This prospective, randomized, double-blind, non-inferiority clinical trial was conducted at BP Koirala Institute of Health Sciences between January 2015 and April 2016. The study was approved by the Institutional Ethical Review Board (Ref: IERB 284/014) and the trial was registered prior to patient enrollment at clinicaltrials.gov (NCT02327923). The study was performed according to the Declaration of Helsinki and it adheres to the guidelines of the CONSORT statement.
Female patients aged 18 to 60 years, American society of Anesthesiologist physical status I and II, scheduled for general anesthesia for elective laparoscopic cholecystectomy were enrolled. Exclusion criteria included those with inability to comprehend VAS (visual analogue scale) or severe mental impairment, difficult intubation, pregnancy, morbid obesity, history of epilepsy or allergy to any drugs used in the study, current use of opioids or beta-adrenergic receptor antagonists, baseline heart rate < 50 beats/min, acute cholecystitis, and chronic pain other than cholelithiasis.
Eligible participants were identified during the pre-anesthetic clinic visit. Informed written consent from the recruited patients was taken in the evening before surgery at the in-patient unit. Patients were also instructed about the use of the 10 cm VAS for pain where 0 was “no pain” and 10 was “worst pain”. Oral diazepam (5mg for ≤50 kg and 10 mg for > 50 kg) was given the night before and 2 h before surgery as premedication.
On the day of surgery at the preoperative holding area, patients were randomly assigned (allocation 1:1) into one of the two groups according to a computer generated random number table. Details of group assignment and case number were kept in a set of sealed opaque envelopes. The anesthesia staff opened the envelope and prepared drugs accordingly. Both the patient and the investigator observing the outcome were blinded to the patient group assignment. The attending anesthesiologist not involved in the study managed the case intraoperatively.
On arrival to the operating room, standard monitoring was applied and baseline heart rate (HR), non-invasive blood pressure, peripheral oxygen saturation and bispectral index (BIS) value (BIS® monitor; Covidien, Boulder, CO, USA) were recorded. General anesthesia was induced with IV fentanyl 1.5µg/kg and propofol 2-2.5 mg/kg until the cessation of verbal response. Tracheal intubation was facilitated with vecuronium 0.1 mg/kg IV. The lungs were mechanically ventilated using the circle system with 50% mixture of oxygen with air to maintain end tidal carbon dioxide between 35 to 45 mm Hg.
During induction, patients in the Lidocaine group received 1.5 mg/kg of lidocaine IV bolus followed by an infusion (Perfusor compact®, B-Braun, Melsungen, Germany) at 1.5 mg/kg/h. Patients in the Esmolol group received an IV bolus of esmolol (0.5mg/kg) during induction followed by an infusion titrated between 5 and 15 µg/kg/min to maintain the HR within 25% of the baseline value. In both groups, paracetamol 1 g IV was infused over 15 min after the induction of anesthesia. Anesthesia was maintained with isoflurane targeting mean arterial pressure (MAP) within 20% of baseline, and BIS value between 50 and 60 in both groups. Neuromuscular blockade was maintained with supplemental doses of vecuronium IV after observing the curare notch in capnogram. Hasson’s surgical technique was used. Each port site was infiltrated with 3 ml of 2% lidocaine before incision. Pneumoperitoneum was achieved with carbon dioxide maintaining the intra-abdominal pressure below 15 mmHg. Episodes of intraoperative hypotension (MAP <65 mmHg) and bradycardia (HR <50 beats/min) were treated with IV ephedrine 5 mg and atropine 0.4 mg respectively.
No supplemental opioids were used during the surgery. All patients received IV ketorolac 30 mg after the removal of the gall bladder. At the end of surgery, the carbon dioxide remaining in the peritoneal cavity was expelled by slow abdominal decompression. Both isoflurane and the study drug infusion were discontinued after the skin closure. Incision site was infiltrated with 10 ml of 0.25% bupivacaine. Residual neuromuscular block was reversed with IV neostigmine 0.05 mg/kg and glycopyrrolate 0.01 mg/kg. When the patients were conscious and had adequate muscle power, thorough oropharyngeal suctioning was done and endotracheal tube was removed. The investigator blinded to the group assignment now entered the operating room to collect data on intraoperative hemodynamics side effects. The patients were then transferred to the PACU after they followed verbal commands.
