Labor induction is a common obstetric practice. According to the National Center for Health Statistics data from 2017, 25.7% of gravid women underwent labor induction, an increase from 23% in 2012.1 Induction rates will likely continue to increase given the rising prevalence in obesity and the findings from the ARRIVE trial, which showed decreased risk for cesarean among nulliparous women undergoing induction compared with expectant management.2–5 Misoprostol, a synthetic prostaglandin E1 analog, is a common medication used for cervical ripening during labor induction and is administered in various doses, frequencies, and routes. In 2014, a Cochrane review on the use of orally administered misoprostol for labor inductions suggested that vaginal misoprostol was less effective than oral misoprostol at achieving vaginal birth, with an increased risk of uterine tachysystole and cesarean delivery with vaginal misoprostol.6 One limitation of this systematic review and meta-analysis was that combined results of different doses of vaginal misoprostol were included in the analysis. As the hyperstimulatory effect of misoprostol is dose-dependent, a subgroup analysis of the safest recommended dose of vaginal misoprostol (25 micrograms) compared with a commonly used oral misoprostol dose (50 micrograms) is necessary to help assess the efficacy and safety of oral compared with vaginal misoprostol for labor induction. In addition, body mass index (BMI, calculated as weight in kilograms divided by height in meters squared) was inadequately evaluated in comparing outcomes of vaginal compared with oral misoprostol used for labor induction.
Given these knowledge gaps, the purpose of this study was to assess the frequency of cesarean delivery with oral misoprostol (50 micrograms) compared with vaginal (25 micrograms) misoprostol in a predominantly overweight population. Among women who successfully achieved a vaginal delivery, we further sought to assess time to vaginal delivery between the two methods.
This retrospective cohort study was performed at a single academic tertiary care center, Boston University Medical Center. Boston University Medical Center is a large urban safety net hospital that cares for a racially diverse and underserved patient population. After publication of a 2014 Cochrane review on oral misoprostol administration, Boston University Medical Center enacted a new protocol administering oral instead of vaginal misoprostol for cervical ripening with the objective of decreasing cesarean delivery rates, minimizing delay in provider administration of vaginal misoprostol, and decreasing the number of necessary vaginal examinations performed during labor inductions.6 To assess the effect of this change, outcomes for women receiving vaginal misoprostol for cervical ripening from 2013 to 2014 during labor induction were compared with women receiving oral misoprostol from 2014 (after the switch to oral misoprostol) through 2015; during each respective time period, vaginal or oral misoprostol were used exclusively for cervical ripening.
The study was approved by the Boston University Institutional Review Board (IRB H-35478, 2016). All women who presented for labor induction during these time periods were identified by the clinical data analyst in the Obstetrics and Gynecology Department. Inclusion criteria consisted of a live singleton pregnancy of at least 34 weeks of gestation, a documented cervical examination with dilatation of 1 cm or less, and initiation of labor induction with either vaginal misoprostol or oral misoprostol. Women were excluded if they had a prior cesarean delivery, if they erroneously received multiple forms of misoprostol during their induction, or if the labor induction was interrupted as a result of systems issues such as floor acuity.
A power analysis with two sample proportions chi-squared test showed that a sample size of 276 (138/group) was required to detect a 15% absolute rate difference in cesarean delivery between the two groups with 80% power. Therefore, the first 138 women receiving vaginal or oral misoprostol who met the study criteria were included in the analysis. To ensure accuracy and consistency of inclusion, two investigators independently screened the charts.
We collected and analyzed a variety of baseline characteristics (Table 1). The diagnosis of failure to progress was obtained from the cesarean delivery indication in the operative note and confirmed by review of labor progress note documentation. In addition, intrapartum clinical actions including cervical Foley catheter use, artificial rupture of membranes, regional anesthesia, and timing between consecutive misoprostol dosing were assessed.
