JP8 exposure and neurocognitive performance among US Air Force personnel

NeuroToxicology 62 (2017) 170–180

Contents lists available at ScienceDirect

NeuroToxicology

Full Length Article

JP8 exposure and neurocognitive performance among US Air Force personnel Kristin J. Heatona,b , Alexis L. Maulea,b,c, Kristen W. Smithb,d, Ema G. Rodriguesb,d, Michael D. McCleanb , Susan P. Proctora,b,e,* a

Military Performance Division, US Army Research Institute of Environmental Medicine, 10 General Greene Avenue, Natick, MA 01760, USA Department of Environmental Health, Boston University School of Public Health, 715 Albany Street, Boston, MA 02118, USA Henry M. Jackson Foundation, 6720A Rockledge Dr. #100, Bethesda, MD 20817, USA d Department of Environmental Health, Harvard T.H. Chan School of Public Health, 677 Huntington Avenue, Boston, MA 02115, USA e Research Service, VA Boston Healthcare System, 150 South Huntington Ave., Boston, MA 02130, USA b c

A R T I C L E I N F O

Article history: Received 1 March 2017 Received in revised form 23 June 2017 Accepted 3 July 2017 Available online 4 July 2017 Keywords: Neurocognitive Total hydrocarbons Naphthalene 1-naphthol 2-naphthol Military

A B S T R A C T

Petroleum-based fuels such as jet propellant (JP) 4, JP5, JP8, and jet A1 (JetA) are among the most common occupational chemical exposures encountered by military and civilian workforces. Although acute toxicity following high-level exposures to JP8 and similar chemical mixtures has been reported, the relationship between persistent low-level occupational exposures to jet fuels and both acute and longerterm central nervous system (CNS) function has been comparatively less well characterized. This paper describes results of neurocognitive assessments acquired repeatedly across a work week study design (Friday to Friday) as part of the Occupational JP8 Exposure Neuroepidemiology Study (OJENES) involving U.S. Air Force (AF) personnel with varying levels of exposure to jet fuel (JP8). JP8 exposure levels were quantified using both personal air monitoring and urinary biomarkers of exposure. Neurocognitive performance was evaluated using an objective, standardized battery of tests. No significant associations with neurocognitive performances were observed between individuals having regular contact and those with minimal/no direct contact with JP8 (measured by average work week levels of personal breathing zone exposure). Also, no significant findings were noted between repeated measures of absorbed dose (multi-day pre-shift urinary 1- and 2-naphthol) and reduced proficiency on neurocognitive tasks across the work week. Results suggest that occupational exposure to lower (than regulated standards) levels of JP8 do not appear to be associated with acute, measurable differences or changes in neurocognitive performance. © 2017 Published by Elsevier B.V.

1. Introduction Petroleum-based fuels such as jet propellant (JP) 4, JP5, JP8, and jet A1 (JetA) are among the most common occupational chemical exposures encountered by military and civilian workforces. Routine occupational exposures involving military and commercial personnel may involve handling of jet fuel on a daily basis over the course of many years. Acute toxicity following high-levels of

* Corresponding author at: Military Performance Division, US Army Research Institute of Environmental Medicine, 10 General Greene Ave., Bldg. 42 Natick, MA 01760, USA. E-mail addresses: [email protected] (K.J. Heaton), [email protected] (A.L. Maule), [email protected] (K.W. Smith), [email protected] (E.G. Rodrigues), [email protected] (M.D. McClean), [email protected] (S.P. Proctor). http://dx.doi.org/10.1016/j.neuro.2017.07.001 0161-813X/© 2017 Published by Elsevier B.V.

exposure to raw fuel products, fuel vapor/aerosol, or products of fuel combustion have been described previously for both animals and humans (Carlton and Smith, 2000; Ritchie et al., 2003; Smith et al., 1997). However, the effects of persistent lower level occupational exposure over time (across a work week or across many years) to jet fuels on central nervous system (CNS) function have been, by comparison, less well characterized. Current exposure guidelines for jet fuel, established by the American Conference of Governmental Industrial Hygienists (American Conference of Governmental Industrial Hygienists (ACGIH), 2013), have been set at an 8-h time weighted average (TWA) of 200 mg/m3 for total hydrocarbons. These limits were established, in part based on known effects on CNS and peripheral nervous system (PNS) observed in both animals and humans exposed to levels of fuels exceeding this threshold (Anger, 1984). JP8 contains more than 200 aliphatic and aromatic solvent

K.J. Heaton et al. / NeuroToxicology 62 (2017) 170–180

compounds and a variety of nonhydrocarbon performance additives (Agency for Toxic Substances and Disease Registry (ATSDR), 2017; National Research Council (NRC), 2003). The organic solvents that are major components of JP8 and other fuels, such as benzene, ethylbenzene, toluene, xylene, and naphthalene, have known adverse effects on the human CNS and PNS resulting in neurobehavioral deficits, including degraded motor, learning, memory, attention, and visual-spatial performance, and reductions in processing speed (Bælum et al., 1985; Echeverria et al., 1991; Foo et al., 1990; Lotti and Bleecker, 2015; Meyer-Baron et al., 2008; Olson et al., 1985; Sainio, 2014; van Valen et al., 2012). The specific effects on CNS function depend largely on the chemical composition of the solvent mixture, the level of exposure and its duration (Sainio, 2014). In some cases, a brain syndrome known as chronic solvent encephalopathy (CSE) has been observed, characterized by impairments in attention, memory, motor performance and information processing speed and is often accompanied by neuropathological changes as observed on brain imaging (van Valen et al., 2012). (Pleil et al., 2000; Puhala et al., 1997). (Meyer-Baron et al., 2008). Although considered a safer fuel alternative than earlier jet fuel formulations (such as JP4 and JP5), JP80 s reduced volatility relative to other jet fuels (National Research Council (NRC), 2003) increases its persistence on skin and clothing, extending the possible duration of exposure and, relatedly, possible exposure dose (Pleil et al., 2000; Puhala et al., 1997). Acute exposures to JP8 at levels near or exceeding 350 mg/m3 and exceeding current exposure guidelines (i.e., 200 mg/m3) have been shown in both animal and in vitro toxicological studies to negatively impact immune, respiratory, and nervous system functions (National Research Council (NRC), 2003). Dizziness, headache, nausea, mental confusion, slurred speech, fatigue and gait instability have also been reported following acute JP8 exposure (Carlton and Smith, 2000; Smith et al., 1997). Moreover, persistent exposure (estimated to be 300 mg/m3 over 17 years on average) in humans to earlier formulations of jet fuel (MC-77, Swedish military fuel equivalent to JP-4) has been associated with dizziness, headache, nausea, neurasthenia, polyneuropathy (e.g., distal paresthesia and numbness, or paresis), and neurocognitive deficits primarily involving psychomotor speed and visual perceptual speed (Knave et al., 1979, 1978, 1976). More recently, subtle (subclinical) shifts in neurocognitive performance were observed between U.S. Air Force (AF) personnel with higher and lower levels of expected exposure to JP8 based on job category at the start of and across a 4–6 h work shift (Anger and Storzbach, 2001). In particular, workers in the higher exposure group compared to lower exposure group performed less well on neurocognitive tasks involving working memory (number sequence recall), psychomotor speed (coding), and motor skills. Across the work shift (from pre- post), those exposed to higher levels of naphthalene demonstrated significant changes in visuospatial memory and motor speed (tapping) performance relative to a comparison group of lower naphthalene exposed individuals. Subtle differences in reaction time and response accuracy have also been noted in Air National Guard personnel exposed to higher levels of JP8 (but below regulated exposure guidelines) as measured by exhaled breath sampling and based on job category (e.g., fuel cell workers) relative to those with expected lower levels of exposure to fuels (e.g., supply workers) (Tu et al., 2004). In this paper we examine the results of neurocognitive performance assessments acquired repeatedly across a standard work week with AF personnel who, as a function of their specific job characteristics, had higher or lower levels of exposure to jet fuel. This study, known as the Occupational JP8 Exposure Neuroepidemiology Study (OJENES), (Proctor et al., 2011) implemented

