Lundgren 1989.

Methods	Study design: randomised Method of randomisation: randomised, balanced by age, sex and presence of diabetes; not blinded Losses to follow up: surgery not performed in 2 participants; exercise not carried out in 4 participants
Participants	Country: Sweden Number of participants: 75 Age: 40 to 80 years Sex: male and female Inclusion criteria: IC > 6 months; MWD < 600 m; BP in first toe > 30 mm Hg Exclusion criteria: rest pain or ulcer
Interventions	Treatment: surgical intervention, including thromboendarterectomy, bypass with synthetic y‐graft, saphenous vein or PTFE graft Control: dynamic leg exercises beyond appearance of pain, supervised by physiotherapist, 30 min sessions 3 times each week; encouraged to exercise at leisure Duration: 12 to 15 months
Outcomes	Treadmill test: pain‐free and MWD (4 km/h at 0 degree slope to maximum of 1000 m) ABI
Notes	A third group of 25 participants received surgery combined with an exercise regimen; these results were discussed but not included in formal meta‐analysis. No source of funding/sponsorship was reported.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	High risk	Quote: "The patients were randomized to one of the three treatment groups with help of an algorithm described by Pocock and Simon, accounting for the distribution of sex, age, and diabetes". Comment: Not appropriate randomisation method.
Allocation concealment (selection bias)	Unclear risk	No information provided.
Blinding (performance bias and detection bias) All outcomes	Low risk	See below.
Blinding of participants and personnel (performance bias) All outcomes	Low risk	No information provided. Most probably blinding was not obtained given the nature of the study. However, the lack of blinding was not likely to influence the outcome and the outcome measures.
Blinding of outcome assessment (detection bias) All outcomes	Low risk	No information provided. Probably no blinding existed, but a set protocol for assessment of follow‐up outcomes was defined, which minimises the risk of bias.
Incomplete outcome data (attrition bias) All outcomes	Low risk	Quote: "Two patients never underwent operations, and complete data at the follow‐up were not available for four patients of Op group. Five patients of the Op + Train group were not treated according to the protocol, and complete follow‐up data were not available for three of these patients. Four patients of the Train group never started their treatment according to the protocol, and follow‐up data for these patients are incomplete." Comment: Missing outcome data balanced in numbers across intervention groups.
Selective reporting (reporting bias)	Unclear risk	Insufficient information provided.
Other bias	Unclear risk	A potential risk of bias in relation to participant compliance with exercise treatment, but insufficient evidence that this problem would introduce bias was available.