Births After Bariatric Surgery in the United States: Incidence, Obstetric Outcomes, and Reinterventions

Structured Abstract

Objective:

To characterize incidence and outcomes for bariatric surgery patients who give birth.

Summary Background Data:

Patients of childbearing age comprise 65% of bariatric surgery patients in the United States, yet data on how often patients conceive and obstetric outcomes are limited.

Methods:

Using the IBM MarketScan database, we performed a retrospective cohort study of female patients ages 18–52 undergoing laparoscopic sleeve gastrectomy or Roux-en-Y gastric bypass from 2011–2017. We determined incidence of births in the first two years after bariatric surgery using Kaplan-Meier estimates. We then restricted the cohort to those with full two-year follow-up to examine obstetric outcomes and bariatric-related reinterventions. We reported event rates of adverse obstetric outcomes and delivery type. Adverse obstetric outcomes include pregnancy complications, severe maternal morbidity, and delivery complications. We performed multivariable logistic regression to examine associations between birth and risk of reinterventions.

Results:

Of 69,503 patients who underwent bariatric surgery, 1,464 gave birth. The incidence rate was 2.5 births per 100 patients in the 2 years after surgery. 85% of births occurred within 21 months after surgery. For 38,922 patients with full two-year follow-up, adverse obstetric event rates were 4.5% for gestational diabetes and 14.2% for hypertensive disorders. 48.5% were first-time Cesarean deliveries. Almost all reinterventions during pregnancy were biliary. Multivariable logistic regression analysis showed no association between post-bariatric birth and reintervention rate (OR: 0.93, 95%CI: 0.78–1.12).

Conclusions:

In this first national U.S. cohort, we find giving birth was common in the first 2 years after bariatric surgery and was not associated with increased risk of reinterventions. Clinicians should consider shifting the dialogue surrounding pregnancy after surgery to shared decisionmaking with maternal safety as one component.

Mini Abstract

Patients of childbearing age comprise 65% of bariatric surgery patients in the United States, yet rates of birth and adverse obstetric outcomes are unknown. Using the IBM MarketScan database, Kaplan-Meier estimates determined an incidence of 2.5 births per 100 female patients in the 2 years after surgery. Giving birth within this purported high-risk timeframe was also not associated with reinterventions.

Introduction:

Patients of childbearing age comprise 65% of bariatric surgery patients in the United States every year.¹ Current consensus guidelines put forth by the American College of Obstetricians and Gynecologists and endorsed by the American Society for Metabolic and Bariatric Surgery recommend patients delay conception by 12–24 months after bariatric surgery due to possible risks to the mother and fetus.² Despite these recommendations, we know little about the incidence of pregnancy and adverse birth outcomes for patients who become pregnant in this proposed higher-risk time frame.

To date, the largest studies of pregnancy after bariatric surgery have only addressed fetal outcomes. Maternal outcomes research after bariatric surgery is limited in a number of ways. Many studies are single center with small numbers, which may lead to biased results.^3–5 Larger, population-based studies examined European^{6, 7} or Australian patients⁸ whose outcomes may not be generalizable to patients in the United States. Patients in the United States have higher pregnancy rates,⁹ greater racial and ethnic disparities in unintended pregnancies,¹⁰ and greater racial/ethnic diversity.¹¹ Lastly, among studies examining maternal outcomes, sleeve gastrectomy, currently the most common bariatric surgery performed,¹² is often missing.¹³

Healthcare providers lack crucial data to counsel reproductive-age patients about when it is safe to conceive after bariatric surgery, including potential obstetric risks for patients and their rates of bariatric-related surgical complications. To address these major gaps, we used a large, national claims database to identify the birth rate after laparoscopic sleeve gastrectomy or Roux-en-Y gastric bypass and to examine obstetric and bariatric surgical outcomes in the two years after their bariatric procedure.

Methods:

Data Source

We performed a retrospective cohort study using IBM MarketScan Commercial, a multi-payer administrative dataset with inpatient, outpatient, and physician claims.¹⁴ This database includes more than 40 million patients employed at large institutions from all 50 states. We examined claims from January 1, 2011 to two years after bariatric surgery or December 31, 2018.

