Gambling 2009.
| Methods | RCT Approved by ethics committees (multi‐centre) and written informed consent obtained Setting: United States of America Funding: industry |
|
| Participants | Adult patients aged 18 years or older with ASA 1 to 3, who were scheduled to undergo total abdominal hysterectomy, prostatectomy or colon resection under general anaesthesia Exclusion criteria were pregnancy or lactation, any present or past disease or condition or prior post‐surgical complication that might increase the risk associated with surgery or risk of post‐surgical complications, sleep disorder (e.g. sleep apnoea, narcolepsy, excessive daytime sleepiness) and history of substance abuse. Patients who underwent treatment with clonidine or daily opioids for longer than 7 days before enrolment or who used any long‐acting opioid, ketorolac or cyclo‐oxygenase 2 inhibitor for 48 hours before surgery were excluded |
|
| Interventions |
Treatment group: epidural catheter at an unspecified level. Test dose with 3 mL of 1.5% lidocaine with epinephrine 5 mcg/mL. Bupivacaine 0.25% 20 mL before surgical incision plus extended‐release morphine 15 mg injected 15 (28 enrolled/28 analysed), 30 (30 enrolled/28 analysed) or 60 minutes (29 enrolled/22 analysed) after bupivacaine dose. Catheters were removed at the end of surgery. IV PCA with fentanyl after surgery Control group: epidural catheter at an unspecified level. Test dose at the discretion of the attending anaesthesiologist. Extended‐release epidural morphine 15 mg. Catheters were removed at the end of surgery. IV PCA with fentanyl after surgery (29 enrolled/27 analysed) General anaesthesia for all participants |
|
| Outcomes | VAS scores on movement at 8, 24 and 48 hours after surgery Vomiting |
|
| Notes | Study also included a group given bupivacaine alone not retained for this review | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Low risk | "computerized randomization system" |
| Allocation concealment (selection bias) | Low risk | "Patient numbers were assigned by the study‐site pharmacist using an interactive voice recognition system" |
| Blinding of participants and personnel (performance bias) All outcomes | Unclear risk | Study medications were prepared and administered by unblinded pharmacists and anaesthesiologists, respectively, none of whom were involved in study assessments |
| Blinding of outcome assessment (detection bias) All outcomes | Low risk | Blinded assessors |
| Incomplete outcome data (attrition bias) All outcomes | Low risk | 16 participants withdrew from the study early because of withdrawal of consent (n = 2), loss to follow‐up (n = 1), change in surgery (n = 2) or other reasons (n = 11, all unique reasons such as bleeding risk, investigator withdrawal and unsuccessful epidural catheterization). 15 of these participants did not receive study drug (extended‐release morphine or placebo) and were excluded from analyses |
| Selective reporting (reporting bias) | Low risk | All results reported |
| Other bias | Unclear risk | Groups well balanced Not in intention‐to‐treat |