Neurodevelopmental and neurological effects of chemicals associated with unconventional oil and natural gas operations and their potential effects on infants and children

Sources of air pollution from UOG

UOG operations emit air pollutants linked to adverse neurological effects throughout their lifecycle. In a study assessing air pollution near natural gas operations in rural Colorado, Colborn et al. (6) identified 35 potential neurotoxins that are associated with the process. Stages of the UOG lifecycle associated with air pollution include the extraction and processing of natural gas, transportation via compressor stations and pipelines, truck transportation of materials to and from well pads, use of vehicular equipment during construction and maintenance, and venting, flaring, production, and leaks from faulty casings (3), (81). These processes release numerous contaminant categories into the air, including our five pollutant categories of concern (1), (3), (4), (6), (12), (77), (79), (80), (82), (83), (84), (85).

Sources of water pollution from UOG

Oil and gas development processes release numerous contaminant categories into surrounding water sources, including our pollutant categories of concern (75), (76), (78), (86), (87), (88), (89), (90). Water pollution can occur during wastewater disposal, transport, and during and after hydraulic fracturing, processing, production and distribution to the market (7), (91). “Flowback” or “produced” water, the water that travels to the surface following the hydraulic fracturing procedure (typically a mixture of water, gas, oil, metals and fracking fluids), can contaminate groundwater (76). The fracturing fluid used in shale gas development contains organic and inorganic chemicals harmful to human health. Although these chemicals are currently unregulated at the federal or state level, many environmental and public health experts, including the EPA, have reported the presence of fracturing fluids in drinking water (75), (92).

People may be exposed to chemical fluids from UOG development processes in a number of ways, including spills, releases from surface tanks, surface leaks, poor well integrity, accidents during transportation, flowback and produced water during hydraulic fracturing, run-off from storms and blowouts and other events (91). There is often inadequate filtering at treatment facilities which can then lead to harmful chemicals entering the local water supplies and/or being used in agriculture (93). These chemicals are then released into landfills or surface waters. In different areas, this water has been used for other purposes such as irrigation, or for spraying roads to reduce dust or melt ice (94). While surface spills and leaks occurring both during and after well stimulation can lead to water contamination, other oil and gas development processes can also lead to water contamination. In California, for example, researchers are investigating the potential health concerns associated with the reuse of oil field produced water for crop irrigation (95).

PM levels near UOG sites

PM is a mixture of solid particles, particle droplets, dust particles, heavy metals, and other organic chemicals that have become suspended in the air, are small enough to be inhaled, and can travel long distances (96), (97), (98).

Diesel PM and PM of 10 μm or smaller in diameter are known to be emitted into surrounding air throughout all of the stages of the UOG lifecycle. Such stages include well-site preparation and road construction (which involves trucks and heavy machinery), hydraulic fracturing, production (gas flaring/venting and maintenance), processing and storage (involving compressors), transmission (involving compressor stations), and well abandonment and site rehabilitation (which includes the use of trucks and heavy machinery) (81), (99).

The U.S. Environmental Protection Agency (EPA) National Ambient Air Quality Standards (NAAQS) estimates that levels posing a potential health risk are 35 μg/m³ for a 24-h average and 15 μg/m³ for an annual average for PM_2.5 (100). One study reported PM levels exceeding the NAAQs around fracking mining operations. In this study, Walters et al. (79) found PM_2.5 levels of 5.82–50.8 μg/m³ in six 24-h samples around fracksand mines and processing sites in Wisconsin and Minnesota. Five of the six air samples exceeded the EPA standard of 12 μg/m³, and researchers pointed out that these may underestimate true levels given that weather conditions can lower PM_2.5 atmospheric concentration levels (Table 2).

Brown et al. (85) used an air exposure model to determine the concentrations of PM_2.5 in the air during different 6-h periods over 24 h near three UOG facilities in Pennsylvania. They reported that peak PM_2.5 exposures occurred 83 times over 14 months of well development with the greatest intensity exposures occurring during well development, drilling, flaring and gas production. Peak exposure levels from compressor stations were reported to take place 118 times and there were 99 peak exposures from a gas processing plant.

Although the EPA has not conducted adequate ambient air quality models of PM_2.5 near UOG sites, it has emphasized the need for their emission reductions given adverse health effects identified in the scientific and medical literature and based on their own regulatory analysis of the industry (103).

Exposure pathways and mechanisms

The primary exposure pathway for particulate matter is through inhalation and the particle size influences where the particles travel in the body. Coarse particles (2.5–10 μm in aerodynamic diameter) are usually deposited in the airways and upper respiratory tract, while fine particles (<2.5 μm) may reach terminal bronchioles and alveoli (104). Compared to large particles, fine particles can remain suspended in the air for longer durations of time and travel over longer distances (104). PM_2.5 (≤2.5 μm) in particular poses a significant health concern, as particles of this size are known to contribute to cardiovascular and respiratory diseases as well as premature death (105), (106).

