Efficacy and safety of carbetocin applied as an intravenous bolus compared to as a short-infusion for caesarean section: study protocol for a randomised controlled trial

This is a prospective, single-centre, randomised, double-dummy, investigator-initiated, non-inferiority trial conducted at the University Hospital Basel, Switzerland. Study recruitment commenced in December 2014 and is expected to last for 16 months. The study was approved by the regional Ethics Committee (Ethikkommission Nordwestschweiz (EKNZ), EKNZ 2014–088, 13 March 2014). The study was registered in a national and an international trial registry (www.​kofam.​ch SNCTP000001197 and clinicaltrials.gov NCT02221531).

Randomisation for a 1:1 allocation of the study treatment is provided through an online-randomiser to ensure allocation concealment. This online-randomiser was integrated into the password-protected database. The randomisation took place immediately before the CS after having controlled the inclusion and exclusion criteria. The randomisation list was generated in Intercooled Stata Version 13.1 (StataCorp, College Station, TX, USA) using variable block sizes between 2 and 8. The list is safely stored in a separate and sealed file, which will only be opened if unblinding becomes necessary. This is considered to be unlikely since the concomitant treatment in case of insufficient uterine tone is independent of the mode of administration of carbetocin.

Subjects are randomised on arrival in the operating theatre to minimise the risk of dropouts after randomisation. To achieve similar numbers, randomisation is stratified for planned or unplanned CS (Fig. 1).

Two members of the study team (ES and SDK) ensure correct administration of the study drug, transfer of the haemodynamic parameters, and collect intraoperative data, neither of whom is involved in the patients’ care. The study nurse (ES) always prepares the study drug. Correspondingly, patients, all care providers (obstetric and anaesthesia team) as well as the statistician (SDK) will remain blinded to group allocation until the data are analysed.

Participants are asked about any side effect including (serious) adverse events immediately and 2 days postoperatively. All these events are recorded and monitored until the patient has completely recovered or until no causal relationship to the study drug administration has been established. All serious adverse events will be reported to the Ethics Committee and the sponsor within the appropriate time frame.

The trial will be terminated early in the event of relevant safety concerns that threaten the study participants in any way. Therefore, we will monitor the incidence of PPH in a blinded manner and stop the study early for safety concerns if the incidence rate should rise above the average value of about 5 % PPH in CS [20].

The aim of this trial is to reach greater haemodynamic stability using the slower administration method (primary safety endpoint) while ensuring non-inferiority with regard to the uterine tone (primary efficacy endpoint) and ultimately the risk of elevated blood loss. Since superiority in haemodynamic endpoints can be achieved with fewer patients, we decided to choose as our primary endpoint non-inferiority in uterine tone.

We consider the choice of the appropriate measurement of our primary endpoint as a main challenge in this trial. The choice of our primary outcome is based on a randomised controlled trial performed by Butwick et al. [11] who also used a LAS ranging from 0 to 10 to assess uterine tone in a similar setting. We expanded the scale from 0 to 100 to avoid the fact that some outcome assessors used decimal places (e.g. 7.5) while others did not. Moreover, manual palpation of the uterus represents the standard method to assess uterine tone in daily clinical practice. To reduce bias, especially when using a subjective primary outcome measurement, we have established and will maintain blinding by using a double-dummy technique for study drug administration in patients and all care providers until the statistical analysis has been performed. We have chosen to abstain from a binary assessment of the uterine tone as it would be too parallel to our secondary outcome variable ‘any additional uterotonics – yes/no’. The objective efficacy outcome of interest would have been intraoperative blood loss. Blood loss during PPH is often estimated visually, but visual assessment has been shown to be imprecise, usually underestimating blood loss particularly concerning larger volumes [21‐24]. We have, therefore, decided to choose the uterine tone as our primary outcome variable. We consider blood loss as a secondary endpoint, calculated by taking into account the pre- and postoperative haematocrit and the estimated blood volume of the parturient [15].

We chose haemodynamic stability, defined as lowest mean arterial pressure and highest heart rate within the first 5 minutes after the administration of the carbetocin. After evaluating the options for continuous invasive and non-invasive blood pressure measurements, we have decided to measure blood pressure non-invasively using a cuff on the upper arm. We have considered the accuracy and the precision of the continuous non-invasive methods insufficient to justify their use in the context of this trial. Although the risk for complications of an intra-arterial catheter is low [25], the potentially severe side effects do not justify invasive blood pressure monitoring.

This sample size allows showing a superiority of the short-infusion over the bolus administration in mean arterial blood pressure by at least 5 mmHg with 80 % power and a type-1 error of 5 % assuming a SD of 10. Thus, even a clinically irrelevant difference in efficacy could be found with our data. However, we need to acknowledge that we are not able to show a minor difference in the incidence of serious drops in blood pressure with our sample size.

This trial will include patients undergoing a planned and an unplanned CS. Patients undergoing a unplanned CS often have already received oxytocin to induce or enhance labour, thus, showing a different response to the intraoperative dosage of carbetocin. Therefore, we are using a stratified randomisation for planned and unplanned CS. Excluding these patients would limit the generalisability of our study results.

Poor recruitment is the most frequently reported cause for discontinuation of a clinical trial [26]. However, due to good interdisciplinary cooperation, adequate personal and financial resources and a high acceptance for clinical research, we are able to exceed the target recruitment rate (Fig. 2). Additionally, an amendment was handed in shortly after the beginning of the recruitment to incorporate patients with a breech position fetus. Within the same amendment, we have also translated the study information and the consent into English to be able to address more patients.

Haemodynamic stability and adequate uterine contractility are important outcomes in CS. The results of this trial may be used to optimise these factors and thereby increase safety in parturients due to reduced cardiovascular side effects.