Ethics and registration

This study was approved by the Medical Ethics Committee in Affiliated Hospital of Yangzhou University, Jiangsu, China (Protocol/serial number: 2022-YKL3-08–002) on 8 March 2022. All study subjects gave informed consent for all treatments and investigations. All procedures in the study involving human participants were performed in accordance with the ethics standards of the institutional and national research committee and with the Helsinki Declaration and its later amendments or comparable ethics standards. The study was conducted and reported in accordance with the Consolidating Standards of Reporting Trials (CONSORT) 2010 statement. This trial was registered at the Chinese Clinical Trial Registry (https://www.chictr.org.cn/edit.aspx?pid=164523&htm=4) on 10/4/2022 (registry number: ChiCTR2200058544).

Patients inclusion and exclusion criteria

Subjects to be enrolled in the study were selected prior to an elective VATS procedure.

Subject inclusion criteria

An elective thoracoscopic lobectomy scheduled,

The surgical procedure to be performed under general anaesthesia,

Age from 18 to 64 years,

Body Mass Index (BMI) from18 to 30 kg/m2,

ASA physical status classification system class I-II.

Subject exclusion criteria

Allergies to the local anesthetics, morphine, or drugs used in the study,

Cardiac diseases, hepatic insufficiency, and renal failure,

Subject’s refusal to participate in the study,

Inability to understand what the study is about,

Previous cardiothoracic surgery,

Any contraindication to intrathecal injection (morphine allergy, coagulopathy, patient refusal),

Patients with a history of chronic pain,

Preoperative opioid use and/or history of opioid abuse,

Psychiatric disease.

Patients were also excluded when failure to operate intrathecal anaesthesia, discontinuation of an intravenous analgesic pump due to a serious adverse reaction, and change of procedure.

Procedures

After patients were admission, oxygen was administered and ECG, HR, NBP and SPO2 were monitored. The right internal jugular vein was punctured and a central venous catheter was placed successfully to establish intravenous access. A radial artery puncture was performed on the healthy side for pressure measurement.

The enrolled patients were randomized by random number method into three groups: Group M (5 μg/kg morphine intrathecal injection), Group B (5 μg/kg morphine combined with bupivacaine 3 mg intrathecal injection) and Group C (intrathecal sham injection). Patients were randomly allocated 1: 1: 1 with three treatment groups using a set of computer-generated random numbers kept in sealed envelopes by an investigator not involved in clinical care. Blocks were performed before induction of general anaesthesia, and patients were not informed of their group assignments. The three groups were intrathecal injection morphine, morphine and bupivacaine, or intrathecal sham injection. The anaesthetist conducting the anaesthetic was not blinded to treatment.

Group M: Under standard monitoring, patients were placed in the left lateral position and the skin of the low back was disinfected with 0.45–0.55% povidone-iodine disinfectant centred on the L2-3 gap and sterile cavity wipes were laid. 1% lidocaine 2 ml was infiltrated locally and a type II lumbar puncture needle (5#) was used to perform an intrathecal puncture in the L2-3 space. The cerebrospinal fluid was drained freely and morphine hydrochloride 5 μg/kg was injected into the subarachnoid space within 15 s. Morphine hydrochloride configuration: dilute 1 mg morphine hydrochloride (1 mg/ml) with 0.9% sodium chloride to 10 ml (0.1 mg/ml) and inject into a 3 ml syringe after drawing at 5 μg/kg using a 1 ml syringe. Retract cerebrospinal fluid diluted to 3 ml for subarachnoid injection.

Group B: The procedure for subarachnoid puncture was the same as that for M group. 5 μg/kg of morphine hydrochloride (morphine preparation was the same as that for M group) combined with 0.75% bupivacaine 0.4 ml (3 mg) was withdrawn and diluted to 3 ml of cerebrospinal fluid for subarachnoid injection.

Group C: After the patient has cooperated in the position, the anaesthetist applies finger pressure to the skin to simulate intrathecal injection.

The 10°Trendelenburg position was used in all three groups. The flat position was resumed 20 min after completion of the operation to achieve a high plane block. Sterile dressings were given to cover the lumbar back after completion of the puncture operation in all three groups. The upper boundary of the plane of anaesthesia was measured with an alcohol swab 10–15 min after administration.

For all three groups, general anaesthesia was induced with 0.05 mg/kg midazolam, 0.4 μg/kg sufentanil and 1.5–2.0 mg/kg propofol. Cisatracurium 0.2 mg/kg was given to facilitate double-lumen bronchial tube intubation. Propofol administration was adjusted to target a Bispectral Index between 40 and 60. Cisatracurium was infused at 0.05 ~ 0.10 mg·kg−1·h−1 and remifentanil was infused at 0.1 ~ 0.25 μg·kg−1·min−1, adjusted as necessary. A single dose of azasetron (10 mg) was permitted for PONV prophylaxis. After surgery, patients were transferred to the postanaesthesia care unit for extubation.