Postoperative pain management included IV paracetamol 1 g and ketorolac 30 mg at 6 h and 8 h respectively. The blinded investigator not involved in the anesthesia management assessed VAS pain scores at rest and during movement at the PACU (on arrival, 15 min, 30 min, 1 h) and surgical in-patient-unit (2 h, 6 h, 12 h and 24 h). If the VAS score for pain exceeded >3 at rest, morphine 1 mg IV was administered in the PACU, and repeated every five min until the VAS score was ≤ 3, or if any adverse effects were noticed. These included increased sleepiness (Ramsay sedation scale (RSS) score > 3), respiratory depression (SpO2 < 90% in room air or respiratory rate < 8/min). The patients were transferred to the in-patient-unit after 1 h of stay in PACU. In the surgical unit, tramadol 50 mg IV was administered and further doses of 50 mg was given every 10 min for maintaining VAS score for pain ≤ 3 (the maximum dose of tramadol was limited to 300 mg in the first 24 h). The tramadol used in surgical unit was converted to morphine equivalent using online calculator (http://clincalc.com/Opioids/).
The primary outcome was the total morphine equivalent consumed in the first 24 h postoperatively. Secondary outcome measures included patient-reported VAS pain scores at rest and movement, postoperative nausea and vomiting (PONV) on a four point scale18 (1 = no nausea, 2 = mild nausea, 3 = severe nausea, 4 = retching and/or vomiting), the 6-point RSS scores 19(1 = patient anxious and restless, 2 = cooperative and awake, 3 = responding to verbal commands, 4 = responding to mild stimulus, 5 = responding to deep stimulus, 6= no response). These parameters were noted in PACU and at 2, 6, 12 and 24 h in the surgical unit. PONV grade 3 & 4 were treated with metoclopramide 10 mg IV. Time to first perception of pain and void, overall patient satisfaction from anesthesia at 24 h based on 5-point Likert scale (1= highly satisfied, 2 = satisfied, 3 = neutral, 4 = dissatisfied, 5 = highly dissatisfied), and occurrence of lidocaine toxicity were also noted. The patients were discharged from the hospital at 24 h after surgery.
Sample size was determined with the aim to reject the inferiority of esmolol infusion compared with lidocaine for the primary outcome of 24 h morphine consumption after surgery. The non-inferiority margin was considered as 2 mg. A sample size of 78 patients (39 per arm) was required to achieve a power of 90%, a one-sided 95% confidence interval, assuming the standard deviation of 3. We finally enrolled 90 patients to allow for possible dropouts or protocol violators (https://www.sealedenvelope.com/power/continuous-noninferior/).
The data collected was entered into excel software and analyzed on STATA version 13.0 (Stata Corporation, College Station, TX, USA). Normality of data was checked using histograms, Skewness-Kurtosis test and Shapiro-Wilk test. Normally distributed data were compared between the two groups using the unpaired Student t-test. Mann-Whitney U-tests were used for continuous non-normally distributed data and ordinal data. The median differences and its 95% CIs for the primary outcome variable were calculated using quantile regression. Comparison of sedation scores between the two groups was performed using a mixed effects model. Fixed effects were time of assessment of sedation scores postoperatively (15 min to 24 h), study-group assignment (esmolol or liodcaine), and participants in the study as a random effect. Interaction between time of assessment of sedation scores and study group was also included in the model and an unstructured covariance matrix was used. For categorical variables, Chi-square test was applied. Time to first perception of pain between the groups was plotted with Kaplan-Meier survival curves and compared with log-rank test. A p value < 0.05 was considered as statistically significant.