Cervicovaginal microbiome in twin vs singleton gestations.

BACKGROUND: The vaginal microbiome diversity profile varies by race and ethnicity and changes considerably from the nonpregnant state to the pregnant state, specifically with a shift to Lactobacillus predominance in singleton gestations. There is a paucity of data that evaluate the cervicovaginal microbiome in women with twin gestations as a distinct population from those with singleton gestations. OBJECTIVE: We sought to characterize the cervicovaginal microbiome diversity profiles among twin gestations in the second trimester of pregnancy. STUDY DESIGN: In this prospective cross-sectional cohort study, women with twin gestations were matched to singleton controls without a history of a short cervix or preterm birth by gestational age ±2 weeks and race. Cervicovaginal lavage samples were collected from 14 to 24 weeks of gestation during prenatal visits followed by a cervical length measurement. Cervicovaginal microbiota were analyzed with 16S RNA gene sequencing and classified into community state types based on Lactobacillus species predominance. Microbiome alpha and beta diversities were compared between twin and singleton gestations. RESULTS: A total of 19 twin gestations and 19 singleton gestations underwent second-trimester cervicovaginal microbiome analysis. The groups were similar in gestational age at sample collection, maternal age, parity, body mass index, preterm birth history, and comorbidity. The cohort was predominantly of Black race (79%). Of twin gestations, 79% were dichorionic and diamniotic and 21% monochorionic and diamniotic. Of note, 3 twin gestations and 1 singleton gestation were complicated by a short cervix (P=.6). The vaginal microbiome of twin gestations had decreased alpha and beta diversities compared with singleton gestations. Twin gestations had lower taxon abundance and decreased variability in taxon abundance than singleton gestations. Overall, there was decreased diversity of community state type groups among twin gestations compared with singleton gestations. Community state types I and III were more prevalent among twin gestations, whereas community state types II and IV were similar among these 2 groups. Community state type IV, which is defined by a lack of Lactobacillus species and the presence of diverse strict anaerobes, was the predominant type among microbiota profiles of twin gestations (55%) and singleton gestations (64%). Community state type V was more prevalent in singleton gestations. When stratified by race, we found similar alpha diversity in Black and non-Black patients with twin gestations. CONCLUSION: In our predominantly Black population of pregnant women, the second-trimester vaginal microbiome in twin gestations showed decreased alpha and beta diversities compared with singleton controls. Our findings increased the understanding of the content of microbial communities in the second trimester of pregnancy in twin gestations and suggested a potential mechanism for preterm birth in twin gestations.

Pulse pressure as a predictor of response to treatment for severe hypertension in pregnancy.

BACKGROUND: Pulse pressure is a proposed means of tailoring antihypertensive therapy for treatment of acute-onset, severe hypertension in pregnancy. OBJECTIVE: This study aimed to determine whether pulse pressure predicts response to the various first-line antihypertensive medications. STUDY DESIGN: This is a retrospective cohort study from a single academic tertiary care center between 2015 and 2018. Patients were screened for inclusion if they had severe hypertension (defined as systolic blood pressure of ≥160 mm Hg or diastolic blood pressure of ≥110 mm Hg) lasting at least 15 minutes and were initially treated with intravenous labetalol, intravenous hydralazine, or immediate-release oral nifedipine. If a patient had multiple episodes of acute treatment during the pregnancy, only one episode was included in the analysis. The primary outcome was time to resolution (in minutes) of severe hypertension. To adjust for factors that may have affected time to resolution, we first compared baseline characteristics on the basis of the antihypertensive agent received. We then assessed the association between baseline characteristics and resolution of severe hypertension within 60 minutes of treatment. Regression analysis incorporated pulse pressure and antihypertensive agents into a model to predict resolution within 60 minutes of onset of severe hypertension. RESULTS: A total of 479 women hospitalized with severe maternal hypertension met the inclusion criteria. Hydralazine was the initial antihypertensive agent administered to 113 women, whereas 233 received labetalol, and 133 received nifedipine. Those who initially received nifedipine had a shorter mean time to resolution of severe hypertension (32.6 minutes vs 46.3 for hydralazine and 50.3 for labetalol; P<.01) and were more likely to have resolution of severe hypertension within 60 minutes (91.0% vs 77.9% for hydralazine and 76.8% for labetalol; P<.01). Nifedipine also resulted in a lower mean posttreatment blood pressure. Regression analysis revealed that a lack of resolution of severe hypertension within 60 minutes was independently associated with 2 measures of hypertension severity (mean arterial pressure of ≥125 mm Hg and the need for ≥2 doses of medication) and pulse pressure of >75 mm Hg at the time of treatment, initial treatment with labetalol, and gestational age of <37 weeks at the time of the hypertensive event (or at delivery if treatment was after delivery). The model's bias-corrected bootstrapped area under the receiver operating characteristic curve was 0.85 (95% confidence interval, 0.79-0.88). Interaction terms between pulse pressure and each antihypertensive agent were not significant and therefore not incorporated into the final model. CONCLUSION: Pulse pressure did not predict response to the various first-line antihypertensive agents. Initial treatment with oral nifedipine was associated with a higher likelihood of resolution of severe hypertension within 60 minutes of treatment than with intravenous labetalol.

