INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Epidemiologic studies have established an association between exposures to air pollution particles and human mortality at concentrations currently found in major American metropolitan areas (1). This association has been documented in numerous investigations around the world and is remarkably consistent (2). Air pollution particles are also associated with an increased incidence of both pulmonary infections and hospitalization for respiratory disease (3).

Despite this impressive association between particles and human mortality, fundamental uncertainties exist as to which physical and/or chemical properties increase risk. In addition, pathophysiologic mechanisms have not been determined. Part of this uncertainty arises from the difficulty of exposing humans and animals to particles that are considered equivalent to those inhaled by populations included in the epidemiologic studies. Particles collected from specific emission sources such as residual oil fly ash (ROFA) or diesel exhaust have been demonstrated to cause an incursion of neutrophils into the lung, an injury, and changes in host defense capability in both animals and humans (6, 7). However, these specific emission sources make up a variable and rather small component of most urban airsheds, making it difficult to determine their exact contribution to human morbidity and mortality after exposure to ambient air particles.

Particles collected from ambient air and instilled into animals can also cause a neutrophilic inflammation and lung injury (6, 8). However, the recovery of the particles on either filters or in baghouses is extremely variable. It is also uncertain whether all components can be retrieved in quantities reflecting the original particle (e.g., some organics and ammonium may be too volatile to be collected). In addition, these particles have been instilled at considerably higher concentrations than would be inhaled by individuals in a real-world setting. Finally, there is considerable controversy as to whether this route of exposure (i.e., instillation) is equivalent to inhalation. All of these limitations make it difficult to extrapolate the results of these studies to epidemiologic findings.

The recent development of ambient particle concentrators has made it possible to perform controlled exposures of animals and humans by inhalation of "real-world" particles. Concentrators allow exposure to particles at high enough concentrations to assess a range of cardiopulmonary responses and therefore testing of specific hypotheses at pertinent masses. Initial findings demonstrate that particles between 0.1 and 2.5 µm are effectively concentrated, whereas gases and smaller particles are not (9). There also does not appear to be appreciable loss of individual particle components such as metal, sulfates, nitrates, acids, elemental and organic carbon, and general classes of organics.

Two recent studies have assessed effects on rats exposed to concentrated ambient air particles (CAPS). Rats exposed to up to 350 µg/m³ of concentrated particles for 3 h did not show evidence of lung inflammation either 3 or 24 h after exposure (10). However, in another study, rats exposed to slightly higher masses of CAPS for 3 consecutive days had slightly elevated bronchoalveolar lavage (BAL) neutrophils (11).

In the Chapel Hill area of North Carolina, ambient particulate mass (PM) is driven primarily by mobile sources such as automobiles and is similar in size distribution and chemical composition to that found in many east coast cities, albeit at lower concentrations. Concentration of particles found in Chapel Hill air 6- to 10-fold allows controlled exposure of humans to higher levels of particles. These may approximate the total levels of particles found in many metropolitan areas of the United States. In this study, we tested the hypothesis that CAPS can cause a neutrophilic inflammation in the lungs of healthy humans. Thirty-eight healthy young volunteers were exposed to either filtered air (n = 8) or CAPS ranging from 23 to 311 µg/m³ (n = 30). Changes in lung function, hematologic parameters, and both lung inflammation and injury were measured after a 2-h exposure to clear air or CAPS.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Ambient Aerosol Exposure System

