Yuan 2019.

Study characteristics
Methods	Study design: parallel‐group randomized controlled trial Sample size: 204 Country: China Setting: secondary care hospital Dates conducted: January 2017 to December 2018 Postoperative opioid used and delivery: IV pethidine Pain score collection: not reported Concurrent postoperative analgesics: none reported
Participants	Inclusion criteria Arthroscopic knee surgery for ligament reconstruction, meniscectomy, synovectomy, intra‐articular fractures reduction or other knee joint diseases Aged 18–65 years old Able to complete the pain visual analog scale and patient global assessment Exclusion criteria Contraindications to the surgery or drug Received analgesics within 1 week before the enrolment Coagulopathy or thromboembolic disease Intra‐articular hyaluronic acid injections within 9 months or corticosteroid within 3 months before the enrolment History of neurologic disease, knee surgery, chronic pain, and/or consumption of daily analgesics Pregnancy or breastfeeding
Interventions	Group EA (102 participants): meloxicam 15 mg oral 1 hour before surgery and 7.5 mg oral at 24 hours after surgery Group PA (102 participants): meloxicam 15 mg oral at 4 hours after surgery and 7.5 mg oral at 24 hours after surgery
Outcomes	Postoperative pain (0‐10 cm VAS from 24 hours before surgery to 48 hours after at rest and flexion) Adverse events (nausea, vomiting, constipation, drowsiness and dizziness at 48 hours) Pethidine consumption (mg at 48 hours)
Notes	Funding: not reported Declarations of interest: "The authors declare that they have no competing interest". Authors contacted: no Other: Early pain not included as not measured once post‐incision dose had taken effect within 6 hours. Nausea and vomiting not reported separately so not included
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Computer‐generated. Quote: "A blocked randomization method was used in this study, and the randomization sequence was created using SAS 9.0".
Allocation concealment (selection bias)	Low risk	Third party randomization. Quote: "...by an independent analyst"
Blinding of participants and personnel (performance bias) All outcomes	High risk	No placebo described
Blinding of outcome assessment (detection bias) All outcomes	High risk	No placebo described
Incomplete outcome data (attrition bias) All outcomes	Low risk	Intention‐to‐treat analysis. Quote: "The analysis in our study was performed according to the intention to treat (ITT) protocol".
Selective reporting (reporting bias)	Unclear risk	No protocol
Other bias	Low risk	Similar groups. Quote: "No difference was found regarding age and gender among the three groups, and BMI, surgery type, operative duration, pain VAS score at rest, pain VAS score at flexion as well as PGA score were also not different, (all P > 0.05)".