171

objective and individual-level measures of JP8 exposure within the context of a prospective research design. Sampling was conducted repeatedly across an entire work week, permitting analysis of whether average exposure dose as well as day-to-day exposure fluctuations influence key neurocognitive performance outcomes. Several papers have been published from the OJENES that provide detailed characterization of the individual-level exposure assessment methods and findings (Maule et al., 2016; Merchant-Borna et al., 2012; Rodrigues et al., 2014; Smith et al., 2010, 2012). In this study, we have two primary aims with respect to the effect of JP8 exposure on neurocognitive performance. First, we examined the relationship between objectively measured personal JP8 exposure and neurocognitive performance, hypothesizing that higher compared to lower or no personal exposure to occupational JP8 was associated with negative neurocognitive performances. Second, we predicted that higher, compared to lower, levels of JP8 repeatedly measured over a work week as determined by urinary biomarker levels of 1- and 2-naphthol would be significantly related to poorer neurocognitive functioning. We also examined the effect of length of AF service and JP8 exposure on neurocognitive performance outcomes to assess whether or not more years of service at higher exposure levels affected performance. 2. Materials and methods The study protocol was approved by institutional review boards at the US Army Research Institute of Environmental Medicine, USAF Research Laboratory at Wright Patterson Air Force Base and Boston University, and was in compliance with human subjects review procedure at the Centers for Disease Control and Prevention. All participants provided written informed consent prior to participation. 2.1. Study design and procedures Proctor and colleagues (Proctor et al., 2011) provide a detailed description of the design and methods of the consecutive 6-day study. In summary, participants were recruited from three AF bases and were eligible to participate if they did not have a selfreported history of loss of consciousness for more than 20 min or known neurological or psychological disorder(s). The recruitment process for this study was designed so that the study sample population included personnel with a range of exposures to JP8 based on their primary job activities. Individuals whose job activities involved routine exposure to JP8 (e.g., fuel cell repair and maintenance) were categorized as high exposure and those whose jobs did not involve routine contact with JP8 (e.g., medical technicians, administration) were categorized as low exposure. In this study, 38 of the 74 participants were considered to be in the high exposure group due to their job description and 36 were classified as being in the low exposure group as they had little-tono regular direct exposure to JP8 as a part of their job tasks. Data collection at each of the three AF base study sites (n = 21 at A, n = 20 at B, n = 33 at C) began at the end of each participant’s work shift on a Friday afternoon (Day 1) and continued Monday morning through Friday morning of the following work week (Days 2–6). All participants were given a brief neurological screening examination by a trained examiner in order to assess the presence of any potential neurologic impairment. As part of the study, questionnaires were administered to gather demographic information (e.g., age, sex, and education level), lifestyle behaviors (e.g., smoking and alcohol use), work history (e.g., current job, length of AF service), daily work activities (e.g., job tasks, workday schedule,

172

K.J. Heaton et al. / NeuroToxicology 62 (2017) 170–180

use of protective equipment), and current health symptoms (e.g. specific neurological health symptoms, functional health). Exposure to JP8 was assessed by the collection of repeated personal breathing zone air samples and urinary biomarkers of exposure (1- and 2-naphthol). Neurocognitive performance was ascertained from performance on a standardized battery of tests administered on Days 1, 2, 4, and 6 of the study. See Fig. 1 for a diagram outlining the study data collection design for the Study Days 1–6. 2.2. JP8 exposure assessment 2.2.1. Personal air sampling Work-shift personal breathing zone air samples were collected on four consecutive work days (Day 2–5). Participants wore a battery-operated personal air sampling pump (Casella Apex Pro IS; Casella USA, Amherst, NH) throughout their entire work shift. The sample pump was turned off during breaks from job tasks in which the participant left the work area (e.g., lunch, cigarette break) and when participants were required to put on respirators to perform certain job tasks (Merchant-Borna et al., 2012). Breathing zone air samples were collected and extracted in accordance with National Institute for Occupational Safety and Health Method (NIOSH) 1550 (National Institute for Occupational Safety and Health (NIOSH), 1994) for naphthalene and Occupational Safety and Health Administration (OSHA) Method 35 for total hydrocarbons (THC) (Occupational Safety and Health Administration (OSHA), 1982). Air samples were analyzed using gas chromatography/mass spectrometry (GC/MS) in selective ion monitoring (SIM) mode for naphthalene and THC. The 8-h time-weighted averages (TWA) for breathing zone air samples were determined for each Study Day measured (Days 2–5); personal air concentrations are presented as mg/m3 for THC and mg/m3 for naphthalene. In this study, to examine the relationship between JP8 exposure levels and neurocognitive proficiencies, we computed the average 4day personal breathing zone exposure level using the average of the daily 8-h TWA levels on Day 2 through Day 5 for THC and naphthalene for each participant and used these values as surrogate indicator metrics of higher or lower JP8 exposure. We chose this approach rather than the a priori high or low exposure group designations based on the results from previous work in this study (Merchant-Borna et al., 2012) that demonstrated while the a priori exposure group designations are significant predictors of measured exposure levels, they explain only 68% (THC) and 74% (naphthalene) of the between-worker objectively measured variability.

2.2.2. Urine sampling Urine samples were collected at the beginning and end of the work shift on all test days, starting at the end of shift on Day 1 and ending at the beginning of shift on Day 6 (Fig. 1). Samples were collected in 120-mL polyethylene cups and analyzed via GC/MS in SIM for 1-naphthol and 2-naphthol as measures of JP8 exposure (Serdar et al., 2003). Because the concentration of compounds measured in urine vary between and within individuals as a result of kidney output and level of hydration, urinary creatinine was analyzed and concentration levels adjusted accordingly (Rodrigues et al., 2014). In this study, creatinine adjusted, pre-shift urinary 1- and 2naphthol concentrations (mg/g creatinine) measured at the time of neurocognitive testing (on the morning of Day 2, Day 4, and Day 6) were examined to address the hypothesis that repeated, daily exposure is associated with reduced neurocognitive proficiency over a work week. Based on the reported elimination half-lives for 1- and 2-naphthols as being between 13 and 16 h (Smith et al., 2012), we rationalized that the pre-shift urine measures would be reflective of cumulative JP8 exposure from the day before. 2.3. Neurocognitive assessment The Day 1 Battery was administered after the work shift on the initial day of the study, and the Repeated Battery was administered at the start of shift on Day 2, Day 4, and Day 6 (Table 1). These batteries are described in greater detail in a previous report (Proctor et al., 2011). Both batteries included traditional examineradministered tests [Auditory Consonant Trigrams (Stuss et al., 1987), Hooper Visual Organization Test (Hooper, 1958), Hopkins Verbal Learning Test-Revised (Benedict et al., 1998; Brandt and Benedict, 2001), Wechsler Adult Intelligence Scale III Digit Span Test (Wechsler, 1997), Grooved Pegboard (Matthews and Klove, 1964)] and computer-based tasks (Automated Neuropsychological Assessment Metrics Version 4) [ANAM4; (Center for the Study of Human Operator Performance (C-SHOP), 2007)]. These tasks were selected to assess general academic knowledge [Shipley Institute of Living Scale, Vocabulary Subscale; (Shipley et al., 1946)] as well as specific domains of neurocognitive functioning including attention, memory, visuospatial ability, and psychomotor speed, each of which has been shown in previous research to be sensitive to the deleterious effects of VOCs/solvents on CNS function. All neurocognitive tasks were standardized, well-validated measures suitable for repeated administration in field settings where both time and environmental constraints limit the use of many standard

Fig. 1. Occupational JP8 Exposure Neuroepidemiology Study (OJENES) Design. Note: Personal air measurements (8-h TWA) from Day 2 to Day 5 were averaged to serve as an indicator of JP8 exposure for the cross-sectional analysis examining the relationship between JP8 exposure and neurocognitive performances. (See Methods Section 2.4.1.) Start of shift urine samples (creatinine-adjusted) on Days 2, 4, and 6 were used in the repeated measures analysis examining the relationship between work shift JP8 exposure and repeated neurocognitive performance over a work week period. (See Methods Section 2.4.2.)

