Patients and data collection
A total of 152 female patients who were 30–60 years of age, were American Society of Anesthesiologists (ASA) class I-II, and presented for three-trocar video-assisted thoracoscopic surgery from February 24, 2020, to April 2, 2020, were enrolled. The Shanghai Jiaotong University Shanghai Chest Hospital (KS1865) Academic Review Board authorised the research procedure for surgeries involving general anaesthesia. We registered this research at Chictr.org (ChiCTR2000030429). Written informed consent was obtained from all participants.
The female patients were selected as follows: 1) capacity to comprehend Chinese verbal and written, 2) American Society of Anesthesiologists physical status I to II, 3) age 30 to 60 years, 4) scheduled for three-trocar VATS under general anesthesia, and 5) received 48 h continual PCIA after surgery.
The exclusion criteria were as follows: 1) age under 30 or over 60 years, 2) refusal to participate, 3) use of psychiatric medications and alcohol abuse, 4) history of cardiovascular disease, 5) acute or chronic liver or kidney disease, 6) cognitive impairment, and 7) pregnancy or lactation.
Patients were randomised with a 1:1 equal allocation ratio to sufentanil and dezocine-based PCIA (group C) or combined dexmedetomidine with sufentanil and dezocine-based PCIA (group D). A clinical trial statistician provided randomisation by using random number table. After written informed consent, the anaesthetist began randomisation of the patients using the online database. Later, in the procedure room, the randomised treatment was started.
The patients and the research team were not informed as to whether the patients were assigned to the sufentanil and dezocine-based PCIA group (group C) or the combined dexmedetomidine with sufentanil and dezocine-based PCIA group (group D) until the end of the study. Only the clinical trial statistician and dispensing nurse know the group allocation. The blinding could be disrupted in case of emergency if the patients’ health or safety were at risk.
Methods of Anaesthesia
According to the clinical research center protocol, when the patient arriveed at the anaesthesia preparation room, a peripheral vein was opened using a catheter, and right internal jugular central venous catheterisation was performed. When the participants were released to the operating room, in addition to a normal evaluation via pulse oximetry, non-invasive blood pressure (NIBP) and electrocardiography (ECG), the arterial pressure was assessed using a radial artery catheter in all patients. During surgery, surface electrodes for the bi-spectral index were placed on the front of the patients; BIS was maintained between 40 and 50. For both groups, a standardised anaesthetic method was used. Induction and upkeep of anaesthesia were the responsibility of the participating anaesthesiologists who were blinded to group assignments. Ten minutes before the induction of anaesthesia, 0.5 μg/kg dexmedetomidine (DEX) was in group D or normal saline in group C was added to a 20-ml syringe for administration over 10 min. Anaesthesia was subsequently induced at an effect-site concentration (Ce) of 4 μg/ml with sufentanil (0.5 μg/kg), cisatracurium (0.2 mg/kg), 50 mg flurbiprofen, and 2% propofol using a target-controlled infusion (TCI). The patients were intubated with a DLT in the lateral position, which required no operation. The correct position was confirmed with the use of a fibreoptic bronchoscope (FOB). Throughout the initial two-lung ventilation (TLV) and one-lung ventilation (OLV) cycles, the tidal volume was 7 ml/kg, which maintained PetCO2 at 35–45 mmHg (I/E = 1:2, f = 12). Anaesthesia was maintained using cisatracurium (0.12 mg/kg/h), remifentanil (0.1–0.3 μg/kg/min) and 2% propofol, with Ce at 2–3 μg/ml titrated to maintain BIS between 40 and 50; the mean arterial blood pressures (MAPs) and heart rates (HRs) were 20% less than the baseline values. The nasopharyngeal temperature was maintained at ≥36.5 °C. FOB was used to validate the correct DLT location after the patient was placed in a lateral decubitus position once more. Using 100% oxygen, anaesthesia induction and OLV were introduced and sustained. At the end of the operation, the inspired concentration of oxygen was decreased by 50% and the positive end expiratory pressure was increased to 5 cmH2O after the lung recruitment manoeuvre. At the end of surgery, intravenous infusion of all anaesthetics were stopped, a palonosetron hydrochloride injection (0.25 mg for a single injection) was given to prevent nausea and vomiting, and an electronic infusion pump (FSQ-11 PCA; Inc., JiangSu AIPENG, ED, China) for PCIA was connected. After surgery, patients were transferred to a post anaesthesia care unit (PACU). When fully awake, the patients were transferred to a regular ward and monitored during the whole study period.
