This was a prospective, single-center, sequential-design clinical trial using Dixon up-and-down methods. The first subject was administered a dose close to ED95, and the CR of the first subject was used to determine the dose for the second subject. When a positive cardiovascular response occurs, the fixed dose will be increased for the next patient. When the cardiovascular response is negative, the next patient dose decreases the fixed dose. This design does not have an accurate sample calculation method, but 20 to 40 sample sizes can form a stable estimate of the target dose [11,12,13]. In the up-and-down process, Dixon [14] suggested that the test could be terminated after a fixed number of crossovers (i.e., crossover between positive and negative reactions) and the results met the need for stable estimation [12]. Researchers generally believe that 6 crossovers are sufficient [12, 15]. Therefore, this trial will be terminated when the seventh crossover occurred. This study was approved by the Clinical Research Center of Shanghai Pulmonary Hospital (ID: 2022LY0416) and the Ethics Review Committee of Shanghai Pulmonary Hospital (ID: L22-261) and was registered on 30/11/2022(NCT05631028) in clinicaltrial.gov before patient recruitment. It was conducted in accordance with the principles of the Declaration of Helsinki. All patients provided written informed consent.

Patients who underwent partial pneumonectomy via thoracoscopy at Shanghai Pulmonary Hospital during December 2022 were included. The inclusion criteria were as follows: (1) Patients who underwent optional video-assisted thoracoscopic surgery (VATS), (2) those aged more than 18 years old, and (3) Those with an American Association of Anesthesiologists grade of I–III. The exclusion criteria were as follows: (1) Patients with a systolic pressure of ≥ 160 mmHg or a diastolic pressure of ≥ 110 mmHg or an HR of ≥ 110 beats/min in the operating room at rest, (2) those with long-term preoperative use of analgesia or sedation drugs, (3) those with pregnancy, lactation, pregnancy possibility, and planned pregnancy, (4) those with an allergy history of the test drug, and (5) those with mental illness or an inability to communicate normally.

The patients were divided into three groups based on the doses of pre-used Dex. The dose of Dex in group A was zero, suggesting that Dex was not used. The patients in group A were divided into the young (A-Y) group (aged 18–64 years) and the elderly (A-O) group (aged ≥ 65 years) by age. Group B received 0.5 µg/kg Dex over a period of 10 min, and group C received 1 µg/kg Dex over a period of 10 min.

To achieve the same anesthetic effect, older patients often require smaller doses. It is necessary for us to define the effective dose of drugs in the elderly population, so the population is divided into elderly and young people at the age of 65 years [16, 17]. Elderly patients often experience drowsiness and even loss of consciousness with group B or C doses of dexmedetomidine. So both groups B and C did not include elderly patients.

The blood pressure, ECG, and pulse oxygen saturation were routinely monitored by Philips Patient Monitor (IntelliVue MX800), venous access was established, and invasive monitoring was performed as needed after entering the operating room. Non-invasive cuff blood pressure was recorded for analysis in this study. The process was initiated with 10 min infusion of Dex [18](Hengrui Co., Jiangsu, China).

After the Dex infusion was completed, anesthesia induction was performed. Sufentanil(Hengrui Co., Jiangsu, China) 0.5 µg/kg was intravenously administered, followed by RT(Hengrui Co., Jiangsu, China) 2 min later. The starting dose of RT was 0.25 mg/kg in group A-Y, 0.15 mg/kg in group A-O, 0.15 mg/kg in group B, and 0.1 mg/kg in group C according to the pretest results. The dynamic adjustment unit was 0.01 mg/kg in each group. The starting dose was different for each group. To achieve the same anesthetic effect, older patients often require smaller doses. And, there is a synergy between the intravenous anesthetic drugs. The higher the dose of Dex, the smaller the required dose of RT. For different populations and scenarios of each group, the starting dose was different. For the interpretation of the pretest, an additional file shows this in more detail [see Additional file 1].

