Study design and setting

Nine clinical sites will participate in this trial. Together, they represent the largest perinatal centres in Switzerland, including all Swiss university hospitals. Sites were involved early on for the development of the protocol. Emphasis was put on designing a study process that is compatible with the national PPH guideline as well as routine processes at each site.

Plasma-derived FXIII (Fibrogammin, CSL Behring) is used in accordance with the indication and dosages described in the summary of products characteristics. Based on the Swiss laws for clinical research in humans, the present trial is a category A trial, meaning a low-risk trial.

Eligibility of patients

Adult, pregnant women, with planned vaginal delivery after 30 weeks of pregnancy and a maternal weight of less than 100 kg (at admission to delivery) are included after they provided written informed consent. The model consent form is provided in online supplemental material. Only vital, singleton pregnancies are considered for preliminary inclusion. The following exclusion criteria apply: antithrombotic therapy (therapeutic dosage) during pregnancy for any indication; pre-eclampsia, eclampsia or haemolysis, elevated liver enzymes, and low platelets (HELLP) syndrome; known history of deep vein thrombosis or pulmonary embolism; known diagnosis of bleeding disorder or thrombophilia; known thrombocytopenia during the second half of pregnancy with thrombocytes<100 G/L; known anaemia during the second half of pregnancy with haemoglobin(Hb)<80 g/L; known sickle cell disease. Additional exclusion criteria are known malignant tumour(s), participation in another interventional study, suspected inability to follow the procedures of the study and non-compliance such as drug or alcohol abuse (figure 1).

• Download figure
• Open in new tab
• Download powerpoint

Figure 1

Eligibility criteria, before and after delivery. Hb, haemoglobin; MBL, measured blood loss; PPH, postpartum haemorrhage; TXA, tranexamic acid.

The final eligibility check comes after delivery because women shall not participate in the trial if the postpartum bleeding remains below 700 mL, if an unplanned caesarean section is to be done or if the woman developed fever≥39°C between admission and delivery. Identifying patients with fever shall exclude women with clinical chorioamnionitis. Information and consent procedures are preferably conducted during routine pregnancy visits at the hospital. However, women who enter the hospital shortly before delivery, but who are not in distress due to labour, receive an abbreviated version of patient information and go through a shortened informed consent process before delivery. Additional information is given after delivery, and women’s willingness to participate is then reconfirmed; if not, data collected may not be used.

Ethical considerations

The study is carried out in accordance with the protocol and with principles enunciated in the current version of the Declaration of Helsinki, the guidelines of Good Clinical Practice issued by International Council for Harmonisatio (ICH), the Swiss Law and Swiss regulatory authority’s requirements. Ethics approval was granted by the central ethics committee (Kantonale Ethikkommission Zürich/Switzerland) on 16 June 2024, reference number: BASEC 2024-00374.

Randomisation

Patients are randomised to receive FXIII in addition to the standard-of-care therapy or standard-of-care therapy only in a ratio of 1:1. Randomisation occurs after delivery if the patient has lost 700 mL of blood and bleeding is continuing despite TXA. Randomisation envelopes are used to allocate the patients to the treatment or control groups. Opening the envelope with the lowest consecutive number available assigns the patient to the respective group.

Study treatments and management of PPH

Patients are closely monitored for bleeding after delivery. If postpartum bleeding passes 300 mL (as described below in detail), a blood collection bag is installed underneath the woman’s pelvis in order to ensure timely and accurate blood loss measurement during the postpartum and study period. If blood loss reaches 500 mL and bleeding continues, all women receive 1 g of TXA, according to the actual standard of care. The woman becomes an actual study participant after the blood loss passes 700 mL with ongoing bleeding. This is the time of randomisation and administration of FXIII (fibrogammin) to the women in the intervention group versus nothing in the control group. Throughout the entire study period, all women receive equivalent obstetric-related treatments irrespective of the assigned group (both control and intervention groups). Obstetric treatments, such as treatment of uterine atony, surgical treatment of trauma-derived bleeding and removal of retained placental tissue, are based on the primary cause of PPH and follow the hospital’s standard-of-care practice.

The women in the intervention group receive either 1250 International Units (IU) of FXIII (body weight of less than 80 kg) or 1500 IU of FXIII (body weight from 80 to 99.9 kg), based on the women’s weight before delivery. FXIII is administered intravenously according to the manufacturer’s instructions. If the woman is assigned to the control group, there will be no treatment in addition to obstetric routine practice for the treatment of PPH (figure 2).

