Obesity is pandemic, resulting in a substantial disease burden worldwide. In the United States, more than 50% of nonpregnant reproductive-aged patients and 35% of preconception or interconception women are overweight or obese; 8% of reproductive-aged women are extremely obese.1,2 Cesarean delivery is one of the most prevalent surgical procedures in the United States.3 Unfortunately, obese patients are at increased risk not only for requiring cesarean delivery, but also for postoperative complications.4–7 Although obesity is known to increase pregnancy complications, the effect of varying degrees of obesity on maternal postoperative risks is not well characterized.4–11 The paucity of information on perioperative risks associated with increasing severity of obesity limits the clinician in counseling and clinical decision-making. Thus, we designed a cohort study to estimate the effect of obesity and extreme obesity on postcesarean maternal complications.
MATERIALS AND METHODS
This is a retrospective cohort study of 585 patients enrolled between 2008 and 2010 performed in a prospective randomized controlled trial designed to evaluate the efficacy of supplemental oxygen for the prevention of postcesarean infectious morbidity. Details of the trial were published previously but are summarized briefly.12 In the initial trial, patients were randomized in a one-to-one ratio to receive either 2 L of oxygen using a nasal cannula during cesarean delivery (standard care) or 10 L of oxygen using a nonrebreather mask (intervention group) during and 2 hours after cesarean delivery. Women received standardized preoperative iodine-based skin antiseptic preparation and perioperative prophylactic antibiotics. During the study period, cefazolin was used as primary preoperative prophylaxis, and clindamycin was used for patients with penicillin allergy. Women undergoing scheduled or intrapartum cesarean delivery with regional anesthesia were eligible for participation and were followed for 1 month postoperatively. We recruited women for study participation from the antepartum service and labor and delivery. Exclusion criteria included emergency surgery, human immunodeficiency virus infection, chronic corticosteroid therapy or other immunosuppressive therapy, general anesthesia, and a diagnosis of extrauterine infection. We obtained institutional review board approval at Washington University School of Medicine before patient enrollment, and written informed consent was obtained from all participants. Washington University institutional review board approval was also granted for the current secondary study before initiation. The primary outcome for the parent study, a composite outcome consisting of endometritis and wound infection, and secondary adverse postoperative complications were defined using strict and distinct diagnostic criteria.12 All outcomes were assessed by the primary care team, and the diagnoses were abstracted from the chart by trained study personnel. The medical record was reviewed at the time of the 2- to 4-week postoperative visit, and all women who did not present for a postoperative visit within 4 weeks postoperatively or who had planned follow-up at an outside clinic were contacted by the research nurse for a standardized telephone interview (23%) to inquire about postoperative complications. All data collection was performed using detailed standardized data collection forms. Data were entered into a database by two trained research team members including a dedicated research nurse and one investigator (C.M.S.). Data were validated prospectively with recurrent research team meetings (usually monthly) to validate collected data, especially with regard to wound outcomes, and to achieve consensus on questionable outcomes. Data were also validated retrospectively with internal validation methods by the two primary investigators (C.M.S. and D.M.S.).
