The Use of Intravenous Tranexamic Acid in Patients Undergoing Total Hip or Knee Arthroplasty: A Retrospective Analysis at a Single Military Institution

Objective: To describe a single institution ’ s experience after initiation of a protocol in which all primary total hip arthroplasty (THA) and total knee arthroplasty (TKA) patients were administered intravenous tranexamic acid (TXA) intraoperatively to decrease perioperative blood loss. Methods: A retrospective review of medical records at a single institution from February 2012 to April 2014. The TXA treatment group was compared to a control group. We reviewed intraoperative blood loss, preoperative hemoglobin (Hb) levels, postoperative day 0 to 2 Hb levels, transfusion rates, postoperative venous thromboembolism, and other complication rates. Results: 259 patients underwent either TKA (165) or THA (94). 121 received perioperative intravenous TXA and 138 did not. There was a statistically decreased rate of allogeneic blood transfusion (0 vs. 10, p = 0.003) as well as a higher postoperative day 2 Hb level (10.8 ± 1.1 vs. 10.2 ± 2.6 g/dL, p = 0.02) in the treatment group. There was no statistical difference in any variable measured in the THA group, though there was a trend toward higher postoperative Hb levels at all-time points measured. Conclusion: Intravenous TXA is a safe and effective drug to decrease perioperative blood loss and allogeneic transfusion in THA and TKA. There was no increased risk of venous thromboembolism or other complications in our review.


INTRODUCTION
The use of tranexamic acid (TXA) has been previously described in the total hip arthroplasty (THA) and total knee arthroplasty (TKA) literature as a safe and effective method to decrease perioperative blood loss and the subsequent need for transfusion. TXA is an antifibrinolytic agent, and is a synthetic analog of lysine. It functions by inhibiting lysine-binding sites on plasminogen, halting the transition of plasminogen to plasmin, effectively inhibiting fibrin clot dissolution. This aids hemostasis, perioperative blood loss, and potentially decreases the need for blood products. There are multiple benefits to avoiding transfusion to include a decreased risk of infection, transfusion reactions, disease transmission, and cost associated with allogeneic blood transfusion. The purpose of this study is to describe a single institution's experience before and after initiation of a protocol in which all primary THA and TKA patients were administered intravenous TXA intraoperatively to decrease perioperative blood loss.

MATERIALS AND METHODS
After institutional review board approval, we retrospectively reviewed all primary TKA and THA procedures performed at our institution from February 2012 to April 2014. During this time period, all fellowship trained adult reconstruction surgeons at our institution implemented the intraoperative use of intravenous TXA. There were 259 patients who underwent either primary TKA (165) or THA (94) and met inclusion criteria. 49 (52.0%) of the THA and 72 (43.6%) of the TKA patients received perioperative intravenous TXA and 138 patients received no perioperative treatment and served as controls. We reviewed each patient's chart to determine demographics, intraoperative blood loss, preoperative hemoglobin (Hb) levels, postoperative day 0 to 2 Hb levels, transfusion rates, postoperative venous thromboembolism, and other complication rates.
One gram of TXA was administered before the skin incision and tourniquet inflation and an additional 1 g was administered during wound closure. All patients underwent primary, unilateral, cementless THA, or cemented TKA procedures. The principal diagnosis was primary osteoarthritis in both groups. A lateral approach was performed in 12 and a posterior approach in 37 of the THA patients in treatment group although a lateral approach was performed in 19 and a posterior approach used in 27 of the control THA patients. All TKA patients received a standard, medial parapatellar approach. The majority of patients underwent a combination of general and regional anesthesia. Postoperative drains were not routinely placed. All patients received Aspirin, lovenox, Coumadin, or rivaroxaban as pharmacologic prophylaxis at the discretion of the treating surgeon. All patients wore sequential compression devices for mechanical deep vein thrombosis (DVT) prophylaxis. Patients received allogeneic blood transfusion if they had a Hb level <7.0 g/dL or were clinically symptomatic according to the treating surgeon. Both DVT and pulmonary embolism (PE) diagnoses were based on ultrasound and CT findings, respectively, if there was clinical suspicion. Table I shows the demographic  information for all TKA patients and Table II shows  treatment and control groups. Data were summarized as number (%) of patients or mean ± SD, as appropriate. Statistical analysis was performed with Student's t-tests for all continuous variables and with Fisher's exact test with mid-P correction for all categorical data. Statistical significance was defined as P < 0.05. The mean ages in the both arms of the TKA group were 59.3 ± 9.9 and 58.0 ± 10.8 ( p = 0.43). There was a greater percentage of males (51.0%) in the TKA treatment group as well as a higher proportion of males in the control group (60.0%), though this was not statistically significant. The body mass index (BMI) between the two groups was 31.8 ± 5.0 and 31.4 ± 5.8 kg/m 2 ( p = 0.61). The American Society of Anesthesiologists (ASA) classification score was roughly proportional and did not reach statistical significance ( p = 0.82).
The mean ages in the both arms of the THA group were 54.8 ± 10.3 and 53.5 ± 13.3 ( p = 0.61). Both groups had more male than female patients, though this difference was not significant ( p = 0.08). The BMI between the two groups was 29.3 ± 4.7 and 29.5 ± 5.9 kg/m 2 ( p = 0.85). There was no significant difference in ASA classification ( p = 0.44). There were 37 and 27 patients who underwent posterior hip approaches in the treatment and control groups, respectively compared to 12 and 19, who underwent lateral approaches in the treatment and control groups, respectively ( p = 0.09).
Preoperative Hb values for TKA and THA patients were obtained not more than 1 month before the procedure. Postoperative Hb values were obtained within 4 hours postoperatively in the postanesthesia recovery room and on postoperative days 1 and 2. Estimated blood loss (EBL) was agreed on by the anesthesiologist and treating surgeon at the conclusion of the each surgery. EBL, transfusion record, and postoperative complications were evaluated in a chart review.

