Home > Results - Bariatric Surgery in Women of...

PubMed Health. A service of the National Library of Medicine, National Institutes of Health.

Maggard M, Li Z, Yermilov I, et al. Bariatric Surgery in Women of Reproductive Age: Special Concerns for Pregnancy. Rockville (MD): Agency for Healthcare Research and Quality (US); 2008 Nov. (Evidence Reports/Technology Assessments, No. 169.)

  • This publication is provided for historical reference only and the information may be out of date.

This publication is provided for historical reference only and the information may be out of date.


Description of the Studies

Our literature search resulted in 998 titles. Reference mining of obtained articles resulted in 37 additional titles. Of these 1,035 titles, 231 titles appeared potentially relevant to our scope and were ordered. We were unable to obtain eight articles before our deadline. Thus, a total of 223 articles were screened using the one-page form described in the Methods section. Figure 1 displays the article flow for the project.

Figure 1. Literature Flow Diagram.


Figure 1. Literature Flow Diagram.

Of the 223 articles screened, 57 were accepted for our report, including 23 case reports, 21 case series, 12 cohort studies, and one case-control study. Details of all accepted articles are presented in the Evidence Tables (Appendix D *). A total of 166 articles were rejected: 88 were not actually on bariatric surgery despite the article title, 60 were not on a procedure of interest or did not include pregnant women, 14 were nonsystematic reviews, two were systematic reviews, one was a background article, and one was in a foreign language where an interpreter was not available.

We identified one case-control study that directly addressed some of the key questions, but no randomized controlled trials or prospective cohort studies. Our findings are based on observational studies, which have a potential for greater bias. Furthermore, many of the studies lacked the necessary design to allow for definite conclusions (i.e., patient selection not defined, no presurgery pregnancy information). Our overall findings are therefore tempered by the limitations in the available data, and are cautious.

Key Question 1. What is the incidence of bariatric surgery in women of reproductive age? What are the trends in incidence of bariatric surgery in women of reproductive age?

For this question, a search for published data on this topic did not find articles reporting data on use of surgery in women of reproductive age. The closest article we found reported use rates overall and by sex, but did not report separately use rates for our target population, women aged 18–45.23 Therefore, with the agreement of AHRQ and our TEP, we performed our own analyses to answer this question. We analyzed data from AHRQ's Healthcare Cost and Utilization Project (HCUP) Nationwide Inpatient Sample for the years 1998-2005 (the last year for which data are present). ICD-9 procedure codes and inclusion criteria of diagnosis of obesity were used to select the cases. Additionally, ICD recodes of bariatric surgery during this time period were also accounted for where appropriate. Table 2 and Figure 2 report our findings. We present estimates of the number of bariatric procedures done, per year, by sex and certain age ranges. These estimates are based on the actual number of procedures in the Nationwide Inpatient Sample database multiplied by the weights used to estimate the total U.S. population. Note that this database will not capture procedures done as an outpatient. As the delivery of outpatient laparoscopic adjustable band procedures increases, this database will increasingly underestimate the use of bariatric procedures. In addition, some gastric bypass procedures have been done as outpatient operations. All estimates are subject to some degree of error due to coding issues.

Table 2. Number of inpatient bariatric procedures*.

Table 2

Number of inpatient bariatric procedures*.

Figure 2. Trend in bariatric inpatient procedures, 1998-2005 *.


Figure 2. Trend in bariatric inpatient procedures, 1998-2005 *. * From Nationwide Inpatient Sample, Healthcare Cost and Utilization Project (HCUP), U.S. Agency for Healthcare Research and Quality. Accessed July 2007.