Experimental studies in both animals and humans have shown that PM can enter the brain through inhalation. In a rodent study, for example, inhalation of particles resulted in translocation of these particles to the brain (107) specifically the brainstem and hippocampus (108). Studies show that high exposure of PM can result in penetration of multiple functional areas of the brain and lead to detrimental effects (109), (110).

Neurological and neurodevelopmental effects

While more research is needed on the mechanisms of particulate matter’s toxicity, the effect on the brain is clear (111). There is direct evidence of various types to support the hypothesis exposure to elevated levels of particulate matter can lead to certain brain diseases and functional clinical impairment [for an extensive review, see (112)].

Neuroinflammation, especially in prenatal and early life appears to have a profound effect on brain development, potentially leading to disorders such as autism, schizophrenia, bipolar disorder and anxiety (67). Immune mechanisms of fighting infection is important for proper brain development and functioning (113). Recent evidence suggests that inflammation and its processes can modify cellular processes in the brain and spinal cord, affecting the synaptic connections and plasticity that is required for learning and memory (64).

In a series of studies comparing the effects of air pollution in Mexico City with those of similar cities with significantly lower pollution levels, researchers have consistently encountered neuroinflammation, damage suggestive of oxidative stress, direct neuronal damage, and poor clinical outcome in both animals and humans (66), (109), (110), (114), (115), (116), (117).

White matter hyperintensities (WMHs) are extreme vascular lesions routinely found in the elderly, predicting increased risk of dementia and death, signal blood-brain barrier (BBB) disruption and neuroinflammation (118). In one study comparing children from Mexico City with children from less polluted areas, magnetic resonance imaging (MRI) prefrontal white matter hyperintensities (WMHs), were found in 56% of clinically healthy children from Mexico City, compared to 7.6% of children that were from the control site. The children residing in Mexico City in Calderón-Garcidueñas et al.’s (65) study with prefrontal WMHs displayed problems with memory and executive functions, and also had large deficits in cognition (measured by the Weschler Intelligence Test for Children and IQ), compared to children from the control sites where there was less pollution.

The same researchers conducting work on children residing in Mexico City have also found evidence of lesions at several levels of the brain, down to the brainstem. They additionally found damage to the nuclei, which are key to basic functions such as auditory processing, balance and autonomic regulation (119). Early lesions such as these can lead to severe developmental problems, although correlating such issues with the lesions is not clinically feasible in most cases. Therefore, this burden is likely underestimated.

In addition to developmental and degenerative disorders, psychiatric disorders may be associated with particulate matter exposures. Fine particulate matter exposure has been associated with high symptoms of anxiety (68). Although largely unexplored, mechanisms such as inflammatory changes and oxidative stress could explain these findings.

PAH levels near UOG sites

PAHs refer to several hundred chemically-related organic compounds. Benzo[a]pyrene (BaP) is most commonly used as an indicator species of PAH pollution and is one of the PAHs that are found in crude oil. PAHs, found in common crude oil, are known constituents in produced water from UOG but are not being monitored in areas impacted by oil and gas operations. PAH compounds are present in fossil fuels and are also products of their combustion. Fossil fuel combustion is one of the primary sources to ambient PAHs (120).

Some studies have reported PAHs in ambient air samples near UOG operations. Colborn et al. (6) for example, found levels of PAHs (PAH16~15.5 ng/m³) near natural gas well pads dangerous to human health. These levels have been associated with significant decreases in IQ and delayed mental development in children exposed in utero (Table 2).

Paulik et al. (102) analyzed for 62 PAHs in rural Ohio where there has been a huge natural gas exploration and production boom and found levels of benzo[a]pyrene, phenanthrene, and other PAH mixtures closest to active natural gas wells to be highest (Table 2).

Exposure pathways and mechanisms

PAHs attach themselves to particles in the air; distribute across air, soil and water and can travel long distances (121). PAHs can enter the body through inhalation, traveling through the lungs into the bloodstream. PAH neurotoxicity is thought to occur indirectly through microglial activation (51). PAHs are thought to cause antiestrogenic effects (122) and DNA damage (123), and can also lead to changes at the cellular level that then affect the exchange of nutrients and oxygen (124). Exposure to PAHs in utero may affect neurological function, immune and metabolic function, and potentially epigenetic programming (44), (121), (125).

Some PAHs are transplacental carcinogens which can cause phenotypic or genotypic changes in cells in the fetus following carcinogenic exposure. In-utero exposure to transplacental carcinogens can result in the production of tumors in the lung, liver, lymphatic and nervous system in children (126).

In a 2005 study conducted by Environmental Working Group (EWG), 217 chemicals known to be toxic to the brain and nervous system were found in the umbilical cord blood of 10 babies born in US hospitals in 2004. Nine of the 217 chemicals were PAHs (127).

Neurological and neurodevelopmental effects

Brain disorders and neuropsychology

Some researchers have followed PAHs’ impairment of normal cognitive function and implicated PAH exposure in the etiology of brain disorders. Perera et al. (49) followed children of nonsmoking African-American and Dominican mothers in a prospective cohort study, and assessed the risk of attention deficit hyperactivity disorder (ADHD) diagnosis through the Diagnostic and Statistical Manual of Mental Disorders (DSM) and found a possible connection between early life exposure to PAHs and childhood ADHD behavior problems.

