This study was a single-center, prospective, randomized, controlled trial comparing Carbetocin and Oxytocin in prevention of PPH after elective cesarean section. A total of 852 pregnant women with one or more PPH risk factors scheduled for an elective cesarean section were recruited between April 2017 and August 2019. This trial was registered with Chinese Clinical Trial Registry (ChiCTR1800015613), and conducted in accordance with Declaration of Helsinki and Good Clinical Practice Guidelines. The protocol had been approved by ethics committee institutional review board of participating center (K2016062), and written informed consent was obtained from each participant.

Women with one or more PPH risk factors, scheduled for cesarean section on term pregnancy were eligible to enter this study. The indications for elective cesarean section and PPH risk factors included scarred uterus (the uterus of a woman that had undergone myomectomy or previous cesarean section), uterine fibroid, breech position, and age $\ge$35 years. The exclusion criteria included age 18 years, multiple pregnancy, placenta praevia, suspected placenta accreta, systematic disease (such as liver dysfunction, kidney dysfunction, heart disease or hypertension need to be treated, endocrine disease except diabetes during pregnancy), abnormal coagulation, or hypersensitivity to carbetocin or oxytocin.

All eligible subjects were randomly assigned to carbetocin group or oxytocin group by a computer-generated coding system (Fig. 1). All women underwent low transverse cesarean section under spinal anesthesia, which performed by an obstetrician team including senior physicians and anesthesiologists led by a department chief to ensure the quality and safety. All doctors in our hospital had undergone strict training and assessment which were similar in the level of expertise. The anesthetic drugs included butorphanol (1 mg, intraspinal injection; H20143106, Jiangsu Hengrui Pharmaceutical Co., Ltd., Lianyungang, China), esmolol (1 g, intravenous injection; H19991058, Qilu Pharmaceutical Co., Ltd., Jinan, China), ropivacaine (20 mg, intravenous injection; H20100103, Naropin, AstraZeneca Pharmaceuticals Ltd., Shanghai, China), flurbiprofen axetil (50 mg, intravenous injection; H20041508, Beijing Taide Pharmaceutical Co. Ltd., Beijing, China) and granisetron hydrochloride (20 mg, intravenous injection; H10970243, Ningbo Tianheng Pharmaceutical Co., Ltd., Ningbo, China). In addition, cefoxitin sodium (2.0 g, intravenous drip; H20057973, Yangzijiang Pharmaceutical Group Co., Ltd., Taizhou, China) was used for infection prophylaxis.

Fig. 1.

Study design.

After delivering the newborn, women in the carbetocin group received 100 $\mu$g carbetocin (H20070013, Ferring Pharmaceutical Co. Ltd., St-Prex, Switzerland) via intravenous injection over one minute. Women in the oxytocin group received 10 IU oxytocin (H31020850, Hefeng Pharmaceutical, Shanghai, China) via uterus body injection, followed by 20 IU oxytocin in 500 mL 5% glucose solution (H11020060, China Resources Zizhu Pharmaceutical Co., Ltd., Beijing, China) via intravenous drip over 1 hour. Uterine atony was evaluated by manual palpation. Decisions relating to administration of additional uterotonics followed the hospital’s protocol. According to the hospital’s protocol, women who had postpartum bleeding more than 800 mL were predicted to develop PPH. Thus, the management of the prevention of PPH was the use of additional uterotonics. The additional uterotonics included oxytocin (H31020850, Hefeng Pharmaceutical, Shanghai, China), misoprostol (H20000668, China Resources Zizhu Pharmaceutical Co., Ltd., Beijing, China), hemabate (H20170146, Pharmacia and Upjohn Co., New York, USA) and ergometrine (H32024525, Chengdu Beite Pharmaceutical Co., Ltd., Chengdu, China).

The baseline characteristics, additional uterotonics usage, blood loss during surgery, PPH, antepartum hemoglobin and postpartum hemoglobin, maternal and neonatal outcomes of patients were recorded and analyzed. Blood loss was calculated by measuring the volume in the suction bottle and the absorption in the surgical drapes, gauzes, and pads.

The primary efficacy endpoint was the proportion of women requiring additional uterotonics after administration of the investigative drug (carbetocin or oxytocin). The primary efficacy endpoint was chosen instead of the occurrence of PPH for the following reasons: firstly, the amount of bleeding was known to be underestimated [13]; secondly, continuous and slow bleeding from the surgical incision contributed to the incidence and severity of PPH [14]; lastly, the administration of several additional uterotonics may be used to successfully avoid bleeding in some cases. Secondary efficacy endpoints included the following: (1) the amount of intrapartum blood loss; (2) the incidence of blood loss $\ge$800 mL intrapartum; (3) blood loss within 2 hours after delivery; (4) blood loss within 24 hours after delivery; (5) the incidence of blood loss $\ge$1000 mL postpartum; (6) the incidence of blood loss 1000 mL within 2 hours and $\ge$1000 mL within 24 hours postpartum; (7) postpartum hemoglobin; (8) rate of hemostatics, defined as the proportion of women requiring hemostatics after delivery; (9) rate of blood transfusion, defined as the proportion of women requiring blood transfusion; (10) rate of additional surgical interventions, defined as the proportion of women requiring additional surgical interventions; (11) rate of uterine massage, defined as the proportion of women requiring additional uterine massage.