Postoperative analgesia: All three groups received PCIA after awakening in the PACU with the same settings: sufentanil 2 μg/kg and azasetron 10 mg diluted to 100 ml with 0.9% NaCl. PCA dose, 2 ml, lockout interval 15 min. If the PCA device was fully used (NRS > 4 after three consecutive effective compressions), then aminotriol ketorolac 30 mg is administered intramuscularly for remedial analgesia. In case of postoperative nausea and vomiting, metoclopramide 10 mg is administered intramuscularly.

Measurement

Patients’ baseline data, surgery and anaesthesia were collected. Mean blood presssure and heart rate at eight time points were recorded: After patient admission (T0), 10 min after intrathecal block (T1), 1 min after double-lumen tracheal intubation (T2), 1 min before surgical cut (T3), 1 min after surgical cut (T4), 5 min after surgical cut (T5), at the end of surgery (T6) and 30 min after extubation (T7). IVPCA sufentanil consumption in 48 h and NRS at 1, 6, 12, 24 and 48 h postoperatively were recorded. Intra-operative consumption of remifentanil and propofol was recorded. Additionally, common side effects (sedative effect, nausea/vomiting, and pruritus) and some recovery indicators were recorded.

For sedation, the Ramsay Sedation Score was used: 1 score: irritable; 2 score: awake, quiet and cooperative; 3 score: drowsy, but able to follow commands and respond with agility; 4 score: in a state of drowsiness, can be awakened quickly; 5 score: in a state of light sleep, unresponsive to calls and responsive to stronger stimuli (e.g., tapping); and 6 score: in a state of deep sleep, not waking up to calls. If the score is greater than 4, the patient is over-sedated, and 15 μg of nalmefene is given intravenously for over-sedation. Nausea and vomiting were classified as three grades: asymptomatic, symptomatic without treatment, and symptomatic with treatment. Gastrografin 10 mg intramuscularly was given, and haloperidol 0.63–1.25 mg intravenously was given for unrelieved nausea and vomiting, which was repeated once every 8 h if necessary.

Pruritus was classified as asymptomatic, symptomatic without treatment and symptomatic with treatment, 10 mg of propofol was given for itching with the maximum dose not exceeding 50 mg. 10 mg of cetirizine tablets were given orally if itching persisted.

All patients were given a preoperative urinary catheter, which was removed on the second day. If urinary retention occurred after removal of the urinary catheter, hot compresses and massage were used, and the catheter was reintroduced if it could not be relieved.

All results were recorded by an anaesthetist who had no knowledge of the subgroups. The scores of Quality of Recovery-15 (QoR-15) is the recently developed to validate short-form postoperative recovery evaluation. QoR-15 questionnaire has 15 questions that assess patient-reported postoperative health status using a 11-point numerical rating scale that leads to a minimum score of 0 (poor recovery) and a maximum score of 150 (excellent recovery). QoR-15 was obtained 24 h and 48 h after surgery in this trail.

The primary outcome was the pain scores at rest and with cough at 48 h after surgery. Secondary outcomes were intraoperative propofol and remifentanil consumption, intraoperative haemodynamic variables, postoperative rescue analgesic requirements and adverse events; and postoperative recovery indicators.

Statistical analysis

Variables were summarized as mean ± SD for normally distributed continuous variables (including demographic profile of the patients, duration of surgery, fluid intake and output, hemodynamic changes, Intra-operative remifentanil use, propofol use, first feeding time, first time out of bed, QoR-15 score), and intergroup comparisons were performed using one-way analysis of variance (ANOVA). Median (25th to 75th percentile) for continuous variables with evidence of nonnormality or for ordinal variables (Hospital stay days, blood loss, pain scores, cumulative amount of IVPCA sufentanil used, time of first analgesic request, post-operative extubation time), comparisons between groups were made using the Kruskal–Wallis test. Frequency counts and percentages for categorical variables were analysed using the Pearson’s chi-squared test (sex, ASA physical status, underlying disease, chest drain days, nausea/vomiting) or Fisher’s exact test (pruritus). Statistical analysis was performed using statistical software SPSS (25.0; IBM Corp., Armonk, NY, USA). The 95% confidence intervals of the median differences were estimated by bootstrap method based on resampling through Python version 2.7.6. The significance level was set at P < 0.05.

Sample size considerations

48 h post-operative coughing on the NRS scores was defined as a clinically meaningful effect. Assuming a standard deviation of 2 for each of the three pain score outcomes on the basis of a preliminary study. The pretest result was an NRS score of 1.3 for group M, 1.2 for group B and 3.3 for group C at 48 h postoperative coughing. A sample size of 20 patients per group had 90% power to detect effectiveness of intrathecal morphine and intravenous analgesic pumps on all three primary outcomes at the 0.05 significance level. To allow for some individuals not completing the trial, 33 patients were recruited in each treatment group.