In this same cohort, Peterson et al. (51) conducted the largest MRI study of the neurological effects of prenatal exposure to PAHs, following urban youth from gestation to school age. The researchers found a dose-response relationship between prenatal PAH exposure and reductions of white matter in the brain that resulted in high scores for ADHD symptoms and conduct disorder problems. They also found an association between PAH exposure and reduced processing speed during intelligence testing (51). Follow-up assessments of postnatal PAH exposure at 5 years of age found disruption in dorsal prefrontal white matter. Although they could not conclude how these postnatal effects on prefrontal white matter impact cognition or behavior, they predicted further difficulty with processing speed, attention, and impulse control functions supported by the prefrontal cortices. Based on their findings, these researchers hypothesized that PAH could alter levels of monoaminergic neurotransmitters, or their receptors, which regulate early brain development and lateralization (51).

Recently, researchers from Columbia Center for Children’s Environmental Health and New York State Psychiatric Institute reported that early life exposure (in utero) to PAH air pollution in 462 children (3–11 years of age) may play a role in the cause of many childhood behavior disorders including ADHD, obsessive-compulsive disorder, Tourette’s syndrome and eating disorders (50). PAH was measured by the presence of PAH-DNA adducts in maternal blood at delivery. Children between 3 and 11 years of age, that had been prenatally exposed to PAHs, failed to improve their self-regulation skills, while children of mothers that did not appear to have the presence of PAH-DNA adducts displayed increased capacity for self-regulation (50). In other words, children with prenatal exposure to PAH had persistent problems with self-regulation, as evidenced through a variety of domains such as managing attention, aggression, anxiety/depression, as well as the ability to get along with others. Based on these findings, researchers concluded that childhood exposure PAH significantly heightens the risk for various disorders (50).

Levels of EDCs near UOG sites

Of the over 750 chemicals used throughout the process of hydraulic fracturing, more than 100 are known or suspected EDCs, and many others have yet to be assessed due to lack of Chemical Abstract Service numbers and/or proprietary information concerns. This can make searching for health data difficult (1), (134). The 2005 Energy Policy Act, exempted hydraulic fracturing from the Underground Injection Control program and the EPA is prohibited from regulating fracking under the Safe Drinking Water Act (1974) (7). Due to these actions, fracturing fluids are excluded from federal disclosure rules and a number of chemicals associated with UOG are not reported to the public.

A mounting number of studies have reported EDC activity in surface and/or groundwater near UOG operations (89), (90), (135). Kassotis et al. (89) found greater EDC activity in surface and ground water from fracturing fluid spill sites in areas where more intensive natural gas development, compared with reference sites with limited UOG activity.

In another study, the same researchers, assessed EDC activity of fracturing chemicals in surface water at a West Virginia injection well disposal site accepting wastewater partly from UOG operations (90). The researchers reported antagonist activities for most of the chemicals that were analyzed. Given that currently in this country, over 95% of end disposal of hydraulic fracturing wastewater from UOG operations occurs via injection wells (with over 140,000 such wells in operation), these environmental impacts and health consequences on surrounding organisms can be thought to be widespread (90).

Exposure pathways and mechanisms

The Endocrine Society defines EDCs as “exogenous chemical(s), or mixtures of chemicals, that interfere with any aspect of hormone action” (52), (55). Once EDCs are absorbed into the body, they act via a range of mechanisms to alter hormone action, thereby causing a spectrum of adverse effects such as hormone-dependent cancers, infertility, miscarriage and birth defects (55), (136), (137), (138). These mechanisms can include direct agonism or antagonism of hormone receptors, disruption of steroidogenic enzymes, and impairment of steroid receptor expression among others (56).

Commonly used UOG chemicals have been shown to be able to act as antagonistic EDCs in animal studies. In one animal study, prenatal exposure (via maternal ingestion) to concentrations equal to and less than those found in three oil and gas wastewater samples (first, from a ruptured pipeline actively leaking; second, an open storage tank, and third, a closed storage tank in Colorado) resulted in adverse male reproductive outcomes in mice. Specifically, there was an increase in weight of testis, body and heart and a decrease in sperm counts (135). Kassotis et al. (139) found that pregnant mice that were prenatally exposed to drinking water contaminated with UOG chemicals from gestational day through birth displayed a host of fertility and endocrine-related adverse health effects including suppressed pituitary hormone concentrations across experimental groups (prolactin, luteinzing hormone (LH), follicle stimulating hormone (FSH), etc.), increased body weights, changes in uterine and ovary weights, disrupted folliculogenesis, among other reported effects. Exposure to EDCs during the perinatal period has been shown to cause permanent changes in the brain and behavior (52), (53), (54).

