Ethics and patients
After obtaining approval by the ethical review board of the Women’s Hospital, School of Medicine, Zhejiang University (approval number 20180095), and the registration in a Chinese Clinical Trial Registry (ChiCTR) (registration number ChiCTR1800018988; principle investigator: Fu Feng, M.D.; date of registration: October 20, 2018), we conducted a randomized, double-blinded dose-response study. Written informed consent was obtained from all subjects participating in the trial. This study adheres to CONSORT guideline.
Eighty parturients, American Society of Anesthesiologists physical status (ASA ≤ 2), with a singleton pregnancy scheduled for elective cesarean delivery were recruited. Patients were excluded if they had pre-existing or pregnancy-induced hypertension, diabetes mellitus, known cardiovascular or cerebrovascular disease, fetal abnormality, or contraindication to spinal anesthesia. The patients, whose body mass index > 30 kg/m2, height < 150 cm or > 170 cm, gestational age < 36 weeks were also excluded from the study.
Patients were randomly assigned into 4 groups based on computer-generated random number sheet. The patient in different group received one of four different fixed-rate prophylactic methoxamine infusion regimens as follows: 1μg/kg/min (group M1), 2μg/kg/min (group M2), 3μg/kg/min (group M3), or 4μg/kg/min (group M4). The range of infusion doses was based on the drug instruction, the consensus of Chinese experts in clinical anesthesiology and the previously published descriptions [11-13]. To maintain blinding, the solutions of methoxamine were prepared in identical 50-mL syringes by a co-investigator (Qian J P), who was not involved in data collection or clinical care of the patients. With the infusion rate predetermined to be set to 50 ml/h, the appropriate concentration of methoxamine was prepared by diluting to a 50 ml mixture with saline for each group. Then the milligram amounts of methoxamine were shown as follow: group M1, weight (kg) × 0.06; group M2, weight (kg) × 0.12; group M3, weight (kg) × 0.18; group M4, weight (kg) × 0.24.
Patients were subjected to 8 h food deprivation and 2 h water deprivation before surgery, without preoperative treatment. After arrival in the operating room, patients were allowed to rest for several minutes before measurements were taken. Standard noninvasive monitoring was applied, including non-invasive BP, pulse oximetry, and electrocardiography. The baseline systolic blood pressure (SBP) and maternal heart rate (HR) were determined from the mean of three readings that fell within 10% of each other, taken at least one minute apart. An IV line was established with an 18-gauge IV cannula in the forearm, and an infusion of lactated Ringer’s (LR) solution was started at a minimal rate to keep the vein open.
Combined spinal-epidural anesthesia was performed with patients in the left lateral position. Epidural anesthesia was performed at the L1-2; 5ml of saline was injected into the epidural space before insertion of the epidural catheter; while spinal anesthesia (midline puncture) was performed at L3-L4 or L4-L5.We injected the mixed intrathecal solution (1.5 ml of 1% hyperbaric Ropivacaine + 1.5 ml of 10% dextrose) into the subarachnoid space at the rate of 1 ml per 10 s. The 15mg dose of hyperbaric ropivacaine was chosen because it is close to the ED95 for cesarean section in Chinese parturients reported by Chen X Z et al. Immediately after the injection of the intrathecal medication, infusion of study drug of methoxamine was started at dosages of 1, 2, 3, and 4μg·kg−1·min−1 (50ml/h) respectively. Simultaneously, a fluid coload with 15 ml/kg of lactated Ringer’s solution over 20 to 30 min was started. Patients were then positioned supine with left uterine displacement and oxygen 5 L/min was administered via a facemask. The sensory block level of anesthesia, assessed by loss of pin prick discrimination, was recorded 5 and 15 min after induction of spinal anesthesia. The tested upper sensitivity block to T6 was considered adequate for surgery. However, the decision to allow surgery to start was based on clinical judgment of the attending anesthesiologist. In this study, we didn’t use the epidural catheter dose during the study period. If patients received the epidural catheter dose, they should be excluded from the study.
Monitoring and interventions
As per our standard practice, SBP, heart rate, and pulse oximetry were assessed every minute, commencing immediately after intrathecal injection until delivery and subsequently at 5-min intervals. Hypotension was defined as a decrease in SBP less than 80% of baseline value or SBP < 90 mmHg. If 90 mmHg ≤ SBP < 100mmHg, it was treated with rapid infusion of lactated Ringer’s solution; if the SBP was <90 mmHg, it was treated with a bolus of 3mg IV methoxamine as rescue medication. Administration of a rescue bolus was repeated at 1-minute intervals if hypotension persisted. We didn’t stop the treatment until hypotension improves or tends to improve [16, 17]. Reactive hypertension, defined as an increase in SBP to ≥ 20% above the baseline, was treated by stopping the infusion of methoxamine. Infusions were restarted only when the SBP decreased to < 120% of baseline. Bradycardia was defined as a heart rate < 50 beats/min; the bradycardia was managed by stopping the methoxamine infusion if no hypotension was present, and if hypotension was present during bradycardia then atropine 0.5mg was given.
As our standard practice, the study was commenced immediately after intrathecal injection, until delivery of the baby. An effective methoxamine infusion dose was defined by the outcomes that patient did not have hypotension throughout the study period.
Additional data collection included maternal demographics: age, weight, height etc. Time of fetal exposure to anesthesia and to surgical procedures were evaluated as induction-delivery interval (time from intrathecal injection to delivery of the fetus) and uterine incision-delivery interval (time from uterine incision to delivery of the fetus). Episodes of hypotension, hypertension, bradycardia, nausea and vomiting and shivering were recorded. Neonatal Apgar scores were measured at 1min and 5min. A segment of the umbilical cord was collected for assessment of blood gases in the umbilical artery. The timeline of the whole study was shown as Fig 1.
Sample size estimation was determined by the Cochran-Armitage Test using PASS® (Version 11.0.7, NCSS, LLC, Kaysville, UT). According to and results from early preliminary data, the frequency of postspinal hypotension (70%, 50%, 30% and 10% respectively) in patients receiving one of the four infusion regimes same to that in the present study. the minimum effect size was 48patients in total (12 patients per group), which was found to have 90% power to detect a linear trend using a Z test with continuity correction and a significance level of 0.05. Allowing for possible dropouts, the sample size was increased to 80 patients (20 patients per group).
Data were expressed as mean and 95% confidence interval [mean (1.96 SD), 95% CI]. For numerical data, the Kolmogorov-Smirnov method was used to test for normal distribution, followed by one-way analysis of variance for normally distributed data and Kruskal-Wallis test for non-normally distributed data among the groups with post Bonferroni tests for pairwise comparisons. For nominal data, statistical analysis was performed by means of the Cochran-Armitage chi-square test for trend.
Dose–response data were analyzed with probit regression. Dose values were entered as x values, Y was the response as a percentage. The regression coefficient was obtained by regression analysis, then the ED50 and ED95 values were obtained from interpolation of the linear probit regression plot, and generation of the sigmoid dose-response plot was obtained secondarily. Serial changes in systolic blood pressure (SBP) of four groups are shown for the first 12 minutes after spinal anesthesia; they were analyzed using two-way ANOVA (include: different time points in the same group and different groups in the same time point). The correlation between duration of hypotension and 1-minute Apgar score, duration of hypotension and umbilical arterial PaCO2 were analysed using Spearman’s rank correlation.
Analyses were performed using IBM SPSS Statistics for Windows version 22.0(IBM Corp, Armonk, NY) and GraphPad Prism version 5.0 (GraphPad Software Inc., San Diego, CA).All tests were two-tailed and P-value<0.05 was considered statistically significant.