K.J. Heaton et al. / NeuroToxicology 62 (2017) 170–180

173

Table 1 Neuropsychological Batteries.

Day 1 Battery Shipley Institute of Living Scale, Vocabulary Hooper Visual Organization Test Hopkins Verbal Learning Test-Revised Repeated Battery Auditory Consonant Trigrams ANAM 4 Match to Sample ANAM4 Simple Reaction Time ANAM4 Continuous Performance Test ANAM4 Finger Tapping WAISIII Digit Span Grooved Pegboard

Domain assessed

Task reference

General academic skills Visual-spatial ability Verbal learning, memory

(Shipley et al., 1946) (Hooper, 1958) (Benedict et al., 1998, Brandt and Benedict, 2001)

Working memory Visuospatial ability, working memory Attention, psychomotor speed Sustained attention Psychomotor speed Attention, working memory Fine motor skill

(Stuss et al., 1987) (Center for the Study of Human (Center for the Study of Human (Center for the Study of Human (Center for the Study of Human (Wechsler, 1997) (Matthews and Klove, 1964)

Operator Operator Operator Operator

Performance Performance Performance Performance

(C-SHOP), (C-SHOP), (C-SHOP), (C-SHOP),

2007) 2007) 2007) 2007)

The Day 1 Battery was administered at the end of shift on the initial day of the study (Friday). The Repeated Battery was administered at the start of shift on Day 2, Day 4, and Day 6.

Fig. 2. A. 1-Naphthol levels (pre- and post- shift) by Exposure Group measured Across the Study Period; B. 2-Naphthol levels (pre- and post- shift) by Exposure Group Measured Across the Study Period.

174

K.J. Heaton et al. / NeuroToxicology 62 (2017) 170–180

Table 2 Participant Demographics and Baseline Functioning Characteristics.

Breathing Zone 4-day Averages 8-h TWA Naphthalene (mg/m3) 8-h TWA THC (mg/m3) Demographics Age Education Years Active AF Service Months worked in current AF job Ave. hours of sleep per night Male Rank E2–E4 E5–E8 Ethnicity White (Caucasian) Non-White Current smoker Current drinker Head injury with LOC Familiarity with computers Not at all Somewhat Moderately Very familiar Shipley Scale, Vocabularyz PANAS Positive Affectz Negative Affecty MOS Cognitive Functioning Scalez SF12 Physical Component Summaryz SF12 Mental Component Summaryz

High exposure+ (n = 38)

Low exposure+ (n = 35)

Mean (SD) [Range]

Mean (SD) [Range]

6.16 (5.7) [0.3–22.4]** 7.62 (7.8) [0.3–33.7]**

0.64 (1.1) [0.3–5.7] 1.19 (1.8) [0.3–9.1]

25.4 (6.2) [18.6–40.8] 12.3 (0.9) [12–16] 5.3 (5.2) [0.5–17.0] 58.4 [6–204] 6.7 [5–8] N (%) 37 (97.4%)**

25.7 (5.9) [19.4–43] 12.5 (1.2) [12–18] 5.9 (5.4) [0.5–20.0] 55.4 [6–240] 6.6 [3–9] N (%) 24 (68.6%)

24 (63.2%) 14 (36.8%)

21 (60.0%) 14 (40.0%)

25 (65.8%) 13 (34.2%) 19 (50%) 26 (68.4%) 1 (2.6%)

23 (65.7%) 12 (34.3%) 13 (37.1%) 24 (68.6%) 1 (2.9%)

0 10 (26.3%) 19 (50.0%) 9 (23.7%) 28.9 (2.6) [22–34]

0 5 (14.3%) 13 (37.1%) 17 (48.6%) 29.2 (3.5) [23–36]

31.4 (6.8) [15–41] 13.6 (3.9) [10–25] 86.3 [10–100] 53.5 [31–61] 52.4 [14–62]

31.8 (6.9) [19–47] 13.9 (3.2) [10–25] 82.1 [25–100] 53.8 [41–61] 53.6 [26–63]

Baseline Functioning (assessed on Study Day 1) Hooper Visual Organization Test, mean # correct responsesz Hopkins Verbal Learning Test-Revised Total recallz Retention (%)z Recognition Discrimination Indexz

25.8 (1.9) [20–29]

26.6 (1.8) [21–30]

24.9 (4.1) [16–31] 86.8 (11.8) [63–100]* 10.4 (1.6) [6–12]

25.4 (3.5) [18–31] 92.3 (9.0) [71–100] 10.5 (1.0) [9–12]

Repeated Day Battery (first session assessment on Study Day 2) ANAM Match to Sample, throughput Continuous Performance Test (CPT) CPT response time (ms)y CPT false positive errorsy CPT non-response errorsy Simple Reaction Time, throughputz Finger Tapping, dominant hand (# of taps)z Finger Tapping, non-dominant hand (# of taps)z Other Tasks Auditory Consonant Trigrams, 36-s delay (#)z Auditory Consonant Trigrams, total (#)z WAISIII Digit Span, forward (total span)z WAISIII Digit Span, backward (total span)z Grooved Pegboard, dominant hand (time to complete, seconds)y Grooved Pegboard, non-dominant hand (time to complete, seconds)y

36.1 (10.4) [20.3–62.9]

37.4 (10.7) [17.6–59.8]

387.0 (75.8) [0–490.9] 0.3 (0.5) [0–2.6]* 0.1 (0.4) [0–2.6] 244.1 (25.3) [181–293] 64.5 (8.3) [52–86] 57.3 (8.3) [45–80]

409.1 (42.5) [312.8–500.0] 0.04 (0.2) [0–0.7] 0.1 (0.3) [0–2.0] 239.6 (27.3) [163–305] 62.1 (7.9) [49–82] 54.5 (7.3) [40–73]

43.6 (7.0) [25–57] 8.8 (3.0) [1–14] 11.3 (2.0) [6–16] 7.2 (2.0) [4–12] 67.2 (10.4) [40–95] 72.0 (10.2) [56–105]

42.4 (7.3) [29–56] 8.7 (3.0) [1–14] 10.4 (2.0) [7–15] 6.5 (1.9) [3–12] 66.5 (9.4) [50–95] 73.7 (12.6) [59–108]

+ High and low exposure groups from a priori categorizations based on job-type activities. *p < 0.05. **p < 0.001. z = higher value indicates better performance. y = lower value indicates better performance.

neurocognitive assessment approaches (White and Proctor, 1992; White et al., 1992, 1994). In addition to the tasks listed in Table 1, participants completed the Test of Memory Malingering (TOMM) (Tombaugh, 1996) to provide a measure of effort on Day 1. Insufficient effort was defined

as scores falling below 38 (out of 50) on the TOMM (O'Bryant et al., 2007). On all days of neurocognitive testing, alertness was assessed using the ANAM4 Sleepiness Scale and current mood was examined using the Positive and Negative Affect Scale (PANAS) (Watson et al., 1988).