Neuroendocrine systems coordinate virtually all homeostatic processes and functions, including growth, stress, energy balance, stress, lactation and reproduction. Importantly, the neuroendocrine system coordinates an organism’s response to the environment including chemical exposures. Although still lacking a formal definition, the term “neuroendocrine disruptors (nEDCs)” has been used to describe chemical impacts on endocrine-related brain development and function including the function of the nervous system (140), (141). Importantly, neuroendocrine disruption is different than neurotoxicity, which defines processes resulting in neuronal cell death and related to up or downstream consequences (e.g. oxidative stress or inhibition of neurotransmission) including peripheral neuropathies. The neuroendocrine system, like other neural systems, is not fully mature at birth and can be extremely sensitive to hormones and nEDCs at many points in the life span, most notably the prenatal window and the pubertal transition (142). It is also, in many aspects, sexually dimorphic, thus resulting in sex-specific vulnerabilities (143).

Cognition, behavior and learning

Although beyond the scope of the present review, numerous prior analyses have linked nEDCs with adverse neural and behavioral outcomes in a variety of animal models, including impaired social interaction/activity, compromised learning and memory, increased anxiety and aggression, modified brain sex differences, altered hippocampal spine density and advanced puberty (55), (56). Other endocrine disruptors (including those not apparently associated with UOG development) such as BPA appear to affect social and maternal behavior and reproductive functions. Low dose oral BPA exposure have been observed on oxytocin and related pathways. Oxytocin is a peptide hormone critical for mediating social and maternal behaviors, such as licking-grooming in animals (56), (144). Developmental exposure to bisphenol A (BPA) has been shown to alter the sexual differentiation of neural circuits involved in the control of reproductive functions and behavior including ovulation and sexual receptivity (143), (145).

Brain disorders and neuropsychology

Exposure to nEDCs that disrupt hormone function during critical periods of prenatal development may enhance susceptibility to sex- and/or hormonally-differentiated behavioral disorders. This is an outcome which has been interpreted to indicate that EDC exposure might contribute to the etiology of disorders with sex-biased prevalence rates such as autism spectrum disorders, ADHD and depression (56). These potential linkages should be viewed with caution, however, because although it is widely postulated that EDCs are contributing to clinical neural and behavioral disorders, specific evidence for such a relationship is sparse, even for well-studied EDCs like BPA. That said, detrimental effects on aspects of cognitive function and reasoning, including IQ, have clearly been shown in both animal models and humans for EDCs such as the PBDEs and PCBs. Effects are particularly evident in arctic populations where exposure is especially high and chronic. These effects can be exacerbated by co-exposure to other contaminants, including heavy metals, emphasizing their significance within a complex exposure paradigm. While difficult to quantify, the economic costs of EDC exposure on neurobehavioral deficits and diseases have been estimated to be in the range of 150 billion euros for the EU (146). These estimates contextualize the real “costs” of EDC exposures even if they cannot be specifically attributed to a clinically diagnosable cognitive or other neural disorder. Currently, the research on specific UOG chemicals that are nEDCs is sparse, but due to what is emerging about well-studied EDCs like BPA, more research exploring the potential link between UOG EDC exposures and disruptions during prenatal neurodevelopment is apparently needed.

Levels of heavy metals near UOG sites

Arsenic has been found in both flowback wastewater and produced water (water that is found in gas formations underground and comes to the surface over the lifetime of the well) from hydraulic fracturing sites on the Marcellus Shale (87). Studies have also confirmed arsenic contamination of ground/drinking water from flowback and produced water being treated and then released back into the environment. For example, Fontenot et al. (75) found higher levels of arsenic in the private wells of homeowners living within 3 km of sites of natural gas extraction in the Barnett Shale region in Texas. These levels exceeded EPA’s maximum contaminant limit for drinking water of 10 μg/L (Table 2) (21), (87). Glenn and Lester (101) found that contamination by arsenic and various other heavy metals of groundwater exceeded maximum contaminant levels and health risk limits in the Gulf Coast aquifer of Texas, a region inhabited by more than 7.5 million people and in the proximity of at least 86,000 facilities (oil, gas and storage wells) This number is most likely an underestimate because certain well types were excluded from analysis (Table 2).

The heavy metal manganese has also been found in flowback water from fracking operations, and in drinking water near UOG sites at levels known to be hazardous to human health (78). Manganese can contaminate groundwater as acids in fracking fluids will cause metal dissolution (147).

Boyer et al. (88) found that after hydraulic fracturing, within 3000 feet of shale gas wells in rural Pennsylvania, there was an increase in concentrations of manganese in drinking water wells. The concentrations spanned from near or below the drinking water standard (0.05 mg/L) to above safety levels after drilling occurred (Table 2) (88). A 2015 report on well water contamination in a rural community within 4 km of unconventional shale gas extraction similarly reported concentrations of manganese found in well water (78). Twenty-five of the reporting households documented levels exceeding the maximum contaminant level (SMCL) of 0.05 mg/L (the highest was 2.627 mg/L) (Table 2) (78). Manganese water contamination has also been reported in other regions across the US such as Pennsylvania (148). Notably, two fracking lawsuits have involved plaintiffs reporting neurological symptoms from exposure to test-verified manganese groundwater contamination in Pennsylvania and West Virginia (149).

