Research www. AJOG.

org

OBSTETRICS
A randomized trial of preinduction cervical ripening:
dinoprostone vaginal insert versus
double-balloon catheter
Antonella Cromi, PhD; Fabio Ghezzi, MD; Stefano Uccella, MD; Massimo Agosti, MD;
Maurizio Serati, MD; Giulia Marchitelli, MD; Pierfrancesco Bolis, MD

OBJECTIVE: We sought to compare the efficacy of a double-balloon 1.26 –3.91). There was no difference in cesarean delivery rates (23.8%
transcervical catheter to that of a prostaglandin (PG) vaginal insert vs 26.2%; odds ratio, 0.88; 95% confidence interval, 0.47–1.65). Oxy-
among women undergoing labor induction. tocin and epidural analgesia were administered more frequently when a
double-balloon device was used. Uterine tachysystole or hypertonus
STUDY DESIGN: In all, 210 women with a Bishop score ⱕ6 were as-
occurred more frequently in the PGE2 arm (9.7% vs 0%, P ⫽ .0007).
signed randomly to cervical ripening with either a double-balloon device
or a PGE2 sustained-release vaginal insert. Primary outcome was vagi- CONCLUSION: The use of a double-balloon catheter for cervical ripen-
nal delivery within 24 hours. ing is associated with a higher rate of vaginal birth within 24 hours com-
pared with a PGE2 vaginal insert.
RESULTS: The proportion of women who achieved vaginal delivery in
24 hours was higher in the double-balloon group than in the PGE2 Key words: double-balloon catheter, labor induction, mechanical
group (68.6% vs 49.5%; odds ratio, 2.22; 95% confidence interval, method, prostaglandins, ripening time

Cite this article as: Cromi A, Ghezzi F, Uccella S, et al. A randomized trial of preinduction cervical ripening: dinoprostone vaginal insert versus double-balloon
catheter. Am J Obstet Gynecol 2012;207:125.e1-7.

O ver the past several decades, the in-

cidence of labor induction has
continued to rise to the point that in de-
orities for research in obstetrics, with a
view to improving the quality of care and
outcomes.
that highly reputed scientific societies
came to opposite recommendations re-
garding the use of balloon catheters for
veloped countries the proportion of in- Methods that have historically been iatrogenic cervical ripening in labor in-
fants delivered following induction of la- applied to induction of labor in the pres- duction guidelines.4,5 Interpretation of
bor can be as high as 1 in 4. Induction of ence of an unripe cervix can be classified comparative data between mechanical
labor has a major health impact on the into 2 categories: (1) mechanical meth- devices and topical PGE2 is hindered by
woman and on her baby, can affect the ods that are thought to work by both di- several factors including often under-
satisfaction with the birth experience, rectly dilating the cervix and by promoting powered trials, use of different measures
and places strain on the organization of endogenous prostaglandin (PG) release; of effectiveness (cervical change vs a va-
care in labor wards. For all of these rea- and (2) application of pharmacologic rip- riety of delivery outcomes), use of dispa-
sons the policies of induction, chiefly in- ening agents such as exogenous PGs. Over rate regimens, and methods of induction
dications and methods, remain key pri- 50 years after discovery of pharmacologic either used alone or in combination.2,3
preparations of PGE2 and several centuries Controlled-release inserts have be-
after the first description of mechanical di- come the preferred vehicle for delivering
From the Department of Obstetrics and
Gynecology, University of Insubria (Drs Cromi, lation of the cervical canal to achieve deliv- vaginal PGs in many settings, probably
Ghezzi, Uccella, Serati, Marchitelli, and Bolis) ery, we continue to search for the optimal due to reduced need of repeated vaginal
and the Department of Neonatology and method that will modulate the unfavor- examination, rapidity and ease of re-
Neonatal Intensive Care Unit, Del Ponte able to favorable cervix, improving the ul- moval when active labor is established or
Hospital (Dr Agosti), Varese, Italy.
timate outcome of labor and ideally elimi- when complications ensue, and the re-
Received Feb. 1, 2012; revised April 15, 2012;
accepted May 24, 2012.
nating risks to the mother and fetus. ported reduction in the need of instru-
The authors report no conflict of interest.
Despite extensive studies, uncertain- mental vaginal deliveries and in the use
ties remain about how best to apply vag- of oxytocin augmentation compared to
Reprints: Antonella Cromi, PhD, Department of
Obstetrics and Gynecology, University of inal PGs in terms of their vehicle, dosage, vaginal PGE2 gel or tablet.1
Insubria, Piazza Biroldi 1, 21100 Varese, Italy. and timing.1 Similarly, the evidence for In view of the increased frequency of
[email protected]. the use of mechanical methods for in- use of sustained-release pessaries and the
0002-9378/$36.00 ducing labor is confused by a large num- very scarce comparative data between
© 2012 Mosby, Inc. All rights reserved.
http://dx.doi.org/10.1016/j.ajog.2012.05.020
ber of small studies using different com- this device and mechanical methods, we
parators and protocols,2,3 to the extent decided to design a randomized study

