Loane 2012.

Methods	Double‐blind (participant, outcome assessor) randomized clinical trial Sequence generation by computer‐generated table Follow‐up: 3 months
Participants	Participants: 69 adult women at university hospital in Vancouver, British Columbia, Canada Operation: elective caesarean delivery with low transverse incision (under SA) 2 groups, size: 33/33 (completed) Age (± SD), group 1, 2: 35 (± 3), 34 (± 5) All female participants Comorbidities: number of multiparous women (group 1, 2): 25/21
Interventions	Group 1 (intrathecal morphine): 100 µg intrathecal morphine at time of spinal insertion. At end of surgery, sham TAP block with capped needle pushing against skin Group 2 (TAP block): no intrathecal morphine was given. At the end of surgery, TAP block 5 mL increments of ropivacaine into transversus abdominis plane on each side (0.5% ropivacaine, 1.5 mg/kg on each side to max of 100 mg (20 mL)) Both groups received standardized SA with 0.75% hyperbaric bupivacaine 11.25 mg + fentanyl 10 µg and at the end of surgery, rectal naproxen 500 mg + paracetamol 975 mg. Both had same post‐op analgesia regimen with 500 mg naproxen every 12 h standing, oral hydromorphone 2 mg‐4 mg every 4 h as needed with IV PCA (bolus 1.5 mg, lockout 7 min, max 10 mg/h) if needed Adjuvants: none Immediate post‐op pain control: pain scores were higher in participants receiving a TAP block at all time points but this was only significant at 10 h; statistically significant increase in morphine consumption 24 h post‐op in TAP group, but not at earlier time point
Outcomes	Dichotomous: pain/no pain "in the operative area" at 3 months Continuous: none Adverse events: incidence of wound infection, nausea/vomiting, pruritus, sedation
Notes	We contacted the study author for clarification on participant flow details, but received no response. Funding sources: "the authors received no external funding for this project." Conflicts of interest: "Dr Joanne Douglas is an Editor of the International Journal of Obstetric Anesthesia. She had no involvement with the editorial process or decision to accept this article."
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Randomly assigned using "computer‐generated table" after consent and enrolment
Allocation concealment (selection bias)	Low risk	Quote: "group allocation was concealed in an opaque envelope until the woman was consented and enrolled"
Blinding of participants and personnel (performance bias) All outcomes	Unclear risk	Quote: "women, postoperative care providers..were blinded to treatment group...The anaesthesiologist caring for the woman, as well as the anaesthesiologist performing the TAP block, were not blinded". Bias during operation by non‐blinded providers possible, e.g. by administering additional morphine, but not very likely
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Quote: "women, postoperative care providers and research staff collecting postoperative data were blinded to treatment group"
Incomplete outcome data (attrition bias) All outcomes	Unclear risk	69 women were randomized, but 1 in intrathecal morphine group and 2 in TAP group were excluded because of protocol violation. 3‐month follow‐up was obtained from 31 (of 33) in group 1 and 28 (of 33) in group 2. Numbers of attrition provided per group, fairly balanced. However, numbers presented in text do not match the numbers presented in the flow chart (reversed groups)
Selective reporting (reporting bias)	Low risk	Primary outcome in protocol fully reported on. Investigator left the study and this led to premature termination of the study before the intended time.
Null bias	High risk	Quote: "pain scores on rest and movement were higher in the TAP block group at all times although this only reached statistical significance at 10 h (P = 0.001)"