Success of labor induction for pre-eclampsia at preterm and term gestational ages

Abstract

Objective:

Determine the impact of gestational age (GA) on vaginal delivery following induction of labor (IOL) for pre-eclampsia, and evaluate factors that influence successful induction.

Study Design:

Population-based retrospective cohort of 1 034 552 live births in Ohio (2006–2012). The rate of vaginal delivery in women with pre-eclampsia who underwent induction was calculated with 95% confidence intervals, stratified by week of GA at birth. Factors associated with the decision to undergo IOL, and success of IOL were evaluated, and multivariable logistic regression estimated the strength of association.

Results:

18 296 (71.3%) of the patients who underwent IOL had a vaginal delivery. The majority achieved vaginal delivery at both preterm (66% at 23–36 weeks) and term GAs (72%). Factors most strongly associated with vaginal delivery following IOL for pre-eclampsia included prior vaginal delivery and young maternal age.

Conclusion:

The majority of women with pre-eclampsia who undergo IOL achieve vaginal birth, even at early GAs.

Introduction

Hypertensive disorders complicate up to 10% of pregnancies worldwide, and the most common of these is pre-eclampsia. Although this disorder most commonly presents near term, in some cases pre-eclampsia develops as early as 20 weeks of gestation.¹ Pre-eclampsia complicated by severe features, including severe hypertension, hemolysis elevated liver enzymes and low platelet syndrome, renal failure, pulmonary edema and eclampsia requires expedited delivery, regardless of gestational age.

In general, cesarean section is reserved for usual obstetric indications in patients with pre-eclampsia. Many women with pre-eclampsia do not labor spontaneously and require induction when delivery is indicated. Whether or not to pursue induction of labor in women with pre-eclampsia has been debated. Some researchers in the field have suggested that labor induction is actually easier in women with pre-eclampsia due to an increase in uterine activity.² Other experts advocate cesarean for delivery for all women with severe pre-eclampsia at less the 30 weeks gestation who are not in labor and have an unfavorable cervical exam.³

The impact of gestational age on success of induction of labor is patients with pre-eclampsia remains controversial. Prior studies have reported that patients with pre-eclampsia have higher rates of cesarean section following induction compared to the general population, regardless of parity or gestational age.⁴ However, there are conflicting reports in the literature of rates of cesarean delivery in patients with pre-eclampsia following induction at preterm and term gestational ages.^{5, 6, 7, 8} The purpose of this study is twofold: to determine the impact of gestational age on vaginal delivery rate following labor induction for pre-eclampsia, and to evaluate the maternal and fetal factors that influence successful labor induction and should be considered when counseling patients about mode of delivery with pre-eclampsia, especially during the preterm period.

Materials and methods

We performed a retrospective population-based cohort study of all Ohio live births over a 7 year period, 2006–2012. The protocol for this study was approved by the Ohio Department of Health and Human Subjects Institutional Review Board, and a de-identified set of birth certificate data was provided for this analysis. This study was exempt from review by the Institutional Review Board at the University of Cincinnati, Cincinnati, OH, USA.

The primary aim of this study was to determine the rate of vaginal delivery at each week of gestational age (23–42 weeks) in patients who underwent induction of labor due to pre-eclampsia. Secondarily, we evaluated maternal and fetal factors associated with the decision to attempt labor induction. The database included 1 034 552 live births in Ohio from 2006–2012. Deliveries that occurred at gestational ages <20 weeks or >42 weeks, as well as those with missing data were excluded from this analysis. Of the remaining 1 016 083 patients, there were 49 920 patients who were diagnosed with pre-eclampsia and were included in this analysis.