The study intervention, supplemental oxygen, had no effect on maternal morbidity; thus, we combined the two intervention groups into one cohort for this analysis. The primary composite outcome for this secondary analysis is maternal infectious morbidity, including endometritis and wound infection. A patient was diagnosed with endometritis if she had an oral temperature higher than 38°C after the first 24 hours postoperatively and either 1) fundal or lower abdominal tenderness greater than expected; or 2) foul-smelling or purulent lochia. Endometritis was diagnosed only if other causes for the patient's signs and symptoms were not identified. Patients had to have been treated with intravenous antibiotics for a diagnosis of endometritis to meet our study definition. The diagnosis of wound infection required a wound opening larger than 1 cm or other surgical intervention such as laparotomy or débridement of tissue plus at least one of the following: 1) purulent drainage from the wound; 2) erythema or induration of the surrounding tissues; 3) maternal oral temperature higher than 38°C; or 4) radiographic evidence of infection. Secondary outcomes include each composite outcome component (endometritis, wound infection) as well as wound opening (larger than 1 cm), wound hematoma or seroma, hospital readmission within 4 weeks, and emergency department visits. The exposure under study is obesity, stratified as normal or overweight (body mass index [BMI, calculated as weight (kg)/[height (m)]2] less than 30), obese (BMI 30–45), or extremely obese (BMI higher than 45). Body mass index was calculated based on weight at the time of study enrollment. All patients were enrolled in the third trimester and within 1–2 days before delivery. Therefore, BMI assessment is most relevant to the time of surgery. Before study initiation, we defined obesity using the BMI (30) cutoff recommended by the World health Organization and extreme obesity using a BMI cutoff derived from a review of the obstetric literature.7,11,13,14
The primary aim for this cohort study is to compare the maternal composite postcesarean infectious morbidity outcome across the three maternal weight exposure strata: nonobese, obese, and extremely obese. We also compared the secondary outcomes across the same BMI-defined patient groups. We first compared the BMI exposure groups for key demographic variables, comorbidities, and outcomes in an unadjusted analysis using analysis of variance and χ2 or Fisher’s exact tests as appropriate for variable type and frequency. We used logistic regression analysis to perform multivariable analysis to estimate the independent effect of obesity on postcesarean complication rate controlling for important confounders of the association between obesity and adverse outcomes. Obesity was included in the multivariable model as an indicator (also known as a “dummy”) variable because it is a trichotomous categorical variable. We selected covariates for inclusion in the initial regression models based on the unadjusted analysis results, biologic plausibility, and prior published research. Candidate variables were selected from the unadjusted analysis if they were associated with the outcome (P<.10). To achieve the final best-fit multivariable models, we used a nonautomated process of backward variable elimination, testing sequentially for significant differences between hierarchical models with the likelihood ratio test or Wald test. We assessed the final explanatory regression model for goodness of fit with the Hosmer–Lemeshow summary statistic. We report adjusted odds ratios (to estimate relative risk) and 95% confidence intervals (CIs) as estimates of the effect of obesity level on postcesarean complication risk. Because all data were collected prospectively, there were very few missing data points. All data collected before hospital discharge were complete with one exception: cord blood gas results were available only for 89.7% of patients. There were 2% of patients (n=11) lost to follow-up for the 4-week infectious morbidity outcomes assessment. These patients were assumed to not have an infectious morbidity in the analysis because patients in this population rarely seek postoperative care outside of our facility.
With a goal to identify attributable factors in the association between obesity and postoperative morbidity, we performed stratified analyses on operative characteristics, including uterine incision type, skin incision type, operative time, and closure technique. These variables were not included in the multivariable analyses because they are likely in the causal pathway and if adjusted for could introduce bias into the effect estimates in the association between obesity and outcome. Preoperative antibiotic administration was not assessed as a potential attributable factor because nearly all patients undergoing cesarean delivery receive this standardized preventive therapy at our institution.
We did not perform a formal sample size calculation because the sample size was fixed at the completion of the parent clinical trial. However, before initiating this study, we estimated that we would have 80% power to detect a 2.2-fold increased risk in the composite morbidity outcome associated with obesity class using the available sample size, a 10% baseline risk of the outcome, an extreme obesity prevalence of 15% within the cohort, and assuming statistical significance at a P value <.05.
In this prospective cohort of 585 patients, postcesarean infectious morbidity, defined as the composite of endometritis and wound infection, occurred in 10.4% of all patients in the cohort. Rates of normal or overweight (BMI less than 30) and obesity class I (BMI 30–34.99), class II (BMI 35–39.99), and class III (BMI 40 or higher) were 25.1%, 30.1%, 14.9%, and 29.9%, respectively. Eighty-five patients (14.5%) had extreme obesity, with BMIs higher than 45.
