Taylor 2005.

Methods	Two centre study Allocation by sealed sequentially numbered opaque envelopes containing computer‐generated random numbers Patients blinded but care providers not blinded Power calculation: performed No losses to follow up, and all patients were analysed in groups to which they were assigned Source of funding: not mentioned
Participants	Setting: two university hospitals, London, UK Inclusion criteria: symptomatic fibroids, uterine size at least 14 weeks, and patient request for myomectomy Exclusion criteria: history of bleeding disorder, concurrent anti‐coagulant therapy, haemoglobin less than 10.5g/dl at the time of surgery, pre‐malignant endometrial histology n = 28 Mean age: 39.5 years (SD 4.7) in control arm and 42.6 years (SD 6.7) in treatment arm Ethnicity: not described Number of fibroids: range 1‐34
Interventions	Treatment arm (n = 14): a number one polyglactin suture tied around the cervix to occlude uterine artery and left there after surgery, plus a polythene tourniquet tied round the infundibulopelvic ligament and removed after the operation Control arm (n = 14): no treatment added to the normal myomectomy procedure Type of operation: laparotomy
Outcomes	Perioperative blood loss, need for blood transfusion, operative morbidity Blood loss was assessed by weighing swabs and measuring blood collected by suction
Notes	Indications for myomectomy: symptomatic uterine fibroids Preoperative GnRH agonists prescribed to anaemic patients to increase preoperative haemoglobin to more than 10.5g/dl Authors contacted July 2015
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Allocation by sealed sequentially‐numbered opaque envelopes containing computer‐generated random numbers
Allocation concealment (selection bias)	Low risk	Adequate
Incomplete outcome data (attrition bias) All outcomes	Low risk	No incomplete outcome data
Selective reporting (reporting bias)	Unclear risk	We do not have access to the study protocol
Other bias	High risk	Statistically significant difference in mean age of the two groups
Blinding of participants and personnel (performance bias) All outcomes	Low risk	Patients blinded but care providers not blinded
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	Unclear if outcome assessment blinded