Fibrates and risk of congenital malformations: a nationwide cohort study in South Korea

Low-density lipoprotein cholesterol is a widely recognized causal factor in the development of atherosclerotic cardiovascular disease and represents the primary focus of lipid-lowering treatment [1]. There is, however, still a substantial risk of atherosclerotic cardiovascular disease (CVD) events despite intensive statin therapy, and the results from clinical trials suggest that an elevated concentration of triglycerides is a marker of residual cardiovascular risk on low-density lipoprotein-lowering therapy. Patients with nonfasting triglycerides concentrations > 443 mg/dL showed 3.4-fold increased risk of CVD (78 events/10,000 person years) compared to those with < 89 mg/dL [2]. Triglyceride levels are a significant risk factor for coronary artery disease in women, more so than in men [3].

Pregnancy-related hypertriglyceridemia is infrequent, but in certain cases, it can pose a life-threatening condition. A systematic review of prevalence of acute pancreatitis in pregnancy suggested that it occurs in 0.02–0.22% of pregnancies [4]. Complications can include acute pancreatitis, hyperviscosity syndrome, and possibly preeclampsia [5‐7]. In diabetic pregnancy, exaggerated hypertriglyceridemia has been found compared to normal pregnancy, which has been correlated to neonatal body weight or fat mass [8]. Fibrates are practical first-line choice for patients with very high triglyceride levels (≥ 500 mg/dL), but only few case reports are available regarding patients been prescribed fibrates in pregnancy and their teratogenic outcomes [9‐11]. The European Society of Cardiology/European Atherosclerosis Society guidelines currently contraindicate the use of fenofibrate in pregnant women due to a lack of substantial evidence [12]. While there's limited data on the usage of fibrates during pregnancy, their application remains a topic of debate because of potential teratogenic risks. No published reports have demonstrated teratogenic effects, and both fenofibrate and gemfibrozil are categorized as pregnancy category C drugs [5, 10, 13‐15]. As of now, such categorizations are no longer widely utilized, and the FDA's stance is that fenofibrate should only be administered during pregnancy if the anticipated benefits outweigh the potential fetal risks [9]. Clinical trials targeting pregnant women are difficult since pregnant women are a vulnerable group. Furthermore, due to very low prevalence of exposure, the analysis of drug exposure cases during pregnancy requires nationwide data which made it difficult to conduct a cohort study.

Prevalence of hypertriglyceridemia increases with age, from 3.9% in women aged 20–29 to 12.6% in women aged 50–59 and the domestic prevalence rate is 15.0% in 2020 [16]. In a Taiwanese cohort study, 19% of pregnant women had fasting plasma TG levels exceeding 140 mg/dL in the first trimester, and the risk of developing adverse pregnancy outcomes (i.e. gestational diabetes mellitus or large for gestational age) was 2.56 times higher [17]. Safety data is required since the possibility of exposure in pregnant women is also increasing. The embryonic period, from the second to the eighth week of gestation, is the critical window of vulnerability for the developing fetus when exposure to harmful agents such as drugs, may cause structural malformations [18]. In pursuit of safe management of this patient group, we investigated the association between fibrates therapy during the first trimester and increased prevalence of congenital malformations.

Pregnancies from 2012 to 2021 in dyslipidemic women aged 19 to 44 years without childbirth records for the preceding year were included (n = 265,739), and 50,139 unlinked cases in the NHIS were excluded. The final study sample comprised 215,600 completed pregnancies linked to live-born infants (Fig. 1). Of those, 260 (0.12%) women received prescriptions for fibrates during the first trimester. Pregnancies in which women who had no prescription claims for fibrates during the first trimester were categorized as a comparison cohort (n = 215,340). Before matching, the fibrate-exposed group exhibited higher medical conditions and concomitant medication use. However, after matching, no significant baseline differences were observed, with standardized mean differences being less than 0.10 for all covariates (Table 1).

