This prospective, randomized, single-blind study was approved by the Medical Ethics Committee of Ondokuz Mayis University in February, 2023 (Approval No. 2022/324). This study was registered at ClinicalTrials.gov (Registration No. NCT05822479) before patient enrollment and was conducted in accordance with the principles of the Declaration of Helsinki (64th WMA General Assembly, Fortaleza, Brazil, October 2013). The Consolidated Standards of Reporting Trials (CONSORT) 2010 Statement was also considered when preparing this manuscript [21].

This study included patients scheduled for elective LSG at a tertiary university hospital between April, 2023 and December, 2023. Written informed consent was obtained from all participants before the study. The inclusion criteria were as follows: patients aged 18–65 years with a body mass index > 35 kg/m² and an American Society of Anesthesiologists physical status score of II-III. Exclusion criteria were as follows: patients with drug allergies (opioids, NSAIDs, LA); neuropsychiatric disorders; cognitive impairment; inability to communicate; history of drug addiction; injection site or systemic infection; use of anticoagulants or bleeding disorders (coagulopathy, abnormal international normalized ratio, thrombocytopenia); significant cardiovascular, hepatic, renal, or endocrine disorders; history of chronic pain syndrome or long-term opioid use (> 4 weeks); obstructive sleep apnea syndrome (OSAS; apnea-hypopnea index > 5/h); chronic obstructive pulmonary disease; and requiring postoperative intensive care follow-up.

Study randomization was performed by an anesthesia assistant who was not involved in patient follow-up. The patients were randomly assigned to the control or EOIB group (n = 30 each). Randomization was performed using a web-based program called Research Randomizer with a 1:1 ratio and a block randomization list (block size = 4) [22]. Sequentially numbered opaque sealed envelopes containing participation numbers were stored in a locked cabinet in the operating room until the day of surgery. One hour before surgery, an anesthesia nurse who was not involved in the randomization process asked the patients to choose one of the envelopes and determine the patient group. The block procedure was performed by an anesthesiologist who was not part of the study in the regional anesthesia room 30 min before the patients underwent surgery. To ensure block quality and standardization, the procedure was conducted by an experienced anesthesiologist who had successfully performed an EOIB at least 30 times. An anesthesia assistant who was not involved in the study performed postoperative pain assessments and data collection. The surgeons, nurses, anesthesiologists, and individuals involved in the analysis of the results were all blinded to patient randomization.

Preoperatively, in the regional anesthesia room, patients scheduled for the EOIB were monitored using electrocardiography, noninvasive arterial blood pressure, and peripheral oxygen saturation. Oxygen was administered at a flow rate of 3 L/min via a nasal cannula. After initiating the intravenous (IV) lactated Ringer’s infusion through a 20–22 gauge IV cannula, intravenous midazolam was administered at a dose of 0.03 mg/kg (ideal body weight [IBW]) to achieve a Ramsey Sedation Score of 2 (awake, calm, and observing their surroundings).

