Trenkwalder 1995.

Methods	Study design: crossover RCT Study duration: not reported Study follow‐up period: 8 weeks (unclear if there was a washout period)
Participants	Country: Germany Setting: Department of Neurology in the Klinikum Grosshadern Health status: participants both with idiopathic RLS and with uraemic RLS were included (criteria for RLS similar to IRLSSG criteria); more than 5 PLM related ‐arousals/h of sleep; sleep latency of more than 25 min; sleep efficiency index < 85% in addition to symptoms of RLS Number: 11 (patients with uraemic RLS out of a total of 32 patients in the study) Mean age ± SD: 49 ± 11 years Sex (M/F): 6/5 Exclusion criteria: patients with signs of any other sleep disorder on polysomnography, especially narcolepsy and sleep apnoea syndrome; receiving neuroleptic or antidepressant medications; any severe additional illness or history of drug abuse; pregnant or lactating women and women without safe contraceptive methods
Interventions	Treatment group Madopar (L‐dopa 100 mg + benserazide 25 mg) given orally 1 hour before bedtime for 4 weeks; dose could be doubled after 2 weeks Control group Placebo given orally 1 hour before bedtime for 4 weeks Co‐interventions None
Outcomes	Polysomnographic studies Actigraphy Subjective rating (using modified 50‐mm Hamburger VAS) Physician's rating (CGI) Safety evaluation for rate of side effects of medications
Notes	Funding source not reported
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	No information on the randomisation procedure
Allocation concealment (selection bias)	Low risk	Double blind
Blinding of participants and personnel (performance bias) All outcomes	Low risk	Double blind
Blinding of outcome assessment (detection bias) All outcomes	Low risk	No clear statement of blinding of outcome assessment, but assessments of outcome (polysomnographic studies, Actigraphy, subjective rating and physician's rating) were carried out at different points of the study while participants and investigators were still blinded
Incomplete outcome data (attrition bias) All outcomes	Low risk	Data regarding different assessment methods were clearly shown and data regarding both groups (idiopathic versus uraemic RLS) were clearly shown
Selective reporting (reporting bias)	Low risk	No protocol but outcomes (improvement in RLS) assessment clearly outlined
Other bias	Unclear risk	Funding source not reported

CGI ‐ clinical global impression; CRP ‐ C‐reactive protein; HD ‐ haemodialysis; IRLSSG ‐ International RLS Study group; M/F ‐ male/female; PD ‐ peritoneal dialysis; QOL ‐ quality of life; RCT ‐ randomised controlled trial; RLS ‐ restless legs syndrome; SD ‐ standard deviation; SF‐36 ‐ short form 36 question survey; TSAT ‐ transferrin saturation; URR ‐ urea reduction ratio; VAS ‐ visual analogue scale