This study was approved by the Tata Medical Center Institutional Review Board (approval no. 2020/TMC/204/IRB62) and written informed consent was obtained from all subjects participating in the trial. The trial was registered prior to patient enrollment at ctri.nic.in (CTRI/2020/12/030076, Principal investigator: Rudranil Nandi, Date of registration: 28/12/2020). The study was performed in accordance with the declaration of Helsinki and has been reported in concordance with CONSORT guidelines.15
All patients aged between 18–75 year., with ASA physical status I-III, scheduled for elective open gastrointestinal oncosurgery were included in the study. Patients posted for hepatobiliary, pancreatic, and esophageal surgery and those on step 3 opioids before the procedure were excluded. A patient who was electively ventilated after the surgery was withdrawn from study. All patients were explained about the methods of analgesia, Numeric Rating Scale (NRS) and satisfaction score while obtaining consent. Randomization was done in the OT waiting area just before the operation by using a computer-generated randomization table by a person who was not involved in the study with an allocation ratio of 1:1. Allocation concealment was done by sequentially numbered opaque sealed envelope technique. On the morning of surgery, the attending anesthesiologist picked up the topmost envelope and randomized the patient accordingly. This was an open labeled study as it was not possible to blind the patient as well as the assessor for postoperative outcome measurements because one group was receiving continuous infusion while the other received intermittent bolus.
Patients were allocated to either group MOR or group ROP. In both the groups, epidural catheters were placed at a level considered congruent to the intended incision before induction of anesthesia. Placement of the catheter was checked by 3 ml of 2% lignocaine with adrenaline (1:200000).
In group MOR, epidural analgesia was provided by morphine 0.05 mg/kg (not more than 3 mg) diluted in 10 ml of bupivacaine 0.1% [for a 60 kg body weight patient, drug was 3 mg morphine (1 ml) + 2ml bupivacaine 0.5%+ 7 ml normal saline] approximately 15 min before surgical incision. The dose was repeated every 8–24 hr. depending on analgesic and other responses during perioperative period till postoperative day 5. Initially the drug was administered every 12 hr., however, in case of inadequate pain relief, epidural morphine (with low dose bupivacaine) was repeated every 8 hr. Pain relief was considered inadequate if fentanyl consumption as rescue analgesia exceeded above 100 mcg in 4 hr. If analgesia was adequate and the patient had some dose limiting side effects like sedation, severe pruritus and respiratory rate below 10/min, the epidural drug was administered every 24-hr.
In group ROP, epidural analgesia was activated with 6 ml bolus of ropivacaine 0.2% with fentanyl 2 mcg/ml, 15 min before surgical incision, followed by infusion at 5 ml/hr. The rate of infusion was titrated between 3 to 10 ml/hr. depending on the quality of analgesia and hemodynamic parameters during the perioperative period. This was based on the clinical judgment of the attending anesthesiologist during the intraoperative period and acute pain service team in the postoperative period.
All patients received general anesthesia after placement of an epidural catheter. In both the groups anesthesia was induced with propofol 2 mg/kg, fentanyl 2 mcg/kg and rocuronium 0.6 mg/kg. During the intraoperative period, even after highest dose of epidural analgesia as per group allocation, if the patient’s heart rate or BP increased to more than 20% of the preinduction values and that rise was clinically judged to be due to inadequate analgesia, rescue analgesia was administered with an intravenous bolus of 25 mcg fentanyl. Anesthesia was maintained with sevoflurane in a mixture of medical air 60% and O2 40%. End-tidal concentration of sevoflurane was adjusted to maintain MAC > 0.7. All patients received balanced salt solution (BSS) at a rate of 4 ml/kg/hour. BP was maintained within 20% of baseline values. Any transient hypotension was treated with a phenylephrine bolus injection. Persistent hypotension was initially treated with a fluid bolus of 5 ml/kg followed by noradrenaline infusion as per institutional protocol. A thermal blanket was positioned over the exposed parts of the body to maintain perioperative normothermia. Thirty minutes before termination of anesthesia, intravenous paracetamol 15 mg/kg was administered.