RESULTS
In the TKA treatment group, there was a statistically decreased rate of allogeneic blood transfusion (0 vs. 10, p = 0.003) as well as a higher postoperative day 2 Hb level (10.8 ± 1.1 vs. 10.2 ± 2.6 g/dL, p = 0.02) compared to controls (Table III). In addition, there was a trend toward equal or higher postoperative Hb at all time points with the TXA treatment group. The EBL was higher in the TXA treatment group, though this difference was not significant ( p = 0.09). Both of these patient population's demographics were evenly matched with no significant differences found in either group.
In the THA group, there was no statistical difference in any variable measured between treatment and control groups (Table IV). The mean and postoperative Hb levels were higher in the treatment group at all time points, though these levels did not reach statistical significance. There was also a trend toward increased EBL in the TXA treatment group 479.7 vs. 414.1 mL ( p = 0.13), though this was not statistically significant.
There was no increased risk of venous thromboembolism compared to controls in either TKA or THA treatment groups. There was 0 DVT and 1 PE diagnosed in the TKA treatment group compared to 1 DVT and 2 PEs in the control group ( p = 0.56, p = 0.77). There was 1 DVT and 1 PE in the THA treatment group compared to 0 DVT and 1 PE

DISCUSSION
The results of this study showed that our TKA patients treated with TXA had a statistically decreased rate of allogeneic blood transfusion, a higher postoperative day 2 Hb level, and a trend toward increased postoperative Hb compared to controls. Likewise, the THA treatment group showed a trend toward higher postoperative Hb levels at all time periods measured, though this difference did not reach statistical significance. There was no statistical difference in EBL in either the TKA or THA groups, nor was there a higher risk of thromboembolisms in TKA or THA treated with TXA.
There have been numerous studies that have investigated the use of both intravenous and topical TXA use during TKA procedures, with most finding a statistical difference in postoperative bleeding and higher postoperative Hb/hematocrit levels when using TXA compared to placebo. [1][2][3][4][5][6][7] Other studies have also showed a decreased incidence and/or amount of blood transfused with TXA vs. placebo, 1,3,5,7,8 though these findings have not been universal. 9 In addition, several meta-analyses have affirmed that the use of TXA in TKA reduced mean blood loss and rates of allogeneic blood transfusion [10][11][12][13][14][15][16] while finding no significant difference in rate of venous thromboembolism.
In addition, there has been considerable investigation into the effectiveness and safety of TXA in THA procedures. As in the TKA literature, TXA has also showed a similar benefit toward decreasing perioperative blood loss, transfusion rates, and transfusion quantity when given both intravenously or topically. [16][17][18][19][20][21][22][23][24][25] One of the strengths and unique aspects of this study was the fact that our patients were generally healthy and younger than most arthroplasty patients reported in the literature. The average age for our THA patient was 54.1 years and our average TKA patient was 58.7 years. In a meta-analysis of   26 found an average age of 65.9 and 65.8 years in the treatment and control groups, respectively. A weakness of this study was its retrospective design. In addition, intraoperative blood loss measurements are often simply an estimate by the operating staff, not as objective as Hb levels measured postoperatively. Our study had relatively few patients in each arm and may not be large enough to detect significant differences in intraoperative blood loss or rare postoperative complications such as PE and DVT. On the whole, however, our data mirrors and further contributes to that of previous studies, which show that TXA limits perioperative bleeding following TKA and THA procedures without increased risk of venous thromboembolism or other major complications. There was no increased risk associated with its use and we will continue its use in our operative protocol.

CONCLUSION
Intravenous TXA is a safe and effective drug to decrease postoperative blood loss in the TKA and THA procedures and may reduce the need for allogeneic blood transfusion. We found no increased risk of venous thromboembolism or other complications in our retrospective review. Our data further contributes to the safety and efficacy of TXA use in TKA and THA.