Tchartchian 2014.

Study characteristics
Methods	Unit of randomisation: women Method of randomisation: computer‐generated allocation schedule randomising 2:1 externally prepared and stored in envelopes Time of randomisation: not clearly stated but apparently intraoperatively Blinding: women blinded; surgeons analysing recordings blinded.
Participants	Women, aged 18 and older, with child‐bearing potential, undergoing laparoscopic myomectomy due to at least one myoma Exclusion criteria: pregnancy, lactation, previous open or closed myomectomy for myomas, active endometriosis, infection, inflammatory of bowel disease, pelvic inflammatory disease, frozen pelvis, hydrosalpinges N = 15 9 randomised to intervention and 6 to control Dropouts: 2 women (one from each group) withdrew consent prior to second‐look laparoscopy Age: mean age 35.8 in intervention group, and 44.3 in control group Baseline characteristics: the control group was significantly older (P = 0.002), had significantly higher gravidity (P = 0.021), and significantly larger number of previous vaginal deliveries (P = 0.038) than the control group. Indication for surgery: myomectomy for removal of at least one myoma Pre‐existing adhesions: 66.7% of the intervention group and 50.0% of the control group had pre‐existing adhesions (P = 0.662) Aetiology of pre‐existing adhesions: not clearly stated Microsurgery: surgery was based on the basic principles of microsurgery Additional surgical procedures: not clearly stated Location: Oldenburg, Germany Timing: November 2008 to May 2009
Interventions	Polyethylene glycol ester amine versus no treatment Other adjuvants used: none stated Second‐look laparoscopy: Timing: 8 to 12 weeks after initial surgery Surgeon blinded: not stated Adhesiolysis: performed in women still wanting to conceive
Outcomes	Adhesions at second‐look laparoscopy: Adhesion score (incidence, severity, and extent) Number of sites with adhesions Area of sites adherent to the uterus Adverse events
Notes	Funding provided by Covidien Video recording performed both at initial surgery and second look laparoscopy Power calculation not performed Clinicaltrials.gov register number: NCT00891657
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Quote: "Subjects were randomly assigned in a 2:1 ratio to the SprayShield Adhesion Barrier group or the control group according to a computer‐generated allocation schedule."
Allocation concealment (selection bias)	Low risk	Quote: "Patient blinding was performed intraoperatively through a third person with externally prepared randomisation using envelopes."
Blinding of participants and personnel (performance bias) All outcomes	Unclear risk	women blinded, operating surgeons not blinded
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Quote: "To blind the reviewer to subject and treatment, all videos were assigned a unique, blind code." "The videos were edited to remove the phase of the SprayShield Adhesion barrier application from treated subjects." "The videos were sent to an independent gynaecological surgeon to obtain standardised, unbiased evaluation of adhesion formation."
Incomplete outcome data (attrition bias) All outcomes	Unclear risk	2 out of 15 women lost to follow‐up
Selective reporting (reporting bias)	Low risk	All outcomes stated in trial registry reported ‐ NCT00891657
Other bias	High risk	Multiple statistically significant differences between the two groups at baseline Quote: "Due to the low number of patients' efficacy, data are not final regarding the performance of SprayShield Adhesion Barrier"