Ethics approval, registration and patient selection

This is a prospective, randomised, double-blind study taking place at Nantong University Hospital and Zhejiang Provincial People's Hospital from May 2020 to April 2021. The study was approved by the Ethics Committee of the Affiliated Hospital of Nantong University (ethical approval number: 2019-K094) and registration was completed with the China Clinical Trials Centre (registration number: ChiCTR2000032693). Written informed consent was obtained from all participants before the conduct of this study. This trial was conducted in accordance with the Declaration of Helsinki. And this manuscript adheres to the Consolidated Standards of Reporting Trials (CONSORT) guidelines.

A total of 120 American Society of Anesthesiologists (ASA) Class I-II surgical patients between the ages of 16 and 65 years were recruited for this study, and the procedure was performed under general anesthesia with nasal intubation for maxillary cyst excision, and the maxillary cysts were all less than 5 cm in diameter. Participants without recent hepatic or renal insufficiency, severe allergic or hypersensitivity reactions to relevant drugs, cardiovascular or neurological disease, pregnant women or patients with airway difficulties, obesity and those taking opioids were excluded. We randomised 120 patients into groups D (dexamethasone group) and C (control group) using a random number table and the results of the randomisation grouping were sealed in opaque envelopes until the pretreatment drugs were prepared. Neither the patients nor the anaesthetists involved in the study were aware of the results of the random grouping.

Study procedures

All patients did not receive any preoperative treatment and were monitored in the operating room for non-invasive blood pressure (BP), Electrocardiogram (ECG), peripheral oxygen saturation with Surgical Pleth Index (SPI) and electroencephalographic bispectral index (BIS). Each patient was induced with sufentanil 0.3 µg·kg−1, propofol 2–2.5 mg·kg−1 and cis-atracurium 0.2 mg·kg−1. Anesthesia was maintained by a combination of intravenous and inhalation methods, 1.0% sevoflurane by inhalation in all patients, and intravenous infusion of remifentanil 6–12 µg·(kg·h)−1 and isoproterenol 3–5 mg·(kg·h)−1, adjusted according to hemodynamic parameters which fluctuated around 20% of basal values of blood pressure and heart rate that the basic value was the average value of blood pressure and heart rate measured three times (5 min apart every time) in a calm state after the patient entered the operating room. During surgery, PetCO2, BIS and SPI values were maintained at 35–45 mmHg (1 mmHg = 0.133 kPa), 40–55 and 30–50 respectively. We adjusted the speed of propofol at 0.5 mg·kg−1 each time if BIS values were out of our target range and the speed of remifentanil at 1 µg·kg−1 each time if SPI values were out of target range, if BIS and SPI values were within normal range but hemodynamic parameters were below or beyond than 20% of the basic values, we adjusted BP by intravenous Norepinephrine 20 µg each time or Urapidil Hydrochloride 5 mg each time and HR by intravenous Atropine 0.5 mg each time or Esmolol Hydrochloride 10 mg each time to our target range with repeated injection. Ten minutes before the start of operation, all patients were given intravenous hydromorphone 0.02 mg·kg−1 to anticipate postoperative analgesia, meanwhile, patients in group D received intravenous dexamethasone 0.2 mg·kg−1 and patients in group C received intravenous equal doses of saline.

Outcome measures

The primary indicators for this study were to evaluate the pain intensity and occurrence of painful event, including the resting pain and active pain in the 48 h postoperative period. Secondary indicators included assessing the facial swelling and restricted mouth opening of the patients in the 48 h postoperative period, and monitoring the patients' adverse effects, such as changes in blood glucose. Another anaesthetist, who was unaware of the intervention, performed the outcome assessment.

VAS (Visual Analogue Scale) pain scale

A 10 cm horizontal line is drawn across the top of the paper, with 0 at one end of the line indicating no pain, 10 at the other end indicating extreme pain, and the middle section indicating varying degrees of pain. The patient is asked to mark a mark on the horizontal line to indicate the degree of pain according to his or her self-perception, and the length is measured. In clinical practice, it was called moderate and severe pain when we measured the patient's VAS score > 3. And in our study, we defined the patient as having a painful event if the score above 3 points after surgery, we calculated for each patient the occurrence of painful event throughout the repeated measurements at 2 h, 6 h, 12 h, 24 h and 48 h postoperatively.