We compared the outcomes of women whose labor induction was initiated with either 50 micrograms of oral misoprostol or 25 micrograms of vaginal misoprostol. The institutional protocol for cervical ripening was to readminister misoprostol every 4–6 hours if there were fewer than four contractions in a 10-minute period at the time of assessment, a category I fetal heart rate tracing, and Bishop score less than 7, for a maximum of six misoprostol doses. Resident physicians and attendings performed the cervical examinations. Cervical Foley catheter balloons were placed for up to 12 hours; cervical Foley catheters were not used with ruptured membranes, and oxytocin was not initiated until the Foley catheter was removed or fell out. The combination of Foley catheter with prostaglandins or oxytocin was not used during the study period.
The primary outcome was cesarean delivery. Secondary outcomes included censored time to vaginal delivery, indications for cesarean delivery, tachysystole, maternal hemorrhage (500-mL blood loss for vaginal delivery and 1,000-mL of blood loss for cesarean delivery), composite neonatal morbidity (5-minute Apgar scores less than 7, unplanned neonatal intensive care unit admission, and neonatal infection requiring postnatal antibiotic administration), number of sterile vaginal examinations performed, oxytocin augmentation requirement, and number of misoprostol doses required for each induction. Dosing interval between subsequent misoprostol doses was recorded and when available, nursing and provider comments explaining any delay in dosing interval were noted.
Failed induction included failure to progress in either the latent or active phase. Latent phase arrest was defined as maximum cervical dilation under 6 centimeters with inadequate contractions, cervical change, or both, despite ruptured membranes and oxytocin administration for 24 hours. Active phase arrest was defined as arrest with ruptured membranes and cervical dilation 6 centimeters or more with 4 or more hours of adequate contractions (more than 200 Montevideo units/10-min interval) or 6 or more hours of inadequate contractions without cervical change. Active labor was defined as cervical dilation of 6 centimeters or more. If a woman had both a nonreassuring fetal heart rate tracing and failed induction as an indication for cesarean delivery, nonreassuring fetal heart rate tracing was categorized hierarchically as the indication for cesarean. Failure to descend was defined as 10-cm dilation and no progress for: 1) 4 or more hours among nulliparous women with an epidural, 2) 3 or more hours among nulliparous women without an epidural, 3) 3 or more hours among multiparous women with an epidural, or 4) 2 or more hours among multiparous women without an epidural.
Demographics, intrapartum outcomes, and labor outcomes were analyzed using unpaired t-tests for continuous data and Chi-square test or Fisher's exact test for categorical data. ANOVA was used to analyze categorical variables with more than one group. Nonparametric k-sample test on the equality of medians was used to compare variables with integer values. Laplace regression for censored time to events, univariable and multivariable logistic and linear regression models were used to control for prognostic covariates in time to event analyses, categorical and continuous variables, respectively. Multivariable analyses included clinically significant factors, those with a P-value .2 or less on univariable comparison, and those considered to be statistically significant (Appendix 1, available online at http://links.lww.com/AOG/B391). Cochrane-Armitage test was used to compare cesarean delivery rates over the years spanning the study period. Results are reported as unadjusted odds ratio (OR) and adjusted OR (aOR). We used an alpha of 0.05 as a measure of statistical significance. All statistical analyses were performed with STATA/MP 15.1.
In the final cohort, 138 women who received vaginal misoprostol and 138 who received oral misoprostol were included (Fig. 1). Baseline demographics did not differ significantly between the two groups (Table 1). The overall cohort was predominantly overweight or obese, with 91% of women having a BMI of 25% or more and 61% with a BMI of 30 or more at the time of delivery.
Cervical Foley catheter balloon use was more frequently used in the vaginal (57%) compared with the oral (41%) misoprostol groups (P=.01, Table 2). The number of women receiving more than one misoprostol tablet for cervical ripening was higher in the oral misoprostol group (P<.01). The median number of vaginal examinations performed was lower in the vaginal misoprostol group (P<.01). Tachysystole occurred more frequently with vaginal misoprostol use (P=.04). Among women receiving multiple misoprostol doses, the time to subsequent dosing was significantly longer in the vaginal compared with oral misoprostol group (P<.01) (Table 2).