For both men and women, and across all age ranges, we found a dramatic increase in the number of procedures performed each year—about 600–800 percent. This observation mirrors recent findings by other researchers.26 An interesting finding of this analysis is that the growth in use of bariatric procedures delivered in the inpatient setting has been even more pronounced in persons over the age of 45. Also of note, there was actually a leveling off of incidence rates in 2003 and a drop in the incidence rate between years 2004 and 2005. One potential explanation for this plateau in the later years, and the lesser rate of increase in younger patients as compared to older patients, is the likely increase in the number of laparoscopic adjustable band procedures being delivered on an outpatient basis and that proportionately more of these procedures are being performed in the older population. Alternatively, the observation could represent a true drop off in the number of cases, perhaps related to changes in insurance coverage.

In summary, in the past 3 years, more than 50,000 women of reproductive age underwent bariatric surgery inpatient procedures annually. Many more women in this age group are also likely undergoing bariatric procedures in the outpatient setting that are not captured in this inpatient dataset. The proportion of these women who subsequently get pregnant is not known.

Key Question 2. What is the evidence that bariatric surgery affects (directly or indirectly) future fertility?

We identified few studies focused specifically on the question of fertility in postbariatric surgery patients. Most studies recruited patients because they were already pregnant; thus they failed to include those patients who remained infertile following the procedure. Additionally, although many studies provided the number of patients who were previously infertile or had undergone fertility treatments in the past, it was usually not clear how many women, out of the total number of women of reproductive potential, desired a pregnancy; therefore, these data should be interpreted with some caution.

Most studies that reported on fertility following bariatric surgery compared patients before and after surgery. These studies are displayed in Table 3. One study compared pre and postoperative reproductive histories of female patients who underwent bariatric surgery and had lost more than 50 percent of excess weight (unclear if consecutive patients). Twenty five percent (29/115) of the women had suffered from infertility prior to bariatric surgery. There were data available on nine women who conceived after surgery, eight were in the group that had had infertility problems preoperatively. Unfortunately, the follow-up time is not stated and it is unknown how many other patients in the postsurgery cohort had tried to conceive.27 This difficulty in determining how many patients, either preoperatively or postoperatively, actually desired pregnancy is common to almost all studies in this group. Another retrospective cohort study compared nine women before and after vertical banded gastroplasty with respect to fertility; two of these patients had not attempted pregnancy prior to surgery. Whereas five out of seven women underwent fertility treatments prior to surgery, only one woman underwent ovulation induction after the surgery. All nine women became pregnant within 5 years after surgery; they represent all female bariatric surgery patients at the site.28 In another retrospective cohort study comparing pregnancies of the same women before and after BPD, of 32 women who had unsuccessfully attempted conception prior to surgery, 15, or 47 percent were able to become pregnant following BPD.29

Table 3. Cohort studies reporting on fertility.

Table 3

Cohort studies reporting on fertility.

Other studies compared a group of patients who had bariatric surgery to those who did not and were not obese. One study compared patients who had undergone bariatric surgery to the general population in Negev, Israel; patients who had undergone bariatric surgery were more likely to have received fertility treatments (6.7 percent vs. 2.3 percent).30 In a study evaluating patients with gestational diabetes who either had bariatric surgery or did not, the bariatric surgery group had higher rates of fertility treatments, which persisted after controlling for obesity.31 However, in both of these studies, it is not entirely clear whether the fertility treatments were “after surgery” or “lifetime.” Similarly, one randomized controlled trial comparing three bariatric procedures (gastric bypass and two kinds of gastroplasty) found that about 10 percent of postsurgical women (<40 y/o) got pregnant in 3 years (21/214), but the number attempting pregnancy was unknown.32

In general, most of the data on need for fertility treatments following bariatric surgery lack information on the number of postoperative patients attempting to get pregnant and number ultimately successful. In general, sample sizes are too small to have statistical power. In addition, most of these data represent convenience samples of women able to get pregnant along with their presurgery fertility histories. The larger studies compare the postsurgery cohorts (although they lost weight still have a higher rate of obesity) to nonobese population, a comparison which is limited since obesity is associated with higher infertility rates. This may explain why some studies report improvements in fertility comparing women before and after bariatric surgery, while other studies report elevated fertility problems in women following bariatric surgery compared to the nonobese general population.

