Marconi 2008
Marconi 2008
Marconi 2008
Abstract
Objective: To compare dinoprostone gel and insert in achieving successful vaginal delivery in nulliparous and multiparous women.
Study design: 220 nulliparous and 100 multiparous with a Bishop score 7 were randomized to receive dinoprostone either gel or insert for
cervical ripening. The main outcome measures were the rate and latency of vaginal delivery.
Results: In nulliparous women no significant differences were found between the gel and insert groups in the rate of vaginal delivery (85.6%
vs. 80.7%) delivery 12 (36.8% vs. 32.9%) and 24 h (85.3% vs. 93.4%) regardless of the preinduction Bishop score. Nulliparous with
Bishop score 4 treated with the insert had a decreased risk ( p < 0.05) of post partum hemorrhage (4.8%) when compared with nulliparous
treated with gel (16.7%). On the contrary, in multiparous the time to delivery interval was significantly shorter in the gel treated group
(9.9 4.9 h vs. 13.1 5 h; p < 0.001) with more patients delivering vaginally 12 h (75% vs. 37.5%, p < 0.001), regardless of the
preinduction Bishop score.
Conclusion: Both dinoprostone gel and insert are efficient in achieving cervical ripening and successful labor in nulliparous and multiparous.
In multiparous, however, the gel significantly reduces the time to vaginal delivery with more patients delivering vaginally 12 h, regardless of
the Bishop score.
# 2007 Elsevier Ireland Ltd. All rights reserved.
0301-2115/$ see front matter # 2007 Elsevier Ireland Ltd. All rights reserved.
doi:10.1016/j.ejogrb.2007.08.009
136 A.M. Marconi et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 138 (2008) 135140
Table 1
Maternal age, gestational age and Bishop score at the time of induction
(mean S.D.) and clinical indication for induction
Gel Insert
Number of patients 161 159
Nulliparous 111 (68.9%) 109 (66.7%)
Multiparous 50 (31.1%) 50 (33.3%)
Maternal age (years) 31 4.7 30.3 5.1
Nulliparous 30.05 4.3 29.4 4.7
Multiparous 33.1 4.9 32.1 5.6
Gestational age (weeks) 39.6 1.3 39.6 1.4
Nulliparous 39.6 1.4 39.6 1.4
Multiparous 39.7 1.3 39.8 1.3
Fig. 1. KaplanMeier plot of time to vaginal delivery stratified by parity
Bishop score at induction 4.1 1.2 4.0 1.1 and treatment group in patients with Bishop score 4 (solid line: M with
Nulliparous 3.9 1.1 3.9 1.2 gel; hatched line: M with insert; solidhatched line: N with gel; dotted line:
Multiparous 4.5 1.1 4.1 1.1 N with insert).
Parity
Nulliparous at first pregnancya 87 (78.4%) 85 (78%) (4.9%) nulliparous had vacuum assisted vaginal delivery
Multiparous with >1 delivery 14 (28%) 9 (18%) regardless of preinduction Bishop score or treatment. The %
Indication for induction
of cesarean section was higher, although not significant, in N
Oligohydramnios 41 35 and M with unfavourable cervix with no treatment based
Premature rupture of membranes 33 22 differences. Overall, only three cesarean sections were
Postdate pregnancy 22 30 performed for failed induction, all in nulliparous with BS
Gestational hypertension 21 24 4 (one in the gel and two in the insert groups).
Intrauterine growth restriction 17 17
Gestational diabetes 9 8
27 patients (25 of which nulliparous) had a 0 change of
Non-reassuring CTG 6 6 BS at the end of the preinduction period (22 in the insert and
Other maternal medical conditions 12 16 5 in the gel group): even though 16/27 remained with a BS
a
Absence of spontaneous or voluntary pregnancy termination.
4, amniotomy/oxytocin was performed and 21/27 deliv-
ered vaginally (15 within 24 h). Only six cesarean sections
were performed in these patients, two for failed induction
significantly higher in the gel than in the insert groups (both in nulliparous).