Study Population

We included adult female patients of childbearing age who underwent laparoscopic sleeve gastrectomy or laparoscopic Roux-en-Y gastric bypass, the two most commonly performed bariatric procedures,¹² from January 1, 2011 to December 31, 2017. Patients were included if they had at least 6 months of continuous follow-up after surgery. Childbearing age was defined as ages 18 to 55 years.^15–17 While 55 is the upper range for childbearing age, we chose 52 years as the limit for our cohort to accommodate full-term pregnancy (40 weeks) and the two-year period after bariatric surgery. We identified bariatric patients using CPT codes 43775, 43644, 43645, 43844, 43846+43659, 43847+43659 for laparoscopic sleeve gastrectomy or laparoscopic Roux-en-Y gastric bypass when they also had ICD-9 and ICD-10 diagnosis codes for morbid obesity. Patients were excluded if they had a diagnosis code reflecting gastric or small bowel cancer associated with surgery to ensure our cohort only included patients undergoing bariatric surgery.

This cohort was used to examine the epidemiology of birth after bariatric surgery. Because we then used logistic regression to examine obstetric outcomes and reinterventions after birth, we further restricted the cohort to patients with full 2-year follow-up in order to account for differential follow-up lengths.

Outcomes

Births

Births were identified by ICD-9 diagnosis code (V27), ICD-10 diagnosis code (Z37), or diagnosis related group codes (765–768, 774, 775, 783–788, 796–798) for still births, live births, cesarean deliveries, and vaginal deliveries. Our primary outcomes were the incidence of births in the first 2 post-operative years and the proportion of these births occurring before 21 months. The mark of 21 months was chosen as this was thought to approximate conception at 12 months, the least conservative recommendation for conception after bariatric surgery, plus 9 months, the usual length of pregnancy. Births rather than pregnancy were used given the more accurate and complete capture of births in claims data. In secondary analyses we examined birth rate across subgroups defined by procedure type and patient age.

Obstetric Outcomes

We then examined obstetric outcomes by procedure type. We first identified obstetric outcomes for the first birth for each patient within the first two years after bariatric surgery. Obstetric outcomes were defined by CPT, ICD-9, or ICD-10 codes based on a review of the literature (Supplemental Digital Content, Table 1). Births were categorized as still births or live births.

Obstetric outcomes include pregnancy complications (gestational diabetes, gestational hypertension, mild/unspecified pre-eclampsia, severe pre-eclampsia, eclampsia, iron deficiency anemia, hyperemesis gravidarum), severe maternal morbidity (postpartum hemorrhage, peripartum hysterectomy, major puerperal infections, anesthesia complications) and delivery complications (PPROM/pre-term delivery, prolonged labor). Patients with more than one hypertensive disorder (gestational hypertension, mild pre-eclampsia, severe pre-eclampsia, and eclampsia) were classified as having only the most severe disorder to avoid counting patients multiple times. We additionally describe rates of delivery types: vaginal, operative, cesarean, vaginal after prior cesarean, and repeat cesarean.

Bariatric Surgery-Related Reinterventions

We examined bariatric reintervention rates in patients who gave birth compared to those who did not. Reintervention CPT codes were obtained from a review of the literature and CPT coding manual in line with the bariatric surgery comparative effectiveness literature.^18–21 (Supplemental Digital Content, Table 2) Reinterventions were categorized as follows:

1. Revision: Operation modifying the index bariatric procedure including included conversion to another bariatric procedure (e.g. sleeve gastrectomy to gastric bypass), any gastrectomy, or anastomotic revision.

2. Enteral access: Surgical, endoscopic, or interventional.

3. Vascular Access: Procedure to reflect the need for parenteral nutrition.

4. Reoperation: Other abdominal operation potentially a sequela of the index bariatric procedure but not affecting bariatric physiology. We examine reoperations as a group as well as separately which include abdominal wall hernia repair, biliary procedure, internal hernia repair, paraesophageal hernia repair, and other abdominal procedure. We decided to examine reoperations separately also because of various physiologic changes of pregnancy which could affect the likelihood of each of these. For example, anatomic changes could increase intra-abdominal pressure and thus the risk of an abdominal wall hernia whereas the mechanism of pregnancy causing increased risk of biliary procedures would be through hormonal changes.

5. Other reintervention: Drainage, aspiration, diagnostic laparoscopy, etc.

Statistical Analysis

We first determined the incidence rate of births in the first two years after bariatric surgery using Kaplan-Meier estimates. We then examined the distribution of births over time and proportion of births occurring before 21 months. Next, we determined birth rates by procedure type (laparoscopic sleeve gastrectomy versus laparoscopic Roux-en-Y gastric bypass) using Cox regression analysis, controlling for age, U.S. state,²² and Elixhauser comorbidities.^{23, 24} Age was treated as a categorical variable: 18–24, 25–29, 30–34, 35–39, 40–44, ≥45 years. U.S. state was the smallest geographic unit available for all patients. Finally, we conducted a Cox regression analysis of the association between age ranges and likelihood to give birth, adjusting for U.S. state and Elixhauser comorbidities. For all Cox regression analyses, we censored patients who did not have a qualifying event at the time of disenrollment from the plan or at 2 years after bariatric surgery.