The frequency of cesarean delivery was significantly higher with oral (32%) compared to vaginal misoprostol (21%) use (P=.04, Table 3). In the adjusted analysis accounting for age, BMI, parity, indication for induction, and Foley catheter use, the odds of cesarean delivery remained significantly higher for women receiving oral misoprostol (aOR 2.01, 95% CI 1.07–3.76). Among nulliparous women the frequency of cesarean delivery was significantly higher in the oral (41%) than the vaginal (28%) misoprostol group (aOR 2.79, 95% CI 1.26–6.19) The overall cesarean delivery rates for all women presenting to labor and delivery did not significantly change between the years spanning the study period (cesarean rates were 28.3%, 31.1%, and 31.6% in 2013, 2014, and 2015, respectively) (P=.27). The median (interquartile range) time to vaginal delivery in the adjusted censored analysis showed significantly shorter delivery time in the vaginal than oral misoprostol group, 31 (19.2–54.8) and 41 (26.0–72.9) hours, respectively (P<.01).
Secondary maternal and neonatal outcomes were generally similar between groups. There were more cesarean deliveries resulting from failed labor inductions in the oral group (18% vs 9%, P=.02) (Table 4).
This retrospective cohort study found that induction of labor with 50 micrograms of oral misoprostol was associated with higher odds of cesarean delivery compared with 25 micrograms of vaginal misoprostol. This association persisted in subgroup analysis restricted to nulliparas. In addition, median [interquartile range] time to vaginal delivery was shorter in the vaginal compared with oral misoprostol groups.
The differential outcomes in this study may be secondary to different pharmacokinetics for oral compared with vaginal misoprostol. Oral misoprostol undergoes rapid absorption from the gastrointestinal tract and rapid and extensive de-esterification during first-pass metabolism to an active metabolite, misoprostol acid, peaking at 15 minutes with a half-life of 20–40 minutes. Misoprostol acid then undergoes early rapid elimination over 120 minutes, followed by slow elimination thereafter. The medication rapidly makes its way to the myometrium. In contrast, after vaginal misoprostol administration, plasma concentrations gradually increase, reaching a maximum level after 70–80 minutes before slowly being eliminated with plasma levels still detectable 6 hours after administration.7 The rapid plasma accumulation and decline after oral misoprostol administration may account for longer times to vaginal delivery and increased cesarean delivery rates given oral misoprostol's more transient steady-state plasma concentrations.
Several randomized trials comparing the use of 25 micrograms of vaginal misoprostol with 50 micrograms of oral have demonstrated superior efficacy with vaginal misoprostol for cervical ripening. A double blind randomized trial comparing the use of 25 micrograms of vaginal and 50 micrograms of oral misoprostol showed similar results to our study, with shorter labor induction times, lower cesarean delivery rates, and fewer required doses of misoprostol to achieve vaginal delivery in the vaginal misoprostol group.8 Other studies have demonstrated shorter induction to delivery times and fewer misoprostol doses with the vaginal route.9,10 On the other hand, a randomized trial including a total of 200 women, using similar misoprostol doses as our study, showed no significant difference in time to vaginal delivery between the two groups, and a lower cesarean delivery rate with oral misoprostol.11 However, this study had more women with hypertensive disease or preeclampsia in the vaginal (43%) compared with oral misoprostol group (20%), which may have confounded the results because preeclampsia was the main indication for cesarean delivery in this trial. Furthermore, the study did not report the baseline demographic BMI of the cohort, and it was not powered to see a difference in cesarean delivery rates between the two groups.
