Influence of exposure to perfluoroalkyl substances (PFASs) on the Korean general population: 10-year trend and health effects

Highlights

• We conducted a time-trend analysis of 13 major PFASs over 10 years for 786 adults.

• This is the longest and largest scale research and the first study in Korea.

• The relationships between PFAS levels and sex, age, and BMI were assessed.

• We observed significant correlations between PFASs and cholesterol levels.

• Uric acid and free thyroxine showed positive correlations with PFAS levels.

Abstract

This study demonstrated the 10-year trend of 13 perfluoroalkyl substances (PFASs) serum levels among 786 adults living in Seoul, Korea. PFAS levels gradually increased from 2006 to 2013, decreasing thereafter. We found that PFAS levels were higher in male than in female participants and were positively correlated with age. PFASs were not significantly correlated with body mass index, although we observed positive correlations with total cholesterol, low-density lipoprotein cholesterol, and triglycerides and negative correlations with high-density lipoprotein cholesterol. Uric acid and free thyroxine (fT4) also showed positive correlations with major congeners while correlations between thyroid stimulating hormone and PFASs were inconsistent. We demonstrated significant correlations between fT4 and perfluorononanoic acid (PFNA), perfluorohexane sulfonate (PFHxS), and perfluorodecanoic acid (PFDA). There were significant differences in PFHxS and perfluorododecanoic acid (PFDoDA) levels between participants with and without diabetes. Furthermore, principal component analysis suggested possible differences in disease manifestation based on the congener distribution of PFASs. This study is the first study of temporal trends of 13 PFAS congeners in serum samples obtained from the Korean general population; it is currently longest and largest scale study of this type.

Introduction

Perfluoroalkyl substances (PFASs), such as perfluoroalkyl sulfonates (PFSAs) and perfluorocarboxylic acids (PFCAs), are industrial chemicals that are widely used in consumer products such as firefighting foam, food packaging, waterproof fabrics, non-stick cookware, cosmetics, pharmaceuticals, surfactants and paints (Kissa, 2001; Use, 1999). Humans are more frequently exposed to these chemicals in daily life than to other persistent organic pollutants (POPs) (Bowman, 2015). The primary routes of human exposure to PFASs include ingestion via the consumption of certain packaged foods or contaminated waters, inhalation of contaminated dust or indoor air and use of some types of kitchenware (D'Hollander et al., 2010; Shoeib et al., 2011; Tittlemier et al., 2007). Like many other chlorinated or brominated POPs, PFASs are persistent, toxic and bioaccumulative. Among PFASs, perfluorooctane sulfonyl fluoride (PFOSF) and perfluorooctane sulfonate (PFOS) were banned under Annex B to the Stockholm Convention on POPs in May 2009 (Wang et al., 2009). According to previous studies, the concentrations of these two PFASs in the environment increased until the 1990s, when efforts to reduce their emissions were initiated (Haug et al., 2009; Zushi et al., 2010). The concentrations of PFOSF and PFOS in the environment have decreased significantly since 2000. Similar changes in PFAS levels have also been confirmed for human exposure to PFASs (Haug et al., 2009; Schröter-Kermani et al., 2013; Zushi et al., 2010).

PFASs are classified as neutral or ionic; the former have high volatility and low solubility and are precursor chemicals whereas the latter compounds have opposite properties and are produced by various degradation processes of precursor chemicals. Therefore, ionic PFASs have both hydrophobic and lipophobic properties unlike other POPs, and these physicochemical properties cause them to easily accumulate in the human body by combining with lipids or proteins (Giesy and Kannan, 2002; Jones et al., 2003). Ionic PFASs detected in the human body consist of a combination of ionic PFASs found in the environment and degradation products of neutral PFASs. Among them, PFOS and perfluorooctanoic acid (PFOA) are the two main PFAS compounds found in human serum, and PFOS has been detected in relatively high concentrations (So et al., 2006). The reasons for this tendency are because many precursor chemicals are degraded into PFOS and PFOA, which have relatively long half-lives of 5.4 and 3.8 years, respectively, and are not easily eliminated from the body (Olsen et al., 2005). In general, the rate of elimination is higher for compounds with shorter carbon chains. In addition, females generally excrete PFOA faster than males owing to menstruation, and they transfer this substance to their infant during pregnancy and breastfeeding (Fromme et al., 2010; Harada et al., 2004; Hinderliter et al., 2006; Lau et al., 2005).

