2.1. Searching and selection
We searched the electronic databases Pubmed (Medline),
Embase and the Cochrane Library from inception until 1st
September 2011. The medical literature was searched for RCTs
and cohort studies studying the effect of the mode of delivery on
neonatal mortality in preterm breech presentation. The following
terms were used: (((‘‘Breech Presentation’’[Mesh]) OR (breech))
AND ((‘‘Obstetric Labor, Premature’’[Mesh]) OR (‘‘Premature
Birth’’[Mesh]) OR (preterm) OR (premature)) AND ((vaginal
delivery) OR (cesarean) OR (caesarean) OR (abdominal delivery)
OR (‘‘Cesarean Section’’[Mesh])) AND ((‘‘Mortality’’[Mesh]) OR
(mortality) OR (death)) AND (Humans[Mesh])). We also performed
a manual search of reference lists from the retrieved studies.
Publication date restriction was imposed on studies published
before 1980, since perinatal care and interventions have greatly
improved over the years. Language restrictions were not applied.
Two authors (LB and LM) independently performed the search and
screened the abstracts of identified studies. Disagreement between
reviewers was resolved by consensus, and if needed the judgement
of a third author was decisive.