Environmental exposure of commuters in Mexico City to volatile organic compounds as assessed by blood concentrations, 1998

Exposición ambiental a compuestos orgánicos volátiles evaluados en concentraciones sanguíneas en usuarios de transporte en la Ciudad de México, 1998

Sharon Lemire, PhD^I; David Ashley, PhD^I; Patricia Olaya, MSc^II; Isabelle Romieu, DSc^III; Susan Welch, PhD^I; Fernando Meneses-González, MSc^II; Mauricio Hernández-Avila, DSc^II

^IDivision of Laboratory Sciences, National Center for Environmental Health/Centers for Disease Control and Prevention. Atlanta, Georgia, EUA
^IICentro de Investigación en Salud Poblacional, Instituto Nacional de Salud Pública. Cuernavaca, Morelos, México ]]> ^IIIPan American Health Organization. México, DF, México

ABSTRACT

OBJECTIVE: To assess the extent of exposure for Volatile Organic Compounds (VOCs) among nonoccupationally exposed commuters in Mexico City.
MATERIAL AND METHODS: Blood concentrations of benzene, toluene, ethylbenzene, m-/p-xylene, o-xylene and methyl tert-butyl ether were determined on samples collected from participants after the morning commute.
RESULTS: Median blood concentrations of benzene (0.11 mg/l), ethylbenzene (0.081 mg/l), m-/p-xylene (0.32 mg/l) and toluene (0.56 mg/l) in the Mexico City participants were all approximately two times higher than in a nonsmoking subset of the Third National Health and Nutrition Examination Survey population of the United States. On the other hand, median VOC blood levels were similar to medians observed in other studies involving commuters in specific U.S. cities, despite the fact that only half the Mexico City study participants commuted by personal vehicles compared with all U.S. commuters.
CONCLUSIONS: These results reflect the extent of the air pollution problem in Mexico City. The surrounding topography exacerbates the problems caused by heavy vehicular traffic, poor emission-control devices on older vehicles, and poor maintenance practices. Elevated levels of gasoline components in the blood of nonoccupationally exposed commuters emphasize the need for regulatory initiatives and mass-transit options to reduce hydrocarbon emissions and thus reduce the risk for nonoccupational exposure for the residents of Mexico City. The English version of this paper is available too at: http://www.insp.mx/salud/index.html

Key words: benzene; blood benzene levels; volatile organic compounds; commuters; Mexico

RESUMEN

Palabras clave: benceno; niveles sanguíneos de benceno; compuestos orgánicos volátiles; usuarios de transporte; México

Occupational exposure to benzene, a human carcinogen, results in an increased risk for developing certain types of cancers,^1-4 most notably leukemia. Additionally, nonoccupationally exposed populations may be at risk because benzene may be carcinogenic at low levels of exposure over the long term.⁵ The primary nonoccupational sources of benzene exposure are vehicle emissions (both vehicular exhaust and evaporative losses of fuel) and tobacco smoke. Studies conducted in the United States⁶ and the United Kingdom⁷ attributed 82% and 97%, respectively, of benzene emissions to vehicular emissions. Although ambient air quality is monitored in Mexico City, pollutant concentrations inside vehicles traveling in slow-moving, high-density traffic are generally higher than concentrations in ambient air.^8-10 Consequently, several studies have been conducted to measure the in-vehicle exposures of commuters to vehicle emissions.^9-16 In a recent study of populations in Mexico City that were occupationally exposed to gasoline fumes and/or to vehicle emissions,¹⁷ the most striking result was that the unexposed control group (n = 10) exhibited higher median blood concentrations for several volatile organic compounds (VOCs) in gasoline than an unexposed US population. To confirm this result, we sampled a larger (n= 93) nonoccupationally exposed population in Mexico City. Blood concentrations of benzene, toluene, ethylbenzene, m-/p-xylene, o-xylene and methyl tert-butyl ether were measured on samples collected from participants after the morning commute into Mexico City. Toluene, ethylbenzene, and the xylenes, though not human carcinogens, are often studied along with benzene because of similarities in chemical structure and sources of exposure. Toluene is present in vehicle emissions and in tobacco smoke and the others are present in vehicle emissions. Methyl tert-butyl ether, a relatively recent additive to gasoline, has also been included because of concern about various health effects that may be associated with its use.