K.J. Heaton et al. / NeuroToxicology 62 (2017) 170–180

2.4. Data analyses Seventy-three out of the 74 consented participants completed each of the four scheduled neuropsychological batteries. None of these participants met the criteria for insufficient effort based on Day 1 performance on the TOMM. Thus, analyses to address the study hypotheses were conducted on 73 participants. A primary goal of this study was to examine potential functional effects of JP8 exposure within the context of a repeat test scenario across a standard work week. For this study, a sample size of approximately 35 persons per group was considered sufficient to detect observable functional differences in neurocognitive performances (defined as 15–25% differences depending on the task) in a repeated measures design. The study was not powered to examine subclinical (less than 10–15%) differences in neurocognitive task performance. Demographic data of the higher and lower exposure groups were compared using Student’s t-test for continuous variables or chi-squared statistics for categorical variables. Personal breathing zone exposure values were log-transformed for data analyses, as the personal air measurements were skewed. Analyses of variance and linear tests for trend were used to examine changes in urinary biomarkers and breathing zone levels of THC and naphthalene over the work week. Graphical presentation of the pre- and post-shift urinary naphthol concentrations are depicted in Fig. 2A and B. As is often applied when analyzing neuropsychological outcomes, test scores were winsorized at 3 standard deviations (SD) from the mean to minimize the effect of extreme values on the results and retain sample size or power (Tabachnick and Fidell,

175

2007). That is, individual scores above or below 3 SD from the mean were replaced with the value of the mean score 3 SD and remained in the analyses (they were not excluded); in these analyses, this was applied to approximately 1% of all individual task performance scores. 2.4.1. Examination of relationship between average JP8 exposure and neurocognitive performance Generalized linear modeling was performed to examine the cross-sectional relationship between average breathing zone exposure levels for THC and naphthalene and neurocognitive task performance outcomes. These analyses address the first study aim to evaluate if higher compared to lower (or no) personal exposure to occupational JP8, determined by average work week exposure, was associated with negative neurocognitive performances. Separate models were performed for each neurocognitive task at the first administration time point (all Day 1 tasks and Day 2 task performance on the Repeated Battery). General intelligence (Shipley score), education, sex, age, years of AF service, enlisted status/rank, and race were considered as possible covariates. Age, education, and sex are known predictors of performance on many cognitive tasks, including those administered in this study (Lezak, 2004). As all participants had a high school education, we selected the Shipley score as a surrogate education measure reflective of general knowledge/intelligence. In addition, the range in age within this AF sample was not wide, thus years of AF service was considered in the analyses. However, years of service could serve as a potential confounder of the relationship between exposure and outcome (healthy worker effect). Thus, we examined the change in

Table 3 Repeated Exposure and Neurocognitive Battery Performance. Monday Day 2 (n = 73)

Wednesday Day 4 (n = 73)

Friday Day 6 (n = 69)

Mean (SD)

Mean (SD)

Mean (SD)

2.42 (2.8) 2.71 (3.6)

4.49 (4.6) 2.79 (3.4)

4.41 (3.8)* 2.92 (3.4)a

5.57 (8.7) 4.62 (4.2)

4.91 (4.1) 4.43 (3.6)

5.09 (3.6) 5.11 (3.9)a

36.76 (10.5)

36.46 (12.3)

38.54 (12.4)

397.64 (62.7) 0.38 (1.5)b 0.26 (1.6) 241.94 (26.2) 63.36 (8.1) 55.93 (7.9)

403.6 (44.9) 0.52 (1.6) 0.33 (1.5) 237.66 (30.9) 63.33 (8.7) 56.08 (8.1)

420.44 (46.5) 0.58 (1.6) 0.32 (1.5) 248.45 (26.9) 63.33 (9.8) 55.78 (8.3)

Other Tasks Auditory Consonant Trigrams, 36-s delay (#)z Auditory Consonant Trigrams, total (#)z WAISIII Digit Span, forward (total span)z WAISIII Digit Span, backward (total span)z Grooved Pegboard, dominant hand (time to complete, seconds)y Grooved Pegboard, non-dominant hand (time to complete, seconds)y

8.74 (3.0) 43.01 (7.1) 10.86 (2.1) 6.85 (2.0) 66.86 (9.9) 72.79 (11.4)

9.92 (3.2) 46.11 (7.6) 11.30 (2.0) 7.78 (2.2) 64.16 (9.1) 70.10 (8.5)

9.87 (3.1) 48.29 (6.5) 11.46 (2.3) 7.86 (2.2) 63.55 (7.4) 68.29 (8.1)

Current mood PANAS, positive affectz PANAS, negative affecty

28.38 (8.3) 13.22 (3.4)

25.5 (9.1) 12.91 (3.2)

26.36 (10.2) 12.04 (2.5)

Pre-shift Urinary Naphthols 1-naphthol (mg/g creatinine) Higher exposure (n = 38) Lower exposure (n = 35) 2-naphthol (mg/g creatinine) Higher exposure (n = 38) Lower exposure (n = 35) ANAM4 Battery Match to Sample, throughputz Continuous Performance Test (CPT) CPT response time (ms)y CPT false positive errorsy CPT non-response errorsy Simple Reaction Time, throughputz Finger Tapping, dominant hand (# of taps)z Finger Tapping, non-dominant hand (# of taps)z

*p < 0.05. n = 32; b n = 72. z = higher value indicates better performance. y = lower value indicates better performance. a

176

K.J. Heaton et al. / NeuroToxicology 62 (2017) 170–180

estimates with and without this variable in the models and found no significant impacts. The final model included the Shipley score, sex, and years of AF service. Within the generalized linear models, the exponeniated beta coefficients represent the change in cognitive performance relative to one unit of exposure (1 mg/m3 THC; 1 mg/m3 naphthalene). If a significant independent effect of exposure on outcome was observed in the initial model, the interaction effect between exposure and years of AF service was examined in a second model to determine whether more years of AF service among those with higher JP8 exposure levels was associated with reduced neurocognitive proficiency. 2.4.2. Examination of relationship between repeated, work week exposure dose and neurocognitive performance We examined the relationship between creatinine-adjusted pre-shift 1- and 2-naphthol levels (separate models) on neurocognitive outcomes via repeated measures, generalized linear modeling with Study Day, sex, and years of AF service as covariates. Pre-shift Day 2 levels of 1- and 2-naphthol were used as the referent in the respective models. The model beta coefficients

represent the change in cognitive performance for every one mg/g creatinine increase in daily pre-shift naphthols. When results showed a significant independent effect of exposure on neurocognitive task performance in the initial model, the interaction between exposure dose and Study Day was tested in a second model to determine if repeated JP8 exposure over a work week was associated with reduced neurocognitive proficiency. 2.4.3. Sensitivity analyses Additional generalized linear modeling was performed examining the relationship between a priori higher and lower exposure group designations and neurocognitive task performance outcomes. Also, examination of the pre- and post-shift 1- and 2naphthol concentration patterns over the study period (Fig. 2A and B) suggest that the naphthol half-lives may have shorter than the anticipated 13–16 h (Smith et al., 2012). Therefore, post hoc repeated measures, generalized linear modeling was performed examining the Repeated Battery, using the post-shift urinary 1and 2-naphthol concentrations (mg/g creatinine) measured at the end of shift on Days 3 and Day 5 and the pre-shift Day 2 level was

Table 4 Average Work Week Exposure and Neurocognitive Performance on Baseline Battery (n = 73). Model 1 Outcome of interest

Hooper Visual Organization Test Mean # correct responsesz

Intercept b (95% CI)

Model 2 Average Daily Naphthalene b (95% CI)

21.65 (17.4, 25.9) 0.21 (0.5, 0.1)

Hopkins Verbal Learning Test  Revised Total recallz 13.58 (4.9, 22.2)

0.29 (0.9, 0.4)

Retention (%)z

1.71 (3.6, 0.1)

Recognition Discrimination Indexz

68.89 (43.8, 94.0) 8.05 (4.8, 11.2)

ANAM4 Matching 2 Sample, throughputz

22.23 (2.3, 46.7) Continuous Performance Test, mean 458.71 (314.4, reaction timey 603.0) 236.09 (179.1, Simple Reaction Time, throughputz 293.1) Finger tapping, dominant hand (# taps)z 57.55 (39.2, 75.9) Finger tapping non-dominant hand (# 51.70 (33.6, 69.8) taps)z Auditory Consonant Trigrams # correct, 36 s delayz Total # correctz

WAISIII Digit Span Forward, # correctz Backward, # correctz

0.03 (0.2, 0.3)

0.29 (2.1, 1.5) 2.39 (8.4, 13.2) 2.22 (2.0, 6.5) 0.17 (1.2, 1.5) 0.17 (1.2, 1.5)

Years of AF Service b (95% CI)

Intercept b (95% CI)

0.03 (0.1, 0.1)

21.65 (17.4, 25.9) 0.22 (-0.5, 0.1)

0.03 (0.1, 0.1)

0.05 (0.2, 0.1) 0.45 (1.0, 0.1) 0.04 (0.1, 0.02)

