Effect of ultrasound-guided superficial versus deep superior costotransverse ligament on haemodynamics during general anaesthesia induction of thoracoscopic lung resection: protocol for a multicentre, double-blind, randomised controlled trial Authors: Yajie Xu, Yi Luo, Wenwen Zhang, Zhaojing Fang, Ziqi Qiu, Honggang Yi, Hongwei Shi, Hao Cheng, Xiaoliang Wang Published: 2026 ABSTRACT AbstractBackgroundThoracic paravertebral block (TPVB) is widely used for analgesia in thoracoscopic lung resection, offering effective pain relief and enhanced recovery. However, preoperative TPVB may block both thoracic somatic nerves and cardiac sympathetic nerves, potentially increasing the risk of hypotension during general anaesthesia induction. This study aims to compare the haemodynamic effects of TPVB performed at the superficial vs deep layers of the superior costotransverse ligament (SCTL).Methods and analysisThis randomised controlled trial will enrol 128 patients undergoing thoracoscopic lung resection under general anaesthesia. Patients will be randomised (1:1) to one of two groups: the Deep SCTL Block Group (Group T), receiving ultrasound-guided deep layer SCTL block 30 min before surgery in the pre-anaesthesia room, or the Superficial SCTL Block Group (Group S), receiving ultrasound-guided superficial layer SCTL block at the same time point. In addition, a supplementary single-centre exploratory 3D-CT imaging substudy will enrol 40 patients requiring clinically indicated preoperative CT-guided lesion localisation. These participants will be randomly allocated at a 1:1 ratio to receive either deep or superficial SCTL block before CT-guided localisation. 3D-CT imaging will be performed 30 min after block administration to compare contrast-enhanced local anaesthetic spread patterns and explore their potential associations with peri-induction haemodynamic changes.The primary outcome is the incidence of hypotension during general anaesthesia induction. Secondary outcomes include the incidence of severe hypotension during induction, as well as Visual Analogue Scale scores at rest and during movement at 4 and 24 hours postoperatively.Ethics and disseminationThis trial was approved by the Ethics Committee of Nanjing First Hospital (No. KY20240924-03-KS-01) on 24 September 2024. The results will be peer-reviewed for publication in a scientific journal.Trial registration numberNCT06841822.Keywords: Adult anaesthesia, Pain management, Blood Pressure STRENGTHS AND LIMITATIONS OF THIS STUDY Multicentre randomised double-blind controlled design. Multicentre randomised double-blind controlled design. Stratified randomisation and blinded outcome assessment reduce bias. Stratified randomisation and blinded outcome assessment reduce bias. Continuous invasive haemodynamic monitoring during anaesthesia induction. Continuous invasive haemodynamic monitoring during anaesthesia induction. 3D-CT imaging used to evaluate local anaesthetic spread. 3D-CT imaging used to evaluate local anaesthetic spread. Imaging substudy conducted in a subset of patients at a single centre. Imaging substudy conducted in a subset of patients at a single centre. Introduction Thoracoscopic lung resection, a minimally invasive surgical approach associated with enhanced recovery, underscores the need for optimised perioperative pain management strategies. Several regional anaesthesia techniques are commonly used for thoracic surgery, including thoracic epidural analgesia, thoracic paravertebral block (TPVB) and erector spinae plane block (ESPB).1 Thoracic epidural analgesia and TPVB provide reliable analgesia but may produce significant sympathetic blockade, which can lead to hypotension, particularly when the block is performed before induction of general anaesthesia. In contrast, ESPB is generally considered haemodynamically more stable because the spread of local anaesthetic to the thoracic paravertebral space is often indirect and variable, although this may result in less consistent neural blockade.2 Because peri-induction hypotension has been associated with an increased risk of organ hypoperfusion and postoperative complications, strategies that maintain effective analgesia while minimising excessive sympathetic inhibition are of clinical interest. TPVB, as one of the most commonly used regional techniques for thoracic surgery, is therefore of particular concern, since its preoperative administration may induce sympathetic blockade and contribute to haemodynamic instability during anaesthesia induction despite improvements in ultrasound-guided safety techniques.3 4 Anatomically, the thoracic paravertebral space is a wedge-shaped compartment bounded by the vertebral column, costotransverse ligaments and pleura, housing critical neurovascular structures. Recent cadaveric studies by Costache et al and high-resolution micro-CT imaging by Cho et al have elucidated two pivotal anatomical insights: (1) the superior costotransverse ligament (SCTL) does not constitute a complete barrier to local anaesthetic diffusion and (2) the paravertebral space exhibits narrow dimensions with direct continuity to adjacent tissue planes.3 5 These findings suggest that the depth of local anaesthetic injection relative to the SCTL may significantly influence the spread of local