Shahrokh‐Tehraninejad 2016.

Study characteristics
Methods	RCT, 2 arms, 120 randomised Setting: Iran, hospital/fertility clinic, 2 centres Recruitment period: January 2013 to December 2014
Participants	Criteria related to previous IVF failure: yes, previous history of at least two failure of IVF/ICSI cycles Inclusion criteria: age < 40 years, presence of at least 4 embryos with good quality (grade 1), normal uterus in hysterosalpingography (HSG), sonography, hystrosonography or hysteroscopy, and at least 7mm endometrium thickness at suppository progesterone administration day. All patients had anatomically normal uterus cavity without any pathology like hyperplasia, malignancy, or endometritis in uterus. No one had received oral contraception agents or gonadotropin‐releasing hormone before FET cycle. Exclusion criteria: submucosal, intramural, and subserous al myoma greater than 5 cm, endometrioma equal to or greater than 3 cm, hydrosalpinx, bilateral obstruction of tube, less than 3‐4 embryos, endometrial tuberculosis, previous history of tuberculosis treatment, Asherman’s syndrome, BMI > 30 kg/m2, active vaginal or cervical infection, and underlying diseases like diabetes or systemic lupus erythematous.
Interventions	Study group: pipelle procedure day 21 of the cycle preceding the frozen embryo transfer cycle Control group: no procedure
Outcomes	Reported in paper: live birth, clinical pregnancy, miscarriage Obtained by author correspondence: ‐ (Authors stated during correspondence that all pregnancies were single however this seems unlikely given the average number of embryos transferred was 3, therefore this data was not used. Authors also stated that pain and bleeding were captured and "there was not any problem" ‐ however this data was not used as unclear how actively this information was captured)
Notes	Trial registration: IRCT201311065181N12 (registered Oct 2015, registered retrospectively) Additional concerns and comments: of 130 women assessed for eligibility, 120 were randomised ‐ which is a higher enrolment rate than usual, but not on its own a source of serious concern. Funding: None stated Author correspondence: emailed the corresponding author with some success (fedyeh_hagh@yahoo.com) Publication: full‐text.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Paper states quote: "Random selection for each method was performed by drawing a piece of printed paper from the plastic bag containing of equal number. Numbers of 1‐59 for treatment group and 60‐120 for control group were selected and By visiting each patient, randomly a number was out of plastic and according to the number, the group was selected."
Allocation concealment (selection bias)	High risk	No description of any safeguards in place to ensure allocation concealment and prevent someone from replacing the paper and selecting another
Blinding of participants and personnel (performance bias) All outcomes	High risk	No blinding employed
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Only objective outcomes reported
Incomplete outcome data (attrition bias) All outcomes	High risk	It is unclear whether any participants were excluded as one of the inclusion criteria appear to relate to a characteristic that would only be known after randomisation (7 mm endometrial thickness on progesterone administration day). It was not possible to confirm this with the authors.
Selective reporting (reporting bias)	High risk	No valid trial registration, and important outcomes such as adverse events not reported
Other bias	High risk	Trial registered retrospectively.