Several case series mentioned prior infertility rates in patients who were able to become pregnant following bariatric surgery. In patients who were able to conceive after bariatric surgery, infertility rates prior to surgery ranged from 15 to 44 percent.3336 In another case series that included 49 postsurgery pregnancies, it was mentioned that no fertility drugs were used.37

In addition to this limited evidence of improved fertility by increased pregnancy rates, there is also evidence of normalization of hormones and menstrual cycles as well as improvement of polycystic ovarian syndrome (PCOS). A prospective case series evaluating hormone levels in women of reproductive age both pre and post BPD demonstrated a normalization of hormones, specifically, a rise in sex hormone binding globulin and decreasing levels of serum testosterone and dehydroepiandrosterone sulfate (DHEA-S).38 In addition to the direct laboratory evidence of hormone normalization, there is also indirect evidence of normalization through the return of normal menstrual cycles in women who had had irregular menses. Of the 40 percent of women of reproductive age who were having irregular menses preoperatively in a cohort study, 89 percent of these patients resumed regular menses following bariatric surgery.27 Lastly, the resolution of PCOS was seen following gastric bypass; in a retrospective cohort study that included 24 women with PCOS, all women resumed normal menstrual cycles in a mean of 3.4 months, and five women were able to conceive without the use of clomiphene. Additionally, of the 23 women with hirsutism, 52 percent had complete resolution of symptoms.39

Key Question 3. What is the evidence that bariatric surgery affects (directly/indirectly) choice of contraception?

There is almost no evidence on this topic. We found only a single study that reported data on the effectiveness of contraceptive methods following surgery. A small case series of 40 patients who had undergone BPD and who were advised to avoid pregnancy for at least 2 years reported two failures for oral contraceptive (OCP) birth control, one at 9 months postoperatively and the other at 24 months. Based on this 5.0 percent failure rate, these authors advised the use of something other than OCP, and called for a large RCT to determine the best method of contraception.38 Given that the failure rate of oral contraceptives in the first year of typical use has been reported at 3 percent for American women,40 the failure rate after BPD may not be higher; clearly more data are needed before conclusions can be drawn.

Key Question 4. In patients who have had bariatric surgery, what is the evidence for prenatal risk factors (e.g., of reduced nutrient absorption, unusual weight gain) that may result in poor pregnancy outcomes?

Key Question 5. What is the evidence that certain management strategies for addressing nutrient absorption and weight gain reduce the risks of poor pregnancy outcomes?

Nutritional Supplementation

It is common practice to recommend nutritional supplementation such as multi-vitamins and iron following bariatric surgery for all patients.41, 42 However, evidence is scarce regarding specific recommendations for pregnancy after bariatric surgery.

Studies evaluating pregnancy following gastric banding or gastric bypass have shown minimal nutritional adverse events; however, most of these studies monitored and ensured that the women complied with vitamin supplementation. In a study where 84 percent of 79 pregnant women reported compliance with multivitamin supplementation following gastric banding, no nutritional problems were reported during pregnancy.43 Three studies describing pregnancy after gastric bypass also describe low rates of anemia requiring either oral or parenteral supplementation (ranging from 0 percent to 4 percent); however these studies did not describe the vitamin supplementation regimen that these women followed.37, 44, 45

