Exposures and cross-shift lung function declines in wildland firefighters

Abstract

Respiratory problems are common among wildland firefighters. However, there are few studies directly linking occupational exposures to respiratory effects in this population. Our objective was to characterize wildland fire fighting occupational exposures and assess their associations with cross-shift changes in lung function. We studied 17 members of the Alpine Interagency Hotshot Crew with environmental sampling and pulmonary function testing during a large wildfire. We characterized particles by examining size distribution and mass concentration, and conducting elemental and morphological analyses. We examined associations between cross-shift lung function change and various analytes, including levoglucosan, an indicator of wood smoke from burning biomass. The levoglucosan component of the wildfire aerosol showed a predominantly bimodal size distribution: a coarse particle mode with a mass median aerodynamic diameter about 12 μm and a fine particle mode with a mass median aerodynamic diameter < 0.5 μm. Levoglucosan was found mainly in the respirable fraction and its concentration was higher for fire line construction operations than for mop-up operations. Larger cross-shift declines in forced expiratory volume in one second were associated with exposure to higher concentrations of respirable levoglucosan (p < 0.05). Paired analyses of real-time personal air sampling measurements indicated that higher carbon monoxide (CO) concentrations were correlated with higher particulate concentrations when examined by mean values, but not by individual data points. However, low CO concentrations did not provide reliable assurance of concomitantly low particulate concentrations. We conclude that inhalation of fine smoke particles is associated with acute lung function decline in some wildland firefighters. Based on short-term findings, it appears important to address possible long-term respiratory health issues for wildland firefighters. [Supplementary materials are available for this article. Go to the publisher's online edition of Journal of Occupational and Environmental Hygiene for the following free supplemental resources: a file containing additional information on historical studies of wildland fire exposures, a file containing the daily-exposure-severity questionnaire completed by wildland firefighter participants at the end of each day, and a file containing additional details of the investigation of correlations between carbon monoxide concentrations and other measured exposure factors in the current study.].

Keywords: exposure assessment; levoglucosan; occupational lung disease.

FIGURE 1.

Normalized personal cascade impactor particle size distributions for (a) airborne particle mass and (b) airborne levoglucosan. Note that airborne particle mass is mostly associated with larger particles, while airborne levoglucosan is bimodally distributed.

FIGURE 2.

Elemental Particle Profile and Electron Micrograph of Particles: (a) samples obtained at base camp indicate a crystal-like particle comprised mostly of titanium, aluminum silicate, and iron; (b) samples obtained at the fire line revealed a spherical, tar-like particle containing mainly carbon; and (c) and other samples obtained at the fire line appear to have an aggregate configuration of many tar-like particles. The gold (Au) and palladium (Pd) peaks are from the sputter coating on the samples and are not constituents of the particles. Active diameter of EDX interrogation was approximately 1 μm and particle composition did not appear to vary with beam location on particle surface.

FIGURE 2.

Elemental Particle Profile and Electron Micrograph of Particles: (a) samples obtained at base camp indicate a crystal-like particle comprised mostly of titanium, aluminum silicate, and iron; (b) samples obtained at the fire line revealed a spherical, tar-like particle containing mainly carbon; and (c) and other samples obtained at the fire line appear to have an aggregate configuration of many tar-like particles. The gold (Au) and palladium (Pd) peaks are from the sputter coating on the samples and are not constituents of the particles. Active diameter of EDX interrogation was approximately 1 μm and particle composition did not appear to vary with beam location on particle surface.

FIGURE 3.

Correlation between paired real-time full-shift time-weighted-average concentrations of CO and particulate (0.1 to 10 μm, as measured by personalDataRAM), stratified by job. Line worker (◆) (eight paired observations); Sawyer (■) (two paired observations); and Swamper (●) (one paired observation).