The management of chronic antithrombotic therapy using antiplatelet agents, vitamin K antagonists, or direct oral anticoagulants in the perioperative setting is a common problem in total knee arthroplasty1,2. Chronic antithrombotic therapy is essential for some patients with cardiovascular and cerebrovascular diseases3. However, balancing the risk of excessive bleeding due to continued treatment and thrombotic risk when discontinued remains an important concern1.
A systematic review and meta-analysis supported continuing antiplatelet agents in surgical procedures with low bleeding risk4. However, the bleeding risk due to continuing antiplatelet agents remains controversial during the perioperative period for total knee arthroplasty5-7. There has been a paucity of studies investigating the impact of continuing vitamin K antagonists and direct oral anticoagulants. Bridging therapy with heparin had been recommended for patients undergoing a surgical procedure1,8. However, in some types of surgical procedures, a strategy of continuing anticoagulants was associated with preferable postoperative results9-11.
A strategy of continuing chronic antithrombotic therapy has been used for patients undergoing total knee arthroplasty in our department. This current study was performed to assess the impact of continuing chronic antithrombotic therapy on perioperative total blood loss and bleeding complications in total knee arthroplasty.
Materials and Methods
The study was performed at the Adult Knee Reconstruction Division of a single general hospital (Hokusuikai Kinen Hospital). The study protocol and publication were approved by the local ethics committee.
We reviewed patients recruited by 1 treating surgeon to undergo total knee arthroplasty between January 2015 and December 2017 in our department. Antithrombotic agents, including antiplatelet agents, vitamin K antagonists, and direct oral anticoagulants, were continued perioperatively in all patients receiving such agents during the study period.
Patients undergoing primary total knee arthroplasty were included in this study. The exclusion criteria were patients undergoing staged bilateral total knee arthroplasty with an interval of <6 months between the knees for a staged bilateral surgical procedure, patients undergoing a single-anesthetic bilateral total knee arthroplasty, and patients undergoing single-anesthetic total knee arthroplasty and total hip arthroplasty.
The 201 included patients were classified into 2 groups: (1) patients receiving chronic antithrombotic therapy with antithrombotic agents, including antiplatelet agents, vitamin K antagonists, and/or direct oral anticoagulants, referred to as the continuing antithrombotic therapy group (n = 32); and (2) patients not taking such agents, referred to as the no antithrombotic therapy group (n = 169).
Chronic antithrombotic therapy with antithrombotic agents, including antiplatelet agents, vitamin K antagonists, and direct oral anticoagulants, was continued during the perioperative period, including the day of the surgical procedure.
Antibiotic prophylaxis with 1 g of Cefamezin (cefazolin; Astellas) was administered intravenously perioperatively.
For both the continuing antithrombotic therapy group and the no antithrombotic therapy group, 1 g of Transamin (tranexamic acid; Daiichi-Sankyo) was administered intravenously just prior to skin incision.
We performed the intraoperative periarticular injection with a solution consisting of 300 mg of ropivacaine, 8 mg of morphine, 40 mg of methylprednisolone, 50 mg of ketoprofen, and 0.3 mg of epinephrine12.
All patients received 4 mg of the oral nonsteroidal anti-inflammatory drug, Lorcam (lornoxicam; Taisho Toyama), 3 times a day.
No thromboprophylaxis was routinely used to prevent venous thromboembolism.
Anesthesia and Surgical Procedure
All patients were managed with general anesthesia by board-certified anesthesiologists. Anesthesia was induced using the short-acting volatile anesthetic, Sevofrane (sevoflurane; Maruishi), and the intravenous anesthetic, Diprivan (propofol; AstraZeneca), and was maintained with sevoflurane and a continuous infusion of the short-acting opioid, Ultiva (remifentanil; Janssen)12. Intravenous fentanyl citrate was used as supplementation when required12. Although a clear target threshold of intraoperative blood pressure was not determined, anesthesiologists controlled intraoperative blood pressure for each patient to balance intraoperative bleeding risk with high blood pressure and ischemic risk with low blood pressure.
