200500078.491_499.tp

• Copyright ᮊ by Walter de Gruyter • Berlin • New York. DOI 10.1515/JPM.2005.087 A clinical evaluation of controlled-release dinoprostone for
cervical ripening – a review of current evidence in hospital
and outpatient settings
Werner Rath*
many clinical circumstances, such as pregnancy-inducedhypertension, prelabor rupture of the membranes, sus- Department Obstetrics and Gynecology, University of pected fetal jeopardy or death, or maternal morbidity (e.g. diabetes), lead to the use of labor induction w1x, itsroutine use in, and recommendation for, post-term preg- Abstract
nancy is becoming more frequent (for example, NICE Labor induction is used in approximately 20% of preg- recommendations in the UK w30x). In many countries, nancies in Europe and North America. Prostaglandins induction is routine for pregnancies over 41 weeks, which and prostaglandin analogs are favored when women account for over 20% of all pregnancies w35x. Recom- undergoing labor induction have an unripe cervix. Con- mendations to induce labor in post-term pregnancy fol- trolled-release dinoprostone, delivered over 24 h from a low the finding that the prevalence of stillbirths increases vaginal insert, results in cervical ripening within 12 h in from 0.35 per 1000 ongoing pregnancies at 37 weeks to most women. It is marginally more effective than imme- 2.12 per 1000 at 43 weeks w17x. Furthermore, routine diate release formulations and has similar efficacy to induction of labor in pregnancies of more than 41 weeks misoprostol, a prostaglandin E analog used off-label for reduces perinatal mortality (OR 0.20 w95% CI 0.06–0.70x) this indication. The controlled-release preparation offers many advantages compared with an immediate-release In attempt to optimize labor induction, there has been formulation: a single application is sufficient; it is less an increase in the number of agents available for cervical invasive; it is easily administered and removed, allowing ripening/labor induction. In this article, we review the lit- greater dose control. The most significant adverse effect, erature on one such agent, a sustained and controlled- uterine hyperstimulation, with and without an effect on release dinoprostone vaginal insert (prostaglandin E ; fetal heart rate, occurs in 5–15% of patients, which is Propessᮋ or Cervidilᮋ). We review the efficacy and safety comparable with other formulations or misoprostol. The of the agent as well as recent data on its use in an out- insert can be removed easily on the first sign of uterine patient setting. With outpatient management in obstetrics hyperstimulation, or as soon as labor starts. The efficacy being continuously expanding, we address the current and safety of controlled-release dinoprostone are com- evidence on this important issue, and its potential impli- parable whether it is used in the outpatient or the inpa- cations in terms of cost and patient convenience.
tient setting. For low-risk women, outpatient use may bea highly attractive option, potentially reducing hospital Aims of labor induction
costs, and improving patient convenience. The ease ofuse of controlled-release dinoprostone and women’s sat- Induction of labor is indicated when a continuation of the isfaction emphasize its benefits over many other agents pregnancy would mean that the risks to the mother or fetus outweigh the benefits from further observation. The Keywords: Controlled-release; dinoprostone; outpatient;
aims of induction are straightforward—vaginal delivery or active labor within 12 h. Undesirable outcomes havebeen clearly defined by the Cochrane CollaborationPregnancy and Childbirth Group and include: vaginal Introduction
delivery not achieved within 24 h; uterine hyperstimula-tion with fetal heart rate wFHRx changes; requirement for The frequency of labor induction has increased in Europe cesarean section; serious neonatal morbidity or perinatal and North America over the past 10 years w25x. Although death; and serious maternal morbidity or death w18, 22x.
In the guideline published by the Royal College of Obste- tricians and Gynaecologists, the term hyperstimulation with and without FHR changes included uterine tachy- systole ()5 contractions in 10 min for G20 min) and uterine hypersystole/hypertonus (a contraction lasting G2 min). However, due to varied reporting of this out- 492 Rath, Clinical efficacy of controlled-release dinoprostone The Bishop score for assessment of cervical ripeness w5x.