The EOIB was performed as described by Elsharkawy et al. [19]. Briefly, the patients were placed in the supine position, and the area between the midclavicular and anterior axillary lines at the sixth to tenth rib levels was prepared following the asepsis/antisepsis rules. After covering the linear ultrasonography probe (8–13 MHz GE LOGIQ V1 Ultrasound System^®) with a protective sterile plastic sheath, the probe orientation marker was directed cranially and positioned sagittally between the midclavicular and anterior axillary lines at the level of the sixth rib. The sixth rib was identified by counting upward from the subcostal margin (tenth rib). After positioning the probe, the cranial end was slightly internally rotated to obtain a paramedian sagittal oblique view. The following structures were identified from superficial to deep: skin-subcutaneous tissue, external oblique intercostal muscle, intercostal muscles, pleura, and lungs. The needle entry site was determined to be cranial to the sixth rib and medial to the anterior axillary line. Subsequently, 2 ml of 2% lidocaine (Aritmal^®, Vem İlaç) was infiltrated into the skin and subcutaneous tissues. A block needle (21-G, 100-mm short bevel; Stimuplex® Ultra 360®, B. Braun) was then advanced in a superomedial-inferolateral direction for 2–3 cm into the external oblique intercostal muscle, reaching the plane between the sixth and seventh ribs. Thereafter, 1–2 ml of 0.9% normal saline was injected to confirm correct placement. After hydrodissection, 30 ml of LA mixture consisting of 0.25% bupivacaine (Marcaine^®, AstraZeneca) + 1:400000 adrenaline was injected into the plane in 5-ml increments with negative aspiration after each increment. LA spread over the sixth rib between the sixth and seventh ribs was observed. The needle was directed toward the eighth rib to enhance the spread of the LA (Fig. 1 and Supplementary Video 1). The same procedure was repeated on the opposite side. The resulting sensory block was checked every 5 min using the pinprick test, covering the T6–T10 dermatomes at the posterior axillary line of the lateral cutaneous nerves and the T6–T9 dermatomes at the midline of the anterior cutaneous nerves (Fig. 2). An anesthesia assistant who was not involved in the EOIB application or patient follow-up performed the dermatome examination. The EOIB was considered successful for patients with a sensory block score ≥ 1 (0 = no sensory block; 1 = touch sensation present, no pain; 2 = no touch sensation or pain) in all dermatomes, and the remaining patients were excluded from the study.

In the control group, no regional block was performed. Patients received standard analgesia consisting of intraoperative and postoperative multimodal analgesic medications as detailed in the 'Analgesia Management' section.

After administration of the block, the patients were taken to the operating room. Anesthesia was administered as follows: induction was performed with propofol (1.5–2.5 mg/kg lean body weight [LBW]), and remifentanil (1 μg/kg LBW IV bolus over 30–60 s, followed by an infusion of 0.25 μg/kg/min). Endotracheal intubation was performed using rocuronium bromide (0.6–1.2 mg/kg LBW IV) to achieve a zero train-of-four (TOF) count. During the operation, neuromuscular blockade was maintained with repeated bolus doses of 0.1–0.2 mg/kg LBW IV rocuronium to keep the TOF count at 1–2 [23]. Maintenance of anesthesia was achieved with O₂/air (inspired oxygen fraction 40%), age-adjusted 0.8–1.2 minimum alveolar concentration sevoflurane, and 0.1–0.25 µg/kg/min IV remifentanil infusion. The heart rate and blood pressure were allowed to vary by no more than 20% from the preoperative values by adjusting the remifentanil infusion rate. At the end of the procedure, extubation was performed after reversing rocuronium with a combination of 0.02 mg/kg IV atropine and 0.04 mg/kg IV neostigmine to achieve a TOF ratio > 0.9.

During the pre-anesthesia visit, patients were informed about the numerical rating scale (NRS), an 11-point scale ranging from 0–10 (where 0 indicates “no pain” and 10 indicates “the worst pain imaginable”). The patients were instructed to select the most appropriate value corresponding to their pain intensity during the postoperative period. Patients were also educated on how to use the patient-controlled analgesia (PCA) device and were informed that they could request opioids using the PCA device if their NRS score at rest was ≥ 4.

Following induction, all patients were administered 20 mg IV tenoxicam and 100 mg tramadol during the intraoperative period. At the time of surgical incision, 0.05 mg/kg IBW IV morphine was administered. Thirty minutes before the end of the procedure, 1 g of IV paracetamol was administered, followed by subsequent doses every 8 h postoperatively. Additionally, postoperative PCA (BodyGuard 575 Pain Manager, Caesarea Medical Electronics GmbH) was initiated in the recovery unit for all groups (requested dose: 20 μg/kg IV morphine, lockout interval: 10 min, 4-h limit: 80% of the total calculated dose, basal infusion: none). For rescue analgesia (patients with an NRS score ≥ 4 at rest despite PCA requests), 25 mg IV meperidine was administered in the recovery unit, and 50 mg intramuscular (IM) meperidine was administered in the ward. The doses of IM and IV meperidine for rescue analgesia were calculated as IV morphine milligram equivalents (MME) using an opioid equivalent analgesic dose calculator (http://www.globalrph.com/narcotic). NRS scores were evaluated at 0 (upon communication with the patient), 3, 6, 12, 18, and 24 h postoperatively, both at rest and during coughing or deep breathing. The calculated MME amounts of IM and IV meperidine were added to the morphine requested from the PCA to record the cumulative opioid consumption at 12 and 24 h.