Reduction of opioid utilization after vaginal delivery with an electronic health record intervention.

BACKGROUND: The national epidemic of opioid misuse has focused its attention on postpartum analgesic usage. Adequate pain control achieved with nonopioid pain medications and conservative measures could reduce the number of opioid pain medications available for misuse and diversion. Interventions that decrease inpatient opioid use after delivery could reduce the potential for chronic dependence in postpartum women. Modification of preloaded electronic order sets to decrease opioid administration has successfully reduced opioid use following a major abdominal surgery, including cesarean delivery. However, interventions to reduce opioid use following vaginal delivery are not well described. OBJECTIVE: We aimed to evaluate the effect of removing opioid medications from postpartum order sets on medication usage following vaginal delivery. STUDY DESIGN: We performed a retrospective cohort study of women undergoing a singleton vaginal delivery at an academic tertiary care center. Our institution removed opioid medications from postpartum order sets in April 2018. We compared the following 2 delivery cohorts: the "preintervention" cohort (April 2016-March 2018) and the "postintervention" cohort (June 2018-July 2018). The primary outcome was postpartum opioid use. The secondary outcomes were nonopioid analgesic use and discharge with an opioid prescription. We compared the demographic and obstetrical data, self-reported pain scores, and postpartum analgesic usage between groups. We determined that a minimum of 138 patients would be needed in each group to identify a 20% decrease in opioid usage (α=.05; ß=.2). RESULTS: We analyzed 276 subjects: 138 in the preintervention group and 138 in the postintervention group. The postintervention group was older and more likely to have an operative vaginal delivery. Otherwise, groups had similar demographic and obstetrical characteristics. Postpartum opioid use decreased from 56% in the preintervention group to 16% in the postintervention group, a 71% reduction (P<.001). The incidence of severe pain score (>7) was similar between groups with a median occurrence of 1 (interquartile range, 0-4) for both (P=.7). The number of opioid discharge prescriptions among those receiving inpatient opioids was significantly lower in the postintervention group than in the preintervention group (18% vs 38%, respectively), a 53% decrease (P<.001). CONCLUSION: Removal of opioids from the postpartum order set was associated with lower rates of opiate usage following vaginal delivery in a single center without changing the frequency of severe pain scores. This simple intervention has the potential to reduce opioid exposure.

Society for Maternal-Fetal Medicine Special Statement: Maternal-fetal medicine subspecialist survey on abortion training and service provision.

Following a collaborative workshop at the 39th Annual Pregnancy Meeting, the Society for Maternal-Fetal Medicine Reproductive Health Advisory Group identified a need to assess the attitudes of maternal-fetal medicine subspecialists about abortion services and the available resources at the local and regional levels. The purpose of this study was to identify trends in attitudes, beliefs, and behaviors of practicing maternal-fetal medicine subspecialists in the United States regarding abortion. An online survey was distributed to associate and regular members of the Society for Maternal-Fetal Medicine to assess their personal training experience, abortion practice patterns, factors that influence their decision to provide abortion care, and their responses to a series of scenarios about high-risk maternal or fetal medical conditions. Frequencies were analyzed and univariable and multivariable analyses were conducted on the survey responses. Of the 2751 members contacted, 546 Society for Maternal-Fetal Medicine members completed all (448 of 546, 82.1%) or some (98 of 546, 17.9%) of the survey. More than 80% of the respondents reported availability of abortion services in their state, 70% reported availability at their primary institution, and 44% reported provision as part of their personal medical practice. Ease of referral to family planning subspecialists or other abortion providers, institutional restrictions, and the lack of training or continuing education were identified as the most significant factors contributing to the respondents' limited scope of abortion services or lack of any abortion services offered. In the univariable analysis, exposure to formal family planning training programs, fewer years since the completion of residency, current practice setting not being religiously affiliated, and current state categorized as supportive by the Guttmacher Institute's abortion policy landscape were factors associated with abortion provision (all P values <.01). After controlling for these factors in a multivariable regression, exposure to formal family planning training programs was no longer associated with current abortion provision (P=.20; adjusted odds ratio, 1.34; 95% confidence interval, 0.85-2.10), whereas a favorable state policy environment and fewer years since the completion of residency remained associated with abortion provision. The results of this survey suggest that factors at the individual, institutional, and state levels affect the provision of abortion care by maternal-fetal medicine subspecialists. The subspecialty of maternal-fetal medicine should be active in ensuring adequate training and education to create a community of maternal-fetal medicine physicians able to provide comprehensive reproductive healthcare services.

