This prospective, single-center, observational, cohort, two-parallel-arm study was conducted in accordance with the principles outlined in the Declaration of Helsinki [8] after obtaining approval from the Ondokuz Mayis University Clinical Research.

Ethics Committee (approval no: 2021/560) and the Ministry of Health of Turkey (22-AKD-181–03.07.2020). The study was registered at ClinicalTrials.gov (NCT05752734) before the enrollment of any participant, and written informed consent was obtained from all participants for the registration and publication of data. In the study, 30 patients who received dexamethasone prior to LSG (group D) and 30 patients who did not (group C) were included using a convenience non-probability sampling method. While both the researcher and outcome assessors were blinded, the patients were not blinded in our study.

Patients who met the following criteria were enrolled between February 2023 and June 2023: scheduled for LSG, body mass index (BMI) > 30 kg/m2, aged 18–65 years, and an American Society of Anesthesiologists (ASA) classification of II–III. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) checklist was followed [9].

Patients with diabetes mellitus, renal, cardiac, and liver diseases, psychiatric and neurological conditions, a history of alcohol and substance abuse, patients who have been using non-invasive positive airway pressure therapy at home (apnea–hypopnea index (AHI) > 5/h with symptoms as diagnosed OSA with polysomnography), and those who did not provide consent were excluded from the study.

During the preoperative visit, each patient was informed regarding the use of the patient-controlled analgesia (PCA) device and QoR-40 questionnaire. Patients were also briefed on the use of numerical rating scale (NRS) for pain severity (0, no pain; and 10, most severe pain).

A standard protocol created for this patient group was applied for management of anesthesia in all patients who underwent LSG in our hospital. Accordingly, certain patients received 8 mg IV dexamethasone 1 h before the surgery in the preoperative waiting area, based on the anesthesiologist’s choice. Patients who fasted overnight were placed in the ramp position on the operating table, and electrocardiography (EKG), oxygen saturation (SpO2), and non-invasive blood pressure monitoring were performed in accordance with ASA standards.

After monitoring, patients were first preoxygenized and then intubated by administering iv remifentanil infusion (0.1–0.25 mcg/kg/min), propofol (1.5–2 mg/kg), and rocuronium bromide (0.6 mg/kg) for muscle relaxation. The drugs were administered according to body weight, calculated using the formula [ABW: ideal body weight (IBW) + 0.4 × (actual body weight-IBW)].

Anesthesia was maintained with an oxygen–air mixture (FiO2 0.50), sevoflurane, and remifentanil infusion 0.1–0.25 mcg/kg/min. Mechanical ventilator settings were tidal volume 6–8 ml/kg IBW, inspiratory/expiratory ratio 1:2, and respiratory rate EtCO2 30–38 mmHg. The remifentanil infusion rate was set to allow up to a 20% change in the patients’ preinduction heart rate and mean blood pressure. After induction, tramadol 100 mg and tenoxicam 20 mg iv were administered, and iv morphine 0.05 mg/kg/IBW was added during the intraoperative period. Thirty minutes before the end of the surgical procedure, an intravenous infusion of paracetamol (1 g) and ondansetron (8 mg) was administered. At the end of the surgery, neostigmine (0.03 mg/kg) and atropine (0.015 mg/kg) were administered to reverse the effects of rocuronium, and the patient was extubated. When there was a change of > 20% in the hemodynamic parameters of the patients during the intraoperative period, sympathomimetic/sympatholytic drugs such as esmolol, nitroglycerin, noradrenaline, and adrenaline were administered.

PONV was evaluated using a verbal descriptive scale (0, none; 1, mild nausea; 2, moderate nausea; 3, vomiting once; 4, multiple vomiting). In the case of verbal scale > 2, metoclopramide 0.15 mg/kg was used as a rescue antiemetic. Patients with a modified Aldrete score of > 9 in the recovery unit were sent to the general surgery ward with a morphine PCA device with the following settings: bolus dose, 1 mg; lock-out time, 15 min, 4-h limit, 80% of the calculated total dose. During the follow-up of these patients, paracetamol 1 g was administered thrice a day at 8-h intervals. Patients were informed that they could request opioids through the PCA device if their NRS score was > 3. Patients with an NRS score > 4 despite opioid administration with the PCA device were administered tramadol 0.5 mg/kg/IBW. The 24-h morphine consumption was defined as the net amount of morphine consumed through PCA. This definition excluded any other opioids used perioperatively. The mobilization time was defined as the period between admission to the ward and when the patient could sit on a chair without assistance. Hospital discharge instructions were used to determine whether patients were ready for discharge.

Patient characteristics, demographic information, and other data were obtained from the electronic medical records and patient files. A resident who was not participated in the study performed all the assessment. The primary objective of our study was to measure the QoR-40 score 24 h postoperatively. The QoR-40 questionnaire consists of a total of 40 questions in 5 subdomains: physical independence (n = 5), patient support (n = 7), comfort (n = 12), emotional status (n = 9), and pain (n = 7). Each question is rated on a Likert scale ranging from 1 to 5, and the sum of these scores yields a total score ranging from 40 (worst recovery quality score) to 200 (best recovery quality score). Two groups were formed on the basis of the use of steroids for multimodal analgesia. Sampling was performed using a convenience non-probability sampling technique, taking into consideration the inclusion criteria. According to this sampling method, patients need to be willing to participate in the study initially. The assignment of consenting patients to the experimental group is entirely determined by the preference of the anesthesiologist in charge of anesthesia management.

All patients were administered the QoR-40 questionnaire at 24 h postoperatively. The secondary objectives of our study were to assess the number of opioid demand count via PCA device and consumption of morphine first 24 h postoperatively, the incidence of patient who need rescue analgesics and antiemetics, intraoperative complications, postoperative wound infection, time to mobilization, and time to discharge.

All patients were provided with analgesia control through a PCA device for 24 h. At the end of 24 h, patients were evaluated using the 40-item QoR-40 questionnaire. A resident who was not participated in the study performed all the assessment.

The sample size for inclusion was determined using power analysis. G Power 3.1.9.2 software was used for the t-test of independent samples, considering a previous study’s global QoR-40 scores (group treatment: 182.1 ± 12 vs. group control: 183.7 ± 9). With a significance level of 5%(α), test power of 80% (1-β), effect size of d = 0.790, and a two-tailed hypothesis, the minimum sample size required for each group was determined to be 27, resulting in a total of 54 participants [10]. Considering potential data loss, 60 patients were included in the study.

Descriptive statistics, frequencies, and other features were used for statistical analysis of the patient data, including all variables. Continuous data are presented as mean ± standard deviation (95% confidence interval (CI)) or median (Q1–Q3). Continuous variables were analyzed using Shapiro–Wilk and Kolmogorov–Smirnov tests to determine the normal distribution of data. Continuous and normally distributed variables were compared using Student’s t-test. Nonparametric tests were used when the data did not follow a normal distribution. Categorical variables were assessed using the chi-square test, and in some cases, Fisher’s exact test was applied. Logistic regression tests were used to analyze risk factors. Analyses were conducted using SPSS Statistics for Windows (version 21.0; IBM Corp., Armonk, NY, USA). All p-values were two-tailed, and p ≤ 0.05 was considered statistically significant. Non-normally distributed quantitative data included subscores for emotions, physical independence, patient support, pain, BMI, morphine consumption in the first 24 h, and morphine demand count.