To investigate obstetric outcomes for the first birth after bariatric surgery, we reported event rates for all patients who gave birth within the first two years after bariatric surgery. For bariatric reinterventions, we utilized the restricted cohort in which all patients had full 2-year follow-up time. Using multivariable logistic regression, we examined the association between giving birth and reinterventions over the entire the study period. In our analysis, we controlled for age, U.S. state, and Elixhauser comorbidities. For specific reinterventions with higher unadjusted event rates for women who gave birth compared to those who did not, we performed multivariable logistic regression adjusting for the same variables.

All statistical tests were two-sided and performed at the 5% significance level. All analyses were performed using Stata 16 (StataCorp LLP, College Station, TX). The study was exempt from Institutional Review Board review due to use of de-identified data.

Results:

Of 69,503 patients of childbearing age who underwent laparoscopic sleeve gastrectomy or Roux-en-Y gastric bypass from 2011–2017, 87% had full 1-year and 56% had full 2-year follow-up. Patient characteristics are included in Table 1. The median age of the entire cohort was 40 years (IQR: 34–46). Patients who gave birth were younger than patients who did not give birth. Patients who gave birth were less likely than patients who did not to have hypertension (29.0% vs 40.3%, P<.001), neurologic disorders (0.4% vs 1.1%, P=0.012), hypothyroidism (7.5% vs 9.4%, P=0.014), depression (11.0% vs 13.3%, P=0.010), and rheumatoid arthritis (0.3% vs 1.1%, P=0.006). There were no other observed differences in comorbidities, including diabetes with chronic complications (1.7% vs 1.6%, P=0.736). The cohort with full 2-year follow-up had 38,922 patients. This restricted cohort had a similar distribution of procedure type and comorbidities.

Table 1:

Baseline Characteristics of Patients of Childbearing Age Undergoing Bariatric Surgery 2011–2017

	Patients Who Did Not Give Birth	Patients Who Gave Birth	P value
	n = 68,039 (97.9%)	n = 1,464 (2.1%)
Age at time of surgery, mean (SD)	39.6 years (8.2)	30.6 years (5.3)	<.001
Age at time of surgery, No. (%)
18–24 years	3,730 (5.5%)	207 (14.1%)	<.001
25–29 years	4,995 (7.3%)	396 (27.0%)	<.001
30–34 years	9,487 (13.9%)	520 (35.5%)	<.001
35–39 years	13,163 (19.3%)	291 (19.9%)	0.61
40–45 years	14,794 (21.7%)	47 (3.2%)	<.001
45–52 years	21,870 (32.1%)	6 (0.4%)	<.001
Year of operation, No. (%)
2011	10,268 (15.1%)	235 (16.0%)	0.31
2012	10,725 (15.8%)	245 (16.7%)	0.31
2013	11,187 (16.4%)	224 (15.3%)	0.24
2014	10,173 (15.0%)	233 (15.9%)	0.31
2015	8,200 (12.1%)	196 (13.4%)	0.12
2016	8,478 (12.5%)	215 (14.7%)	0.01
2017	9,008 (13.2%)	116 (7.9%)	<.001
Comorbidities*, No. (%)
Hypertension	27,428 (40.3%)	424 (29.0%)	<.001
Chronic pulmonary disease	6,095 (9.0%)	124 (8.5%)	0.52
Diabetes without chronic complications	12,076 (17.8%)	236 (16.1%)	0.11
Hypothyroidism	6,406 (9.4%)	110 (7.5%)	0.01
Liver disease	7,815 (11.5%)	163 (11.1%)	0.68
Depression	9,057 (13.3%)	161 (11.0%)	0.01

Note: Percentages are reported by column.

^*Elixhauser comorbidities with a prevalence greater than 5% are listed.

