Bozkurt 2006.

Methods	Design: randomised parallel‐group multicentre controlled clinical trial Setting : 12 major vascular centres Country: Turkey Loss to follow‐up: complete data from 162 participants presented in the analysis after a follow‐up of 4 and 24 weeks (19% of participants missing from analysis; 16% in iloprost group and 22% in sympathectomy group) Intention‐to‐treat: not stated
Participants	Number randomised n = 200 (iloprost n = 100; sympathectomy n = 100); number analysed at 24 weeks (iloprost n = 84; sympathectomy n = 78) Average age (range): 40.8 years (25 to 66) Gender: male: 97.6% No systemic comorbidities Inclusion criteria: Buerger's disease diagnosed by Shionoyas criteria Critical ischaemia ‐ rest pain/ischaemic ulcer Exclusion criteria: not mentioned
Interventions	Prostanoid iloprost; intravenous infusion 1 ng/kg/min, six hours/day for 28 days Open surgical unilateral sympathectomy at lumbar levels 2, 3, 4
Outcomes	Reported in paper, used in review relief of rest pain ulcer healing avoidance of major amputation mortality adverse events Reported, not used change in ulcer size 50% reduction in ulcer size analgesic requirements clinical improvement by SVS/ISCVS Sought from authors: not reported claudication distances ABPI, tissue oxygenation (TcPO2), toe pressure progression to minor amputation quality of life and functional status analysis of cost‐effectiveness Duration of follow‐up: 24 weeks
Notes	Funding: research fund of Istanbul University Comments: no conflicts of interest Trial was not registered
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Quote from report: "independent statistician prepared the randomisation list by the method of computer generated random numbers."
Allocation concealment (selection bias)	Low risk	Use of randomly generated list prepared by an independent statistician
Blinding (performance bias and detection bias) Subjective outcome (pain)	Unclear risk	Blinding of participants was not possible due to the nature of intervention ‐ a drug was being compared to a surgical procedure Pain is a subjective outcome and no data on objective recordings, e.g. using pain scores is reported; it was not reported if a blinded assessor or adjudication committee was used
Blinding (performance bias and detection bias) Objective outcomes (amputation, healing)	Low risk	Amputation and ulcer healing are definite clinical outcomes, evidence of risk of bias for these outcomes is minimal. Data on objective documentation of ulcer healing are presented, which also reduces the risk of bias
Incomplete outcome data (attrition bias) All outcomes	Unclear risk	162 of 200 randomised participants included in analysis; both groups had similar numbers of dropouts: iloprost n = 16, sympathectomy n = 22; study quoted missing participant data as "insufficient data for the remaining participants" with no further information on reasons so we judged attrition bias as unclear
Selective reporting (reporting bias)	Low risk	Although the trial was not registered, all proposed outcome measures were reported adequately
Other bias	Low risk	None detected by the review authors

ABPI: ankle brachial pressure index
 SVS/ISCVS: Society for Vascular Surgery and the North American Chapter of the International Society for Cardiovascular Surgery