Schellong 2012.

Study characteristics
Methods	Study design: double‐blind, randomised controlled trial; parallel assignment Country: Germany Setting: multi‐site Intention‐to‐treat: yes: participants who received ≥ 1 dose of trial medication and who had ≥ 1 valid measurement of PFWD under therapy
Participants	Number of participants randomly assigned: 561 (pentoxifylline 285, alprostadil 276) Number of participants analysed: 541 (pentoxifylline 272, alprostadil 269); completed study: 458 (pentoxifylline 233, alprostadil 225) Exclusions postrandomisation: 103 (pentoxifylline 52, alprostadil 51) Losses to follow‐up: 4 (pentoxifylline 3, alprostadil 1) Age (mean): overall 66.5 (SD 8.7) years; pentoxifylline 66.8 (SD 8.8) years, alprostadil 66.3 (SD 8.6) years Sex: male:female: 173:368 (pentoxifylline 89:183, alprostadil 84:185) Inclusion criteria: people with PAOD of the lower extremity in Fontaine stage II; maximum walking distance on treadmill (12%, 3 km/h) 30–150 m; stable IC ≥ 6 months standing with no acute shortening of walking distance over the past 3 months; stenoses or occlusions below femoral bifurcation (above‐knee or below‐knee type) confirmed by duplex ultrasound or angiography; ABI ≤ 0.90 with a decrease in systolic ankle pressure ≥ 10% after maximum loading (maximum walking distance on the treadmill at 3 km/h: 12%); person physically and mentally capable of participating in the trial; aged > 40 years, men and women; participant informed and given ample time and opportunity to think about her/his participation and provided written informed consent; participant willing and able to comply with all trial requirements Exclusion criteria: surgical or other interventional measures performed on affected extremity and prostaglandin treatment within the 6 months immediately before the trial; rest pain and necroses; systolic ankle pressure < 50 mmHg; change in maximum walking distance during 1‐week run‐in phase > ± 25% of baseline; successful physical walking training within the 6 months immediately before the trial; inflammatory vascular disease; polyneuropathy in diabetes mellitus; disease limiting walking distance (arthrosis, inflammatory disease of the joints, neurological disease, disease of the vertebral column, cardiopulmonary disease); history of pulmonary oedema; MI within previous 6 months; pregnancy or breastfeeding; known hypersensitivity to any components of trial medication or comparative drug; renal insufficiency, compensated retention (creatinine > 2.0 mg/dL); severe retinal haemorrhage; massive haemorrhage; known existing malignant disease; vasoactive concomitant medication (e.g. naftidrofuryl, pentoxifylline, buflomedil, cilostazol) or other prostaglandins; untreated or uncontrolled hypertension (systolic blood pressure ≥ 180 mmHg, diastolic blood pressure ≥ 110 mmHg); previous participation in the present trial
Interventions	Treatment: alprostadil (PGE1): 8 weeks total; 4 weeks of daily treatment (once daily IV infusion of 3 ampoules (20 μg) PGE1 in 50–250 mL physiological saline solution over 2 hours); 4‐week interval treatment period (twice weekly IV infusion of 3 ampoules (20 μg) of PGE1 in 50–250 mL physiological saline solution over 2 hours); received placebo tablets mimicking schedule of pentoxifylline Control: pentoxifylline: Trental, 8 weeks of 600 mg tablets bid; received placebo infusions of saline mimicking the schedule of alprostadil Duration: 8 weeks
Outcomes	PFWD, TWD, QoL (PAVK), side effects
Notes	Treadmill test: 12% grade and 3 km/h All data were retrieved from the ClinicalTrials.gov website, which offered no actual walking distances – only ratios – and no statistical analyses. A full report of the study including outcomes is currently being worked on by trialists and should provide additional information on bias issues and outcome data
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Insufficient information given to determine adequate random sequence generation.
Allocation concealment (selection bias)	Unclear risk	Insufficient information given to determine adequate allocation concealment.
Blinding of participants and personnel (performance bias) All outcomes	Low risk	Double‐blind (participants and investigator) using adequate techniques to maintain blinding.
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	Blinding of outcome assessors not discussed in abstract.
Incomplete outcome data (attrition bias) All outcomes	Low risk	All randomly assigned participants were accounted for, and intention‐to‐treat analysis included nearly all participants; detailed table given to describe exclusions and loss to follow‐up, although additional information should be provided regarding when these participants dropped out of the study.
Selective reporting (reporting bias)	Low risk	All initially indicated outcomes were reported.
Other bias	Unclear risk	Authors of the study reported that limitations of the study included early termination, leading to small numbers of participants analysed, and technical problems with measurement, leading to unreliable or uninterpretable data. Although the work was sponsored by UCB Pharma, it was indicated that the principal investigator of the study was not employed by the sponsor, and that the sponsor could not change communications or publications about the project.