All total knee arthroplasties were performed or were supervised by 1 surgeon using a cemented, posterior-stabilized prosthesis. No pneumatic tourniquet was used during the study period. No drain was placed for any of the patients.
Blood Management Strategies
We did not use any predeposited autologous transfusion or intraoperative blood salvage techniques for the patients included in the study. We planned additional allogeneic blood transfusion for patients who had a hemoglobin level of <7.0 g/dL and were asymptomatic or patients who had a hemoglobin level of <10.0 g/dL and had symptoms related to anemia13.
Preoperative and Postoperative Screening for Deep Vein Thrombosis
As routine preoperative laboratory testing, the plasma D-dimer level was measured in all patients scheduled for total knee arthroplasty. Patients with a plasma D-dimer level of >0.5 μg/mL were routinely tested for deep vein thrombosis by skilled clinical laboratory technicians using ultrasonography.
At 1 day after total knee arthroplasty, all patients were screened for the presence of deep vein thrombosis by ultrasonography. At 7 days after the total knee arthroplasty, patients were screened using the clinical model of Wells et al.14, and those with a score of ≥3 were again tested by ultrasonography. These measurements were performed during hospitalization because the current Japanese universal health insurance system allows >7 days of hospitalization for patients undergoing total knee arthroplasty.
The primary outcome was the volume of perioperative blood loss, which was measured using the calculated blood volume and change in hemoglobin from preoperatively to postoperative days 1, 3, and 715,16. The blood volume of each patient was calculated using the formula reported by Nadler et al.16.
Intraoperative blood loss was compared between the continuing antithrombotic therapy group and the no antithrombotic therapy group. The number of patients requiring allogeneic blood transfusion was recorded. Bleeding and thrombotic events up to 3 months after total knee arthroplasty were also investigated with special reference to deep vein thrombosis.
Sample Size Calculation and Statistical Analysis
We used the following parameters to calculate the sample size: an overall 2-sided significance level of 0.05, 80% power, a between-group difference for the change in the mean perioperative calculated blood loss score of 200 mL, and a within-group standard deviation of 287 mL. Using these parameters, we estimated that 33 patients would be needed per group. As the minimum unit of allogeneic transfusion was made from 200 mL of whole blood in Japan, we considered a 200-mL difference to be a reasonable estimate of the minimal clinically important difference with respect to perioperative blood loss. The standard deviation of 287 was based on estimates of variability from our prior work in which the perioperative calculated blood loss was compared between patients with and without antiplatelet agents, in whom total knee arthroplasty was performed under lumbar anesthesia without the use of a pneumatic tourniquet6.
The differences in the mean value and the 95% confidence interval (CI) were analyzed with the unpaired Student t test for comparison of perioperative total blood loss between groups. Although we planned to exclude participants with missing data with regard to the primary outcome from the analysis, the primary outcome data were available for all participants included in this study.
The unpaired Student t test was used to assess continuous variables, and the chi-square test was used to compare categorical variables.
All statistical tests were performed at a 2-sided 5% significance level.
A total of 321 patients were screened for eligibility. We excluded 1 patient who underwent staged bilateral total knee arthroplasty with a 3-month interval, 102 patients who underwent single-anesthetic bilateral total knee arthroplasty, and 17 patients who underwent single-anesthetic total knee arthroplasty and total hip arthroplasty. The remaining 201 patients were included in the study (Fig. 1). Data with regard to the primary outcome were available for all 201 patients, and these patients were followed for >3 months.
Of the 201 patients, 32 (15.9%) received chronic antithrombotic therapy, and 169 did not. Table I summarizes the demographic characteristics of the patients in the 2 groups. The agents administered to the 32 patients receiving chronic antithrombotic therapy are shown in Table II.
TABLE I -
Patient Demographic and Baseline Clinical Characteristics
||Continuing Antithrombotic Therapy (N = 32)
||No Antithrombotic Therapy (N = 169)
||80 ± 6
||74 ± 9
||153 ± 7
||153 ± 9
||60.8 ± 9.5
||61.0 ± 11.9
|Body mass index*(kg/m
||26.0 ± 3.4
||26.1 ± 4.4
| Rheumatoid arthritis
|History of diabetes mellitus‡
||13.1 ± 1.0
||12.9 ± 1.3
|Preoperative knee flexion angle*(deg)
||123 ± 12
||119 ± 17
|Preoperative knee extension angle*(deg)
||−10 ± 5
||−11 ± 8
|Duration of operation*(min)
||100 ± 16
||100 ± 14
|Deep vein thrombosis detected by preoperative screening‡
The values are given as the mean and the standard deviation.
