Efficacy of Different Dose of Dexmedetomidine Combined With Remifentanil in Colonoscopy

A Randomized Controlled Trial

Li Jia; Meng Xie; Jing Zhang; Jingyu Guo; Tong Tong; Yuying Xing


BMC Anesthesiol. 2020;20(225) 

In This Article


Study Design and Patients

This was a prospective trial that was carried out in patients who were scheduled to undergo colonoscopy at the Fourth Hospital of Hebei Medical University between February 2018 and October 2018. All patients were inpatients. This study was approved by the Medical Ethics Committee of the Fourth Hospital of Hebei Medical University (2017MEC113) and written informed consent was obtained from all subjects participating in the trial. The trial was registered at the Chinese Clinical Trial Registry (ChiCTR2000029105, Principal investigator: Li Jia, Date of registration: 2020–01-13). This study adheres to CONSORT guidelines.

The inclusion criteria were: 1) ASA grade was I-II; 2) 18–75 years of age; 3) weight of 50–80 kg; 4) no obvious abnormalities in preoperative ECG, blood routine, electrolytes, and other tests; 5) no history of allergies to narcotic drugs; 6) no history of sedation, analgesics, or alcohol abuse; and 7) no mental illness. The exclusion criteria were: 1) emergency patients; 2) severe abnormalities in heart, lung, kidney, liver, and other functions; 3) sleep apnea syndrome or difficult airways; 4) bronchial asthma; or 5) recent respiratory infections.

Grouping and Intervention

The patients were randomly divided into three groups using the random number table method: group I (DEX 0.2 μg·kg− 1), group II (DEX 0.3 μg·kg− 1), and group III (DEX 0.4 μg·kg− 1) (Figure 1). Patients, surgeon and postoperative observes were blind to group allocation. The routine preoperative preparation was performed. The patient was placed on the left side with the knees bent after entering the room. The Bene View T5 monitor (Mindray Biomedical Electronics Co., Shenzhen, China) and Aspect 2000 EEG monitor (Aspect Medical Systems, Inc., Newton, MA, USA) were connected to monitor the systolic blood pressure (SBP), diastolic blood pressure (DBP), heart rate (HR), pulse oximetry (SpO2), respiratory rate (RR), and bispectral index (BIS). Oxygen mask inhalation was given at 5 L/min. An upper limb venous access was opened. For group I, 0.2 μg·kg− 1 DEX (batch number: 10122334, Jiangsu Hengrui Pharmaceutical Co., Ltd.) and a loading dose of 1 μg·kg− 1 remifentanil (batch number: 6120721 Yichang Renfu Pharmaceutical Co., Ltd.) was injected successively, both within 2 min with an intravenous pump. Then, remifentanil was given at a maintenance dose of 0.1 μg·kg− 1·min− 1. Colonoscopy started after 2 min. The infusion of remifentanil was stopped after the end of colonoscopy. For group II, the patients were injected intravenously with 0.3 μg·kg− 1 DEX. The patients in group III were injected intravenously with 0.4 μg·kg− 1 DEX. The dose of remifentanil was the same in all three groups. If bradycardia occurred during the examination (HR < 50 beats/min), atropine 0.5 mg was injected intravenously. Ephedrine 5–10 mg was injected intravenously if hypotension (SBP < 90 mmHg) appeared. And after the colonoscopy, patients were monitored in PACU for 30 min and transferred to the ward.

Figure 1.

Flow Diagram

Data Collection

The data of patients, including SBP, DBP, HR, SpO2, and RR at admission (T0), immediately after giving DEX (T1), at the beginning of the examination (T2), 5 min after the beginning of the examination (T3), 10 min after the beginning of the examination (T4), and at the end of the examination (T5), and the BIS value at T0, T1, and T2, and the minimum value of BIS during the examination were recorded. We investigated whether the patient was cooperating quietly and whether there were body movements during the examination; whether the patient could wake up; if the patient was needed to turn and whether the patient could do it on his own to complete the examination; number of patients with slight limb activity who did not need additional medications; number of patients with great limb mobility who were unable to cooperate quietly and complete the examination and required additional medications (intravenous injection of 50 mg propofol); number of patients who could wake up; and number of patients who could turn over the body on their own and cooperate to the examinations were recorded. Adverse reactions such as bradycardia, hypotension, and nausea and vomiting were recorded. The examination duration was also recorded. At the end of the examination, the satisfaction degree of the surgeon on the anesthesia effect, which was divided into three grades of excellent, good, and poor, were investigated. Excellent: the patient was quiet during the examination, had no limb movement, was able to wake up during the operation, and was able to turn over the body to cooperate with the examination. Good: the patient had only slight limb movement, which did not affect the examination, was able to wake up during the operation and was able to turn over the body to cooperate with the examination. Poor: the patient had a large degree of limb activity, and it was difficult to complete the examination quietly or cooperatively, or the patient had no limb activity, but could not wake up, or could wake up but could not turn over the body to cooperate with the examination. All examinations were performed by a senior doctor in the endoscopy department of our hospital.


The primary outcomes of this study were the patient's body movements during the procedure and adverse events such as bradycardia, hypotension, nausea, and vomiting. The secondary outcomes were the duration of colonoscopy and the satisfaction of the surgeon to the anesthesia effect.

Statistical Analysis

The sample size was calculated based on the patient's movements during the examination. Our preliminary study found that the incidence of patient movements in 11 patients with remifentanil for analgesia was 55%. A reduction of 25% after combined with dexmedetomidine was considered clinically significant. Therefore, a minimum sample size of 46 patients for each group would be required with a significance level of 5% to achieve a power of 80%. Taking into consideration a potential dropout rate of 10%, we recruited 50 patients per group. Normally distributed continuous variables were presented as mean ± standard deviation and were analysed using Student's t test. Mann-Whitney U test was used for non-normally distributed continuous variables, which were presented as median (interquartile range) [M(Q)]. Categorical variables were expressed as frequency (percentage) and were analysed using the Pearson chi-square test. Wilcoxon rank sum test was used for comparison of rank variables. p < 0.05 was statistically significant.