Associations Between Polycystic Ovary Syndrome and Adverse Obstetric and Neonatal Outcomes

A Population Study of 9.1 Million Births

Ginevra Mills; Ahmad Badeghiesh; Eva Suarthana; Haitham Baghlaf; Michael H. Dahan


Hum Reprod. 2020;35(8):1914-1921. 

In This Article

Materials and Methods

We conducted a retrospective population-based study utilising data from the HCUP-NIS over 11 years from 2004 to 2014. The HCUP-NIS is the largest inpatient sample database in the USA and is comprised of hospital inpatient stays submitted by hospitals in 48 states and the District of Columbia. Each year, the database provides information relating to 7 million inpatient stays, including patient characteristics, diagnosis and procedures. The data are representative of ~20% of admissions to US hospitals and geographically represents over 96% of the American population.

We evaluated deliveries between 2004 and 2014 inclusively, by the using international classification of diseases, ninth edition, clinical modification (ICD-9-CM) codes for delivery-related discharge diagnoses (650.xx, 677.xx, 651.xx-676.xx where the fifth digit is 0, 1 or 2), and birth-related procedural diagnosis (72.x, 73.x, 74.0–74.2). Within this group, all women with PCOS were identified using ICD-9 diagnostic code 256.4, and the remaining deliveries were categorised as non-PCOS births and comprised the reference group. Women with anovulatory infertility (n = 29) were also excluded from the reference group. Although, this (n = 29) was verified twice, they may be under-represented in the database. ICD-9 codes were also used to identify demographic characteristics, maternal baseline characteristics and delivery and neonatal outcomes of all deliveries included in the study population. Baseline clinical characteristics included: patient age, race, income, insurance type, hospital type, previous C/S, multiple gestation, smoking history, obesity (BMI ≥ 30), pre-existing hypertension (HTN), pre-existing diabetes and pre-existing thyroid disease. Delivery outcomes include PPROM (preterm pre-labour rupture of membranes in the third trimester), PTD (i.e. before 37 completed weeks of gestation), placental abruption, C/S, maternal infection and chorioamnionitis. Neonatal outcomes included SGA, intrauterine foetal death (IUFD) and congenital anomalies. There were 9 096 788 births between 2004 and 2014, included in the study. Of these pregnancies, 14 882 women, or 163.6 per 100 000, had a documented diagnosis of PCOS.

An initial analysis was performed to identify the prevalence of pregnant women with PCOS over the entire duration of the study. We then contrasted baseline clinical and demographic characteristics between women with PCOS to those without PCOS. Subsequently, logistic regression analyses were conducted to explore associations between PCOS and delivery and neonatal outcomes through the estimation of odds ratio (OR) and 95% confidence intervals (CI). The regression models were adjusted for the potential confounding effects of maternal demographic, pre-existing clinical characteristics and concurrently occurring characteristics. All analyses were performed using SPSS 23.0 (IBM Corporation, Chicago, USA) software for all analyses. This study used exclusively publicly accessible, anonymised data; hence, according to articles 2.2 and 2.4 of Tri-Council Policy Statement (2010), institutional review board approval was not required.

Results and analysis regarding the development of gestational diabetes and pregnancy-associated hypertensive disorders in this study population are presented separately to ensure adequate discussion of these findings (Mills et al., 2020) .