Student t test.
The values are given as the number of patients.
TABLE II -
Medication Used for Antithrombotic Therapy
|1. Antiplatelets (n = 22)
| 1A. Single agent for antiplatelet therapy (n = 18)
| Aspirin (n = 9)
| Cilostazol (n = 4)
| Clopidogrel (n = 2)
| Ethyl icosapentate (n = 2)
| Beraprost (n = 1)
| 1B. Combination antiplatelet therapy (n = 4)
| Aspirin and cilostazol (n = 1)
| Aspirin and ticlopidine hydrochloride (n = 1)
| Cilostazol and clopidogrel (n = 1)
| Clopidogrel and sarpogrelate (n = 1)
|2. Vitamin K antagonist (n = 3)
| Warfarin (n = 3)
|3. Direct oral anticoagulants (n = 6)
| Dabigatran (n = 3)
| Rivaroxaban (n = 2)
| Apixaban (n = 1)
|4. Antiplatelet and direct oral anticoagulant (n = 1)
| Rivaroxaban and cilostazol (n = 1)
The calculated perioperative total blood loss is shown in Table III. There were no significant differences in perioperative total blood loss between the 2 groups.
TABLE III -
Perioperative Total Blood Loss Calculated from Blood Volume and Change in Hemoglobin
||Continuing Antithrombotic Therapy* (N = 32)
||No Antithrombotic Therapy* (N = 169)
||95% CI of the Difference (mL)
|Postoperative day 1
||448 ± 213
||495 ± 345
||−172 to 77
|Postoperative day 3
||841 ± 308
||826 ± 328
||−108 to 139
|Postoperative day 7
||855 ± 313
||861 ± 245
||−122 to 108
The values are given as the mean and the standard deviation in milliliters.
Student t test.
The intraoperative blood loss (and standard deviation) was 178 ± 127 mL in the continuing antithrombotic therapy group and 172 ± 103 mL in the no antithrombotic therapy group. The difference between the groups was not significant (95% CI, −35 to 47 mL; p = 0.78).
No patients in the continuing antithrombotic therapy group and 2 patients (1.2%) in the no antithrombotic therapy group required blood transfusion (p = 1).
Postoperative deep vein thrombosis screening using a pulse-wave Doppler ultrasound system indicated that 1 patient (3.1%) in the continuing antithrombotic therapy group and 17 patients (10.1%) in the no antithrombotic therapy group had deep vein thrombosis (p = 0.32). No symptomatic pulmonary embolism was observed.
With regard to major bleeding complications, 1 patient in the no antithrombotic therapy group had subarachnoid hemorrhage at 4 weeks after the total knee arthroplasty. This was the only bleeding complication in either group.
A strategy of continuing antiplatelet agents, vitamin K antagonists, and direct oral anticoagulants during total knee arthroplasty was not associated with increasing perioperative blood loss in patients receiving chronic antithrombotic therapy. Moreover, the rate of required allogeneic blood transfusion did not differ between patients continuing chronic antithrombotic therapy and those with no antithrombotic therapy.
The amount of perioperative blood loss was not significantly different between patients with continuing chronic antithrombotic therapy and patients not receiving antithrombotic therapy in this study. Although postoperative bleeding is one of the most important concerns in the decision whether to continue chronic antithrombotic therapy1, the amount of blood loss has not been well investigated in total knee arthroplasty. With regard to antiplatelet agents, 2 previous studies measured total blood loss in patients undergoing total knee arthroplasty. These 2 studies included patients taking aspirin and compared perioperative blood loss between patients continuing and discontinuing antiplatelet use. Meier et al. reported that there were no significant differences in perioperative blood loss between 17 patients continuing aspirin and 79 patients discontinuing aspirin5. Tsukada and Wakui reported that there was no significant difference in perioperative blood loss between 31 patients continuing several types of antiplatelet agents and 20 patients discontinuing these agents6. To our knowledge, there have been no studies quantifying total blood loss after total knee arthroplasty in relation to the continued use of anticoagulants, including vitamin K antagonists and direct oral anticoagulants. Our current study suggested that continuing chronic antithrombotic therapy using anticoagulants would not be associated with blood loss in the setting of individually controlled blood pressure during total knee arthroplasty without pneumatic tourniquet use.
