P–634 The effect of bariatric surgery on PCOS patients’ obstetrical and neonatal outcomes: A population based study

Abstract

Abstract Study question Does bariatric surgery (BS) decrease the occurrence of adverse obstetrical and neonatal outcomes in morbidly obese women with polycystic ovarian syndrome (PCOS)? Summary answer Patients with PCOS who underwent BS were at lower risk for gestational diabetes mellitus (GDM) and pregnancy-induced hypertension (PIH), than other women with PCOS. What is known already Studies have shown that PCOS in pregnancy is associated with the occurrence of GDM, preeclampsia, PIH, preterm birth (PTB), cesarean section, miscarriage, hypoglycemia, and neonatal death. Patients with PCOS may have an increased risk of obesity compared to the general population, which magnifies the inherent insulin resistance many fold. PCOS patients who underwent bariatric surgery have a marked improvement in menstrual irregularities, hirsutism, T2DM, hypertension, and dyslipidemia. The benefit of bariatric surgery for PCOS patients and the improvement of pregnancy related complications has not been adequately studied, with most studies being small. Study design, size, duration This is a retrospective study using the Health Care Cost and Utilization Project-Nationwide Inpatient Sample (HCUP-NIS) database from 2004 to 2014. Using multivariate logistic regression analysis, we compared women with PCOS who underwent BS with a control group consisting of pregnant patients with PCOS of all BMIs who had not had weight reduction operations regarding pregnancy, delivery, and neonatal outcomes. Participants/materials, setting, methods The study group included pregnant PCOS patients who underwent BS and the control group consisted of pregnant patients with PCOS; who delivered between 2004 and 2014. Demographic and clinical characteristics, pregnancy, delivery, and neonatal outcomes were compared. Multivariate logistic regression analysis was performed to control for all statistically different (P < 0.05) confounding effects. Each subject was included once per delivery. Main results and the role of chance We identified 9,096,788 pregnancies during the study period. 141 patients had a history of PCOS and underwent BS. The control group was composed of 14,741 patients with PCOS who didn’t undergo BS. Prevalence of PCOS patients who underwent BS increased from 0/1,000 in 2004 to 14.6/1,000 in 2014(p = 0.001). Those who underwent BS were more likely to be older than 35-years(42.6% vs. 18.7%, p < 0.0001), obese at delivery(44.7% vs. 22%, p < 0.0001) and have a history of previous cesarean sections (24.8% vs. 18.2%, p = 0.045). Differences in pre-gestational diabetes (7.1% vs. 4.1%, p = 0.086) and hypertension (12% vs. 8.4%, p = 0.125) .The BS group was less likely to develop PIH (aOR–0.39, 95%CI–0.21–0.72) and GDM( aOR–0.40, 95%CI–0.23–0.70) when compared to the control group. When comparing the PCOS with and without BS; differences in gestational hypertension(95%CI–0.22–1.30), preeclampsia(95%CI–0.19–1.15), spontaneous vaginal delivery (95%CI 0.58–1.3), operative vaginal delivery (95%CI 0.34–1.8), Cesarean section (95%CI 0.79–1.79), post partum hemorrhage (95%CI–0.12–1.94), transfusion (95%CI–0.1–5.22), preterm delivery (95%CI–0.56–1.75) and maternal infection (95%CI–0.27–2.07) were similar. Regarding neonatal outcomes of PCOS with and without BS: small for gestational age babies (95%CI–0.26–2.68), and congenital anomalies (95%CI–0.09–4.52) were similar. Limitations, reasons for caution Limitations of our study include its retrospective design. Information regarding the time interval between the surgical intervention and pregnancy wasn’t available. Nor was information on the type of gastric bypass performed. Wider implications of the findings: Our study demonstrated an association between bariatric surgery in the setting of PCOS patients and a reduced risk of GDM and PIH. Although no differences were noted in other delivery and neonatal outcomes, data was not compared to a group of strictly obese PCOS controls. Trial registration number Not applicable