Postoperative PCIA strategy
In the PCIA protocol, the sufentanil and dezocine levels were determined based on body weight in the study timeframe. As per the study design, the PCIA base regimen was 1.5 μg/kg sufentanil and 0.3 mg/kg dezocine diluted to 100 ml with 0.9% normal saline. In the D group, in addition to the sufentanil and dezocine, 3.0 μg/kg dexmedetomidine was added to the PCIA pump. The PCIA was set to a background infusion rate of 2 ml/h, lock-out interval of 15 min and 0.5 ml bolus on demand. All study patients received a 0.5 ml i.v. of PCIA solution while connected to a PCIA pump. The PCIA was set to a continuous background infusion of 0.03 μg/kg/h sufentanil with a bolus of 0.02 μg/kg sufentanil and allowed a continuous background infusion of 0.06 μg/kg/h dexmedetomidine with a bolus of 0.03 μg/kg dexmedetomidine. The PCA was used continuously for the first 48 h postoperatively. If patients complained of extreme discomfort, higher than a VAS level of 5, Once released to the general ward, the physicians advised them to push the PCA button, and if comfort was not achieved, then the nurse would administer a rescue analgesic of 50 mg flurbiprofen intravenously during the 48 h study period.
The primary endpoint measures was the satisfaction degree in the two groups within 48 h after surgery. The satisfaction degree scores were from 1 to 4 (4 = very satisfied, 3 = satisfied, 2 = moderately satisfied, and 1 = not satisfied).
The secondary endpoint measures were postoperative nausea and vomiting (PONV), PCA bolus, requirement for rescue analgesia, drug-related adverse effects, the visual analogue scale (VAS) pain scores at rest and while coughing, and the Ramsay sedation score (RSS) at 6, 12, 24, 36 and 48 h postoperatively. The operative times and the incidences of hypotension, atrial fibrillation and sinus bradycardia.
The nausea and vomiting (PONV) ratings were on a 4-point scale (1 = without nausea and vomiting, 2 = nausea without vomiting, 3 = less than twice vomiting, 4 = extreme vomiting more than twice), and the sedation score was on a 6-point sedation scale (1 = fully awake; 2 = cooperative, calm; 3 = response only to verbal commands; 4 = vigorous response to light stimulation while asleep; 5 = sleeping without light reaction; 6 = unarousable) . The degree of resting and coughing pain from the operation was assessed at 6, 12, 24, 36 and 48 h. A numerical rating scale from 0 to 10 (0 indicates no pain at all and 10 reflects the worst pain known) was used to measure pain severity .
The 48-h analgesia pump PCA bolus and rescue analgesia were recorded, and the occurrences of nausea and vomiting, pruritus, hypotension (MAP < 60 mmHg), sinus bradycardia (HR < 60 beats/min), excessive sedation (RSS ≥3), hypoxaemia (SpO2 < 93%), and respiratory depression (respiratory rate < 8 bpm) were recorded for 48 h postoperatively.
Quantitative variables are reported as the mean ± standard deviation. Categorical data are defined in terms of frequencies and fractions. The statistical study was carried out using SPSS version 22 (SPSS, Chicago, IL). The χ2 test or Fisher’s exact test was performed to compare all categorical variables. Continuous variables were tested with the non-parametric Mann-Whitney U test or t-test depending on the distribution of the data. Two-sided p-values < 0.05 were considered as statistically significant. The sample size was estimated based on an anticipated 20% drop in the satisfaction degree ratio for 48 h post-surgery. For an 80% power (α = 0.05, β = 0.2), the sample size needed for each group was estimated to be 69, with a dropout rate of 10%. There were 76 patients per group for eligibility.