There are reports in the literature describing the need for supplementation and parenteral nutrition in pregnancy following BPD. Three reports (one in abstract form) from the same investigators and institution describe nutritional problems in patients following BPD.35, 46, 47 In the largest of these three reports, 1136 women who received BPD surgery between 1976 and 1994 at a single institution had 245 pregnancies occurring two to 17.3 months following the surgery. There were 73 abortions, two for nutritional causes (no additional data provided). In 21 percent of patients, parenteral nutritional support was required (criteria unstated), with about a third of these requiring hospitalization. For all other patients “the usual supplements were given.”47 In a small case series evaluating the efficacy of contraception following BPD, four pregnancies occurred; one woman suffered from anemia, while another suffered from an unspecified vitamin deficiency. While one of these pregnant women was taking nutritional supplementation, the other was not.38 In another small case series that included nine pregnancies following BPD, all four patients who were tested suffered from nutritional deficiencies, requiring blood transfusions, parenteral nutrition, or parenteral iron supplementation.48 Lastly, there have been case reports of adverse events in pregnancy following BPD. In one case, dehydration and malnutrition as a result of vomiting and diarrhea led to an emergent caesarian section despite vitamin supplementation and multiple hospitalizations to administer intravenous fluids.49 Another case report demonstrated vitamin A deficiency in pregnancy following BPD; late in the pregnancy, the patient was hospitalized 5 days weekly for parenteral nutrition. The baby was still found to have symptoms of vitamin A deficiency, such as micropthalmia, at birth.50 As the risk of nutritional adverse events in pregnancy following BPD is appreciable, even with good compliance with supplementation, it is logical that there are reports of adverse events following noncompliance with supplementation. For example, there is a case report demonstrating neonatal vitamin A deficiency with maternal night blindness during the third trimester associated with refusal of nutritional treatment during pregnancy following BPD.51

Studies on pregnancy following bariatric surgery rarely describe the specific supplementation regimens employed; however, some case reports document adverse maternal or neonatal nutritional outcomes following poor compliance with supplementation. Two studies report neural tube defects in patients who underwent gastric bypass prior to pregnancy. Each of these studies, one in Maine, and the other in Iowa, found three neonates affected by neural tube defects; none of the six pregnant women were compliant in taking nutritional supplements.52, 53 Unfortunately, the amount of folic acid included in the nutritional supplements was rarely mentioned in the reports of studies of bariatric surgery. Also, among two gastric bypass case reports, one report documented failure to thrive in a neonate, which was thought to be caused by low fat content in the maternal breast milk, and a second case report documented neonatal vitamin B12 deficiency. Neither of the women in these cases was compliant with their recommended supplementation.54, 55 There are also case reports of maternal and neonatal nutritional deficiencies during pregnancy in patients following gastric bypass despite compliance with supplementation. One case study reported on neonatal B12 deficiency despite the use of prenatal vitamins during pregnancy and lactation.56 Another case report documented iron and vitamin B12 deficiencies starting at six weeks gestation; the patient required parenteral B12 and blood transfusion due to the anemia being refractory to parenteral iron.57

We conclude that published reports of adverse nutritional outcomes in pregnant patients who underwent gastric banding or gastric bypass and subsequently received standard nutritional supplementation are rare. There are more reports of severe malabsorption and nutritional deficiencies following BPD, occasionally requiring parenteral nutrition in pregnant patients. Although some maternal and neonatal adverse events occurred among women who had undergone bariatric surgery even with maternal vitamin supplementation, many of these adverse outcomes were attributed to maternal noncompliance with vitamin supplementation. An important caveat is that, in general, vitamin, mineral, and trace element levels were not monitored in mother or baby, and if clinical manifestations of these deficiencies are subtle and thus difficult to detect, they may be higher than reported.

Adjustable Gastric Band Management

There is no consensus on band management in pregnancy following gastric band placement; in fact, while there were studies that either deflated the band or did not, no studies compared different methods of band management. Studies that systematically deflated the bands early in the pregnancy did so in order to allow for optimal nutrition during fetal development and to decrease vomiting in the first trimester.33, 58, 59 Other studies, which did not deflate the band routinely, did so only if there were symptoms of nausea and vomiting, or by request of the patient. We identified three case series of pregnant women who had received adjustable gastric banding. Among 67 potentially fertile women who had the procedure performed at a single institution, 21 women had 25 pregnancies, of which 18 of these went to term. Deflation of the band was required in two women (11 percent) for nausea and vomiting.36 In another single-institution study, 49 pregnancies in 44 women were identified from a database of all women who received adjustable band surgery. Eight women (18 percent) required band deflating.60 Finally, among 359 women who enrolled in two clinical trials of adjustable gastric banding, 256 were fertile, and in this group, there were 20 women with 23 pregnancies. Among the 18 deliveries, six (33 percent) had adjustments to the band; three patients required it for nausea and vomiting, one patient had band deflation “to prevent vomiting,” and two patients had band adjustment or removal at their own request.34, 61 Across all three studies, with 84 deliveries, 16 patients (19 percent) had adjustment or removal of their band.34, 36, 61 In a large case series, two women presented in the second trimester with severe vomiting, dehydration, and electrolyte abnormalities and were found to have band slippage; the band was removed in both patients, without any further complications.62 We judge the evidence is insufficient to reach conclusions regarding band management in pregnant patients.

