This study was designed adhering to the Declaration of Helsinki and the CONSORT checklist, and was approved by the internal review board (03/30/2017; P2017/074; Comité Ethique, Erasme hospital, Brussels; Chairman: Jean-Marie Boeymans). The trial was registered on clinicaltrials.gov (NCT03110653; PI: Luc Barvais; date of registration: 03/31/2017). Written informed consent was obtained from each patient participating in the study.
Inclusion criteria were the following: male or female patients aged 18–65 undergoing thyroid surgery at the Erasme hospital, Brussels, ASA physical status of I-III, and a knowledge of French, English, or Dutch.
Exclusion criteria were the following: pregnancy, hypo or hyperthyroidism, gastro-duodenal ulcer, allergy or contraindications to one of the study drugs, renal insufficiency, liver insufficiency, neuropsychiatric disturbance, BMI > 30, history of drug and alcohol abuse, and preoperative analgesic drug use.
The main objective of this study was to evaluate immediate post-operative pain levels in patients who underwent two types of remifentanil withdrawal methods. The timing of the first demand of post-operative analgesic was compared in both groups. In parallel, post-operative morphine consumption was also compared between both groups, as well as numeric rating scale (NRS) values at rest and after a small head flexion. In this way, pain was evaluated after 0, 15, 30, 45, 60, 75, 90, 105, 120 min, and 4 h after admission to the post-anesthesia care unit (PACU), as well as 24 h post-operatively. A quality of recovery (QoR-40) questionnaire was given to each patient pre- and post-operatively to evaluate patient satisfaction . In the PACU, sedation was assessed in both groups via the Ramsay Sedation Scale .
Patients were randomized to one or the other group using a computer-generated randomization list (QuickCalcs program-GraphPad Software Inc., La Jolla, USA). For premedication, all patients received alprazolam (0.5 mg P.O.) one hour pre-operatively. Multi-parameter monitoring was used according to our institution’s protocol (ECG, pulse oximetry, non-invasive blood pressure every three minutes, Bispectral Index (BIS), and neuromuscular blockade monitoring utilizing the train-of-four ratio). In the case of blood pressure dropping 20% or more from initial baseline measurement, a bolus of intravenous ephedrine was administered.
Anesthesia was induced intravenously in both groups with remifentanil TCI (Minto model; 5 ng ml− 1; approximately 0.15 mcg kg− 1 min− 1) and propofol TCI (Schnider model; starting at 3 mcg ml− 1; adjusted to BIS levels between 45 and 55). Rocuronium (0.6 mg kg− 1) was then administered to facilitate tracheal intubation. After tracheal intubation, but pre-incision, remifentanil levels were lowered to 2 ng ml− 1 (approximately 0.065 mcg kg− 1 min− 1). Just before incision, remifentanil levels were increased to 5 ng ml− 1, and acetaminophen (1000 mg), diclofenac (1 mg kg− 1) and morphine (0.15 mg kg− 1) were administered as co-analgesics. Dexamethasone (10 mg) was administered to both groups. By the end of the surgery, ondansetron (4 mg) was given to all patients. Upon skin closure, remifentanil levels were progressively lowered to 3.5 ng ml− 1 (approximately 0.13 mcg kg− 1 min− 1). By the end of the surgery, propofol TCI was discontinued and remifentanil was kept at 2 ng ml− 1. Tracheal extubation was accomplished after adequate spontaneous ventilation was ensured, as well as the patient’s responsiveness to verbal commands.
The patients were then transferred to the PACU with the remifentanil TCI infusion pump running at 2 ng ml− 1. Back check valves and continuous saline infusion were used in the IV line in order to avoid backflow, dead space syndrome, or unintentional bolus.
According to pre-operative randomization, the remifentanil infusion syringe was then switched in the PACU to a new syringe, which was prepared by a colleague (WT) who was independent of the study. This post-operative syringe was either remifentanil diluted to 20 mcg/ml or NaCl 0.9%. PACU nurses and the primary anesthesiologist did not know whether their patients had received a newly prepared remifentanil or NaCl 0.9% syringe.
A gradual decrease of this infusion was then started in the PACU according to the following protocol: a reduction in infusion rate of 30% every 15 min (2 - > 1.4 - > 1 - > 0.7- > 0.5 - > 0.35 - > 0.25 - > 0 ng ml− 1). This gradual decrease was achieved in two hours.
While in the PACU, intravenous morphine (2 mg every five minutes) was administered if a patient’s pain was more than 3/10 on the NRS. All patients were monitored and under immediate medical surveillance, and therefore no limit was imposed on total morphine consumption. Additionally, acetaminophen (1000 mg every 6 h) and diclofenac (1 mg kg− 1 every 12 h) were administered to all patients. Patients stayed a minimum of four hours in the PACU and were discharged from the PACU once they were comfortable (e.g. NRS < 3/10, no post-operative nausea and vomiting, Ramsay scores of 2–3, no hematoma at the surgical site).
The total quantities of each drug used, as well as the different timings (induction, incision, end of surgery, and extubation times), were recorded. The primary anesthesiologist collected data pre-operatively and post-operatively (the first 24 h post-operation).
As the primary outcome of the study was to compare the first post-operative demand of analgesic drugs, we considered a difference of 35 min or more to be of clinical relevance. With a two-sided alpha level of 0.05 and power of 90% (standard deviation (SD) of 27 min), the study needed to be conducted on 26 patients with each group consisting of 13 patients.
Statistix 9.0 (Analytical Software, Tallahassee, USA) was used for statistical analyses. The results were presented as mean ± standard deviation. P values < 0.05 were considered to be statistically significant. The Mann-Whitney U test was used for the analysis of the first demand of analgesic drug data, morphine consumption data, QoR-40 data, and NRS score data at 24 h. NRS values and Ramsay scale scores in the PACU were analyzed using a two-way ANOVA for repeated measures.