DEBELLUM 2012.

Methods	Study design: randomized controlled trial Method of randomization: participants were randomized (1:1) without stratification using computer‐generated assignments when they entered the angiographic suite Blinding: participants but not operators were blinded to the assigned intervention Exclusions postrandomization: 4 Losses to follow‐up: 0 Study enrollment period: September 2010 to March 2011 Cross‐over: 0
Participants	Country: Italy Setting: hospital ‐ single center Number of participants: 54 randomized, 50 analyzed, 28 participants meeting study criteria Age (mean ± SD): 66 ± 4 years Gender: male 74% Fontaine class: DEB: class IIb (64%), class III (28%), class IV (8%); PTA: class IIb (60%), class III (28%), class IV (12%) ABI (± SD): DEB: 0.55 ± 0.06, PTA: 0.57 ± 0.05 Inclusion criteria: Single or multiple lesions (stenosis or occlusion 3 cm to 30 cm) in the native SFA, the popliteal, or the BTK arteries, or with concomitant multilevel disease Exclusion criteria: Instent restenosis, aneurysms, acute thrombosis, pregnancy, life expectancy < 1 year, and absence of a patent crural artery Requiring provisional or bailout stenting after angioplasty as a result of flow‐limiting dissection or residual stenosis > 50%
Interventions	Uncoated balloon angioplasty: 27 participants Uncoated balloon angioplasty device: noncoated IN.PACT Admiral (SFA) and noncoated IN.PACT Amphirion (BTK), Medtronic DEB: 27 participants DEB device: IN.PACT Admiral (SFA) and IN.PACT Amphirion (BTK), Medtronic Drug used: paclitaxel (dose unknown) Vessels treated: femoropopliteal (75.4%) and BTK (24.6%) vessels Anticoagulation/platelets: ASA 100 mg/day and clopidogrel 75 mg/day for 3 days preprocedure. If participant was not receiving antiplatelet therapy preoperatively then clopidogrel 300 mg loading dose was administered. Heparin 5000 U administered after sheath insertion. Clopidogrel continued for 4 weeks Predilation before DEB: yes
Outcomes	Primary: Late lumen loss at 6 and 12 months in the SFA and BTK vessels Secondary: Binary restenosis (> 50%) Acute thrombotic occlusion of an artery within 48 hours of the procedure Any reintervention performed for thrombosis or restenosis (> 50% DS) of the target lesion after documentation of recurrent ischemic symptoms (TLR) Amputation at 6, 12, and 24 months
Notes	Late lumen loss was the difference in millimeters between the MLD immediately after the procedure and the MLD during follow‐up One third of participants received stents Substantial number of participants with TASC II C (33.6%) and D (10.6%) lesions Participants were followed‐up at 6, 12, and 24 months Duplex follow‐up in femoropopliteal region and angiography in the BTK arterial region Sponsor: no industry support
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	"Patients were randomized (1:1) without stratification using computer‐generated assignments when they entered the angiographic suite"
Allocation concealment (selection bias)	Low risk	"Patients were randomized (1:1) without stratification using computer‐generated assignments when they entered the angiographic suite"
Blinding of participants and personnel (performance bias) All outcomes	High risk	"Patients, but not operators, were blinded to the assigned intervention"
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	"Postoperative evaluation was deferred to different physicians not informed about the assigned intervention". Unclear what type of physicians performed those evaluations and what type of qualifications they had
Incomplete outcome data (attrition bias) All outcomes	Low risk	0 participants reported as lost to follow‐up and no missing data
Selective reporting (reporting bias)	Low risk	Restenosis rate not reported at 1 year. While this omission was concerning, the remaining outcomes were reported and as such this single omission does not, in our opinion, place the study at a high risk of reporting bias
Other bias	High risk	"Patients requiring provisional or bailout stenting […] were excluded from the study". "The decision to implant a nitinol stent in the SFA territory was left to the judgment of the operator and typically driven by lesion length and presence of severe calcification" The stent deployment criteria are unclear and approximately 37% of the treated lesions were also stented