Balasch 1997

Methods	Prospective randomised controlled trial. Pilot study
Participants	Infertile women with asymptomatic minimal or mild endometriosis (n = 60). Mean age: Treatment: 31.2 ± 3.8; Control: 32.4 ± 3.1 Inclusion criteria: at least 12 months of primary infertility, no previous pelvic surgery, minimal or mild endometriosis confirmed by laparoscopy Exclusion criteria: any previous pelvic surgery, pelvic disorders such as adhesions and tubal obstructions, in addition to endometriosis
Interventions	1. Pentoxifylline 400 mg: 1 tablet twice a day for 12 months (n = 30) 2. Placebo (n = 30). 12‐month duration and 12‐month follow‐up. During this time, participants received treatment for infertility problems (i.e. male problems, ovulatory problems, cervical mucus abnormalities, IUI, ovulation induction)
Outcomes	Pregnancy rates confirmed by ultrasound Miscarriage rate
Notes	1 dropout from the treatment group and 3 from the control group－all due to refusal to start treatment after randomisation. Number reported is 56. ITT is used for meta‐analysis Trial held from November 1993 to December 1995 Single‐centre study conducted in Spain Ethical approval and all consents obtained Funding source not reported
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	The investigators describe a random component in the sequence generation process that was using a computer random number generator
Allocation concealment (selection bias)	Unclear risk	Allocation described as being "designated". Authors contacted regarding this and confirmed concealment "computerised allocation"
Blinding (performance bias and detection bias) All outcomes	Low risk	Women are described as being blinded. Authors contacted regarding other blinded persons. They confirmed that participants were blinded, but investigators, outcome assessors and clinicians were not
Incomplete outcome data (attrition bias) All outcomes	Low risk	Only a small number of dropouts－4 participants lost; 1 in treatment, 3 in control. All explained－1 due to refusal and 3 due to failure to continue taking the medication. No ITT carried out
Selective reporting (reporting bias)	Unclear risk	Pregnancy rates were stated as the outcome of interest in the Methods section of the paper. However, miscarriage rates were given in the Results and were not mentioned in the Methods. 1 participant in each study group became pregnant, then miscarried, then became pregnant again. The first 2 pregnancies were not included in the analysis. Live births not reported
Other bias	High risk	Some women with other fertility issues apart from endometriosis were treated for these additional conditions (i.e. male factor (receiving bromocriptine), oligo‐ovulation (receiving ovulation induction and some additional IUI) poor post‐coital test, hyperprolactinaemia). Numbers of women in treatment and numbers of controls in each of these categories are given. However, these treatments may bias the results, as nearly double the control women in the additional treatment group received IUI compared with the treatment group