(1.9 1.04 vs. 1.3 0.5; p < 0.001). At vaginal delivery, we found no differences in the mean
Two patients in the gel group and two patients in the blood loss (273 230 ml in the gel group and 253 232 in
insert group had complications. In the insert group, uterine the insert group), however more patients with an unfavour-
tachysystole with abnormal fetal heart rate tracings able cervix in the gel treated group had a post partum
developed within 2 h from induction, the ripening agent hemorrhage (PPH) and this was mainly due to nulliparous.
was removed and, after the fetal heart rate had recovered, The median hospital stay was not different in the gel and
amniotomy was performed and both patients delivered insert groups (4 days) regardless of parity: 26/30 patients in
vaginally after approximately 3 and 5 h, respectively. In the the gel group and 26/35 in the insert group with a hospital
gel group, one multiparous patient had uterine tachysystole stay >4 days had an unripe cervix (BS 4) (Table 3).
with abnormal fetal heart rate: vaginal washing was
performed, fetal heart rate recovered and she delivered
vaginally approximately after 4 h. In the other patient, fetal
heart rate abnormalities appeared with no evidence of
uterine tachysystole: she underwent a cesarean section and
the umbilical cord was found having two true knots.
No maternal side effects (nausea, vomiting, fever and
diarrhea) were reported in any patient.
Time to vaginal delivery is presented as a KaplanMeier
plot stratified by parity and treatment in Fig. 1 (patients with
Bishop score 4) and Fig. 2 (Bishop score 57). When Coxs
regression analysis was performed we found that the time to
vaginal delivery interval was independently correlated to
treatment ( p < 0.02), parity ( p < 0.001) and preinduction
Fig. 2. KaplanMeier plot of time to vaginal delivery stratified by parity
Bishop score ( p < 0.001) (Table 2). Multiparous with BS 57 and treatment group in patients with Bishop score 57 (solid line: M with
treated with gel had the shortest time to delivery interval gel; hatched line: M with insert; solidhatched line: N with gel; dotted line:
(Table 3); no differences were present in nulliparous. 9/183 N with insert).
138 A.M. Marconi et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 138 (2008) 135140
Table 2
Cox regression analysis
Hazard ratio Standard error z p > [z] [95% Confidence interval]
Treatment 0.7706115 0.0883825 2.27 0.023 0.6154744 0.9648527
Parity 2.002239 0.2592057 5.36 0.000 1.553535 2.580541
Bishop score 2.034914 0.2627192 5.50 0.000 1.579978 2.620844
Neonatal outcome is presented in Table 4: there were no with an unripe cervix went into labour after the first
differences in the mean birthweight and in the percentage of intracervical gel.
babies 2500 grams between the gel and insert treated Furthermore, once in progress, the preinduction is
patients. 19 babies were admitted in the NICU, only 5 of brought until delivery has been accomplished; amniotomy
which for delivery related problems. and/or oxytocin is started either when the BS is >7 or when
regular uterine contractions are not present at the end of the
ripening process, independent of the BS. Other authors
4. Comment chose to repeat the preinduction after 24 h if labor did not
start or the BS did not improve [5,912] even though the
This is the largest prospective randomized study ripening process should determine changes in histologic
comparing two dinoprostone agents, gel and insert, in characteristics and biochemical composition of the cervix
achieving cervical ripening and successful delivery in which not necessarily imply the onset of uterine contractions
nulliparous and multiparous patients analyzed separately. [16]. These authors report repetition of induction rates
We studied 320 patients, mostly (68.7%) nulliparous. varying from 3% (10) to 53% (9) with the insert and from
Our protocol does not allow more than one insert 16% (10) to 34% (9) with the gel without any evident benefit.
