Gabbott 1997.

Study characteristics
Methods	Study design: parallel‐group randomized controlled trial Sample size: 65 Country: UK Setting: secondary care hospital Dates conducted: not reported Postoperative opioid used and delivery: PCA morphine Pain score collection: trained nurse Concurrent postoperative analgesics: none reported
Participants	Inclusion criteria ASA 1 or 2 Female Elective total abdominal hysterectomy with or without salpingo‐oophorectomy for non‐malignant disease Exclusion criteria > 100 kg weight > 60 years old Renal impairment Peptic ulceration Asthma Coagulopathy or anticoagulants Allergy to NSAIDs Drug or alcohol abuse Current NSAID use Lithium
Interventions	Group D (34 participants): IM ketorolac 30 mg 45‐90 minutes preoperatively and placebo at incision Group B (31 participants): IM ketorolac 30 mg at incision and placebo 45‐90 minutes preoperatively
Outcomes	Morphine consumption (mg consumed at 1, 2, 4, 8 and 12 hours) Postoperative pain (VRS of nil, mild, moderate, severe and asleep at 1, 2, 4, 8 and 12 hours) Nausea (nil, mild, moderate, severe and asleep at 1, 2, 4, 8 and 12 hours) Sedation (nil, mild, moderate, severe and asleep at 1, 2, 4, 8 and 12 hours) Respiratory rate (at 1, 2, 4, 8 and 12 hours)
Notes	Funding: not reported Declarations of interest: not reported Authors contacted: yes Other: unable to use any data due to different time points, lack of reporting and inability to convert the scales used
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Unclear whether a truly random process. Quote: "Randomization was achieved by allocating each patient the next numbered pair of study ampoules".
Allocation concealment (selection bias)	Unclear risk	Pharmacy‐controlled but in view of above unclear. Quote: "The code determining the contents of each ampoule was kept by the pharmacy department and not released to the investigators until after the study was terminated".
Blinding of participants and personnel (performance bias) All outcomes	Low risk	Double‐dummy placebo. Quote: "All ampoules were identical and packaged in pairs labelled 'pre‐operative' and 'intra‐operative'. Each boxed pair was sequentially coded allowing the study to be carried out in a double‐blind fashion".
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Likely blinded from above
Incomplete outcome data (attrition bias) All outcomes	Low risk	Large number of dropouts, although these individuals had been randomized and thus intention‐to‐treat was not followed but the intervention was assigned double‐blinded; the total dropout rate was only 14%, and there were similar numbers of dropouts between the groups (ranging from 10‐23%). Quote: "One hundred and sixty patients agreed to take part in the study. Of these, 23 were withdrawn before induction of anaesthesia. These patients were either not given the study drug or more than 90 min had elapsed between the preoperative injection and the time of arrival in the anaesthetic room".
Selective reporting (reporting bias)	High risk	Some outcomes not reported or not fully reported
Other bias	Low risk	Similar baseline characteristics. Unclear role for Selecta UK in trial conduct