Particles between the sizes of 0.1 and 2.5 µm present in the Chapel Hill air were concentrated using a Harvard/EPA Ambient Fine Particle Concentrator (HAPC) (Harvard and EPA) consisting of three-stage virtual impactors. The principles by which this concentrator works have been previously described (9, 12). A schematic of the concentrator and human exposure chamber are provided (Figure 1). The concentrator employs the inertial separator technique, thus concentrating particles only, not gases. Briefly, outside air is first drawn through an Anderson high-volume conventional impactor with a 2.5 µm cutoff size at a rate of 5,000 L/min. The exit flow from the Anderson impactor, which contains particles mainly less than 2.5 µm in diameter, is drawn into the first stage of the concentrator in which five virtual impactor slits (1,000 L/min per slit) are arranged in parallel. The virtual impactor consists of two parts: the upper part is in the form of a rectangular nozzle through which airflow is accelerated, and the lower part is in the form of a sharp-edged slit which receives incoming particles. Each virtual impactor operates at a minor to total flow ratio of 0.2, so that 80% of airflow leaving the rectangular nozzle is deflected to the side stream (i.e., major flow) and 20% of the flow is extracted straight down into the receiving slit (i.e., minor flow). In this design, particles larger than 0.1 µm in diameter achieve a sufficient momentum to cut across the deflecting major flow stream and enter into the receiving slit, whereas particles less than 0.1 µm follow the major stream and are exhausted. For this reason, particles smaller than 0.1 µm are not concentrated. Ideally, if all particles between 0.1 and 2.5 µm are condensed into the minor flow, particle concentration in the minor flow will increase by 5 times. In the present system, particles in the size range of 0.1 to 2.5 µm are concentrated approximately 2.5- to 3-fold in the first stage and a combined flow from five receiving slits (minor flow) is drawn into the second stage at the rate of 1,000 L/ min. The second stage consists of a single virtual impactor identical to those in the first stage. Here, particles are concentrated 2.5- to 3-fold again and drawn into the third stage at the rate of 200 L/min. A single virtual impactor in the third stage operates at a minor to total flow ratio of 0.4 and concentrates particles approximately 2-fold at a flow rate of 80 L/min. Finally, the concentrated aerosols leaving the third stage are mixed with 120 L/min of clean and conditioned air (20^o C and 50% relative humidity), and the resulting conditioned aerosols are delivered into the exposure chamber at the rate of 200 L/min. The addition of the conditioned air dilutes the concentrated aerosols, but provides consistent temperature and humidity and allows sufficient airflow for subjects to exercise during exposure. In this study, particles were concentrated 6- to 10-fold at the inlet of the chamber.

View larger version (17K): [in this window] [in a new window]   Figure 1.   Schematic of the Harvard/EPA Ambient Fine Particle Concentrator.

A controlled exposure to air with no particulate matter was required. Subsequently, filtered air (containing no metals, carbon, sulfates, or nitrates) was employed. Sham exposures were conducted using 200 L/min of conditioned air and no air from the HAPC. Outside air was drawn in across beds of activated charcoal and potassium permanganate on alumina. After heating to 550^o F to remove bound organics, the air was passed over cooling coils to a final temperature of 58° F. After a series of high-efficiency particulate absolute (HEPA) filters, the air was introduced into the particle chamber.

The maximal concentration of aerosols to be delivered to the chamber varied depending on concentrations of naturally occurring particulates with a mass median diameter of less than 2.5 µm (PM_2.5) in the Chapel Hill air (which usually ranges 5 to 30 µg/m³). The exposure chamber is 4.0 × 6.7 × 7.5 feet in size and constructed with aluminum panels and heavy-duty clear plexiglass for doors and windows. Because the air pumping units are located downstream of the chamber and HAPC, the chamber is operated under a slightly negative pressure (10 to 12 inches of water). Aerosols enter the chamber via a 6-inch-diameter curved duct positioned on the top and middle of the chamber and exit via an exhaust duct positioned in the middle of one of the vertical walls (Figure 1). The subject sits between the inlet and exit duct with his or her head located less than 18 inches away from the inlet duct. A series of tests conducted in the present study have shown that the particle concentration at the subject's head position is at least 90% of that at the inlet duct.