Incidence of Births

Within the first 2 years after surgery, 1,464 (2.1%) patients gave birth. Five patients had two births within this time period. There were 17 still births and 1,452 live births. Examining time to first birth, the incidence rate was 0.0000347 births per person-day, or 0.025 births per 730 person-days. Thus, for every 100 patients of childbearing age, the incidence rate was 2.5 births in the first two years after bariatric surgery. The distribution of time to first birth by months after bariatric surgery is shown in Figure 1. Thirty-two births (2.2%) occurred prior to post-operative month 9, corresponding to conception during the first month after surgery. The majority of births (85%), occurred before 21 months after bariatric surgery, corresponding to conception during the first 12 months after surgery.

Figure 1:

Cumulative Distribution of Births Over Time after Bariatric Surgery

The number of births within every three months after bariatric surgery is shown. The total cumulative number of first births is 1,464 for women who underwent bariatric surgery from 2011–2017. The dashed line corresponds to births that would have had a conception date before 12 months after bariatric surgery.

Factors Associated with Birth

We examined the association between procedure type and age group with likelihood to give birth. Cox regression analysis showed no association between procedure type and likelihood to give birth (HR: 0.92, 95%CI: 0.82–1.03). The age group with the highest likelihood of birth was age 25–29 (HR: 1.46, 95%CI: 1.24–1.73, ref: age 18–24, Supplemental Digital Content, Table 3). The group least likely to give birth was age 45–52 (HR: 0.004, 95%CI: 0.0002–0.010).

Obstetric Outcomes

For the 1,079 patients with full two-year follow-up who gave birth, we examined rates of obstetric outcomes which are shown in Figure 2. Event rates for all obstetric outcomes are found in Table 2. Of the patients who gave birth there were 496 (46.0%) vaginal deliveries, 21 (2.0%) operative deliveries, 523 (48.5%) cesarean deliveries, 12 (1.1%) vaginal deliveries after prior cesarean delivery, and 27 (2.5%) repeat cesarean deliveries.

Figure 2:

Obstetric Outcome Event Rates for Women Who Gave Birth Within 2 Years After Bariatric Surgery

Event rates for women who gave birth within the first two years after bariatric surgery. The event rates for eclampsia, prolonged labor, peripartum hysterectomy, and major puerperal infections were all 0%. The event rates for hyperemesis gravidarum was 0.1%, for prolonged labor was 0.2%, and for anesthesia complication was 0.5%.

Table 2:

Event Rates for Obstetric Outcomes in Patients Who Gave Birth 2 Years After Bariatric Surgery

	Patients Who Underwent Bariatric Surgery (n=1,079)
Outcome	%
Gestational Diabetes	4.5%
Hypertensive Disorders of Pregnancy	14.2%
Gestational Hypertension	7.7%
Mild-unspecified Pre-eclampsia	3.5%
Severe Pre-eclampsia	3.3%
Eclampsia	0%
Iron Deficiency Anemia	1.1%
Hyperemesis Gravidarum	0.1%
Postpartum Hemorrhage	1.8%
Peripartum Hysterectomy	0%
Major Puerperal Infections	0%
Anesthesia Complications	0.5%
PPROM/Pre-term Labor	9.6%
Prolonged Labor	0.2%

Bariatric Reinterventions

There were 38,922 patients with full 2-year follow-up. Of these, 1,079 patients gave birth. Among patients who did not give birth, there were 4,575 (12.1%) patients who underwent a reintervention during the study period compared to 146 (13.5%, P=0.15) patients who gave birth. Of patients who gave birth and experienced a reintervention, 82 (56%) had a first reintervention before giving birth, 15 (10%) during pregnancy (9 months before birth), 3 (2%) the same day of birth (all abdominal wall hernia repairs), 4 (3%) within 6 months postpartum, and 42 (29%) patients had their first reintervention beyond 6 months postpartum. For reinterventions during pregnancy, there were 11 biliary operations, 1 vascular access procedure, 1 enteral access procedure, and 1 other operation. Table 3 shows unadjusted rates of each type of reintervention in patients who did not give birth and patients who gave birth. Multivariable logistic regression showed patients who gave birth were not more or less likely to experience a reintervention compared to patients who did not give birth (OR: 0.93, 95%CI: 0.78–1.12). Giving birth was also not associated with likelihood to undergo biliary procedures (OR: 1.15, 95%CI: 0.93–1.42).

