Investigating the Effect of Intravenous and Oral Tranexamic Acid on Blood Loss After Primary Hip and Knee Arthroplasty

Although numerous studies have demonstrated efficacy in using tranexamic acid (TXA) in primary hip and knee arthroplasty, it has not yet become a recognised standard of care in the United Kingdom (UK). In part, this is because trials to date have excluded many of the patients who could most benefit from a reduced peri-operative blood loss. This trial aims to maximize the potential of TXA by continuing administration post-operatively, at the time of greatest loss. In doing so, the aim is to establish a standardised protocol that can be adopted by other units. Also, despite any published evidence and the excellent safety profile of TXA, there have been concerns about a potential increased risk of venous thrombo-embolism (VTE) which in itself has been a contentious issue in the UK. The exclusion criteria have been adapted in this trial to address these safety concerns.This trial will contribute to establishing the effect of TXA on perioperative blood loss.

Reducing blood loss after hip and knee arthroplasty helps patients avoid anaemia and allogenic blood transfusion with their incumbent risks. TXA is effective intravenously and topically at reducing perioperative bleeding in various surgeries, the evidence in joint arthroplasty is compelling with calculated blood losses reduced by 30% and the need for a transfusion by 50%. However, it remains unclear as to whether extending a dosing regime of TXA beyond the immediate perioperative period would lead to further reductions in blood loss. Oral TXA would be a cheaper and less labour intensive mode of delivery post-operatively than either intravenous bolus or intravenous infusion regimes.

Patients undergoing joint arthroplasty are often elderly with co-morbidities; this frailty makes them particularly vulnerable to the adverse effects of major blood loss. Reducing bleeding in this group is likely to be particularly beneficial to patient outcomes.The majority of blood loss occurs in the first 24 hours post-operatively. A dosing regimen of intravenous TXA eight hourly for twenty-four hours has been suggested. Replacing this with an oral dose would be cheaper, less labour intensive for nursing staff and less disruptive for patients. One gram of oral TXA given 8 hourly for four doses following arthroplasty is consistent with prescribing guidance in the Summary of Product Characteristics (SPC) for local fibrinolysis.There is no evidence in the context of joint replacements to favour either 8 hourly or 6 hourly dosing. Both are licensed regimens as per SPC and therefore an 8 hourly regime is considered less intrusive to the patient.

Intravenous (IV) TXA has an elimination half-life of 3 hours (almost completely excreted unchanged in urine). Five hours after a dose of 10mg/kg IV bolus, the serum level is 5µg/L and it is thought that a level of 10µg/L provides 80% inhibition of fibrinolysis and is considered clinically effective. Oral TXA has a peak level 2 hours following ingestion. By administering the dose in the recovery ward 2 hours following knife to skin (KTS), this will give its peak effect just over 4 hours after the IV dose is administered in both hip and knee arthroplasty.

The aim is to minimise sub therapeutic TXA levels at the time of the highest bleeding risk. Nilsson (1987) noted that after total hip arthroplasty (THA), blood loss from drains was lower following a single 10mg/kg bolus intra-operatively in the first 2 hours but that the drain outputs then equilibrated between the TXA group and the control group after that for the following 10 hours. This supports the current hypothesis that administering further doses of TXA early post-operatively could further reduce blood loss. Also it will be easier to ensure accurate time of the first dose administration in recovery rather than on the ward due to greater staff availability.

This study initially used two tranexamic acid intervention groups (Groups 1 \& 2) and one no treatment comparator group (Control Group 3). Although the benefits of using TXA in elective hip and knee arthroplasty to reduce blood loss has been shown, it has not become standard of care in the UK and hence the initial need for the Control Group 3. Following an interim analysis, it was recommended to stop randomisation to Control Group 3.

In addition, this study will incrementally contribute to the growing data relating to use of TXA in patients with a history of VTE. However, to definitively determine what impact TXA has on VTE in the peri-operative period would require a large appropriately powered multi-centre study.

Study Aim and Objectives

This is a Phase IV, single centred, open label, parallel group, randomised controlled trial involving primary hip and knee arthroplasty patients.

The overall aim of this study is to reduce blood loss in THA and Total Knee Arthroplasty (TKA) patients using TXA. While TXA has been shown to reduce blood loss in THA/TKA patients, the optimum method of delivery, optimum number of doses and optimum period of use is not clear. A previous audit of patients under the care of the Chief Investigator demonstrated that 5/6 of the blood loss occurs post wound closure. Therefore, hypothetically if TXA is given to a patient for over 24 hours following surgery it should further reduce blood loss as compared to TXA dosing in the immediate peri-operative period.

-Primary Objective: The primary objective is to determine if the use of oral TXA post-operatively for up to 24 hours will confer a reduction in the calculated blood loss at 48 hours beyond an intra-operative intravenous bolus alone for patients undergoing unilateral primary total hip or knee replacement.

-Secondary Objective: The secondary objective is to determine if the addition of oral TXA post-operatively to an intra-operative intravenous bolus of TXA produces any change in other measurable parameters as compared to those observed either with an intra-operative intravenous bolus alone or no TXA for patients undergoing unilateral primary THA/TKA.

While this study is not powered to answer questions such as whether the use of TXA can lower complications, it is important that complications are reported on. If it was found that using TXA increased the number of complications, this would be a clinically important outcome which should be reported.

Eligible participants were initially allocated to Intervention Group 1 (IV TXA peri-operatively plus 1g oral TXA every 8hrs for up to 24hrs), Intervention Group 2 (IV TXA peri-operatively) or Control Group 3 (standard care) with an allocation ratio of 2:2:1.

Patients with a renal impairment receive a reduced dose dependent on pre-operative serum creatinine.

Following the interim analysis it was recommended to stop randomisation to Control Group 3. Eligible participants are allocated to Intervention Group 1 or Intervention Group 2 with an allocation ratio of 1:1.

Anaesthetists, surgeons, other theatre, recovery and ward staff will not be blinded to the treatment, nor will the study investigators or the patient themselves.

However, the number of patients falling below the transfusion trigger will be blinded.The transfusion trigger will be set as per standardised protocol after consent and prior to randomisation. Therefore, the decision about whether or not to transfuse will not be blinded but the transfusion trigger will be blinded.

Staff in an off-site laboratory who are processing the blood samples will be blinded to the treatment Group allocation.

The TRAC-24 trial initially required a minimum of 1166 patients to be recruited. Recruitment was due to stop when 583 participants had been recruited for both the TKA and THA patients. However, following the interim analysis, randomisation of patients to Control Group 3 was stopped. Recruitment will now continue until a minimum of 932 patients, comprising of 466 TKA and 466 THA, is reached.