Exposure pathways and mechanisms

Arsenic ranks as number one in the Agency for Toxic Substances and Disease Registry (ATSDR)’s list of 275 environmental substances that pose the most significant threat to human health (150). In addition to gastrointestinal, respiratory, cardiovascular, and other diseases, acute and chronic arsenic exposure is associated with neurotoxicity, nerve inflammation (neuritis) and neuropathy, with possible long-lasting effects (151), (152).

Consuming contaminated drinking water is generally the main route of human exposure to arsenic, in addition to consuming food that is irrigated with contaminated water or from food prepared with contaminated groundwater (21). Infants are more vulnerable to exposure because of their greater consumption than adults on a body-weight basis. After ingestion, 60–90% of organic and inorganic arsenic is absorbed into the bloodstream (153). Mechanisms of arsenic neurotoxicity include interference with neurotransmitter release and metabolism, and induction of oxidative stress (21).

Arsenic impairs the developing brain by inhibiting neuron growth both in the central and peripheral nervous system. It additionally interferes with various stages of neurodevelopment including synapse formation and neuronal migration (19). Arsenic has been shown to cross the placenta and may also cross the BBB and directly affect the central nervous system (21). One study reported that in-vitro arsenic exposure can induce disturbances in neuronal development and apoptosis of neuronal cells in human fetal brain explants (154).

Manganese, like arsenic, is a known neurotoxin that can produce cognitive, neuropsychological and motor deficits in both humans and animals. Chronic exposure has been linked to Parkinson’s disease (155), (156). According to a recent review of the literature, potential mechanisms include dysregulation of the dopaminergic system and alteration of the working memory network due to impacts on the striatum, frontal and parietal cortex (the region integral to sensory information and working memory performance) (155). Notably, both animal and human studies have shown that concurrent exposure to arsenic and manganese can have additive effects on newborns intellectual functioning. In one study, 11–13 year olds with exposure to both arsenic and manganese showed greater deficits in intellectual functioning (29). More research is needed on both the mechanisms by which manganese induces cognitive impairment and on the psychiatric effects of chronic manganese exposure.

Neurological and neurodevelopmental effects

BTEX levels near UOG sites

BTEX – benzene, toluene, ethylbenzene and xylene – are a tetrad of volatile organic compounds and have routinely been found near UOG sites (165). These compounds are emitted in every stage of the UOG lifecycle – from machinery used in well site preparation; well drilling, fracking, and well completion processes; flowback or produced water; gas flaring/venting and maintenance during production; separators, condensate tanks, and compressors used in processing and storage; pipelines, compressor stations and gas venting used in transmission; and finally, from machinery used in well abandonment and site rehabilitation (7), (77), (82), (83), (84).

A number of studies have found elevated BTEX concentrations in ambient air samples and water samples near UOG sites that exceed regulatory standards and/or minimum health risk levels established by the Agency for Toxic Substances and Disease Registry (ATSDR), Center for Disease Control and Prevention (CDC), and the Environmental Protection Agency (EPA) (Table 2) (3), (4), (76), (77), (80), (166).

Researchers Rich and Orimoloye (80) reported elevated concentrations of ambient BTEX compounds in residential areas at different distances from a natural gas facility in six counties in Texas. Concentrations of benzene were elevated when compared to the U.S. EPA’s Urban Air Toxics Monitoring Program. Twenty-four hour benzene concentrations ranged from 0.6 parts per billion by volume (ppbv) to 592 ppbv. One hour benzene concentrations ranged from 2.94 ppbv to 2900.20 ppbv (Table 2) (80).

McKenzie et al. (3) measured hydrocarbons near drilling sites in Colorado and found benzene levels in the air to be as high as 22 ppbv, many magnitudes higher than the 0.4 ppb concentration in ambient air that the EPA has estimated could put 1 in every 100,000 exposed people at an increased risk of cancer (Table 2) (3), (12). All four BTEX compound concentrations increased with proximity to the well site.

Macey et al. (77) found similarly elevated concentrations in air samples collected near UOG operations in five states, including Arkansas, Colorado, Ohio, Pennsylvania, Wyoming. The air samples contained varying National Ambient Air Quality Standards (NAAQS), including benzene concentrations exceeding health risk levels by several orders of magnitude at up to 885 feet, and toluene and xylene samples exceeding ATSDR risk levels (Table 2). Elevated concentration levels of BTEX chemicals (including benzene) have also been found in groundwater near UOG sites (86). Gross et al. (76), for example, reported in 90% of the groundwater samples contaminated by surface spills from UOG wells in Colorado, benzene measurements exceeded the US EPA National Drinking Water maximum contaminant level of 5 ppb (Table 2).

Exposure pathways and mechanisms

Exposure to BTEX can occur through inhalation, dermal contact, transplacental or oral exposure from drinking water (167). While several studies have established the interactive effect of BTEX neurotoxicity, more research is needed on its mechanisms. Proposed mechanisms include oxidative DNA damage, apoptosis and fragmentation resulting in changes in signaling pathways, and cellular homeostasis (62), (168). Being highly lipophilic, these substances have easy access to the CNS.

