Perfluoroalkyl substances follow inverted U-shaped distributions across various stages of glomerular function: Implications for future research

Highlights

• PFAS follow an inverted U-shaped distribution across declining stages of glomerular function.

• Shape of the inverted U-shaped distributions for PFAS differ by gender.

• Points of inflection are located at GF-2 for males and GF-3A for females.

Abstract

Data (N = 6844) from National Health and Nutrition Examination Survey for US adults aged ≥ 20 years for the years 2007–2014 were analyzed to evaluate distributional characteristics of selected perfluoroalkyl substances (PFAS) - perfluorooctanoic acid (PFOA), perfluorooctane sulfonate (PFOS), perfluorodecanoic acid (PFDA), perfluorohexane sulfonate (PFHxS) and perfluorononanoic acid (PFNA) with declining glomerular function. The population was stratified according to the estimated glomerular filtration rates (eGFR) that accompany the stages of kidney disease, designated as glomerular function-1 (GF-1, eGFR>90 mL/min/1.73 m²); GF-2 (eGFR 60–89 mL/min/1.73 m²), GF-3A (eGFR 45–59 mL/min/1.73 m²), and GF-3B and 4 combined (eGFR 15–44 mL/min/1.73 m²). Unadjusted as well as adjusted geometric means for serum PFOA, PFDA, PFHxS, and PFNA increased as expected through stage GF-3A but decreased below the concentrations associated with GF-1 for those who were in GF-3B/4. For example, unadjusted geometric means for PFOA were 2.59, 3.02, 3.01, and 2.22 ng/mL for GF-1, GF-2, GF-3A, and GF-3B/4 respectively. Adjusted geometric means for PFOA were 2.34, 2.83, 2.83, and 1.81 ng/mL for GF-1, GF-2, GF-3A, and GF-3B/4 respectively. Thus, PFAS were found to follow inverted U-shaped distributions across different stages of glomerular function. For females, decreases in adjusted PFAS serum levels were initiated at GF-3A, while decreases for males began as early as GF-2. Usually, females are known to have lower levels of PFAS but when in GF-3A and GF-3B/4, females were found to have higher levels of PFAS than males. Thus, inverted U-shaped curves for males and females intersected between GF-2 and GF-3A for PFOA and PFHxS and at GF-3A for PFOS and PFNA. Associations between PFAS and biomarkers of kidney function may be modified in both magnitude and even in direction as kidney function deteriorates. These findings have implications for studies that evaluate associations between PFAS and disease states that affect kidney function, as well as outcome biomarkers known to be affected by kidney function.

Introduction

Per- and polyfluoroalkyl substances (PFAS) are man-made chemicals with a wide variety of useful applications. PFAS have long-term environmental persistence attributed to their strong carbon-fluorine bonds. A subset of PFAS including PFAS alkyl acids with chain length ≥ C7 and alkyl sulfonic acids with chain length ≥ C6 have half-lives of several years or more in humans (Li et al., 2018) and are commonly detected in people around the world. The bioaccumulation of these long-chain PFAS in humans is attributed, in part, to their slow renal elimination (Han et al., 2012). In this context, it is natural to ask if declining kidney function may affect the rate of excretion of long chain PFAS. Watkins et al. (2013) found an interquartile increase in perfluorooctanoic acid (PFOA) to be associated with a 0.75 mL/min/1.73 m² decrease in estimated glomerular filtration rate (eGFR) among children aged < 18 years of age, and the study model assumed a linear, unchanging association between PFOA and eGFR, a marker of kidney function. PFAS excretion rates are sexually dimorphic in rats and therefore, it is not clear if there are also small gender differences even in some primates (Worley and Fisher, 2015, Butenhoff et al., 2004, Kudo et al., 2002).). There is limited evidence of sexually dimorphic renal excretion in humans. Gomis and colleagues found relatively faster excretion for PFOA and PFOS among females beyond what could be explained by the well-known effect of menstruation, cord-blood transfer, and breast feeding (Gomis et al., 2017).

Given the importance of renal excretion to PFAS serum concentrations and the uniquely long serum half-lives of the longer chain PFAS species, there is surprisingly little data published about how serum PFAS are affected by kidney disease. As kidney function deteriorates, we sought to consider the possibility that the distributions of selected PFAS may change with eGFR in magnitude, direction or both. For this investigation, we utilized publicly available data from National Health and Nutrition Examination Survey (NHANES, www.cdc.gov/nchs/nhanes/index.htm) conducted by US Centers for Disease Control and Prevention. NHANES data on both PFAS and renal failure have continuously been made available since 2003 in two-year cycles. The human serum concentrations of PFAS chemicals have changed over time, largely decreasing for the longer chain PFAS. Geometric mean concentrations of PFOA in NHANES data for the years 2003–2004, 2007–2008, and 2013–2014 were 3.96, 4.15, and 1.98 ng/mL respectively (Centers for Disease Control, 2018). Geometric mean concentrations of PFOS for the years 2003–2004, 2007–2008, and 2013–2014 were 20.9, 13.5, and 5.22 ng/mL respectively (Centers for Disease Control, 2018). It should be noted that in addition to PFOA and PFOS, CDC does provide data on other PFAS, for example, PFHxS and PFNA. In the interest of estimating PFAS distributions that will be most relevant for the currently existing concentrations of PFAS, a decision was made to use data for the years 2007–2008 to 2013–2014. In addition, this study was limited to US adults aged ≥ 20 years and to the commonly detected perfluoroalkyl- and perfluoroalkyl sulfonic acids in NHANES data. In this work, we therefore investigate data for long chain acids but retain the naming convention of PFAS (which includes these common acids and many other PFAS chemicals).

Our aim was to evaluate the association of estimated glomerular filtration rates (eGFR) that correlate with defined Kidney Disease Improving Global Health Outcomes (Inker et.al, 2014) stages of renal failure to PFAS serum concentrations.