* Station of fetus relative to the ischial spine come, possible subjective bias in interpretation exists venous infusion of oxytocin, which rapidly stimulates w30x. Where possible, we used these outcomes in uterine contractions but has no direct effect on the con- dition of the cervix w20x. Prostaglandins (dinoprostoneand dinoprost wprostaglandin F x), however, are used for Methods of labor induction
cervical ripening w21, 42x. Dinoprostone is favored overdinoprost for several reasons, including relative potency A pre-requisite for successful labor induction is cervical ripening (or cervical maturation). Many clinicians regardlabor induction and cervical ripening as essentially the Dinoprostone formulations
same process. If labor is induced when the cervix isunripe and unable to yield in response to contractions, The risk of systemic side effects with prostaglandins has high rates of failure are observed w9x.
led to the development of various formulations for local Because of the importance of cervical ripening, meth- application. These include, intra-vaginal tablets or sup- ods of evaluating the status of the cervix have been positories, intra-cervical or intra-vaginal gels in a single- developed. The most widely used is the Bishop score w5x use syringe, and sustained release intra-vaginal inserts.
or the modified Bishop score w8x. The Bishop score rates Tablets or suppositories are the simplest formulation to the dilatation, consistency, and length of the cervix, and administer, but outcome can be unpredictable w27x. In the position of the fetus. A combined score gives a addition, a second dose of the tablets is often adminis- measure of cervical ripeness on a scale from 0 to 13 tered 6–8 h after the first (if the single dose is considered (Table 1). The state of the cervix is not favorable enough ineffective), but since the onset of labor may be 8–12 h to commence labor induction when the Bishop score is after induction, this double dosing increases the risk of below a certain level. No consensus exists on the precise uterine hyperstimulation. Gel formulations of dinopros- score that indicates the need for cervical ripening, but tone provide more predictable results, but administration generally a Bishop score -6 are considered too low to (particularly of intra-cervical gels) requires an experi- enced clinician and can be uncomfortable for the woman.
Approximately, 50% of women undergoing induction of Spillage from the cervical canal can reduce efficacy or, if labor have an unfavorable cervix and, therefore, require the dose passes into the extra-amniotic space, cause cervical ripening w2x. The optimal method of labor induc- uterine hyperstimulation w27x. Both intra-vaginal and tion should ideally also ripen the cervix. Dilatation and intra-cervical gels may also cause uterine hyperstimula- labor can be induced by mechanical means, such as tion by delivering dinoprostone too quickly (i.e. dose- membrane sweeping, artificial rupture (amniotomy), dig- dumping). In the event of such an adverse reaction, gel ital stretching of the cervix, or insertion of catheters, but formulations are quite impossible to remove.
most mechanical methods can traumatize the cervix and A sustained and controlled release of dinoprostone is are associated with discomfort and bleeding in a sub- needed to overcome the potential problems with gel and stantial number of women w49x. In addition, women tablet formulations. A retrievable vaginal insert is availa- undergoing labor induction tend to favor less invasive ble that releases a continuous and predictable dose of pharmacological methods to mechanical methods, such dinoprostone at a rate of approximately 0.3 mg/h for 24 h as artificial rupture of membranes w41x. Pharmacological w28x. The controlled-release formulation of dinoprostone, compounds for inducing labor are therefore preferred— minimizes the risk of uterine hyperstimulation, is easy to the most widely used agents are oxytocin, prostaglan- apply, and can be removed quickly when labor starts, or in the event of an adverse reaction. Furthermore, the pro- The ideal agent must effectively induce labor and con- gressive cervical ripening induced by the controlled vert an unfavorable cervix to one receptive to delivery, release of dinoprostone may be more acceptable to but also needs to be safe, easy to administer, and patients than the rapid onset of contractions observed acceptable to the patient. The most frequently used with some agents w44x. The controlled release of dino- pharmacological method for inducing labor is an intra- prostone from the vaginal insert continues for up to 24 h Rath, Clinical efficacy of controlled-release dinoprostone w28x allowing longer induction, if required, with only one 18.32 w95% CI 9.49–35.38x; P-0.0001), and overall treatment success (OR 4.93 w95% CI 3.36–7.24x;P-0.0001) compared to placebo w11x. Controlled-releasedinoprostone was also associated with a significantly Efficacy of controlled-release dinoprostone
lower rate of oxytocin use (OR 0.14 w95% CI 0.06–0.32x;P-0.0001) and a non-significantly lower rate of cesarean Placebo-controlled trials
delivery. While parity did not affect most of the outcomemeasures, multiparous women had 50% shorter labors Three randomized, double-blind, clinical trials have assessed the efficacy of controlled-release dinoprostone The efficacy of controlled-release dinoprostone was vs. placebo in 485 women at term (referred to as studies reviewed by Kelly et al. w23x. The primary aim of the 1, 2 and 3, respectively) w34, 47, 48x. A subsequent meta- review was to determine the effects of vaginal prosta- analysis including these trials and other studies compar- glandin E for cervical ripening or induction of labor in ing dinoprostone with different cervical ripening agents comparison with placebo, no treatment, or other vaginal were performed by Crane and Bennett w11x.
