Study Protocol
This prospective, double-blind, randomized controlled study was approved by the Institution Review Board of the Eye & ENT Hospital of Fudan University. Written informed consent were obtained from all participants. The trial was registered before patient enrollment at the Chinese Clinical Trial Registry (Clinical Trial Number: ChiCTR-IPR-16007927).
From May 2016 through November 2017, 109 patients undergoing glaucoma or retina (vitrectomy) surgeries were initially recruited in this study. Patients with the age between 65 and 79 years old who were classified as American Society of Anesthesiologists (ASA) Physical Status Class 1 or 2 were eligible to participate. Exclusion criteria included long-term chronic pain, long-term use of opioid analgesics, sedatives, or antidepressants, recent intake of opioid analgesics, sedatives, or antidepressants, current alcohol or other substance abuse, having allergy to study drugs, with liver or renal dysfunctions, or any medical conditions that were contradicted for laryngeal mask airway (LMA) anesthesia.
After a thorough explanation to eligible patients regarding the surgical procedure, general anesthesia, and the purpose of this study, informed consent was obtained from 107 patients. Patients were then randomized using a computer randomization program into one of the three groups: Group O1: oxycodone 0.1 mg/kg (Hamol Limited; Nottinghamshire, UK), Group O2: oxycodone 0.2 mg/kg, or Group F: fentanyl 2 mcg/kg (Yichangrenfu; Hubei, China). The designated drug for each patient was prepared into a 10-ml solution with normal saline, and labelled with the patient’s name by a nurse. It was then delivered to the anesthesia team blindly right before the surgery.
Standard ASA and bispectral index (BIS) monitors were applied to patients upon their arrival to the operating room, without any pre-medication. After confirmation of each patient’s baseline vital signs and adequate pre-oxygenation, general anesthesia was induced with intravenous propofol (2-2.5 mg/kg, AstraZeneca; Cambridge, UK), cisatracurium (0.15 mg/kg, Shanghai Pharma; Shanghai, China), and 10 ml of test drug. All patients were intubated with LMAs 3 minutes after induction of anesthesia. General anesthesia was maintained with sevoflurane and 50% air in oxygen with a constant fresh gas flow of 2 L/min to keep the BIS readings between 40 and 60. Intravenous ephedrine (3-5 mg) was administered for each event of hypotension (mean arterial pressure<65 mmHg).
After surgery completed, sevoflurane was turned off, and patients were transferred to the Post-Anesthesia Care Unit (PACU) for recovery. Patients were extubated at the PACU once criteria met. The severity of post-operative pain was evaluated using the Numeric Rating Scale (NRS: 0=no pain, 10=worst pain imaginable). When patients complained of moderate to severe pain (NRS>4), intravenous oxycodone (3 mg) or fentanyl (30 mcg) was administered as the rescue analgesic for Group O1/O2 or Group F patients respectively.
Data Collection
Age, sex, height, weight, ASA physical status classification, surgical time and anesthesia time were collected for each patient. Surgical time was defined as the duration between surgical incision and closure, and anesthesia time was defined as the duration between administration of induction agents and patient transport out of the operating room [19]. Heart rate (HR), systolic blood pressure (SBP), diastolic blood pressure (DBP), and mean arterial pressure (MAP) were documented and analyzed during the following time periods: baseline upon the arrival to the operating room; during induction (0-10 min after administration of induction agents); during the intraoperative period; and during PACU stay. Clinically significant hypotension was defined as either: MAP decrease of >40% and MAP<70 mmHg, or MAP<65 mmHg [15]. The recovery time of spontaneous breathing (the duration between discontinuation of anesthetics and returning of spontaneous breathing), the emergence time (the duration between discontinuation of anesthetics and extubation), and the PACU length of stay were recorded for each patient. The severity of postoperative pain was evaluated after extubation and at 0, 2, 6, 24 hours after discharge from PACU. The administration of rescue analgesics for each patient was recorded and analyzed. The Ramsay Sedation Scale (RSS) was used to evaluate the sedation level at PACU discharge (1=anxious, agitated, restless; 2=cooperative, oriented, tranquil; 3=responding to commands only; 4=brisk response to light glabellar tap or loud noise; 5=sluggish response to light glabellar tap or loud noise; and 6=no response). The incidents of perioperative complications including coughing during induction, postoperative nausea/vomiting, and hypoxia (pulse-oximetry reading below 90%) at PACU were also recorded for analysis.
Statistical Analysis
The main efficacy endpoint of our study was the PACU length of stay. Based on the data of our pilot study (33.4±8.12 min in Group F, 30.9±6.48 min in Group O1, 38.9±12.93 min in Group O2), the sample size of 28 patients in each group was calculated at α=0.05, β=0.2 using the sample size software (NCSS-PASS, Kaysville, UT, USA). We aimed to recruit a total of 109 patients to allow for potential loss to follow-up.
We used the SPSS 23.0 software to perform data analysis. Continuous data with normal distribution was presented as mean±standard deviation. One-factor analysis of variance was used to analyze the baseline data among the groups, followed by SNK method for post hoc pairwise test. Analysis of variance of repeated measures was applied for hemodynamic data. Continuous data without normal distribution was presented as median (interquartile range). Kruskal Wallis test was used for analyzing the data among the groups, followed by Man-Whitney U test for post hoc pairwise test. Nominal data were presented as number of patients. Chi square test and Fisher exact test was applied for comparison of the data among the groups, followed by partitions of chi square test for post hoc pairwise test. The type-I error of multiple comparisons of continuous data without normal distribution and nominal data was corrected by Holm-Bonferroni method. P<0.05 was regarded as statistically significant.