This was a prospective, double-blind, randomised, BCUD-based, sequential dose-finding study was approved by China Ethics Committee of Registering Clinical Trials, West Hospital, Sichuan University, Sichuan, China on 29 September 2020 (Chairperson Taixiang Wu, Ethical No.: ChiECRCT20200301) and registered at http://www.chictr.org.cn (ChiCTR2000040375; 28 Nov 2020). A total of 85 women, aged 22–40 years, and with a full-term (> 37 weeks of gestation) singleton pregnancy, the American Society of Anaesthesiologists (ASA) physical status II, and undergoing an elective cesarean delivery were recruited during Jan 2021 and April 2021 at the International Peace Maternity and Child Health Hospital. After providing signed informed consent, the women were allocated by a research assistant to one of two groups, using a random number generator in SPSS for Windows version 18.0 (SPSS Inc., Chicago, IL, USA), specifically, the ropivacaine (R) and ropivacaine with intravenous S-ketamine (RS) groups. All patients were blinded to group allocation.
Exclusion criteria were as follows: contraindications to combined spinal and epidural (CSE) anesthesia; the ASA score of III–IV; allergy to ketamine, S-ketamine, or ropivacaine; history of gastroesophageal reflux, full stomach, placenta previa, placental abruption, pre-existing or pregnancy-induced hypertension, preeclampsia, cardiovascular or cerebrovascular disease; foetal distress or abnormalities; multiple gestations; refusal to participate, and emergent cesarean delivery. Due to differences in delivery time and uterine contraction pain levels after surgery, women who had previously undergone a cesarean section and multiparas were excluded.
Patients were recruited for the study on the day before surgery in the obstetrical wards, when they were asked to complete the Edinburgh Postnatal Depression Scale (EPDS), which is a useful screening instrument for detection of women at risk for depression in the perinatal period.14 All patients were instructed to fast from midnight on the day of surgery. After entering the operating room, an IV line was established with an 18G IV catheter in the left forearm, and lactated Ringer’s solution was infused at a rate of 1 ml.min− 1 to maintain the vein open.
Routine monitoring was performed continuously, including electrocardiography, non-invasive blood pressure, and pulse oximetry. Systolic blood pressure (SBP), heart rate (HR), and pulse oximetry were assessed every 2 min. The first two resting SBP and HR measurements obtained in the supine position were recorded, and their average values were treated as the baseline values. If the baseline SBP was of > 140 mmHg, the patient was excluded from the study due to hypertension.
The CSE puncture was performed routinely at the level of L3-4 with the patient in the right lateral decubitus position. A 17G Tuohy needle was used to perform the epidural puncture with a paramedian approach. After identifying an entrance into the epidural space, a 27G Whitacre needle was inserted through the epidural needle. Once the cerebrospinal fluid was detected, a research isobaric ropivacaine dose of 0.75% was injected through the Whitacre needle. After injecting the ropivacaine, an epidural catheter was inserted into the epidural space by advancing it 3 cm through the Tuohy needle. The patient was immediately moved to the supine position with the left uterine displacement created by placing a wedge under the right hip.
In addition, an infusion of 1 ml.kg− 1.h− 1 lactated Ringer’s solution was administered until delivery. A rescue 5-ml bolus of 2% lidocaine was administered through the epidural catheter 2 min before the surgery for patients that could still feel the cold sensation of ice being placed on the skin below the level of T6 and those that complained of pain and whose VAS score was of > 2 points during the surgery. A 5-ml IV dose of 0.15 mg.kg− 1 S-ketamine with normal saline or 5 ml normal saline solution was administered with a 5-ml syringe 1 min before the surgery, depending on group assignment. Hypotension was defined as SBP of < 80% of the baseline value. A rescue dose of 50 µg phenylephrine was administered whenever hypotension was detected. Bradycardia was defined as HR of < 50 beats per minute (bpm) and treated with 0.5 mg atropine. The level of sedation was assessed using a 5-point scale (1 = agitated; 2 = alert; 3 = calm; 4 = drowsy; 5 = asleep).15
Immediately after delivery, 1 ml of umbilical vein blood was collected by the obstetrician, and blood gas assessments were performed, using a blood gas analyser (iSTAT1 Analyzer MN:300-G, Abbott Point of Care Inc., USA) with an iSTAT CG7 + test cartridge. A patient-controlled intravenous analgesia (PCIA) pump (AM3300, ACE MEDICAL EQUIPMENT INC., Korea) was connected to the patient, and a 2-mL PCIA bolus was given at the end of the surgery; meanwhile, the patient was taught how to use the PCIA pump. The PCIA fluid consisted of 10 mg hydromorphone and 90 ml normal saline at the total volume of 100 ml. A 2-ml background infusion and a 2-ml patient-controlled analgesia (PCA) bolus was set to a 15-min lockout time. The patient was followed-up at bedside for the post-operative analgesic effect and recovery of lower limb mobility 24 and 48 h after surgery. The EPDS was administered at discharge.