Facial swelling grading

Reference and improvement of Daniel Lim [9] study method, specific approach: first measure the distance from the corner of the mouth to the earlobe on the extraction side (a), the distance from the earlobe to the mandibular angle (b) and the distance from the external canthus to the mandibular angle (c) respectively (Fig. 1), calculate the facial measurement distance X = [(a + b)/2 + c]/2, and then calculate the facial swelling percentage, the calculation formula is [postoperative facial measurement distance (X1)—preoperative facial measurement distance (X0)] / preoperative facial measurement distance (X0)*100%. The facial swelling was assessed according to the facial swelling percentage, and the criteria: Grade 0, swelling area < 3%; Grade I, swelling area 3 to 6%; Grade II, swelling area 6 to 12%; Grade III, swelling area > 12%. In statistics, we considered Grade 0 and Grade I as light facial swelling while Grade II and Grade III as heavy.

Fig. 1

Diagram of facial swelling measurement. a The distance from the corner of the mouth to the earlobe on the extraction side. b The distance from the earlobe to the mandibular angle. c The distance from the external canthus to the mandibular angle. Calculating the facial measurement distance (X) according to formula X = [(a + b)/2 + c]/2, the X value was taken as the average value of facial measurements in millimeters in three times. And then calculating the facial swelling percentage = [postoperative facial measurement distance (X1)—preoperative facial measurement distance (X0)] / preoperative facial measurement distance (X0) * 100%

The facial swelling was assessed according to the facial swelling percentage, and the criteria were: Grade 0, swelling area < 3%; Grade I, swelling area 3 to 6%; Grade II, swelling area 6 to 12%; Grade III, swelling area > 12%. we considered Grade 0 and Grade I as light facial swelling while Grade II and Grade III as heavy.

Grading criteria for trismus

The distance between the incisal margins of the maxillary and mandibular central incisors is measured with vernier calipers. Grade 0, spacing > 2.5 cm; Grade I spacing 2–2.5 cm; Grade II, spacing 1–2 cm; Grade III, spacing < 1 cm. In statistics, we considered Grade 0 and Grade I as light trismus while Grade II and Grade III as heavy.

Sample size

The sample size estimation was based according to the active pain intensity 24 h after operation. Our preliminary study found that the mean VAS of the group C and the group D were 4.55 ± 0.98 and 3.90 ± 0.74 (10 patients each group). A power analysis was done with the use of G* Power 3.1.9.7 software and an effect size of d = 0.75 was calculated. A sample size of 48 per group to achieve a power of 95% and a type I error of 5%. To compensate for the possibility of dropout, we eventually recruited a total of 120 patients.

Statistical analyses

All statistical analyses were performed with SPSS version 20. The Shapiro–Wilk test was applied to assess the normality of the data. To verify the homogeneity of variance, a Levene test was conducted. Quantitative variables with Gaussian distribution were presented as the mean ± standard deviation (SD) or mean with 95% confidence intervals and nonparametric data as a median and interquartile range (IQR). The statistical significance of differences between groups was analysed using the independent t-test to the variables with Gaussian distribution, and analysis of Mann–Whitney U test was used to analyse the non-parametric values. We conducted a linear mixed model (LMM) analysis to compare the difference of postoperative pain between two groups as we performed repeated measurements at different time points postoperatively. Occurrence of painful event was performed with a median with range by Graphpad prism 6.0 and analysed by Mann–Whitney U test. Categorical variables were expressed as number (proportion) and analysed by Pearson χ2 test or Fisher exact test, such as ASA classifications, gender, facial swelling and mouth opening level. All figures were plotted with Graphpad prism 6.0 statistical software. A statistically significant difference was determined at a P value < 0.05.