We display demographics, clinical factors, and surgical characteristics compared across obesity classifications in Table 1. Rates of maternal black race, chronic hypertension, diabetes, and gestational diabetes increased with increasing obesity severity. Operative duration increased with increasing obesity severity, and the mean operative time in extremely obese patients was 15 minutes longer than in nonobese patients. Rates of vertical skin incision and nontransverse uterine incision increased with increasing obesity. As anticipated, rates of preoperative antibiotic use were high and similar across obesity categories. Exposure groups were also similar with regard to the rates of cesarean delivery after labor and cesarean delivery after rupture of membranes.
After controlling for relevant confounders, extremely obese patients had a nearly threefold increase in the rate of the composite postoperative infectious morbidity outcome compared with nonobese patients (Table 2). We also observed a two- to fourfold increase in other postoperative complications in extremely obese patients, although the associations with endometritis, wound opening, seroma and hematoma development, and hospital readmission within 4 weeks were not statistically significant (Table 2). Extremely obese patients had statistically significant increases in the rates of wound infection and emergency department evaluation. Obese and extremely obese patients combined appeared to be readmitted to the hospital within 4 weeks after delivery two times more frequently than nonobese patients; however, the difference in these point estimates was not statistically significant (relative risk 2.5, 95% CI 0.8–8.2, P=.1). Point estimates for rates of most adverse outcomes in patients with obesity not characterized as extreme appeared higher, but the relative risks were not statistically significant (Table 2).
We performed stratified analyses on skin and uterine incision types among only obese patients. Neither skin nor uterine incision type was associated with morbidity in all obese (data not shown) or extremely obese patients. However, this subanalysis is limited by the restricted sample size and the consequentially low power. Among the 85 extremely obese patients, we observed an excess of the primary composite infection outcome in those who had transverse (21.9%) compared with vertical (0%) uterine incisions, but this was not statistically significant (P=.07). Among extremely obese patients, the observed rate of seroma or hematoma was higher with vertical (20%) compared with transverse (5.3%) skin incisions, but this was not statistically significant (P=.09). We also performed stratified analyses on other key operative characteristics. We explored the effect of long operative times on maternal postoperative morbidity. Although obese patients had longer operative times on average, a long operative time—defined as higher than the 90th percentile for nonobese patients (83 minutes)—was not associated with an increase in the composite infectious morbidity outcome in obese, extremely obese, or nonobese patients (data not shown). Lastly, we explored the effect of placing sutures in the subcutaneous tissue on postoperative morbidity. Subcutaneous suture placement did not affect the postcesarean infectious morbidity rate in obese, extremely obese, or nonobese patients; relative risks (95% CI) for the primary infectious morbidity outcome with subcutaneous suture use were 0.8 (0.3–1.8), 1.4 (0.5–4.2), and 1.0 (0.3–3.4) for the respective weight categories. The lack of observed associations between various operative characteristics and infectious morbidity could alternately be attributed to low statistical power rather than a proven equivalence.
We assessed the effect of obesity severity on coarse measures of neonatal outcome. Birth weight increased with increasing severity of maternal obesity; mean birth weight (standard deviation) was 2,878 g (827), 3,126 g (805), and 3,310 g (702) for nonobese, obese, and extremely obese patients, respectively (χ2 for trend P<.001). Mean umbilical cord blood pH was lower for extremely obese patients compared with nonobese and obese patients (7.23 compared with 7.27, P=.003). However, cord blood base deficit and 5-minute Apgar score were similar across maternal weight classifications (data not shown). Unexpectedly, the rate of admission to the neonatal intensive care unit or special care nursery was significantly lower (P<.001) in extremely obese patients (18.8%) compared with obese (30.9%) or nonobese patients (36.7%). In logistic regression analysis, longer operative time did not confound or modify the association between extreme obesity and neonatal intensive care unit admission.
