Over the past few decades, obesity has emerged as one of the most common medical problems in the world. In the United States, the rates have more than doubled in the past 25 years from 15% in 1980 to 32.9% in 2004.1 Moreover, nearly one third of women of reproductive age are obese, and approximately 6% are extremely obese.1 In addition to the usual health implications, obesity increases the rate of complications associated with pregnancy. Numerous studies have demonstrated an increased risk of gestational diabetes, pregnancy-induced hypertension, and fetal macrosomia in obese patients. There is also evidence that labor tends to progress more slowly in obese patients.2 In addition, overweight women are often admitted at an earlier and less favorable stage of labor and subsequently have higher rates of labor induction.2,3 Given the fact that induction of labor has become a common obstetric intervention in the United States, the issue affects a considerable portion of the population. Recent data revealed that the rates of induction more than doubled from 9.5% in 1990 to 22.5% in 2006.4 This increase has been attributed to a variety of social and legal reasons as well as an increase in medical and obstetric indications, many of which are proportionally more frequent in obese women. The increased likelihood of obstetric complications, higher rates of induction, and slower labor progress have all been independently implicated in the increased likelihood of cesarean delivery.2,5
The objective of this study was to estimate the effects of maternal obesity, stratified by body mass index (BMI), on progress and outcomes of labor induction.
MATERIALS AND METHODS
The study was a secondary analysis of the data collected from the Misoprostol Vaginal Insert randomized comparator-controlled trial.6 The original trial was a multisite, double-blind, randomized study comparing the efficacy and safety of three sustained-release vaginal inserts containing dinoprostone 10 mg (Cervidil; Forest Laboratories, Inc., New York, NY), misoprostol 50 micrograms, or misoprostol 100 micrograms. Forty-nine institutions participated in the trial, and 1,308 patients were enrolled from April 26, 2006, through August 7, 2007.
Enrollment criteria for this study can be found in the original publication of the Misoprostol Vaginal Insert Trial.6 In brief, the participants were aged at least 18 years, of low parity (three or less) with singleton pregnancies, and at least 36 weeks 0 days of gestation. They had no conditions requiring urgent delivery and required cervical ripening defined as a baseline modified Bishop score of 4 or less. Participants were excluded if labor or vaginal delivery was contraindicated, delivery was deemed to be urgent, or there was a known sensitivity to either misoprostol or dinoprostone. Random assignment was ensured by providing sequentially numbered kits for each parity cohort (nulliparous and parous). The randomization sequence was created using a computer-generated randomization scheme and was stratified by center with a 1:1:1 allocation for each arm of the study. The trial demonstrated similar rates of cesarean delivery and safety profiles for the three vaginal insert preparations. In addition, 100-microgram misoprostol vaginal inserts and dinoprostone vaginal inserts had similar median intervals from insertion of the study drug to vaginal delivery that were significantly shorter than the time interval for 50-microgram misoprostol vaginal inserts.
The original Misoprostol Vaginal Insert Trial presented the data on 1,308 patients using an intent-to-treat analysis. The current analysis was confined to patients who completed the study and delivered during their first hospital admission. The study sponsor (Controlled Therapeutics [Scotland] Ltd., East Kilbride, United Kingdom) used all the available data collected in the course of the trial to construct a comprehensive database that was made available to the current authors for secondary analysis. University of California, Irvine institutional review board approval was obtained for the use of these data in a secondary analysis.
Maternal height and weight were collected at the time of admission to labor and delivery and were used to derive the BMI for each patient. Maternal BMI was classified based on the World Health Organization categories.7 Given the small number of women (n=70) meeting the criteria for “normal range” BMI (18.5–24.9), meaningful comparison could not be performed using this category as the reference group. Instead, these participants were combined with the “overweight” (BMI 25.0–25.9) category and were collectively referred to as lean or BMI less than 30. Labor characteristics and outcomes of these participants were then compared with those of obese (BMI 30–39.9) and extremely obese (BMI 40 or higher) participants.
