El‐Sayed 2004.

Methods	RCT ‐ with cross‐over for some unsuccessful ECVs.
Participants	Inclusion criteria: term singleton pregnancy with breech presentation. N = 59. Exclusion criteria: Maternal exclusion criteria included chronic hypertension, preeclampsia, placental abruption, placenta praevia, maternal cardiac disease, chorioamnionitis and previous uterine surgery. Fetal exclusion criteria included ruptured membranes, IUGR (EFW < 10th centile for GA by US), decreased AFI or oligohydramnios, fetal anomalies incompatible with life and an extended fetal head.
Interventions	Intervention:tocolytic: nitroglycerine ‐ nitric oxide donor (A3) ‐ parenteral. IV nitroglycerin (100 μg IV × 2). N = 30. Comparison:tocolytic: terbutaline ‐ beta stimulant (A1) ‐ parenteral. Terbutaline (0.25 mg SQ). N = 29. After successful ECV, the decision to induce then or wait for spontaneous labour was left to the doctor. After failed ECV, the options were intervention with the other drug in the trial, discharge with appointment for CS or immediate CS; the decision was left to the doctor.
Outcomes	Successful ECV; difficult ECV; palpitations; headaches; method of delivery; light‐headedness; flushing; reversion (back to breech after ECV).
Notes	We have used only data on initial "Failed ECV" because of the cross‐over element of this study. We are contacting study authors to clarify the other outcome data.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Randomisation done by a third party not involved in the trial. 30 labels bearing the word 'nitroglycerin' and 30 labels bearing the word 'terbutaline.' Labels were placed on 60 unmarked opaque envelopes, which were sealed, shuffled thoroughly and numbered sequentially.
Allocation concealment (selection bias)	Low risk	Labels were placed on 60 unmarked opaque envelopes, which were sealed and numbered sequentially.
Blinding of participants and personnel (performance bias) All outcomes	High risk	No blinding of participant or doctor. Differing routes of administration of drugs, IV or SQ, meant that people would know which drug was being administered.
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	Unclear whether the assessor was blinded.
Incomplete outcome data (attrition bias) All outcomes	Low risk	1 woman was assigned to terbutaline before it was confirmed that the baby was breech; excluded as fetus had a cephalic presentation. This was considered insufficient to influence the analysis.
Selective reporting (reporting bias)	Unclear risk	We did not assess the trial protocol.
Other bias	Low risk	No statistically significant differences in pretreatment maternal or fetal characteristics (see Table 1, on p 2053). These were maternal age, GA at ECV, multiparity, EFW,  body mass index, anterior placenta and ECV by maternal‐fetal medicine attending.