Side‐effects of carbetocin to prevent postpartum hemorrhage: A systematic review and meta‐analysis of randomized controlled trials

Abstract Postpartum hemorrhage (PPH) increases the risk of maternal death worldwide. Heat‐stable carbetocin, a long‐acting oxytocin analog, is a newer uterotonic agent. Clinicians do not fully understand its side‐effects, particularly the unanticipated side‐effects. The aim of this study is to investigate the side‐effects of carbetocin to PPH. The Cochrane Library, Web of Science, PubMed, Elsevier ScienceDirect, Embase, and ClinicalTrials.gov were searched from the inception to September 2020. Randomized controlled trials (RCTs) that considered pregnant women who received carbetocin before delivery and provided at least one adverse event were included. Statistical analysis included random or fixed‐effect meta‐analyses using relative risk. Stratified analyses and sensitivity analyses were also performed. Begger's and Egger's test and funnel plots were used to assess the publication bias. Seventeen RCTs involving 32,702 women were included, and all these studies ranked as medium‐ to high‐quality. Twenty‐four side‐effects were reported. The use of carbetocin had a lower risk of vomiting in intravenously (0.53, 0.30 to 0.93) and cesarean birth (0.51, 0.32 to 0.81) women, and had a slightly higher risk of diarrhea (8.00, 1.02 to 62.79) compared with oxytocin intervention. No significant difference was found among other side‐effects. Evidence from our systematic review and meta‐analysis of 17 RCTs suggested that the risk of vomiting decreased with carbetocin use in the prevention of PPH after delivery.


| INTRODUC TI ON
Postpartum hemorrhage (PPH) caused a significant number of maternal deaths worldwide. About 27.1% of all maternal deaths were caused by hemorrhage, and these data can reach 36.9% in most low-income countries and regions. 1 It has already been confirmed that prophylactic administration of uterotonic agents is the most important component in terms of reducing the risk of PPH and preventing the irreversible functional consequences in the stage of labour. 2 Oxytocin, a short half-life uterotonic agent, is recommended by the World Health Organization (WHO) as the first line for the prevention and treatment of PPH in 2012. 3 However, it is sensitive to heat and requires cold storage and transport in usage. The active ingredient and purity are mostly affected in low-resource settings where the cold chain is not commonly available. Because of its heat sensitivity, it does not possess satisfactory real-world efficacy, particularly in hot low-and middle-income countries and regions. 4 Meanwhile, the short half-life required frequently or continuously repeated administration.
Heat-stable carbetocin, a long-acting oxytocin analog, is a newer uterotonic agent. Its effects of uterine contractions can start within two minutes, and the rhythmic contractions can last for 60 to 120 minutes in intravenous and intramuscular injection, respectively. 5,6 What is important is that it has high thermal stability, and it can be transported and stored at normal temperature and even in hot and humid environments without compromising quality. The heat-stability data showed that it maintained for a minimum of 36 months at 30°C and 75% relative humidity and at extreme temperatures, such as 50°C, for three months. 7 Hence, it would be advantageous and even a significant breakthrough for maternal health in hot environments lacking cold chain routes to use carbetocin. Recently, a multicenter clinical trial, including 23 hospitals in 10 countries, indicated that the intramuscular administration of 100 ug of heat-stable carbetocin was noninferior to the administration of 10 IU oxytocin for the prevention of PPH after vaginal birth. 8 Meanwhile, systematic reviews and metaanalysis demonstrated that carbetocin significantly reduced postpartum blood loss, additional uterotonics, and transfusion. 9,10 The use of carbetocin is recommended for the prevention of PPH for all births by WHO, particularly in settings where oxytocin is unavailable or its quality cannot be guaranteed. 2 Side-effects were also an important concern when choosing uterotonic agents. Although carbetocin seems to be an ideal agent compared to other uterotonic agents, some side-effects, such as vomiting, nausea, and dysarteriotony, are still concerned. There are many trials of carbetocin use to prevent PPH, and side-effects are always as secondary outcomes in these trials. Clinicians do not fully understand the side-effects of carbetocin to PPH, particularly the unanticipated ones. There seems to be a gap to detailed presentation of the side-effects of carbetocin. Therefore, this study aims to assess the side-effects of prophylactic carbetocin to PPH among the previous randomized clinical trials.

| MATERIAL S AND ME THODS
This systematic review was pre-registered online in the PROSPERO registry (CRD42019134522). The perform of the current study followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines. The ethical approval was not required for this study.

| Inclusion and exclusion criteria
Studies were considered if the following criteria were met: 1) randomized controlled trials (RCTs) design, 2) pregnant women received the prophylactic carbetocin before delivery, 3) compared carbetocin with oxytocin or placebo interventions, and 4) provided the frequency of at least one side-effect. Cluster-or quasi-RCTs, ongoing trials, cross-sectional studies, case series, abstracts without full text, or studies without sufficient data were excluded.

