Tenenbein 2008.

Methods	Parallel RCT Ethics committee: approved by the Ethics Review Board of the University of Manitoba Health Sciences Centre Informed consents: all participants gave written informed consent Site: University of Manitoba, Canada Setting: university hospital Dates of data collection: between July 1, 2003, and June 30, 2004 Funding: supported by the Health Sciences Centre Research Foundation Registration: not registered
Participants	50 participants; mean age: 60.5 years; sex distribution: not reported Inclusion criteria Patients younger than 80 years of age Deemed appropriate for their facilitated recovery programme Undergoing CABG surgery Exclusion criteria Age 80 years or older Previous cardiac surgery Combined procedures Serum creatinine > 150 mmol/L Pre‐existing coagulopathy Use of antiplatelet agents other than aspirin Active liver disease Severe spinal deformity Ejection fraction < 30% Body mass index > 35 kg/m²
Interventions	Intervention Epidural analgesia (N = 25) Control Systemic analgesia (N = 25) Premedication: diazepam Induction: sufentanil, sodium thiopental, or propofol and rocuronium Maintenance: isoflurane Surgery: CABG with CPB
Outcomes	Relevant to this review Risk of mortality Risk of pulmonary complications (respiratory depression, pneumonia) Risk of atrial fibrillation or atrial flutter Risk of neurological complications (cerebrovascular accident, serious neurological complications from epidural analgesia) Tracheal extubation Pain scores Others Rescue analgesia Lung function tests
Notes	Correspondence: information received from study authors Conflict of interest: no conflicts of interest DOI: 10.1007/BF03021489
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Quote: "prospective, randomized, controlled trial"; "Randomization occurred immediately after enrolment"; "assigned a sealed envelope that contained the group assignment"
Allocation concealment (selection bias)	Low risk	Quote: "assigned a sealed envelope that contained the group assignment"
Blinding of participants and personnel (performance bias) All outcomes	High risk	Quote: "this was not a blinded study"
Blinding of outcome assessment (detection bias) All outcomes	High risk	Quote: "this was not a blinded study"
Incomplete outcome data (attrition bias) All outcomes	Low risk	Quote: "Due to incomplete data collection, two patients were excluded from each group"; "unable to insert an epidural in one patient, and the epidural was not use, postoperatively in another patient, because of quadriparesis on emergence"
Selective reporting (reporting bias)	Low risk	All results reported
Other bias	Low risk	Quote: "intention to treat principle" Preoperatively, groups were similar, except for higher mean ejection fractions in the control group (59.1 ± 8.9% vs 52.9 ± 7.5%; P < 0.01)