Peyton 2003.

Methods	RCT Approved by the human investigation committee at each of 25 participating centres, and written informed consent was obtained from all randomized participants Setting: Australia Funding: governmental
Participants	888 patients (920 randomized) undergoing elective, open and major abdominal surgery (including oesophagectomy) at high risk of an adverse outcome from having 1 or more important pre‐existing co‐morbidities
Interventions	Treatment group: TEA/LEA epidural block 2 spinal segments above the upper end of the participant's wound, intraoperative loading with local anaesthetic (bupivacaine or ropivacaine) and continuation by infusion of local anaesthetic and opioid (pethidine or fentanyl) after surgery for 72 hours or longer (n = 447) Control group: participant‐ or physician‐controlled IV opioid infusions initially (n = 441) All participants had general anaesthesia
Outcomes	VAS at rest at 24, 48 and 72 hours (morning) VAS on movement (coughing) at 24, 48 and 72 hours (morning) Hospital LOS
Notes	225/447 fully compliant with protocol (183 removed before 72 hours)
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	"By permuted random blocks with stratification by study centre"
Allocation concealment (selection bias)	Low risk	Eligible patients were identified preoperatively by nurses or anaesthetists in collaborating hospitals and, after informed consent had been obtained, they were allocated by a central 24‐hour randomization service to control or epidural groups by permuted random blocks with stratification by study centre
Blinding of participants and personnel (performance bias) All outcomes	High risk	No sham catheters; "deemed masking to be unethical in very sick patients"
Blinding of outcome assessment (detection bias) All outcomes	High risk	Relevant data were collected by study nurses at each participating institution, but whether particular endpoints had occurred was defined by a computer algorithm at the time of data entry by staff of the Trial Secretariat who were blinded to treatment allocation
Incomplete outcome data (attrition bias) All outcomes	Low risk	"23 patients whose surgery was cancelled after randomisation and four who were randomised for an ineligible procedure were also excluded from analysis. 19 patients who were listed for an eligible procedure at the time of randomisation subsequently underwent a non‐eligible operation. By the intention‐to‐treat principle, these patients were included in the primary analysis"
Selective reporting (reporting bias)	Low risk	All results reported
Other bias	Low risk	Comparison of treatment groups showed no statistically significant differences in the distribution of any inclusion criterion "23 patients whose surgery was cancelled after randomisation and four who were randomised for an ineligible procedure were also excluded from analysis. 19 patients who were listed for an eligible procedure at the time of randomisation subsequently underwent a non‐eligible operation. By the intention‐to‐treat principle, these patients were included in the primary analysis"