Carli 2002.

Methods	RCT Approved by the ethics committee and informed consent obtained Setting: Canada Funding: charity
Participants	64 adult patients undergoing elective colorectal surgery for non‐metastatic conditions
Interventions	Treatment group: TEA (T8 or T9 interspace) with bupivacaine 0.1% and fentanyl 2 mcg/mL or morphine 0.1 mg/mL for 4 days (n = 32) Control group: IV PCA with morphine for 3 to 4 days (n = 32) All participants received general anaesthesia
Outcomes	VAS at rest at 24, 48 and 72 hours VAS on movement at 24, 48 and 72 hours Time to transit taken at time to first faeces Anastomotic leak Hospital LOS
Notes
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	"randomly assigned", no details
Allocation concealment (selection bias)	Low risk	"On the morning of the surgical procedure, the group to which the subject had been randomly assigned was revealed"
Blinding of participants and personnel (performance bias) All outcomes	High risk	"The patients were not blinded"
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	Not mentioned
Incomplete outcome data (attrition bias) All outcomes	Low risk	No loss to follow‐up
Selective reporting (reporting bias)	Low risk	All results reported
Other bias	Low risk	Groups well balanced If the epidural block did not provide adequate analgesia, the participant continued to be included in the intention‐to‐treat analysis but was excluded from analysis addressing efficacy