In this study, pregnant women exposed to fenofibrate during the first trimester were followed up to assess the risk of congenital malformations. To the best of our knowledge, this is the first large cohort study using the NHIS’s mother–child linked data to confirm the fetal safety of fenofibrate. A total of 0.7 million pregnancies were reviewed, and our study found that fibrates exposure during the first-trimester, a critical window of vulnerability for congenital malformations, was not associated with a substantially increased risk of overall congenital malformations.

The observed null associations of overall malformations with early prenatal fibrate use were consistent with existing case reports. A case report showed that fenofibrate could be used to treat high triglyceride-associated pancreatitis in the last trimester of pregnancy [11]. In another case report, gemfibrozil was successfully used to ameliorate the patient‘s hypertriglycerides, with no adverse pregnancy complications [14]. As a result, if deemed necessary, fibrates can be administered cautiously to pregnant women. It is not known whether fibrates cross the human placenta, however, the molecular weight of the metabolite (~ 319 Da) and the prolonged elimination half-life suggest that the drug would cross the placental barrier [22]. Alternatively, the high serum protein binding might limit the amount available for transfer to the embryo or fetus.

In our study, the analysis of duration of fenofibrates use is noteworthy. Note that a longer duration of exposure (> 60 days) increased the risk of overall malformations. This same pattern was observed in malformations excluding those of the circulatory system. Fenofibrate has been shown to be embryocidal and teratogenic in rats when given in doses 7–10 times the maximum recommended human dose [23] and embryocidal in rabbits when given at 9 times the maximum recommended human dose, suggesting potential harmful effects of fenofibrate exposure during the first trimester on fetal development. Therefore, our research results do not necessarily mean that fenofibrate is absolutely safe. There’s a possibility that cumulative effects could appear with continuous use, so caution should be exercised.

According to the results revealed in our study, the overall malformation rate is 9.7% in the control group and 10.8% in the exposure group, which appears to be higher than the 5.5% reported by another study [24]. To determine the effect of patient characteristics of the cohort, prevalence of overall malformations were confirmed in the entire cohort of 215,600 patients. When analyzed in the entire cohort, teratogenicity was 4.9% suggesting high prevalence was due to underlying maternal condition of the study subjects (data not shown). In our analysis, fibrates appeared to be relatively safe in terms of effects on the circulatory system when compared to other potential areas of concern. One possible explanation for this observation might lie in the mechanism of action of fibrates. These drugs primarily act to reduce triglyceride levels and increase HDL cholesterol [25]. The impact of fibrates on the circulatory system, particularly during fetal development, might be mitigated by their specific pharmacodynamic properties. However, there may be aspects of the fibrate mechanism that haven’t been fully explored in relation to fetal development, especially concerning the circulatory system.

This study also has a few limitations, which should be carefully taken into account when interpreting our findings. First, we were unable to determine whether patients actually took the fibrates prescribed. Therefore, we applied a strict exposure definition of  ≥ 2 prescriptions of fibrates to address for this uncertainty. It was assumed that if a woman refilled the respective drug’s prescription, she probably took them. Second, while we used the entire population of South Korea, there werent enough cases to ensure statistical significance when analyzing all malformations by category. Further studies are needed regarding organ-specific malformations of fibrates in the future. Last, residual confounding by indication or from unmeasured confounders may still be present due to inherent limitations of the observational nature of this study using claims data.

Our analysis did not identify a significant teratogenic effect from the use of fibrates in the first trimester. This suggests that inadvertent use during this period might not be a major concern. However, exposure exceeding 60 days showed increased teratogenicity. More information on the long-term effects of in utero exposure to fibrates is essential. The impact on other neonatal outcomes also needs investigation. Moreover, replicating our findings in other large datasets—with detailed information on fibrate use, confounders, and outcomes—is crucial. Only with these further studies can fibrate use during pregnancy be deemed safe.