Maternal demographic, social and prenatal factors were compared between women with pre-eclampsia who¹ underwent induction of labor and delivered, versus those who delivered without labor induction, and² those who underwent labor induction and were successful (vaginal delivery), versus those who had a failed labor induction (cesarean delivery). The 2003 version of the United States birth certificate was utilized for all live births in Ohio during the study period. Induction of labor was recorded in the birth certificate defined as ‘initiation of uterine contractions by medical and/or surgical means for the purpose of delivery before the spontaneous onset of labor’, as specified in the National Vital Statistics System Guide for Completing the Facility Worksheets for the Certificate of Live Birth in the US.⁹

The primary exposure variable for this study, gestational age at delivery, was defined using the variable combined estimate of gestational age, which takes into account a combination of last menstrual period, ultrasound and clinical dating–as is commonly defined in clinical practice.⁹ The US birth certificate does not have an individual category for ‘pre-eclampsia’; therefore in the present study ‘pre-eclampsia’ included patients identified as having gestational hypertension, pregnancy induced hypertension, pre-eclampsia or eclampsia, but not those with only chronic hypertension.⁹ Pregnancy weight gain categories were defined by the Institute of Medicine (IOM) recommended weight gain according to their pre-pregnancy BMI class.¹⁰ Gestational weight gain was adjusted for gestational age at delivery for assignment of gestational weight gain categories. Intrauterine growth restriction was defined as estimate fetal weight below the 10th percentile for gestational age.¹¹

Statistical analyses were performed using STATA Release 12 software (StataCorp, College Station, TX, USA). Demographic characteristics were analyzed using unpaired Student’s t-tests for continuous variables and Χ² tests for categorical variables. A two-sided P-value of <0.05 was considered statistically significant. Multivariable logistic regression was performed to estimate the relative influence of these characteristics on rate of vaginal delivery in women whose labor was induced. Adjusted odds ratio and the 95% confidence interval were calculated using variables that were most significant. Covariates were selected based on significance of differences noted in univariate comparisons and those with biologic plausibility including maternal age, race, prior vaginal delivery, birth weight, maternal obesity, prior cesarean section, malpresentation and gestational weight gain greater than IOM guidelines. The rate of successful vaginal delivery following induction of labor and the 95% confidence interval were calculated for each week of gestational age at birth.

Results

Characteristics of women with pre-eclampsia who underwent induction of labor (n=25 773 51.6%) and those who did not (n=24 147 48.4%) are presented in Table 1a. By gestational age, number and percent of patients induced were as follows: <27 weeks (66,11%), 27–30 weeks (244,18%), 31–34 weeks (1378,33%) and >34 weeks (24085,55%). The rate of induction in patients greater than 34 weeks was found to be significantly higher compared to those ⩽34 weeks (P<0.005). Median and interquartile range of continuous variables including maternal age, gestational age at delivery and birth weight were found to be non-normally distributed and can be found in Tables 1a and b. Women who underwent induction were more likely to be white and under age 35. The average birth weight in the induction group was higher and these pregnancies were more likely to be term. Women carrying a growth-restricted fetus and those with diabetes or chronic hypertension were less likely to be induced. Other characteristics associated with a higher rate of induction of labor included being married, prior vaginal delivery, non-obese, nulliparity and gestational weight gain greater than the IOM guidelines. Of all patients who were not induced, 23.2% had at least one prior cesarean section, but of those with one prior cesarean who were induced, 48% had a successful vaginal delivery. None of the patients with two or more prior cesarean sections successfully delivered vaginally following induction. Characteristics associated with a lower rate of induction included being a high school graduate, medicaid, underweight, obese, limited or no prenatal care, prior cesarean section, macrosomia, malpresentation and gestational weight gain equal to or less than the IOM guidelines.

Table 1a Characteristics of patients selected for induction of labor versus no induction of labor in pregnancies complicated by pre-eclampsia

Table 1b Characteristics of patients who had a successful induction of labor versus failed induction of labor in pregnancies complicated by pre-eclampsia

Table 1b summarizes the characteristics of women who underwent induction of labor and subsequently had a successful vaginal delivery (n=18 296) versus those who underwent labor induction and delivered by cesarean section (7366; 111 had unknown mode of delivery). Successful induction, defined as vaginal birth following labor induction, was more likely in women who were younger and white. Successful labor induction was also increased in women who had gestational weight gain equal to or less than the IOM recommendations. Diabetes, chronic hypertension and obesity were associated with decreased success rates. The rate of vaginal delivery was increased in patients with a term pregnancy. Other factors associated with a higher rate of successful induction of labor included being married, having finished high school, non-obese and having a prior vaginal delivery. Cigarette use, nulliparity, intrauterine growth restriction, malpresentation and prior cesarean section were associated with a lower rate of vaginal delivery following induction.

