INTRODUCTION

Top Abstract Introduction Methods Results Discussion References

DEPOSITION SITE AND DOSE of inhaled particles within the lung vary widely depending on particle size, breathing pattern, and lung structure. Generally, particles deposit more in the proximal airways with an increase in particle size and breathing rates, whereas enhanced pulmonary deposition takes place with small-size particles and slow breathing rates (9, 10, 30). Within the bronchial airways, particle deposition tends to localize at and near the carina, particularly with particles >1 µm in diameter (17, 19). In patients with obstructive airway disease, deposition patterns are very heterogeneous and marked by a pronounced deposition in the central airways and various focal "hot" spots in different regions of the lung (22, 26, 29). Overall lung deposition has also been shown to increase in the lungs with obstructed airways and abnormal geometry (15, 20). Therefore, regional deposition dose, or local tissue burdens, can be significantly different among individuals, even if total lung deposition values are comparable. It is also possible that particle burdens could reach the threshold limits at local lung regions under exposure conditions that are normally acceptable, particularly in individuals with compromised lungs. This may have significant implications in health risk assessment of pollutant particulates and in identifying subpopulations prone to overexposure to particulate matter.

Previous studies investigated regional lung deposition primarily by inhalation of radiolabeled aerosols and subsequent measurements of radioactivity in the lung by either a gamma camera or whole body counter. The results have been reported for three large lung compartments, e.g., extrathoracic (larynx and above), tracheobronchial (TB), and alveolar regions (AL) (23, 27, 28). Because these measurements are closely linked to particle clearance from the TB airways, the results have been instrumental for estimating a short- and long-term retention of particles in the lung. However, the three lung compartments are not detailed enough to assess local variations of deposition dose in the lung and to differentiate deposition characteristics among individuals. The reported data were also based largely on measurements in healthy male adults, and it is not warranted for these data to be applied to population groups with different ages and genders.

There are many differences in the anatomy of the lung between men and women. For instance, the size of the lung is smaller in women compared with men, particularly in the upper airways (UA; pharynx and larynx) and the large conducting airways (7, 24). The dynamics of the larynx and vocal cord are also different, as evidenced by different pitches in sound that result partly from different tissue densities and muscle tension. These anatomic and dynamic differences can be significant factors for altering deposition characteristics in the lung. However, how these factors affect deposition characteristics in women and the potential impact of the results in health risk assessment have not been investigated thoroughly. Earlier studies showed that total lung deposition values are comparable between men and women (2, 21, 25) or slightly greater in women than men, depending on breathing patterns (2). However, Pritchard et al. (25) found that deposition distribution within the lung was more proximal in women than men under various breathing conditions and suggested that there is a potential for excessive deposition in the UA regions in women. Presently, there are no systematic data available about the magnitude and specific lung regions of the potential dose difference between genders.

Previously, our laboratory has reported a new method that can measure detailed regional lung deposition by means of serial bolus delivery technique and have shown distinctive deposition patterns in young adults (18). In the present study, we measured regional lung deposition in each of 10 equally divided volumetric compartments in healthy male and female adults and compared the results with respect to different particle sizes and breathing patterns. The purpose of this study was to verify the potential effects of gender on deposition characteristics of inhaled particles and to obtain accurate exposure-dose relationships that can be used to develop improved health risk assessment models for inhaled particulates.

	METHODS

Top Abstract Introduction Methods Results Discussion References

Subjects. Healthy young subjects, both men and women in equal number (n = 11 in each group), were recruited locally. Ages of the subjects ranged from 21 to 32 yr and were matched between men and women. The subjects had no history of smoking, hay fever, or asthma. All subjects underwent a screening procedure that included a complete medical history, physical examination, SMA-20 blood chemistry screen, and complete blood count with differential. Those who passed the initial screening had their basic lung functions measured by both spirometry and body plethysmography. All subjects were asked to read and sign a consent form approved by the institutional review board of the University of North Carolina School of Medicine at Chapel Hill. Subject characteristics and lung function test results are given in Table 1.

