Iglesia 2010.

Methods	Trial design: multi‐centre randomised controlled trial involving 6 surgeons at 3 sites
Participants	Number of participants randomised: 65 (mesh group 32, no mesh group 33) Number of participants analysed: 65 Mean age: (mean ± SD): mesh group = 64.4 ± 10.8, no mesh group = 63.5 ± 8.9 Inclusion criteria: ≥ 21 years, grade 2 to 4 (POPQ) uterovaginal or vaginal prolapse and agreed to undergo vaginal surgery, available for 12 months' review, can complete questionnaires Exclusion criteria: multiple medical contraindications, short vagina, uterus > 12 weeks in size, desire future fertility, postpartum Setting: 3 University hospitals in USA Timing: January 2007 to August 2009
Interventions	Intervention: anterior Prolift or total vaginal mesh (Prolift) if point C or D on POPQ ≥ 3. No T incisions were performed and hysterectomy was performed if uterus was present (n = 32). Comparison: anterior colporrhaphy with uterosacral colpopexy with polytetrafluoroethylene sutures or sacrospinous colpopexy with Goretex sutures and hysterectomy performed if uterus was present (n = 33)
Outcomes	Primary outcome/s: objective failure rate at 1 year (any stage 2 or greater prolapse) Secondary outcome/s: subjective failure, reoperation for prolapse, surgery for mesh exposure, de novo dyspareunia, de novo SUI, responses to a range of quality of life questionnaires, postoperative complications, long‐term complications, serious adverse events
Notes	Intention‐to‐treat analysis: Manuscript states that 1 participant assigned to the mesh group did not receive mesh and was analysed as a member of the no‐mesh arm. This participant was analysed in the mesh group for 3‐year follow‐up. Sample size calculation: yes, calculated to require 90 participants (45 per arm) to provide 80% power to detect a 20% difference in primary outcome Trial registration: yes. Clinicaltrials.gov NCT00475540 Funding: supported by a grant from the AUGS Foundation and the MedStar Health Research Institute Intramural Grant Program. Prolift mesh kits used in this trial were donated by the mesh manufacturer (Ethicon Women Health and Urology, Somerville, New Jersey, USA). Conflicts of interest: study authors did not report any potential conflicts of interest. The ethics committee stopped the study before completion owing to predetermined stopping criteria of mesh erosion rate >15% being reached, with 65 of the desired sample size of 90 having undergone interventions.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Computer‐generated randomisations
Allocation concealment (selection bias)	Low risk	Opaque sealed envelopes were opened in the operating theatre after participant had received anaesthesia.
Blinding of participants and personnel (performance bias) All outcomes	Low risk	Trial co‐ordinator at each site and participants were blinded to treatment by use of sham dressings. Trial personnel not blinded (e.g. operating theatre staff, inpatient and office personnel) were instructed to not disclose treatment assignment to participants.
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Three‐ and 12‐month follow‐up examinations were conducted by evaluator blinded to group allocation.
Incomplete outcome data (attrition bias) All outcomes	Low risk	No losses to follow‐up 3 months postoperatively 27/32 in mesh group and 33/33 in no mesh group underwent 1‐year follow‐up. 25/33 in mesh group and 26/32 in no mesh group underwent 3‐year follow‐up.
Selective reporting (reporting bias)	Low risk	Prespecified outcomes appear reported on.
Other bias	Unclear risk	Before surgery, all demographic details were similar between the 2 groups, except group B had lower POPDI‐6 score than group A. The ethics committee stopped the study before completion owing to predetermined stopping criteria of mesh erosion rate > 15% being reached, with 65 of the desired sample size of 90 having undergone interventions.