• Download figure
• Open in new tab
• Download powerpoint

Figure 2

Trial flow. Women who signed the informed consent are monitored for blood loss right after delivery. After blood loss of 300 mL and ongoing bleeding: blood collection bag for quantitative measurement of blood loss is installed. After blood loss of 500 mL and ongoing bleeding: 1 g of tranexamic acid (TXA) is administered. After blood loss of 700 mL and ongoing bleeding: randomisation with allocation to study or control group. Volume and coagulation management follows a standardised procedure until measured blood loss (MBL) of 1500 mL (or 1100 mL for women<60 kg); after that threshold, coagulation management is released to local practice. Blood withdrawals (BWs) before, during and 48 hours after postpartum haemorrhage (PPH) treatment are applicable to a subgroup of patients only. FXIII, factor XIII; IU, International Units; iv, intravenous

For the benefit of standardising the clinical procedures while guaranteeing the patient’s safety at all times, the participating sites are asked to administer no additional coagulation factors (such as fibrinogen, FXIII, rFVIIa), blood products (such as fresh frozen plasma, red blood cell or thrombocyte concentrates) or a second round of TXA. Regarding volume management, no colloids shall be administered, and administration of crystalloids should be restricted to 2000 mL until the blood loss has reached 1500 mL (or 1100 mL in women with a body weight of less than 60 kg). The restrictions can be bypassed at any time by a so-called safety net, which allows deviation from the restrictions above, if a laboratory parameter or point-of-care testing indicates a significant lack of one of the measured parameters, bleeding continues and the patient life is deemed in imminent danger. The safety net values are defined as fibrinogen<2.0 g/L, FXIII<60%, platelets<50 G/LHb<70 g/L or suspect point-of-care testing (ROTEM: MCF A10Fibtem≤ 6 mm; or TEG functional fibrinogen:≤10 mm).

Once the measured blood loss (MBL) reaches 1500 mL (or 1100 mL if the woman is below 60 kg), the coagulation and volume management are released to standard local practice.

Primary outcome: MBL (in mL, within 24 hours after delivery)

Accurate and continuous blood loss measurement is a prerequisite for the analysis of the primary outcome; therefore, the SWIFT sites use a previously validated quantitative technique of real-time blood loss measurement with a proven correlation between blood loss and postpartum Hb reduction.24 The participating sites receive standardised blood collection bags as well as working instructions (instructions in written form and a video tutorial, please see video in the online supplemental file 2) for the purpose of achieving the maximum possible accuracy, comparability and consistency in measuring blood loss.

Blood loss measurement starts immediately after delivery by placing a fresh absorbent pad underneath the woman’s pelvis. After the delivery of the baby, any additional vaginal fluid is blood and not amniotic fluid, which is absorbed by the pad. The pad is regularly checked, and if continued bleeding is observed, the pad is weighed on a neonatal scale in the delivery suite. If the weighed blood reaches 300 mL, a standardised blood collection bag with a quantitative scale is installed under the woman’s pelvis24 (figure 3).

• Download figure
• Open in new tab
• Download powerpoint

Figure 3

Blood collection bag with a quantitative scale for blood loss measurement in vaginal deliveries.24

Blood loss is continuously observed, quantified and protocolled. If the bag must be removed before the end of PPH (ie, for transfer to the operation room), the additional blood loss is documented either by visual estimation, weighing or other methods (such as collected blood from suction tubes in the operation room). Visual estimations (such as blood on the floor, in bedsheets) shall always be recorded with a range of minimum and maximum of observed blood loss, taking into account the uncertainty of mere blood loss observations without a measuring tool (above all in situations when blood is spilled on the floor, soaked bed linen, etc). Using a range helps provide a more realistic assessment of blood loss, reflecting the impossibility to determine exact values in these situations. Total postpartum blood loss at the end of acute PPH is used for the analysis of the primary endpoint and can be documented, consequently, as a range. This helps correct for the level of uncertainty during final statistical analysis by means of interval-censored outcome values and is definitely more realistic than a supposedly exact specification of a specific blood loss quantity. Secondary PPH, that is, increased blood loss needing medical care later than 24 hours after delivery, is recorded as a separate event and will not enter the primary outcome calculations.

Secondary outcomes

Secondary outcomes will be compared between the intervention group and control group; time point of variable assessment will be 48 (range 36–60) hours postpartum if not stated otherwise. Composite of adverse, PPH-related maternal include MBL≥2000 mL (within 24 hours), admission to intensive care unit, blood transfusion, need for embolisation of the pelvic arteries, laparotomy with surgical measures such as compression sutures or ligatures, or hysterectomy during hospitalisation. Furthermore, changes in haematological standard values from before delivery to 48 hours after delivery such as Hb, leucocytes and thrombocyte count are assessed. (Blood samples for prepartum analysis are taken at admission for delivery, and blood samples for postpartum analysis are taken 48 hours (range 30–60) after delivery.) In specific centres, the association of antepartum and postpartum laboratory parameters with PPH and the dynamics of coagulation parameters during the course of PPH and their effect on MBL is analysed.