AUGUST 2012 American Journal of Obstetrics & Gynecology 125.e1

Research Obstetrics www.AJOG.org

comparing this vaginal PGE2 prepara- then pulled snugly back against the os. The achieve 7 contractions in 15 minutes or
tion with a double-balloon device specif- second (vaginal) balloon was then inflated up to a maximum infusion of 30 mIU/
ically developed and engineered for rip- with 50 mL of saline to apply pressure on min. Once in active labor (cervix at least 5
ening the cervix and licensed for use in the vaginal side of the cervix. The external cm dilated), standardized intrapartum
obstetric care. end of the device was taped without trac- management was carried out by the staff
tion to the medial aspect of the woman’s members in charge of the labor and deliv-
thigh. After completion of the device ery unit according to institutional proto-
M ATERIALS AND M ETHODS placement, patients underwent continu- cols. Slow progress of labor was defined as
Patients with unfavorable cervices, sched- ous fetal heart rate monitoring for 30 ⱕ1 cm of cervical progress in 2 hours. If the
uled to undergo labor induction from Au- minutes and then were allowed to ambu- membranes were intact, amniotomy was
gust 2010 through October 2011 at the late. The double-balloon device was left first performed. If there was still no prog-
Obstetrics Department of University of In- in place for approximately 12 hours, as ress 1 hour after amniotomy, augmenta-
subria, Varese, Italy, were screened for per manufacturer’s recommendation. tion by oxytocin infusion was started. Oxy-
study inclusion. All recruited women pre- Reasons for removing the catheter in- tocin was initiated immediately if the
sented with a singleton gestation, vertex cluded: (1) the maximal time allowed membranes were already ruptured and the
presentation, Bishop score ⱕ6, intact for cervical ripening to take place had cervix remained unchanged on 2 consecu-
membranes, gestational age ⱖ34 weeks, elapsed; (2) spontaneous rupture of mem- tive pelvic examinations conducted 2
and reassuring fetal heart tracing on branes occurred; (3) the balloon was ex- hours apart. Oxytocin was also adminis-
admission. Women with antepartum pelled spontaneously; (4) patients entered tered when the second stage of labor was
bleeding, intrauterine fetal death, prior labor (defined as rhythmic, firm, ade- ⬎2 hours in nulliparous and 1 hour in
uterine scars, positive vaginal or rectal quate-quality uterine contractions occur- multiparous women.
group B streptococcus screening cultures, ring at a frequency of ⱖ4 in 30 minutes and Tachysystole was identified when there
placenta previa, or any other contraindica- lasting ⱖ40 seconds, with an effaced cervix
were ⬎5 contractions per 10 minutes for at
tion to vaginal delivery were excluded. All and a cervical dilatation ⱖ3 cm); or (5) fe-
least 20 minutes. Hypertonus was defined
the participants gave written informed tal distress was suspected.
as a single contraction lasting at least 2
consent and local institutional review In the group randomly assigned to
minutes. Failed induction was diagnosed
board approval was obtained before the pharmacologic ripening, the PGE2 slow-
when women did not progress into the ac-
beginning of the study. release vaginal insert was placed high
tive phase of labor despite adequate con-
Once the decision to induce labor was in the vaginal fornix and patients were
traction pattern, after amniotomy and a
made, women who whished to partici- monitored for uterine activity and fetal
minimum of 10 hours of oxytocin infu-
pate in the study were recruited by a staff heart rate for at least 1 hour and then
sion. Failure to progress was defined as un-
physician. Participants were randomly allowed to ambulate. Primary reasons
allocated to preinduction cervical ripen- for discontinuation of the PGE2 insert changed cervical dilatation in a 4-hour in-
ing with either a double-balloon catheter included: (1) completion of maximum terval despite oxytocin augmentation and
or a 10-mg controlled-release dinopros- recommended dosing period (24 hours); a sustained uterine contraction pattern or
tone vaginal insert. The randomization (2) onset of labor; or (3) uterine contrac- no descent after 1 hour during the second
sequence was created using a computer- tile abnormalities or nonreassuring fetal stage of labor. Bishop score was calculated
generated randomization scheme with heart rate patterns that prompted clini- prior to labor induction and after removal
1:1 allocation for each arm of the study. cal intervention. of the ripening devices by the attending
The random allocation sequence was Soon after expulsion or removal of the physician or a member of the resident staff.
concealed from those responsible for re- double-balloon device or 1 hour (as per Blood loss at vaginal delivery was esti-
cruiting participants into the study (at- the manufacturer’s recommendation) mated using an underbuttocks drape with
tending physicians) by keeping it in a file after completion of maximum recom- a graduated pouch for measurement.
cabinet with access restricted to research mended dosing period of the PG pessary, Blood loss at cesarean delivery was esti-
staff. A research assistant disclosed the oxytocin was administered to those mated from the content of suction devices.
nature of the assignment only after women who were not in labor. It is our All outcome data were obtained con-
enrollment. policy to perform amniotomy before ini- current with patient care and recorded
In the group assigned to mechanical rip- tiating induction of labor with oxytocin, by the investigators team. The primary
ening, a double-balloon catheter (Cook unless the fetal station is considered too outcome measure was vaginal delivery
Cervical Ripener Balloon; Cook OB/GYN, high to safely perform amniotomy or the within 24 hours of the initiation of rip-
Spencer, IN) was inserted into the cervical cervix is closed. Oxytocin was adminis- ening. Other outcome variables included
canal under direct visualization during a tered using a standard dose regimen in improvement in the Bishop score after
sterile speculum examination. Once both all patients. The induction protocol at ripening, cesarean delivery rates, ripening-
balloons entered the cervical canal, the first our institution specifies starting oxyto- to-delivery interval, oxytocin administra-
(uterine) balloon was filled with 50 mL of cin at 5 mIU/min increasing incremen- tion, epidural request, and neonatal
saline above the level of the internal os and tally by 5 mIU/min every 15 minutes to outcomes (admission to the neonatal in-

