Prepregnancy obesity has become an increasing problem among women of reproductive age.1,2 A recent study examining the prevalence of prepregnancy obesity in nine states found that its prevalence increased a total of 69.3% in the decade of analysis, from 13% in 1993–1994 to 22.0% in 2002–2003.3 Women are not only beginning pregnancy at a higher body mass index (BMI, calculated as weight (kg)/[height (m)]2) but also they are also gaining more than has been recommended by the Institute of Medicine, and they are more likely to retain this excess weight after delivery.4–7
Obesity has been associated with multiple medical and reproductive health problems, including increased rates of gestational diabetes and preeclampsia.8–11 Children born to obese mother are also at risk, with higher rates of fetal growth restriction, macrosomia, stillbirth, and childhood obesity.11,12
Given these substantial risks, there has been support for lifestyle interventions before pregnancy.13–15 However, sustained weight loss with exercise, diet, and behavior modification is difficult to maintain; thus, more women have turned to bariatric surgery as an alternative.14,16 Several studies have been performed evaluating patients who undergo bariatric surgery and their subsequent pregnancy outcomes; the majority are cohort studies with small numbers of patients.16–20 A recent systematic review culled the results of many cohort studies and concluded that the rates of adverse maternal and neonatal outcomes may be lower in women who become pregnant after having undergone bariatric surgery.17 This study aims to estimate rates of pregnancy outcomes of women after bariatric surgery relative to those in a control group with BMIs similar to their presurgery BMIs and also to determine whether the outcomes are comparable with those of patients starting pregnancy at a similar BMI without surgery.
MATERIALS AND METHODS
A chart review was conducted with Northwestern University Institutional Review Board approval using the electronic medical record at Northwestern Memorial Hospital from December 1, 2005 to December 1, 2009. Bariatric surgery patients were identified with International Classification of Diseases, 9th Revision, Clinical Modification codes documenting weight loss surgery before a subsequent pregnancy. These included: 649.2, bariatric surgery status complicating pregnancy, child birth, or the puerperium; 44.95, laparoscopic gastric restrictive procedure; 44.96, laparoscopic revision of gastric restrictive procedure; and 44.68, laparoscopic gastroplasty. Patients were also identified if they had a surgical history that identified bariatric surgery, gastric bypass, lap-band, or weight loss surgery in their admission notes. Additional patients were included from the long-term outcomes of bariatric surgery patients database, maintained by the bariatric surgery department at Northwestern Memorial Hospital, and were cross-referenced with subsequent obstetric delivery notes. Further information was obtained about each patient's prenatal course and delivery by abstracting data from a perinatal database, created from an electronic medical record, which was completed by the patient's providers.
Seventy-five pregnant women, ranging from 18 to 50 years old who delivered after 20 weeks of gestation at Northwestern Memorial Hospital, were identified as having their first pregnancy after bariatric surgery. Two patients were subsequently removed before analysis because no documentation of bariatric surgery could be found. Three additional patients were eliminated because they carried multiple gestations. Self-reported prepregnancy weights were ascertained from the prenatal record. Prebariatric surgery weights were obtained from surgical reports when available, the prenatal record, or calculated based on the patient-reported number of pounds lost after bariatric surgery and adding those pounds to their prepregnancy weight. For those who had no identifiable prebariatric surgery weight (n=20 or 28.6%), we made an assumption that their BMI was 40, because this is the BMI at which otherwise healthy patients are eligible for bariatric surgery.21,22 For those patients with BMI exceeding 40 even after a bariatric procedure, we assumed no weight loss. Given the constraints of our data, we believed these assumptions to be conservative and more likely to show a true effect, if one were to exist.
Patients were identified for two separate control groups: those patients whose prepregnancy BMIs were within 6 of the mean BMI of bariatric surgery patients before they underwent surgery and those patients whose prepregnancy BMIs were within 6 of the mean of the postbariatric surgery prepregnancy BMI. Seeking twice as many control participants for case participants, the control participants were selected randomly from a pool of patients meeting these BMI criteria who delivered during the same time period. Thus, control group A comprised 140 patients with a BMI within 6 of the mean presurgery BMI of the case group (mean 49.31), and control group B similarly contained 140 patients with a BMI within 6 of the mean prepregnancy BMI of the case group (mean 33.66). Those patients in case and control groups with pre-existing chronic hypertension or diabetes were removed from the denominators for calculation of the outcomes of gestational diabetes and gestational hypertension, respectively. Demographic information, medical comorbidities, and delivery outcomes were collected for all patients. Neonatal outcomes, including birth weight, neonatal intensive care unit admission, selected perinatal morbidities, and mortality were also collected.