application (unless accidentally removed before the start In our study, >80% of the patients delivered vaginally
of uterine contractions with a BS < 7) and more than three regardless of parity, treatment and preinduction BS. Other
administration of gel: intracervical vs. intravaginal route is authors have reported similar high rate both for the insert
dependent on the BS, based upon the observation that [2,5,7,8] and the gel [5,7,8,17]. However, when compared to
intracervical administration is particularly effective when other studies, we report higher rates of vaginal delivery
the cervix is unfavourable (BS 4) [15]. As a matter of within 12 h regardless of parity [10,17], treatment
fact, 34% of the nulliparous and 44% of the multiparous [6,10,11,17] or preinduction BS [11,12]: this was mainly
Table 3
Data at delivery and hospital stay according to parity and to the Bishop score at the time of induction in the gel and insert groups
BS 4 p BS 57 p
Gel Insert Gel Insert
Number of patients (%) 111 (68.9%) 109 (68.6%) NS 50 (31.1%) 50 (31.4%) NS
Nulliparous 87 (78.4%) 80 (73.4%) 24 (48%) 29 (58%)
Multiparous 24 (21.6%) 29 (26.6%) 26 (52%) 21 (42%)
Vaginal delivery (%) 96 (86.5%) 89 (81.6%) NS 48 (96%) 47 (94%) NS
Nulliparous (% of N) 72 (82.8%) 62 (77.5%) 23 (95.8%) 26 (89.7%)
Multiparous (% of M) 24 (100%) 27 (93.1%) 25 (96.1%) 21 (100%)
Hours from induction to VD 15.3 7.7 15.6 7.1 NS 9.6 4.4 12.3 4.9 0.006
Nulliparous 16.5 8 16.1 7.7 11.5 4.8 12.9 5.6
Multiparous 11.9 5.7 14.5 5.5 7.8 3.1 11.5 3.8 0.001
Cesarean section 15 (13.5%) 20 (18.3%) NS 2 (4%) 3 (6%) NS
Nulliparous (% of N) 15 (17.2%) 18 (22.5%) 1 (4.2%) 3 (10.3%)
Multiparous (% of M) 0 2 (6.9%) 1 (3.8%) 0
Amniotomy/Oxytocin for induction 45 (40.5%) 43 (39.4%) NS 7 (14%) 10 (20%) NS
Nulliparous (% of N) 36 (41.4%) 35 (43.8%) 5 (20.8%) 6 (20.7%)
Multiparous (% of M) 9 (37.5%) 8 (27.6%) 2 (7.7%) 4 (19%)
Post partum hemorrhage (500 ml)a 14 (14.6%) 4 (4.5%) 0.03 6 (12.5%) 7 (15.2%) NS
Nulliparous (% of N) 12 (16.7%) 3 (4.8%) 0.05 3 (13%) 6 (23.1%)
Multiparous (% of M) 2 (8.3%) 1 (3.7%) 3 (12%) 1 (4.8%)
Hospital stay >4 days 26 (23.4%) 27 (24.8%) NS 3 (6%) 9 (18%) NS
Nulliparous 25 (28.7%) 23 (28.7%) 3 (12.5%) 7 (24.1%)
Multiparous 1 (4.2%) 4 (13.8%) 0 2 (9.5%)
a
At vaginal delivery.
A.M. Marconi et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 138 (2008) 135140 139
Table 4
Birthweight, and neonatal outcome in the study groups
Gel Insert p
Birthweight (grams) [range] 3238 475 [17804300] 3286 509 [19304630] NS
Nulliparous 3204 463 3259 489
Multiparous 3316 495 3346 551
Birthweight 2500 g 13 (8.1%) 9 (5.7%) NS
Cesarean section for fetal heart rate abnormalities 9 (5.6%) 6 (3.8%) NS
0
Apgar 7 at 1 7 (4.3%) 6 (3.8%) NS
Apgar 7 at 50 1 (0.6%) 1 (0.6%)
Umbilical arterial pH < 7.00 1a (0.6%) 1a (0.6%) NS
Meconium stained amniotic fluid 3 (1.9%) NS
Transfer to neonatal intensive case unit 8 (5%) 11 (6.9%) NS
For delivery related problems 3 (1.9%) 2 (1.2%)
a
Delivered by vacuum.
