Lee 2017b.

Methods	Study design: randomised controlled trial Study grouping: parallel Ethics and informed consent: yes Follow‐up period: 90 days Sample size estimate: yes ITT analysis: no number randomised: 102 number analysed: 102 Funding: not company funded Pre‐registration: yes
Participants	Location: Canada Intervention group: n = 53control group: n = 49 Mean age: intervention group = 69 ± 10control group = 68 ± 10 Inclusion criteria: patients with 1 of the following 3 risk factors for SSIs were enrolled in the trial: obesity defined as a BMI of > 30 kg/m², previous femoral artery exposure, or presence of minor or major ischaemic tissue loss. Exclusion criteria: patients with pre‐existing groin infection, a known allergy to dressing material, or those who could not be followed postoperatively were excluded from the study.
Interventions	Aim/s: to perform an RCT to study the role of NPWT on SSI in primarily closed groin incisions after lower extremity revascularisation in vascular surgery patients Group 1 (NPWT) intervention: NPWT remained on until either hospital discharge or postoperative day 8, whichever occurred earlier. Group 2 (control) intervention: standard gauze dressing (the dressing removed on postoperative day 2, and then had daily dressing changes with inspection of the wound) Study date/s: August 2014 to December 2015
Outcomes	the incidence of SSI within 30 days of revascularisation duration of hospital stay SSI within 90 days reoperation and readmission rate owing to SSI within 90 days mortality within 90 days Validity of measure/s: SSI was diagnosed using the CDC guideline as a superficial or deep infection. The Szilagyi classification of vascular wound infection was also used to classify the infection. Time points: once discharged, both groups were followed up in the clinic at 30 and 90 postoperative days.
Notes
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Eligible patients were randomised to NPWT or a standard sterile gauze dressing using an internet‐based software, sealedenvelope.com (London, UK), using block randomisation.
Allocation concealment (selection bias)	Unclear risk	Not reported
Blinding of participants and personnel (performance bias) All outcomes	Low risk	Evidence for participants: not possible Comment: unlikely to affect outcomes Evidence for personnel: not possible Comment: unlikely to affect outcomes
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Wounds were inspected at each clinic visit by a wound specialist nurse who was blinded to the treatment groups. If she was uncertain, the staff physician determined the presence or absence of an SSI. An SSI could also be diagnosed by the patient care team if there were clinical signs and symptoms of infection.
Incomplete outcome data (attrition bias) All outcomes	Low risk	102 participants were enrolled and analysed.
Selective reporting (reporting bias)	Low risk	Planned outcomes reported. Protocol registered on ClinicalTrials.gov (NCT02084017).
Other bias	Low risk	No other biases detected.