125.e2 American Journal of Obstetrics & Gynecology AUGUST 2012

www.AJOG.org Obstetrics Research

tensive care unit, Apgar score ⱕ7 at 5 min-

FIGURE 1
utes, umbilical artery pH ⬍7.00).
Diagram of flow of participants through trial
Planned sample size for this investiga-
tion was based on detecting a clinically Assessed for eligibility (n=508)
significant increase in vaginal delivery
achieved within 24 hours when a balloon Excluded (n=298)
catheter is used for preinduction cervical Not meeting inclusion criteria
(n=137)
ripening. Previously published institu- favorable cervix (n=117)
tional data from inductions with a PGE2 known colonization with GBS (n=27)
vaginal insert indicated a vaginal delivery (n=5)
(n=12)
rate in 24 hours of 48.5%.6 We assumed Declined to participate (n=0)
that a double-balloon device, left in place
for a maximum of 12 hours, would allow Randomized (n=210)
achieving a 20% increase in the rate of
vaginal delivery within 24 hours. With
alpha ⫽ 0.05 and beta ⫽ 0.20, a sample
size of 102 women per group would be Allocated to double-balloon catheter (n=105) Allocated to prostaglandin E2 vaginal insert
Received allocated intervention (n=103) (n=105)
required to detect an increase from 48.5- Received allocated intervention (n=105)
Did not receive allocated intervention
68.5% in the proportion of women who (device insertion failure, n=2) Did not receive allocated intervention (n=0)
deliver vaginally within 24 hours. Sam-
ple size was calculated using a 2-tailed
test in G*Power 3 software (Institut für p
Experimentelle Psychologie, Dusseldorf, Discontinued intervention (n=0) Discontinued intervention (

Germany).7 To account for a protocol prostaglandin E2

, n=2)
violation rate of 3%, 105 patients were
enrolled in each arm of the study.
Analy
Statistical analysis of outcomes data was
Analysed (n=105) Analysed (n=103)
performed with software (GraphPad, Ver- Excluded from analysis (protocol deviation,
Excluded from analysis (n=0)
sion 5; GraphPad Software, San Diego, n=2)
CA). Normality testing (D’Agostino and
Pearson test) was performed to determine FHR, fetal heart rate; GBS, group b strep.
whether data were sampled from a gauss- Cromi. PGE2 vaginal insert vs double-balloon catheter. Am J Obstet Gynecol 2012.