How Long To Delay Pregnancy After Surgery

Expert opinion is that patients should not attempt pregnancy within the period of rapid weight loss (first year) following bariatric surgery. We identified little published evidence that assessed the evidence on this issue. One letter to the editor reported on 18 women who had 21 babies after gastric bypass surgery; ten of these women conceived within the first year after surgery.63 The authors state that there were no statistical differences between babies conceived within the first year and those conceived later on with respect to rates of cesarean section, other delivery complications, neonatal jaundice, low birth weight, or congenital abnormalities. However, the small sample size limited the study's power to detect anything other than very large differences between the two groups. Furthermore, the results of statistical tests were not reported in the letter. Another study compared 21 pregnancies beginning within the first year following gastric bypass to 13 that began later.45 Again, no statistically significant differences were found between groups, and again, the small sample sizes limits the conclusions that can be drawn. In a study comparing birth outcomes in women before and after laparoscopic adjustable gastric banding, the authors report on the 20 pregnancies (out of a total of 79 that were included) where conception occurred within the first year after the procedure.43 While maternal weight gain during pregnancy was lower in these 20 pregnancies than in pregnancies occurring later, the birth weight of babies did not differ, and there were no statistically significant differences in complications of pregnancy or preterm deliveries. Another study investigated the characteristics of pregnancies that occurred within 18 months of BPD compared to pregnancies after 18 months postoperatively; a higher rate of spontaneous abortion was seen in the early group (31 percent vs. 18 percent). However, birth weights showed no difference.64 One study investigating pregnancies within the first 2 years after gastric bypass found a high rate of premature births (18 percent), but, unfortunately, this study had no comparison group.44 When extending the period of time to 2 years following adjustable gastric banding, a study found an increased spontaneous abortion rate of 29 percent as well as two band-related complications, including slippage and balloon leakage. As there was no comparison group in this study, nothing was concluded regarding relative birth weights or complications.58 Lastly, a case report documented a pregnancy that was determined to have begun one day prior to gastric stapling surgery. Although the woman experienced minor dehiscence of the gastric wound by endoscopy and minor liver enzyme elevation, the remainder of the pregnancy and birth were uneventful.65 We conclude there is scant evidence of pregnancy outcomes upon which to make recommendations about how long to delay pregnancy following surgery.

Key question 6. For women who have had bariatric surgery, what is the evidence for morbidity and mortality risks for: a) mother and b) neonate?

Laparoscopic Adjustable Gastric Band

We identified two cohort studies and five case series that reported on morbidity and mortality outcomes following laparoscopic adjustable gastric banding. These studies are summarized in Tables 4 and 5. In the two cohort studies, samples sizes of pregnancies were small (22 and 79). Both studies compare pregnancy outcomes in the same women before and after laparoscopic adjustable band placement, and also to community outcomes or to the outcomes of obese patients who did not undergo bariatric surgery. Before surgery, patients had rates of pregnancy complications such as gestational diabetes, preeclampsia, and hypertension that were similar to other obese pregnant women. After laparoscopic adjustable gastric band placement, the rates of these pregnancy complications were similar to rates seen in the community. However, due to small sample sizes, it is not possible to determine whether the rates of these complications may still be elevated following bariatric surgery. For example, in the study by Dixon,43 the rate of gestational diabetes in postsurgery pregnancies was 6.3 percent (as compared to 15 percent in presurgery pregnancies), while in the community it is 5.5 percent. The difference in these two rates is - 0.8 percent, but the 95-percent confidence interval of the difference is - 6.8 percent to 5.2 percent. This means that the rate of gestational diabetes in past surgery pregnancies could conceivably still be twice as high as community rates. Therefore, it is premature to conclude that bariatric surgery reduces the rates of these complications to those of the average woman. One stillbirth and one case of duodenal atresia occurred in pregnancies following bariatric surgery; sample sizes were too small to draw conclusions. The five case series articles included 141 pregnancies in total. Rates of pregnancy complications were low. These data support the cohort data that rates of pregnancy complications following laparoscopic adjustable gastric band placement are low. Data are insufficient to comment on rare outcomes.