Particle Characterization

Air was sampled just before entering the HAPC and again just before entering the chamber from the inlet duct. Particles were collected on preweighed 47-mm Teflon filters (2-µm pore; Gelman Sciences, Ann Arbor, MI) at a flow rate of 10 L/min for 2 h during the exposure. Filters were weighed on an electrobalance (Mettler UMT2, Columbus, OH) in a temperature (20° C) and humidity (45%) controlled room. This balance has a capacity to reliably weigh masses as low as 1 µm. The end net filter weight, sampling time, and flow rate were used to calculate the particle concentration in µg/m³. Filters with sequestered PM (both before and after concentration) were analyzed for a number of components, including transition metals, elemental and organic carbon, sulfates, and nitrates. There were no appreciable differences in chemical composition of particles before and after concentration.

The particle size distribution was obtained using a micro-orifice uniform deposit impactor (MOUDI; MSP Corporation, Minneapolis, MN), which is an eight-stage cascade impactor containing a series of micro-orifices that collect particles onto preweighed 37-mm Teflon filters (2-µm pore; Gelman Sciences, Ann Arbor, MI). Aerosols were sampled from the inlet duct at a flow rate of 30 L/min for 2 h. The filter substrates from the impactor were weighed under a controlled environment following the same procedure described previously for total filter samples. The weights from each stage were used to determine the mass median aerodynamic diameter (MMAD) and the geometric standard deviation (GSD). These data are reported in Table 1. In addition, ozone was measured inside the exposure chamber; concentrations did not exceed 0.05 parts per million (ppm).

Study Population

Volunteers responding to a newspaper advertisement were prescreened over the telephone using the following criteria: age between 18 and 40 yr old; nonsmokers for at least 5 yr prior to study; no history of allergies or respiratory diseases (food allergy, hay fever, dust allergies, rhinitis, asthma, chronic bronchitis, chronic obstructive pulmonary disease, tuberculosis, hemoptysis, or recurrent pneumonia); and not presently on any medication prescribed by a physician (except birth control pills). A urine pregnancy test was performed on all female subjects, and a positive result excluded the subject from further participation in this study.

Before participating in the study, subjects were informed of the procedures and potential risks and each signed a statement of informed consent. The protocol and consent form were approved by the University of North Carolina School of Medicine Committee on the Protection of the Rights of Human Subjects. The screening procedures for each subject included a Minnesota Multiphasic Personality Inventory, medical history, physical examination, chest radiograph, and routine hematologic and biochemical tests.

Exposure to CAPS

Each volunteer had a single exposure to either filtered air or CAPS. Subjects were monitored continuously using telemetry and arterial saturation by oxygen. Total exposure time was 2 h. Subjects entered the exposure chamber and sat on a recumbent bicycle ergometer. Subjects exercised for 30 min of each hour. The schedule of exercise was 15 min on a cycle ergometer, 15 min rest, and this was repeated four times. Exercise intensity, i.e., cycle ergometer workload, was adjusted so that subjects breathed at a ventilatory rate, normalized for body surface area, of 25 L/m² · min. In most subjects, this will be approximately 50 L/min (i.e., oxygen consumption [O₂] of approximately 1.0 L/m). A cycle ergometer work setting of 75 to 100 watts achieved such a physiological response. During the 2-h exposure, particle concentrations were monitored continuously at the inlet duct of the chamber by using Tapered Element Oscillating Microbalance (TEOM, Series 1400a; Rupprecht & Patashnick, Inc., Albany, NY). TEOM was used to monitor a consistency or short-term excursion of exposure concentration. The average exposure concentrations were determined by filter samples as described previously.

Venipuncture and Pulmonary Function Testing

Venous blood was sampled from an antecubital site immediately before and 18 h after the exposure. Measurements were performed by LabCorp (Burlington, NC) and included a complete blood count, ferritin, blood viscosity, and fibrinogen.

Spirometry and plethysmographic measurement of airway resistance (Raw) were similarly measured before and immediately after the exposure as described earlier (13). The subject inhaled completely and then exhaled rapidly and completely via a tube into a spirometer. From this maneuver, FVC, FEV₁, and peak expiratory flow (PEF) were derived. For the plethysmographic measurement of Raw, subjects sat in a body box and performed a brief (20-s) panting maneuver at 1.5 Hz through a pneumotachograph. Airflow was occluded midway through the maneuver. By measurement of unoccluded airflow, pressure at the mouth during occlusion, and pressure in the box under both conditions, both thoracic gas volume and airway resistance to airflow were quantified.

