Katz 1996.

Methods	Triple‐blind (participants, providers, outcome assessors), sham/placebo‐controlled, randomized clinical trial Sequence generation was by random number tables Follow‐up: 18 months
Participants	Participants: 30 adults in a university setting in Toronto, Ontario, Canada Operation: lateral thoracotomy for pulmonary or oesophageal disease 2 groups, size: 15/15 Age (group 1, 2): 54.6 years (range 19‐75), 58.9 (range 46‐72) Men/women (group 1, 2): 5/10, 8/7
Interventions	Group 1 (preincision intercostal block): placebo rectal suppository, intramuscular midazolam (0.05 per kg), GA (fentanyl 1 µg/kg), preincision intercostal nerve block with bupivacaine (0.5% with epinephrine (1:200.000), 3 mL/interspace) 2 spaces above and below planned incision, post‐op for 72 h PCA morphine (demand 1.5 mg‐2 mg, lockout 6 min, max dose 30 mg/4 h) Group 2 (sham/placebo block): IM morphine (0.15 mg/kg) and perphenazine (0.03 mg/kg), indomethacin (100 mg, rectal suppository), GA (fentanyl 1 µg/kg), preincision sham intercostal nerve block with normal saline (3 mL/level) 2 spaces above and below planned incision, post‐op for 72 h PCA morphine (demand 1.5 mg‐2 mg, lockout 6 min, max dose 30 mg/4 h) Adjuvants: none Immediate post‐op pain control: initial analgesic consumption reduced
Outcomes	Dichotomous: pain and analgesic consumption at 18 months Continuous: verbal rating scale at 18 months Secondary: allodynia at 6 and 12 months
Notes	We contacted the study author for missing information. He provided a data table with unpublished data from the follow‐up study to Kavanagh 1994, the second manuscript reporting on (Katz 1996). Funding sources: "this study was supported by a research scholarship from the Medical Research Council of Canada (MRC) and by MRC grant MT‐12052 to Dr Katz." Conflicts of interest: a conflict of interest statement was not given.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Quote: "a table of random numbers was used to allocate patients."
Allocation concealment (selection bias)	Low risk	Quote: "..investigator (who had no further involvement with that patient) who administered the medications in accordance with the instructions in the envelope...".
Blinding of participants and personnel (performance bias) All outcomes	Low risk	Quote: "the patients and all other personnel involved in subsequent patient management and assessment were completely blinded as to group allocation, ...thus maintain the blind and (patients) also received a placebo rectal suppository."
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Quote: "other personnel involved in subsequent patient management and assessment were completely blinded as to group allocation...,thus maintain the blind..."
Incomplete outcome data (attrition bias) All outcomes	High risk	Attrition was described with regards to group allocation. Per‐participant analysis was performed, with no ITT analysis considered. Bias is unlikely, as an ITT analysis would not alter the lack of the statistical significance.
Selective reporting (reporting bias)	Unclear risk	Primary outcomes fully reported on
Null bias	High risk	Quote: "in the original study, use of preemptive multimodal analgesia during surgery was not found to be more effective than the placebo in reducing the intensity of acute postoperative pain"