The safety was evaluated by the maternal and neonatal outcomes. The organ damage (heart, lung, brain, liver or kidney), need for respirator support and hemofiltration or plasmapheresis were used to evaluate the maternal outcomes. Meanwhile, Apgar score and neonatal intensive care unit (NICU) admission were selected to evaluate the neonatal outcomes.

The sample size calculation for this trial was based on primary endpoint according to the parameter of previous study [15]. The investigator anticipated a better performance of carbetocin on the efficacy at a significant level of $\alpha$ = 0.05 and $\beta$ = 0.2, therefore the test validity was 80%. Accordingly, 374 women per group were sufficient with a level of significance of 95% ($\alpha$ = 0.05) and a power of 80% ($\beta$ = 0.2). Considering drop-out rate of 10%, a total of 832 samples were needed at least. Therefore, the investigator planned to enroll 416 cases in each group.

Efficacy analysis was performed based on the intention-to-treat population. All results were analyzed in full analysis set (FAS) and per-protocol set (PPS). SPSS version 22.0 (SPSS Institute, Chicago, IL, USA) was used to perform the statistical analyses. Quantitative data were expressed as means $\pm$ standard deviations (means $\pm$ SD) and were compared using independent samples t test. Qualitative data were expressed as number and percentage, and were evaluated using $\chi$${}^2$ test. Statistical significance was set at p 0.05.

852 eligible pregnant women were randomly assigned to receive carbetocin (n = 442) or oxytocin (n = 410). A total of 841 women were included in FAS analysis (carbetocin group: n = 440; oxytocin group: n = 401). The reasons for not including in FAS analysis was a need for emergency surgery during night shift (n = 11). 827 patients were included in the PPS analysis (carbetocin group: n = 439; oxytocin group: n = 388). 14 patients who were treated with oxytocin mistakenly due to PPH were excluded from PPS analysis (Fig. 2). Baseline characteristics of patients were shown in Table 1. The two groups were generally well balanced for the maternal age, body mass index (BMI), gestational age, parity, antepartum hemoglobin, antepartum hematocrit, pregnancy complications, birth weight and PPH risk factors.

Fig. 2.

Patients flowchart.

The primary and secondary efficacy endpoints were analyzed in FAS and PPS. When the conclusions of the two were inconsistent, the FAS analysis results were mainly considered.

The primary efficacy endpoint of the use of additional uterotonics (FAS analysis) occurred in 81 (18.4%) cases in the carbetocin group, as compared with 98 (24.4%) in the oxytocin group (Table 2). Compared with oxytocin, carbetocin showed the superiority in terms of the primary efficacy endpoint (FAS analysis, p = 0.03, odds ratio = 0.7, 95% CI 0.5–0.9).

The amount of intrapartum blood loss was 370.3 $\pm$ 177.4 mL in carbetocin group and 386.6$\pm$191.6 mL in oxytocin group (FAS analysis), which was no statistically significant difference (p $>$ 0.05). Either the amount of blood loss within 2 or 24 hours after delivery, the incidence of blood loss $\ge$800 mL intrapartum, the incidence of blood loss $\ge$1000 mL postpartum, or the incidence of blood loss 1000 mL within 2 hours and $\ge$1000 mL within 24 hours postpartum were identical between two groups (Table 2, all p $>$ 0.05 in FAS and PPS analysis). In addition, there was no difference between the two groups in the postpartum hemoglobin, rate of hemostatics, blood transfusion, additional surgical interventions or uterine massage (Table 2, all p $>$ 0.05 in FAS and PPS analysis).

The maternal and neonatal outcomes were listed in Table 3. Regarding maternal outcomes, no poor outcomes such as organ damage (heart, lung, brain, liver or kidney), need for respirator support and hemofiltration or plasmapheresis were observed in carbetocin and oxytocin groups. For neonatal outcomes, mild asphyxia (Apgar score 4–7) was occurred in 9 (2.1%) cases in the carbetocin group, as compared with 5 (1.3%) cases in the oxytocin group. No other poor neonatal outcomes such as severe asphyxia (Apgar score 1–3) or NICU admission were observed in two groups.