Using a well-characterized cohort of general obstetric patients who required cesarean delivery, we observed that obese patients, particularly those with extreme obesity, have a considerable increase in risk for postoperative complications compared with nonobese patients. Patients with lesser degrees of obesity appeared to have estimates for adverse outcome rates that were higher than nonobese patients but risks were not statistically increased compared with nonobese patients. In our cohort, postcesarean complication risks associated with obesity were not significantly worsened by other comorbidities such as diabetes mellitus or hypertension. Thus, risks were mostly attributable to obesity rather than chronic disease. Although previous studies have shown a decrease in risk for postoperative wound complications with the use of subcutaneous adipose sutures,15 we did not observe a risk reduction with the use of this wound closure approach. This may have been the result of limited power to detect a difference or the fact that the subcutaneous suture approach was not used consistently because the technique was not randomized or prescribed by the study protocol. Surprisingly, subcutaneous closure was used in only 14% of extremely obese patients and the frequency that surgeons used the subcutaneous closure technique decreased as obesity increased. This counterintuitive practice pattern may be an indication that the technique could be used more universally at our institution to reduce postoperative risk among obese patients.
Unfortunately, there is limited clinical research on the effects of increasing severity of obesity on perioperative cesarean complications and risks.14,16 Most recently, Connor and colleagues17 performed a retrospective cohort study that revealed a positive dose–response relationship between obesity severity and cesarean wound complication rate. Our findings corroborate those of Connor and colleagues.
The advantages of our study design enabled us to better characterize perioperative complications as well as validate findings of previous observational studies. Potential for misclassification bias is minimal because the cohort data were collected prospectively for a clinical trial designed to study postcesarean wound complications. Thus, wound complications were strictly defined and both exposure and outcome data are accurate and complete. Data on surgical technique are also robust because they were a focus of the primary study.
Although our study has advantages, it is important to acknowledge its limitations. Because the parent study was designed primarily to assess maternal risks, data on neonatal outcomes are limited to gross measurements of neonatal outcome. This study population is derived from a metropolitan referral center; therefore, findings may not be generalizable to other populations. The sample size is restricted for the purpose of subanalyses of surgical technique and stratified analyses such as skin and uterine incision types. Furthermore, because surgical technique was not randomly allocated, there is a distinct possibility for selection bias or confounding by indication with regard to the association between various surgical techniques and complications. Specifically, we observed no relationship between skin incision type or uterine incision type and wound complications within obese or extremely obese patients, but the study has limited power to detect an association. Data are conflicting, but most observational studies of obese patients do not reveal a difference in postcesarean wound complication rates between vertical and transverse skin incisions.18–21
In conclusion, extremely obese pregnant patients are at increased risk for postcesarean infectious morbidity and other postoperative complications. These data can be used to counsel extremely obese patients on operative risks. Our findings suggest that clinicians should consider obesity in delivery mode decisions and may need to alter the threshold for using cesarean delivery in extremely obese patients compared with nonobese patients. Additionally, clinicians should consider earlier or more frequent postoperative follow-up in extremely obese patients to monitor for wound complications. Our results confirm an urgent need for clinical trials to research novel strategies to reduce the chance of cesarean delivery and to develop perioperative methods to reduce the risks of cesarean delivery in extremely obese patients.
1. Flegal KM, Carroll MD, Kit BK, Ogden CL. Prevalence of obesity and trends in the distribution of body mass index among US adults, 1999-2010. JAMA 2012;307:491–7.
2. D'Angelo D, Williams L, Morrow B, Cox S, Harris N, Harrison L, et al.. Preconception and interconception health status of women who recently gave birth to a live-born infant—Pregnancy Risk Assessment Monitoring System (PRAMS), United States, 26 reporting areas, 2004. MMWR Surveill Summ 2007;56:1–35.
4. Obesity in pregnancy. Committee Opinion No. 549. American College of Obstetricians and Gynecologists. Obstet Gynecol 2013;121:213–7.