anaesthetic into the thoracic paravertebral space and thereby affect the extent of sympathetic blockade. Specifically, superficial SCTL injections—where the local anaesthetic is deposited without penetrating the ligament—may facilitate gradual diffusion into the paravertebral space, potentially reducing sympathetic blockade compared with deep SCTL injection. This study aims to evaluate whether a superficial SCTL block can maintain haemodynamic stability during anaesthesia induction while providing analgesic efficacy comparable to conventional deep SCTL approaches. To further explore the mechanism underlying potential differences, 3D-CT imaging will be used to visualise the spread of local anaesthetic after superficial and deep SCTL block. By combining clinical outcomes with imaging findings, this trial seeks to determine whether modification of the injection plane relative to the SCTL can improve peri-induction haemodynamic stability without compromising analgesic efficacy. Methods Study design This prospective randomised controlled trial is carried out at five medical centres. A total of 168 patients will be recruited, including 128 participants for the main multicentre randomised clinical trial and 40 participants for the single-centre exploratory 3D-CT imaging substudy conducted at Nanjing First Hospital. The 40 patients in the imaging substudy will not contribute to the primary efficacy analysis of the main clinical trial. figure 1 shows the study flowchart. Study flow diagram. CI, cardiac index; SCTL, superior costotransverse ligament; SV, stroke volume; SVI, stroke volume index; VAS, Visual Analogue Scale; QoR-15, 15-item Quality of Recovery questionnaire. Participating centres A total of five centres are going to attend this multicentre trial which is listed as the following: Department of Anesthesiology, Perioperative and Pain Medicine, Nanjing First Hospital, Nanjing Medical University, Nanjing, China. Department of Anesthesiology, Perioperative and Pain Medicine, Nanjing First Hospital, Nanjing Medical University, Nanjing, China. Department of Anesthesiology and Pain Medicine, Geriatric Hospital of Nanjing Medical University. Department of Anesthesiology and Pain Medicine, Geriatric Hospital of Nanjing Medical University. Department of Anesthesiology, Yancheng No.1 People’s Hospital, Yancheng, China. Department of Anesthesiology, Yancheng No.1 People’s Hospital, Yancheng, China. Department of Anesthesiology, Suzhou Dushu Lake Hospital, Suzhou, China. Department of Anesthesiology, Suzhou Dushu Lake Hospital, Suzhou, China. Department of Anesthesiology, Huai'an First People’s Hospital, Huai'an, China. Department of Anesthesiology, Huai'an First People’s Hospital, Huai'an, China. Inclusion criteria Participants scheduled for elective double-port thoracoscopic lung resection. Participants scheduled for elective double-port thoracoscopic lung resection. Aged ≥18 years. Aged ≥18 years. American Society of Anesthesiologists(ASA) classification of I–III. American Society of Anesthesiologists(ASA) classification of I–III. Body Mass Index(BMI) of 18 to 30 kg/m². Body Mass Index(BMI) of 18 to 30 kg/m². For 3D-CT imaging subgroup: scheduled for preoperative CT-guided lesion localisation. For 3D-CT imaging subgroup: scheduled for preoperative CT-guided lesion localisation. Exclusion criteria Severe hypertension (systolic blood pressure (BP) ≥180 mm Hg or diastolic BP ≥110 mm Hg). Severe hypertension (systolic blood pressure (BP) ≥180 mm Hg or diastolic BP ≥110 mm Hg). Mean arterial pressure (MAP) <70 mm Hg before induction of general anaesthesia. Mean arterial pressure (MAP) <70 mm Hg before induction of general anaesthesia. Emergency surgery. Emergency surgery. Severe cardiovascular disease (history of brain or thoracic/abdominal aortic aneurysm). Severe cardiovascular disease (history of brain or thoracic/abdominal aortic aneurysm). Congestive heart failure, classified as New York Heart Association Class III or IV. Congestive heart failure, classified as New York Heart Association Class III or IV. Untreated or unstable ischaemic heart disease. Untreated or unstable ischaemic heart disease. History of severe aortic valve or mitral valve disease. History of severe aortic valve or mitral valve disease. Pregnant or breastfeeding women. Pregnant or breastfeeding women. Coagulation dysfunction. Coagulation dysfunction. Bacteremia, septicemia or infection at the puncture site. Bacteremia, septicemia or infection at the puncture site. Severe liver and kidney dysfunction. Severe liver and kidney dysfunction. Allergy to any of the drugs involved in the study; for participants in the 3D-CT imaging substudy, known allergy to iodinated contrast media will also be an exclusion criterion. Allergy to any of the drugs involved in the study; for participants in the 3D-CT imaging substudy, known allergy to iodinated contrast media will also be an exclusion criterion. Concurrent neurological disorders, spinal diseases (malformations or trauma), history of spinal surgery or abnormal skin sensation in the chest or back. Concurrent neurological disorders, spinal diseases (malformations or trauma), history of spinal surgery or abnormal skin sensation in the chest or back. Difficulty or anticipated difficulty in airway management. Difficulty