Prospective, Blinded Evaluation of Template-Based Cesarean Documentation Error in an Obstetric Training Program.

OBJECTIVE: Cesarean operative report accuracy impacts postoperative care and future obstetric decision-making. The impact of electronic health record template use on cesarean documentation error remains unknown. The aim of the study was to describe the incidence of resident physician documentation error in cesarean operative reports using electronic health record templates. METHODS: Attending physicians completed a standardized audit form after cesarean deliveries, which was compared with the resident operative report. Resident physicians were blinded to the auditing process. Errors were classified as none, major, or minor using predefined definitions. Author and operative characteristics were collected for comparative and predictive analyses. Data were analyzed by presence or absence of error. RESULTS: We reviewed 100 cesarean operative reports. Major and minor errors were encountered in 33% and 53% of operative notes, respectively. Advancing training level was associated with lower incidence of major error (50%, for postgraduate year [PGY] 1, 33% for PGY 2, and 0% for PGY 3/4, P = 0.02), but minor errors were similar among training level, P = 0.48. Operative duration, documentation interval, and shift characteristics were similar in cases with and without documentation errors. In multiple logistic regression, PGY was predictive of major documentation error (adjusted odds ratio = 0.39, 95% confidence interval = 0.17-0.92). CONCLUSIONS: A high incidence of clinically significant documentation error was observed in a residency training program using standardized templates for cesarean operative reports. Incidence of major error decreased with increasing training, but minor error was similar across levels of training. These data suggest that measures should be taken to improve documentation accuracy in medical training to provide optimal obstetric care.

Pregnancy duration with use of 17-α-hydroxyprogesterone caproate in a retrospective cohort at high risk of recurrent preterm birth.

BACKGROUND: The use of 17-α-hydroxyprogesterone caproate for the prevention of recurrent spontaneous preterm birth has become widespread, yet there are conflicting data regarding its efficacy. OBJECTIVE: We sought to determine whether administration of 17-α-hydroxyprogesterone caproate was associated with pregnancy prolongation in women at a high risk of recurrent spontaneous preterm birth. STUDY DESIGN: This is a retrospective cohort study of women with singleton pregnancies and a history of spontaneous preterm birth at <37 weeks' gestation who received care at our academic tertiary care center between 2009 and 2019. We included women with gestations that progressed beyond 16 weeks. We excluded those who underwent history-indicated cerclage placement. We first examined the characteristics of women who received 17-α-hydroxyprogesterone caproate and those who did not. Covariates with a P value of ≤.2 on this univariate analysis were considered for incorporation into a Cox proportional hazards model to assess the association between 17-α-hydroxyprogesterone caproate use and pregnancy prolongation up to 35 weeks. RESULTS: Of 861 women included in the study, 570 (66.2%) reported non-Hispanic black racial identity, 237 (27.5%) lived in zip codes with a high infant mortality rate (≥12.1/1000 infants), 287 (33.3%) had more than 1 previous spontaneous preterm birth, 372 (43.2%) had previous spontaneous preterm birth at ≤32 weeks' gestation, and 242 (28.1%) were smokers. Here, 152 pregnancies (17.6%) were complicated by spontaneous preterm birth at <35 weeks' gestation. Factors independently associated with pregnancy duration up to 35 weeks included weight gain of <0.2 kg (0.5 lb) per week, first recorded weight of <98 kg (215 lb), obstetrical history, non-Hispanic white racial identity, lack of prenatal care, and vaginal bleeding. Gestational age at delivery was also independently associated with interventions typically employed for midtrimester cervical shortening and/or dilation, including ultrasound- and examination-indicated cerclage, pessary placement, and vaginal progesterone administration. The use of 17-α-hydroxyprogesterone caproate was not associated with pregnancy prolongation (adjusted hazard ratio, 0.83; 95% confidence interval, 0.60-1.15). CONCLUSION: The risk profile of our cohort is similar to that of women enrolled in the landmark trial that led to the Food and Drug Administration's approval of 17-α-hydroxyprogesterone caproate. Despite the high-risk nature of the pregnancies examined, we found no association between use of the medication in daily clinical practice and pregnancy prolongation up to 35 weeks. This finding adds to the mounting evidence that calls into question the drug's efficacy in reducing the risk of recurrent spontaneous preterm birth.