13.50 (4.8, 22.2)

0.19 (-0.8, 0.4)

0.04 (0.2, 0.1) 0.42 (1.0, 0.1) 0.04 (0.1, 0.03)

Average Daily Total Hydrocarbons b (95% CI)

68.52 (43.1, 93.9) 1.16 (-3.0, 0.7) 7.99 (4.8, 11.2)

0.10 (0.6, 21.51 (3.0, 0.4) 46.0) 1.85 (1.2, 4.9) 458.81 (314.5, 603.1) 0.23 (1.0, 1.4) 236.72 (179.6, 293.8) 0.17 (0.2, 0.6) 58.11 (39.8, 76.4) 0.21 (0.2, 0.6) 51.78 (33.7, 69.9)

0.13 (0.3, 7.87 (0.9, 14.8) 0.02) 0.04 (0.3, 0.4) 39.56 (23.2, 55.9)

0.08 (0.2, 0.3)

2.34 (8.5, 13.2)

0.08 (0.6, 0.4) 1.86 (1.2, 4.9)

1.87 (2.4, 6.2)

0.22 (1.0, 1.4)

0.15 (1.5, 1.2) 0.12 (1.2, 1.5)

0.16 (0.2, 0.5) 0.21 (0.2, 0.6)

0.10 (0.4, 0.6)

0.13 (0.3, 0.01) 0.04 (0.3, 0.4)

0.13 (1.7, 2.0)

7.79 (0.9, 14.7)

0.14 (0.4, 0.7)

39.67 (23.3, 56.0)

0.40 (0.8, 1.6)

8.88 (4.4, 13.4) 4.43 (0.2, 8.7)

0.31 (0.02, 0.7) 0.17 (0.2, 0.5)

0.02 (0.1, 0.1) 0.08 (0.01, 0.2)

1.26 (2.9, 0.3)

0.10 (0.4, 0.6) 62.17 (40.7, 83.6) 1.34 (3.0, 0.3)

2.32 (4.2, 0.5)*

0.19 (0.3, 0.7) 91.41 (66.4, 116.5)

Grooved Pegboard Dominant hand, mean time to completey 62.05 (40.6, 83.5) 91.77 (66.9, Non-dominant hand, mean time to completey 116.6)

8.94 (4.4, 13.5) 4.37 (0.1, 8.6)

Years of AF Service b (95% CI)

0.47 (0.8, 1.7)

0.28 (0.1, 0.6) 0.21 (0.1, 0.5)

2.13 (4.0, 0.2)*

*p < 0.05. Model 1 examines Naphthalene exposure and performance. Model 2 examines Total Hydrocarbon exposure and performance. Both Models 1 and 2 are adjusted for sex, Shipley scale summary score (general intelligence), and years of AF service. z = higher value indicates better performance. y = lower higher value indicates better performance.

0.02 (0.1, 0.1) 0.08 (0.01, 0.2)

0.09 (0.4, 0.5) 0.19 (0.3, 0.7)

K.J. Heaton et al. / NeuroToxicology 62 (2017) 170–180

used as the referent given that no measurements were collected during the weekend. 3. Results Participants who completed the neuropsychological testing batteries in the higher and lower JP8 exposure groups were similar in age, education, Shipley Scale score, years of AF service, rank, and ethnicity (Table 2). There were significantly more males in the higher JP8 exposure group compared to the lower JP8 exposure group. And, approximately 20% (n = 15) of the participants had 10 or more years of AF service. Individuals in the high JP8 exposure group had significantly higher 4-day average personal breathing zone exposure (i.e.,

177

current exposure) for both naphthalene and THC (p < 0.001) relative to those categorized into the low JP8 exposure group (Table 2). Measured levels were well below the regulated guideline of 200 THC mg/m3. Across the work week, significant increases in personal air levels of THC (F = 10.29, p = 0.002) were found among the higher exposure group but not among the low exposure group. No significant increases in personal air levels of naphthalene were observed in either a priori exposure group over the work week. Pre-shift 1-naphthol levels significantly increased (F = 4.70, p = 0.03) among the higher exposure group across the work week, but no significant increases were observed in the low exposure group (Table 3). In addition, no significant changes across the work week were observed for pre-shift 2-naphthol levels in either the higher or lower JP8 exposure groups.

Table 5 Repeated, Pre-shift Urinary Naphthols and Repeated Neurocognitive Performance. Model 1 Outcome of interest

Model 2 &

Intercept

1-naphthol

Study Day

Intercept

2-naphthol

Study Day&

b (95% CI)

b (95% CI)

b (95% CI)

0.05 (0.3, 0.2)

Day 4: 0.38 (2.2, 1.4)

b (95% CI)

b (95% CI)

b (95% CI)

ANAM4 Matching 2 Sample, throughputz

23.86 (0.3, 47.4)

0.07 (0.4, 0.5)

Day 4: 0.44 (2.3, 1.4) 24.51 (0.9, 48.2)

Continuous Performance Test, mean reaction timey

511.17 (422.9, 599.5)

Simple Reaction Time, throughputz

Finger tapping, dominant hand (# taps)z

Finger tapping non-dominant hand (# taps)z

Auditory Consonant Trigrams # correct, 36 s delayz

Total # correctz

Day 6: 1.81 (0.8, 4.4) 0.21 (1.5, 1.9) Day 4: 4.55 (6.7, 15.8) 510.42 (420.7, 600.2) Day 6: 22.59 (10.6, 34.6)** 221.82 (172.0, 0.02 (1.2, 1.2) Day 4: 4.79 (11.5, 220.88 (170.7, 271.7) 1.9) 271.0) Day 6: 4.85 (2.5, 12.1) 56.21 (38.1, 74.4) 0.20 (0.5, Day 4: 0.44 (0.6, 1.5) 53.84 (35.0, 72.7) 0.1) Day 6: 0.40 (1.1, 1.9) 47.10 (30.0, 64.2) 0.13 (0.4, Day 4: 0.47 (0.6, 1.5) 47.14 (29.8, 64.5) 0.1) Day 6: 0.19 (1.1, 1.5)

8.25 (2.8, 13.7)

0.07 (0.2, 0.1)

36.63 (23.2, 50.1) 0.11 (0.4, 0.2)

Day 4: 1.41 (0.6, 2.2)**

8.18 (2.6, 13.7)

Day 6: 1.23 (0.4, 2.1)** Day 4: 3.61 (2.0, 5.2)**

35.71 (22.1, 49.3)

0.14 (1.4, 1.7)

Day 6: 22.85 (11.1, 34.6) ** 0.11 (0.6, 0.9) Day 4: 4.7 (11.3, 1.9)

0.14 (0.02, 0.3) 0.03 (0.2, 0.1)

7.60 (3.4, 11.8)

0.0004 (0.1, 0.1)

Backward, mean # correctz

4.35 (0.4, 8.3)

0.06 (0.02, 0.2)

0.01 (0.1, 0.1)

7.85 (3.6, 12.1)

0.02 (0.1, 0.04)

Day 6: 0.57 (0.1, 1.0)* Day 4: 0.91 (0.4, 1.4)**

4.23 (0.3, 8.2)

0.03 (0.03, 0.1)

Non-dominant hand, mean time to completey

(4.9,

Day 4: 0.37 (0.01, 0.7)* Day 6: 0.57 (0.1, 1.0)* Day 4: 1.00 (0.5, 1.5)** Day 6: 1.06 (0.6, 1.6)**

66.91 (50.5, 83.3) 0.09 (0.3, 0.2)

(5.7, (5.0,

Day 4: 1.32 (0.5, 2.1)**

Day 6: 5.37 (3.6, 7.1)**

Day 4: 0.38 (0.01, 0.8)*

64.84 (48.7, 80.9) 0.16 (0.2, 0.5) Day 4: 2.80 0.7)** Day 6: 3.79 1.9)** 81.62 (64.0, 99.2) 0.11 (0.2, 0.5) Day 4: 2.82 0.6)* Day 6: 4.83 2.5)**

Day 6: 0.15 (1.3, 1.6) Day 4: 0.31 (0.7, 1.3)