The primary outcomes of the study were gestational diabetes and hypertensive disorders of pregnancy. We chose these outcomes because previous studies have demonstrated a reduction in these maternal morbidities with bariatric surgery, and they are concerning maternal morbidities more frequently seen with morbid obesity.13,16–21,23,24 We defined preeclampsia by the American College of Obstetricians and Gynecologists criteria as elevated blood pressure of 140 mm Hg systolic or 90 mm Hg diastolic with proteinuria exceeding 0.3 g on a 24-hour collection or more than 100 mg/dL on urinanalysis. We diagnosed gestational hypertension in women who met these blood pressure criteria but without proteinuria.25 All patients had documented testing for gestational diabetes with a 1-hour glucola test and 3-hour glucose tolerance test, if necessary. The number of unscheduled visits to labor and delivery before their delivery encounter was tracked. Patients were labeled as having the complications of chorioamnionitis or shoulder dystocia if these were diagnosed and charted by their primary physician. A postpartum hemorrhage was defined as more than 500 mL of estimated blood loss for a vaginal delivery and more than 1,000 mL estimated blood loss for a cesarean delivery; additionally, postpartum blood transfusions were recorded. Postpartum emergency department visits within 6 weeks of delivery in the Northwestern Memorial Hospital emergency department were also tracked. Wound complications were assessed either in the inpatient setting or on additional presentation to the hospital and encompassed wound seromas, wound separations, or wound infections.
In calculating our sample size, we assumed an incidence of 4% of gestational diabetes and an incidence of 5% of preeclampsia in the general obstetrics population.25–28 Based on previous studies, we estimated the risks of gestational diabetes and hypertensive disorders of pregnancy in the morbidly obese population at 18% and 17%, respectively.13,16,19,23,24 With α of 0.05 and a ratio of 2:1 for control group participants to case group participants, 70 bariatric surgery case group participants would have 80% power to demonstrate a reduction in the risk of gestational diabetes and 70% power to demonstrate a reduction in the risk of hypertensive disorders of pregnancy to the background rates of these conditions.
Statistical analysis was performed using SAS 9.3 and Open Epi 2.3.1 to calculate χ2 or Fisher exact tests for categorical variables, Wilcoxon rank sum tests to compare medians, and Student t tests for the means of continuous variables. Minitab15 was used to perform logistic and linear regressions. Results are presented as means, medians, frequencies, and odds ratios with 95% confidence intervals. P<.05 was considered significant.
A total of 70 patients who delivered after bariatric surgery and who met inclusion criteria were included in this analysis. Of these 70 patients, nine (14.3%) had undergone laparoscopic banding, whereas 60 (85.7%) had undergone either open or laparoscopic Roux-en-Y gastric bypass and one procedure was not specified. Table 1 includes demographic and presentation characteristics for the case group and both control groups. The bariatric patients were significantly older than participants in control groups A and B. The mean BMI for control group A was similar to the mean presurgery BMI of the bariatric cohort, and the mean BMI for control group B was similar to the mean postsurgery BMI of the bariatric cohort. Bariatric patients delivered an average of 3.61 years after surgery, with only one patient conceiving in the first year. The patients were well-distributed across races, with no significant differences among white, African American, or Latina patients. Antenatal weight gain did not vary significantly across groups, and there was no significant difference in unscheduled hospital visits during the antepartum period.
Selected pre-exisiting medical comorbidities are outlined in Table 2. Patients were similar in most categories; however, there were significant differences across groups for hypertension, depression, and anemia. Bariatric surgery patients were significantly more likely to have hypertension than control group B participants, and significantly more likely to have depression than both comparison group participants. Additionally, there were significantly more anemic patients in the bariatric surgery group, possibly because of malabsorptive iron and vitamin B12 deficiencies. Bariatric surgery patients had more pre-existing diabetes and asthma than control group B participants, but this was not significant.
