management in elective knee and hip arthroplasty in Europe. Transfusion 2003;43(4):459. 37. Well PS, Anderson DR, Bormanis J, et al. Value of assessment of pretest probability of deep-vein thrombosis in clinical management. Lancet 1997;350:1795. 38. Rabinov K, Paulin S. Roentgen diagnosis of venous thrombosis in the leg. Arch Surg 1972;104(2):134. 39. Lin PC, Hsu CH, Huang CC, et al. The blood-saving effect of tranexamic acid in minimally invasive total knee replacement: is an additional pre-operative injection effective? J Bone Joint Surg Br 2012;94(7):932. 40. Levine MN, Raskob G, Beyth RJ, et al. Hemorrhagic complications of anticoagulant treatment: the seventh ACCP conference on antithrombotic and thrombolytic therapy. Chest 2004;126(3 Suppl):287S. 41. Shaieb MD, Watson BN, Atkinsom RE. Bleeding complications with enoxaparin for deep venous thrombosis prophylaxis. J Arthroplasty 1999;14:432. 42. Ricket AC, Stewart DW, Wood RC, et al. Comparison of postoperative bleeding in total hip and knee arthroplasty patients receiving rivaroxaban or enoxaparin. Ann Phyarmcother 2016;50:270. 43. Trkulja V, Kolundzic R. Rivaroxaban vs dabigatran for thromboprophylaxis after joint-replacement surgery: exploratory indirect comparison based on metaanalysis of pivotal clinical trials. Croat Med J 2010;51(2):113. 44. Galat DD, McGovern SC, Hanssen AD, et al. Early return to surgery for evacuation of a postoperative hematoma after primary total knee arthroplasty. J Bone Joint Surg Am 2008;90(11):2331.

6

J.-W. Wang et al. / The Journal of Arthroplasty xxx (2016) 1e6

45. Patel VP, Walsh M, Sehgal B, et al. Factors associated with prolonged wound drainage after primary total hip and knee arthroplasty. J Bone Joint Surg Am 2007;89(1):33. 46. Yen SH, Lin PC, Kuo FC, et al. Thromboprophylaxis after minimally invasive total knee arthroplasty: a comparison of rivaroxaban and enoxaparin. Biomed J 2014;37:199. 47. Charters MA, Firsch NB, Wessel NM, et al. Rivaroxaban versus enoxaparin for venous thromboembolism prophylaxis after hip and knee arthroplasty. J Arthroplasty 2015;30:1277.

 A, Fuzilleau F, Dumser D, et al. Efficacy of tranexamic acid on blood loss 48. Clave after primary cementless total hip replacement with rivaroxaban thromboprophylaxis: a case-control study in 70 patients. Orthop Traumatol Surg Res 2012;98(5):484. 49. Wood AM, Smith R, Keenan A, et al. Using a combination of tranexamic acid and rivaroxaban in total knee replacements reduces transfusion requirements: a prospective cohort study. J Arthrosc Joint Surg 2014;1(2):76.