5. Fyfe EM, Anderson NH, North RA, Chan EH, Taylor RS, Dekker GA, et al.. Risk of first-stage and second-stage cesarean delivery by maternal body mass index among nulliparous women in labor at term. Obstet Gynecol 2011;117:1315–22.
6. Magriples U, Kershaw TS, Rising SS, Westdahl C, Ickovics JR. The effects of obesity and weight gain in young women on obstetric outcomes. Am J Perinatol 2009;26:365–71.
7. Weiss JL, Malone FD, Emig D, Ball RH, Nyberg DA, Comstick CH, et al.. Obesity, obstetric complications and cesarean delivery rate—a population-based screening study. Am J Obstet Gynecol 2004;190:1091–7.
8. Bonnesen B, Secher NJ, Moller LK, Rasmussen S, Andreasen KR, Renault K. Pregnancy outcomes in a cohort of women with a preconception body mass index >50 kg/m2
. Acta Obstet Gynecol Scand 2013;92:1111–4.
9. O'Dwyer V, Farah N, Fattah C, O'Connor N, Kennelly MM, Turner MJ. The risk of caesarean section in obese women analysed by parity. Eur J Obstet Gynecol Reprod Biol 2011;158:28–32.
10. Pallasmaa N, Ekblad U, Aitokallio-Tallberg A, Uotila J, Raudaskoski T, Ulander VM, et al.. Cesarean delivery in Finland: maternal complications and obstetric risk factors. Acta Obstet Gynecol Scand 2010;89:896–902.
11. Robinson HE, O'Connell CM, Joseph KS, McLeod NL. Maternal outcomes in pregnancies complicated by obesity. Obstet Gynecol 2005;106:1357–64.
12. Scifres CM, Leighton BL, Fogertey PJ, Macones GA, Stamilio DM. Supplemental oxygen for the prevention of postcesarean infectious morbidity: a randomized controlled trial. Am J Obstet Gynecol 2011;205:267.e1–9.
13. Cedergren MI. Maternal morbid obesity and the risk of adverse pregnancy outcome. Obstet Gynecol 2004;103:219–24.
14. Perlow JH, Morgan MA. Massive maternal obesity and perioperative cesarean morbidity. Am J Obstet Gynecol 1994;170:560–5.
15. Dahlke JD, Mendez-Figueroa H, Rouse DJ, Berghella V, Baxter JK, Chauhan SP. Evidence-based surgery for cesarean delivery: an updated systematic review. Am J Obstet Gynecol 2013;209:294–306.
16. Alanis MC, Goodnight WH, Hill EG, Robinson CJ, Villers MS, Johnson DD. Maternal super-obesity (body mass index > or = 50) and adverse pregnancy outcomes. Acta Obstet Gynecol Scand 2010;89:924–30.
17. Conner SN, Verticchio JC, Tuuli MG, Odibo AO, Macones GA, Cahill AG. Maternal obesity and risk of postcesarean wound complications. Am J Perinatol 2014;31:299–304.
18. Bell J, Bell S, Vahratian A, Awonuga AO. Abdominal surgical incisions and perioperative morbidity among morbidly obese women undergoing cesarean delivery. Eur J Obstet Gynecol Reprod Biol 2011;154:16–9.
19. Houston MC, Raynor BD. Postoperative morbidity in the morbidly obese parturient woman: supraumbilical and low transverse abdominal approaches. Am J Obstet Gynecol 2000;182:1033–5.
20. McLean M, Hines R, Polinkovsky M, Stuebe A, Thorp J, Strauss R. Type of skin incision and wound complications in the obese parturient. Am J Perinatol 2012;29:301–6.
21. Wall PD, Deucy EE, Glantz JC, Pressman EK. Vertical skin incisions and wound complications in the obese parturient. Obstet Gynecol 2003;102:952–6.