Multivariable logistic regression identified factors most strongly associated with vaginal delivery following induction (Table 2). Factors with the most robust positive association with successful labor induction were prior vaginal delivery (aOR, 5.91; 95% CI, 5.44–6.43) and young maternal age (aOR 1.44; 95%CI, 1.30–1.59). Factors negatively associated with vaginal delivery after labor induction for pre-eclampsia include black race, birth weight >4000 g or <2500 grams, maternal obesity, prior cesarean, fetal malpresentation and pregnancy weight gain greater than IOM recommendations. Increasing GA was only modestly associated with higher odds of vaginal birth (aOR 1.03; 95%CI 1.01–1.05).

Table 2 Factors associated with outcome of induction of labor in women with pre-eclampsia

The primary outcome of rate of successful induction of labor at each week of gestational age is displayed in Figure 1. The majority of patients who underwent induction for pre-eclampsia achieved vaginal delivery at all gestational ages >30 weeks. The rate of successful induction was lowest at 25 weeks of gestation (30%; 95% CI, 8.3–62.0%) and highest at 38 weeks (73.5%; 95% CI 72.3–74.7%). Overall, the rate of successful induction was 66% at preterm gestational ages (23–36 weeks) and 72% at term gestational ages (⩾37 weeks). By gestational age, number and percent of patients induced were as follows: <27 weeks (66, 11%), 27–30 weeks (244, 18%), 31–34 weeks (1378, 33%) and >34 weeks (24,085, 55%). Prior to 30 weeks, 16% of patients (n=310) were offered induction, and 51% (n=157) delivered vaginally. Beyond 30 weeks, 53% (n=25,463) had an induction of labor and 71% (n=18,139) had a vaginal delivery. Success increased with each successive week of preterm gestational ages from 30 to 39 weeks, then decreased beyond 40 weeks. Induction was successful in >45% at all gestational ages >23 weeks, and exceeded 50% in all gestational ages except for 25 weeks (30%), 28 weeks (47.8%; 95% CI, 33.75–62.2%) and 30 weeks (45.5%; 95% CI, 36.5–54.8%).

Figure 1

Rate of vaginal birth (with 95% confidence interval) following labor induction for preeclampsia, stratified by week of gestational age.

Comment

This study describes the success rate of labor induction in patients with pre-eclampsia at each gestational age from 23 to 42 weeks. It is known that there is a higher rate of failed induction in patients diagnosed with pre-eclampsia when compared to the general population. One smaller retrospective study reported a higher rate of cesarean delivery in pregnancies complicated by pre-eclampsia undergoing induction of 26.8%, compared to 15.4% in those without pre-eclampsia (P<0.001), and this was true at both term and preterm gestational ages.⁴ In patients with other concurrent risk factors for failed induction, such as unfavorable cervical exam, morbid obesity and nulliparity, physicians may recommend cesarean section for delivery. However, the results of our study demonstrate that in a large cohort of women, premature gestational age does not have a strong association with failed labor, and therefore induction may be offered as a reasonable option in patients without contraindication to labor who require delivery at premature gestational ages.