After study enrollment, an examination was performed to confirm vertex fetal presentation and to determine baseline modified Bishop score. The design of the trial called for a dosing period of up to 24 hours with the study drug unless there was a reason for earlier removal. Primary reasons for removal of the study drug insert included onset of active labor, study drug falling out of the vagina, completion of the 24-hour dosing period, and maternal–fetal complications. A cervical examination conducted 12 hours after study drug insertion was used to calculate the 12-hour modified Bishop score. Active labor was defined as at least three firm, rhythmic contractions lasting 45 seconds or longer within a 10-minute period with progressive cervical dilatation, or achievement of 4-cm dilatation. Frequency, duration, and amount of predelivery oxytocin use was recorded and subcategorized into oxytocin for induction and oxytocin for augmentation based on the absence or presence of active labor, respectively. Time to delivery was determined from insertion of the study drug to delivery of the fetus.
Statistical analyses were performed using JMP 7.0 (SAS Institute Inc., Cary, NC) statistical software, and all tests were conducted at the 0.05 significance level. Comparisons are reported using the BMI less than 30 group as the referent, and separate analyses were performed comparing the BMI 30–39.9 with the BMI less than 30 group and the BMI 40 or higher with the BMI less than 30 group. Normality of continuous data was assessed by the Shapiro-Wilk test. Normally distributed data were compared using analysis of variance and Student t tests, whereas continuous data that were not normally distributed were compared with the Kruskal-Wallis rank sum test. Categorical variables were evaluated using the χ2 test or Fisher exact test. Multivariable regression analysis was performed on all outcomes of interest, adjusting for race, parity, and treatment group allocation.
A total of 1,308 women were enrolled in the original trial from April 26, 2006, through August 7, 2007. Of 1,297 women who completed the study, 1,274 delivered during their first admission. Anthropometric information was not available for 1 study participant, leaving 1,273 participants for the final analysis. Demographic and baseline characteristics were comparable between study groups (Table 1). Four hundred eighteen study participants were classified as lean or BMI less than 30, 644 were classified as obese or BMI 30–39.9, and 211 were classified as extremely obese with BMI 40 or higher. Maternal age, proportion of nulliparous women, gestational age at induction, maternal height, and treatment group allocation were similar between the study groups. Obese and extremely obese groups had a higher proportion of African-American and Hispanic women compared with the BMI less than 30 group. Finally, reasons for induction were statistically different between the three BMI groups. This was particularly true in regard to conditions generally associated with obesity, such as hypertension/preeclampsia, diabetes, and suspected macrosomia.
Comparison of the 1,273 study participants revealed similar baseline modified Bishop scores at enrollment (Table 2). After 12 hours of labor induction with the study drug, the difference between groups in modified Bishop score was statistically but not clinically significant (P=.004). Evaluation of delivery (regardless of route) within 24 hours after initiation of induction demonstrated a significant difference, with 54.6% of lean participants delivering, compared with 47.5% of obese participants (odds ratio [OR] 0.75, 95% confidence interval [CI] 0.59–0.97, P=.025) and 41.7% of extremely obese women (OR 0.60, 95% CI 0.43–0.83, P=.002). Confining the analysis to vaginal delivery participants revealed a similar, albeit not statistically significant, trend, with 58.4% of lean participants delivering within 24 hours, compared with 52.9% of obese participants (OR 0.80, 95% CI 0.60–1.07, P=.13) and 49.3% of extremely obese women (OR 0.69, 95% CI 0.46–1.04, P=.07). However, multiple logistic regression analysis adjusting for race, parity, and treatment group allocation demonstrated a significant correlation between maternal obesity and vaginal delivery within 24 hours (Table 2).