Table 4. Cohort studies reporting morbidity and mortality with laparoscopic adjustable band.

Table 4

Cohort studies reporting morbidity and mortality with laparoscopic adjustable band.

Table 5. Case series reporting morbidity and mortality with laparoscopic adjustable band.

Table 5

Case series reporting morbidity and mortality with laparoscopic adjustable band.

Of note, we identified one case report following open nonadjustable gastric band where the woman developed severe vomiting secondary to pouch outlet obstruction.66 Subsequent weight loss led to significant fetal growth retardation, and enteral nutrition via feeding tube was required to normalize the weight gain for the fetus. Following delivery, the women's outlet obstruction resolved.

Gastric Bypass and Vertical Banded Gastroplasty

We identified one case-control study, six cohort studies and six case series that reported on morbidity and mortality outcomes following gastric bypass or vertical banded gastroplasty. Some studies reported combined outcomes for patients who had received gastric bypass or some other bariatric procedure; those studies are included here.

The case-control study included women who had received bariatric surgery and surveyed them on their pregnancy histories and outcomes.67 Authors compared postoperative pregnancies (n=57) to presurgery pregnancies (n=57) that were matched for presurgery weight, age, parity at index pregnancy, and delivery year. They reported no difference in gestational diabetes (3 cases in postsurgery pregnancies compared to 6 in presurgery control group, p=NS), less hypertension, which included chronic and pregnancy-induced (9 percent v 46 percent, p<0.001). The average neonatal birth weight was lower in the postsurgery group (3,205 versus 3,604 grams, p<0.001) and there were less large-for-gestational-age neonates (16 percent versus 36 percent, p<0.02). They found no statistical difference in small for gestation age neonates (4 vs 2, p=NS), premature deliveries at <37 weeks (7 vs 4, p=NS), or perinatal deaths (4 vs 4, p=NS).

The cohort studies are summarized in Tables 6 and 7. Like the data for the laparoscopic adjustable gastric band, the findings of these cohort studies suggest that rates of gestational diabetes, preeclampsia, and hypertension are decreased following bariatric surgery. As with the laparoscopic adjustable band data, the number of cases studied is relatively small—across all studies, data are reported on a total of 188 pregnancies following surgery. There were no differences seen in average birth weight, proportion with low birth weight, or premature delivery between babies born before or after bariatric surgery. In the only article to report data, the proportion of babies with congenital malformations was 7.1 percent in patients with gestational diabetes after bariatric surgery, compared to 4 percent in pregnant women with gestational diabetes who did not receive bariatric surgery. The sample sizes were too small to draw conclusions.

Table 6. Cohort studies reporting morbidity and mortality with gastroplasty and gastric bypass.

Table 6

Cohort studies reporting morbidity and mortality with gastroplasty and gastric bypass.

Table 7. Case Series reporting morbidity and mortality with gastroplasty and gastric bypass.

Table 7

Case Series reporting morbidity and mortality with gastroplasty and gastric bypass.

Table 8. Cohort Studies reporting morbidity and mortality for biliopancreatic diversion.

Table 8

Cohort Studies reporting morbidity and mortality for biliopancreatic diversion.