Society for Maternal-Fetal Medicine (SMFM) Consult Series #48: Immediate postpartum long-acting reversible contraception for women at high risk for medical complications.

Reproductive planning is essential for all women and most important for those with complex health conditions or at high risk for complications. Pregnancy planning can allow these high-risk women the opportunity to receive preconception counseling, medication adjustment, and risk assessment related to health conditions that have a direct impact on maternal morbidity and mortality risk. Despite the need for pregnancy planning, medically complex women face barriers to contraceptive use, including systemic barriers, such as underinsurance for women at increased risk for complex medical conditions as well as low uptake of effective postpartum contraception. Providing contraceptive counseling and a full range of contraceptive options, including immediate postpartum long-acting reversible contraception (LARC), is a means of overcoming these barriers. The purpose of this document is to educate all providers, including maternal-fetal medicine subspecialists, about the benefits of postpartum contraception, and to advocate for widespread implementation of immediate postpartum LARC placement programs. The following are Society for Maternal-Fetal Medicine recommendations: we recommend that LARC be offered to women at highest risk for adverse health events as a result of a future pregnancy (GRADE 1B); we recommend that obstetric care providers discuss the availability of immediate postpartum LARC with all pregnant women during prenatal care and consult the U.S. Medical Eligibility Criteria for Contraceptive Use guidelines to determine methods most appropriate for specific medical conditions (GRADE 1C); we recommend that women considering immediate postpartum intrauterine device insertion be counseled that although expulsion rates are higher than with delayed insertion, the benefits appear to outweigh the risk of expulsion, as the long-term continuation rates are higher (GRADE 1C); we recommend that obstetric care providers wishing to utilize immediate postpartum LARC obtain training specific to the immediate postpartum period (BEST PRACTICE); for women who desire and are eligible for LARC, we recommend immediate postpartum placement after a high-risk pregnancy over delayed placement due to overall superior efficacy and cost-effectiveness (GRADE 1B); we recommend that women considering immediate postpartum LARC be encouraged to breastfeed, as current evidence suggests that these methods do not negatively influence lactation (GRADE 1B); for women who desire and are eligible for LARC, we suggest that early postpartum LARC placement be considered when immediate postpartum LARC placement is not feasible (GRADE 2C); and we recommend that contraceptive counseling programs be patient-centered and provided in a shared decision-making framework to avoid coercion (BEST PRACTICE).

Time from neuraxial anesthesia placement to delivery is inversely proportional to umbilical arterial cord pH at scheduled cesarean delivery.