Although there was no difference in overall oxytocin requirements among the BMI categories, a stratification of oxytocin use by induction and augmentation revealed significant differences. Overall, obese (39.4%) and extremely obese (48.3%) women were more likely to require predelivery oxytocin for induction (OR 1.46, 95% CI 1.12–1.89, P=.004 and OR 2.10, 95% CI 1.49–2.95, P<.001, respectively) compared with participants in the BMI less than 30 group (30.9%). Moreover, many women in the two heavier groups never achieved active labor with prostaglandin alone; thus, augmentation with oxytocin was relatively less frequent with increasing BMI. Regardless of the indication, significant differences in duration and total amount of predelivery oxytocin administered were observed between the BMI categories (Table 3). Extremely obese participants received significantly higher amount of oxytocin (5.0 [0.4–20.9] units) than either obese women, who received 3.5 (0.3–15.8) units, or lean participants, who received 2.6 (0.4–13.5) units (P<.001). Furthermore, the median duration of predelivery oxytocin in the BMI 40 or higher group was 8.5 (2.1–22.6) hours, compared with 7.7 (1.9–18.6) hours in the BMI 30–39.9 group and 6.5 (1.9–11.4) hours in the BMI less than 30 group (P=.008).
In addition to the increased requirements for oxytocin, maternal obesity was associated with a significant increase in duration of labor. Specifically, lean women had a significantly shorter duration of active labor (14.9 [6.4–32.0] hours) than either the obese (16.0 [6.8–33.0] hours) or the extremely obese (19.3 [7.2–38.5] hours) women (P=.001). The difference persisted even after confining the analysis to vaginal delivery participants: 14.4 (6.4–29.7) hours for the BMI less than 30 group compared with 15.2 (7.1–32.6) hours for the BMI 30–39.9 group and 17.8 (7.1–33.7) hours for the BMI 40 or higher group (P=.03). Furthermore, the median time from insertion of the study drug to delivery was significantly longer in the obese and extremely obese women compared with the lean study participants (P<.001). Overall, women in the lowest BMI category delivered 2 hours faster than women in the BMI 30–39.9 group and more than 4 hours earlier than women in the BMI 40 or higher group. Analysis of vaginally delivered participants again proved to be statistically different, with mean delivery times of 21.9 (11.1–37.2) hours for lean participants, 23.0 (11.6–40.4) hours for obese participants, and 24.3 (12.1–47.0) hours for extremely obese participants (P=.03). Overall, more than 90% of women achieved active labor during the course of their inductions. Nevertheless, a higher percentage of women in the lowest BMI category reached active labor (96.4%), compared with their counterparts in the obese (92.1%) and extremely obese (91.5%) groups (OR 0.43, 95% CI 0.24–0.78, P=.004 and OR 0.40, 95% CI 0.20–0.81, P=.009, respectively).
The overall rate of cesarean delivery in the original Misoprostol Vaginal Insert Trial was approximately 28%. While 21.3% of participants in BMI less than 30 group required a cesarean delivery, the incidence increased to 29.8% in the BMI 30–39.9 group (OR 1.57, 95% CI 1.18–2.10, P=.002) and 36.5% in the BMI 40 or higher group (OR 2.12, 95% CI 1.47–3.06, P<.001). This difference was amplified in the analysis of proportion of cesarean deliveries performed in the first stage of labor. Nearly 62% of cesarean deliveries occurred in the first stage of labor in the BMI less than 30 group, compared with 74% in the BMI 30–39.9 group (OR 1.76, 95% CI 1.03–3.0, P=.04) and 83% in the BMI 40 or higher group (OR 3.04, 95% CI 1.46–6.34, P=.002). Indications for cesarean delivery were varied and generally similar between the BMI categories (Table 3). There was a marginally increased likelihood of cesarean deliveries performed for failed induction and failure to progress in the higher BMI categories. Multivariate analysis demonstrated that certain obstetric reasons for induction were predictive of increased cesarean delivery rate. This was not true for medical indications for induction such as diabetes or hypertensive disorders of pregnancy, perhaps due to small samples in each of these categories. In fact, only elective induction demonstrated an independent correlation with rates of cesarean delivery. Overall, the associations between maternal obesity and increased likelihood of cesarean delivery remained significant.