or anticipated difficulty in airway management. Other conditions deemed unsuitable for inclusion by the investigator. Other conditions deemed unsuitable for inclusion by the investigator. Drop-out criteria Participants who do not follow the study protocol. Participants who do not follow the study protocol. Unwilling to continue with examination or treatment. Unwilling to continue with examination or treatment. Those who concurrently use prohibited medications or drugs that may affect the evaluation of efficacy. Those who concurrently use prohibited medications or drugs that may affect the evaluation of efficacy. Those who experience serious adverse events during treatment and require emergency unblinding. Those who experience serious adverse events during treatment and require emergency unblinding. Participants whose pre-existing conditions worsen to the extent that they are unable to complete the study. Participants whose pre-existing conditions worsen to the extent that they are unable to complete the study. Randomisation and blinding This study will use stratified block randomisation to allocate participants into two groups at a 1:1 ratio. Randomisation will be stratified by study centre and performed by an independent statistician using a computer-generated randomisation sequence. Allocation will be concealed using sequentially numbered, sealed, opaque envelopes, which will be opened only by the anaesthesiologist performing the nerve block immediately before the procedure. Because of the nature of the intervention, the anaesthesiologist performing the SCTL block cannot be blinded to group allocation. However, this anaesthesiologist will not be involved in anaesthesia induction, intraoperative management, postoperative care, data collection or outcome assessment. After completion of the block, this investigator will have no further role in perioperative care. To minimise performance bias, the block procedure will be performed in the pre-anaesthesia room, and the intraoperative anaesthesia team will not be present during block performance. An independent anaesthesiologist, blinded to group allocation, will be responsible for anaesthesia induction, haemodynamic management and intraoperative anaesthesia according to the predefined protocol. The group assignment will not be recorded in the anaesthesia chart or any clinical document accessible to the intraoperative anaesthesia team. Sensory blockade assessment will be performed by a separate blinded investigator before anaesthesia induction. The results of sensory assessment will be recorded in the case report form but will not be disclosed to the anaesthesia management team, because the extent of sensory blockade may potentially suggest treatment allocation. Outcome assessors responsible for recording peri-induction haemodynamic data and postoperative outcomes will remain blinded to group allocation throughout the study. Participants will also remain unaware of their allocated intervention. To prevent inadvertent unblinding, communication regarding treatment allocation will be restricted to the block performer only. Study personnel involved in anaesthesia induction, haemodynamic management, postoperative follow-up, outcome assessment and statistical analysis will be instructed not to discuss allocation information. Any accidental disclosure of group allocation, emergency unblinding or protocol deviation related to blinding will be documented in the case report form and reported to the study coordinating centre. These cases will be considered in the per-protocol and sensitivity analyses when appropriate. For the supplementary 3D-CT imaging substudy conducted exclusively at Nanjing First Hospital, eligible patients will be allocated into two groups using simple randomisation without stratification factors. Radiologists performing the scans and imaging assessment will remain blinded to group allocation. Anaesthesia Preoperative routine follow-up is conducted 1 day prior to surgery for all patients. The patient’s age-adjusted Charlson Comorbidity Index and Clinical Frailty Scale are recorded, along with the patient’s BP in the ward. All patients are required to fast for 8 hours prior to the surgery. After entering the anaesthesia preparation room, peripheral venous access is established, and non-invasive monitoring of BP, ECG and oxygen saturation is performed. All patients undergo invasive BP monitoring with radial artery catheterisation under local anaesthesia and the FloTrac/Vigileo (EV 1000 system, Edwards Lifesciences, Irvine, California, USA) haemodynamic monitoring system is established. Midazolam is administered at 0.01 mg/kg to maintain a calm environment for the patient for at least 5 min, and heart rate (HR) and MAP are measured every minute for three consecutive times, with the average value recorded as the baseline.5 During the nerve block, the anaesthesiologist opens a sealed opaque envelope containing the patient’s group assignment and performs the nerve block accordingly. Group S receives the superficial layer of SCTL, while Group T receives the deep layer. After 30 min, the patient is transferred to the operating room, and general anaesthesia is administered. No fluids are given prior to anaesthesia induction. During the study period, all