Bronchoscopy with Lavage

Using a standard protocol (14), the volunteers underwent bronchoscopy with lavage 18 h after exposure to either the filtered air or CAPS. Previous investigation in both animals and humans indicated that an inflammatory response resulted between 18 and 24 h after exposure to particles (6, 7). The fiberoptic bronchoscope was wedged into a segmental bronchus of the lingula. Four aliquots of sterile saline were instilled and immediately aspirated. The first was 20 ml, and this fraction was labeled the bronchial lavage (BL) sample. The remaining three aliquots were 50 ml each, and the return from this BAL is considered to reflect the environment of the distal respiratory tract. These were designated the alveolar sample. The procedure was repeated on the right middle lobe again using 170 ml saline. Samples were put on ice immediately after aspiration and centrifuged at 300 × g for 10 min at 4° C. Cells were washed once with RPMI medium and viability determined via trypan blue exclusion. Viability exceeded 85% in all samples and there was no difference between air-exposed and CAPS-exposed individuals. Cell numbers were determined using a hemocytometer. Cell differentials were performed on cytocentrifuged slides stained with a modified Wright Stain (Leukostat Solution; Fisher Scientific, Pittsburgh, PA). At least 200 cells per slide were counted.

There was no difference in recovery of BL or BAL fluid (BALF) between air-exposed and CAPS-exposed individuals, and all fluid recoveries were within 10% of one another. Consequently, soluble components were normalized per milliliter of fluid. Interleukin-8 (IL-8) and IL-6 concentrations were measured using ELISA kits purchased from R&D Systems. Prostaglandin E₂ (PGE₂) was quantified with a radioimmunoassay (RIA) kit purchased from New England Nuclear. Fibronectin and ₁-antitrypsin were measured using ELISA methodology as described earlier (13). Protein and fibrinogen concentrations were determined using a centrifugal chemical analyzer purchased from Roche Laboratories (Nutley, NJ).

Statistics

Data are expressed as mean values ± standard error. Differences between air-exposed and CAPS-exposed groups were tested using the T test of independent means. For those variables that were significantly altered, the population was divided into quartiles and differences between groups compared using one-way analysis of variance (ANOVA) (15). The post hoc test employed was Scheffe's test. Two-tailed tests of significance were employed. Significance was assumed at p < 0.05.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Study Population and Exposure

The subject population included 38 volunteers (26.2 ± 0.7 yr old; 36 males and two females). There were eight exposures to filtered air (PM mass by measured filtered weights of 2.9 ± 1.9 µg/m³) and 30 exposures to CAPS (PM mass of 120.5 ± 14.0 µg/m³). Those volunteers exposed to filtered air only included seven 7 males and one female. The mean age was 27.3 yr (no significant difference with the remainder of the study population); the standard error was 1.4. There was a substantial range in CAPS exposures reflecting the variation in particles outside the facility (Table 1) with individual exposures ranging from 23.1 to 311.1 µg/m³. Taking into account time of exposure and ventilation rates, we estimate that individual lung exposures approximated a total dose of 1,200 µg on those days with the highest PM mass. The study population was divided into quartiles with the eight individuals exposed to filtered air defining the first quartile (Quartile 1) and the remaining 30 exposures arranged into groups of 10 with ascending PM mass (Quartiles 2, 3, and 4). Differences between the quartiles in PM mass were significant (F = 41.2; p < 0.0001). Excluding air exposures, the concentration factor was 6.5 ± 0.9. There were differences (F = 3.6; p = 0.04) between the concentration factors with post hoc testing revealing significance only between the second and third quartiles. Measurement of iron, zinc, and sulfur by X-ray fluorescence (XRF) verified concentration factors that approximated the value for total mass (8.5 ± 4.4, 10.8 ± 3.9, and 6.8 ± 1.4, respectively). The size distribution of exposure aerosols was approximately log-normal with the values of MMAD and GSD being 0.65 ± 0.03 and approximately 2.35, respectively. There was a slight increase in MMAD from 0.54 to 0.72 µm with an increase in mass concentration from Quartile 2 to Quartile 4.