The average fetal birth weight for the entire study population (1,273 participants) was 3,376 g; however significant differences were noted across the BMI categories despite similar gestational ages at delivery. The mean fetal birth weight in the BMI 40 or higher group (3,489±515) was approximately 100 g greater than in the BMI 30–39.9 group (3,399±463) and 200 g more than neonates in the BMI less than 30 group (3,286±456) (P<.001). Although differences in maternal and neonatal duration of hospital stay were also statistically significant, these were minor and clinically irrelevant.
Multiple logistic regression analysis was performed adjusting for race, parity, and treatment group allocation to estimate the true effect of maternal obesity on the aforementioned outcomes. Although parity proved to be the most important variable, maternal obesity remained a significant and independent predictor of oxytocin requirements, duration of labor, and mode of delivery (Table 2).
The current analysis supports the findings of previous studies in regard to adverse effects of maternal obesity on labor characteristics and outcomes. Numerous investigators have demonstrated a proportional increase in rates of cesarean delivery corresponding to the level of maternal obesity.3,8–11 Some have attributed this finding to a higher likelihood of pregnancy-related complications in obese women and a subsequent increase in labor inductions.12 However, other investigators have suggested that the increased levels of soft tissue in the maternal pelvis may be the underlying reason for higher cesarean delivery rates.9,13 The current analysis demonstrated a difference in indications for labor induction between the BMI categories, particularly in regard to conditions generally associated with obesity, such as hypertensive disorders of pregnancy, diabetes, and suspected macrosomia. However, the prolonged duration of active and total labor in obese women is difficult to attribute entirely to the increased incidence of pregnancy-related complications. Several prior reports have found a similar association between maternal obesity and prolonged duration of labor.2,11,14 These findings, along with the increased requirements of predelivery oxytocin, may be explained by the pharmacokinetic and pharmacodynamic properties of drugs. The relative increase in volume of distribution found in obese women has a dilutional effect on both the ripening agent and oxytocin during the course of labor induction. This could potentially result in a reduced tissue response and a subsequent need for increased doses and duration of drug administration.
There are limitations to the current analysis that should be acknowledged. The variability in obstetric practices, physician behavior, and baseline cesarean delivery rates among the sites could have influenced the results of the current analysis. However, these differences could not be addressed in either the original protocol or this secondary analysis in a practicable manner. Another limitation of the current analysis is in the lack of consensus regarding BMI classifications in pregnancy. In fact, the World Health Organization, Institute of Medicine, and National Heart, Lung, and Blood Institute each use different values in categorizing BMI.7,15 Furthermore, these organizations acknowledge the limitation and need for further research in BMI in the setting of pregnancy.16 The relatively rapid weight gain in the short interval of pregnancy can have a significant effect on BMI at the time of delivery and can easily result in a woman being classified as a higher (ie, obese) category compared with the prepregnancy BMI. Nevertheless, numerous investigations have demonstrated a significant effect of maternal obesity on obstetric outcomes regardless of when the BMI is calculated. Additional information, including prepregnancy BMI and maternal weight gain during pregnancy, would have been helpful; however, these data were not collected during the original study. This limitation of the current analysis is common to all retrospective studies, because only the information gathered in the course of the original trial is available for subsequent analysis.
The strength of this investigation is that the data were collected from 49 diverse sites in one of the largest prospective, double-masked trials to date on prostaglandin induction of labor using a standardized protocol. Furthermore, the current analysis contains one of the largest cohorts of obese and extremely obese women and offers convincing evidence of the effects of maternal obesity on the duration, characteristics, and outcomes of labor. Last, a review of baseline demographics reveals heterogeneous maternal and pregnancy characteristics increasing the external validity of this analysis.