Material and Methods

As in the previous study,¹⁷ blood samples were analyzed at the Centers for Disease Control and Prevention (CDC) in Atlanta, GA. Vacutainer tubes used in sample collection were previously treated to remove VOC contaminants and examined to verify that the contaminants were adequately removed.¹⁸ Blood samples were maintained at 4 ⁰C during storage and shipment. Samples can be stored in this manner for up to 7 weeks without measurably affecting the results of the analyses.¹⁹ The analysis was performed by purge-and-trap gas chromatography isotope dilution mass spectrometry as described by Ashley et al.²⁰ Briefly, samples, heated to 30^–0 C, were purged for 15 min with helium and trapped on Tenax (Tekmar-Dohrmann, Cincinnati, OH). After dry purging for 6 min to remove absorbed water, the trap was thermally desorbed at 180 ^{– 0}C for 4 min. The VOCs were cryogenically trapped at the gas chromatograph (GC) injection port, then injected onto the GC column by rapidly heating the cryogenic trap. Separation was obtained on a DB-624 (J&W Scientific, Inc., Folsom, CA) capillary GC column. High-resolution mass spectrometry was performed at full scan over a mass range of 40-200 amu at 1 scan/s on a VG 70S sector mass spectrometer (Micromass, Inc., Beverly, MA).

Before analysis, each sample was spiked with stable isotopic analogs of the native compounds of interest. The responses of specific ions from the unknown sample relative to the labeled analog were measured against a six-point calibration curve to quantitate the results.

Spiked blood samples were analyzed to determine the accuracy and precision of the method, resulting in an estimated precision of <20% relative standard deviation. Method blanks were prepared from a water source free of VOCs. The limits of detection (LODs) for the compounds reported here are: benzene 0.010 mg/l; ethylbenzene 0.031 mg/l; o-xylene 0.050 mg/l; m-/p-xylene 0.019 mg/l; toluene 0.016 mg/l; styrene 0.0083 mg/l; and methyl tert-butyl ether (MTBE) 0.021 mg/l.

The purge-and-trap method described here and a similar method based on headspace analysis are the two methods commonly used to measure VOCs in blood. Ashley et al have participated in inter-laboratory comparisons to ensure comparable results between the two methods.*The purge-and-trap method results in improved LODs as a consequence of more complete removal of the VOCs from the sample matrix.

Questions designed to eliminate participants who may have been exposed to the components of gasoline fumes from sources other than the morning commute did not lead to the removal of any subjects from the data set. Data were analyzed using SAS statistical software (SAS Institute, Inc., Cary, NC). The Shapiro and Wilk test indicated that the data were not normally distribut ed (p= 0.0001) but were skewed to the right for most analytes. Neither the square root nor the log transform successfully produced a normal distribution for all analytes. For consistency, nonparametric methods utilizing the Wilcoxon Rank Sum Test were used for all comparisons and significance was determined at the 95% confidence level. Stratification of the data by smoking status indicated a statistically significant difference (p < 0.05) for most VOCs, even for VOCs not associated with smoking. However, dimethylfuran, which is considered a biomarker for smoking, was not significantly different. These results and reports by most participants of smoking less than 10 cigarettes/day suggested that the differences were not associated with smoking. Therefore, the results presented are for all commuters, regardless of smoking status. Some VOC values are missing for some of the participants because either the sample volume was insufficient for analysis or the analysis was out of control for the missing analyte(s) at the time the analysis was performed. The blood concentrations are given in terms of the maximum, minimum, median, 90^th and 95^th percentiles for benzene, ethylbenzene, m-/p-xylene, o-xylene, toluene and MTBE (Table I). For statistical purposes, values listed as below the detection limit were given values of half the LOD according to the method by Nehls and Akland.²¹