prostaglandins (except misoprostol). This was reflected The three placebo-controlled trials were similar with by an increase in successful vaginal delivery rates within respect to design. Inclusion criteria were singleton preg- 24 h, no increase in operative delivery rates, and signif- nancy G37 weeks’ duration, cephalic presentation, Bish- icant improvement in cervical condition within 24 to 48 h.
op score F4 at admission, medical or obstetric reasonfor induction, and F3 previous viable deliveries. Exclu- Comparative trials
sion criteria included previous uterine scar, vaginal bleed-ing, ruptured membranes, FHR abnormalities, and A number of trials have examined the efficacy of con- medical conditions precluding dinoprostone administra- trolled-release dinoprostone compared with vaginal or tion. The primary outcome measure for these trials was intra-cervical gel formulations of dinoprostone, and the cervical ripening with treatment success defined as vag- prostaglandin E analog, misoprostol. However, most of inal delivery within 12 h, or a Bishop score G6, or an the comparative trials involve small numbers of patients, increase from baseline in Bishop score G3 at 12 h. A reducing the statistic power. The latter agent is widely number of secondary outcomes were also measured used off-label in both the USA and Europe for cervical including time to vaginal delivery, time to onset of labor, ripening. Misoprostol oral tablets are indicated for the need for oxytocin (reflecting induction failure) and cesa- prevention and treatment of NSAID gastric and duodenal ulcers and were not approved for induction of labor.
The baseline characteristics were similar in each trial Three prospective trials and one retrospective study and did not differ between placebo and active-treatment have assessed the comparative efficacy of controlled- groups, with the exception that in study 3, the mean age release dinoprostone and vaginal or intra-cervical gel of women in the multiparous group receiving placebo dinoprostone w10, 31, 33x. The aggregated data from the was significantly higher than for those receiving active two prospective trials were also analyzed in the Crane treatment (28.2"5.1 vs. 25.6"5.3 years; Ps0.03) w48x.
and Bennett meta-analysis w11x. In one trial, 73 women In all three trials, the most frequent indication for labor who had an indication for induction of labor for a variety of reasons were randomized to receive either intra-cer- Dinoprostone was significantly better than placebo for vical gel (ns36) or a controlled-release (ns37) dinopros- all primary outcomes (Table 2), and in general, the results tone insert w10x. Although changes in Bishop scores were were unchanged when nulliparous women were analyzed similar in the two groups, the mean times to cervical rip- separately. However in study 3, the response to con- ening, active labor, and delivery were significantly shorter trolled-release dinoprostone in nulliparous women was with the use of controlled-release dinoprostone (Figure primarily a cervical change (Bishop score was improved 1). Hospital stay was also shorter with the controlled- but there were no vaginal deliveries in either placebo release formulations than with the gel; 3.7 vs. 4.4 days, recipients or dinoprostone recipients), whereas in multi- respectively (Ps0.03). If active labor had not already parous women the response was mainly higher rates of begun, oxytocin infusion was initiated 6 h after the final delivery within 12 h w48x. Furthermore, in both nulliparous dose of gel or 30 min after removal of the insert. Oxytocin and multiparous women, the time to onset of labor and was administered in 97% of patients receiving dinopros- to vaginal delivery was significantly shorter in those tone gel and in only 76% of patients receiving the con- women receiving controlled-release dinoprostone w34, trolled-release dinoprostone insert (Ps0.014) w10x. In a previous study, this figure was even less for the dino- The meta-analysis of these three trials confirmed that prostone insert group whereby 54% of patients required controlled-release dinoprostone was associated with a labor to be initiated or augmented with oxytocin w47x. At significantly higher rate of cervical ripening (OR 3.99 least 50% of women required three doses of the gel to w95% CI 2.71–5.86x; P-0.0001), onset of labor (OR 494 Rath, Clinical efficacy of controlled-release dinoprostone Primary efficacy outcomes from the three placebo-controlled trials of controlled-release dinoprostone w34, 47, 48x.