Delivery outcomes are described in Table 3. Median gestational age at delivery appeared similar between bariatric surgery patients and control group A participants, but there was a small but statistically significant difference compared with control group B participants. Additionally, the incidence of preterm birth was higher in the bariatric surgery group as compared with control group B. Control group A had higher rates of suspected macrosomia than did the bariatric surgery group. Hypertensive disorders of pregnancy were lower in the bariatric surgery group when compared with both control groups A and B, but this finding was not significant. However, the incidence of gestational diabetes was significantly decreased in the bariatric group compared with both control groups. Whereas there were no differences in mode of delivery, there were significantly more operative vaginal deliveries in the bariatric group than in control group A. There were also no differences in occurrences of shoulder dystocia among vaginal deliveries, chorioamnionitis, venous thromboembolism, or wound complication. The rate of postpartum hemorrhage, however, was significantly lower in the bariatric surgery group, but there was no significant difference in blood transfusion requirements. Further, the rate of postpartum endometritis was significantly lower than that in control group A. There were fewer postpartum emergency department visits among bariatric surgery patients, also compared with control group A participants, although this was not significant.
Table 4 illustrates the neonatal outcomes. There was a significant decrease in mean birth weight among bariatric surgery patients, and the incidence of macrosomia (defined as birth weight more than 4,000 g) was lower as well compared with both control groups. When birth weight was stratified by gestational age to evaluate fetal growth, there was a significant increase in small-for-gestational-age (SGA) neonates within the bariatric surgery group compared with both control groups. Other assessed neonatal morbidities were not significantly different among the groups, except for respiratory distress syndrome and perinatal mortality. The number of deaths was surprisingly high in the bariatric surgery group; all were fetal deaths of nonanomalous fetuses between 22 and 40 weeks of gestation. Further investigation revealed that all of these patients had normally grown fetuses, no evidence of hypertensive disorders of pregnancy or diabetes, and no significant pre-existing medical comorbidities. Three were preterm deliveries, one with cervical insufficiency at 22.5 weeks; this patient was the only patient who conceived within 1 year of her surgery. The fourth was a term intrauterine fetal demise without a laboratory or pathologic diagnosis.
A logistic regression was performed to analyze the effect of potential confounders on maternal and neonatal outcomes; these results are detailed in Table 5. Age, depression, and anemia had no effect on the outcome of pregnancy-related hypertension. Additionally, gestational diabetes could not be modeled using logistic regression because there were no cases of it in the bariatric surgery group. Neonatal outcomes of birth weight and SGA neonates remained significant when controlling for the same variables. Odds ratios revealed a decreased risk of macrosomia compared with control group A but an increased risk of SGA neonates in the bariatric surgery group compared with both control groups. The rate of stillbirth was no longer significantly increased compared with control group A, but it was still higher than the prepregnancy BMI-matched control group.
Although several studies have examined the rates of pregnancy-induced hypertension and gestational diabetes in the bariatric surgery population, many have had small numbers of bariatric surgery patients. Our study attempted to examine a larger case group of bariatric surgery patients who had a subsequent pregnancy and to compare them with morbidly obese patients who represent the outcomes in patients who had not had bariatric surgery and with obese patients who represent the outcomes in patients of similar prepregnancy BMI but who had not had bariatric surgery. Our study found a significant decrease in the occurrence of gestational diabetes in the bariatric surgery group as compared with both of their counterpart control cohorts. However, a decrease in the rates of hypertensive disease of pregnancy was not found to be significant. It is noteworthy that no cases of gestational diabetes were identified after bariatric surgery.
Our study supports the findings of other studies in the literature that have documented reductions in rates of gestational diabetes after bariatric surgery.16–21,29 Further, the significant decrease in the incidence of gestational diabetes in this sample as compared with both control groups was perhaps attributable to the biologically plausible possibility of metabolic or absorptive changes that occur in response to bariatric surgery. Although the literature does suggest a significant decrease in the rates of preeclampsia and gestational hypertension with bariatric surgery, our analysis was unable to detect a significant reduction. Our sample initially was powered to detect this difference based on estimations from the literature, but a post hoc calculation of our effect size demonstrated only 24% power, suggesting a type II error. Other secondary outcomes were also notable and not previously reported. There were no significant differences in route of delivery, except for an increased operative vaginal delivery rate compared with that of morbidly obese control group participants. The reduced rate of operative vaginal delivery in morbidly obese control patients may have been secondary to providers' suspicions of macrosomia or concerns for risk of shoulder dystocia in this cohort. Additionally, there was a decrease in rates of postpartum hemorrhage after bariatric surgery compared with both control groups. Overall, it appears that bariatric surgery lowers the risk of maternal morbidities for the morbidly obese to the risk of obese patients (and perhaps even lower) who have similar prepregnancy BMIs but who did not undergo surgery.