The concentrations of the compounds related to gasoline exposure that were measured in blood samples from Mexico City commuters are presented in Table I. The primary means of commuting was by automobile for 50% (45% by personal vehicles and 5% by taxi) of the subjects and by bus for 45%; the remaining 5% designated either another (4%) or walking (1%) as their primary mode of transportation. The levels of gasoline-related compounds in the blood of automobile versus bus commuters would be expected to differ on the basis of exposure to different fuel types (gasoline versus diesel). However, in Mexico City there are two types of buses: diesel and gasoline-powered microbuses. Because the study design did not allow for a distinction between these two types of buses, an unknown number of bus commuters were also exposed to gasoline-related compounds. Consequently, a comparison of the blood VOC concentrations between participants who commuted by bus (median commute 75 min) and participants who commuted by automobile (median commute 60 min) revealed no significant differences (p = 0.05). Additionally, the blood concentrations for these compounds might be expected to be highest for those participants with the longest commutes; however, regression analyses performed on each of the analytes showed no correlation (R² <0.01) for either mode of travel. The one-way commute to the workplace by automobile ranged from 2 to 120 min, and by all other means, from 5 to 180 min.

Mexico City commuters had significantly lower blood levels of benzene, ethylbenzene, m-/p-xylene and o-xylene (p < 0.05) than the control group of beginning-shift office workers in the previous Mexico City study.¹⁷ A comparison of toluene and MTBE between commuters and office workers, however, indicated no significant differences. Because blood samples were taken after the morning commute from office workers who were not occupationally exposed, VOC blood levels would be expected to be similar to those of the commuters in the current study. However, the results indicate that, even though commuters and office workers may have been exposed to similar sources of toluene and MTBE during the commute, office workers may have been exposed to additional sources of the other four compounds. Alternatively, the differences may simply be due to analysis of a different population or to the small sample size of the office workers.

The Third National Health and Nutrition Examination Survey (NHANES III) data are used as a reference range for the general U.S. population.²² A comparison of the current data to a nonsmoking subset of the NHANES III data showed a significant difference in the distributions of the two data sets (p < 0.01) for all the VOCs, except for MTBE, which was not measured in NHANES III. Table II shows the median blood levels of benzene, ethylbenzene, m-/p-xylene, o-xylene, toluene and MTBE for the Mexico City commuters compared with a nonsmoking subset of the NHANES III population. For benzene, ethylbenzene, m-/p-xylene and toluene, the medians for the commuters are approximately two to three times higher than the medians for the NHANES III population, while the o-xylene results were approximately the same. All the Mexico City participants are considered nonsmokers, because the few who did smoke, smoked fewer than 10 cigarettes/day.

Discussion

Median blood VOC levels for Mexico City commuters are consistent with medians observed in other studies involving commuters (Table II). Mexico City medians for benzene, ethylbenzene, and m-/p-xylene are approximately midway between the median blood levels for commuters in Connecticut and Alaska. For o-xylene, the medians are similar for Mexico City and Connecticut commuters; the median for Alaska commuters is 2.5 times higher. The median blood toluene level for commuters from Mexico City is 1.5 times higher than from Alaska and three times higher than from Connecticut. Mexico City commuters also had median blood MTBE levels two times higher than Connecticut commuters, even though the percentage of MTBE added to gasoline in Mexico²³ (5%) is one third that added to gasoline in the US (15%). The results for toluene and MTBE indicate the possibility of additional sources of exposure in Mexico City. Additionally, the topography surrounding Mexico City contributes to higher concentrations of air pollutants in the city and this increased concentration may be responsible for the elevated blood levels in commuters. Gasolines, formulated for maximum performance according to weather conditions and temperature extremes, differ seasonally as well as regionally. Thus, differences in the gasolines used in the Alaska (winter), Connecticut (winter), and Mexico City (spring) studies may partially account for the differences in the blood.

