Casini 2006.

Methods	Parallel‐group, randomised, controlled, single‐centre trial Power calculation not reported Approved by the hospital's ethics committee No source of funding or conflict of interest reported
Participants	Country: Italy Setting: AGUNCO Obstetrics and Gynecology Centre, Rome Population: women referred to the centre from January 1998 to April 2005 for fertility problems were examined for inclusion. All women underwent routine examinations including the study of ovarian function (FSH, luteinising hormone, oestradiol and progesterone concentrations); prolactin, free triiodothyronine, free thyroxine and thyroid‐stimulating hormone concentrations; postcoital test; TVUS; HSG and analysis of the partner's semen. The TVUS was performed to diagnose the presence of uterine fibroids. After these examinations, all women found to be affected by uterine fibroids excluding all other causes of infertility were asked to participate in the study. Type of subfertility: infertility for ≥ 1 year (range: 1 to 5 years); no further clarification on primary versus secondary subfertility Mean age: women with submucous fibroids alone: 31.4 ± 2.5 years; women with mixed submucous‐intramural fibroids: 32.2 ± 2.5 years Number recruited: 193 women Number participants: 181 women Number participants with submucous fibroids only: 52 women Number participants with mixed submucous‐intramural fibroids: 42 women Inclusion criteria: aged ≤ 35 years; infertility for ≥ 1 year; presence of 1 knot or fibroid of diameter ≤ 40 mm (or both) and absence of other causes of infertility at the performed examinations Exclusion criteria: presence of ≥ 2 knots or fibroids of diameter > 40 mm (or both); body weight > 20% of normal weight; and use of medication containing oestrogens, progestins or androgens within 8 weeks prior to study Duration of study: 86 months; conducted January 1998 to April 2005
Interventions	Intervention group: hysteroscopic surgery to remove fibroids (52 women) Control group: not treated (42 women) Participants were examined by TVUS 3 months after surgery. Participants who underwent surgery were suggested to abstain from having sexual intercourse for 3 months and then to start having regular fertility‐oriented intercourse. Participants who did not undergo surgery were asked to immediately start having regular fertility‐oriented intercourse (intercourse during the 6‐day fertile interval ending on the day of ovulation). Participants were monitored for up to 12 months after study commencement.
Outcomes	Clinical pregnancy defined by the visualisation of an embryo with cardiac activity at 6–7 weeks of pregnancy Miscarriage classified as clinical loss of an intrauterine pregnancy between the 7th and 12th weeks of gestation
Notes	Authors stated that the differences in pregnancy rates between the comparison groups were statistically significant for the women with submucous fibroids (P < 0.05), which is in contrast with the calculation of the results in Review Manager 5. The definition of knot was unclear: it could not be clarified since we were unable to contact the study authors. Unclear whether a hysteroscopy was done in all women to confirm the exact position of the ultrasonically detected fibroids.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Quote: "Subsequently, women of each group were randomized into two subgroups, according to a randomisation table." Comment: low risk of selection bias related to random sequence generation.
Allocation concealment (selection bias)	Unclear risk	Method not stated: no further clarification obtained from the study authors. Comment: unclear risk of selection bias related to allocation concealment.
Blinding of participants and personnel (performance bias) All outcomes	Low risk	Method not stated: no further clarification obtained from the study authors. Comment: not applicable as this is a surgical study with unequivocal outcomes.
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Method not stated: no further clarification obtained from the study authors. Comment: not applicable as this is a surgical study with unequivocal outcomes.
Incomplete outcome data (attrition bias) All outcomes	Low risk	Quote: "One hundred and ninety‐three patients were diagnosed as affected by uterine fibroid excluding all other causes of infertility and met the requirements of the inclusion and exclusion criteria. Of these, 181 decided to participate in the study. Among the 181 patients, 52 had submucosal fibroids (SM group) while 45 had intramural fibroids (IM group), 11 had subserosal fibroids (SS group), 42 had a mix of submucosal–intramural (SM‐IM group) and 31 patients had a mix of intramural–subserosal fibroids (IM‐SS group)." Quote: "Out of 181 women, 68 become pregnant." Comment: low risk for attrition bias.
Selective reporting (reporting bias)	High risk	All specified outcomes reported in the results section. Nevertheless, the published report failed to include results for the live birth rate, which was the primary outcome of interest that would be expected to have been reported for a trial on fertility treatment conducted over a 7‐year period.
Other bias	Unclear risk	The mean ages and duration of infertility in the intervention and control group of women with submucous fibroids were not reported. No further clarification by the authors was obtained. It was unclear whether there might have been imbalance in the baseline characteristics between the comparison groups. Failure to do a hysteroscopy in all women to confirm the position of the ultrasonically detected fibroids could have caused information bias.