This study had several limitations. We retrospectively reviewed consecutive patients who underwent total knee arthroplasty and classified these patients into 2 groups according to whether they had received chronic antithrombotic therapy. The surgeons and anesthesiologists were aware of whether the study subjects had received chronic antithrombotic therapy. A double-blinded randomized controlled study, in which patients with chronic antithrombotic therapy are allocated to continuing the anticoagulation or receiving a placebo, would be preferable to strictly investigate the impact of continued chronic antithrombotic therapy on perioperative blood loss.
Although this current study fulfilled the sample size required to compare perioperative blood loss between the group continuing chronic antithrombotic therapy and the group with no antithrombotic therapy, the number of patients was too small to assess the impact of continuing chronic antithrombotic therapy on the occurrence of thrombotic events. For example, although the results of our study showed a possibility of a decreased risk of deep vein thrombosis in the continuing antithrombotic therapy group (1 [3.1%] of 32 patients in the continuing antithrombotic therapy group and 17 [10.1%] of 169 patients in the no antithrombotic therapy group), the difference was not significant.
Our study patients did not receive thromboprophylaxis. The lengths of hospital stays in our study were longer than those in North America and Europe, which may delay postoperative ambulation. These are possible reasons for the high rate of deep vein thrombosis in our study, in which 18 (9.0%) of 201 patients developed deep vein thrombosis. One study using the National Inpatient Sample of the United States showed that the rate of patients who developed deep vein thrombosis was 0.45% in 201117. We should note that this current single-center study also had other specific features, including surgical procedures without a pneumatic tourniquet and the routine use of intravenous tranexamic acid, that could affect the amount of perioperative blood loss.
In our study, the continuing antithrombotic therapy group included patients who received different types of anticoagulation treatment such as single antiplatelet therapy, dual antiplatelet therapy, single anticoagulant therapy, and combined antiplatelet and anticoagulant therapy. The anticoagulation mechanism differs between types of anticoagulation treatment. Moreover, types of anticoagulation treatment were determined based on the pathologic condition of each patient. Caution is required in interpretation of the results of our study because these heterogeneities can affect the study results.
Our results suggest that the strategy of continuing chronic antithrombotic therapy may be one option for patients who are receiving chronic antithrombotic therapy and are scheduled for total knee arthroplasty. We believe that this strategy would reduce the risk of thrombotic events compared with the interruption of chronic antithrombotic therapy prior to the surgical procedure.
In conclusion, the calculated perioperative blood loss did not significantly increase in the continuing antithrombotic therapy group compared with the no antithrombotic therapy group in the postoperative period after total knee arthroplasty. A strategy of continuation of antiplatelet agents, vitamin K antagonists, and direct oral anticoagulants may be an alternative for patients receiving chronic antithrombotic therapy during total knee arthroplasty.
1. Baron TH, Kamath PS, McBane RD. Management of antithrombotic therapy in patients undergoing invasive procedures. N Engl J Med. 2013 May 30;368(22):2113-24.
2. Spyropoulos AC, Al-Badri A, Sherwood MW, Douketis JD. Periprocedural management of patients receiving a vitamin K antagonist or a direct oral anticoagulant requiring an elective procedure or surgery. J Thromb Haemost. 2016 May;14(5):875-85. Epub 2016 Apr 7.
3. Antithrombotic Trialists’ Collaboration. Collaborative meta-analysis of randomised trials of antiplatelet therapy for prevention of death, myocardial infarction, and stroke in high risk patients. BMJ. 2002 Jan 12;324(7329):71-86.