Day 6: 1.14 (0.3, 2.0)* 0.06 (0.1, 0.3) Day 4: 3.51 (1.9, 5.1)**

Day 6: 0.98 (0.5, 1.5)** Grooved Pegboard Dominant hand, mean time to completey

Day 6: 4.87 (2.3, 12.0) Day 4: 0.27 (0.7, 1.3)

Day 6: 0.03 (1.2, 1.3)

Day 6: 5.51 (3.7, 7.3)** WAISIII Digit Span Forward, mean # correctz

Day 6: 1.90 (0.7, 4.5) Day 4: 4.85 (6.3, 16.0)

81.20 (63.3, 99.1)

(7.2,

Urinary naphthol levels from Pre-shift Day 2, Pre-shift Day 4, Pre-shift Day 6. & Study Day 2: Reference group. *p < 0.05. ** p < 0.01. Model 1 examines 1-Naphthol (mg/g creatinine) and performance. Model 2 examines 2-Naphthol (mg/g creatinine) and performance. Both Models 1 and 2 are adjusted for sex, Shipley scale summary score (general intelligence), and years of AF service. Neurocognitive battery repeated on Day 2 (n = 73), Day 4 (n = 73), and Day 6 (n = 69) of study protocol. z = higher value indicates better performance. y = lower higher value indicates better performance.

Day 4: 2.64 0.6)* Day 6: 3.59 1.8)** 0.07 (0.2, 0.4) Day 4: 2.66 0.5)* Day 6: 4.69 2.4)**

(4.7, (5.4, (4.8, (7.0,

178

K.J. Heaton et al. / NeuroToxicology 62 (2017) 170–180

On Day 1 and on all subsequent neurocognitive test days, current mood on the PANAS was not significantly associated with JP8 exposure. In addition, self-reported alertness, assessed via the ANAM4 Sleepiness Scale as part of the repeated day neurocognitive battery (pre-shift ratings Day 2: 2.95 (1.2); Day 4: 2.95 (1.3); Day 6: 2.78 (1.3)) did not change to a great extent over the work week (F = 0.42, p = 0.655) and was not significantly associated with JP8 exposure. Table 3 presents the repeated day neurocognitive testing results over the work week (mornings on Mon (Day 2), Wed (Day 4), and Fri (Day 6)). Overall, although not significant, improvements in performance over the repeated testing days (from Day 2 to Day 6) were observed on most tests (Match to Sample, Simple Reaction Time, Auditory Consonant Trigrams, WAIS III Digit Span, and Grooved Pegboard). However, performance on the Continuous Performance Task demonstrated slower response times over the work week. 3.1. Examination of the relationship between average work week exposure and neurocognitive performance There were no significant associations between the average JP8 exposure level and reduced proficiency in neurocognitive performance (Table 4), although higher average levels of both naphthalene and THC were significantly related to faster time to completion with the non-dominant hand on the Grooved Pegboard task. For example, Grooved Pegboard time to completion decreased 0.28 s (dominant hand) and 0.10 (non-dominant hand) for a 1 mg/m3 increase in average naphthalene exposure and 0.26 s (dominant hand) and 0.12 s (non-dominant hand) for every 1 mg/ m3 increase of average THC exposure. Number of years of AF service was not significantly related to neurocognitive performance. General academic (verbal) intelligence, measured using the Shipley Institute of Living Scale, was significantly related to better performance on the Hooper Visual Organization Test (# correct: bNAPmodel = 0.16, p < 0.05; bTHCmodel = 0.17, p < 0.05) and total recall on the Hopkins Verbal Learning TestRevised (bNAPmodel = 0.40, p < 0.05; bTHCmodel = 0.40, p < 0.05). Males compared to females were observed to have slower times to complete the Grooved Pegboard Test (dominant hand: bNAPmodel = 9.89, p < 0.01; bTHCmodel = 9.91, p < 0.01; non-dominant hand: bNAPmodel = 7.82, p < 0.05; bTHCmodel = 7.31, p < 0.05). As no significant effects were observed between average JP8 exposure levels and reduced proficiency in neurocognitive performance, the interaction effect between years of AF service and exposure was not examined. 3.2. Examination of relationship between repeated work day exposure dose and neurocognitive performance across a work week There were no significant associations between repeated measures of absorbed dose (pre-shift urinary 1- and 2-naphthol) and reduced proficiency in neurocognitive performance (Table 5). Results show that the strongest predictor of performance on the repeated battery tasks was Study Day, with results improving on most tasks on Day 4 and Day 6 of testing compared to Day 2 performance, specifically for the Auditory Consonant Trigrams total # correct, WAISIII Digit Span forward and backward, and Grooved Pegboard time to completion for both dominant and nondominant hands. For example, for the latter, in model 1, time to completion was 3.79 s and 4.83 s faster on Day 6 compared to Day 2 for the dominant and nondominant hands, respectively. However, a significantly poorer performance was observed for the Continuous Performance task over the work week as the mean reaction time increased by 22.59 ms and 22.85 ms on Day 6 compared to Day 2. But, as no significant effects were observed between repeated exposure and reduced proficiency in neurocognitive

performance, the interaction between Study Day and dose level was not examined. Number of years of AF service was not significant in the repeated exposure models. Repeated neurocognitive performance results showed that males performed slower on the Grooved Pegboard compared to females, but results were only significant for the dominant hand (b1-naphtholmodel = 6.03, p < 0.01; b2naphtholmodel = 6.33, p < 0.01). 3.3. Sensitivity analyses The results from the post hoc examination of the relationship between a priori higher and lower exposure groups and neurocognitive task performance outcomes demonstrated a significant association between JP8 exposure and reduced neurocognitive proficiency with respect to the mean number of correct responses on the Hooper Visual Organization Test (bhighexposure = 0.91, p < 0.05). There were significant associations between exposure and better performances on the WAISIII Digits total number of correct responses Forward and Backward (bhighexposure = 1.09, p < 0.05 and bhighexposure = 1.24, p < 0.05, respectively) (Supplement Table 4X). A significant association between repeated measures of absorbed dose and reduced proficiency in neurocognitive performance was observed between post-shift 2-naphthol and mean number of taps for the non-dominant hand on the Finger Tapping test (bhighexposure = 0.13, p < 0.001) (Supplement Table 5X). 4. Discussion In this study, we examined the effects of occupational exposure to jet fuel (JP8) on neurocognitive performance outcomes in AF personnel and found minimal evidence that JP8 exposure at levels below regulated guidelines significantly affects neurocognitive performances. Previously, we noted the differences in group-level neurocognitive performances at the beginning of the week-long assessment period compared to clinical test norms (Proctor et al., 2011) on a verbal fluency task (Hopkins Verbal Fluency Test). However, in the current analyses, when we examined the relationships between objective JP8 exposure levels and performances, we did not detect a pattern of evidence indicative of degraded neurocognitive performance associated with average exposure levels or with personal pre-shift JP8 exposure measures in this work-week study design. In post hoc analyses, we did observe reduced performances of visual memory (Hooper Visual Orientation Test) among the higher compared to the lower exposure group and motor speed (Finger Tapping, non-dominant hand) tasks related to higher exposure over a work week period. While visual memory and motor speed have been found to be associated with solvent exposure in the literature, it is possible the findings were due to chance in this study. Several factors may have contributed to the observed findings. For one, measured JP8 exposure levels, both area and individuallevel sampling measures, were within permissible occupational exposure limits across the study period and AF personnel were observed to follow AF safety guidelines for the safe handling of JP8 (e.g., use of personal protective equipment). Under such conditions, changes in neurocognitive performance across the work week would likely be subtle, if present. As previously mentioned, this study was not powered to detect subtle, subclinical shifts or differences in neurocognitive performance. However, both study groups (higher and lower exposure) were similar in terms of key demographic and work history variables. Thus, it is less likely that factors such as age, gender, and years served in the AF may have obscured detection of subtle shifts in neurocognitive function across the study assessment period (Proctor et al., 2011). Also,