Several studies^{8,11,12,28-30} have measured commuter exposure indirectly by measuring in-vehicle VOCs and comparing these values to ambient air measured at fixed monitoring stations. Chan et al¹¹ and Lawryk et al²⁸ reported in-vehicle medians for benzene (14, 15 mg/m³), ethylbenzene (11, 8.7 mg/m³), m-/p-xylene (40, 34 mg/m³), o-xylene (15, 13 mg/m³), and toluene (59, 54 mg/m³) that were similar for Raleigh, North Carolina and New York City/New Jersey area commuters, respectively. Ambient concentrations were also similar except for the xylenes, which were approximately two times higher in Raleigh. In-vehicle/ambient air concentration ratios for these five VOCs indicated that in-vehicle concentrations ranged from 7 to 16 times higher than ambient air concentrations. Jo and Park³⁰ measured median in-vehicle benzene and MTBE levels of 45 and 49 mg/m³ in Korea that produced in-vehicle/ambient air ratios of 9 and 14, respectively. By comparison, the ambient air concentration of benzene in Mexico City was 45 mg/m³ (annual hourly mean) in 1995, similar to in-vehicle concentrations in Korea, Duffy et al²⁹ provide an extensive table of in-vehicle benzene concentrations from previous studies. High blood benzene levels in Mexican commuters may increase their risk of benzene-related cancers; a study by Serrano³¹ reported that the environmental concentration of benzene as measured by personal monitoring was 17.5 mg/m³ in Mexico City and the risk of acquired cancer was 1 x 10.⁴

In all these studies, in-vehicle concentrations greatly exceed concentrations measured at monitoring stations that are often situated several feet above and away from the roadway. The main sources of in-vehicle exposure are the exhaust of nearby vehicles and vehicle self-contamination.⁸ Several factors, including the design and condition of engines and fuel-distribution systems, ventilation conditions, traffic density and velocity, meteorological factors, and gasoline compositions, can influence local VOC emissions and/or commuter exposure levels from these sources. Lawryk et al²⁸ found roadway air to be the primary VOC source in vehicles with fuel injection and properly functioning carburetors. On the other hand, both Lawryk et al²⁸ and Jo and Park³⁰ found that most in-vehicle VOCs were higher for carbureted vehicles under low-ventilation conditions, indicating that carbureted vehicles may be more self-contaminating than fuel injected vehicles. Lawryk et al²⁸ found that all substituted aromatics were two times higher for carbureted vehicles; and Jo and Park³⁰ found that benzene and MTBE were 3.5 times greater. In Mexico City, the microbuses and the majority of automobiles are carbureted, thus vehicle self-contamination contributes to the high levels of VOCs found in Mexico City commuters.

Measurements of gasoline-related compounds inside vehicles have shown that models based on ambient air concentrations at monitoring stations underestimate VOC exposure to commuters. This study assesses commuter exposure by directly measuring the concentrations of these VOCs in the blood of exposed persons. Median VOC blood levels in the Mexico City commuters were two to three times higher than in the general US population represented by NHANES III but were similar to blood levels for commuters in specific US cities. These results confirm the conclusion of previous in-vehicle studies that an increase in exposure to VOCs occurs during the commute. Lofgren et al have predicted that commuters receive approximately 50% of their daily VOC exposure during the commute,⁹ and Wallace has stated that approximately 25% of benzene exposure results from personal activities, primarily driving or riding in automobiles, and approximately 25% from outdoor sources, primarily vehicle exhaust.¹⁵ In view of the health risks associated with exposure to gasoline-related compounds, reducing the amount of exposure to these compounds during the commute is important. As Duffy et al²⁹ and Chan et al¹¹ observed, the most effective means for an individual to reduce personal exposure to exhaust is by driving with the vents closed or with the air conditioner operating. In Mexico City, buses do not have air conditioning, and although newer-model automobiles do, few commuters use it, so that exposure to vehicle exhaust also contributes to the increased levels of VOCs in the blood of the commuters. Providing air-conditioned buses and promoting the use of air conditioners in automobiles so equipped would reduce exposure from the exhaust of nearby vehicles. Additionally, Chan et al¹¹ found no significant difference in VOC levels between two well-maintained vehicles of different ages, and Lawryk et al²⁸ observed much higher in-vehicle VOC levels in a malfunctioning vehicle than in a correctly functioning one. These results suggest that routine maintenance and regular inspections could significantly reduce in-vehicle exposures to gasoline-related compounds. Chan et al¹² found in-vehicle VOC concentrations for urban routes to be about 1.5 times higher than on interstate routes. Thus, reducing urban traffic density by improving mass-transportation options or traffic flow through congested areas would also improve the quality of in-vehicle air.