In the postoperative period, if the patient’s pain score became greater than 3 even after the highest dose of epidural analgesia (epidural infusion at 10 ml/hr. in group ROP, epidural morphine bolus every 8 hr. in group MOR), i.v. fentanyl bolus 25 mcg was administered. If fentanyl consumption exceeded 100 mcg (4 bolus dose of 25 mcg fentanyl) in 4 hr., i.v. fentanyl infusion was started at 20 mcg/hr. and titrated by 5 mcg/hr. in every 30 min till adequate pain relief was achieved. In the postoperative period all patients received maintenance fluid with BSS at the rate of 1.5 ml/kg/hr. If there was hemodynamic compromise [hypotension (SBP < 90mm of Hg), urine output < 0.5ml/kg/hr.] intravenous fluid bolus at 5 ml/kg was administered. If hypotension persisted even after fluid bolus, noradrenaline infusion was started as per institutional protocol. Intraoperative blood loss was estimated, and transfusion was planned as per institutional protocol.
Postoperative fentanyl consumption as rescue analgesia was recorded as primary outcome, whereas postoperative pain scores at different time points, vasopressor requirement, volume of fluids administered, time to enteric resumption and length of hospital stay (LOHS) were noted as secondary outcomes. Intraoperative HR, systolic blood pressure (SBP) and diastolic blood pressure (DBP) were recorded before induction of anesthesia, before surgical incision, 2 min after surgical incision and thereafter every 5 min till completion of surgery. All the values of HR, SBP, DBP taken during the intraoperative period of a particular patient were averaged and noted down as average intraoperative HR, SBP or DBP, respectively of that patient. Total consumption of noradrenaline required to maintain normal hemodynamics was noted during the intraoperative and postoperative period. Noradrenaline solution was prepared by diluting 4 mg (4 ml) of noradrenaline in normal saline to make a 40 ml solution and the administered volume of that solution was noted. The proportion of minimal alveolar concentration (MAC) of sevoflurane administered during the intraoperative period was noted down in every 5 min during surgery and the values were averaged to get the average proportion of MAC requirement of that patient.
Postoperative HR, SBP, DBP, pain score on NRS on rest and movement and sedation level using modified Ramsay sedation score (RSS) were recorded immediately after surgery and 1 hr., 4 hr., 12 hr., 24 hr., 48 hr., 72 hr. after surgery. Sedation level was graded as follows- (1) anxious, agitated, or restless, (2) oriented, calm, and co-operative, (3) responsive to commands only, (4) exhibiting brisk response to light glabellar tap or loud auditory stimulus, (5) exhibiting a sluggish response to light glabellar tap or loud auditory stimulus, and (6) unresponsive. Patient satisfaction score was recorded using the 7-point Likert scale where 1 means extremely dissatisfied and 7 means extremely satisfied. Postoperative nausea and vomiting (PONV) were graded as (0)-no nausea and vomiting, (1)-nausea without vomiting, (2)-nausea with vomiting < 3 episodes, (3)-nausea with vomiting > 3 episodes. The time when a patient tolerated 500 ml enteric liquid was documented as time to enteric resumption. Treating surgeons were asked about satisfaction with postoperative analgesia on a scale of 0 to 10, where 0 is not at all satisfied and 10 means very much satisfied. Any complication in the postoperative period was noted. The patients were discharged after satisfactory mobilization, bowel opening and tolerating oral diet.
Standard deviations of postoperative rescue fentanyl consumptions (primary outcome), calculated from the hospital’s own unpublished data, were 120 mcg and 110 mcg after epidural ropivacaine-fentanyl infusion and epidural morphine (with low dose bupivacaine) bolus respectively. Based on our clinical experience, the difference of fentanyl consumption of less than 100 mcg in the first 24 hr. was defined as an equivalence margin. Assuming 2.5% type I error, 90% power and 10% drop out sample size of the study was 86 (43 subjects in each group).
Continuous data was expressed as mean and 95% confidence interval (CI) and qualitative data was expressed as number of events (%). Nominal data was compared using a chi-squared test. Ordinal qualitative variables or quantitative variables were compared using Wilcoxon − Mann–Whitney sum of ranks test and Student’s t respectively. Results with p < 0.05 were considered as statistically significant. All analyses were performed by applying a bilateral hypothesis using statistical software package R (version-4.3.1).