4. Biondi-Zoccai GG, Lotrionte M, Agostoni P, Abbate A, Fusaro M, Burzotta F, Testa L, Sheiban I, Sangiorgi G. A systematic review and meta-analysis on the hazards of discontinuing or not adhering to aspirin among 50,279 patients at risk for coronary artery disease. Eur Heart J. 2006 Nov;27(22):2667-74. Epub 2006 Oct 19.
5. Meier R, Marthy R, Saely CH, Kuster MS, Giesinger K, Rickli H. Comparison of preoperative continuation and discontinuation of aspirin in patients undergoing total hip or knee arthroplasty. Eur J Orthop Surg Traumatol. 2016 Dec;26(8):921-8. Epub 2016 Sep 9.
6. Tsukada S, Wakui M. Continuing versus discontinuing antiplatelet drugs, vasodilators, and/or cerebral ameliorators on perioperative total blood loss in total knee arthroplasty without pneumatic tourniquet. Arthroplast Today. 2017 Aug 10;4(1):89-93.
7. Jacob AK, Hurley SP, Loughran SM, Wetsch TM, Trousdale RT. Continuing clopidogrel during elective total hip and knee arthroplasty: assessment of bleeding risk and adverse outcomes. J Arthroplasty. 2014 Feb;29(2):325-8. Epub 2013 Jul 12.
8. Keeling D, Tait RC, Watson H; British Committee of Standards for Haematology. Peri-operative management of anticoagulation and antiplatelet therapy. Br J Haematol. 2016 Nov;175(4):602-13. Epub 2016 Oct 7.
9. Birnie DH, Healey JS, Wells GA, Verma A, Tang AS, Krahn AD, Simpson CS, Ayala-Paredes F, Coutu B, Leiria TL, Essebag V; BRUISE CONTROL Investigators. Pacemaker or defibrillator surgery without interruption of anticoagulation. N Engl J Med. 2013 May 30;368(22):2084-93. Epub 2013 May 9.
10. Khadim MF, Bell PR, Rashid A, Lewis HG. A postal survey of UK practice on discontinuation of anticoagulant/antithrombotics therapy before minor cutaneous surgery of the head and neck. J Plast Reconstr Aesthet Surg. 2011 Aug;64(8):e213-5. Epub 2011 Apr 22.
11. Edmunds I, Avakian Z. Hand surgery on anticoagulated patients: a prospective study of 121 operations. Hand Surg. 2010;15(2):109-13.
12. Tsukada S, Kurosaka K, Maeda T, Iida A, Nishino M, Hirasawa N. Early stage periarticular injection during total knee arthroplasty may provide a better postoperative pain relief than late-stage periarticular injection: a randomized-controlled trial. Knee Surg Sports Traumatol Arthrosc. 2018 Sep 20. [Epub ahead of print].
13. Iseki T, Tsukada S, Wakui M, Yoshiya S. Intravenous tranexamic acid only versus combined intravenous and intra-articular tranexamic acid for perioperative blood loss in patients undergoing total knee arthroplasty. Eur J Orthop Surg Traumatol. 2018 Oct;28(7):1397-402. Epub 2018 Apr 24.
14. Wells PS, Anderson DR, Rodger M, Forgie M, Kearon C, Dreyer J, Kovacs G, Mitchell M, Lewandowski B, Kovacs MJ. Evaluation of D-dimer in the diagnosis of suspected deep-vein thrombosis. N Engl J Med. 2003 Sep 25;349(13):1227-35.
15. Tsukada S, Wakui M. Combined intravenous and intra-articular tranexamic acid in simultaneous bilateral total knee arthroplasty without tourniquet use. JBJS Open Access. 2017 Apr 18;2(2):e0002.
16. Nadler SB, Hidalgo JH, Bloch T. Prediction of blood volume in normal human adults. Surgery. 1962 Feb;51(2):224-32.
17. Dua A, Desai SS, Lee CJ, Heller JA. National trends in deep vein thrombosis following total knee and total hip replacement in the United States. Ann Vasc Surg. 2017 Jan;38:310-4. Epub 2016 Aug 12.