Fuel Vapors May Be Silently Stealing Gas Station Workers’ Sense of Smell

Nearly six in ten gas station attendants tested in a new Brazilian study showed measurable damage to their sense of smell, and most of them had no idea anything was wrong. The research, published in the journal iScience, used a portable digital scent-testing device to evaluate 116 fuel pump attendants in a metropolitan region of Brazil, and the results paint a striking picture of an occupational hazard that has largely flown under the radar of workplace health regulators. Among the workers examined, 58.6 percent met the criteria for olfactory dysfunction, with 46.6 percent classified as hyposmic, meaning a reduced ability to detect odors, and 12.1 percent classified as anosmic, meaning a profound or complete loss of smell. Only 41.4 percent of the tested workers scored within the normal range for odor identification.

The findings are particularly alarming because the impairment appears to accumulate silently over time. The researchers, led by José Maria Viana dos Santos and Márcio Nakanishi, found that odor identification scores declined steadily with years on the job. Workers with olfactory impairment had an average tenure of 10.2 years at the pumps, compared with just 5.9 years among those with normal smell. A statistical correlation analysis confirmed a moderate negative relationship between years of exposure and olfactory performance, and a multivariable regression model that accounted for age and smoking history revealed that each additional year of work increased the odds of olfactory dysfunction by roughly 15 percent. Notably, age itself was not significantly associated with impairment, suggesting that the effect is driven by exposure rather than simple aging.

What makes the study especially compelling is the gap between objective test results and subjective awareness. Only 27 of the 116 participants, or 23.3 percent, believed that their work was affecting their sense of smell. Yet among the 68 workers with objectively confirmed dysfunction, 48, or 70.6 percent, reported no such perception. Several participants were reportedly surprised by their results and attributed any perceived changes to aging or recurring nasal problems rather than to their workplace. This pattern of underrecognition echoes earlier hypotheses in occupational health literature that chemical-induced olfactory disorders are frequently subclinical and go unreported unless they become severe, which limits the effectiveness of surveillance programs that rely solely on workers noticing and reporting symptoms.

The exposure at issue is a complex cocktail of volatile organic compounds. Gasoline vapors contain BTEX, a group of aromatic hydrocarbons comprising benzene, toluene, ethylbenzene, and xylene. Benzene is the most notorious member of this group: it is classified as carcinogenic to humans, with strong evidence linking it to acute myeloid leukemia, and it is present in gasoline as an inherent component that cannot simply be substituted away. Although humans can typically detect benzene by its sweet smell at concentrations of roughly one to five parts per million, the study’s authors emphasize that odor perception is an unreliable safety indicator. Individual sensitivity varies widely, olfactory adaptation can dull the warning signal after repeated exposure, and workers cannot depend on their noses to alert them to hazardous airborne concentrations.

The biological plausibility of fuel-vapor-induced smell loss is well grounded in toxicology. The olfactory epithelium, the specialized tissue lining the upper nasal cavity where odor detection occurs, is remarkably metabolically active and rich in cytochrome P450 enzymes, which can bioactivate inhaled xenobiotics into reactive metabolites. Chronic inhalation of solvents may damage olfactory receptor neurons and their supporting cells through oxidative stress, inflammation, and mucosal irritation. Benzene and its metabolites are known to induce oxidative stress and DNA damage, and biomonitoring studies of Brazilian gas station workers have documented genotoxic effects consistent with such exposure. The authors caution, however, that their findings should not be interpreted as direct evidence of DNA damage in the olfactory mucosa; for solvent mixtures and volatile organic compounds, irritation and inflammation of the nasal lining are considered more plausible mechanisms than direct nerve uptake, a pathway more clearly established for ultrafine particles. The irritant properties of petrochemical vapors may also provoke chronic rhinitis, and workers in Brazilian biomonitoring studies have frequently reported nasal irritation and nosebleeds, consistent with mucosal toxicity.

Importantly, the pattern of errors on the smell test suggests that the damage is not simply a matter of becoming desensitized to fuel odors. While gasoline was correctly identified by 69.8 percent of participants, the odors most frequently misidentified by impaired workers were apple, eucalyptol, and orange, none of which are fuel-related, while garlic, clove, and cinnamon were recognized by most participants. This non-selective pattern of loss indicates that selective adaptation to gasoline odor alone is unlikely to explain the dysfunction, strengthening the case that the vapors are causing genuine injury to the olfactory system rather than mere familiarity effects.

