Kaczmarzyk 2010.

Study characteristics
Methods	Study design: parallel‐group randomized controlled trial Sample size: 64 Country: Poland Setting: secondary care hospital Dates conducted: not reported Postoperative opioid used and delivery: no opioid used Pain score collection: self‐report Concurrent postoperative analgesics: paracetamol PRN
Participants	Inclusion criteria Healthy individuals Surgical extraction of the lower wisdom teeth Exclusion criteria Age under 18 or over 60 years Pregnancy Allergy to ketoprofen, aspirin or any other NSAID Lactose intolerance Gastrointestinal disease Inflammation in the area of the tooth to be extracted Any antibiotic or analgesic intake within the previous 7 days
Interventions	Group Pre (34 participants): 100 mg ketoprofen PO 60 minutes preoperatively, followed by 100 mg placebo PO 60 minutes postoperatively Group Post (30 participants): 100 mg placebo PO 60 minutes preoperatively, followed by 100 mg ketoprofen PO 60 minutes postoperatively
Outcomes	Postoperative pain (0‐100 mm VAS every hour for 12 hours postoperatively) Paracetamol consumption (number of doses consumed) Time to first and second analgesic request (minutes)
Notes	Funding: "Grant of the Jagiellonian University (K/ZDS/00519)" Declarations of interest: none declared Authors contacted: no Other: pain data extracted from graphs and SD estimated. Pain score from 2 hours to allow postoperative dosing time to take effect
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Random number table. Quote: "Each envelope contained the group assignment for one patient, which was determined in advance by a random number table".
Allocation concealment (selection bias)	Low risk	Sealed and opaque envelopes. Quote: "One hundred opaque, sequentially numbered envelopes were used for the concealed allocation of patients to trial groups".
Blinding of participants and personnel (performance bias) All outcomes	Low risk	Double‐dummy placebo. Quote: "The patients, the statistician and the surgeon performing the qualification, operative procedure and follow‐up examination were all blinded with regard to which patients had received which form of treatment...Identical, nonmarked capsules with 100 mg ketoprofen or 100 mg placebo were prepared and coded in a professional pharmaceutical laboratory".
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Likely blinded from above as self‐report
Incomplete outcome data (attrition bias) All outcomes	Low risk	Only two dropouts from post group, unlikely to cause bias. Quote: "one hundred patients entered the trial, of whom 4 did not check‐in for the follow‐up examination. In all, complete data sets from 96 patients were statistically analysed".
Selective reporting (reporting bias)	Unclear risk	No protocol
Other bias	Low risk	Similar baseline characteristics