Serial bolus delivery technique. This method has been described in detail elsewhere (18). Briefly, the method is based on the notion that the conventional aerosol inhalation, with a tidal volume (VT), may be divided into a series of inhalations of smaller volume aerosol. Here, VT is divided into a number of smaller compartments with equal volume and a series of inhalations is performed with the same VT in which the aerosol fills only one volumetric compartment in each inhalation, as shown in Fig. 1. Total lung deposition fraction (TDF) will then be obtained by

(1)

where n is the total number of volumetric compartments, and RC_j (= N_ex/N_in) is the recovery of bolus aerosol from the jth compartment. Here, N_in and N_ex are the total number of particles inhaled and exhaled, respectively. If particle deposition efficiency in the jth compartment is defined by x_j or local deposition efficiency (LDE_j), and if deposition efficiencies are the same on inspiration and expiration, a unique functional relationship exists between RC and x as

(2)

Values of x_j can then be obtained by the ratio of RC values from two adjacent compartments as

(3)

(4)

where x_j is defined by the ratio of the amount of aerosol deposited in a volumetric compartment j to the amount entering the same compartment. Once x_j values have been obtained, the deposition fraction of a bolus aerosol in each volumetric compartment (BDF_ij) can be obtained by combining the inspiratory and expiratory deposition in each compartment, where the subscript i represents the number of sequential bolus inhalation. The local deposition fraction in the jth compartment (LDF_j) can then be obtained by

(5)

and subsequently TDF of the entire VT aerosol can be obtained by the summation of LDF_j, where n is the total number of volumetric compartments of VT or the number of sequential bolus inhalations. Therefore, total as well as regional lung deposition can be determined by measuring the RC values of aerosols from a series of bolus inhalations. The method is illustrated schematically in Fig. 1.

View larger version (25K): [in this window] [in a new window]   Fig. 1.   Illustration of calculation procedures for deposition efficiency in jth compartment (xj) and deposition fraction in local lung regions in a serial compartment lung model. Top: 3-compartment model (j = 1-3). Aerosol recovery (RCj) and bolus deposition fraction in each compartment (BDFij) are shown for each sequential bolus inhalation (i = 1-3). Values of xj were assumed to be same on inspiration and expiration. Expressions for inspiratory and expiratory BDF are shown on top and bottom of each compartment, respectively. Aerosol fractions remaining at end inspiration are shown on right of each diagram. Nin and Nex, total no. of particles inhaled and exhaled, respectively. Right: expressions for RC. Expressions for both RC and BDF are shown as a fraction of inhaled aerosol entering the mouth.

Generation of test aerosols. Monodisperse DES (di-2-ethylhexyl sebacate) oil aerosols were generated by an evaporation-condensation-type aerosol generator (MAGE, Lavoro E Ambiante, Bologna, Italy). The performance characteristics of the MAGE generator have been described previously (11). In the present study, the original MAGE generator was modified to improve the quality of aerosols and to generate large-size particles. Briefly, aqueous solutions of NaCl (5-10 mg/l) were nebulized by a Collison-type atomizer that was operated with compressed nitrogen gas (20 lbs./in.²). Liquid aerosols generated initially were passed through a drying column filled with silica gel, and the resulting dry nuclei aerosols were passed at a rate of 1-3 l/min through a "boiler" in which DES oil was heated and vaporized at temperatures of 170-250°C. The mixture of nuclei and DES oil vapor from the boiler was passed through a reheater maintained at 280-320°C and subsequently through a vertical condensation column that was designed to induce condensation of vapor on the surface of nuclei particles. Monodisperse DES aerosols emerging from the condensation column were diluted with clean air (20-100 l/min) by using a two-stage diluter. By changing the concentration of nuclei and the temperatures of the boiler and the reheater, we generated monodisperse aerosols with 1-, 3-, and 5-µm diameter (geometric SD < 1.15) particles. Particle size was measured by an aerodynamic particle sizer (APS 3310, TSI, St. Paul, MN) equipped with an online aerosol diluter (1:100 ratio, model 3302, TSI). Concentration of aerosols was maintained at a level of 2 × 10³ to 40 × 10³ particles/cm³, depending on particle size, with higher concentrations for smaller particle size and vice versa.