ian distribution. The Student t test and the

Mann-Whitney U test were used to com- the sustained-release insert fell out acci- sert was removed due to uterine hyper-
pare groups of continuous normally and dentally during the ripening process. tonus without fetal heart rate changes in
not-normally distributed variables, re- Both research subjects were withdrawn 4 (3.9%) patients, while tachysystole
spectively. The ␹2 test was used to analyze
as a result of this protocol deviation. The with associated fetal heart rate decelera-
proportions. Analysis of the proportions of
double-balloon catheter was not suc- tions occurred in 6 (5.7%) cases, 2 of
women who remained undelivered over cessfully placed in 2 women randomized which required an emergency cesarean
time was performed by plotting Kaplan- to mechanical ripening, who had a delivery due to persistent abnormal trac-
Meier survival curves. Delivery by emer- closed cervix that did not admit the rip- ing after insert withdrawal. No case of
gency cesarean section was taken as cen- ening device. In all cases of catheter in- uterine hypertonus/tachysystole occurred
soring. A P value ⬍ .05 was used as the sertion failures, patients received a PGE2 in the transcervical catheter arm (9.7% vs
cut-point for significance. vaginal insert for cervical ripening. None 0%; P ⫽ .002; odds ratio [OR], 22.76; 95%
of these women were excluded and an confidence interval [CI], 1.31–394.0).
R ESULTS intent-to-treat analysis was performed. The proportion of women who achieved
Over the study period, 2224 women The demographics and baseline char- vaginal delivery within 24 hours was signif-
delivered at our institution and 508 acteristics were similar across both treat- icantly higher in the double-balloon cath-
(22.8%) deliveries were a result of labor ment groups (Table 1). Table 2 shows the eter group than in the PGE2 vaginal insert
induction. In all, 210 women met the in- details of the ripening process and labor group (OR, 2.22; 95% CI, 1.26 –3.91).
clusion criteria and were entered into the induction outcomes. Accidental expul- Thirty-five (33.3%) patients in the double-
study. Flow of participants through the sion of the vaginal insert occurred in 2 balloon group, and 56 (54.4%) in the
randomized clinical trial is displayed in women and a new device was reinserted PGE2 vaginal insert arm went into labor
Figure 1. In the PGE2 insert group, 2 as soon as the loss of the ripening agent during the ripening process, without any
women received PGE2 vaginal gel after has been noticed. The vaginal PGE2 in- additional intervention (P ⫽ .003). Oxyto-

