Therapeutic and pharmaco-biological, dose-ranging multicentre trial to determine the optimal dose of TRAnexamic acid to reduce blood loss in haemorrhagic CESarean delivery (TRACES): study protocol for a randomised, double-blind, placebo-controlled trial

Postpartum haemorrhage (PPH) remains the leading cause of maternal mortality worldwide [1]. PPH mostly occurs during caesarean section (CS) due to atony and abnormal placental insertion. Haemorrhagic CS represents more than half of severe PPH. PPH-induced coagulopathy can be potentially major and contributes to the severity of PPH [2]. Coagulopathy has been observed as a major component of maternal morbidity after vaginal delivery due to uterine atony [3, 4], but also in patients experiencing intensive care admission after massive PPH [5, 6]. Early hyper-fibrinolysis can be observed at the onset of PPH and becomes maximal 2 h after the beginning of bleeding when no prohaemostatic treatment has been given [7].

Fibrinogen decrease is a central component of PPH-induced coagulopathy concomitant to D-dimer increase and factor II decrease [3, 4, 7]. Active fibrinogenolysis remains a major component of amniotic fluid embolism, fœtal death or abruptio placentae [8, 9]. Tranexamic acid (TA) (Exacyl® Sanofi, Paris, France) reduces D-dimers and the plasmin-antiplasmin (PAP) complex peak observed at PPH onset [7], as well as in abruptio placentae [10].

TA reduces bleeding and transfusion need in medical and gynaecological haemorrhage, major surgery and trauma, without major side effects [11, 12].

In obstetric settings, available TA therapeutic efficiency and safety data are limited. Eight placebo-controlled, blinded, randomised trials demonstrated that the prophylactic use of TA reduced postoperative bleeding in elective non-haemorrhagic CS or after vaginal delivery [13‐20]. In antepartum anaemic patients, this mild reduction was associated with a decrease in transfusion need [20].

Only one randomised controlled study (EXADELI trial) tested TA efficiency in active PPH following vaginal delivery [21]. A high dose of 4 g was administered, followed by a continuous infusion of 1 g/h. The EXADELI trial demonstrated a mild, but significant, reduction of the bleeding volume (173 mL vs 221 mL in the control group) and a significant reduction of the delay for PPH cessation measured as a blood flow of less than 50 mL per 10 min (Fig. 2) [21]. Although limited, the bleeding volume reduction was associated with a reduction of the percentage of patients developing severe morbidity, such as anaemia or haemorrhagic shock [21]. Mild side effects were observed, including nausea and visual disturbances, but no acute kidney injury nor seizure or thromboembolism were observed. Renal function was similar or even better in the TA group. Deep vein thrombosis complicated catheter insertion in two patients in the control group and three in the TA group [21]. The major point of the EXADELI trial was the strict measurement of PPH bleeding on special graduated bags and the strict respect to the French timed PPH management [22]. Concomitant administration of prohaemostatic treatments was avoided, except for untraceable PPH, enabling the observation of the natural course of PPH-induced coagulopathy. TA clinical efficiency to reduce bleeding in the haemorrhagic caesarean context has not been previously published. In the only (unpublished) Tunisian randomised, placebo-controlled trial in haemorrhagic CS, TA administration induced a reduction of blood loss and transfusion need (p = 0.03). Because the absence of academic literature on CS and the low level of evidence after vaginal delivery, current 2014 French guidelines for PPH management restricted the use of TA to late administration in massive haemorrhage: ‘Tranexamic acid may be of interest in PPH treatment, even if its clinical interest has not been demonstrated in an obstetrics setting. Every clinician is free to use TA. If used, the expert group suggests the administration of a 1-g dose only if PPH resists prostaglandins’ [22]. The WOMAN trial aimed to assess the impact of the early administration of TA on death and hysterectomy in women experiencing PPH [23, 24]. Between 2010 and 2016, 20,060 women with a clinical diagnosis of PPH in 21 low- and middle-income countries were included in this randomised, double-blind, placebo-controlled trial to intravenously receive either 1 g TA (n = 10,051) or placebo (n = 10,009) in addition to standard care. The WOMAN trial found that TA decreased death due to bleeding (155 (1.5%) vs 191 (1.9%), RR 0.81; 95% CI 0.65–1.00; p = 0.045), especially in women given treatment within 3 h of giving birth (89 (1.2%) vs 127 (1.7%), RR 0.69; 95% CI 0.52–0.91; p = 0.008). Laparotomy for bleeding was also reduced (82 (0.8%) vs 127 (1.3%), RR 0.64; 95% CI 0.49–0.85; p = 0.002) after CS and vaginal delivery. The original composite primary endpoint was not reduced (534 (5.3%) vs 546 (5.5%), RR 0.97; 95% CI 0.87–1.09; p = 0.65). Adverse events were not increased in the TA group vs placebo: thromboembolic events (30 (0.3%) vs 34 (0.3%)), seizures (33 (0.3%) vs 43 (0.4%)), and renal failure (129 (1.3%) vs 120 (1.2%)). One could suggest that the TA benefit on mortality may be higher in women with severe PPH, then aim to reserve TA for the more severe situations. However, the data showed that the delayed administration of TA was associated with potential harm. Thus TA, if used, should be given early and empirically as soon as a PPH is diagnosed, and before severe haemorrhage. As the study was based on robust clinical endpoints, neither blood loss estimation nor haemoglobin drop were available. There was no subgroup analysis of TA efficacy according to the volume of bleeding. Laboratory results were not included in the primary analysis but were assessed in a few centres as a part of a secondary analysis [25]. The biological effect of the high TA dose administered in the EXADELI trial was observed and compared to a non-treated haemorrhagic group and a non-haemorrhagic postpartum group. Significant excessive fibrinolysis was noted in the untreated haemorrhagic group vs the non-haemorrhagic group at PPH onset. This excessive fibrinolysis was inhibited in the TA-treated group 30 min and 2 h after TA administration, as demonstrated by the inhibition of the D-dimers and PAP increases (Fig. 3) [7].