One of the limitations of our study is the potential for bias, as it is retrospective. To address this issue, charts of women who presented for induction were assessed in chronologic order by time of induction and the first 138 women meeting criteria in each group were included in the study to limit selection bias. In addition, two investigators independently screened the charts to ensure accuracy and consistency of inclusion. Another potential limitation of the study is the different time periods during which each route of misoprostol was used. The years were evaluated in immediate succession when there was minimal other change in general obstetric management. Although resident physicians both graduated and entered the training program during the study period, there were no large-scale changes in leadership or attending physician faculty during this period. Given the stability of the clinical faculty, resident supervision was consistent over the study period. Interrogation for differences in the patient population demonstrated no statistically significant findings in comparison of demographics between groups. Additionally, the study was not powered to evaluate many secondary outcomes, including more detailed BMI categories. Lastly, although this study showed significant differences with route of misoprostol administration for a number of outcomes in both adjusted and unadjusted models adjusted estimates could be affected by some degree of overfitting, particularly if conservative assumptions related to model construction are applied.12
Strengths of this study include the racial and ethnic diversity and the pragmatic retrospective analysis of an institutional change. Given the before and after study approach, significant selection bias with regards to route of administration is less likely. The overall clinical volume and cesarean delivery rates did not change significantly throughout the years spanning the study period, making confounding secondary to temporal trends less likely. Furthermore, the doses of oral (50 micrograms) and vaginal (25 micrograms) assessed in this analysis are among the most commonly used in the United States, and considered the safest doses per route, adding to the generalizability of the study. Additionally, given that inductions in this study were performed outside a research protocol, the results may be more generalizable to more obstetric practices.
In summary, we found increased risk for cesarean delivery with oral compared with vaginal misoprostol along with longer induction times in a diverse and predominantly overweight population. Future studies in different settings are required to assess the replicability and generalizability of these findings.
1. Osterman MJ, Martin JA. Recent declines in induction of labor by gestational age. NCHS Data Brief 2014;155:1–8.
2. Grobman WA, Rice MM, Reddy UM, Tita ATN, Silver RM, Mallett G, et al. Labor induction versus expectant management in low-risk nulliparous women. N Engl J Med 2018;379:513–23.
3. Mendy VL, Vargas R, Cannon-Smith G, Payton M. Overweight, obesity, and extreme obesity among Mississippi adults, 2001–2010 and 2011–2015. Prev Chronic Dis 2017;14:E49.
4. Ogden CL, Carroll MD, Kit BK, Flegal KM. Prevalence of obesity among adults: United States, 2011–2012. NCHS Data Brief 2013:1–8.
5. Wang YC, McPherson K, Marsh T, Gortmaker SL, Brown M. Health and economic burden of the projected obesity trends in the USA and the UK. Lancet 2011;378:815–25.
6. Alfirevic Z, Aflaifel N, Weeks A. Oral misoprostol for induction of labour. The Cochrane Database of Systematic Reviews 2014, Issue 6. Art. No.: CD001338.
7. Bygdeman M. Pharmacokinetics of prostaglandins. Best Pract Res Clin Obstet Gynaecol 2003;17:707–16.
8. Jindal P, Avasthi K, Kaur M. A comparison of vaginal vs. oral misoprostol for induction of labor-double blind randomized trial. J Obstet Gynaecol India 2011;61:538–42.
9. Ezechukwu PC, Ugwu EO, Obi SN, Chigbu CO. Oral versus vaginal misoprostol for induction of labor in Enugu, Nigeria: a randomized controlled trial. Arch Gynecol Obstet 2015;291:537–44.
10. Wing DA, Ham D, Paul RH. A comparison of orally administered misoprostol with vaginally administered misoprostol for cervical ripening and labor induction. Am J Obstet Gynecol 1999;180:1155–60.
11. Komala K, Reddy M, Quadri IJ, B S, V R. Comparative study of oral and vaginal misoprostol for induction of labour, maternal and foetal outcome. J Clin Diagn Res 2013;7:2866–9.
12. Vittinghoff E, McCulloch CE. Relaxing the rule of ten events per variable in logistic and Cox regression. Am J Epidemiol 2007;165:710–8.