patients receive balanced crystalloids at a rate of 10 mL/kg/hour. The anaesthesiologist administers general anaesthesia induction as follows: two target-controlled infusion (TCI) pumps are used, delivering 1% propofol (4.0 µg/mL) for the initial target concentration (effect-site, modified Marsh model, ke0=1.21 /min) and remifentanil (5 ng/mL) at a fixed target concentration (plasma-site, Minto model). Remifentanil’s target concentration (5 ng/mL) is reached 5 min after administration. Propofol is titrated in increments of 0.1 µg/mL to maintain a BIS of 40–50. Following neuromuscular blockade monitoring, rocuronium (0.6 mg/kg) is given after loss of consciousness for muscle relaxation. After confirming that remifentanil has reached the target concentration (5 ng/mL) and Train-of-Four (TOF)=0, tracheal intubation is performed. HR and MAP are measured every minute from the start of induction to 5 min post-induction. Both groups undergo tracheal intubation 5 min after anaesthesia induction to compare the haemodynamic effects of different nerve blocks. Video laryngoscopy is used for left double-lumen bronchial intubation, and mechanical ventilation is established with fibreoptic bronchoscopy for positioning. The ventilation mode is set to pressure-controlled ventilation–volume guaranteed, with tidal volume (VT) 6–8 mL/kg, FiO2 40%, positive end-expiratory pressure 5 cm H2O, I:E 1:2 and RR adjusted to maintain partial pressure of end-tidal carbon dioxide at 35–45 mm Hg. After induction, sufentanil (0.3 µg/kg) is given, and anaesthesia is maintained with TCI pumps for propofol (1.5–3.5 µg/mL) and remifentanil (4.0–8.0 ng/mL) based on the surgical situation. Rocuronium is intermittently administered to maintain muscle relaxation, and anaesthesia depth is monitored to maintain a bispectral index (BIS) of 40–60. After one-lung ventilation begins, VT is adjusted to 4–6 mL/kg while other respiratory parameters remain unchanged. Throughout surgery, the surgeon assesses for any nerve block-related complications, including bleeding or accidental pleural puncture. 30 min before the end of surgery, a single dose of oxycodone (0.03 mg/kg) is administered. Postoperatively, all patients are transferred to the Post-Anaesthesia Care Unit for recovery. Extubation occurs once the muscle relaxation test (TOF≥0.9) is passed and the patient is awake. Postoperative pain management is provided with patient-controlled intravenous analgesia, consisting of oxycodone (20 mg), tropisetron (6 mg) and saline to 100 mL, with a background dose of 0.5 mL/hour and a bolus dose of 5 mL every 8 min. Hypotension is categorised as a MAP reduction of 20% or an absolute MAP below 65 mm Hg, while severe hypotension is defined as a MAP reduction of 30% or an absolute MAP below 55 mm Hg. For hypotension management, if the HR exceeds 40 bpm, 50 µg of phenylephrine is administered. If HR is under 40 bpm, 6 mg of ephedrine is given. In cases where systolic arterial pressure rises more than 20% above baseline, 300 µg of nicardipine may be administered. Bradycardia (HR<40 bpm) is treated with 0.5 mg of atropine, and tachycardia (HR>100 bpm) is managed with esmolol at a dose of 0.5 mg/kg. Study interventions Table 1 presents the process of enrolment, intervention and assessment conforming to the Standard Protocol Items: Recommendations for Interventional Trials (SPIRIT) statement. The Group S receives ultrasound-guided the superficial layer of SCTL, while Group T receives the deep layer of SCTL. Both groups receive neural block injections consisting of 20 mL of 0.375% ropivacaine at the T4 and T6 levels, respectively. Table 1Diagrammatic representation of trial processesStudy periodEnrolmentAllocationPost-operation (hours)TimepointPre-anaesthesiaInductionOperation42472 Enrolment: Eligibility screening×Informed consent×Allocation×Demographics× Interventions: The Deep SCTL(Group T)×The Superficial SCTL (Group S)× Assessments Testing the nerve block plane×3D-CT imaging×Adverse events related to puncture×Hypotension×SV, CI, SVI××Use of vasoactive drugs×Other indicators during general anaesthesia××VAS score××Analgesic pump usage and rescue analgesia×QOR-15 score×Other complications×Patient Satisfaction Survey×CI, cardiac index; QoR-15, 15-item Quality of Recovery questionnaire; SCTL, superior costotransverse ligament; SV, stroke volume; SVI, stroke volume index; VAS, Visual Analogue Scale. Diagrammatic representation of trial processes CI, cardiac index; QoR-15, 15-item Quality of Recovery questionnaire; SCTL, superior costotransverse ligament; SV, stroke volume; SVI, stroke volume index; VAS, Visual Analogue Scale. During the nerve block procedure, the patient is placed in the lateral position on the unaffected side with full exposure of the back. Ultrasound is used to locate and mark the T4 and T6 spinous processes, and sterile technique is applied for disinfection. A portable colour Doppler ultrasound device with a low-frequency probe is used to identify the SCTL, gradually moving the probe along the posterior axillary line of the affected side until the SCTL is visible. The needle is then inserted based on group assignment, either to the superficial (non-penetrating) or deep (penetrating) layer of the SCTL. After confirming no blood, cerebrospinal fluid, or