This study was reviewed and approved by ethics committee (full name of the committee: China Ethics Committee of Registering Clinical Trials; reference No. ChiECRCT-20190007). Study protocol was registered in the Chinese Clinical Trial Registry prior to patient enrollment (Registration no. ChiCTR1900021281, Principal investigator: Dr. Jieshu Zhou, Date of registration: 11 February 2019, weblink for the online registration: http://www.chictr.org.cn/showproj.aspx?proj=35234). Informed consent was obtained from all patients before enrollment. This manuscript adheres to the applicable CONSORT guidelines (figure 1). The authors assert that all procedures contributing to this work comply with the ethical standards of the relevant national and institutional committees on human experimentation and with the Helsinki Declaration of 1975, as revised in 2008.
Parturients with twin pregnancies were screened according to following inclusion criteria: American Society of Anesthesiologists physical status I-III, gestational age between 32 and 38+6 weeks, going to receive cesarean section with spinal anesthesia, and estimated body weight ≥ 1500g for both fetuses. Parturients with cardiovascular disease, hypertensive disorder, onset of labor, contraindications to spinal anesthesia or known severe fetal abnormality would be excluded. Patients who underwent cesarean section due to non-reassuring fetal status or received general anesthesia for the cesarean section would also be excluded.
Upon entering the operating room, patients were positioned on the operating table in supine position tilted to the left for 15° to avoid possible aortocaval compression as per our department protocol. Patients were also positioned in this left tilt position during the cesarean procedure. Standard monitoring was applied, including 5-lead electrocardiography, non-invasive blood pressure, and pulse oximetry. Blood pressure and heart rate were measured and recorded at 1-min interval till the difference between 3 consecutive measurements was less than 10%. The mean values of these 3 measurements were recorded as baseline values.
Two 18-G intravenous catheters were inserted into forearm veins while no fluid preload was given. Then, patients were positioned in left lateral position while spinal anesthesia was performed. 2.5ml of 0.5% plain isobaric bupivacaine were injected intrathecally at L3-L4 or L2-L3 intervertebral space via a 25-G pencil-point spinal needle. Intrathecal administration of opioids was spared in this study.
Sensory blockade level was assessed by pinprick and surgery would not begin unless the sensory block reached T6 level. Patients with inadequate sensory blockade would receive general anesthesia for cesarean delivery, who would then be excluded from this study.
Infusion of study drug was initiated, when spinal injection started. According to a computer-generated random number table, sequentially numbered opaque sealed envelopes were prepared by one of the investigators (Z.C.). The envelop was opened before spinal anesthesia. Following the instruction of the envelop, the syringe was prepared by J.Z. and labelled as “study drug”, which containing either 8 µg/ml norepinephrine or 100 µg/ml phenylephrine. The infusion rate was 24 ml/hour for both vasopressors, corresponding to 3.2 µg/min for norepinephrine and 40 µg/min for phenylephrine. Phenylephrine infusion at 40 µg/min for is our standard protocol. The infusion rate of norepinephrine was then determined based on the estimated potency of 11.3 [11] to 13.1:1 [12] for norepinephrine: phenylephrine. Study drug was flushed into the blood stream by carrier lactated Ringer’s solution infused at the rate of 1 ml · kg−1 · hour−1. Infusion of study drug was stopped when the second twin was delivered, and the study was then terminated. Both the patients and the anesthesiologists were blinded for the patients’ assignments. Ringer’s solution co-loading via another peripheral venous catheter was initiated as infusion of the study drug began. The infusion bags were suspended about 1.5 meter above the parturient’s right atrium with the clamp fully opened till the study ended. Then, the fluid infusion rate was adjusted at the discretion of the anesthesiologist.
Blood pressure was measured and recorded every minute till delivery of the second twin. Hypotension, defined as the systolic blood pressure (SBP) < 90mmHg or 80% of the baseline SBP, was rescued by 1-ml bolus of the study drug, i.e. norepinephrine 8 µg for parturients with norepinephrine infusion and phenylephrine 100 µg for parturients with phenylephrine. These boluses were prepared beforehand in a 10-ml syringe labelled as “rescuing drug”. Hypertension, defined as SBP > 140mmHg or 120% of the baseline, was managed by temporarily stopping study drug infusion, which would be restarted as soon as SBP was within the normal range. Heart rate was monitored continuously and recorded at 1-min intervals. Bradycardia was defined as heart rate < 60 beats/min and 0.5 mg atropine would be given when heart rate further decreased below 50 beats/min. Oxygen 5L/min was given by a facemask during the study period.
Apgar scores were assessed by midwives 1, 5 and 10 minutes after delivery, who were unaware of the patients’ assignment. Umbilical arterial (UA) and venous blood (UV) were sampled from a double-clamped segment of umbilical cord by using Pulset arterial 3 cc blood gas syringes (Westmed, AZ85706, USA). Within 20 minutes after clamping, umbilical blood gas was analyzed with the bedside epoc® Blood Analysis System (Siemens Healthcare GmbH, Erlangen, Germany) in the operating theatre.
The primary outcome was the change in heart rate and blood pressure during the study period, from intrathecal injection to delivery of the second twin, which was calculated as the standard area under the curve of heart rate (please find below, in the section of statistical analysis). The secondary outcomes included the incidences of hypotension, hypertension, bradycardia, nausea and vomiting, umbilical blood gas status, Apgar scores, both maternal and neonatal length of post-operative stays, and rates of maternal and neonatal intensive care unit (NICU) admission.
Statistical analysis
Continuous data such as pH values were assessed for normal distribution using Kolmogonov-Smirnov test and were then compared using the Student’s t-test or Mann-Whitney U test, as appropriate. Categorical data such as the incidence of bradycardia were analyzed using chi-square test. A survival analysis with log-rank test was used to compare the requirement of physician interventions for correcting maternal hemodynamic abnormalities (including injection of rescuing bolus of vasopressors or atropine and temporary cessation of vasopressor infusion). The area under the curves (AUC) for SBP and heart rate plotted against time were calculated using the trapezium rule. As the surgical time varied among parturients, a variable number for heart rate and blood pressure measurements were recorded for each parturient. Each AUC value was divided by the number of data points recorded to give the standardized AUC for each patient, which were compared between two vasopressors using the Student’s t-test. Analysis was performed with Prism (version 8.4.3, GraphPad Software, San Diego, California USA). A p value < 0.05 was considered statistically significant.
The sample size was calculated based on our unpublished preliminary data, in which parturients with twin pregnancy received norepinephrine or phenylephrine infusion following spinal anesthesia. The means of standardized AUC for heart rate were 79 ± 12 and 72 ± 8 beats/min for norepinephrine and phenylephrine, respectively. A minimal sample size of 46 per group was required to detect a significant difference (10% of the baseline heart rate) between two groups, with 90% power and a two-sided type I error of 0.05. Considering the potential withdraw of consent or loss of follow-up, a final sample size of 50 per group was determined.