Amir 2015.

Study characteristics
Methods	Parallel design, single‐centre, randomised controlled trial Computer‐based randomisation Cabergoline vs no intervention Setting: Israel
Participants	40 high‐risk women undergoing IVF/ET or IVF‐PGD, aged 18–40 years, serum E2 > 4000 pg/mL or the development of > 20 follicles > 12 mm in diameter Exclusion criteria: systemic disease and participating in other research studies Cabergoline group: 20 women Control group: 20 women
Interventions	Cabergoline group: cabergoline tablet 0.5 mg/day for 8 days from the day of hCG injection Control group: no cabergoline
Outcomes	Moderate and severe OHSS identified by the modified classification of Golan and colleagues (Golan 1989) assessed at day of ET, ET+7, ET+12 Severe OHSS (cabergoline group vs control group): 0/20 vs 2/20 Moderate OHSS (cabergoline group vs control group): 3/20 vs 10/20 Live birth rate: not reported Miscarriage rate (cabergoline group vs control group): 0/20 vs 1/20 Clinical pregnancy rate (fetal heartbeat) (cabergoline group vs control group): 2/20 vs 5/20 Multiple pregnancy rate (cabergoline group vs control group): 0/20 vs 1/20 Any other adverse effects of the treatment: not stated
Notes	Did apply coasting to both groups in about 50% of women if serum E2 > 5000 pg/mL
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Computer‐generated random numbers.
Allocation concealment (selection bias)	Unclear risk	Lack of sufficient data to permit judgement.
Blinding of participants and personnel (performance bias) All outcomes	High risk	Neither participants nor physicians blinded.
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Ultrasound experts were blinded.
Incomplete outcome data (attrition bias) All outcomes	Low risk	No missing data.
Selective reporting (reporting bias)	Unclear risk	No exclusions (no live birth rate mentioned).
Other bias	Unclear risk	Lack of sufficient data to permit judgement.