due to multiparous treated with gel, independently of the differences in operative vaginal delivery, rate of oxytocin
Bishop score. Similarly, almost 90% of the patients who augmentation or other post partum complications.
delivered vaginally, delivered within 24 h regardless of We are aware that misoprostol is much more cost-
parity and this is the highest percentage reported thus far in effective (same efficacy with very low costs) however its use
similar studies. Also, the time to delivery interval is shorter is still associated with the observation of significant side
in our study when compared to others both in the gel [2,5] effects [2022]. Furthermore, compared to the results of our
and insert [2,18] groups. Both observations might be study it does not seem to offer big advantages in terms of
partially explained by the fact that others allowed repetition cesarean section rate and time to delivery intervals [19,23]
of the preinduction [5,11,12] which most likely has even though only a randomized trial could confirm this
lengthened the whole process. However, it is worth noting observation.
that the rate of vaginal delivery within 24 h is higher also In conclusion, our study shows that both dinoprostone gel
than that reported by Stewart et al. [17] in a gel group treated and insert are efficient in achieving cervical ripening and
with immediate oxytocin (76% in nulliparous and 89% in successful vaginal delivery both in nulliparous and multi-
multiparous) and by Bolnik et al. [19] in an insert group parous with negligible maternal and fetal/neonatal side
treated with immediate oxytocin (81%). effects. In multiparous, however, the gel significantly and
Conversely, we had a low cesarean section rate, correctly powered reduces the time to vaginal delivery with
particularly in multiparous (2% in the gel group and more patients delivering vaginally within 12 h, regardless of
4% in the insert group), regardless of the BS. Even though the Bishop score. The insert seems more convenient in
this study is underpowered to detect the cesarean section nulliparous patients, especially when the preinduction BS is
rate differences, we found (to achieve a significant result 4, given its handiness: one administration, possible
we should have enrolled 1555 patients in each arm) the application overnight, easy removal in case of side effects,
cesarean section rate in these patients is comparable with less vaginal examinations and, most important, decreased
rates obtained in low risk patients in spontaneous labour at risk of post partum hemorrhage.
our institution (13.7%). A possible explanation for this is
the large use of amniotomy/oxytocin: on the whole, 63%
of our patients received either further induction or Acknowledgments
augmentation and this might have increased the number
of successful vaginal delivery on one side, and reduced the The authors wish to thank Camilla De Gasperi, MD,
number of cesarean section for failed induction on the Maria Nobile De Santis, MD and Simona Vailati, MD for
other side. their precious help in collecting the data forms.
Nulliparous with an unfavorable cervix treated with gel
had an increased rate of post partum blood loss 500 ml
when compared to insert treated women. Even though References
induction of labor with PGE2 has been reported as a risk
factor for PPH when compared to placebo [1], to our [1] Kelly AJ, Kavanagh J, Thomas J. Vaginal prostaglandin (PGE2 and
knowledge this is the first time a difference among PGF2a) for induction of labour at term. Cochrane Database Syst Rev
2003 (4). Art. No.: CD003101. doi:10.1002/14651858.CD003101.
treatments is reported. We do not have a clear explanation [2] Chyu JK, Strassner HT. Prostaglandin E2 for cervical ripening: a
for this finding: in both groups patients received the same randomized comparison of Cervidil versus Prepidil. Am J Obstet
prophylaxis (oxytocin 10 iu im) and there were no Gynecol 1997;177:60611.
140 A.M. Marconi et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 138 (2008) 135140
[3] Ottinger WS, Menard MK, Brost BC. A randomized clinical trial of [14] Freeman RK. Problems with intrapartum fetal heart rate monitoring
prostaglandin E2 intracervical gel and a slow release vaginal pessary interpretation and patient management. Obstet Gynecol
for preinduction cervical ripening. Am J Obstet Gynecol 2002;100:81326.
1998;179:34953. [15] Zanini A, Ghidini A, Norchi S, Beretta E, Cortinovis I, Bottino S. Pre-
[4] Wieland D, Friedman Jr F. Comparing two dinoprostone agents for induction cervical ripening with prostaglandin E2 gel: intracervical
preinduction cervical ripening at term. J Reprod Med 1999;44:7248. versus intravaginal route. Obstet Gynecol 1990;76:6813.