There were several important findings within the neonatal outcomes. Neonates of bariatric surgery patients had reduced mean birth weight and occurrence of macrosomia compared with both control groups. This was likely related to reduced rates of gestational diabetes and changes in nutritional status among bariatric surgery patients. A concerning finding was the significant increase in SGA neonates after bariatric surgery. This may be secondary to malnutrition or malabsorption, although a recent study comparing types of bariatric surgery discovered no difference in birth weight between purely restrictive and restrictive plus malabsorptive procedures.17,21 And although there were no increased rates of neonatal intensive care unit admissions or hypoglycemia in the neonates of bariatric surgery patients, there was an increased rate of perinatal mortality in the bariatric surgery group compared with control group B that persisted after controlling for potential confounders. Even though it is known that obese women have an increased risk of stillbirth, an increased risk in the bariatric surgery population has not previously been reported.16–21,29 Future studies will need to clarify whether this is a consistent finding and an important concern. Our findings would suggest that bariatric surgery confers an improvement in maternal outcomes, but possibly at the expense of an increase in adverse neonatal outcomes, including SGA neonates, preterm delivery, and perinatal mortality.
Limitations of this study include that it is a retrospective cohort study and is subject to confounding, information, and recall bias. Outcomes were estimated by controlling for potential confounders with logistic and linear regression, and data were abstracted directly from patient medical records, thus reducing ascertainment bias. However, not all records were complete, because presurgery BMIs were not available for 29% of patients; conservative estimates were used to define those missing BMIs. These estimates may bias our outcome toward the null, obscuring either larger effects on gestational diabetes or a significant effect on all hypertensive disorders of pregnancy. This study also reflects one institution's experience, which may have different practice patterns and which may not be representative of a larger population. Finally, the study was not powered to detect differences in all primary outcomes, most specifically preeclampsia and gestational hypertension.
There are clear health benefits of weight loss for morbidly obese women of reproductive age, and bariatric surgery has an important role to play in this population. Our data support the findings of previous investigators with regard to improved pregnancy outcomes for women after weight loss surgery, but they do raise new concerns regarding selected neonatal outcomes. In general, patients may be reassured that there are significant health benefits of bariatric surgery before pregnancy, but they also should be cautioned that there may be consequences for the pregnancy and the fetus. With the numbers of women of reproductive age seeking weight loss surgery increasing, it is important to consider the physiologic consequences of both restrictive and malabsorptive procedures and to work more closely with patients and their surgeons to optimize both prepregnancy and gestational nutritional status. This may help to reduce the risk of growth restriction and perinatal mortality in neonates born to mothers having undergone bariatric surgery. Finally, long-term follow-up of these neonates will determine the effect of prepregnancy weight loss surgery on what might be the most important consequences, metabolic health and rates of childhood obesity and diabetes.
1. Flegal KM, Carroll MD, Ogden CL, Curtin LR. Prevalence and trends in obesity among US adults, 1999–2008. JAMA 2010;303:235–41.
2. 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.
3. Kim SY, Dietz PM, England L, Morrow B, Callaghan WM. Trends in pre-pregnancy obesity in nine states, 1993–2003. Obesity (Silver Spring) 2007;15:986–93.
4. Chu SY, Kim SY, Bish CL. Prepregnancy obesity prevalence in the United States, 2004–2005. Matern Child Health J 2009;13:614–20.
5. Rasmussen KM, Abrams B, Bodnar LM, Butte NF, Catalano PM, Maria Siega-Riz A. Recommendations for weight gain during pregnancy in the context of the obesity epidemic. Obstet Gynecol 2010;116:1191–5.
6. Cedergren MI. Optimal gestational weight gain for body mass index categories. Obstet Gynecol 2007;110:759–64.
7. Olson CM, Strawderman MS, Hinton PS, Pearson TA. Gestational weight gain and postpartum behaviors associated with weight change from early pregnancy to 1 y postpartum. Int J Obes Relat Metab Disord 2003;27:117–27.
8. Ramos RG, Olden K. The prevalence of metabolic syndrome among US women of childbearing age. Am J Public Health 2008;98:1122–7.
