Objective
The purpose of this study was to determine whether maternal obesity is associated with cesarean delivery and decreased contraction strength in the first stage of labor.
Study Design
We studied a retrospective cohort of women who delivered within a single healthcare system from 2007-2009; we included 5410 women with an intrauterine pressure catheter during the last 2 hours of the first stage of labor and who either had a vaginal delivery or cesarean delivery for dystocia. Logistic regression was used to determine how body mass index was associated with cesarean delivery or mean Montevideo units of ≥200.
Results
Although obese women were at significantly greater odds of cesarean delivery than normal-weight women (odds ratio, 2.4; 95% confidence interval, 1.9–3.1), they were equally able to achieve Montevideo units of ≥200. Among women with a vaginal delivery, obese women had a longer first stage of labor compared with normal-weight women (597 vs 566 min; P = .003).
Conclusion
Obese women have longer labors but are equally able to achieve adequate Montevideo units as normal-weight women.
Obesity is an epidemic in many developed countries. More than one-third of all women in the United States are obese (body mass index [BMI], ≥30 kg/m 2 ), with approximately 1-in-5 being obese at the onset of pregnancy. Maternal obesity is associated with multiple labor abnormalities that include increased risks for induction of labor, postdates pregnancy, prolonged labor, oxytocin augmentation, postpartum hemorrhage, and cesarean delivery (CD). There is a dose-dependent relationship between maternal BMI and a risk for CD that is independent of parity and infant birthweight. A recent study found that, for every 1 unit increase in BMI, the risk for CD increased by 5% for both nulliparous and multiparous women without a previous CD. The increased rate of CD among obese women appears to be confined to the first stage of labor and is, in large part, due to failure to progress or cephalopelvic disproportion. Furthermore, progression in the first stage of labor is particularly slow among obese women.
Given that the initiation and maintenance of normal labor is complex (and incompletely understood), the cause of the increased CD rate among obese women is likely multifactorial. However, given the multiple labor abnormalities that are seen more frequently among obese women, a leading theory is that obesity may be associated with impaired myometrial contractility. This theory is supported by multiple ex vivo and animal studies. Zhang et al found that myometrial strips obtained from obese women who undergo an elective CD at term contracted with less force and frequency than strips that were obtained from normal-weight women; however, a subsequent study was unable to confirm these findings. In 2 other studies, leptin and low-density lipoprotein cholesterol were found to inhibit the contractility of human myometrial strips. In a recent study that used a rat model, a high-fat, high-cholesterol diet resulted in decreased expression of certain myometrial contractile-associated proteins.
Although these ex vivo and animal studies indicate that myometrial contractility may indeed be impaired in the setting of maternal obesity, data as to whether these results translate into clinically measurable or meaningful differences in contractile strength during human labor are limited. Buhimschi et al found that obese women in the second stage of labor generate equivalent intrauterine pressures with pushing compared with normal-weight women. Nuthalapaty et al examined whether uterine contraction strength in the first stage of labor varies with BMI. Although maternal BMI was related inversely to the rate of cervical change in the first stage of labor, neither uterine responsiveness (as measured in Montevideo units [MVUs]) nor rates of oxytocin administration differed by BMI.
Using a large and detailed electronic obstetric database from a single integrated healthcare system, we sought to determine whether maternal obesity is associated with (1) an increased risk for CD in the first stage of labor and (2) decreased uterine contractility in the first stage of labor, as measured by intrauterine pressure catheters (IUPCs) and MVUs.
Materials and Methods
Study population
University of Utah and Intermountain Healthcare institutional review board approval was obtained to access electronic data on women who delivered from 2007-2009 at 10 hospitals within Intermountain Healthcare (a single vertically integrated health care system). Electronic data were from an Intermountain Healthcare-specific database (StorkBytes) that captures maternal history, labor progression (fetal and uterine tracings, cervical examinations, oxytocin doses, and nurse charting), and delivery data.