air on aspiration, 0.375% ropivacaine (20 mL) is injected at the T4 and T6 target sites. Following the block, patients will remain in the supine position. After 30 min, a blinded assessor will evaluate the extent of sensory loss using a cold stimulus at the anterior chest wall (midclavicular line), lateral chest wall (posterior axillary line) and posterior chest wall (paravertebral region). Cold sensitivity testing with a temperature method is performed at the relevant blockade areas along the midaxillary line at 5, 10, 20 and 30 min after the block, and the blockade level is recorded. In the supplementary 3D-CT imaging substudy, patients who require clinically indicated preoperative CT-guided lesion localisation at Nanjing First Hospital will be screened for eligibility. This substudy will include 40 patients, with 20 patients allocated to the deep SCTL block group and 20 patients allocated to the superficial SCTL block group. 30 min before CT-guided lesion localisation, patients will receive either an ultrasound-guided deep SCTL block or an ultrasound-guided superficial SCTL block on the surgical side at the T4 and T6 levels. A total of 20 mL of injectate will be administered at each level, consisting of 0.375% ropivacaine mixed with iohexol, with a final volume of 20 mL per injection site. Patients will then undergo standard CT-guided lesion localisation, and 3D-CT reconstruction will be performed to assess the spread of contrast-enhanced local anaesthetic. The 3D-CT imaging substudy is designed as a single-centre exploratory mechanistic component rather than a confirmatory efficacy analysis. It will not contribute to the primary sample size calculation or the primary efficacy analysis of the main multicentre clinical trial. The rationale for this substudy is that the different haemodynamic effects of superficial and deep SCTL block may be related to differences in the anatomical spread of local anaesthetic and the consequent extent of sympathetic blockade. We hypothesise that deep SCTL block may allow more direct or extensive spread of local anaesthetic into the thoracic paravertebral space and adjacent neural structures, potentially resulting in greater sympathetic blockade and a higher risk of BP reduction during anaesthesia induction. In contrast, superficial SCTL block may produce a more limited or gradual spread pattern, thereby attenuating sympathetic inhibition while preserving postoperative analgesic efficacy. Because sympathetic blockade cannot be directly visualised by CT imaging, the distribution of contrast-enhanced injectate will be used as an anatomical surrogate for the potential extent of neural and sympathetic involvement. Imaging outcomes will include spread to the paravertebral space, neural foramina, epidural space, intercostal space and the craniocaudal extent of distribution. These imaging outcomes will be analysed descriptively and comparatively between the two groups. Exploratory analyses will be performed, where feasible, to examine whether predefined imaging spread patterns are associated with peri-induction haemodynamic variables, including hypotension occurrence and changes in mean arterial pressure (MAP), stroke volume (SV), cardiac index (CI) and stroke volume index (SVI). However, given the small sample size and single-centre nature of the imaging substudy, these imaging-haemodynamic association analyses will be interpreted as exploratory and hypothesis-generating rather than confirmatory evidence. Outcome measures Primary outcome The incidence of hypotension during the general anaesthesia induction period (from the start of induction to 5 min post-induction). Secondary outcomes The incidence of severe hypotension during the general anaesthesia induction period (from the start of induction to 5 min post-induction); severe hypotension was defined as a decrease in MAP≥30% from baseline or an absolute MAP<55 mm Hg. The incidence of severe hypotension during the general anaesthesia induction period (from the start of induction to 5 min post-induction); severe hypotension was defined as a decrease in MAP≥30% from baseline or an absolute MAP<55 mm Hg. Visual Analogue Scale at rest and during movement (coughing) at 4 and 24 hours postoperatively. Visual Analogue Scale at rest and during movement (coughing) at 4 and 24 hours postoperatively. Use of vasoactive drugs during induction; Use of vasoactive drugs during induction; HR, MAP, SV, CI and SVI were recorded before nerve block, during induction (per minute) and at the occurrence of hypotension. Minimum values of SV, CI and SVI during induction were also recorded. HR, MAP, SV, CI and SVI were recorded before nerve block, during induction (per minute) and at the occurrence of hypotension. Minimum values of SV, CI and SVI during induction were also recorded. The amount of propofol and remifentanil used during the surgery. The amount of propofol and remifentanil used during the surgery. Anaesthesia duration, surgery duration, intraoperative intravenous fluid volume, intraoperative urine output, intraoperative blood loss and extubation time. Anaesthesia duration, surgery duration, intraoperative intravenous fluid volume, intraoperative urine output, intraoperative blood loss and extubation time. The use of the pain pump within 24 hours