Next, 0.6 mg/kg Rocuronium(Xianju Co., Zhejiang, China) was administered following the loss of consciousness (MOAA/S ≤ 1). When the muscle relaxant was completely effective, double-lumen endotracheal intubation(Covidien Co.,Minneapolis,USA) was performed under direct vision, the catheter position was determined using fiberoptic bronchoscopy. Either a prolonged intubation process or an intubation failure will affect the results. Tracheal intubation was performed by an experienced anesthesiologist, with the aid of visualization. If the intubation process for more than 2 min or the intubation failed, the test will be terminated immediately.

Propofol(Fresenius Kabi Co., Beijing, China) was pumped at 4–8 mg/kg/h, Remifentanil(Hengrui Co., Jiangsu, China) was pumped at 0.05–0.1 µg/kg/min, and Rocuronium was intermittently added for anesthesia maintenance. On the ventilation side, a protective ventilation strategy with a low tidal volume (4–6 mL/kg) was used along with positive end-expiratory pressure ventilation and a pulmonary resuscitation strategy. The intraoperative infusion used sodium acetate Ringer’s solution and hydroxyethyl starch with the ratio was 2:1, and the speed was 5–8 mL/kg/h.

Ondansetron(Qilu Co., Shandong, China) 4 mg and Flurbiprofen axetil(Taide Co., Beijing, China) 50 mg were scheduled to be administered 20 min preoperatively. Following anesthesia, the patient-controlled analgesia pump was connected to the vein, and a volume of 2 mL of 1 µg/mL Sufentanil with a background dose of 2 µg/h was given. Patient-controlled analgesia with a locking time of 15 min and 0.5 mL per patient-controlled analgesia administration. Paracetamol tablets (Qilu Co., Shandong, China) 0.5 g were administered orally for remedial analgesia.

The primary outcome of this study was ED50 and ED95 of RT to inhibit the positive CR which was induced by double-lumen bronchial intubation during anesthesia induction after Dex be pre-used. The inhibition of positive CR by RT means that patients will not experience positive CR induced by intubation stimulation after the use of RT.

The secondary outcome was a change in CR with the evaluation time included the start of endotracheal intubation until three minutes after completion. A positive CR was defined as follows: MAP or elevation in the HR of ≥ 15% of the baseline or tachycardia (> 120 beats/min) or hypertension (systolic blood pressure > 180 mmHg) during intubation. Non-invasive blood pressure (systolic, diastolic, and MAP) and HR were measured at the following four time points: the baseline time (T1) when patients entered the operating room at rest; (T2) at the end of Dex pumping; (T3) at the end of RT injection, and (T4) 3 min within tracheal intubation, wherein MAP = diastolic + 1/3 (systolic-diastolic) and MAP and HR were calculated and compared. Sedation success was defined as success when MOAA/S ≤ 1, indicating that the patient was losing consciousness. Hypotension was defined as a systolic blood pressure < 90 mmHg or a MAP < 60 mmHg. Low HR was defined as a heart rate < 50 beats per minute.

Normally distributed data were expressed as mean ± standard deviation, non-normally distributed data were expressed as median (interquartile range) and counting data as number (percentage). SPSS 25.0 software(IBM Co., New York, USA ) was used for statistical analysis, and the Kolmogorov–Smirnov test was used to test the normality of the data. The measurement data were compared in pairs using analysis of variance (least significant difference) or a nonparametric method. The counting data were tested using the χ2 test based on whether the data met the normality and homogeneity of variance requirements.

Using the Dixon up-and-down method, terminate the trial when the seventh crossover occurred. Probit regression was used to fit the trial results, with drug dose as the dependent variable and the occurrence of positive results as the outcome. Fit the equation that is closest to the actual situation and use it to estimate the ED50, ED95, and corresponding 95% confidence interval(CI) for each group. This has also been used in previous studies [12, 19]. P < 0.05 was considered statistically significant.