AUGUST 2012 American Journal of Obstetrics & Gynecology 125.e3

Research Obstetrics www.AJOG.org

3-arm, randomized study in which no

TABLE 1 outcome measure was prespecified and
Demographic and baseline characteristics statistical power was inadequate to de-
Double-balloon PGE2 vaginal tect any difference in treatment effective-
Characteristic catheter, n ⴝ 105 insert, n ⴝ 103 P value ness.8 For the current trial we chose as a
Maternal age, y 34 (19–42) 33 (20–45) .18 measure of effectiveness a time-based
..............................................................................................................................................................................................................................................
Admission BMI, kg/m 2
28.1 ⫾ 4.4 29.2 ⫾ 5.4 .14 definition of a successful induction such
.....................................................................................................................................................................................................................................
as the rate of vaginal delivery achieved in
Parity 0 (0–5) 0 (0–5) 1.0
............................................................................................................................................................................................................................ 24 hours. Since the purpose of an induc-
Nulliparous 82 (78.1%) 75 (72.8%) .42 tion method is to cause a nonlaboring
............................................................................................................................................................................................................................
Parous 23 (21.9%) 28 (27.2%) woman to enter labor, a reasonable
..............................................................................................................................................................................................................................................
Gestational age on admission, wk 40.4 (34–41.9) 40.6 (34–42.4) .78 working definition could be achieving
..............................................................................................................................................................................................................................................
active labor as a measure of success.
Gestational age ⬍37 wk 8 (7.6%) 9 (8.7%) .80
.............................................................................................................................................................................................................................................. However, as the downstream effect of in-
Baseline Bishop score 2 (0–5) 2 (0–5) .08 duction (ie, the potential effect on the
..............................................................................................................................................................................................................................................
Reason for induction mode of delivery) is equally important,
.....................................................................................................................................................................................................................................
Postterm pregnancy (ⱖ41 5/7 wk) 45 (42.8%) 46 (44.7%) .89 many investigators use cesarean delivery
.....................................................................................................................................................................................................................................
rate as a metric.
Hypertensive disorders 20 (19.0%) 17 (16.5%) .72
..................................................................................................................................................................................................................................... Since nonclinical factors seem to play
Oligohydramnios 7 (6.7%) 7 (6.8%) 1.0 an increasingly important role in the
.....................................................................................................................................................................................................................................
Intrauterine growth restriction 8 (7.6%) 5 (4.8%) .57 clinical decision-making process that
.....................................................................................................................................................................................................................................
Polyhydramnios 8 (7.6%) 7 (6.8%) 1.0 leads to a surgical birth, assessing how an
.....................................................................................................................................................................................................................................
induction method will truly affect cesar-
Intrahepatic cholestasis 4 (3.8%) 2 (1.9%) .68
..................................................................................................................................................................................................................................... ean delivery rates is a challenging task. In
Suspected fetal macrosomia 3 (2.9%) 6 (5.8%) .33 a randomized trial comparing single- vs
.....................................................................................................................................................................................................................................
Maternal medical disorders 5 (4.8%) 4 (3.9%) 1.0 double-balloon catheter vs vaginal PGE2
.....................................................................................................................................................................................................................................
Others 5 (4.8%) 9 (8.7%) .28 gel for cervical ripening in nulliparous
.............................................................................................................................................................................................................................................. women, the researchers enrolled just
Data are presented as number (%), median (range), or mean ⫾ SD.
BMI, body mass index; PG, prostaglandin. enough women to be able to detect an
Cromi. PGE2 vaginal insert vs double-balloon catheter. Am J Obstet Gynecol 2012. improbable difference between groups
of ⱖ50% in the rate of cesarean deliv-
ery.9 The assumption of such a dramatic
cin for induction/augmentation of labor mission to the neonatal intensive care unit) reduction in cesarean section seems un-
(OR, 5.04; 95% CI, 2.58 –9.84) and epidu- were similar in both study groups (Table realistic within a context where abdom-
ral analgesia (OR, 3.07; 95% CI, 1.61–5.84) 3). No newborn was admitted to the neo- inal delivery rates ranged from 36-43%
were administered more frequently natal intensive care unit due to suspected across study arms. It would be wholly
when a double-balloon device was sepsis or had a culture-proven sepsis. implausible to expect that a single inter-
used than when a sustained-release di- vention (a new ripening agent) could
noprostone vaginal insert was admin- C OMMENT impact delivery mode in such a huge
istered for cervical ripening. This study was undertaken to compare way.
Postpartum hemorrhage, defined as the efficacy of a double-balloon catheter We believe that the time taken for cer-
blood loss ⬎1000 mL, occurred in 8 vs a sustained-release vaginal PGE2 in- vical ripening is an important consider-
(7.6%) and 7 (6.8%) deliveries in the dou- sert among women with an unfavorable ation when a method for preinduction is
ble-balloon and PGE2 insert groups, re- cervix undergoing labor induction. Our chosen, and authors of labor induction
spectively (P ⫽ 1.0). The median (range) findings indicate that the risk of not reviews as well as members of guidelines
number of maternal hospitalization days achieving vaginal birth within 24 hours development groups identified vaginal
subsequent to delivery was similar between was reduced in the group of women in- delivery within 24 hours as being most
groups (3 [2– 6] vs 3 [2–7], P ⫽ .73). duced with a double-balloon catheter representative of the clinically relevant
Kaplan-Meier survival curves illus- compared with those who received a di- measures of effectiveness for trials of
trating the fraction of women who gave noprostone device. The proportion of methods of labor induction.2,4,10 Safety
birth vaginally at a given time after initi- women achieving active labor, mode of and efficacy being equal, patient satisfac-
ation of cervical ripening in the study delivery, and length of time to achieve tion continues to be the major objective
groups are shown in Figure 2. each of these outcomes were comparable with economic evaluations now becom-
Neonatal outcomes (including birth- between the groups. ing a significant factor in the search for
weight, rate of macrosomic fetuses, 5-min- These ripening methods have been the ideal ripening method. Qualitative
ute Apgar scores, fetal acidemia, rate of ad- previously compared in a single small, studies of how women perceived their