postoperatively (oxycodone usage, first press time, effective press count, actual press count) and rescue analgesic administration. The use of the pain pump within 24 hours postoperatively (oxycodone usage, first press time, effective press count, actual press count) and rescue analgesic administration. The QOR-15, a validated 15-item patient-reported outcome measure assessing postoperative recovery across five domains (physical comfort, emotional state, physical independence, psychological support, and pain), at 24 hours postoperatively. The QOR-15, a validated 15-item patient-reported outcome measure assessing postoperative recovery across five domains (physical comfort, emotional state, physical independence, psychological support, and pain), at 24 hours postoperatively. The observation of puncture-related complications within 72 hours postoperatively (hemothorax, pneumothorax, local haematoma, local anaesthetic poisoning, epidural anaesthesia, total spinal anaesthesia, etc). The observation of puncture-related complications within 72 hours postoperatively (hemothorax, pneumothorax, local haematoma, local anaesthetic poisoning, epidural anaesthesia, total spinal anaesthesia, etc). Using a Likert scale to assess patient satisfaction 72 hours postoperatively; Using a Likert scale to assess patient satisfaction 72 hours postoperatively; In the supplementary 3D-CT imaging substudy, the spread of contrast-enhanced local anaesthetic will be evaluated in predefined anatomical regions, including the paravertebral space, neural foramina, epidural space, intercostal space and the craniocaudal extent of distribution. Sensory blockade will be assessed 30 min after the nerve block using cold stimulation in the anterior chest wall (midclavicular line), lateral chest wall (posterior axillary line) and posterior chest wall (paravertebral area). These imaging outcomes will be analysed as exploratory mechanistic endpoints. In the supplementary 3D-CT imaging substudy, the spread of contrast-enhanced local anaesthetic will be evaluated in predefined anatomical regions, including the paravertebral space, neural foramina, epidural space, intercostal space and the craniocaudal extent of distribution. Sensory blockade will be assessed 30 min after the nerve block using cold stimulation in the anterior chest wall (midclavicular line), lateral chest wall (posterior axillary line) and posterior chest wall (paravertebral area). These imaging outcomes will be analysed as exploratory mechanistic endpoints. Data collection An independent investigator screens patients according to the eligibility criteria and collects the demographics and baseline characteristics. The outcome measurements and perioperative data are documented by independent research investigators via postoperative patient visits and reviewing electronic medical records. All data are recorded in the case report forms and registered in the electronic database. An independent data monitoring committee conducts an ongoing data review throughout this study. Patients’ confidentiality will be ensured and deidentified data will be analysed. Sample size estimation Preliminary data from a single-centre pilot study conducted at Nanjing First Hospital showed that the incidence of hypotension during anaesthesia induction was 76.2% (16/21) in the deep-layer SCTL block group and 52.9% (9/17) in the superficial-layer SCTL block group. The pilot study used the same anaesthesia induction protocol, nerve block technique, haemodynamic monitoring method, hypotension definition and outcome assessment procedures as the present trial. Based on these preliminary data, with a two-sided significance level of 0.05 and 80% power, Power Analysis and Sample Size software estimated that 57 participants would be required in each group. After allowing for an anticipated 10% dropout rate, 128 participants will be recruited for the main clinical trial. Because this is an individually randomised multicentre trial, the original sample size calculation was performed at the individual-patient level. Randomisation will be stratified by study centre, and the anaesthesia protocol, nerve block technique, haemodynamic management strategy, monitoring method and primary outcome definition will be standardised across all participating centres. Therefore, substantial between-centre clustering is not expected. Nevertheless, we acknowledge that centre-level variability may reduce the effective sample size and should be explicitly considered. To evaluate the potential impact of centre-level clustering, we estimated the design effect using the formula: DE=1+(m−1)×Intraclass Correlation Coefficient (ICC), where m represents the average number of participants per centre and ICC represents the intraclass correlation coefficient. With 128 participants recruited from five centres, the average cluster size is approximately 25.6. Because reliable published ICC estimates are unavailable for this specific endpoint and intervention setting, sensitivity calculations were performed using plausible ICC values of 0.01 and 0.02. The corresponding design effects were 1.246 and 1.492, resulting in effective sample sizes of approximately 103 and 86 participants, respectively. These calculations suggest that the trial may be underpowered