Table 9. Case Series reporting morbidity and mortality for biliopancreatic diversion.

Table 9

Case Series reporting morbidity and mortality for biliopancreatic diversion.

The six case series studies reported on about 300 pregnancies in total. The most notable finding was that two case series reported more babies with neural tube defects than expected. One report was about three women who had four pregnancies, all of which had neural tube defects,53 and the other report was a case series of 110 pregnancies in 87 women who had received gastric bypass; three babies had neural tube defects.

Biliopancreatic Diversion

We identified two cohort studies and four case series that reported on morbidity and mortality following biliopancreatic diversion. The BPD cohorts were larger than either the laparoscopic adjustable gastric band or gastric bypass cohorts - a total of 490 pregnancies were reported. In the first cohort, out of 1,136 women who had received BPD over the past 18 years, 129 women had 239 pregnancies, of which 152 were brought to term.47 The authors report that 32 patients (21 percent of pregnancies) required parenteral nutritional support, including 10 patients who needed to be hospitalized. One woman developed pregnancy-induced hypertension. Before BPD surgery, among these 129 women, there were seven cases of preeclampsia, three cases of gestational diabetes, and two cases of pregnancy-induced hypertension. In pregnancies after BPD, 40 newborns (26 percent) were small for gestational age, although all had good Apgar scores. Four babies died at or shortly after birth, two for unknown reasons and one each after surgery to try to correct diaphragmatic eventration and meconium obstruction. In the other cohort study, a mailed survey to 916 women who had received BPD surgery yielded 783 responses. There were 251 pregnancies in 132 women, resulting in 166 infants born to 109 mothers. Compared to pregnancies prior to surgery, the number of miscarriages remained about the same (21.6 percent prior to surgery compared to 26.0 percent after). There was no significant difference in the incidence of stillbirths or malformation.29

The four case series were small, reporting on a total of 108 pregnancies. The most notable was a report of nine adverse neonatal outcomes all associated with severe nutritional deficiencies.48 Another reported that out of 84 pregnancies after BPD, 21 were voluntarily aborted and another ended in miscarriage.35 They report that parenteral nutrition is safe and could be of benefit for mother and infant.46

The number of reports of severe nutritional problems during pregnancy is higher for BPD than for the other types of bariatric surgery. Since data for all procedures come from a limited number of providers and patients, extrapolation of the results from such limited samples to the larger population may not be justified. Still, the increased number of reports of severe nutritional deficiencies in pregnant women is consistent with the greater degree of malabsorption caused by BPD relative to the other bariatric procedures.

We identified two studies that reported on growth and development of children born to mothers who had undergone BPD. The first study was reported68 only in abstract form and consisted of a large case series from a single practice. This study, which included data on 100 pregnancies (from among 2,500 patients having had bariatric surgery), found that some of the children were now over 10 years of age, and found no statistical differences in development. No additional details were reported. The second report compared 172 children, aged 2 to 18 years, who were born to 113 obese mothers who had undergone BPD surgery to 45 same-age siblings born to these mothers prior to surgery, and also compared the outcomes to current population standards. On average, the mothers had a presurgery BMI of 48 and a postsurgery BMI of 31. All patients underwent surgery at the same center. The authors found that the proportion of children who were overweight or obese was much higher for those born to the mother prior to bariatric surgery (60 percent versus 35 percent). The proportion of children who were underweight did not differ statistically between those born before or after maternal bariatric surgery (4.4 percent vs. 7.5 percent, p=0.742).70

Maternal and neonatal surgical adverse events following bariatric surgery

While there are many potential benefits of bariatric surgery for women considering future pregnancy, there are also risks. We identified over a dozen reports of complications requiring surgical intervention during pregnancy following bariatric surgery, many with deleterious effects for the neonate and mother. There were 13 case reports that described 14 complications requiring surgical intervention (one report discussed two patients): eight small bowel obstructions due to internal hernia, two mid-gut volvulus (one from adhesions), one perforated gastric ulcer, two band complications (including erosion and bleeding), and one staple line stricture.7183 Eight of these bariatric procedures were performed laparoscopically and six were performed in an open fashion. Time from surgery to pregnancy ranged from 1.5 to 108 months (median=24 months).

