Senagore 2003.

Methods	RCT Approved by the Institutional Review Board and informed consent obtained Setting: United States of America Funding: unspecified
Participants	38 patents scheduled for segmental laparoscopic colectomy
Interventions	Treatment group: TEA (T8‐T9 or T9‐T10) with bupivacaine 0.1% and fentanyl 20 mcg/mL for 18 hours (n = 18) Control group: IV PCA with morphine for 18 hours (n = 20) General anaesthesia, diclofenac 50 mg orally every 8 hours beginning the evening before surgery and continued after operation, ketorolac 30 mg within 30 minutes of completion of surgery and antiemetic prophylaxis of dexamethasone 8 mg and ondansetron 4 mg for all participants
Outcomes	VAS scores at movement at 6 and 24 hours Hospital length of stay (Criteria for discharge from hospital in both groups of participants included tolerance of 3 consecutive general meals without nausea or vomiting, adequate pain control with oral analgesics and passage of flatus. Length of hospital stay was defined as the number of nights spent in hospital from day of operation until discharge)
Notes	Investigators reported no instances of TEA catheter malfunction or complications that required early removal of the catheter
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	"computer‐generated random number after giving informed consent"
Allocation concealment (selection bias)	Low risk	"computer‐generated random number after giving informed consent"
Blinding of participants and personnel (performance bias) All outcomes	Unclear risk	Not mentioned
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	Not mentioned
Incomplete outcome data (attrition bias) All outcomes	Low risk	"There were no instances of thoracic epidural catheter malfunction or complications that required early removal of the catheter" "Forty‐seven patients were randomized during the study but four (two thoracic epidural and two IV opioids) had no resection or a second surgical procedure during their hospital stay, and five (three TEA and two PCA) had protocol violations and were excluded from subsequent analysis"
Selective reporting (reporting bias)	Low risk	All results reported
Other bias	Unclear risk	Groups well balanced Not in intention‐to‐treat: "whose operation was converted to open surgery after randomization were excluded from analysis"