Bolus aerosol inhalation system. The core of the system consisted of a laser aerosol photometer, an aerosol bolus injection module, and an on-line data-acquisition system (Fig. 2). Test aerosols were introduced into the aerosol photometer as a small bolus (half-width = 45 ml) by activating a solenoid valve in the bolus injection module. When the valve was open, an aerosol was ejected into the inspiratory airstream via four narrow slits (1.6 mm wide and 18 mm long) positioned across the diameter of the stream. The aerosol chamber upstream of the solenoid valve was maintained at slightly above room pressure (1-5 cmH₂O) to help inject the aerosol. The multiple-slit system was designed to ensure a rapid mixing of aerosol with the inspiratory airflow, thereby producing a well-defined small bolus. In the laser photometer a laser beam (15 mW He-Ne, Melles Griot, Carlsbad, CA) was expanded into a thin sheet via a cylindrical lens and shone through an aerosol detection cell, where the laser beam was scattered by aerosol particles. The scattered light was collected on a photomultiplier tube (model 9798B, EMI Gencom, Plainview, NY), and the signals from the photomultiplier tube were amplified to the range of 0-10 V with a current amplifier (model 427, Keithly Instruments, Cleveland, OH) and subsequently transmitted to the data-acquisition system. The aerosol detection cell was heated to 40°C by an electric resistor imbedded in the metallic block of the cell to prevent moisture condensation on the lens during exhalation. Flow rates through the laser photometer were measured by a pneumotachograph (Fleisch no.1) in conjunction with a pressure transducer (model 239, ±1.27 cmH₂O range, Setra Systems, Acton, MA) connected directly to the inspiratory inlet of the detector cell. The data-acquisition system consisted of a signal modulator and a personal computer (model 326, Dell Computer, Austin, TX) equipped with a high-speed data-acquisition board capable of sampling signals at a rate of up to 27 kHz (DT2801A, Data Translation, Marlboro, MA). Both flow and aerosol signals were displayed digitally in the signal modulator in which a built-in integrator circuit provides volume signals for inhaled and exhaled air. The volume signals were used to activate the solenoid valve and to deliver an aerosol bolus to a prescribed lung depth. For online data acquisition, both flow and aerosol signals were acquired at a rate of 200 Hz, and flow signals were smoothed by passing them through a 50-Hz low-pass filter to eliminate spikes generated by the activation of the solenoid valve. Smoothing of aerosol signals was not necessary. All of the controls for bolus inhalation, data acquisition, and analysis were programmed with ASYST software (ASYST Software Technologies, Rochester, NY).

View larger version (33K): [in this window] [in a new window]   Fig. 2.   Experimental system used for aerosol bolus inhalation. HV, high voltage; PMT, photomultiplier tube; PC, personal computer; MAGE, evaporation-condensation-type aerosol generator.

Bolus aerosol inhalation procedure. After a few practice breaths, the subject inhaled filtered air via a laser aerosol photometer (25-ml dead space volume) from functional residual capacity (FRC) after a prescribed breathing pattern displayed on the computer screen. The subject then activated the data-acquisition mode by pressing a handheld switch during expiration, followed by inhalation of a prescribed volume and exhalation to residual volume (RV) at a constant flow rate () (Fig. 3). During the data-acquisition mode, a small aerosol bolus (~45 ml half-width) was introduced into the inspiratory stream by opening an aerosol valve for a predetermined duration of time. The duration of valve opening was adjusted between 50 and 250 ms, depending on , to maintain a consistent bolus volume; the faster the , the shorter the duration. The peak concentration of bolus was maintained at a level between 6 and 9 V; the voltage level of 1 V was equivalent to ~5,000 particles/cm³ for 1-µm-diameter particles. The bolus was delivered to a lung depth ranging from 100 to 500 ml in 50-ml increments. This procedure was repeated with monodisperse aerosols of three different particle sizes (1-, 3-, and 5-µm diameter), and for each particle size three different (150, 250, and 500 ml/s) were used. In all tests the same was used for both inspiration and expiration, and the inspiratory volume was maintained at 500 ml from FRC. For a given bolus-delivery condition, at least five repeated measurements were made.

View larger version (13K): [in this window] [in a new window]   Fig. 3.   Schematic diagram depicting an inhalation maneuver for bolus aerosols. Inhalation starts (left dashed line) at functional residual capacity (FRC) with 500-ml tidal volume, and exhalation continues to level of residual volume with a constant flow rate (). An aerosol bolus (45-ml half-width) is injected at prescribed time points during inspiration. Bolus injection (middle dashed line) at beginning of inspiration delivers aerosol more deeply into lung and vice versa. Conc, concentration. Right dashed line, exhalation bolus.

Data analysis. For each breath the total number of particles inhaled (N_in) and exhaled (N_ex) were calculated by integrating the product of aerosol number concentration, C(t), and volumetric flow rate, (t), over the inspiratory (TI) and expiratory (TE) period, respectively. In the calculation the baseline concentration was set at a level of 3% of the peak value of the exhaled bolus and was subtracted from the acquired signals. RC of bolus aerosol (i.e., RC_j) was plotted as a function of penetration depth (Vp), where Vp was defined as the inhaled air volume from the mean concentration of the bolus to the end of inspiration. RC data were then grouped for 10 Vp values starting from 50 to 500 ml with an interval of 50 ml. RC values (means ± SD) for each Vp were obtained by averaging all RC data falling within Vp of ±25 ml. Volumetric lung regions (V_j; j = 1-10) were defined by the regions confined between two adjacent Vp values. For example, V₁ is the region between Vp = 0 and 50 ml, V₂ is the region between Vp = 50 and 100 ml, and so on. The mean RC vs. Vp data were used to calculate values of local deposition efficiency (x or LDE) and LDF for each volumetric region by Eqs. 4 and 5, respectively.