Acknowledgments

We thank S. Caudill whose advice and guidance were invaluable during statistical analysis of the data.

References

1. Dean BJ. Recent findings on the genetic toxicology of benzene, toluene, xylenes and phenols. Mutat Res 1985;154:153-181.         [ Links ]

2.Hayes BR, Yin SN, Dosemeci M, Li GL, Wacholder S, Travis LB et al. Benzene and the dose-related incidence of hematologic neoplasm in China. J Natl Cancer Inst 1997; 89:1065-1071.         [ Links ]

3. Lynge E, Andersen A, Nilsson R, Barlow L, Pukkala E, Nordlinder R et al . Risk of cancer and exposure to gasoline vapors. Am J Epidemiol 1997;145:449-458.         [ Links ]

4. Rinsky RA, Young RJ, Smith AB. Leukemia in benzene workers. Am J Ind Med 1981; 2:217-245.         [ Links ]

5. US Public Health Service, Agency for Toxic Substances and Disease Registry. Toxicological Profile for Benzene. Oak Ridge, (TN): Oak Ridge National Laboratory, 1989. ATSDR/TP-88/03.         [ Links ]

6. Wallace L. Major sources of exposure to benzene and other volatile organic chemicals. Risk Anal 1990;10:59-64.         [ Links ]

7. UK Department of the Environment. First Report: Urban air quality in the United Kingdom. London, England: Quality of Urban Air Review Group, 1993.         [ Links ]

8. Leung PL, Harrison RM. Roadside and in-vehicle concentrations of monoaromatic hydrocarbons. Atmospheric Environ 1999;33:191-204.         [ Links ]

9. Löfgren L, Persson K, Strömvall AM, Petersson G. Exposure of commuters to volatile aromatic hydrocarbons from petrol exhaust. Sci Total Environ 1991; 108:225-233.         [ Links ]

10. Serrano-Trespalacios P. Indicadores ambientales de compuestos orgánicos volátiles en el aire ambiente de la Ciudad de México. En: CONSERVA 2, Salud. México, DF: Departamento del Distrito Federal,1998:154-269.         [ Links ]

11. Chan CC, Ozkaynak H, Spengler JD, Sheldon L. Driver exposure to volatile organic compounds, CO, ozone, and NO₂ under different driving conditions. Environ Sci Technol 1991;25:964-972.        [ Links ]

12. Chan CC, Spengler JD, Ozkaynak H, Lefkopoulou M. Commuter exposures to VOCs in Boston, Massachusetts. J Air Waste Manag Assoc 1991; 41:1594-1600.         [ Links ]

13. Chan CC, Lin SH, Her GR. Student's exposure to volatile organic compounds while commuting by motorcycle and bus in Taipei City. J Air Waste Manag Assoc 1993; 43:1231-1238.         [ Links ]

14. Lawryk NJ, Weisel CP. Concentrations of volatile organic compounds in the passenger compartment of automobiles. Environ Sci Technol 1996;30:810-816.         [ Links ]

15. Wallace L. Major sources of benzene exposure. Environ Health Perspect 1989; 82:165-169.         [ Links ]

16. Fernández-Bremauntz AA, Asmore MR. Exposure of commuters to carbon monoxide in Mexico city II, Comparison of in-vehicle and fixed-site concentrations. J Exp Anal Environ Epidemiol 1995;5:497-510.         [ Links ]

17. Romieu I, Ramírez M, Meneses F, Ashley D, Lemire S, Colome S et al. Environmental exposure to volatile organic compounds among workers in Mexico City as assessed by personal monitors and blood concentrations. Environ Health Perspect 1999:107:511-515.         [ Links ]

18. Cardinali FL, McCraw JM, Ashley DL, Bonin M, Wooten J. Treatment of vacutainers for use in the analysis of volatile organic compounds in human blood at the low parts-per-trillion level. J Chromatogr Sci 1995;33:557-560.         [ Links ]

