This was an investigator-initiated, prospective, single-center, double-blind, randomized, placebo-controlled trial. The trial protocol was approved by the Ethics Committee of Tongji Medical College, Huazhong University of Science and Technology (TJ2020S143; date of approval: 20/09/2020) and the written informed consent was obtained from all subjects participating in this trial. The trial was registered prior to patient enrollment at http://www.chictr.org.cn (number: ChiCTR2000038798; date of registration: 02/10/2020). An independent data and safety monitoring committee oversaw the study conduct and reviewed blinded safety data. This study was conducted between October 12, 2020 and September 20, 2022 at Tongji Hospital affiliated to Tongji Medical College of Huazhong University of Science and Technology according to the Helsinki manifesto. This manuscript adheres to the applicable CONSORT guidelines.

Patients were eligible for participation if they were aged between 18 and 64 years, and scheduled to craniotomy. American Society of Anesthesiologists (ASA) Physical Status of class I, II, or III was included. The exclusion criteria included emergency operation, coagulation disorders, neuropathic disorders, allergy to local anesthetics, any chronic pain conditions, chronic opioid usage, alcohol or drug addiction, psychiatric disorders, severe liver and kidney dysfunction, history of craniotomy, pregnant and lactating women, inability to understand pain scoring, preexisting nausea and vomiting, abnormal surgical scalp sensation, and participating with other studies within 30 days.

Enrolled patients were randomly assigned to either the LOGAB group or the control group in a 1:1 ratio. Randomization numbers were computer-generated and concealed in sealed, opaque envelopes. One researcher opened the envelope after patient enrollment, and prepared the drugs (0.5% ropivacaine or normal saline), and delivered them to the anesthesiologist for block who was blinded. Physicians responsible for intraoperative management, data collection and postoperative follow-up was blinded to group assignments.

The ultrasound-guided unilateral lesser occipital nerve combined with great auricular nerve block was performed by experienced J Liu. After induction, superficial cervical plexus (C4 level) was scanned transversely by ultrasonography (Sonosite Edge II, Fujifilm, USA) with a 6 to 13 MHz ultrasound probe (linear array HFL38xi) at the midpoint of the posterior margin of sternocleidomastoid muscle (Fig. 1B). Then, the probe slid toward the cephalic side 1 ∼ 2 cm to acquire clear image of both lesser occipital nerve and greater auricular nerve (Fig. 1C). Under aseptic conditions, a 22-gauge 50-mm needle (B. Braun Stimuplex D, Meisungen, Germany) was advanced twice into the subcutaneous space using “in-plane” method, outside the investing fascia (Fig. 1D and E). Water separation was utilized throughout the insertion to avoid nerve damage. Correct insertion of needle was verified by enhancement of nerve bundles after injection. A repeat negative aspiration test was performed during injection.

5 ml of 0.5% ropivacaine in the LOGAB group and an equal volume of 0.9% normal saline in the NSB group was injected through the posterolateral approach. The blocking procedure was completed within 10 min in both groups. Patients in both groups received local infiltration with 0.5% ropivacaine at head fixation sites before surgery.

No premedication was given before surgery. Patients were fitted by standard monitoring including heart rate (HR), continuous pulse oximetry, invasive arterial blood pressure (IABP) via arteria dorsalis pedis, continuous capnography, body temperature and bispectral index (BIS). General anesthesia was induced with intravenous propofol (2 mg ∙ kg–1) and sufentanil (0.4 μg ∙ kg–1). Rocuronium (0.9 mg ∙ kg–1) was intravenously administrated to facilitate endotracheal intubation. Dexamethasone (10 mg) and flurbiprofen axetil (50 mg) were intravenously administrated immediately after induction. Mechanical ventilation was initiated with a tidal volume of 6 ∼ 8 ml ∙ kg–1 in air-oxygen mixed gas (60% oxygen) and a respiratory rate of 12 ∼ 15 per minute to maintain EtCO2 between 35 and 40 mmHg. General anesthesia was maintained with sevoflurane inhalation at 0.5 minimal alveolar concentration (MAC) and a continuously intravenous infusion of remifentanil (0.1 ∼ 0.5 μg ∙ kg–1 ∙ min–1) to maintain BIS values between 40 and 60. The remifentanil rate was adjusted when MAP and/or HR fluctuated beyond 20% of baseline levels. Sufentanil (0.1 ∼ 0.2 μg ∙ kg–1) was remedially administered to relieve potent surgical stimulations. Clinical data including operation and anesthesia time, urine volume, blood loss, incision length were recorded. Tropisetron (2 mg) was intravenously administered before extubation. The intensity of postoperative pain was assessed using numeric rating scale (NRS) score after surgery. If NRS score > 4, a bolus of anal diclofenac (12.5 mg) was used as the remedial medication to avoid postoperative opioid interference with neurological evaluation. Time of diclofenac administration and the total dosage were recorded.