K.J. Heaton et al. / NeuroToxicology 62 (2017) 170–180

improvements in neurocognitive test performance were noted across repeated test administrations over the work week on some tasks (working memory, motor control/reaction time). Improved performance with repeated task exposures, or practice effects, are common to most performance-based tasks, including neurocognitive tests. Although participants in this study were provided an opportunity to become familiar with and practice the neurocognitive tasks in a single practice session prior to the start of the study, this familiarization session would not be expected to eliminate expected improvements in performance associated with practice entirely (Calamia et al., 2012; Collie et al., 2003). In an appropriately designed repeated measures study, one in which the exposed and comparison groups are similar in all respects except for level of exposure, it would be expected that those with no to low exposure would display anticipated practice effects while those in the higher exposed group may not and thus function with reduced proficiency by comparison. In this study, we observed improvements in performance irrespective of exposure levels and did not observe significant performance differences related to JP8 exposure. Given limited access to participants due to work schedules and other logistical constraints, detailed, qualitative evaluations of individual exposure histories were not feasible in this study and the neurocognitive assessment battery was, by necessity, kept quite brief. Although abbreviated neurocognitive batteries can provide valuable information regarding general neurocognitive proficiencies across broad domains of function, they typically lack sensitivity and specificity to illuminate subtle differences in neurocognitive performance that a more detailed clinical neuropsychological evaluation can provide. In addition, previous research suggests that subtle deficits in neurocognitive proficiency may be revealed when neurocognitive resources are sufficiently challenged but are difficult to detect under less challenging conditions (Foo et al., 1990; Hänninen et al., 1976). It is possible that the neurocognitive tasks used in this study did not sufficiently tax neurocognitive resources to reveal potential exposure-related deficits. The OJENES project was designed to address a critical gap in current understanding of the effects of persistent, low-level exposures to solvent mixtures on CNS function. By conducting repeated, objective, standardized assessments of neurocognitive performance across a work week, OJENES provides an important extension of previous studies utilizing cross-sectional approaches (Tu et al., 2004). Carefully designed cross sectional studies, such as the 2000 AF study (Anger and Storzbach, 2001; Tu et al., 2004), coupled with detailed exposure level data can provide valuable information regarding current health status as a function of exposure history and current exposure levels. However, by design, they cannot characterize fluctuations in exposure level and neurocognitive function that likely occur over time. Repeated assessments address this shortfall by characterizing exposure and performance over time. In addition, although several studies have utilized inhalation and dermal exposure in combination with biological markers of JP8 exposure dose measured in urine and exhaled breath samples (Egeghy et al., 2003; Serdar et al., 2004), few studies have provided concurrent, standardized, repeated, objective measurement of neurocognitive performance together with reported information pertaining to potential exposuremodifying lifestyle factors (e.g., smoking and alcohol use (Söderkvist et al., 1996)). In the OJENES, participants completed measures of neurocognitive and general neurologic function, provided blood and urine samples to permit analyses of biological markers of JP8 exposure and answered questions regarding pertinent lifestyle factors, such as prior pesticide use, tobacco use, work history, and exposures to solvents via work activities or hobbies.

179

5. Conclusions In this study, no significant decrements in neurocognitive function were observed as related to JP8 exposure among individuals exposed to JP8 below regulated occupational exposure limits as a function of their AF job activities. Conflicts of interest statement The authors declare that there are no conflicts of interest. Acknowledgements We thank the U.S. Air Force personnel for their support and generous participation in the project. We are grateful to the additional Boston area study team personnel (N. Longcore, E. Kryskow, K. Merchant-Borna, H. MacDonald, and A. Graefe) and other personnel from local universities for their assistance in the data collection process at the respective US Air Force base locations. We acknowledge and appreciate the personnel at the Organic Chemistry Analytical Laboratory at the Harvard School of Public Health (HSPH), Centers for Disease Control and Prevention (CDC, Combustion Products and Persistent Pollutants Biomonitoring Laboratory and VOC and Perchlorate Laboratory) for the timely analyses of collected environmental and personal samples. We appreciate the assistance of Ms. Amanda Winkler in helping to prepare this manuscript for submission. This work was supported by the US Army Medical Research and Materiel Command through a grant award (W81XWH-06-1-0105; PI: S. P. Proctor) to the Henry M. Jackson Foundation for the Advancement of Military Medicine, Inc. The opinions or assertions contained herein are the private views of the author(s) and are not to be construed as official or as reflecting the views of the Army or the Department of Defense or Department of Veterans’ Affairs. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j. neuro.2017.07.001. References Agency for Toxic Substances and Disease Registry (ATSDR),, 2017. Toxicological Profile for Jet A, JP-5, and JP-8 Fuels. US Department of health and Human Services, Public Health Service, Atlanta, GA. American Conference of Governmental Industrial Hygienists (ACGIH), 2013. Kerosene/jet Fuels (CASRN 8008-20-6). Threshold Limit Values for Chemical Substances and Physical Agents and Biological Exposure Indices. ACGIH, Cincinnati, OH. Anger, W., Storzbach, D., 2001. Results and Discussion: Neurobehavioral  Interim Report. The Institute of Environmental, Human Health, Lubbock, TX, pp. 65–70. Anger, W.K., 1984. Neurobehavioral testing of chemicals: impact on recommended standards. Neurobehav. Toxicol. Teratol. 6 (2), 147–153. Bælum, J., Andersen, I., Lundqvist, G.R., Mølhave, L., Pedersen, O.F., Væth, M., Wyon, D.P., 1985. Response of solvent-exposed printers and unexposed controls to sixhour toluene exposure. Scand. J. Work Environ. Health 271–280. Benedict, R.H.B., Schretlen, D., Groninger, L., Brandt, J., 1998. Hopkins Verbal Learning Test Revised: normative data and analysis of inter-form and test-retest reliability. Clin. Neuropsychol. 12 (1), 43–55. Brandt, J., Benedict, R.H.B., 2001. Hopkins Verbal Learning Test-Revised. Psychological Assessment Resources, Inc., Lutz, FL. Calamia, M., Markon, K., Tranel, D., 2012. Scoring higher the second time around: meta-Analyses of practice effects in neuropsychological assessment. Clin. Neuropsychol. 26 (4), 543–570. Carlton, G.N., Smith, L.B., 2000. Exposures to jet fuel and benzene during aircraft fuel tank repair in the US Air Force. Appl. Occup. Environ. Hyg. 15 (6), 485–491. Center for the Study of Human Operator Performance (C-SHOP), 2007. ANAM4: Software User Manual. University of Oklahoma, Norman, OK.

