S
imões
R
et
al
.
490
R
ev
A
ssoc
M
ed
B
ras
2015; 61(6):489-494
bor, preterm OR labor, premature OR premature obs-
tetric labor).
Cochrane
Strategy:
cesarean section AND premature birth.
Studies retrieved (7/10/14)
TABLE 1
Number of studies retrieved with the search
strategies used for each scientific database.
Database
Number of studies
Primary
PubMed-Medline
4,816
Cochrane
124
Inclusion criteria for studies retrieved
Selection of studies, assessment of titles and abstracts ob-
tained from the search strategy in the consulted databas-
es was conducted by two researchers with skills in the
preparation of systematic reviews, both independent and
blinded, strictly observing the inclusion and exclusion
criteria previously established. All potentially relevant
studies were identified. Whenever the title and the sum-
mary were not enlightening, researchers sought the full
article.
Study design
Narrative reviews, case reports, case series and studies pre-
senting preliminary results were excluded from the as-
sessment. Systematic reviews and meta-analyzes were used
with the basic purpose of recovering references that per-
haps had been lost at first, from the initial search strate-
gy. Studies designed as cohort or controlled clinical tri-
als (randomized or not) were included.
Cohort study was defined as those with follow-up of
patients, the same history, and analysis of prognostic out-
comes.
Controlled clinical trials were evaluated according to
the Jadad score.
3
P.I.C.O. components
•
•
Patient:
nulliparous or multiparous women in labor
of a preterm singleton live fetus in cephalic position.
•
•
Intervention:
cesarean-section.
•
•
Comparison:
vaginal delivery.
•
•
Outcome:
the outcomes were divided into maternal
and newborn outcomes. The maternal outcomes in-
clude: maternal death or severe maternal morbidity (ad-
mission to ICU, sepsis and organ failure); bleeding com-
plications (postpartum hemorrhage, anemia, need for
blood transfusion after childbirth) and complications
of surgical wound (wound infection, dehiscence or pain).
Late maternal outcomes were also included, such as
complications in breastfeeding, perineal pain, abdomi-
nal pain, dyspareunia, urinary incontinence, fecal in-
continence, perineal trauma, and genital dystopia.
Newborn outcomes, in turn, include: perinatal or neo-
natal death (excluding cases of death related to fatal
fetal abnormalities), neonatal morbidity, such as sei-
zures (occurring within the first 24 hours of birth or
that require two or more drugs to control), Apgar score,
birth asphyxia, respiratory complications, infection,
need for admission into neonatal intensive care unit,
neonatal encephalopathy, trauma at birth (bone frac-
tures, subdural hematoma, cerebral or intraventricu-
lar hemorrhage), spinal cord injury, peripheral nerve
injury (e.g., brachial plexus injury), disabilities in child-
hood, hypotonia, intubation or need for ventilation
for at least 24 hours, and need for tube feeding for four
days or longer.
Language
We included studies available in Portuguese, English,
French or Spanish.
According to publication
Only studies with full text available were considered for
critical assessment.
Studies selected in the first assessment
After entering the search strategy in the primary databas-
es (PubMed-Medline and Cochrane), the assessment of
titles and abstracts led to the selection of nine studies.
Evidence selected in critical evaluation and exhibition of results
The studies considered for full text reading were critical-
ly assessed according to inclusion and exclusion criteria,
study design, P.I.C.O., language and availability of the
full text.
Results pertaining clinical status will be displayed in-
dividually, showing the following items: clinical question,
number of studies selected (according to inclusion crite-
ria), description of the studies (Table 2), results and sum-
mary of the available evidence. References related to the
studies included are shown in Table 4.
After applying the inclusion and exclusion criteria,
the evidence selected in the search and defined as ran-
domized controlled trials (RCT) were subjected to an ap-
propriate checklist for critical assessment (Table 3). Crit-
ical assessment of RCTs allows to classify them according