Analysis of differential treatment effects by means of a data-driven approach for the identification of relevant subgroups is performed.25

The total medical hospital costs per patient, taken from the hospital’s internal billing system, from time of admission until hospital discharge at the end of the hospital stay, will provide a first, simplified impression on the costs generated by FXIII treatment and potential savings related to fewer and/or milder adverse outcomes.

Patient and public involvement

Given the increasing importance of patients and public involvement in research, we involved the service of the Swiss Patient Organization for the review of the patient information sheet. PPH and the management thereof often lead to stressful situations, and the woman’s feeling could be disregarded while providing emergency treatment affecting her psychosocial and emotional well-being. Recent publications point to the risk of post-traumatic stress disorder and depression in women who had a primary PPH.26 27 Therefore, a subgroup of patients in the early phase of the trial is asked for voluntary patient survey by providing feedback on the atmosphere in the delivery room during treatment of PPH in the main study centre. A questionnaire, designed in collaboration with the psychology department of the University of Zurich, asks for the woman’s personal sensations such as safety, anxiety, distress or insecurity. The survey is distributed and collected at discharge from the hospital, in general, 3–5 days after delivery. Once the feedback of the subgroup has been analysed, potentially negative behavioural patterns of the staff can be addressed to increase the well-being of future patients in identical situations.

Patient follow-up

6–9 weeks after the delivery, women are contacted in order to assess the occurrence of adverse events, particularly thromboembolic events. Also, the number of women who breastfeed their babies after PPH will be recorded since PPH and its related adverse outcomes may affect the women’s ability and/or willingness to breastfeed their newborn child.28

Statistical considerations

For the primary analysis, the blood loss distributions of the intervention and control group are compared by means of the proportional OR in a proportional odds model allowing continuous and interval-censored MBLs. The model corresponds to a series of binary logistic regression models for all possible cut-off points in MBL featuring a common OR treatment effect parameter. A corresponding estimate, along with a two-sided 95% CI and two-sided p value against the null hypothesis OR=1, is obtained from the Continuous Outcome Logistic Regression (Colr) model.29 30 Interval censoring and right-censoring in the primary endpoint are explicitly allowed and handled by the corresponding likelihood in the Colr model.

The null hypothesis of no treatment effect (OR=1) of replenishment of coagulation factor XIII in women with PPH (blood loss≥700 mL) on MBL is formally rejected when the p value is <0.05.

Determination of sample size

Previous research quantified the effect of coagulation factor XIII by a partial OR of 1.011 (95% CI 1.006 to 1.015).21 This means that increasing factor XIII by one unit increases the odds of losing less than a specific amount of blood by 1.1% for constant values of prepartum Hb, prepartum fibrinogen and prepartum factor II. The marginal OR (in a model only conditioning on coagulation factor XIII, but not on prepartum Hb, prepartum fibrinogen and prepartum factor II) is almost identical to the partial OR (1.01094 vs 1.01091). It is assumed that replenishment of coagulation factor XIII leads to an increase of 30 units, which is equivalent to an expected treatment effect of OR=1.01091³⁰ = 1.38491.

Sample size was calculated on the subset of the data,21 which suggested that the primary analysis aiming at the demonstration of a positive treatment effect (with type I error of 5% and type II error of 20%) would reasonably require 2×494 women with PPH≥700 mL to be randomised in this trial. At this sample size, the power to detect a significant difference in the occurrence of major PPH (defined as binary event MBL>1000 mL) would be about 0.64; this explains why, in the case of a cut-off variable as the primary outcome, the required number of patients would be significantly higher compared with the continuous outcome logistic regression model without achieving any additional scientific benefit.

Current trial status (updated 27 March 2025)

More than 800 patients have already been screened, and 147 patients have been randomised since the official start of recruitment on 9 July 2024 in Zurich.

Ethics and dissemination

The findings of this study will be published open-access in appropriate medical journals, as well as presented at national and international conferences. The data-sharing plan follows the Findable, Accessible, Interoperable and Reusable (FAIR) principles31 for scientific data and is provided as online supplemental material.

Data collection, maintenance and management are supported by the use of electronic case report forms, which are generated by members of the research team with support from the Data Management Department of the Clinical Trials Centre, University Hospital Zurich, Switzerland (CTC Zurich). Appropriate coded identification is used to enter participant data into the database. The Electronic Data Capture software REDCap (www.project--‐redcap.org) is used for data processing and management. We use a risk-based monitoring approach based on the categorisation as a low-risk trial.