[5] Vollebregt A, vant Hof DB, Exalto N. Prepidil compared to Propess [16] Arias F. Pharmacology of oxytocin and prostaglandins. Clin Obstet
for cervical ripening. Eur J Obstet Gynecol Reprod Biol Gynecol 2000;43:45568.
2002;104:1169. [17] Stewart JD, Rayburn WF, Farmer KC, Liles EM, Schipul AH, Stanley
[6] Miller AM, Rayburn WF, Smith CV. Patterns of uterine activity after JR. Effectiveness of prostaglandin E2 intracervical gel (Prepidil), with
intravaginal prostaglandin E2 during preinduction cervical ripening. immediate oxytocin, versus vaginal insert (Cervidil) for induction of
Am J Obstet Gynecol 1991;165:10069. labor. Am J Obstet Gynecol 1998;179:117580.
[7] Green C, Pedder G, Mason G. A randomized trial of Propess against [18] Hennessey MH, Rayburn WF, Stewart JD, Liles EC. Pre-eclampsia
prostin gel for the induction of labour at term. Br J Obstet Gynaecol and induction of labor: a randomized comparison of prostaglandin E2
1998;105(Suppl. 17):82. as an intracervical gel, with oxytocin immediately, or a sustained-
[8] Rabl M, Joura EA, Yucel Y, Egarter C. A randomized trial of vaginal release vaginal insert. Am J Obstet Gynecol 1998;179:12049.
prostaglandin E2 for induction of labor. Insert vs tablet. J Reprod Med [19] Bolnick JM, Velazquez MD, Gonzalez JL, Rappaport VJ, McIlwain-
2002;47:1159. Dunivan G, Rayburn WF. Randomized trial between two active labor
[9] Mukhopadhyay M, Lim KJH, Fairlie FM. Is propess a better method of management protocols in the presence of an unfavorable cervix. Am J
induction of labour in nulliparous women? J Obstet Gynaecol Obstet Gynecol 2004;190:1248.
2002;22:2945. [20] Papanikolaou EG, Plachouras N, Drougia A, et al. Comparison of
[10] DAniello G, Bocchi C, Florio P, et al. Cervical ripening and induction misoprostol and dinoprostone for elective induction of labour in
of labor by prostaglandin E2: comparison between intracervical gel nulliparous women at full term: a randomized prospective study.
and vaginal pessary. J Matern Fetal Neonatal Med 2003;14:15862. Reprod Biol Endocrinol 2004;2:707.
[11] Facchinetti F, Venturini P, Verocchi G, Volpe A. Comparison of two [21] Ramsey PS, Meyer L, Walkes BA, et al. Cardiotocographic abnorm-
preparations of dinoprostone for pre-induction of labour in nulliparous alities associated with dinoprostone and misoprostol cervical ripening.
women with very unfavourable cervical condition: a randomized Obstet Gynecol 2005;105:8590.
clinical trial. Eur J Obstet Gynecol Reprod Biol 2005;119:18993. [22] Megalo A, Petignat P, Hohlfeld P. Influence of misoprostol or pros-
[12] Strobelt N, Meregalli V, Ratti M, Mariani S, Zani G, Morana S. taglandin E(2) for induction of labor on the incidence of pathological
Randomized study on removable PGE2 vaginal insert versus PGE2 CTG tracing: a randomized trial. Eur J Obstet Gynecol Reprod Biol
cervical gel for cervical priming and labor induction in low-Bishop- 2004;116:348.
score pregnancy. Acta Obstet Gynecol Scand 2006;85:3025. [23] Sanchez-Ramos L, Kaunitz AM. Misoprostol for cervical ripening and
[13] ACOG technical bulletin: induction of labor. Int J Gynaecol Obstet labor induction: a systematic review of the literature. Clin Obstet
1996; 53:6572. Gynecol 2000;43:47588.