Sykes 2000.

Methods	Study design: RCT. Method of randomisation: Sealed envelopes on day of operation. Blinding of assessment: Unclear. Exclusions post‐randomisation: None stated. Losses to follow‐up: None stated.
Participants	Country: UK. Setting: Hospital. No. of participants: 50. Age mean (years): 49 (32 to 65) no tourniquet; 47 (29 to 80) tourniquet. Gender: M:F 1:4 no tourniquet; 1:2 tourniquet. Inclusion criteria: Unilateral SFJ ligation strip and avulsions. Exclusion criteria: History of DVT or deep venous insufficiency, extensive anterior thigh veins, saphenopopliteal reflux. Groups "similar" for age, sex and VV grade.
Interventions	Treatment: Sterile Loquist (Boazal cuff) inflated to 120 mm Hg. Control: no Boazal cuff. All patients had SFJ ligation strip using disposable Astra vasistrip and avulsions. 30° head down tilt. Cotton wool and crepe bandages applied for 48 hrs then TED stockings for 2 weeks.
Outcomes	Median blood loss: Treatment: 0 ml. Control: 125 ml (range 20 to 300); P < 0.01. Mean operative time: Treatment: 30 mins (range 11 to 47). Control: 37 mins (range 18 to 50); P < 0.01. Median area of bruising thigh at 7 days: Treatment: 77 cm² (range 30 to 429). Control: 179 cm²; P < 0.01. Pain (VAS 0 to 10): No significant difference. Patient activity over first week (VAS 0 to 7): No significant difference. Cosmetic appearance at 6 weeks: No significant difference.
Notes	3 patients had temporary sensory neuralgia ‐ one in treatment group and two in control group. Two patients had wound complications (one infection and one haematoma ‐ both in I2.
Risk of bias
Bias	Authors' judgement	Support for judgement
Allocation concealment (selection bias)	Unclear risk	B ‐ Unclear