In conclusion, the degree of maternal obesity is a major factor in predicting outcomes after prostaglandin cervical ripening and labor induction. In addition to increasing the likelihood of cesarean delivery, maternal obesity is inversely associated with oxytocin requirements and progress and duration of labor. These factors need to be borne in mind whenever labor induction is proposed for an obese woman, particularly a nulliparous one. Because there is no end in sight to the cesarean epidemic, physicians should be cautious in their management decisions, which could contribute to an increased likelihood of adverse perinatal outcomes.
1. Ogden CL, Carroll MD, Curtin LR, McDowell MA, Tabak CJ, Flegal KM. Prevalence of overweight and obesity in the United States, 1999–2004. JAMA 2006;295:1549–55.
2. Vahratian A, Zhang J, Troendle JF, Savitz DA, Siega-Riz AM. Maternal prepregnancy overweight and obesity and the pattern of labor progression in term nulliparous women. Obstet Gynecol 2004;104(pt 1):943–51.
3. Weiss JL, Malone FD, Emig D, Ball RH, Nyberg DA, Comstosk CH, et al; FASTER Research Consortium. Obesity, obstetric complications and cesarean delivery rate: a population-based screening study. Am J Obstet Gynecol 2004;190:1091–7.
4. Martin JA, Hamilton BE, Sutton PD, Ventura SJ, Menacker F, Kirmeyer S, et al. Births: final data for 2006. Natl Vital Stat Rep 2009;57:1–102.
5. Rayburn WF, Zhang J. Rising rates of labor induction: present concerns and future strategies. Obstet Gynecol 2002;100:164–7.
6. Wing DA; Misoprostol Vaginal Insert Consortium. Misoprostol vaginal insert compared with dinoprostone vaginal insert: a randomized control trial. Obstet Gynecol 2008;112:801–12.
7. World Health Organization. BMI classification. Available at: http://www.who.int/bmi/index.jsp?introPage=intro_3.html
. Retrieved May 14, 2009.
8. Johnson JW, Longmate JA, Frentzen B. Excessive maternal weight and pregnancy outcome. Am J Obstet Gynecol 1992;167:353–70.
9. Kaiser PS, Kirby RS. Obesity as a risk factor for cesarean in a low-risk population. Obstet Gynecol 2001;97:39–43.
10. Brost BC, Goldenberg RL, Mercer BM, Iams JD, Meis PJ, Moawad AH, et al. The Preterm Prediction Study: association of cesarean delivery with increases in maternal weight and body mass index. Am J Obstet Gynecol 1997;177:333–7.
11. Nuthalapaty FS, Rouse DJ, Owen J. The association of maternal weight with cesarean risk, labor duration, and cervical dilatation rate during labor induction [published erratum appears in Obstet Gynecol 2004;103(pt 1):1019]. Obstet Gynecol 2004;103:452–6.
12. Cnattingius R, Cnattingius S, Notzon FC. Obstacles to reducing cesarean rates in a low-cesarean setting: the effect of maternal age, height, and weight. Obstet Gynecol 1998;92(pt 1):501–6.
13. Crane SS, Wojtowycz MA, Dye TD, Aubry RH, Artal R. Association between pre-pregnancy obesity and the risk of cesarean delivery. Obstet Gynecol 1997;89:213–6.
14. Jensen H, Agger AO, Rasmussen KL. The influence of prepregnancy body mass index on labor complications. Acta Obstet Gynecol Scand 1999;78:799–802.
15. Institute of Medicine. Influence of pregnancy weight on maternal and child health: workshop report. Washington, DC: National Academies Press; 1990.
16. World Health Organization. Physical status: the use and interpretation of anthropometry. Report of a WHO Expert Committee. WHO Technical Report Series 854. Geneva: World Health Organization; 1995.