Effect of acidic vaginal pH on the efficacy of dinoprostone (PGE₂) vaginal tablet for labor induction in full term pregnant women: a randomized controlled trial

Study design

This was a double-blind, parallel-group, with balanced randomization study, conducted between October 29th 2021 and December 30th 2022 at Alexandra General Hospital, Athens, Greece. Local Scientific Committee (an Institutional Review Board) approval was obtained (approval number: 68/7/22-9-21). The study protocol was in compliance with the Declaration of Helsinki and a signed informed consent was obtained from all the participants before their enrollment in the study. The study was registered in ClinicalTrials.gov (ID: NCT05100264) and reported according to the CONSORT statement [11].

Participants

Women that were scheduled for labor induction were considered for recruitment. Prior to enrollment, all women were subject to detailed history taking, complete general and obstetric examination. Inclusion criteria included singleton pregnancy, full term (>37 weeks), cephalic presentation, reactive non-stress test (NST), absence of spontaneous uterine contractions and no contraindications to vaginal delivery. Women with multifetal pregnancy, fetal malpresentation, preterm delivery (<37 weeks), advanced maternal age (>40 years), Bishop score >7, fetal macrosomia (>4,500 g), non-reassuring NST, suspected chorioamnionitis or previous cesarean delivery or other uterine surgery were excluded from this study.

Randomization – blinding

Participants were divided into two groups: Group A, who received acidic vaginal wash and Group B, who received a normal saline vaginal wash. Randomization was computer generated. Allocation concealment was ensured by opaque sealed envelopes. Blinding of outcomes assessor, principal as well as other investigators was ensured as randomization was performed by an external investigator that did not participate in the clinical process of the study or the statistical analysis. The ward midwife opened the randomization envelope after patient enrollment and different types of vaginal washing was applied as per randomization. Normal saline solution was readily available in 10 mL bottles NaCl 0.9 %, while acetic acid (CH₃COOH) was available in 5 % preparation. Syringes containing 10 cc of each respective solutions were prepared and the two different types were marked as A or B. Participant, care giver and outcomes assessor (physician that evaluated the progress of labor as well statistician involved in the analysis) were blinded to the procedure.

Interventions

After enrollment, Bishop score was evaluated and recorded. A non-stress test (NST) was performed and if reassuring, labor induction proceeded. Initially, vaginal washing was applied via a syringe containing 10 cc of the respective solution. After that, a vaginal tablet of 3 mg dinoprostone (Prostin E2^®, Pfizer) was inserted into the posterior vaginal fornix. If needed, vaginal washing and a second dose was repeated in 6 h, after assessing Bishop score and confirming a reassuring NST.

Six hours after the last dose of dinoprostone, a standardized oxytocin augmentation protocol was initiated if needed and dose titrated accordingly. Continuous fetal heart monitoring and uterine activity monitoring were performed in all patients. Progress of labor was monitored and charted in partograph by the labor room doctor. Epidural anesthesia was available on request and recorded. Mode of delivery, complications of induction (tachysystole, postpartum bleeding), reason for cesarean section were also recorded. Tachysystole was defined as >5 contractions in 10 min, averaged over a 30 min window [12].

Outcomes

Primary outcome was duration from induction to delivery. Secondary outcomes included duration from induction to active labor, duration from induction to second stage of labor, failure of induction, mode of delivery and Bishop score changes. Active phase was defined as cervical effacement and dilatation ≥5 cm when effective regular, coordinated uterine contractions were present [13].

Sample size

The available research articles were retrieved and an extrapolation to the effect of dinoprostone, as opposed to the effectiveness of misoprostol, was considered [14]. Specifically, we considered as optimal the level of statistical significance to be set at 0.05 and the power of the study at 0.80 using a 1/1 allocation. The effect size was predetermined to be set 0.25 and the sample size in each group was calculated to result in 93 women. Considering an estimated 10 % of patient withdrawal due to the acidification of the vagina that might prove to be poorly tolerable, we chose to include at least 100 women in each arm. The G power tool v. 3.1.9.2 was used in all cases.