BACKGROUND: Neuraxial block-related hypotension and maternal obesity contribute to uterine hypoperfusion and decreased umbilical arterial pH at cesarean delivery. Between the time of anesthesia placement and delivery, the fetus may be exposed to a hypoperfused uterine environment without surgeon awareness of fetal compromise. OBJECTIVE: We sought to evaluate neonatal umbilical arterial pH according to predelivery time intervals at scheduled term cesarean. STUDY DESIGN: We performed a retrospective cohort study of cesarean deliveries between September 2014 and February 2017. Singleton gestations undergoing scheduled cesarean delivery under spinal anesthesia between 37 and 41 weeks with a reassuring preoperative nonstress test were included. Time intervals between operative room entry, spinal anesthesia placement, skin incision, uterine incision, and delivery were calculated. The primary outcome was umbilical arterial pH. Demographic data, maternal blood pressures, predelivery time intervals, and delivery outcomes were analyzed according to umbilical arterial pH intervals of <7.0, 7.01-7.10, 7.11-7.20, 7.21-7.30, and >7.30. Umbilical cord gas analytes and neonatal outcomes were analyzed by spinal to delivery time. Stepwise linear regression was performed to identify predictors of decreasing umbilical arterial pH. Receiver-operator characteristic curves were calculated for spinal to delivery time and umbilical arterial pH <7.0 and 7.1. RESULTS: Among 527 included participants, median umbilical arterial pH was 7.27 [interquartile range, 7.23-7.29] and body mass index was 35 kg/m2 [interquartile range, 30-41]. Both maternal body mass index and hypotensive episodes increased with decreasing umbilical arterial pH (P <.001, P ≤ .02). All predelivery time intervals (operative room to delivery, spinal to skin, spinal to delivery, and uterine incision to delivery) increased as umbilical arterial pH interval decreased (P < .05 for all). In a stepwise linear regression, maternal body mass index, noncephalic presentation, spinal start to delivery interval, uterine incision to delivery interval, and maximum reduction in blood pressure from baseline were predictive of decreasing umbilical arterial pH after controlling for confounding variables (F [5,442] = 17.7, P = .0001], adjusted R2 of 0.157. When evaluated by spinal to delivery time, both umbilical arterial and venous pH and partial pressure of carbon dioxide decreased (P < .001 for all), but base deficit and neonatal outcomes were similar (P ≥ .7 for all). There were 2 cases of hypoxic-ischemic encephalopathy (0.38%). A receiver-operating characteristic curve demonstrated that a spinal start to delivery time greater than 27 minutes was associated with an umbilical arterial pH <7.1 (area under the curve, 0.74, 100% sensitivity, 21% specificity), and an interval greater than 30 minutes was associated with an umbilical arterial pH <7.0 (area under the curve, 0.80, 100% sensitivity, 33% specificity). CONCLUSION: Longer spinal-to-delivery and uterine incision-to-delivery time intervals were associated with decreasing umbilical arterial pH at scheduled term cesarean delivery. Efforts to minimize predelivery time following spinal placement could reduce the frequency of unanticipated neonatal acidemia.

Emerging understanding and measurement of plasma volume expansion in pregnancy.

Plasma volume expansion is an important component of a successful pregnancy. The failure of maternal plasma volume expansion has been implicated in adverse obstetric outcomes such as pre-eclampsia, fetal growth restriction, and preterm birth. Altered iron homeostasis and elevated maternal hemoglobin concentrations have also been associated with adverse pregnancy outcomes; limited data have suggested that these effects may be mediated by inadequate plasma volume expansion. In addition, it has been noted that pregnant, obese women, compared with lean subjects, have decreased plasma volume expansion along with impaired iron homeostasis and increased inflammation. Current estimates of plasma volume expansion are outdated and do not necessarily reflect contemporary obstetric populations. Moreover, the validation of clinically applicable methods of plasma volume determination as well as enhanced methodologies should be a priority. Further study is needed to characterize diminished plasma volume expansion during pregnancy and to understand the potential role of impaired iron homeostasis and inflammation in adverse obstetric outcomes, especially in obese women.

Effect of ambient light on the time needed to complete a fetal biophysical profile: A randomized controlled trial.

OBJECTIVE: The objective of this study is to determine whether ambient light serves as a fetal stimulus to decrease the amount of time needed to complete a biophysical profile. STUDY DESIGN: This is a randomized controlled trial of singleton gestations undergoing a biophysical profile. Patients were randomized to either ambient light or a darkened room. The primary outcome was the time needed to complete the biophysical profile. Secondary outcomes included total and individual component biophysical profile scores and scores less than 8. A subgroup analysis of different maternal body mass indices was also performed. RESULTS: 357 biophysical profile studies were analyzed. 182 studies were performed with ambient light and 175 were performed in a darkened room. There was no difference in the median time needed to complete the biophysical profile based on exposure to ambient light (6.1min in darkened room versus 6.6min with ambient light; P=0.73). No difference was found in total or individual component biophysical profile scores. Subgroup analysis by maternal body mass index did not demonstrate shorter study times with ambient light exposure in women who were normal weight, overweight or obese. CONCLUSION: Ambient light exposure did not decrease the time needed to complete the biophysical profile. There was no evidence that ambient light altered fetal behavior observed during the biophysical profile.

Blood volume determination in obese and normal-weight gravidas: the hydroxyethyl starch method.