Only women who had a term (≥37 weeks’ gestation) vaginal delivery (VD), or a CD in the first stage of labor (<10 cm dilation) and who had an IUPC for at least 2 hours before complete dilation or CD were included. Because we sought to focus on contraction strength in the first stage of labor (at which time progression is particularly slow among obese women) and because maternal pushing efforts might make the calculation of MVUs inaccurate, women who had a CD in the second stage of labor were excluded. Women with a multiple gestation, previous CD, major fetal anomaly, diagnosis of chorioamnionitis, or diagnosis of preeclampsia that required magnesium sulfate treatment were excluded. To eliminate potential coding errors, women with gestational weight gain (GWG) recorded as <0 or >75 lbs were excluded; women with a gestational age at delivery of ≥43 weeks were excluded, and women with a recorded time in labor that was a negative value were excluded. Women with a listed CD indication that included failed induction/augmentation, failure to progress, arrest of dilation/descent, or cephalopelvic disproportion were included in a category that was designated as “dystocia.” Some women had >1 indication listed for their CD (for example, arrest of dilation and non-reassuring tracing). If one of the indications was dystocia as defined earlier, then they were placed in the dystocia category for the purposes of analysis. Women with no recorded indication for CD and women with a listed indication of cord prolapse, malpresentation, compound presentation, or marginal previa were excluded.
Measures
The primary predictor was prepregnancy BMI category that was defined in the following manner: normal weight (BMI, ≤25 kg/m 2 ), overweight (BMI, 26-30 kg/m 2 ), or obese (BMI, ≥30 kg/m 2 ). Prepregnancy BMI was calculated on the basis of self-reported prepregnancy weight and height at the time of admission to Labor and Delivery. The primary outcome variables were CD and mean MVUs of ≥200 (typically considered adequate in clinical practice) in the last 2 hours of the first stage of labor (2 hours preceding complete dilation for women with a VD and 2 hours preceding delivery for women with a CD). When prompted, a computer algorithm calculates MVUs over a 10-minute period. These MVU values are then recorded electronically by nurses. Variables that were evaluated as potential confounders of the association between prepregnancy BMI and CD were maternal age, parity, induction of labor, gestational age, admission cervical dilation, excessive GWG as defined by Institute of Medicine 2009 criteria, and infant birthweight. In addition to these variables, maximum oxytocin dosage in the last 2 hours of the first stage of labor was included as a potential confounder when we evaluated the association between BMI and mean MVUs.
Statistical analysis
One-way analysis of variance or the nonparametric Kruskal-Wallis test was used in bivariate analyses. To evaluate for potential colinearity among predictors, correlation coefficients were calculated. Multivariable logistic regression was used first to determine how prepregnancy BMI category was associated with CD. Logistic regression was then used to determine how prepregnancy BMI category was associated with mean MVUs of ≥200 in the last 2 hours of the first stage of labor. Because women who ultimately had a CD may have had a more dysfunctional labor course than those with a VD and because the phase of labor during which MVUs were recorded would differ for women with a CD vs VD, this analysis was stratified by mode of delivery. We hypothesized that women whose only indication for CD was dystocia (ie, not also including nonreassuring tracing) represent a group for which clinicians would be particularly aggressive in trying to achieve adequate MVUs. For this reason, we performed a subanalysis of the association of BMI with MVUs among this group.
The logistic regression models are summarized by odds ratios (ORs) with corresponding 95% confidence intervals (CIs). All statistical tests used a 2-sided alpha of .05. Analysis was performed with SAS software (version 9.2; SAS Institute Inc. Cary, NC).
Results
From a total of 69,857 women who delivered at these 10 hospitals during the study period, 7611 women were identified who had a term singleton gestation that resulted in a VD or primary CD in the first stage of labor and who had an IUPC during the last 2 hours of the first stage of labor. Three women with no listed indication for CD, 6 women who had a listed indication of cord prolapse, malpresentation, compound presentation, or marginal previa, and 443 women who had a listed indication of a major fetal anomaly, chorioamnionitis, or preeclampsia that required magnesium sulfate treatment were excluded. After additional exclusions for potential coding errors (as described earlier), a total of 5410 women were included in the analysis (VD, 4772; CD in the first stage of labor, 638). Most of the women (n = 509; 79.8%) who had a CD had a recorded indication classified in the dystocia category. Four hundred ten women who had a CD had only a dystocia indication. Two thousand eight hundred eighteen of the women (52.1%) were normal weight; 1398 of the women (25.8%) were overweight, and 1194 of the women (22.1%) were obese. As compared with normal-weight women, obese women were significantly older and significantly more likely to be parous, to undergo an induction of labor, to have larger infants, and to have excessive GWG ( Table 1 ). Admission cervical dilation did not differ by BMI category.