Statistical methods

Statistical analysis was performed using the SPSS 20.0 program (IBM Corp., released 2011. IBM SPSS Statistics for Windows, Version 20.0., Armonk, NY, USA). Evaluation of the normality of distributions was performed with graphical methods and the Kolmogorov–Smirnov analysis. The differences of continuous variables were assessed using the Mann–Whitney and Kruskal–Wallis test (due to the abnormal distribution that was observed during the evaluation of normality), whereas dichotomous variables were analyzed with the chi-square test. Fisher’s exact test was applied wherever the number of observations was lower than five in the case of dichotomous variables. The level of significance for all analyses was set at p <0.05.

Main findings

Our study suggests that the process of vaginal acidification has no effect on the efficacy of the dinoprostone vaginal tablet. There were no statistically significant differences in mode of delivery, duration of different labor stages, Bishop score changes, or possible complications. However, participants in the acidification group less often needed labor augmentation with oxytocin and epidural anesthesia. Even though this finding may be a product of chance alone, one possible explanation is that vaginal acidification may result in better drug absorption which could stimulate more adequate uterine contractions, minimizing the need for oxytocin. This difference in oxytocin augmentation could act as a confounding factor and explain the difference in the need for epidural anesthesia, as these two are often correlated.

Comparison to existing literature

The process of vaginal acidification for improved absorption of prostaglandins has been studied with regards to misoprostol. While there is evidence that moistening vaginal misoprostol tablets with normal saline or acetic acid results in better absorption compared to dry tablets [15], studies investigating clinical outcomes have had inconsistent findings. Three studies reported better outcomes with the use of acetic acid in women with first [16] and second trimester abortions [17, 18], while two others showed no statistically significant results in second trimester abortions [19, 20]. Regarding the efficacy of misoprostol in labor induction, Ramos et al. reported that tablet moistening with 3 % acetic acid solution had comparable outcomes with dry misoprostol [21], while a more recent study by Rashwan et al. showed that women with 5 % acetic acid vaginal douching prior to misoprostol administration required shorter time to reach active phase of labor, compared to those with normal saline or alkaline vaginal douching [5].

There is controversial evidence that vaginal pH may influence the absorption and efficacy of prostaglandins. Regarding dinoprostone, some studies suggest that this has a significant effect on cervical ripening [22], [23], [24] and even on duration and mode of delivery [24], while others show insignificant findings [6, 7]. Results from a recent systematic review and meta-analysis from our department support that the rates of successful vaginal delivery and the intervals between induction and active phase, second stage and delivery seem to be unaffected by vaginal pH [14].

Limitations

To the best of our knowledge, this is the first study investigating the impact of an acidic medium on dinoprostone efficacy. A limitation of this study is that we did not measure the vaginal pH before and after vaginal washing. Therefore, we could not assess if there was a difference in initial vaginal pH between the two groups or the actual impact of vaginal washing in altering the vaginal pH. This limitation is related to financial difficulties that arose from a lack of funding. Another limitation could be the variance in the oxytocin augmentation, as the dosage was titrated individually by each labor room doctor. Nevertheless, it should be stressed that in our institution the protocol for oxytocin augmentation is rather strict, permitting only subtle variations in oxytocin intake in otherwise uncomplicated pregnancies. Therefore, we believe that this variance is less likely to result in significant differences in the reported outcomes. Moreover, the process of randomization and blinding minimizes the impact of this limitation, by ruling out the possibility of patient selection.

Implications

Future research is deemed necessary to further investigate the possible impact of vaginal acidification on the efficacy of dinoprostone as currently the available evidence is mainly focused on misoprostol. This is essential as it will externally validate our findings and the evidence must be provided from studies that will also include a substantial number of patients to meet the appropriate sample size. Further randomized trials could include pH measurement before and after intervention and assess the impact of vaginal washing with greater acetic acid concentration or douching volume. A pharmacokinetic study to investigate the actual absorption of dinoprostone following such interventions would also be valuable.