RAMB Novembro/Dezembro 2015

esarean

delivery

and

prematurity

ssoc

ras

2015; 61(6):489-494

491

to the Jadad score, so that Jadad < 3 trials are considered

inconsistent (

), and those with scores ≥ 3, consistent (

For critical analysis of non-randomized studies, among

them prospective observational studies, we used the New-

castle-Ottawa scale.

For results with available evidence, wherever possible

the following specific items are defined: population, in-

tervention, outcomes, the presence or absence of benefit

and/or damage and controversies.

Cost issues will not be included in the results.

The results will be presented preferably in absolute

data, absolute risk, number needed to treat (NNT), or

number needed to harm (NNH), and occasionally in mean

and standard deviation.

TABLE 2

Worksheet used for description of studies

included, and exposure of the results.

Evidence included

Study design

Population selected

Time of follow-up

Outcomes considered

Expression of results: percentage, risk, odds, hazard ratio

TABLE 3

Critical assessment script for randomized

controlled trials (Checklist).

Study data

Reference, study design,

Jadad, strength of

evidence

Sample size calculation

Estimated differences, power,

significance level, total number of

patients

Patient selection

Inclusion and exclusion

criteria

Patients

Recruited, randomized, prognostic

differences

Randomization

Description and blinded

allocation

Patient follow-up

Time, losses, migration

Treatment protocol

Intervention, control and

blinding

Analysis

Intention to treat, analyzes of intervention

and control

Outcomes considered

Primary, secondary,

measuring instrument of

the outcome of interest

Result

Benefit or harm in absolute data, benefit

or harm on average

esults

Clinical question

Is the performance of a C-section in singleton preterm

pregnancies in cephalic presentation related to lower ma-

ternal, perinatal and neonatal morbidity and mortality

compared with vaginal delivery?

Evidence selected

TABLE 4

Selection process.

Type of publication

Included

Non-concurrent cohort studies

5-15

The main reasons for the exclusion of works were: the un-

availability of the full text; a study design other than lon-

gitudinal observational (retrospective or prospective) or

experimental (controlled clinical trials, randomized or

not) studies.

Results of the evidence selected

Of the 4,940 articles initially retrieved, nine were select-

ed to support the summary of evidence concerning ma-

ternal, perinatal and neonatal morbidity and mortality,

according to mode of delivery chosen for labor resolu-

tion of preterm fetuses in cephalic presentation. Studies

included are shown in Table 4.

1. Malloy MH, et al. (

•

Design: non-concurrent observational longitudi-

nal study.

•

Population: Women who gave birth (through C-

-section or vaginally) to fetuses (n=1,765) weigh-

ing less than 1,550 g in seven neonatal intensive

care centers.

•

Outcome: neonatal morbidity and mortality.

•

Result: the cesarean delivery rate for newborns

weighing 501-750 g was 32.5% and for those weigh-

ing between 751 to 1.000 g, 52.4%. With respect to

neonatal mortality in cases of C-section, there was

a 53% rate for newborns weighing 501-750 g, com-

pared to 64% among those born vaginally. Howev-

er, analyzing the newborns weighing between 751

to 1,000 g, the mortality rate for those born by C-

-section was 14.4% compared to 7.8% for births that

occurred vaginally. The incidence of intraventric-

ular hemorrhage was significantly lower among

newborns weighing between 1,251 and 1,500 g

born by cesarean delivery compared to vaginal

births (11.8

versus

18.9%, respectively). After adjust-

ment performed using logistic regression (consid-

ering gestational age, breech presentation, pres-

ence or absence of labor), no difference was found

in neonatal mortality and intraventricular hem-

orrhage between the two modes of delivery; OR=1