180

K.J. Heaton et al. / NeuroToxicology 62 (2017) 170–180

Collie, A., Maruff, P., Darby, D.G., McStephen, M., 2003. The effects of practice on the cognitive test performance of neurologically normal individuals assessed at brief test-retest intervals. J. Int. Neuropsychol. Soc. 9 (3), 419–428. Echeverria, D., Fine, L., Langolf, G., Schork, T., Sampaio, C., 1991. Acute behavioralcomparisons of toluene and ethanol in human-subjects. Br. J. Ind. Med. 48 (11), 750–761. Egeghy, P., Hauf-Cabalo, L., Gibson, R., Rappaport, S., 2003. Benzene and naphthalene in air and breath as indicators of exposure to jet fuel. Occup. Environ. Med. 60 (12), 969–976. Foo, S.C., Jeyaratnam, J., Koh, D., 1990. Chronic neurobehavioral effects of toluene. Br. J. Ind. Med. 47 (7), 480–484. Hänninen, H., Eskelinen, L., Husman, K., Nurminen, M., 1976. Behavioral effects of long-term exposure to a mixture of organic solvents. Scand. J. Work Environ. Health 240–255. Hooper, H.E., 1958. Manual: Hooper Visual Organization Test. Western Psychological Services (WPS). Knave, B., Persson, H., Goldberg, M., Westerholm, P., 1976. Long-term exposure to jet fuel: an investigation on occupationally exposed workers with special reference to the nervous system. Scand. J. Work Environ. Health 2, 152–164. Knave, B., Olson, B.A., Elofsson, S., Gamberale, F., Isaksson, A., Mindus, P., Persson, H. E., Struwe, G., Wennberg, A., Westerholm, P., 1978. Long-term exposure to jet fuel: II. A cross-sectional epidemiologic investigation on occupationally exposed industrial workers with special reference to the nervous system. Scand. J. Work Environ. Health 19–45. Knave, B., Mindus, P., Struwe, G., 1979. Neurasthenic symptoms in workers occupationally exposed to jet fuel. Acta Psychiatr. Scand. 60 (1), 39–49. Lezak, M.D., 2004. Neuropsychological Assessment. Oxford University Press, USA. Lotti, M., Bleecker, M., 2015. Neurotoxicity of solvents. Occup. Neurol. 131, 93. Matthews, C., Klove, K., 1964. Instruction Manual for the Adult Neuropsychological Test Battery. University of Wisconsin Medical Center, Madison, WI. Maule, A.L., Proctor, S.P., Blount, B.C., Chambers, D.M., McClean, M.D., 2016. Volatile organic compounds in blood as biomarkers of exposure to JP-8 jet fuel among US air force personnel. J. Occup. Environ. Med. 58 (1), 24–29. Merchant-Borna, K., Rodrigues, E.G., Smith, K.W., Proctor, S.P., Mcclean, M.D., 2012. Characterization of inhalation exposure to jet fuel among U.S. air force personnel. Ann. Occup. Hyg. 56 (6), 736–745. Meyer-Baron, M., Blaszkewicz, M., Henke, H., Knapp, G., Muttray, A., Schaper, M., van Thriel, C., 2008. The impact of solvent mixtures on neurobehavioral performance  Conclusions from epidemiological data. Neurotoxicology 29 (3), 349–360. National Institute for Occupational Safety and Health (NIOSH), 1994. Naphthas: Method 1550, NIOSH Manual of Analytical Methods, 4th ed. National Institute for Occupational Safety and Health, Cincinnati, OH. National Research Council (NRC), 2003. Toxicologic Assessment of Jet-propulsion Fuel 8. The National Academies Press, Washington, DC. O'Bryant, S.E., Engel, L.R., Kleiner, J.S., Vasterling, J.J., Black, F.W., 2007. Test of memory malingering (TOMM) trial 1 as a screening measure for insufficient effort. Clin. Neuropsychol. 21 (3), 511–521. Occupational Safety and Health Administration (OSHA), 1982. Method 35. Analytical Methods Manual. OSHA Organic Methods Evaluation Branch, Salt Lake City, UT. Olson, B.A., Gamberale, F., Iregren, A., 1985. Coexposure to toluene and para-Xylene in man  central nervous functions. Br. J. Ind. Med. 42 (2), 117–122. Pleil, J.D., Smith, L.B., Zelnick, S.D., 2000. Personal exposure to JP-8 jet fuel vapors and exhaust at air force bases. Environ. Health Perspect. 108 (3), 183. Proctor, S.P., Heaton, K.J., Smith, K.W., Rodrigues, E.R., Widing, D.E., Herrick, R., Vasterling, J.J., McClean, M.D., 2011. The occupational JP8 exposure neuroepidemiology study (OJENES): repeated workday exposure and central nervous system functioning among US air force personnel. Neurotoxicology 32 (6), 799–808.

Puhala, E., Lemasters, G., Smith, L., Talaska, G., Simpson, S., Joyce, J., Trinh, K., Lu, J., 1997. Jet fuel exposure in the United States air force. Appl. Occup. Environ. Hyg. 12 (9), 606–610. Ritchie, G.D., Still, K.R., Rossi, J., Bekkedal, M.Y.V., Bobb, A.J., Arfsten, D.P., 2003. Biological and health effects of exposure to kerosene-based jet fuels and performance additives. J. Toxicol. Environ. Heal B 6 (4), 357–451. Rodrigues, E.G., Smith, K., Maule, A.L., Sjodin, A., Li, Z., Romanoff, L., Kelsey, K., Proctor, S., McClean, M.D., 2014. Urinary polycyclic aromatic hydrocarbon (OHPAH) metabolite concentrations and the effect of GST polymorphisms among US air force personnel exposed to jet fuel. J. Occup. Environ Med. 56 (5), 465–471. Söderkvist, P., Ahmadi, A., Åkerbäck, A., Axelson, O., Flodin, U., 1996. Glutathione Stransferase M1 null genotype as a risk modifier for solvent-induced chronic toxic encephalopathy. Scand. J. Work Environ. Health 360–363. Sainio Sr, M., 2014. Neurotoxicity of solvents. Handbook Clin. Neurol. 131, 93–110. Serdar, B., Egeghy, P.P., Waidyanatha, S., Gibson, R., Rappaport, S.M., 2003. Urinary biomarkers of exposure to jet fuel (JP-8). Environ. Health Perspect. 111 (14), 1760. Serdar, B., Egeghy, P.P., Gibson, R., Rappaport, S.M., 2004. Dose-dependent production of urinary naphthols among workers exposed to jet fuel (JP-8). Am. J. Ind. Med. 46 (3), 234–244. Shipley, R.A., Dorfman, R.I., Buchwald, E., Ross, E., 1946. The effect of infection and trauma on the excretion of urinary cortin. J. Clin. Invest. 25 (5), 673–678. Smith, L.B., Bhattacharya, A., Lemasters, G., Succop, P., Puhala, E., Medvedovic, M., Joyce, J., 1997. Effect of chronic low-level exposure to jet fuel on postural balance of US Air Force personnel. J. Occup. Environ. Med. 39 (7), 623–632. Smith, K.W., Proctor, S.P., Ozonoff, A., McClean, M.D., 2010. Inhalation exposure to jet fuel (JP8) among US Air Force personnel. J. Occup. Environ. Hyg. 7 (10), 563–572. Smith, K.W., Proctor, S.P., Ozonoff, A.L., McClean, M.D., 2012. Urinary biomarkers of occupational jet fuel exposure among air force personnel. J. Expo. Sci. Env. Epid. 22 (1), 35–45. Stuss, D., Stethem, L., Poirier, C., 1987. Comparison of three tests of attention and rapid information processing across six age groups. Clin. Neuropsychol. 1 (2), 139–152. Tabachnick, B., Fidell, L., 2007. Multivariate analysis of variance and covariance. Using Multivariate Statistics, vol. 3. , pp. 402–407. Tombaugh, T.N., 1996. Test of Memory Malingering: TOMM. Multi-Health Systems Inc, New York. Tu, R.H., Mitchell, C.S., Kay, G.G., Risby, T.H., 2004. Human exposure to the jet fuel, JP8. Aviat. Space Environ. Med. 75 (1), 49–59. Watson, D., Clark, L.A., Tellegen, A., 1988. Development and validation of brief measures of positive and negative affect: the PANAS scales. J. Pers. Soc. Psychol. 54 (6), 1063. Wechsler, D., 1997. WAIS-III: Administration and Scoring Manual: Wechsler Adult Intelligence Scale. Psychological Corporation. White, R., Proctor, S., 1992. Research and clinical criteria for development of neurobehavioral test batteries. J. Occup. Med. 34 (2), 140–148. White, R.F., Feldmann, R., Proctor, S.P., 1992. Neurobehavioral Effects of Toxic Chemicals, Clinical Syndromes in Adult Neuropsychology: The Practitioner's Handbook. Elsevier Science Ltd.. White, R.F., Gerr, F., Cohen, R.F., Green, R., Lezak, M.D., Lybarger, J., Mack, J., Silbergeld, E., Valciukas, J., Chappell, W., 1994. Criteria for progressive modification of neurobehavioral batteries. Neurotoxicol. Teratol. 16 (5), 511– 524. van Valen, E., van Thriel, C., Akila, R., Nilson, L.N., Bast-Pettersen, R., Sainio, M., van Dijk, F., van der Laan, G., Verberk, M., Wekking, E., 2012. Chronic solventinduced encephalopathy: european consensus of neuropsychological characteristics, assessment, and guidelines for diagnostics. Neurotoxicology 33 (4), 710–726.