	RESULTS

Top Abstract Introduction Methods Results Discussion References

Bolus aerosol RC. Typical bolus RC data from one male subject are shown in Fig. 4 for three different particle sizes with a fixed of 250 ml/s, and the summary of all subject data for men and women is shown in Fig. 5 for three different values of particle diameter (D_p; 1, 3, and 5 µm) and (150, 250, and 500 ml/s). RC values were very repeatable and consistent in a given subject, as shown in Fig. 4, and the intersubject variability was small, as indicated by small error bars shown in Fig. 5. RC decreased monotonically with increasing Vp for each particle size and used, but the decrease of RC was greater with particles of larger size in both men and women. Figure 5 also shows that RC values were smaller with lower (i.e., longer residence time) for a given value of D_p. This flow effect was consistent for all regions of Vp except for very shallow regions, Vp < 100 ml, in which the effect of was small or negligible. It is also noted that the flow effect was smaller with D_p = 5 µm than with D_p = 1 and 3 µm, and the effect was minimal particularly in the group of women.

View larger version (18K): [in this window] [in a new window]   Fig. 4.   Typical bolus aerosol RC data from 1 male subject as a function of penetration volume (Vp) for particles 1 (), 3 (), and 5 µm in diameter () with a fixed of 250 ml/s. Vertical dashed line, system dead space. Note that measurements are very repeatable, as indicated by compact clusters of data points, and RC values decrease consistently with an increase in Vp.

View larger version (28K): [in this window] [in a new window]   Fig. 5.   Bolus aerosol RC in men (solid symbols) and women (open symbols) as function of Vp for particles 1 (dashed lines), 3 (dotted-dashed lines), and 5 µm in diameter (solid lines) with 3 different rates: 150 (top), 250 (middle), and 500 ml/s (bottom). Error bars, ±SE. Vertical dashed line, system dead space.

LDE. Figure 6 shows LDE values for each 50-ml volumetric region as a function of Vp for different particle sizes and values for both men and women. It can be seen that LDE increases consistently with an increase of Vp for all experimental conditions, with increases seen particularly in men; the larger the Vp, the greater the LDE. However, in female subjects, there was a transient peak in the region of Vp = 100 ml with particles of large D_p or high values. LDE increased with an increase in D_p or a decrease in in all Vp regions in both male and female subjects.

View larger version (39K): [in this window] [in a new window]   Fig. 6.   Comparison of local deposition efficiency between men () and women () as function of Vp for 3 different sizes of particles [particle diameter (Dp) = 1 (top), 3 (middle), and 5 µm (bottom)] and 3 different rates [150 (left), 250 (middle), and 500 ml/s (right)]. Error bars, ±SE. For Dp = 1 µm, P = not significant (NS) between men and women for all Vp except for * P < 0.05; for Dp = 3 and 5 µm, P < 0.05 for all Vp except for + P = NS.

LDF. Deposition values in each 50-ml volumetric region are shown in Figs. 7 and 8, in which the effects of D_p and are illustrated for both male and female subjects. The aforementioned figures show a wide variation of LDF with Vp; LDF increased with increasing Vp initially, reached a peak value, and then decreased with a further increase in Vp. For D_p = 1 µm, the regional variation of LDF was small and the peak deposition was found in the middle to distal regions of the lung. However, with an increase in D_p, the peak height increased rapidly and the region of the peak deposition shifted proximally regardless of (Fig. 7). For men, peak deposition was found in the Vp = 300-ml region for D_p = 1 µm, Vp = 200-ml region for D_p = 3 µm, and Vp = 100-150-ml region for D_p = 5 µm. For women, the pattern was similar to that in men, but the peak height was greater and the region of the peak deposition was more proximal than those in men. Figure 7 also shows that in the shallow regions of the lung (Vp < 250 ml) LDF increases with an increase in D_p regardless of values used (P < 0.05). However, in the deeper regions (Vp > 250 ml) LDF was greater with D_p = 3 µm than with 5 µm (P < 0.05) in both men and women except for  = 500 ml/s, with which the difference was small (e.g., in women) or negligible (P = NS in men). At a low of 150 ml/s, LDF was comparable among all three D_p values in the deep lung regions, Vp > 400 ml.