Neurological and neurodevelopmental effects

Analyzing the neurotoxic and developmental effects of these compounds is challenging, especially in humans. Frequently, exposures are not specific, and exposure to one or more BTEX compounds might also accompany exposures to other neurotoxic compounds such as formaldehyde. Most of the human studies have employed occupational settings [which are generally airborne concentrations averaged over a period, either a long-term exposure limit (8 h time weighted average) or a short-term exposure limit (15 min period) and inhalant abusers (episodic binge exposures to high concentrations] (169), (170).

Low to moderate concentrations of toluene have been linked with impaired cognitive, auditory and neuromuscular functions. Studies in experimental animals have found that exposure to both benzene and toluene can affect spatial learning and memory, even at levels considered subtoxic (60). Toluene exposure has been shown to cause intellectual, psychomotor, and neuromuscular impairment at moderate concentrations (80–150 ppm) (60). Infants of mothers that abused toluene during pregnancy showed delayed development of speech, motor function and had low scores on developmental tests. At high levels of exposure (1000–20,000 ppm), humans experience severe CNS dysfunction and in some cases can lead to permanent damage and death (60). In occupational settings “clinical neurobehavioral deficits” may be absent or low level, however when tested neurobehavioral performance is impaired (63). These performance tests give a continuous dimensional reading to the exposure and may be used to identify the beginnings of pathology not yet apparent (171). Some of these dimensions include motor performance, audiometry or tests of color vision. These dimensions were reflected in a study on young healthy volunteers given a single exposure to 200 ppm of toluene (172). A robust decline in a perceptual learning task was found amongst the study participants when a “distractor” was also present, a result that was not seen without the distractor (172). This illustrates the fact that relatively sophisticated testing must be conducted to document the subclinical effects of these toxicants (172).

Occupational exposure to xylene at the level of 14 ppm has been shown to cause anxiety, forgetfulness, difficulty with concentration, and other neurologic dysfunction (173). At more extreme (accidental) levels of exposure such as 10,000 ppm, amnesia, brain hemorrhage, unconsciousness and seizures have been reported (173). According to a 2015 review of approximately 150 peer-reviewed animal and human studies relevant to xylene toxicity, neurological effects include delayed nerve signaling, while neurobehavioral effects include tremors, altered vision, and numbness (62).

Xylene and toluene have also been associated with impaired neurodevelopment. Children exposed to toluene in utero have been reported to be born with small head circumference, serious facial deformations, and general growth retardations. A follow-up study at 3 years of age showed that these same children had hyperactivity, language impairment, developmental delays, postnatal growth retardation and cerebellar problems (57), (58). In another study of these same children, a high incidence of postnatal microcephaly and developmental delay was reported (59). It should be noted that these studies were primarily examining inhalant abusers and it is likely that the dosing was very high. Similar effects have been reported for xylene exposure (62).

Ethylbenzene’s neurotoxicity has been reported in animal and human studies. Neurologic effects associated with inhalation exposure to ethylbenzene has included hearing loss, with worsening of auditory response and changes to cochlear anatomy (174). Notably, these effects occurred even when exposed levels of ethylbenzene were lower than health standards established by OSHA and National Institute for Occupational Safety and Health (NIOSH) (174).

Neural tube defects (NTDs)

Studies have linked maternal benzene exposure to neural tube defects in both experimental animals and human infants. In Texas, a state that has very high ambient benzene levels, mothers living in areas with high benzene levels were reported to more likely have children with spina bifida than women living in areas with lower levels (61). In Sweden, women exposed to benzene and other organic solvents in biomedical research laboratories were found to be at an elevated risk for neural crest (cells which migrate through the embryo and engender several cell populations, including the peripheral nervous system) malformations during pregnancy (175). At a Marine Corps Base Camp in North Carolina, maternal exposure to benzene and trichloroethylene-contaminated drinking water was associated with NTDs (176). In a retrospective study of 124,842 births between 1996 and 2009 in rural Colorado, McKenzie et al. (12) found that the prevalence of NTDs were possibly associated with the highest exposure of mothers to natural gas development (based on density and proximity of natural gas wells within a 16-km radius of residence).

Health risks identified

Our review found that five pollutant categories are associated with increased neurological and neurodevelopmental problems in developing children: heavy metals (arsenic and manganese), particulate matter, BTEX, PAHs and EDCs. We found that these five pollutant groups are associated with neurotoxicity, neuroinflammation, psychomotor effects and neuromuscular effects. Some of these pollutant categories are also linked with neural tube defects and neurodevelopmental effects such as impaired memory, intellectual function, learning and cognitive function. Finally, we also found that some of these pollutants are associated with brain disorders, adverse neuropsychological effects, and behavioral effects including impulsivity, aggression, hyperactivity and ADHD (Table 1).

Given the profound sensitivity of the developing brain and central nervous system, it is reasonable to conclude that young children who experience frequent exposure to these pollutants are at particularly high risk for chronic neurological diseases.