Primary outcome was the NRS scores at rest and motion within 48 h after surgery. The NRS score was assessed using a measuring ruler of 10 cm length. All patients received NRS scoring training during the preoperative consultation: 0 for completely painless and 10 for unbearable worst pain. Postoperative pain was assessed at extubation, and at six different time points (2, 4, 6, 12, 24, 48 h) after surgery.

Secondary outcomes included MAP and HR data from several intraoperative time points: 1 min before and after anesthesia induction; 1 min before and after skin incision; 3 min, 5 min, 10 min after skin incision; 1 min before and after dural incision; 3 min, 5 min, 10 min after dural incision. Besides, the intraoperative opioid consumption including the average remifentanil rate, total dosage of remedial sufentanil the time of diclofenac administration, the total dosage of drug and RSS score within 48 h after surgery were recorded as secondary outcomes either. RSS score was documented through the following grading: (1) anxiety; (2) awake and quiet; (3) lethargy, but able to follow instructions; (4) falling asleep, but arousable; (5) falling asleep, slow response to verbal stimuli; (6) deep sleep, not arousable. Adverse events including hematoma, infection, dyspnea, and local anesthetics toxicity, postoperative nausea and vomiting were conducted at the same time as the postoperative NRS evaluation.

The sample size was estimated using Power Analysis and Sample Size Software V23.0 (PASS, Kaysville, Utah, USA). Previous studies have shown that within 48 h after surgery, pain scores with and without block were 2.0 ± 1.6 and 3.7 ± 2.4, respectively [22]. Therefore, using a unilateral α level of 0.05 and 80% power, we calculated that 28 patients per group would be needed. Considering 20% loss of follow-up or consent withdrawals, 62 patients were required.

Data was analyzed on an intention-to-treat basis using the Statistical Package for Social Sciences for Windows software version 23 (IBM SPSS, USA). No patients withdrew their consent before surgery. Therefore, patients who met the inclusion criteria were included in the final analysis.

The demographic and clinical characteristics of patients were described. Normally distributed continuous variables (age, weight, height, BMI, etc.) were expressed as mean with standard difference, and analyzed using the One-Way Analysis of Variance (ANOVA) or Mann-Whitney U test. The normality of data was analyzed using the Kolmogorov-Smirnov test. Categorical variables (American Society of Anesthesiologists Physical Status, diclofenac count, postoperative nausea and vomiting, etc.) were expressed as frequency (%), and analyzed using the pearson chi-square test or Fisher’s exact test. The NRS scores at rest and motion within 48 h were plotted for homogeneity of variances test. Other variables (time, volume, rate, length, drug dosage, etc.) were processed with normality test. The ANOVA was used if data was distributed normally, or nonparametric test Mann-Whitney U test was used. Then the area under the receiver operating characteristic curve (AUC) was estimated for assessing pain intensity and duration using GraphPad Prism V.8 (GraphPad Software, La Jolla, California, USA). The mean AUCs (0 ∼ 6 h, 6 ∼ 12 h, 12 ∼ 24 h, 24 ∼ 48 h) were compared using the ANOVA according to randomization. Changes in MAP, HR and its increment rate (growth rate after skin or dural incision) were estimated by AUCs and compared by ANOVA. MAP and HR difference at time point were further analyzed. A Kaplan-Meier survival curve was generated to evaluate the difference in the first administration of diclofenac after surgery, and hazard ratio (HR) was used to evaluate the benefit using log-rank test. All tests were two-tailed, and P value < 0.05 was considered as significant.