9. Brown WV, Fujioka K, Wilson PW, Woodworth KA. Obesity: why be concerned? Am J Med 2009;122(4 Suppl 1):S4–11.
10. Metwally M, Saravelos SH, Ledger WL, Li TC. Body mass index and risk of miscarriage in women with recurrent miscarriage. Fertil Steril 2010;94:290–5.
11. Bariatric surgery and pregnancy. ACOG Practice Bulletin No. 105. American College of Obstetricians and Gynecologist. Obstet Gynecol 2009;113:1405–13.
12. Eerdekens A, Debeer A, Van Hoey G, De Borger C, Sachar V, Guelinckx I, et al.. Maternal bariatric surgery: adverse outcomes in neonates. Eur J Pediatr Feb 2010;169:191–6.
13. Obesity in pregnancy. ACOG Committee Opinion No. 315. American College of Obstetricians and Gynecologists. Obstet Gynecol 2005;106:671–5.
14. Satpathy HK, Fleming A, Frey D, Barsoom M, Satpathy C, Khandalavala J. Maternal obesity and pregnancy. Postgrad Med 2008;120:E01–9.
15. Power ML, Cogswell ME, Schulkin J. US obstetrician-gynaecologist's prevention and management of obesity in pregnancy. J Obstet Gynaecol 2009;29:373–7.
16. Wax JR, Cartin A, Wolff R, Lepich S, Pinette MG, Blackstone J. Pregnancy following gastric bypass surgery for morbid obesity: maternal and neonatal outcomes. Obes Surg 2008;18:540–4.
17. Maggard MA, Yermilov I, Li Z, Maglione M, Newberry S, Suttorp M, et al.. Pregnancy and fertility following bariatric surgery: a systematic review. JAMA 2008;300:2286–96.
18. Patel JA, Patel NA, Thomas RL, Nelms JK, Colella JJ. Pregnancy outcomes after laparoscopic Roux-en-Y gastric bypass. Surg Obes Relat Dis 2008;4:39–45.
19. Ducarme G, Revaux A, Rodrigues A, Aissaoui F, Pharisien I, Uzan M. Obstetric outcome following laparoscopic adjustable gastric banding. Int J Gynaecol Obstet 2007;98:244–7.
20. Bennett WL, Gilson MM, Jamshidi R, Burke AE, Segal JB, Steele KE, et al.. Impact of bariatric surgery on hypertensive disorders in pregnancy: retrospective analysis of insurance claims data. BMJ 2010;340:c1662.
21. Sheiner E, Balaban E, Dreiher J, Levi I, Levy A. Pregnancy outcome in patients following different types of bariatric surgeries. Obes Surg 2009;19:1286–92.
22. Gastrointestinal surgery for severe obesity. Proceedings of a National Institutes of Health Consensus Development Conference. March 25–27, 1991, Bethesda, MD. Am J Clin Nutr 1992;55:487–619S.
23. Weiss JL, Malone FD, Emig D, Ball RH, Nyberg DA, Comstock CH, et al.. Obesity, obstetric complications and cesarean delivery rate–a population-based screening study. Am J Obstet Gynecol 2004;190:1091–7.
24. Kumari AS. Pregnancy outcome in women with morbid obesity. Int J Gynaecol Obstet 2001;73:101–7.
25. Diagnosis and management of preeclampsia and eclampsia. ACOG Practice Bulletin No. 33. American College of Obstetricians and Gynecologists. Obstet Gynecol 2002;99:159–67.
26. Chu SY, Callaghan WM, Kim SY, Schmid CH, Lau J, England LJ, et al.. Maternal obesity and risk of gestational diabetes mellitus. Diabetes Care 2007;30:2070–6.
27. O'Brien TE, Ray JG, Chan WS. Maternal body mass index and the risk of preeclampsia: a systematic overview. Epidemiology 2003;14:368–74.
28. Clinical management guidelines for obstetrician-gynecologists. ACOG Practice Bulletin No. 30. American College of Obstetricians and Gynecologists. Obstet Gynecol 2001;98:525–38.
29. Josefsson A, Blomberg M, Bladh M, Frederiksen SG, Sydsjo G. Bariatric surgery in a national cohort of women: sociodemographics and obstetric outcomes. Am J Obstet Gynecol 2011;205:206.e1–8.