Susceptibility during prenatal, postnatal and reproductive years

The entire nervous system, consisting of the brain, spinal cord and peripheral nerves, is highly vulnerable to environmental toxicants especially during development (74), (177), (178). During critical stages of cellular growth, migration and synaptic organization, even subtle disruptions can have profound and reorganizational effects on neurodevelopment and can result in permanent brain damage or neural impairments (74), (177), (178). Numerous studies have repeatedly shown that chemical exposures that produce little to no perceptible adverse effects in an adult brain can significantly impact neurodevelopment (71), (178). The human brain undergoes substantial growth and development during the prenatal period. Beginning with the formation of the neural tube at 2 weeks of life and continuing through until birth, the development of the central nervous system involves the formation of 100 billion neurons and more than 100 trillion synaptic connections (179). It is therefore not surprising that the prenatal window is considered one of the periods of the greatest vulnerability to neurotoxicants, neuroendocrine disruptors and other environmental insults (179), (180), receptivity (143). Although the developing fetus was once thought fully protected from exogenous exposures, it is now clear that the placenta and fetus are unable to block the passage of many of the environmental toxicants to which the mother is exposed. Within the developing brain, billions of cells must be precisely located, interconnected and specialized. To achieve this, neurons need to develop and migrate along precise pathways. The ability to repair any anomaly is unlikely, if at all possible, which can lead to permanent deficits (72), (73), (181), (182).

There are many mechanisms by which typical development can be interrupted. Mounting research has detected more than 200 such chemicals in umbilical cord blood, showing that babies are being born “pre-polluted” with a host of chemicals, some of which are neurotoxic (71). Routes of fetal exposure include transplacental transmission, wherein compounds pass through the placental barrier and reach the embryo and/or fetus by mimicking or binding to essential compounds (183), through ingestion of amniotic fluid and via transdermal exposure (184), (185). At least some of these compounds reach the fetal brain. The BBB which imparts some protection from toxicant exposure, is not fully formed until approximately 6 months after birth, leaving the developing brain particularly susceptible to exposure during prenatal and neonatal life (72).

The period after birth is also considered an enhanced window of vulnerability, though for different reasons. First, the brain is still remodeling and organizing, which it will continue to do long into early adulthood. It is thus still susceptible to developmental disruption, but via potentially different mechanisms and in distinct regions (186). Second, exposure pathways for neurotoxicants are heightened during infancy and childhood compared to later in life. Children experience greater internal exposures to toxicants than adults because they eat, drink and breathe more toxicants per pound of body weight than adults (187). This is the result of having greater interaction with chemical toxicants (from spending more time on the floor and putting objects in mouths), and having different morphometry compared to adults (187). Children are often not able to detoxify, metabolize or excrete toxicants as efficiently as adults can due to their underdeveloped metabolic systems (187). Even as children grow into adolescents, their developing central nervous system remains vulnerable to environmental toxicants. This is because synaptic remodeling, myelination and additional developmental processes continue in cortical regions as well as other areas of the brain fundamental to cognitive reasoning, executive function, impulse control and other high-level behaviors (188). Chronic exposures, even at low levels, are particularly concerning for children because they are still developing and have years of life remaining over which to be exposed. Thus, there is heightened risk of developing chronic diseases later in life (189).

In our discussion of neurotoxicity, it is important to keep in mind that the factors of dose, duration, and frequency of exposure to neurotoxicants interplay to influence their ultimate effect in the developing brain (190). One reason why the pandemic of neurotoxicity was first called “silent” is that some neurotoxin effects are subtle and thus discerned only through special testing rather than standard examination, which is designed to detect clinical neural disorders.

Finally, it is important to understand the distinction between neurotoxicity (which results in cell death) and neurodevelopmental impact (which can constitute organizational and other changes within the CNS without obvious pathology). Although the specific mechanisms by which chemicals alter brain organization and function remain largely unresolved, subclinical effects may reflect a dose-dependent continuum of toxic effects, wherein low doses may cause surprisingly large functional decrements (72). The rapidly growing body of literature on the neurodevelopmental consequences of neurotoxicant exposures in early life emphasizes their likely role in the etiology of neurodevelopmental as well as neurodegenerative disorders later in life (71), (179), (191).

When provided by the literature, we discuss in this review the biological mechanisms of neurotoxicity and neurodevelopmental disorders (oxidative stress, DNA damage, apoptosis, etc.). Measurements of cognitive and neurobehavioral deficits in the literature include the IQ test, the Wechsler Scale of Intelligence for Children (WISC-IV), and the Child Behavior Checklist (CBCL), a widely used method for identifying “problem” behavior in children, such as aggressiveness (192).

The significant impact of pollutants on neurodevelopment

The Collaborative on Health and the Environment’s 2008 Consensus Statement on Environmental Agents also examined the role of environmental agents on neurodevelopmental disorders (193). They concluded that the existing animal and human studies suggest a greater proportion of development is environmentally influenced than has been generally believed and note the serious implications for families, schools, local communities and society at large (193). Additionally, in another paper Bellinger (194) assessed the risk of different factors on neurodevelopment and full-scale IQ, and determined that environmental chemical exposures had a greater impact on the brain than pediatric conditions like traumatic brain tumors, brain injury and congenital heart disease.