125.e4 American Journal of Obstetrics & Gynecology AUGUST 2012

www.AJOG.org Obstetrics Research

TABLE 2 FIGURE 2
Induction and intrapartum outcomes Survival curves analysis for
interval to delivery
Double-balloon PGE2 vaginal
Variable catheter, n ⴝ 105 insert, n ⴝ 103 P value
Improvement in Bishop score 3 (0–9) 4 (0–10) .09
..............................................................................................................................................................................................................................................
Onset of active labor 96 (91.4%) 93 (90.3%) .81
..............................................................................................................................................................................................................................................
Time to onset of active labor, h 15.6 ⫾ 4.5 16.6 ⫾ 8.8 .71
..............................................................................................................................................................................................................................................
Mode of delivery
.....................................................................................................................................................................................................................................
Vaginal–overall 80 (76.2%) 76 (73.8%) .75
.....................................................................................................................................................................................................................................
Vacuum 6 (5.7%) 1 (1.0%) .12
.....................................................................................................................................................................................................................................
Cesarean 25 (23.8%) 27 (26.2%) .75
..............................................................................................................................................................................................................................................
Time to delivery, h 19.7 ⫾ 5.9 20.4 ⫾ 10.3 .83
..............................................................................................................................................................................................................................................
Time to vaginal delivery, h 18.8 ⫾ 5.4 19.9 ⫾ 9.6 .86
..............................................................................................................................................................................................................................................
Vaginal delivery within 24 h 72 (68.6%) 51 (49.5%) .007 Kaplan-Meier survival curves illustrating fraction
..............................................................................................................................................................................................................................................
of women who gave birth vaginally at given time
Oxytocin administration a
90 (85.7%) 56 (54.4%) ⬍ .0001
.............................................................................................................................................................................................................................................. after initiation of cervical ripening. Black line ⫽
Indication for cesarean section
..................................................................................................................................................................................................................................... prostaglandin E2 vaginal insert; red line ⫽ dou-
Failed induction 6 (5.7%) 4 (3.9%) .75 ble-balloon device.
.....................................................................................................................................................................................................................................
Cromi. PGE2 vaginal insert vs double-balloon catheter.
Failure to progress 8 (7.6%) 3 (2.9%) .21 Am J Obstet Gynecol 2012.
.....................................................................................................................................................................................................................................
Nonreassuring FHR tracing
............................................................................................................................................................................................................................
During ripening 3 (2.9%) 6 (5.8%) .33 ing little, if any, uterine activity, and
............................................................................................................................................................................................................................
During induction/active labor 8 (7.6%) 10 (9.7%) .63 most deliveries occur only after a formal
.....................................................................................................................................................................................................................................
Others 0 (%) 4 (3.9%) .06 process of induction is embarked upon.
..............................................................................................................................................................................................................................................
Thus, we refute the claim that a regimen
Epidural rate 87 (82.9%) 63 (61.2%) .0006
.............................................................................................................................................................................................................................................. that allows more time for cervical ripen-
Data are presented as number (%), median (range), or mean ⫾ SD. ing to take place unavoidably translates
FHR, fetal heart rate; PG, prostaglandin.
a
Use of oxytocin both for labor induction and augmentation.
into a lower proportion of women who
Cromi. PGE2 vaginal insert vs double-balloon catheter. Am J Obstet Gynecol 2012. are delivered vaginally within 24 hours.
Likewise, the increased rate of oxytocin
use in the catheter group reflects the dif-
birth experience in the setting of induc- cise patience, we deem that in the social ferent nature of mechanical and pharmaco-
tion of labor indicated that a long time context of many contemporary Western logic ripening agents, rather than a misuse of
interval required to achieve delivery was societies, acceptance of an induction oxytocin augmentation that ultimately af-
a significant determinant of dissatisfac- method tightly depends on its ability to fected the primary outcome.
tion with the birth process.11 When act within a reasonable time frame. Moreover, defining the success of an
pregnant women are interviewed as to It can be argued that the difference be- induction of labor by the amount of time
their expectations regarding childbirth, tween the study arms in the maximal it takes to achieve vaginal delivery is use-
they identify short duration and man- time allowed for ripening to take place ful when approximating the economic
ageable pain as the main hopes for their before formal induction could have impact of labor induction. Cost-effec-
labor.12 Opponents of the use of time- skewed results. PGs, unlike mechanical tiveness analysis of ripening methods
based definitions of a successful induc- methods, blur the line between cervical should assess whether additional costs
tion could argue that in settings where ripening and induction of labor and of- from higher need of oxytocin and epidu-
time thresholds are used, this could lead ten produce a significant uterine con- ral in the double-balloon group are
to an increase in unnecessary cesarean tractility, such that labor ensues during counterbalanced by cost savings from
deliveries simply because of a lack of pa- ripening and oxytocin is often not re- faster induction processes. Moreover, we
tience by both the providers and patients quired. Several studies showed that uter- have to acknowledge that the use of a
and that, in cases where the need for de- ine contractions become apparent since Foley catheter as a mechanical cervical
livery is not urgent, cervical ripening the first hour of administration and per- dilator would have resulted in significant
may take place over several days, with pe- sist with an adequate pattern in approx- cost savings when compared with a dou-
riods of rest. Although maternity health imately half of all cases.13,14 On the con- ble-balloon device. The goal of balloon
care providers need to help women de- trary, transcervical balloon catheters catheters is to ripen the cervix causing
velop realistic expectations and to exer- promote cervical changes while provok- release of local PGs by separation of the