aTreatment success is defined as a change in Bishop score of G3, or a Bishop score of G6 at 12 h or delivery within 12 h of insertplacement.
*In nulliparous and multiparous women separately. In all other cases, figures are combined.
In a more recent comparison by Ottinger et al. w31x, 90 prostone gel (ns50) (29.8"22.0 h vs. 62.0"78.8 h, women with an indication for labor induction were ran- respectively wPs0.039x) w43x. Additionally, controlled- domized to receive either intra-cervical gel (ns45) or release dinoprostone was associated with a higher rate controlled-release dinoprostone (ns45), and the advan- of cervical ripening and a higher rate of deliveries within tages of controlled-release formulations were noted. The 24 h than dinoprostone gel (80% vs. 56% and 62% vs.
investigators found that controlled-release dinoprostone was associated with a greater mean change in Bishop In a more recent study, controlled-release dinopros- score (1.8"1.9 vs. 3.2"3.1, Ps0.01) than the gel. In this tone had generally equivalent efficacy to an intra-cervical study, there was no significant difference between treat- gel formulation of dinoprostone w13x. In this trial of 115 ments in the percentage of deliveries within 24 h, and women, the majority needing labor induction for post- there was a non-significant decrease in the application- term pregnancies, there was a slightly higher occurrence to-delivery interval (28.3 and 24.0 h for gel and con- of vaginal delivery in the group receiving controlled- trolled-release formulations, respectively wPs0.19x) w31x.
release dinoprostone than in those receiving the gel, but A prospective multicenter German study found similar this difference did not reach statistical significance.
results. In this study, 158 pregnant women with a Bishop Other studies have shown advantages of a controlled- Score F4 were randomized to receive either controlled- release formulation over vaginal gels in terms of initiating release dinoprostone (ns83) or intra-cervical application sustained high-amplitude contractions w29x or initiating of 0.5 mg prostaglandin E gel (ns75) w33x. The study labor w39x. However, a meta-analysis of nine published found that there was no significant difference in the per- studies concluded that there were no clinically significant centage of vaginal deliveries within 24 h between the differences between controlled-release dinoprostone and women given controlled-release dinoprostone and those alternatives, such as gels, used for cervical ripening in the intra-cervical gel group, and no significant differ- (however, no sub-analysis was done on trials with intra- ence in the mean induction to delivery interval (28 h for cervical or vaginal gels specifically) w19x.
the gel and 21.5 h for the controlled-release formulation).
A recent study by Perry and Leaphart w32x reported There was no significant difference in perinatal outcome.
that intra-cervical placement of the sustained-release The investigators concluded that controlled-release dino- dinoprostone insert decreased time to delivery without prostone is appropriate for use across the whole range increasing the cesarean delivery rate, infectious morbi- of Bishop Scores, from the unripe to the ripe cervix. This dity, or other complications of labor. However, when is of benefit to the obstetrician as only one formulation induction of labor is undertaken with prostaglandins, for induction is required, independent of cervical status.
intravaginal prostaglandin E should be used in prefer- The study also monitored FHR in both sample groups ence to intracervical preparations, because despite being w33x. Fetal heart rate was monitored for 20–30 min in the effective, administration of the vaginal insert is less inva- controlled-release dinoprostone group in accordance sive w30x. In addition, it should be noted that the vaginal with product guidelines, and for 2 h in the intra-cervical insert is not licensed for intra-cervical administration and gel group. No significant difference in fetal outcome as

Source: http://www.medpharm.com.cn/data/upload/download/012_so8NRT.pdf