Any discussion on the topic of environmental effects of pollutants on neurodevelopment must carefully address the developmental status of the fetus or young child and the effects of exposure on certain stages of brain development, the variation in exposure routes between children and adults, the endpoints used to measure effects, the time frame used to measure effects factoring in the “downstream” effect of exposures on the developing organism, and various other issues (178). Unfortunately, regulatory standards have not been developed to sufficiently account for these questions.

The significant impact of neurotoxicants on IQ is attributed not to the magnitude of the effect on the individuals, but rather to the prevalence of exposure across the population (194), (195). Small decrements in IQ could go unnoticed, depending on the specific study method. Since some disorders such as diabetes, depression, and hypertension can be clearly defined using “cutoff values of continuously distributed measurements”, some researchers argue that it is not always a question of whether an individual has a disorder but the extent to which the disorder is evident (194), (195). Neurodevelopmental disorders fit into this “continuously distributed dimensional category”. IQ is easily measured and has been studied for a long time, but other more complex disorders can be harder to quantitate (194), (195). Some of these are more common than easily measured and recognized disorders. Disorders that are more difficult to measure include social awareness and sensory integration. In an effort to strengthen the diagnostic practices of more subtle disorders, the National Institute of Mental Health has advocated for changes in the field of nosology, or the classification of diseases, and has termed the effort the Research Domain Criteria (RDoC) (196). This will focus on intrinsic biologically based symptoms and is likely to be independent of clinically defined disorders (197). Current epidemiology has been largely based on correlating risk factors with disorders, thus grossly underestimating the effects of many risks on the many aspects of brain development that are likely to be the building blocks of disorders (196).

General policy recommendations

Improved exposure assessment

While we strongly support a precautionary approach that prevents children’s exposure, we recommend that well-designed biomonitoring studies should be undertaken to measure existing exposures to pollutant groups associated with UOG. Currently, only a small number of studies document a causal relationship between pollution created by UOG operations and undesirable health outcomes. Better population exposure assessment is needed to document these relationships. The most accurate way to obtain information about human exposures from environmental pollution is through well-designed biomonitoring studies.

Mental health monitoring before and after UOG development

There should be the requirement by legislation or executive mandate that states monitor the mental health impacts of UOG development and operations, with an appointed external advisory panel of health experts, and paid for by a dedicated commonwealth revenue source such as a severance tax on UOG. Research should be integrated with the creation of an Unconventional Natural Gas and Oil Development Health Registry. The procedural guidelines should specify not only that children be included in all health studies, but as a vulnerable population they should be given specific attention as President Clinton directed in his Executive Order 13045 to reduce environmental health risks and safety risks to children.

Lack of transparency and research barriers

The 2005 Energy Policy Act, exempted hydraulic fracturing from the Underground Injection Control program and the EPA is prohibited from regulating fracking under the Safe Drinking Water Act (1974) (7). Due to these actions, a number of chemicals associated with UOG are not reported to the public. Disclosure of chemicals is critical in understanding the full scope of neurological health effects for infants and children. In a 2011 review about the human health effects of the 353 chemicals used in natural gas operations (as identified by Chemical Abstract Service numbers), Colborn et al. (1) determined that approximately 40–50% of the chemicals used could affect the brain and nervous system. However, the study was limited because of the lack of transparency about chemical mixtures used in the UOG process. The nondisclosure of these chemicals creates barriers to efficient research practices and contributes to the lack of knowledge concerning UOG and its potential health effects (7). Since a number of chemicals associated with UOG are not disclosed to the public, there exists uncertainty about the chemical makeup of UOG fluids. In many states, companies are not required to disclose information about the concentrations or what chemicals are used in the process because of trade secret protections.

Maximum contaminant levels

Researchers have reported that it is difficult to measure health risks from many of the compounds used in oil and gas development because many of them lack scientifically based maximum contaminant levels (7).

Review limitations considerations

This review is not exhaustive in scope. To make the review manageable, we focus on five particular pollutant groups of concern and do not discuss a number of other air and water pollutants that are known to cause neurodevelopmental harm, such as cadmium and methylene chloride, among others. Thus, the review is not comprehensive, but rather representative of a major issue.

The studies we reviewed evaluated exposures in a variety of settings. For example, some of these studies assessed atmospheric and water concentrations from UOG operations while others assessed emissions from oil and gas refineries as well as urban traffic. We also examined exposures from some of these pollutants in both indoor and outdoor settings. In some cases, where literature is lacking, we examined studies focusing on both conventional and unconventional sources of oil and gas development. The relevance to exposures near UOG sites varies.

This review is not intended to provide a formal risk assessment that would characterize the exposure levels among children to our pollutants of concern. Instead, we review studies that measure UOG air emissions and atmospheric concentrations of our five pollutants categories of concern.

Additionally, UOG is a recent development, and the most effective epidemiological studies will take a long time to complete. Further, there is still much that is unknown about neurological health effects and their relevance for children living near UOG areas. Though there is far more research than there was 5 years ago, there are still only a number of epidemiological studies that explore the associations between risk factors and health outcomes in people living close to UOG development.