Changes in Symptoms and Lung Function

Subjects did not report any symptoms either before or immediately after exposure to air or CAPS, nor were any abnormalities observed upon physical examination of the subjects. There were no significant differences in FEV₁, FVC, Raw, or PEF in the subjects exposed to CAPS (Table 2). Using Knudson's criteria to provide predicted values (16), all spirometric indices were normal.

Changes in Peripheral Blood

A recent epidemiologic study has reported an association between ambient air particles and increased blood viscosity (17) Therefore, in this study, a number of hematologic parameters involved in inflammation, acute-phase response, and blood viscosity or clotting were measured (Table 3). Exposure to filtered air or to CAPS had no effect on numbers of neutrophils, lymphocytes, monocytes, red blood cells (RBCs), or platelets in the blood 18 h after exposure. This was true whether the data were analyzed as air versus all CAPS or broken into quartiles for analysis. Similarly, there were no differences in hemoglobin, blood viscosity, or ferritin between groups exposed to air and CAPS. However, significant differences were observed in the concentration of blood fibrinogen between air-exposed and CAPS-exposed subjects (p = 0.009). All three CAPS quartiles showed a similar change between postexposure and preexposure values (38 to 43 µg/dl), indicating there was no dependence on dose (Figure 2).

View larger version (28K): [in this window] [in a new window]   Figure 2.   Concentrations of blood fibrinogen after exposures to CAP and filtered air. Blood fibrinogen concentrations increased with particle inhalation relative to filtered air, but this elevation did not reach significance. The concentration of the acute-phase reactant appeared to be dependent on exposure and increased with PM mass.

Changes in BL Fluid and Cells

Total cells present in the BL fraction were not affected by CAPS exposure (Table 4). There were no CAPS-induced changes in percentages of macrophages, lymphocytes, and epithelial cells in the bronchial fraction. However, there was a significant CAPS-induced increase in percent neutrophils in the BL, with particle-exposed individuals having 8.1 ± 2.7% and air-exposed individuals having 2.7 ± 0.6% neutrophils (p = 0.001). In addition, there was a statistically significant increase in monocytes (0.76 ± 0.12% after CAPS exposures and 0.36 ± 0.13% for air exposures; p = 0.010). There was also a significant increase in the number of neutrophils present in BL fluid (6.6 ± 1.7 × 10⁵ for CAPS exposures and 1.8 ± 0.05 × 10⁵ for air exposures; p = 0.011). Post hoc testing after ANOVA revealed significant differences only between the first and third quartiles in total numbers of neutrophils (Figure 3).

View larger version (24K): [in this window] [in a new window]   Figure 3.   Neutrophil numbers in the bronchial fraction after inhalation of particles and filtered air. The total number of neutrophils in the bronchial fraction of the lavage increased with particle inhalation. Percentage neutrophils were comparably increased.

Concentrations of IL-8, IL-6, PGE₂, ₁-antitrypsin, and fibronectin in the bronchial fluid were not changed after CAPS inhalation (Table 5). The concentration of protein in the bronchial fluid was decreased after exposure to CAPS; BL of air-exposed subjects contained 57.5 ± 5.7 µg/ml, whereas BL of particle-exposed subjects contained 35.9 ± 2.5 µg/ml (p = 0.0006). Post hoc tests after ANOVA revealed significant differences between the first quartile (filtered air) and the other three quartiles.