Most women presented with nonspecific abdominal pain, nausea, and/or vomiting; in fact, two patients were treated for pancreatitis days prior to the correct diagnosis. Given the vague, rather common symptoms, there were often delays of up to several days prior to surgical intervention. Eight patients ultimately underwent Computed Tomography (CT), which often prompted surgery immediately upon receipt of the results. Several patients were in florid septic shock at the time of the intervention.

At the time of the adverse event, gestational age ranged from 25 to 36 weeks (median = 29.5 weeks). Most required urgent surgical intervention. Examples of findings at exploration included pulling-through of sutures placed to close a mesenteric defect, internal hernia through Peterson's Space, and band erosion. An emergent cesarean section or premature rupture of membranes occurred in six of 13 cases (46 percent). Overall five of 13 (38.5 percent) neonates died (one natal outcome was not reported). Five were delivered at full-term. There were three maternal deaths (21.4 percent).

It is not known if pregnancy increases risk for surgical complications, as these types of reports are also found for nonpregnant patients and there are relatively uncommon. However, the maternal and fetal mortality and morbidity rates associated with these types of complications are high. Early diagnosis and surgical intervention are key to addressing bowel obstructions, band erosions, and gastrointestinal hemorrhage. Although the data are retrospective and self-reported, CT scan was helpful in diagnosing many of the patients with internal hernias. These data suggest that these types of complications should be considered early in the work-up of pregnant women who present with signs and symptoms of intestinal obstruction, perforation, or hemorrhage.

Key question 7. What is the evidence that cesarean section for women who have had bariatric surgery affects the risks of morbidity and mortality for: a) mother and b) neonate?

One case-control study assessed effects of cesarean section on women who have had bariatric surgery.67 Postoperative pregnancies (n=57) were compared to presurgery pregnancies (n=57) matched for presurgery weight, age, parity at index pregnancy, and delivery year. They reported no difference in rate of primary cesarean section (7 patients vs 8 patients, p=NS) or rate of repeat cesarean section (7 vs 1, p=0.07). For the observational studies, the rates of cesarean section vary greatly from study to study and the difference between postsurgery deliveries and presurgery also varied with some showing higher rates, others lower, and other no difference. In one study, bariatric surgery was found to be an independent risk factor for cesarean delivery.30 However, the author indicated there is no known physiologic reason for performing more cesarean deliveries among patients who have had bariatric surgery, and it is difficult to ascribe the increased use of cesarean delivery in patients with previous bariatric surgery to anything but caregiver bias. In general, most of the observational studies reporting on cesarean section rates did not have details on number of previous cesarean sections, breech position of fetus, or maternal choice in delivery method, thus making comparisons with the population difficult.

The case-control study reported on specific delivery complications.67 They found no differences in need for transfusions (5 patients versus 1 patient, p=NS), peripartum need for intravenous antibiotics (4 versus 4, p=NS), or peripartum thromboembolic events requiring heparin (1 versus 1, p=NS). Blood transfusions were associated with cesarean section in 3/5 cases in the postsurgery group and 1/1 for the control group. Receipt of antibiotics was associated with cesarean section in 2/4 cases for postoperative group and 3/4 for the control.

We did not find any data specifically looking at possible operative injury complications following cesarean section.



Appendixes cited in this report are provided electronically at http://www​.ahrq.gov/clinic/tp/barireptp​.htm

Cover of Bariatric Surgery in Women of Reproductive Age
Bariatric Surgery in Women of Reproductive Age: Special Concerns for Pregnancy.
Evidence Reports/Technology Assessments, No. 169.
Maggard M, Li Z, Yermilov I, et al.

AHRQ (US Agency for Healthcare Research and Quality)

PubMed Health Blog...

read all...

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...