Periconceptional Use of Phthalate-Containing Medications and Secondary Sex Ratio

Lauren A. Wise,1 Thomas P. Ahern,2 Anne Broe,3,4* and Per Damkier3,5* Department of Epidemiology, Boston University School of Public Health, Boston, Massachusetts, USA Department of Surgery, Larner College of Medicine at the University of Vermont, Burlington, Vermont, USA Department of Clinical Biochemistry and Pharmacology, Odense University Hospital, Odense, Denmark Department of Clinical Pharmacology, Pharmacy, and Environmental Medicine, University of Southern Denmark, Odense, Denmark Department of Clinical Research, University of Southern Denmark, Odense, Denmark


Introduction
Secondary sex ratio (SSR) is defined as the ratio of number of males to females at birth and is considered a meaningful end point for indicating adverse effects of environmental exposures (Organisation for Economic Co-operation and Development 2018). The SSR in humans has been relatively stable over time in showing a slight male excess in births (51.4%) (James 2008). However, in recent years, some industrialized nations have witnessed a decline in SSR (Chao et al. 2019). Reasons for variability in the SSR are unclear. A prevailing hypothesis is that exposure to endocrine-disrupting chemicals in animals and humans influences SSR via changes in periconceptional hormonal concentrations (James 1987). Nevertheless, human studies of periconceptional exposure to endocrine-disrupting compounds and SSR have produced inconsistent results (Terrell et al. 2011).
We evaluated the association between periconceptional use of phthalate-containing medications and SSR among Danish women.

Methods
Using personal identifiers assigned to all Danish residents, we linked data from several Danish health registries (Laugesen et al. 2021). Data on medication excipients in pharmaceuticals marketed during the period 2004-2017 from the Danish Medicines Agency were linked with individual-level prescription data from the Prescription Register. We used the Anatomical Therapeutic Chemical Classification System (ATC) and unique product codes to identify phthalate content of prescriptions (Ennis et al. 2018), classified as ortho-phthalates or phthalate polymers. We defined preconceptional exposure as prescription redemption in the 3 months before conception (14 d of gestation) and early pregnancy exposure as prescription redemption during the first trimester (conception until 90 d of gestation).
We obtained data on offspring sex and potential confounders from the Medical Birth Register and National Patient Register, including maternal age, calendar year at delivery, parity, firsttrimester smoking, and first-trimester body mass index (BMI). The in vitro fertilization (IVF) register contained data on use of assisted reproductive technologies (ART) and prescriptions filled for fertility medications in the month before the index pregnancy, including gonadotropins and other ovulationinducing drugs (ATC codes: G03G, H01CA, H01CC, L02AE, N04BC).
During the period 2004-2018, we identified 894,547 liveborn singleton births. We excluded pregnancies without known gestational duration (8,714; 1%), women not residing in Denmark continuously for at least 1 y before pregnancy until delivery (13,456; 2%), women with preconceptional exposure before 2004 (59,012; 7%) or first trimester exposure ending after 2017 (28,510; 3%), and women without prescriptions redeemed during either exposure window (332,521; 42%), leaving 452,334 singletons born to 339,876 mothers. To minimize confounding by indication, we further restricted analyses to women exposed to medications available in both phthalate-containing and phthalatefree versions (78,165 singletons born to 51,397 mothers). Thus, we compared SSR among women taking phthalate-containing medication relative to women who took the same medication that did not contain phthalates. Clinicians and pharmacists were considered blinded as to phthalate content of the prescribed medications.
We estimated odds ratios (OR) and 95% CI using generalized estimating equations to account for multiple births per mother. We selected confounders based on the literature and causal diagrams. These included maternal age at birth (<25, 25-29, 30-34, 35-39, ≥40 years of age), calendar year of infant's birth (2004-2008, 2009-2013, 2014-2017), parity (1, 2, ≥3 births), and history of infertility prior to current pregnancy (ever vs. never). We performed secondary analyses further adjusted for paternal phthalate-containing medication use during preconception and analyses restricted to first births during the study period (regardless of parity) or births conceived without use of ART or fertility medications. We assigned missing values for smoking and BMI to the modal value for each variable. We used Stata software (version 16.1; Stata Corp.) for statistical analyses.

Results
Exposed and unexposed women were comparable with respect to age, BMI, parity, smoking status, and concomitant medication use. Exposed women were less likely to conceive in the later years of the study, to have an infertility history, or to have used ART to conceive their index pregnancy (Table 1).

Conclusion
In this Danish registry-based cohort study, maternal use of phthalate-containing medications during preconception or early pregnancy was not appreciably associated with SSR. Table 1. Baseline characteristics of Danish women with and without exposure to phthalate-containing medications during preconception or early pregnancy (2004-2017).
Exposed Unexposed

Total exposed a
Ortho-phthalate exposed Phthalate-polymer exposed Phthalate unexposed (exposed to study medication) b Unexposed to study medication but exposed to other medications c Note: Expressed as numbers and column percents or means and SD. ART, assisted reproductive technology; IVF, in vitro fertilization; SD, standard deviation. a Exposed in preconception or first trimester of pregnancy. Exposure group for all subsequent analyses. b Exposed study medication in a phthalate-free version.
c Excluded from analysis. Unexposed to study medications but exposed to other prescription medications (Ennis et al. 2018).