Table 3:

Unadjusted Rates of Reinterventions in Patients After Bariatric Surgery with Full 2-Year Follow-up

	No Birth (n=37,843)	Gave Birth (n=1,079)	P value*
Outcome
Revisions	132 (0.3%)	1 (0.1%)	0.28
Enteral Access	62 (0.2%)	2 (0.2%)	0.86
Vascular Access	381 (1.0%)	8 (0.7%)	0.39
Reoperation
Abdominal Wall Hernia Repair	495 (1.3%)	8 (0.7%)	0.10
Biliary Procedure	2,203 (5.8%)	102 (9.5%)	<.001
Internal Hernia Repair	102 (0.3%)	1 (0.1%)	0.54
Paraesophageal Hernia Repair	116 (0.3%)	2 (0.1%)	0.78
Other Abdominal Procedure	930 (2.5%)	19 (1.8%)	0.14
Other Reintervention	154 (0.4%)	3 (0.3%)	0.81

^*P values are shown for bivariate comparisons using Chi-square test if n > 5 or Fisher Exact test if n ≤ 5.

Discussion:

This is, to our knowledge, the first national cohort of commercially-insured patients of childbearing age in the United States. Despite consensus guidelines outlining the potential serious adverse effects of conception close to bariatric surgery on mothers,² our findings show a relatively high incidence of births after bariatric surgery (2.5 births per 100 patients), at approximately half the birth rate of the general U.S. population (5.9 births per 100 patients).²⁵ Overall, our data provide preliminary evidence that giving birth after bariatric surgery may be safe for mothers. Obstetric outcomes were generally comparable to U.S. population-level estimates, and there was no association between giving birth and having a bariatric reintervention.

Our study is the first to identify the birth rate of patients after bariatric surgery in the United States. At 2.5 births per 100 patients, this rate is relatively high considering current national recommendations to avoid conception during this purported high-risk period. Lack of concordance with guidelines brings up questions regarding efficacy of pre-conception counseling as well as alignment with patient preferences. Previous work in the Longitudinal Assessment of Bariatric Surgery-2 cohort showed pre-operatively rating pregnancy as important was associated with conception within 18 months.²⁶ Researchers using the same cohort also found one third of patients ages 18–44 years who intended to become pregnant planned to do so within 2 years of bariatric surgery.²⁷ 4.3% of patients actively tried to conceive within the first year.²⁶ Thus, pregnancy may be something patients value highly and may outweigh theoretical risks related to giving birth closely after bariatric surgery. Additionally, some patients may not be counseled that one of the effects of weight loss is increased fertility.

We only included full-term pregnancies in our study due to the limitations of identifying early pregnancy losses in claims data. Thus, the results of our study only explore complication rates in pregnancies resulting in birth and cannot be used to draw conclusions about overall fertility rates, pregnancy loss, or complications in early pregnancy losses. Data have shown mixed results for patients who undergo bariatric surgery regarding fertility, with some showing reduction in fertility²⁸ and others showing increased fertility.^29–31 Studies examining miscarriage rates also show mixed results, though data are from small studies.^{31, 32} Future work is needed to better understand this important period in pregnancy. Additionally, the method of conception (natural conception, medication-assisted conception, in vitro fertilization, etc) cannot be assessed in this dataset, potentially underestimating pregnancy complications such as preterm birth, placental disorders, and preeclampsia.

Our findings suggest that in patients with a history of bariatric surgery who give birth two years after their procedure, there is not an increased risk of surgical reinterventions, and obstetric outcomes are equivalent. We found patients who gave birth were not more likely than those who did not to undergo surgical reinterventions. Even when comparing surgical techniques with different risks, sleeve gastrectomy and gastric bypass, we found no elevated risk with procedures thought to cause more disruption to intra-abdominal anatomy. A recent study from Auger et al in a Canadian cohort also found no elevated obstetric risks in severe preeclampsia, sepsis, cardiac complications, and other morbidities for patients giving birth after bariatric surgery compared with patients without obesity.³³

To place our obstetric findings into context, we turn to rates established in the literature. Gestational diabetes and pre-eclampsia rates are reported to be 11.2%³⁴ and 9.1%³⁵ for women with BMI ≥ 35kg/m², respectively. Interestingly, the rate of iron deficiency anemia in the general U.S. population is estimated to be 18%³⁶ regardless of BMI. Our findings of lower rates of iron deficiency anemia could indicate appropriate monitoring and treatment of pregnant patients after bariatric surgery or a limitation of claims data. Previous work has shown higher BMI is associated with more severe maternal morbidity and mortality.³⁷ Thus, given more time to work, bariatric surgery may reduce obesity-related risks in pregnancy even further. Of note, a study using 113,818 U.S. patients with BMI ≥ 35 kg/m² by Butwick et al found the cesarean delivery rate to be 45.5%.³⁸