19. Ashley DL, Bonin MA, Cardinali FL, McCraw JM, Wooten JV, Needham LL. Important considerations in the ultratrace measurement of volatile organic compounds in blood. In: Applications of Molecular Biology in Environmental Chemistry. Boca Raton (FL):CRC Press, 1995;135-146.         [ Links ]

20. Ashley D, Bonin MA, Cardinali FL, McCraw JM, Holler JS, Needham LL et al .Determining volatile organic compounds in human blood from a large sample population by using purge and trap gas chromatography/mass spectrometry. Anal Chem 1992; 64:1021-1029.         [ Links ]

21. Hornung RW, Reed LD. Estimation of average concentration in the presence of nondetectable values. Appl Occup Environ Hyg 1990; 5:46-51.         [ Links ]

22. Ashley DL, Bonin MA, Cardinali FL, McCraw JM, Wooten JV. Blood concentrations of volatile organic compounds in a nonoccupationally exposed U.S. population and in groups with suspected exposure. Clin Chem 1994;40:1401-1404.         [ Links ]

23. Petróleos Mexicanos (Pemex refinación). Calidad de combustibles y proyectos ambientales. México, DF: Pemex,1996.         [ Links ]

24. White MC, Johnson CA, Ashley DL, Buchta TM, Pelletier DJ. Exposure to methyl tertiary-butyl ether from oxygenated gasoline in Stamford, Connecticut. Arch Environ Health 1995; 50:183-189.         [ Links ]

25. Backer LC, Egeland GM, Ashley DL, Lawryk NJ, Weisel CP, White MC et al. Exposure to regular gasoline and ethanol oxyfuel during refueling in Alaska. Environ Health Perspect 1997;105:850-855.         [ Links ]

26. Brugnone F, Perbellini L, Faccini GB, Pasini F, Maranelli G, Romeo L et al. Breath and blood levels of benzene, toluene, cumene and styrene in nonoccupational exposure. Int Arch Occup Environ Health 1989; 61:303-311.         [ Links ]

27. Brugnone F, Perbellini L, Maranelli G, Romeo L, Guglielmi G, Lombardini F. Reference values for blood benzene in the occupationally unexposed general population. Int Arch Occup Environ Health 1992;64:179-184.         [ Links ]

28. Lawryk NJ, Lioy PJ, Weisel CP. Exposure to volatile organic compounds in the passenger compartment of automobiles during periods of normal and malfunctioning operation. J Exp Anal Environ Epidemiol 1995;5:511-531.         [ Links ]

29. Duffy DL, Nelson PF. Exposure to emissions of 1,3-butadiene and benzene in the cabins of moving motor vehicles and buses in Sydney, Australia. Atmosph Environ 1997;31:3877-3885.         [ Links ]

30. Jo WK, Park KH. Exposure to carbon monoxide, methyl tertiary-butyl ether (MTBE) and benzene levels inside vehicles traveling on an urban area in Korea. J Exp Anal Environ Epidemiol 1998;8:159-171.         [ Links ]

31. Serrano-Trespalacios P. Evaluación de la exposición a compuestos orgánicos volátiles y riesgo de cáncer en la Ciudad de México. En: CONSERVA 2, Salud. México, DF: Departamento del Distrito Federal,1998:282-291.        [ Links ]

Address reprint requests to: Sharon Lemire, National Center for Environmental Health, Centers for Disease Control and Prevention, 4770, Buford Highway NE, Bldg 17 Loading Dock, Atlanta, GA 30341-3724, E-mail: SGL4@CDC.GOV

Received on: November 28, 2002
Accepted on: September 11, 2003
Grants: Mexican Ministry of Health, the Commission and Technology, Mexico (CONACyT project 3786P-M); U.S. EPA; Department of the Federal District, Mexico; Pan American Health Organization; National Center for Environmental Health, Centers for Disease Control and Prevention; Oak Ridge Institute for Science and Education.

* Ashley D. Quality Assurance Project/ BTX in Blood. Human Blood Samples for Interlab Comparison. Atlanta (GA): Centers for Disease Control, Public Health Services, US Department of Health and Human Services, 1997. Unpublished data.
** Mannino DM, Schreiber J, Aldous K, Ashley D, Moolenaar R, Almaguer D. Unpublished data.