if the true between-centre variability is higher than anticipated. To minimise the influence of imbalanced centre recruitment, enrolment will be monitored throughout the study and distributed across centres as evenly as feasible. No single centre is expected to contribute more than 50% or fewer than 10% of the main trial population. In the primary analysis, study centre will be included as a random effect in a generalised linear mixed model to account for potential between-centre variability. The robustness of the primary findings will also be assessed using ICC-based sensitivity analyses. If the observed centre-level variability is greater than anticipated, this limitation will be acknowledged when interpreting the trial results. In addition, the supplementary 3D-CT imaging substudy will include 40 patients, with 20 patients allocated to each group. This substudy is designed as a single-centre exploratory mechanistic component to compare local anaesthetic contrast spread patterns between superficial and deep SCTL block. It is not powered for confirmatory efficacy analysis and will not contribute to the primary sample size calculation of the main clinical trial. Statistical analysis The primary outcome, defined as the incidence of hypotension during anaesthesia induction, will be analysed according to the intention-to-treat principle. A per-protocol analysis will be performed as a supportive analysis. In case of discrepant findings, the intention-to-treat analysis will be considered the primary inference. Because this is a multicentre randomised trial, the primary analysis will use a generalised linear mixed model with a logit link. Treatment group will be included as the main fixed effect, and study centre will be included as a random intercept to account for potential clustering by centre. An adjusted GLMM will also be performed by including prespecified baseline covariates, including age, sex, BMI, ASA class and baseline MAP, as fixed effects. These covariates were selected a priori based on clinical relevance rather than data-driven selection. Sensitivity analyses will be performed to assess the robustness of the primary findings, including per-protocol analysis, missing-data sensitivity analyses and ICC-based sensitivity analyses assuming plausible ICC values of 0.01–0.02. Missing data will be handled using multiple imputation under the missing-at-random assumption, and extreme-case best-case and worst-case scenarios will also be explored. Continuous secondary outcomes will be compared using independent t-tests or Mann-Whitney U tests, as appropriate. Categorical secondary outcomes will be analysed using χ² tests or Fisher’s exact tests. Time-to-event outcomes, if applicable, will be analysed using Cox proportional hazards models or mixed-effects Cox models. Exploratory analyses of hypotension burden, including the duration of hypotension and area under the MAP threshold of 65 mm Hg during induction, will be performed using continuously recorded arterial pressure data. No formal adjustment for multiplicity will be applied to secondary or exploratory outcomes; therefore, these analyses will be interpreted cautiously. A two-sided p value<0.05 will be considered statistically significant for the primary outcome. Statistical analyses will be conducted using R software. A detailed statistical analysis plan will be finalised before database lock and before completion of patient enrolment. Patient and public involvement Neither patients nor members of the public participated in the design, conduct, reporting or dissemination plans of this research. Ethics and dissemination This trial was approved by the Ethics Committee of Nanjing First Hospital (No. KY20240924-03-KS-01) on 24 September 2024, and then registered on the Clinical Trial Registry (https://clinicaltrials.gov, NCT06841822) on 21 February 2025. We expect to complete patient recruitment before December 31, 2025. All patients will provide their written informed consent (see online supplemental file 1). Patients can withdraw their consent at any time during the study. This study follows the Declaration of Helsinki. The study results will be published in a peer-reviewed journal. bmjopen-16-7-s001.pdf Discussion In this multicentre, randomised double-blind, prospective, parallel controlled trial, the objective of our study is to evaluate the effects of the deep layer SCTL block and the superficial layer SCTL block on hypotension during general anaesthesia induction in patients undergoing thoracoscopic pulmonary resection and to compare their postoperative analgesic efficacy. Our primary hypothesis is that the superficial-layer SCTL block, characterised by reduced initial diffusion of local anaesthetic agents into the thoracic paravertebral space and a slower pharmacokinetic profile, may attenuate sympathetic inhibition, thereby mitigating the risk of hypotension during anaesthesia induction. Furthermore, we hypothesise that, despite differences in diffusion kinetics, gradual permeation of local anaesthetics into the thoracic paravertebral space via the superficial-layer SCTL block is hypothesised to achieve postoperative pain control comparable to that of the deep-layer SCTL block. Emerging evidence underscores the critical association between intraoperative