Fetal outcomes in births after bariatric surgery are also vital to consider. A study from the Swedish Medical Birth Register found pregnancies after bariatric surgery were associated with lower risk of large-for-gestational age infants but higher risk for small-for-gestational age infants and still birth compared to matched counterparts though this study included just 650 bariatric surgery patients.³⁹ An Australian study examined infants born to the same patient before and after bariatric surgery as well as multiple births in the general population and found bariatric surgery was associated with fewer large-for-gestational age infants and fewer admissions to neonatal special or intensive care.⁸ A systematic review by Maggard et al found lower rates of macrosomia and neonatal complications after gastric bypass.³ The review’s findings also suggested biliopancreatic diversion was associated with lower adverse neonatal outcomes but higher miscarriage rates.³ More data are needed to ensure fetal safety and health.

Our preliminary data suggest it is time to rethink our counseling for pregnancy after bariatric surgery. Current recommendations are based on theoretical concerns for maternal and fetal risks: our data show births after bariatric surgery are not associated with increased rates of surgical intervention and obstetric outcomes are similar to other patients with obesity. Further, existing data demonstrate equivalent to improved fetal outcomes after bariatric surgery,.⁸ Given these findings, perhaps we need to readjust paradigms towards shared decisionmaking. Shared decisionmaking is a central component of the choice to pursue bariatric surgery; this should be mirrored in considerations of when to pursue pregnancy after bariatric surgery. Components that perhaps should weigh more heavily in the decisionmaking process are the risk of becoming pregnant too early (not achieving as much weight loss or comorbidity resolution), family planning desire (wanting a baby now), and the risk of delaying pregnancy (miscarriage and genetic disorders). Already established shared decisionmaking aids for reproductive goals⁴⁰ and pregnancy clinical decisions can be adapted to explore these aspects of pregnancy and bariatric surgery. Many aids have already been effective in helping patients make other types of clinical decisions during pregnancy.⁴¹ Future work is needed to incorporate fertility and early pregnancy outcomes into the decisionmaking process.

One major component of this complex decisionmaking process we do not have is efficacy data surrounding weight loss after pregnancy. Studies show obesity is related to early or recurrent miscarriage,^{42, 43} gestational hypertension,⁴⁴ pre-eclampsia,^{44, 45} gestational diabetes, cesarean delivery,^{46, 47} fetal growth disorders,⁴⁸ and infants who develop childhood obesity.⁴³ However, patients may be willing to accept less weight loss and not fully realize these benefits if they are able to have a pregnancy when younger or before reaching advanced maternal age. It is also possible patients may be seeking bariatric surgery in order to become pregnant³ due to obesity-related infertility.^{49, 50} If there is no hazard to the mother, this tradeoff between weight loss and other values and concerns, may be acceptable.

Our study is not without limitations. First, because we examine births rather than pregnancy, so it is possible that conception rates, and thus guideline discordance, are even higher than reported findings. Second, BMI prior to pregnancy, at conception, and at birth as well as clinical severity of comorbidities are not found in claims data, so we are unable to examine weight loss outcomes. Third, there may be important differences we were unable to account for between patients who did and did not give birth in this observational study, precluding causal inference. Fourth, race, ethnicity, and sociodemographic data were not available. Given the higher maternal mortality and morbidity among patients of color and low-income mothers, this is a vital part of future research. Additionally, these differences may represent sources of unmeasured confounding in our analysis of reinterventions. Fifth, prior cesarean delivery may have been undercounted since we relied on codes reflecting prior births. Sixth, time-to-event analysis which would account for censoring could not be used in our analysis because births were a rare event. Lastly, our study examines commercially-insured patients. Commercially-insured patients make up 73% of bariatric surgery patients,⁵¹ but findings may not be applicable to Medicare-, Medicaid-, and un-insured patients. Additionally, insured patients may have easier access to healthcare and be more likely to receive appropriate preventive care as well as timely diagnosis and management of pregnancy complications. They may also face fewer adverse social and structural determinants of health which are known to contribute to adverse maternity care and general health outcomes. Despite these limitations, this is the largest, nationwide cohort study examining births after bariatric surgery in the modern era of sleeve gastrectomy.

Conclusion

Our study provides important safety data to inform shared decisionmaking for pregnant patients and their healthcare providers. We found a high number of patients become pregnant after surgery. The decisionmaking process for childbirth after bariatric surgery is a complex one. We need to share these findings with patients to better inform them and perhaps revisit how we counsel patients.