AUGUST 2012 American Journal of Obstetrics & Gynecology 125.e5

Research Obstetrics www.AJOG.org

methods of preinduction cervical ripening

TABLE 3 over pharmacologic agents.2,3,15 The evi-
Neonatal outcomes dence is extremely relevant to current clinical
Double-balloon PGE2 vaginal practice, since PGs are the main induction
Characteristic catheter, n ⴝ 105 insert, n ⴝ 103 P value method recommended by authoritative
Birthweight, g 3268 ⫾ 582 3314 ⫾ 572 .56 guidelines on induction of labor.4 To im-
..............................................................................................................................................................................................................................................
Macrosomia (⬎4000 g) 11 (10.5%) 8 (7.8%) .63 provethequalityofcareaffordedtopregnant
..............................................................................................................................................................................................................................................
women undergoing induction of labor, fu-
Umbilical artery blood pH ⬍7.00 1 (0.9%) 0 (0%) 1.0
.............................................................................................................................................................................................................................................. ture larger trials should assess the risk of
5-min Apgar score ⬍7 1 (0.9%) 0 (0%) 1.0 induction failure against time and plug
..............................................................................................................................................................................................................................................
NICU admission 8 (7.6%) 5 (4.8%) .57 important knowledge gaps such as patient
..............................................................................................................................................................................................................................................
Data are presented as number (%) or mean ⫾ SD. perception and satisfaction with different ap-
NICU, neonatal intensive care unit; PG, prostaglandin. proaches to cervical ripening. Finally, further
Cromi. PGE2 vaginal insert vs double-balloon catheter. Am J Obstet Gynecol 2012. research should clarify the usefulness of tar-
geting a subgroup of patients who would
amnion from the decidua, as well as by ported higher pain scores on insertion of most likely benefit from mechanical meth-
mechanical gradual stretching of the cer- the ripening device in the double-bal- ods, such as those whose indication for in-
vix. Potential advantages of double-bal- loon group compared with the PGE2 ductionimpliesahighersusceptibilitytofetal
loon devices, which have been specifi- vaginal gel, but higher pain levels during distress. f
cally developed for inducing labor, the entire ripening phase in women who
include that the pressure is applied at the receive PGE2. These disparities did not
REFERENCES
level of the cervix from both the external translate into a significant difference be-
1. Kelly AJ, Malik S, Smith L, Kavanagh J,
and internal os and that the balloon in tween groups in the overall satisfaction Thomas J. Vaginal prostaglandin (PGE2 and
the extraamniotic space is held in place with induction of labor. Patient satisfac- PGF2a) for induction of labor at term. Cochrane
by the vaginal balloon when softening tion questionnaires were administered Database Syst Rev 2009;4:CD003101.
and distensibility of the cervix, as a result 24-48 hours after delivery, and this may 2. Boulvain M, Kelly AJ, Lohse C, Stan CM, Irion
of ongoing ripening process, can lead to cast doubts on the extent to which the O. Mechanical methods for induction of labor.
Cochrane Database Syst Rev 2001;4:
early spontaneous expulsions of the de- whole birth experience impacts recall of CD001233.
vice.15 A study of single- vs double-bal- the phenomenon of interest. Compari- 3. Vaknin Z, Kurzweil Y, Sherman D. Foley cath-
loon catheters for labor induction sug- son of patient satisfaction with cervical eter balloon vs locally applied prostaglandins for
gested that both were equally effective, ripening using sustained-release pessar- cervical ripening and labor induction: a system-
but hinted at trend toward more opera- ies and balloon catheters would be inter- atic review and metaanalysis. Am J Obstet Gy-
necol 2010;203:418-29.
tive deliveries with the double-balloon esting since both methods, unlike other
4. National Institute for Health and Clinical Ex-
catheter.16 The investigators suggested vehicles for delivering PG, do not imply cellence (NICE). Induction of labor: clinical
that the different degree of balloon infla- repeated vaginal examinations. Second, guideline no. 70. London: National Institute for
tion (60 mL for the single- vs 80 mL the nature of balloon catheter treatment Clinical Excellence; 2008.
in the double-balloon group) and the means that it would not have been pos- 5. American College of Obstetricians and Gyne-
cologists. ACOG committee on practice bulle-
traction applied on the catheters (pres- sible to conceal treatment allocation,
tins– obstetrics. ACOG practice bulletin no.
ent in the single- and absent in the dou- therefore managing obstetricians could 107: induction of labor. Obstet Gynecol
ble-balloon group) may explain the have inadvertently influenced factors re- 2009;114:386-97.
trend toward increased operative deliv- lated to time to delivery or decision to 6. Cromi A, Ghezzi F, Agosti M, et al. Is tran-
eries with the double-balloon catheter. A perform cesarean section simply in re- scervical Foley catheter actually slower than
prostaglandins in ripening the cervix? A ran-
20-mL difference in balloon inflation sponse to the fact that they knew that
domized study. Am J Obstet Gynecol 2011;
from 60-80 mL translates into a 2.5-mm outcomes would be analyzed. Third, the 204:338.e1-7.
increase in the balloon diameter and study lacks sufficient power to show sig- 7. Faul F, Erdfelder E, Lang A-G, Buchner A.
whether this subtle change might impact nificant treatment differences in second- G*Power 3: a flexible statistical power analysis
on the labor outcome is somewhat ques- ary outcomes and to address safety is- program for the social, behavioral, and biomed-
ical sciences. Behav Res Methods 2007;39:
tionable. Moreover, approximately 1 in 3 sues. A recent metaanalysis of studies
175-91.
participants underwent ripening with a comparing outcomes after induction 8. Yuen PM, Pang HY, Chung T, Chang A. Cer-
balloon catheter as a rescue method after with an intracervical Foley and with vical ripening before induction of labor in pa-
PG failure and this raises doubts as to the PGE2 gel showed that hyperstimulation, tients with an unfavorable cervix: a comparative
generalizability of results. postpartum hemorrhage, and umbilical randomized study of the Atad Ripener Device,
prostaglandin E2 vaginal pessary, and prosta-
The current study has limitations that artery blood pH were in favor of use of a
glandin E2 intracervical gel. Aust N Z J Obstet
are worth mentioning. First, we did not balloon catheter.17 Gynaecol 1996;36:291-5.
address patient satisfaction with the cer- This study adds to previous literature sup- 9. Pennell CE, Henderson JJ, O’Neill MJ, Mc-
vical ripening process. Pennell et al9 re- porting potential advantages of mechanical Chlery S, Doherty DA, Dickinson JE. Induction