In both RS and R groups, a research spinal dose of ropivacaine in a 5-ml syringe was prepared and given to the patient through the Whitacre needle by a specialist anaesthesiologist, who knew the exact administered dose. The remaining anaesthesiologists and all patients were blinded to the administered doses.
The spinal ropivacaine dose of 12 mg was used for the first patient in both groups, based on findings from Tang et al.16 and Mei et al.17. The dose administered to each subsequent patient was determined by the response of the immediately preceding patient. Satisfactory anaesthetic effect was defined as the upper sensory level of spinal anesthesia not below T6 before surgery and the VAS score of ≤ 2 points before delivery. Provided a satisfactory anaesthetic effect was achieved, the spinal dose of ropivacaine was considered suitable, and the next patient was 1/9-times as likely to receive a lower dose (decreased by 1.5 mg) or 8/9-times as likely to receive the same dose as the previous patient. When satisfactory anaesthetic effect was not achieved, the following patient received a dose that was increased by 1.5 mg. Minimum and maximum doses of spinal ropivacaine were 9 mg and 18.5 mg, respectively. The dose assignment was prepared by a statistician, using the BCUD function in Microsoft Excel 2016; only the specialist anaesthesiologist had access to this information, ensuring the double-blind nature of this study.
The primary outcomes were satisfactory anaesthetic effect until delivery and the incidence of hypotension. Secondary outcomes included maternal hypotension-related symptoms and their management, upper sensory level, total co-load IV fluid volume, blood loss volume, level of sedation during surgery, recovery of lower limb mobility, analgesic effect 24 and 48 h after surgery, and EPDS scores at discharge; neonatal outcomes included induction-delivery interval, umbilical vein blood gas values (including pH, pO2, pCO2, base excess (BE), SO2, and lactate levels), neonatal weight, Apgar scores measured at 1 min and 5 min post-delivery, and the incidence of neonatal intensive care unit admission. Maternal demographic characteristics such as age, weight, height, gestational weeks, gravida, para, baseline SBP, baseline HR, and EPDS scores at admission were recorded. All primary and secondary outcomes were observed by anaesthesiologists blinded to patient group and dose assignments.
Statistical analysis and sample size calculation
This dose finding study was based on BCUD design and simulation studies that suggest that a sample size of 20–40 patients may provide stable estimates of the target dose for most realistic cases.18 The present study included 40 patients per group.
The ED90 was defined as the spinal dose of ropivacaine associated with a 90% anaesthetic success rate and was estimated, using the isotonic regression method18. The corresponding 95% CI was obtained, using the bias-corrected percentile method with 2000 bootstrap replications.19 Isotonic regression and bootstrapping analyses were performed in R software version 3.4.4. (R Foundation for Statistical Computing, Vienna, Austria)
Demographic characteristics and secondary outcome estimates were reported as means ± standard deviations, medians (interquartile range), and counts and proportions. Parametric data were analysed with the t-test; nonparametric data were analysed with the Mann-Whitney test. Proportions were compared, using the Chi-square and Fisher exact tests, as suitable. The peri-operative haemodynamic parameters were assessed by multiple t-tests and estimated with a two-stage linear step-up procedure of Benjamini, Krieger, and Yekutieli, with Q = 1%. Statistical comparisons were made, using SPSS for Windows version 18.0 (SPSS Inc., Chicago, IL, USA) and GraphPad Prism 8 for windows (GraphPad Software, San Diego, California, USA). Statistical significance was defined as p-values of < 0.05.