Changes in BALF and Cells

Total cells found in BALF were increased in those individuals exposed to CAPS (Table 6). Individuals exposed to air had 15.9 ± 1.9 × 10⁶ cells, whereas those exposed to CAPS had 21.4 ± 1.3 × 10⁶ cells (p = 0.04). The percentages of macrophages, lymphocytes, monocytes, and epithelial cells were not increased after CAPS exposure. However, the percentage of neutrophils significantly increased after particle exposure (2.5 ± 0.6 for CAPS and 0.8 ± 0.3 for air; p = 0.016). In addition, absolute numbers of neutrophils were increased in BALF after CAPS exposure (0.56 × 10⁶) as compared with air exposure (0.09 × 10⁶) (p = 0.0013). Neutrophil influx appeared to be dependent on dose with the greatest elevations occurring in those subjects exposed to the highest concentration of particles (F = 2.9; p = 0.05) (Figure 4). There was also an increase in the total number of monocytes found in the BAL of CAPS-exposed individuals (2.7 ± 0.4 × 10⁵) compared with air- exposed subjects (1.8 ± 0.3 × 10⁵), although the increase did not quite reach statistical significance (p = 0.08).

View larger version (25K): [in this window] [in a new window]   Figure 4.   Neutrophil numbers in the BAL after inhalation of particles and filtered air. The total number of neutrophils in the BAL increased with particle inhalation. Percentage neutrophils were comparably increased.

Concentrations of IL-6, IL-8, PGE₂, ₁-antitrypsin, and fibronectin in BALF did not change with exposure to CAPS (Table 7). Interestingly, IL-8 concentrations were considerably lower in the most heavily exposed individuals (66.4 ± 21.7 pg/ml) compared with those exposed to air (189.3 ± 52.1 pg/ml), although these differences did not reach statistical significance. In contrast with the bronchial fraction, protein concentrations in BALF were not significantly different after inhalation of CAPS relative to air. Somewhat surprisingly, fibrinogen concentrations in BALF were decreased in CAPS-exposed individuals (15.2 ± 1.4 mg/dl), compared with air- exposed subjects, who had 23.2 ± 2.1 mg/dl in their BALF (p = 0.009).

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

This study demonstrates that young, healthy volunteers exposed to ambient air particles develop an inflammation in the lower respiratory tract. The highest concentration of PM_2.5 employed in this investigation (200 µg/m³) would be an uncommon level of fine particles to be encountered in this country. However, the total exposure for individuals living in environments with much lower PM_2.5 levels will still be greater than that to which these volunteers were exposed. Whereas a resident of any urban area in the United States will be exposed to elevated particle levels almost all through the day, the volunteers in this study inhaled CAPS for only a short duration of time (i.e., 2 h).

Increased neutrophils were present in both the bronchial and bronchoalveolar fractions after CAPS exposure. The influx of neutrophils into the lung of CAPS-exposed individuals was dose-dependent, with those subjects exposed to the highest concentration of CAPS having the most neutrophils. The number of neutrophils present in BL or BALF was comparable to that found in healthy young volunteers exposed to low concentrations (0.10 ppm) of ozone for several hours (13). Neutrophils present in the BL of humans exposed to CAPS were also similar quantitatively to those found after human exposure to 300 µg/m³ diesel exhaust for 1 h (7). Although the latter study did not observe increased neutrophils in the alveolar fraction, this disparity may be explained by the differences in particle composition or source, timing of bronchoscopy (6 h after diesel exhaust exposure versus 18 h after CAPS exposure), or discrepancies in the total particle dose delivered to the lung in the two studies.

Rats exposed to up 350 µg/m³ CAPS for 3 h did not show evidence of lung inflammation either 3 or 24 h after exposure (10). This difference between human and rat response to CAPS may be reflective of different sensitivities of the two species or reflect the higher ventilatory rate of exercising humans during exposure in this study. However, in another study rats were exposed to considerably higher levels of CAPS (5 h/d for 3 consecutive days to concentrated particles ranging from 206 to 733 µg/m³), and had slightly elevated BAL neutrophils (11). Taking into account particle concentration, exposure duration, minute ventilation, deposition rate, and lung surface area of humans and rats in these two studies, we estimate the dose of particles delivered to rat lungs was at least 10 times greater than that for humans. Because the neutrophil levels were similar in both studies, these data lend credence to the notion that humans may be more sensitive to the effects of CAPS, at least as measured by lung inflammation.

