Tandoc 2011.

Methods	Parallel group RCT.
Participants	In the USA, 78 participants aged 18 to 78 years undergoing elective arthroscopic shoulder surgery were included. Participants with coagulopathy, allergy to local anaesthetics, hypertension, peripheral neuropathy or chronic obstructive pulmonary disease were excluded.
Interventions	Block All participants underwent nerve stimulator‐guided brachial plexus block with bupivacaine 0.5% with epinephrine 1:200,000 40 ml. Dexamethasone/placebo Dexamethsone group: dexamethasone 8 mg perineurally. Placebo group: normal saline 2 ml perineurally. Intraoperative anaesthesia/analgesia Midazolam 1‐2 mg and/or fentanyl 50‐100 micrograms was administered before the block. Anaesthesia was induced with propofol 2 mg/kg and maintained with sevoflurane 1.0‐1.5 MAC. Postoperative analgesia Acetaminophen 325 mg + hydrocodone 7.5 mg 1‐2 tablets was administered if pain score was greater than 3. If pain persisted, ibuprofen 400 mg was administered.
Outcomes	Otcomes of interest for review Duration of sensory block defined as time of discharge to the time the patient experienced pain at or greater than 3. Duration of motor block defined as the time from discharge to the time when the patient was able to abduct the arm at least 2 inches away from the body. Participant satisfaction measured on a 5‐point scale. Other outcomes Number of acetaminophen 325 mg + hydrocodone 7.5 mg tablets taken in the first 72 hours after surgery. Incidence of adverse events.
Notes	This was a three‐arm study comparing dexamethasone 8 mg, dexamethasone 4 mg and placebo. In order to avoid unit of analysis errors, we chose to include the dexamethasone 8 mg arm and exclude the dexamethasone 4 mg arm since 8 mg is the dose most commonly used in clinical practice. Funding: provided by Buffalo Anesthesiology Associates. Conflicts of interest: no information provided.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Participants were randomized using a randomization table.
Allocation concealment (selection bias)	Unclear risk	No indication of how group allocation was concealed.
Blinding of participants (detection bias)	Unclear risk	No indication whether participants were blinded.
Blinding of personnel (detection bias)	Unclear risk	No indication whether personnel were blinded.
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	Nurse who assessed the outcomes after discharge was blinded. Unclear whether the anaesthesiologist assessing the incidence of postoperative adverse events was blinded.
Incomplete outcome data (attrition bias) All outcomes	Low risk	No participants were excluded from the dexamethasone group and two were excluded from the placebo group.
Selective reporting (reporting bias)	Low risk	No protocol available. All outcomes reported as stated in the methods section.
Other bias	Low risk	Appears to be free of any other bias.