| Study selection and data extraction
Two authors (Wen Ai and Dazhi Fan) independently identified eligible articles on title and abstract first, and then on the full text. The data extraction was also independently performed by the same authors using a prespecified Excel form, and the extracted variables from each study included study and participant characteristics (first author, year of publication, region, trial registration number, funding source, risk of PPH, and mode of delivery), arms and treatment regimens (dose and route), and the type and frequency of side-effects. Disagreement between authors was solved by discussion between the two authors. Meanwhile, we also entered the data into the EpiDate software to check the accuracy.

| Risk of bias assessment
Using the Cochrane Handbook, 11 two authors (Yanfei Zeng and Yubo Ma) independently assessed the quality of the included studies. Evaluation criteria included the following major domains: randomization, implementation of blind, data reporting, and other bias, such as funding source, and potential conflicts of interest.

| Data analyses
The assessment was the relative risk (RR) with 95% confidence intervals (CIs) comparing the frequency of side-effects between carbetocin and control groups, which was calculated using frequentist pairwise meta-analysis. Forest plots were used to present the results of RR and 95%CIs. Heterogeneity across studies was assessed using Tau 2 , I 2 , and Chi 2 statistics. The selected effect models, fixed or random, were based on the heterogeneity result.
For all side-effects, if they were provided by ten or more trials, we stratified analyses by the route of carbetocin administration (intravenously versus intramuscularly), mode of delivery (vaginal versus cesarean birth), prior risk of PPH (high, low or none), trial register (yes or no), funding source (company, researcher, or none), and control-intervention ways to identify the main sources of heterogeneity between trials. To assess the dose effect, a sensitivity analysis was performed to exclude the trial in which the carbetocin dose was not 100 micrograms. To assess the publication bias, the Begger and Egger tests were used. Data analysis and graphing were conducted using the R software, the Review Manager software, and Microsoft Excel.

| Study identification and characteristics
We first identified 221 potentially eligible articles, and 136 articles were scrutinized for the full text. Ultimately, a total of 17 RCTs 8,12-27 involving 32,702 women were included (Figure 1). Two articles 12  Four articles were funded by pharmaceutical company, four articles were funded by organizations, university, or hospital, and nine articles did not disclose funding sources. (Table 1).

| Risk of bias
The quality of the included articles varied. In each of the domain, most of the articles were rated with low or unclear risk of bias. High risk of bias was mostly attributed to incomplete outcome data and other bias, such as granting by the pharmaceutical company and the failure to declare potential conflicts of interest. Meanwhile, performance and detection bias existed in one article. 15 In general, the quality was good, with five high-quality articles, twelve moderate quality articles, and none of the articles was classified as low quality (Figures 2, 3).

| Quantitative analysis
The use of carbetocin had a slightly higher risk of diarrhea (8.00;

| Sensitivity analysis and publication bias
Sensitivity analysis by excluding a dose of 125 ug carbetocin trial 15 showed that the results were not substantial influenced, and they were just slight changes. Funnel plots and Begger's and Egger's tests found no significant publication bias.

| DISCUSS ION
In this systematic review and meta-analysis, which included 17 RCT studies covering 32,702 women, we found that vomiting, nausea, to high-quality evidence.
Fewer clinical trials have focused on the side-effects of carbetocin to prevent PPH. Mannaerts D et al 16 have compared the adverse effects between carbetocin and oxytocin in their trial. They found the incidence of nausea was lower after carbetocin, but there was no statistical difference. A previous review, only included several studies, also showed that carbetocin is associated with fewer adverse effects. 10 Furthermore, our study demonstrated that the administration of carbetocin in delivery is associated with the lower incidence of vomiting. We therefore suggested that carbetocin might be considered as an appropriate choose for pregnant women with vomiting intolerance for the prevention of PPH.
Studies suggested that carbetocin is superior in terms of additional uterotonics when compared with other uterotonic agent at cesarean delivery. 25

| CON CLUS ION
The overall results of our systematic review and meta-analysis study may raise concerns about the potential side-effects of uterotonic agents use for preventing postpartum hemorrhage. It may help clinicians better understand the side-effects, particularly the unanticipated side-effects, of carbetocin use during labor and delivery. These results provide insights toward optimizing clinical decision-making strategies, which should consider the potential benefits of using carbetocin to prevent PPH in parts of the world where a lack of cold chain transported and stored.

ACK N OWLED G M ENTS
We appreciate the efforts of all the researchers whose articles were included in this study. In addition, this work was not supported by any funding.

CO N FLI C T O F I NTE R E S T
The authors declare that they have no competing interests.

AUTH O R S' CO NTR I B UTI O N S
DF and YZ participated in the design and coordination of the study.
WA conceived the study and drafted the manuscript. YZ, YM, and LL searched for the studies, collected, and analyzed the data. DF participated in the design of this study and edited the manuscript. DF, SW, YZ, and YM did the data management and analyzed the data. All authors read and approved the final manuscript.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data used to support the findings of this study are included within the article.