For postoperative nausea and vomiting (PONV) prophylaxis, patients were routinely administered 8 mg IV dexamethasone before induction and 0.15 mg/kg IBW IV ondansetron 20 min before the end of the procedure. PONV was evaluated using a verbal descriptive scale (0 = none; 1 = mild nausea; 2 = moderate nausea; 3 = vomiting once; 4 = vomiting more than once). Patients with a score ≥ 3 were administered an additional 4 mg IV ondansetron. If the score remained ≥ 3 despite the administration of ondansetron, 10 mg IV metoclopramide was administered.

Patient satisfaction and the quality of pain management in all patients from both groups were assessed using the Turkish version of the American Pain Society Patient Outcome Questionnaire (APS-POQ-R-TR). The questionnaire consists of 18 primary and 3 secondary questions. These questions aim to evaluate pain intensity, frequency of severe pain, impact of pain on daily activities and emotional well-being, side effects of treatment, patients’ perceived analgesic benefit, level of engagement in treatment, non-pharmacological methods used to alleviate pain, and overall patient satisfaction [24].

After the induction of anesthesia, the patients were placed in the supine position. The first trocar (12 mm) was inserted approximately 20 cm below the xiphoid process, and CO₂ gas at a pressure of 13–15 mmHg was used to expand the abdominal cavity. Subsequently, four additional trocars, ranging in size from 5 to 15 mm, were placed, and the surgical procedure was performed transperitoneally with the patient in the reverse Trendelenburg position. Following surgery, a drain was inserted through a 5 mm port in the left anterior axillary line (Supplementary Fig. 2).

The primary outcome was the cumulative IV MME consumption in the first 24 h postoperatively. The secondary outcomes included cumulative IV MME consumption in the first 12 h postoperatively; NRS scores for rest/activity at 0, 3, 6, 12, 18, and 24 h; intraoperative remifentanil consumption; number of patients requiring rescue analgesia; time to first analgesic request from the PCA; nausea and vomiting scores; number of patients requiring antiemetics; and revised APS-POQ-R-TR scores at 24 h postoperatively. Hemodynamic data, block-related complications (hematoma, infection, pneumothorax, and LA toxicity), and opioid-related side effects (itching, fatigue, sedation, and respiratory depression) were also recorded.

The sample size was determined by considering the cumulative opioid consumption (in MME) within the first 24 h postoperatively in a preliminary pilot study that included 10 patients per group. The sample size was calculated using G*Power software, version 3.1.9.7. The minimum number of patients to be included in each group was 25 with a 95% CI, 80% test power (1-β), and 0.8253421 effect size. The mean ± SD were 20.58 ± 4.75 mg in the control group and 15.72 ± 6.84 mg in the EOIB group. Considering the possibility of missing data, the sample size was increased by 20%. Thirty patients were thus included in each group, resulting in a total of 60 patients.

Statistical analyses were performed using SPSS software, version 26 (IBM Inc.). Descriptive statistics for categorical variables were presented as frequencies (n) and percentages (%). Comparisons between categorical variables were performed using Pearson’s chi-square or Fisher’s exact tests, as appropriate. The Shapiro–Wilk test was used to assess the normality of the distribution of the numerical variables. Descriptive statistics for numerical variables were presented as the mean ± SD and 95% CIs for normally distributed data and median values with interquartile ranges (Q1, Q3) for non-normally distributed data. An independent sample t-test was used to compare two independent groups of normally distributed data.

The Mann–Whitney U test was used to compare two independent groups with non-normally distributed data. For pain scores at rest and with activity, two-way repeated-measures analyses of variance were performed, which accounted for each group, the measurement times (0, 3, 6, 12, 18, 24 h), and the interaction term between the group and measurement time. For pairwise comparisons with further testing, a Bonferroni correction was applied, and the significance level was accepted as P < 0.0083. Statistical significance was defined as P < 0.05 with two-tailed.