125.e6 American Journal of Obstetrics & Gynecology AUGUST 2012

www.AJOG.org Obstetrics Research
of labor in nulliparous women with an 12. Gibbins J, Thomson AM. Women’s expec- double balloon device: experience with 250
unfavorable cervix: a randomized controlled trial tations and experiences of childbirth. Midwifery cases. Br J Obstet Gynaecol 1997;104:29-32.
comparing double and single balloon catheters 2001;17:302-13. 16. Salim R, Zafran N, Nachum Z, Garmi G,
and PGE2 gel. BJOG 2009;116:1443-52. 13. Miller AM, Rayburn WF, Smith CV. Patterns Kraiem N, Shalev E. Single-balloon compared
10. World Health Organization, Department of of uterine activity after intravaginal prostaglan- with double-balloon catheters for induction of
Reproductive Health and Research. WHO rec- din E2 during preinduction cervical ripening. labor: a randomized controlled trial. Obstet Gy-
ommendations for induction of labor. Geneva: Am J Obstet Gynecol 1991;165:1006-10. necol 2011;118:79-86.
World Health Organization; 2011. 14. Wikland M, Lindblom B, Wiquist N. Myome- 17. Jozwiak M, Oude Rengerink K, Benthem M,
11. Shetty A, Burt R, Rice P, Templeton A. trial response to prostaglandins during labor. et al; PROBAAT Study Group. Foley catheter
Women’s perceptions, expectations and satis- Gynecol Obstet Invest 1984;17:131-6. versus vaginal prostaglandin E2 gel for induc-
faction with induced labor–a questionnaire- 15. Atad J, Hallak M, Ben-David Y, Auslender tion of labor at term (PROBAAT trial): an open-
based study. Eur J Obstet Gynecol Reprod Biol R, Abramovici H. Ripening and dilatation of the label, randomized controlled trial. Lancet 2012;
2005;123:56-61. unfavorable cervix for induction of labor by a 378:2095-103.

AUGUST 2012 American Journal of Obstetrics & Gynecology 125.e7