It was somewhat surprising that CAPS did not induce increases in soluble BAL components such as IL-8, IL-6, PGE₂, protein, or fibronectin. All of these compounds were seen in BAL 24 h after exposure of humans to low levels of ozonea regimen which induced similar polymorphonuclear leukocyte (PMN) levels as seen in this study. This discordance suggests the possibility that CAPS and ozone may exert their effects via different mechanisms. This possibility is further strengthened by the observation that low concentrations of ozone, but not CAPS, induce changes in lung function (e.g., FEV₁ and FVC). Another possibility is that CAPS may induce cytokine release earlier than 18 h.

Another noteworthy outcome of this study is that fibrinogen concentrations in the blood may be increased after inhalation of CAPS. Fibrinogen is an extracellular, dimeric glycoprotein (molecular weight of 340,000 daltons) synthesized by the liver. It is the circulating precursor of the fibrin clot and contributes significantly to blood viscosity and coagulation, cell to cell adhesion, and platelet aggregation. Concentrations are associated with season of the year, dietary factors, geography, race, age, obesity, pregnancy, oral contraception, menopause, cigarette smoking, stress, and hypercholesterolemia (18). Fibrinogen is an acute-phase reactant and its concentrations increase during inflammation, infection, and stress. Plasma elevations in disease (e.g., hypertension, diabetes, malignancy, surgery, and inflammatory disorders) can be proportionally related to activity and tissue damage. These elevations may represent an activation of cells by an inflammatory stimulus (i.e., CAPS) with release of cytokines which induces synthesis of fibrinogen in the liver. However, the elevation of this glycoprotein did not appear to be a nonspecific elevation of an acute-phase reactant as other proteins participating in this host response were not similarly increased (i.e., ferritin). Decrements in fibrinogen in the BAL of individuals exposed to CAPS support a consumption of the protein with the activation of coagulation.

Elevation in blood fibrinogen after CAPS may reflect a disruption in the normal equilibrium between coagulation and fibrinolysis and present a thrombogenic risk. This disturbance in the normal equilibrium was not sufficient to be reflected by changes in prothrombin time and partial thromboplastin time (data not shown). Independent of other measures of coagulation, increases in blood fibrinogen concentrations can be associated with ischemic heart disease, including myocardial infarction and sudden death (19). This may reflect a mechanism to explain elevated cardiovascular deaths without a significant lung injury (22).

The lack of a significant finding in certain specific measures after exposure to CAPS is also remarkable. There were no changes in red blood cell counts, hematocrit, and hemoglobin that can be associated with particle inhalation (23). Similarly, and in contrast to exposure to diesel exhaust, there was no increase in either blood neutrophils or platelets (7). Correlations between elevations in blood viscosity and particle exposure have also been previously noted (17), but this measure did not vary with inhalation of CAPS.

Finally, there is evidence of a relationship between lung inflammation after particle exposure and small transient decreases in pulmonary function among children and asthmatics (24). However, comparable to exposures to diesel exhaust (7), there was no evidence of a CAPS-induced lung injury after healthy volunteers in this study. This is a consistent result whether injury is quantified using either a biochemical index (i.e., protein) or a physiologic index (i.e., lung function). It is possible that the response of the lung to CAPS, including the neutrophilic influx, constitutes a defense of the tissue and functions to prevent such an injury.

We conclude that exposure of healthy volunteers to CAPS can be associated with a mild influx of neutrophils into the lower respiratory tract. This did not appear to be associated with a lung injury in these individuals. Populations with decreases in cardiopulmonary function will be studied for evidence of lung injury after CAPS, including patients with asthma, chronic obstructive pulmonary disease, and ischemic heart disease.