| Characteristic | Normal weight | Overweight | Obese | P value |
|---|---|---|---|---|
| Age, y a | 25.8 ± 5.0 | 26.6 ± 5.2 | 27.3 ± 5.3 | < .0001 |
| Nulliparous, n (%) | 1620 (57.5) | 707 (50.6) | 535 (44.8) | < .0001 |
| Parous, n (%) | 1198 (42.5) | 691 (49.4) | 659 (55.2) | |
| Induction of labor, n (%) | 1444 (51.2) | 824 (58.8) | 735 (61.6) | < .0001 |
| Spontaneous labor, n (%) | 1374 (48.8) | 574 (41.1) | 459 (38.4) | |
| Infant birthweight, g a | 3359.4 ± 416.3 | 3430.1 ± 429.4 | 3443.6 ± 425.2 | < .0001 |
| Excessive gestational weight gain, n (%) | 1153 (40.9) | 952 (68.1) | 719 (60.2) | < .0001 |
| Admission dilation (per 1 cm) a | 2.4 ± 1.3 | 2.4 ± 1.3 | 2.3 ± 1.3 | .44 |
| Gestational age, wk a | 39.1 ± 1.1 | 39.1 ± 1.1 | 39.0 ± 1.1 | .08 |
| Cesarean delivery, n (%) | 277 (9.8) | 171 (12.2) | 190 (15.9) | < .001 |
After adjustment for maternal age, gestational age, birthweight, excessive GWG, admission dilation, induction of labor, and parity, overweight and obese women were at significantly increased odds of having a CD in the first stage of labor compared with normal-weight women (overweight women: odds ratio, 1.3 (95% CI, 1.0–1.6); obese women: odds ratio, 2.4 (95% CI, 1.9–3.1).
The maximum dose of oxytocin did not differ by BMI category either for women who had a CD or for women who had a VD. There was a mean of 1.9 (SD, ± 1.1) MVU values recorded per woman, which did not vary by BMI category. Among women who had a CD, 281 women (44%) had mean MVUs of ≥200 in the 2 hours preceding delivery. Among women who had a VD, 2287 women (47.9%) had mean MVUs of ≥200 in the last 2 hours of the first stage of labor. The results of the final logistic regression models that evaluated the association between BMI category and MVUs are detailed in Table 2 . Among women who had a CD, the odds of mean MVUs of ≥200 were actually greater for obese women compared with normal-weight women (OR, 1.76; 95% CI, 1.11–2.81). Among women who had a VD, there was no significant association between BMI category and mean MVUs of ≥200. In both models, each 5 mU/min increase in maximum oxytocin dose was associated with decreased odds of mean MVUs of ≥200.
| Predictor | Cesarean delivery, a OR (95% CI) | Vaginal delivery, a OR (95% CI) |
|---|---|---|
| Body mass index category | ||
| Obese vs normal weight | 1.76 (1.11–2.81) | 1.16 (0.98–1.37) |
| Overweight vs normal weight | 1.12 (0.73–1.73) | 1.08 (0.93–1.26) |
| Maximum oxytocin dose (per 5 mU/min) | 0.79 (0.69–0.90) | 0.87 (0.84–0.90) |
| Excessive gestational weight gain (yes vs no) | 1.31 (0.84–2.06) | 1.03 (0.89–1.20) |
| Maternal age (per 5 y) | 0.87 (0.74–1.03) | 0.79 (0.74–0.84) |
| Gestational age (per 1 wk) | 0.96 (0.82–1.11) | 1.00 (0.94–1.07) |
| Infant birthweight (per 100 g) | 0.98 (0.94–1.01) | 0.96 (0.95–0.98) |
| Parity (parous vs nulliparous) | 1.10 (0.66–1.81) | 1.07 (0.93–1.23) |
| Induction of labor (spontaneous vs induction) | 0.85 (0.59–1.21) | 0.67 (0.59–0.77) |
| Admission dilation (per 1 cm) | 1.11 (0.98–1.27) | 0.86 (0.82–0.90) |
a Deliveries that were included in the final models because of missing data for some predictor variables. Cesarean delivery, 623 women; vaginal delivery, 4675 women.
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