Few other cases or case series have previously suggested the efficiency of TA to reduce maternal mortality and morbidity in placental abruption, third trimester placental bleeding or amniotic fluid embolism [10, 26].

The TRACES clinical and pharmacokinetics dose-ranging study objective will be a better understanding of the mechanisms of TA efficacy and the optimal selection of the TA dose.

The minimal effective dose of TA is not known. In major cardiac surgery, the 4-g dose followed by a continuous infusion of 1 g/h is the known efficient therapeutic dosage. However, in trauma, as in major surgery, fibrinolysis is an expected part of coagulopathy. In PPH associated with CS, although fibrinolysis appears early and is constant, its intensity is variable and most of the PPH stops after the first step of management. We hypothesis that a 0.5-g dose is sufficient to inhibit the most common fibrinolytic activity in haemorrhagic CS, while some of the CS-associated fibrinolysis is so intense that neither 0.5 g nor 1 g will be enough the inhibit the coagulopathic process.

The higher dose of 4 g and continuous infusion was not selected because of the minor side effects observed in the EXADELI trial [21]. The TA 2-g dosage inhibited fibrinolysis after the 30th min in the treated patients in the EXADELI trial. The dose of 1 g or 10 mg/kg is commonly used prophylactically before CS, but has not yet been tested as a treatment in an actively bleeding haemorrhagic CS. Because of the lack of data on lower doses and TA pharmacokinetics, a low 0.5-g dose should also be tested.

The aim of the study is to determine, out of two doses (a standard and a low dose) compared to placebo, the optimal and minimal dose of an intravenously administered single bolus of TA to reduce blood loss when administered during haemorrhagic CS.

The originality of the trial can be described as follows: no previous study has been published evaluating the clinical impact of TA on blood loss and transfusion need reduction during haemorrhagic CS. The determination of the optimal dose to obtain this clinical impact is of interest for clinicians in the current obstetric practice. TA concentration and innovative direct plasmin peak assessment by a simultaneous generation thrombin plasmin assay (SGTPA) are expected to provide scientific and objective data to determine the optimal dose. The dose-ranging study with a low dose is expected to describe the minimal dose able to inhibit fibrinolysis and to contribute to bleeding reduction while minimising side effects. The first pharmaco-biological observational results allows the choice of this dose ranging. No previous study has tested the low dose in the obstetric literature in either trauma-induced massive haemorrhage or major surgery.

The TRACES study is a multicentre, randomised, double-blinded, placebo-controlled therapeutic and pharmaco-biological dose-ranging study to evaluate the clinical efficiency of TA on blood-loss reduction in patients experiencing PPH during elective or non-emergency CS (Fig. 4).

The TRACES trial obtained approval from the competent national authorities (ANSM 201500249926) and the Ethics Committee (CPP 15/50 020216) before beginning the study, in accordance with article L1121-4 of the Public Healthcare Code. This trial has been declared on the clinical trials registration under the number CT 02797119. Registration will be performed in accordance with decree dated 14 November 2006 about gathering data in the national register of individuals participating in biomedical research.

The five medical centres are known for their strict multidisciplinary management of PPH, including systematic measurement of blood loss, and for their ability to perform research programmes, such as the EXADELI trial. Anaesthetic and obstetric staff of each centre agreed to their participation to this study and were involved in writing the French guidelines for PPH management [22]. The study will be sponsored by The University and Regional Hospital Centre of Lille, France.

The inclusion criteria for the experimental haemorrhagic group concern each patient experiencing a bleeding volume of more than or equal to 800 mL due to surgery or to uterine atony during an elective or non-emergency CS. This experimental group will receive a low dose or a standard dose of treatment (treatment groups) or placebo (control group). The non-haemorrhagic group, where patients experience a bleeding volume of less than 800 mL, is a reference group for biology and will be sampled to obtain a reference profile for the direct plasmin generation test in CS. All patients receive complete information, give their written consent and are covered by social security.

Non-inclusion criteria are as follows: hypersensitivity to the product or its excipient, previous or ongoing arterial or venous thrombosis, disseminated intravascular coagulopathy, except disseminated intravascular coagulopathy (DIC) associated with a predominant fibrinolytic profile, renal failure, previous seizures, severe HELLP syndrome, emergency CS, administration of TA before inclusion, inherited haemorrhagic diseases and low-molecular-weight-heparin administration within 24 h before inclusion, previous inclusion in an interventional trial for 2 months, or patient being unable to consent.

The coordinating team includes the coordinator, the promotor and the major investigators of each of the centres. A call conference will be organised every 2 months between the coordinator and the centres, supported by a newsletter reporting inclusion rates and protocol information. Data and consent collection will be monitored by the promotor. An independent safety monitoring committee is recruited to observe blinding safety issues and allow trial continuation. Data management and statistical analysis will be done by an independent unit. Plans are written in advance for investigators and sponsor to communicate trial results to participants, healthcare professionals, the public, and other relevant groups (e.g. via publication, reporting in results databases, or other data-sharing arrangements).

TRACES inclusions have already begun. In the first opened centre (promoter centre), 64 patients have been included (57 haemorrhagic and seven non-haemorrhagic). The four other centres are opened and included 13 patients. Three Safety and Monitoring Committee meetings monitored these first data and allowed the continuation of the trial.