We analyzed demographic and pregnancy factors associated with induction versus elective cesarean or spontaneous labor, and the rate of success in those who were induced. Prior studies have shown that reasons for planned cesarean section rather than induction of labor in women with pre-eclampsia include unfavorable cervix, malpresentation, non-reassuring fetal heart pattern, prior cesarean section, placenta previa and worsening maternal condition.⁷ We found that certain demographic factors were associated with a significantly increased rate of induction of labor. Prior cesarean delivery accounted for almost 25% of the patients who were not induced, and appears to be one of the most compelling reasons that providers proceed immediately to cesarean delivery. Of the patients who had one prior cesarean and were induced, 49% had a successful vaginal birth after cesarean. And of the 23 patients with two or more prior cesarean deliveries, 11 (48%) delivered vaginally. Women were more likely to undergo induction, however, if they had a prior vaginal delivery or delivered at term. The average gestational age at delivery was significantly lower in the group who were delivered electively by cesarean section compared to those were offered induction (36 weeks gestation compared to 38 weeks, respectively; P<0.001). Both of these factors have been shown to increase the rate of vaginal delivery following induction for multiple factors in prior studies. Recently, Feghali et al. found that in the general population, the rate of vaginal delivery following induction increased with increasing preterm gestational age, from 35% at 24–27+6 weeks gestation to 76% at 34–36+6 weeks. They also found that the strongest predictor of vaginal delivery was increased parity.¹² Nulliparity and obesity are two known factors that increase the risk of both cesarean following induction and risk of pre-eclampsia. One recent study of 273 women confirmed nulliparity as an independent risk factor for developing pre-eclampsia (aOR 1.73, 95% CI 1.26–2.38).¹³ This study also found obesity be the strongest risk factor for pre-eclampsia, with an attributable risk of 64.9%. A dose-response effect between BMI and pre-eclampsia was shown; patients with Class 3 obesity had the highest risk of pre-eclampsia (aOR 6.04, 95% CI 3.56–10.24.¹³ In our study, obese patients were slightly less likely to be induced (39.8%, compared to 41.5% who were not induced, P<0.001) however were far less likely to have a successful induction (36.5%, compared to 63.5% who were non-obese, P<0.001). In our large cohort of patients with pre-eclampsia, who are more likely to possess these characteristics, we found that each of these factors was negatively associated with the likelihood of vaginal birth following induction.

Outcomes of induction of labor in women with pre-eclampsia at preterm and term gestational ages are valuable for appropriate counseling of patients about rates of success. Several small studies reporting vaginal birth following labor induction in women with pre-eclampsia at preterm gestational ages have been published. One study of 145 women reported outcomes of labor induction in those <34 weeks gestation.⁷ Another study of women with hemolysis elevated liver enzymes and low platelet who underwent induction included 189 patients.¹⁴ Other recent studies including a larger number of women reported induction outcomes by gestational age, however, these studies did not look at rates in women with pre-eclampsia.¹⁵ There are no prior studies to our knowledge that were large enough to have sufficient power to report outcomes of induction at specific gestational ages in women with pre-eclampsia, similar to ours. Because of the large population-based nature of our study population (n=49 920) and its representation of a variety of hospital types, we feel that our findings are generalizable and relevant to most patient populations. Regarding the accuracy of recorded labor induction, the use of birth certificate data for determining the rate of induction was previously found to be highly specific in cases which were deemed medically indicated.¹⁶ This type of study is limited, however, by the limited clinical information available from the data source regarding the delivery. We were unable to analyze factors such as cervical dilation, method of induction, labor duration and delivery complications as those are not included in the US birth certificate. The Bishop score has also been shown in many studies to be one of the strongest predictors of vaginal delivery following induction.^{7, 15} This concern may be most relevant at early gestational ages, when the provider may choose only to induce women with a more favorable cervix. Although it is unlikely that a significant number of women would have had a favorable cervix at a very early gestational age, a detailed chart review would be required to determine whether or not this may have influenced the decision to induce labor in these patients. Although these factors influence mode of delivery and obstetric outcomes, the primary aim of this study was to analyze induction outcomes in a large number of women with pre-eclampsia to produce generalizable gestational-age specific information. Prospective studies are needed to determine the optimal mechanism of induction, and the utility of the Bishop score in these women. We were also limited by the lack of standardized approach to mechanism of induction, labor management and guidelines for when to perform a cesarean section in patients who were induced, which can vary greatly between individuals and institutions. This is especially significant in the setting of preterm pre-eclampsia as providers may be less likely to aggressively pursue vaginal delivery.

Pre-eclampsia remains the most common medical indication for preterm induction of labor.^{15, 17} Factors most strongly associated with pre-eclampsia are also known risks for failed labor induction, creating a difficult task for obstetricians when deciding how to proceed with delivery. In our large cohort of patients who were delivered because of pre-eclampsia, induction was successful the majority of the time regardless of gestational age. This information may be encouraging for patients when faced with preterm labor induction even at early gestational ages, and may ultimately avoid unnecessary primary cesarean deliveries and their associated complications.