hypotension and postoperative morbidity. Both the severity and cumulative duration of hypotension exhibit progressive correlations with major complications, including myocardial injury, acute kidney injury and mortality.6–8 Notably, up to 34% of intraoperative hypotension episodes occur during the vulnerable window between anaesthesia induction and surgical incision, a period characterised by abrupt sympathetic inhibition and haemodynamic instability.7 This early-phase hypotension may disproportionately contribute to organ hypoperfusion due to the absence of compensatory surgical stimuli, amplifying its pathophysiological impact.9 Consequently, optimising haemodynamic stability during induction has emerged as a key preventive strategy in enhanced recovery protocols. We hypothesise that the superficial-layer SCTL block, by modulating local anaesthetic pharmacokinetics to achieve gradual thoracic paravertebral diffusion, may preserve sympathetic tone more effectively than the deep-layer approach. This mechanism-driven design aims to address the clinically urgent need to mitigate induction-phase hypotension while maintaining comparable postoperative analgesia—a dual objective aligned with current recommendations for haemodynamic precision management.9 The supplementary 3D-CT imaging substudy is intended to provide anatomical evidence to support the hypothesised mechanism underlying the different haemodynamic effects of the two SCTL block approaches. We speculate that deep SCTL block may allow more direct and extensive spread of local anaesthetic into the thoracic paravertebral space and adjacent neural structures, which may result in greater sympathetic blockade and more pronounced BP reduction during anaesthesia induction. In contrast, superficial SCTL block may produce a more limited or gradual spread pattern, thereby attenuating sympathetic inhibition while maintaining postoperative analgesic efficacy. Importantly, CT imaging cannot directly demonstrate sympathetic blockade. Therefore, the distribution of contrast-enhanced injectate will be interpreted as an anatomical surrogate for the potential extent of neural and sympathetic involvement. Any associations between imaging spread patterns and peri-induction haemodynamic outcomes, including hypotension occurrence and changes in MAP, SV, CI and SVI, will be considered exploratory and hypothesis-generating rather than confirmatory evidence. This study has several limitations. First, due to concerns regarding radiation exposure and contrast administration, 3D-CT imaging will not be performed in all participants in the main clinical trial. Instead, the imaging substudy will be restricted to patients who already require clinically indicated preoperative CT-guided lesion localisation at Nanjing First Hospital. Therefore, the imaging substudy is single-centre in nature and includes a relatively small sample size. Second, because the 3D-CT substudy is designed as an exploratory mechanistic component, it is not powered to provide confirmatory evidence regarding imaging-haemodynamic associations. Any observed relationship between contrast-enhanced local anaesthetic spread and peri-induction haemodynamic changes should therefore be interpreted cautiously and considered hypothesis-generating. Third, CT imaging can visualise the anatomical distribution of contrast-enhanced injectate, but it cannot directly demonstrate the functional extent of sympathetic blockade. Thus, imaging findings will serve only as an anatomical surrogate for potential neural and sympathetic involvement. Supplementary Material Reviewer comments Author's manuscript Footnotes Contributors: YX, YL and HS contributed to conception and study design. YX, WZ and Hc drafted the manuscript. YL and XW made critical revision of the manuscript. ZF, ZQ and HY contributed to sample size calculation and statistic plan. XW and Hc acquired the research fundings. XW is the primary investigator and the guarantor. All authors agreed to be accountable for all aspects of the work and gave their final approval of this version to be published. Limited AI use for language editing. The authors used artificial intelligence–assisted language tools (ChatGPT) solely to improve the clarity and grammar of the manuscript written in English. The AI tool was not used for study design, data analysis, data interpretation, or generation of scientific content. All scientific decisions, data verification, and final responsibility for the manuscript remain with the authors. Funding: This work was supported by the Scientific Research Project of the Talent Development Fund of Geriatric Hospital of Nanjing Medical University, IR2024104, and the Nanjing Medical Science and Technology Development Key Project, ZKX21038. Competing interests: None declared. Patient and public involvement: Patients and/or the public were not involved in the design, or conduct, or reporting or dissemination plans of this research. Provenance and peer review: Not commissioned; externally peer reviewed. Supplemental material: This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise. Ethics statements Patient consent for publication Not applicable. References Associated Data Supplementary Materials bmjopen-16-7-s001.pdf Reviewer comments Author's manuscript