The study’s methodology deserves attention because it showcases a new generation of field-deployable sensory testing. Conventional standardized smell assessments, such as the University of Pennsylvania Smell Identification Test and Sniffin’ Sticks, are well validated but require trained personnel, consume considerable time, and are costly to administer at scale. The researchers instead used the MultiScent-20, an automated digital device developed in São Paulo, Brazil, which delivers 20 microencapsulated odor stimuli through a tablet interface in a four-alternative forced-choice format. Each participant taps a start button, receives a five-second burst of scent carried by a controlled airflow, and selects the matching option from four choices on the touchscreen. The entire test takes roughly 10 to 15 minutes, scores itself automatically, and can be self-administered. Validation studies have shown high test-retest reliability and strong concordance with Sniffin’ Sticks, with a correlation of approximately 0.88, making the device well suited for workplace screening campaigns that would be impractical with traditional methods.

The Brazilian context amplifies the significance of the findings. National legislation in Brazil prohibits self-service fueling, which means attendants handle every refueling and bear the full burden of cumulative vapor inhalation, eight hours a day. Regulatory guidelines such as NR-09 mandate biological monitoring of benzene exposure using urinary trans,trans-muconic acid, and Brazilian studies have shown that even at ambient benzene concentrations below one part per million, workers exhibit elevated biomarkers of benzene-related biological effects. Yet protective measures remain strikingly inadequate: in the study, only 3 of 116 participants, or 2.6 percent, reported using personal protective equipment, even though 62.8 percent of those asked believed such equipment was necessary. Workers cited discomfort, inconvenience, interference with customer service, and lack of employer enforcement as reasons for non-use, and 63 percent were aware of benzene-related health risks.

The safety implications extend beyond health into hazard detection itself. Gas station workers may rely on smell as one of several warning cues to detect fuel leaks and spills, thereby reducing fire and explosion risks, and case reports have documented incidents of ignition near gasoline spills involving individuals with anosmia. Reduced smell sensitivity also diminishes the ability to detect smoke, gas leaks, or spoiled food in daily life. In some countries, such as Germany, intact olfactory function is formally required for specific occupations involving gas detection, but no such requirement exists for gas station workers in Brazil, leaving workers with impaired smell effectively less able to perceive odor-related dangers on the job.

The authors argue that their results should prompt a broadening of occupational surveillance frameworks, which currently focus almost exclusively on hematological and carcinogenic endpoints of benzene toxicity, toward sensory outcomes. They recommend periodic olfactory screening for solvent-exposed workers, engineering controls such as stage II vapor recovery systems at fuel pumps, job rotation to limit continuous exposure, worker education about the sensory effects of fuel vapors, and longitudinal studies to track long-term outcomes. They also suggest that respirators with organic vapor cartridges should be available for high-exposure tasks like tank cleaning and spill response, even if sustained routine use remains impractical without cultural change. The study does carry limitations: its cross-sectional design precludes definitive causal inference, individual exposure levels were not measured, and only current workers were evaluated, raising the possibility that the most affected individuals had already left the job. Still, the consistency of the findings with international evidence, including an Indonesian study that found 77.5 percent of gas station employees with olfactory dysfunction compared with none of the unexposed controls, suggests the problem transcends borders. For a workforce that keeps the world’s vehicles fueled, the message is sobering: the very vapors they breathe every day may be eroding one of the senses they need to stay safe, long before they notice anything is wrong.

Subject of Research: Olfactory dysfunction among gas station workers chronically exposed to fuel vapors containing benzene and other volatile organic compounds

Article Title: Olfactory dysfunction among gas station employees exposed to fuel vapors using digital odor identification testing

Article References: Viana dos Santos, J. M., & Nakanishi, M. (2026). Olfactory dysfunction among gas station employees exposed to fuel vapors using digital odor identification testing. iScience, 29(11), Article 117786. https://doi.org/10.1016/j.isci.2026.117786

Image Credits: AI Generated

DOI: Not provided

Keywords: olfactory dysfunction, gas station workers, benzene, BTEX, fuel vapors, occupational health, volatile organic compounds, smell testing, hyposmia, anosmia, MultiScent-20, Brazil