According to previous epidemiologic studies, chronic exposure to PFASs can cause hepatotoxicity, neurotoxicity, immunotoxicity, behavioral disability and genetic defects (Biegel et al., 2001; Cook et al., 1992; Lau et al., 2007; Sibinski et al., 1987). In addition, PFASs are strong peroxisome proliferators and can increase activation of peroxisome proliferator-activated receptor (PPAR) isoforms (Takacs and Abbott, 2007). This peroxisome proliferation and activation of PPARs have been shown to change levels of lipids, hormones and enzymes (Cook et al., 1992; Kennedy et al., 2004; Lau et al., 2007). PFAS levels in serum have strong correlations with age and sex, and older people generally have higher PFAS concentrations than younger people (Calafat et al., 2007; Kim et al., 2014). As touched on above, PFAS concentrations in serum collected from males tend to be higher than those of females (Calafat et al., 2007; Kim et al., 2014). Similarly, body mass index (BMI) is related to PFAS levels in the human body, and it is known that the levels of blood cholesterol, triglyceride, and low-density lipoprotein (LDL) have positive correlations with PFOA exposure levels (Costa et al., 2009; Ji et al., 2012; Nelson et al., 2010; Olsen and Zobel, 2007).

Metabolic syndrome refers to a clustering of several medical conditions such as abdominal obesity, hypertension, dyslipidemia and insulin resistance. The syndrome is associated with a risk of developing type 2 diabetes and cardiovascular disease (Felizola, 2015; Kaur, 2014). Uric acid is a natural metabolic product of purine nucleotides in the human body. Abnormal uric acid levels are a known risk factor for hypertension, cardiovascular disease, kidney disease and diabetes (Hayden and Tyagi, 2004; Shankar et al., 2006). It has also been shown that an excess accumulation of uric acid in the body can cause gout (Heinig and Johnson, 2006). Thyroid hormones are known to increase the basal metabolic rate, help with the regulation of bone growth and affect protein synthesis (Welcker et al., 2013). In previous studies, PFOS exposure was revealed to have a negative correlation with thyroid stimulating hormone (TSH) and a positive correlation with free thyroxine (fT4) (Bloom et al., 2010; Costa et al., 2009; Dallaire et al., 2009; Olsen et al., 2001). In other words, the persistence of PFASs in the body can affect thyroid hormone homeostasis (Butenhoff et al., 2002; Lau et al., 2007) and several previous studies have reported increased mortality rates owing to diabetes, based on PFAS exposure (Leonard et al., 2008; Lundin et al., 2009).

There have been several studies investigating the relationship between some diseases and exposure to PFASs. However, the data on PFASs in the human body have been limited, except for among a few studies with inconsistent results (Ji et al., 2012; Nelson et al., 2010; Steenland et al., 2010a). In addition, very little research has been conducted on the time trends and health effects of various PFAS congeners (Glynn et al., 2012). Therefore, this study aimed to demonstrate the changes of PFAS levels in human serum over time and the possible health effects of various PFAS congeners, using a large data set. In this study, we conducted time-trend analysis of 13 major PFASs over 10 years for 786 adults living in Seoul, South Korea. In addition, the health risks of PFASs were assessed by examining the relationship between serum concentrations of PFASs and sex, age, BMI and various adverse health effects such as increased cholesterol, elevated uric acid levels, hypothyroidism and diabetes mellitus. To the best of our knowledge, this is the longest and largest scale study to date and also the first to investigate the temporal trends of 13 PFASs in human serum conducted in Korea. In addition, unlike most previous studies that used pooled serum samples, individual serum samples were used to provide variability and to improve reliability of the results.