OBJECTIVE: The impact of obesity on maternal blood volume in pregnancy has not been reported. We compared the blood volumes of obese and normal-weight gravidas using a validated hydroxyethyl starch (HES) dilution technique for blood volume estimation. STUDY DESIGN: Blood volumes were estimated in 30 normal-weight (pregravid body mass index [BMI] <25 kg/m(2)) and 30 obese (pregravid BMI >35 kg/m(2)) gravidas >34 weeks' gestation using a modified HES dilution technique. Blood samples obtained before and 10 minutes after HES injection were analyzed for plasma glucose concentrations after acid hydrolysis of HES. Blood volume was calculated from the difference between glucose concentrations measured in hydrolyzed plasma. RESULTS: Obese gravidas had higher pregravid and visit BMI (mean [SD]): pregravid (41 [4] vs 22 [2] kg/m(2), P = .001); visit (42 [4] vs 27 [2] kg/m(2), P = .001), but lower weight gain (5 [7] vs 12 [4] kg, P = .001) than normal-weight women. Obese gravidas had similar estimated total blood volume to normal-weight women (8103 ± 2452 vs 6944 ± 2830 mL, P = .1), but lower blood volume per kilogram weight (73 ± 22 vs 95 ± 30 mL/kg, P = .007). CONCLUSION: Obese gravidas have similar circulating blood volume, but lower blood volume per kilogram body weight, than normal-weight gravidas near term.

Epidural-associated hypotension is more common among severely preeclamptic patients in labor.

OBJECTIVE: To determine whether severe preeclampsia is associated with increased maternal hypotension or fetal heart rate abnormalities after epidural anesthesia placement during labor. STUDY DESIGN: Retrospective cohort study of 100 women with severe preeclampsia and 100 normotensive controls who underwent epidural anesthesia during labor from May 2008 to July 2011. Blood pressures during 2-minute time intervals for 20 minutes postepidural were compared with baseline measurements. Fetal heart rate abnormalities and related interventions were evaluated according to study group. RESULTS: Severe preeclampsia was associated with a greater percent decrease in blood pressure across all intervals and more episodes of systolic and diastolic hypotension after dosing. Severely preeclamptic women received more intravenous pressor support and developed more Category II fetal heart rate tracings, minimal variability, and late decelerations after dosing. CONCLUSION: Severe preeclampsia is associated with more frequent hypotension, need for pressor support, and fetal heart rate abnormalities after epidural anesthesia placement during labor.

Impact of morbid obesity on epidural anesthesia complications in labor.

OBJECTIVE: We sought to determine whether morbid obesity is associated with increased maternal hypotension or fetal heart rate (FHR) abnormalities after epidural anesthesia placement during labor. STUDY DESIGN: This was a retrospective cohort study of women undergoing epidural anesthesia during labor at term from April 2008 through July 2010. RESULTS: A total of 125 morbidly obese patients were matched for age and race with 125 normal-weight patients. Morbidly obese patients had more frequent persistent systolic (16% vs 4%, P = .003) and diastolic (49% vs 29%, P = .002) hypotension and more prolonged (16% vs 5%, P = .006) and late (26% vs 14%, P = .03) FHR decelerations. Increasing body mass index was associated with persistent systolic (odds ratio, 1.06; 95% confidence interval, 1.02-1.10) and diastolic (odds ratio, 1.04; 95% confidence interval, 1.01-1.06) hypotension after controlling for epidural bolus dose and hypertensive disorders. CONCLUSION: Morbidly obese women have more hypotension and prolonged FHR decelerations following epidural anesthesia during labor at term.

Anesthesia complications during scheduled cesarean delivery for morbidly obese women.

OBJECTIVE: We sought to estimate the morbidity associated with regional anesthesia in morbidly obese women undergoing scheduled cesarean delivery. STUDY DESIGN: This was a retrospective cohort study of women undergoing elective scheduled cesarean delivery from September 2004 through December 2008. RESULTS: A total of 142 morbidly obese, 251 overweight and obese, and 185 normal-weight women met inclusion criteria. Differences between groups were identified regarding: complicated placement (5.6%, 2.8%, and 0%, respectively; P = .007), failure to establish (2%, 0%, and 0%, respectively; P = .047), and insufficient duration (4%, 0%, and 0%, respectively; P = .02) of regional anesthesia. The groups differed in the frequency of general anesthesia (6%, 0%, and 0%, respectively; P = .003), intraoperative hypotension (3%, 0%, and 0%, respectively; P = .01), and overall anesthetic complications (8.4%, 0%, and 0%, respectively; P < .0001). Prepregnancy body mass index > or = 40 kg/m(2) (receiver operating characteristic area under the curve, 0.856; positive likelihood ratio, 4.0) and delivery body mass index > or = 45 kg/m(2) (receiver operating characteristic area under the curve, 0.877; positive likelihood ratio, 4.1) were predictive of anesthetic complications. CONCLUSION: Morbidly obese women have significant risk for anesthesia complications during cesarean delivery.