Ibrahim 2012a.

Study characteristics
Methods	Double‐blind randomised controlled trial
Participants	Diagnosis: DSM‐IV major depressive disorder; score ≥ 22 score on MADRS; an inadequate response to at least two therapeutic trials of an antidepressant N: 42 Age: riluzole group M = 47.2 (SD = 13.3); placebo group M = 47.2 (SD = 13.0) Sex: riluzole group 38% female; placebo group 38% female Baseline depression severity: riluzole group MADRS = 32.7 (SD = 3.7); placebo group MADRS = 32.7 (SD = 5.7)
Interventions	4 weeks of treatment All patients first received a single open‐label IV infusion of 0.5 mg/kg of ketamine hydrochloride over 40 minutes, then were randomised to riluzole or placebo Riluzole (N = 21) dose titrated; initiated and maintained at 100 mg/day (50 mg twice daily). Dose could be flexibly increased in increments of 50 mg to a maximum of 200 mg/day Dose escalations continued on a weekly basis until treatment‐limiting side effects or completion of the study Placbeo (N = 21) daily capsules Concomitant treatment: No, patients had a 2‐week drug‐free period before commencing treatment
Outcomes	MADRS Response rate (≥ 50% reduction in MADRS scores) HRSD BDI VAS HAMA BPRS CADSS YMRS SSI Adverse events
Notes
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	No details given on randomisation procedure other than quote: ''patients were randomised''
Allocation concealment (selection bias)	Low risk	Quote: ''All study investigators, staff, and patients were blind to riluzole or placebo assignment''
Blinding of participants and personnel (performance bias) All outcomes	Low risk	Quote: ''All study investigators, staff, and patients were blind to riluzole or placebo assignment''
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	No details given on blinding of outcome assessment other than 'quote:'double blind''
Incomplete outcome data (attrition bias) All outcomes	Low risk	n reported alongside means at each time point in data provided by author through correspondence
Selective reporting (reporting bias)	High risk	Protocol unavailable and only one of 8 secondary measures reported
Other bias	Low risk	No other potential sources of bias identified