View larger version (39K): [in this window] [in a new window]   Fig. 7.   Effects of particle size on deposition fraction in local volumetric regions for men (left) and women (right). Comparisons are shown for 3 different sizes of particles [Dp = 1 (), 3 (), and 5 µm ()] under 3 different rates [150 (top), 250 (middle), and 500 ml/s (bottom)]. P < 0.05 between data points at all Vp except for + P = NS.

View larger version (39K): [in this window] [in a new window]   Fig. 8.   Effects of rate on deposition fraction in local volumetric regions for men (left) and women (right). Comparisons are shown between 3 different rates [150 (), 250 (), and 500 ml/s ()] and each of 3 test aerosols: Dp = 1 (top), 3 (middle), and 5 µm in diameter (bottom). P < 0.05 between different values for all Vp except for + P = NS between 2 data points and ++ P = NS among 3 data points.

View larger version (35K): [in this window] [in a new window]   Fig. 9.   Comparison of local deposition fraction between men () and women () as function of lung depth from mouth for 3 different sizes of particles [Dp = 1 (top), 3 (middle), and 5 µm in diameter (bottom)] with varying rates [150 (left), 250 (middle), and 500 ml/s (right)]. * P < 0.05 between women and men.

Three-compartment regional deposition. Three-compartment regional lung deposition values are summarized in Table 2, in which percent deposition in the UA, TB, and AL regions as well as in the total lung (TDF) are compared between men and women for D_p = 1, 3, and 5 µm and =150, 250, and 500 ml/s. In women, UA deposition was in the range from 1.8 to 3.3 and 8.6 to 9.3% for D_p = 3 and 5 µm, respectively, and these values were greater than those in men (P < 0.05). UA deposition was negligible for D_p = 1 µm. TB deposition ranged from 15.1 to 24.8% for D_p = 3 µm and from 40.8 to 42.8% for D_p = 5 µm within the range of used, and these TB values were greater by 62-104% and 41-69%, respectively, compared with those in male subjects (P < 0.05). For D_p = 1 µm, TB values were small in both men and women, but the values tended to be greater in women than men (P = NS). AL deposition was in the range from 4.3 to 16.3% for D_p = 1 µm, 30.1 to 40.6% for D_p = 3 µm, and 28.1 to 31.6% for D_p = 5 µm in women. Compared with values in men, in women these values were greater by 11-23% for D_p = 1 µm (P = NS) but smaller by 21% for D_p = 5 µm (P < 0.05). TDF values were greater in women than men by 14-31% for D_p = 3 µm and 9-20% for D_p = 5 µm with  = 150-500 ml/s (P < 0.05), and, for a given D_p, the difference in men vs. women was greater with an increase in . For D_p = 1 µm, TDF tended to be greater (9-28%) in women than men, but the difference was not significant.

	DISCUSSION

Top Abstract Introduction Methods Results Discussion References

We compared regional deposition of inhaled particles in young adult men and women in situ under controlled breathing conditions by using a serial bolus aerosol delivery method. This was the first study to investigate the effects of gender on detailed dose distribution of inhaled particles within the lung. The results demonstrate that 1) deposition distribution within human lungs is highly uneven along the volumetric depth of the lung; and 2) there are distinct differences in deposition distribution within the lung between men and women, namely, a greater deposition in the proximal airways in women than men.

We used a bolus aerosol method to assess regional lung deposition in situ with inert aerosols. Because the method does not require radiolabeled aerosols and aerosol monitoring can be achieved on-line continuously during inhalation, a large number of repeated measurements can be achieved with varying inhalation conditions in a short period of time. However, with the method regional deposition is inferred on the basis of the delivery depth of bolus aerosol and, as such, certain limitations are applied to the method, as discussed in a previous report (18). Briefly, one of the key factors affecting the accuracy of the bolus aerosol method is the distribution pattern of bolus aerosol within the lung as the aerosol flows in and out of the lung. If the aerosol flow is not distributed evenly within the lung, the actual anatomic regions to which bolus aerosol is delivered may not be determined accurately by the inhaled volume of air. This would also result in a wide variation of bolus RC or deposition data expressed by Vp because uneven distribution patterns are not likely to be consistent among different subjects, or even in the same subjects, inhaling aerosols with different breathing patterns at different times. However, many ventilation distribution or imaging studies indicate that ventilation patterns in normal subjects are even and reversible (4, 13). In our preliminary study using radiolabeled bolus aerosol, the deposition pattern was found to be fairly even between the right and left lung in healthy persons when the bolus was inhaled from the near-FRC level (1). In the present study bolus RC data are very consistent with respect to lung depth, as shown in Fig. 4. In every set of measurements in all subjects, RC values decreased monotonically with an increase in Vp, indicating that a series of bolus aerosols was indeed delivered to regions of successively greater depth in an orderly fashion. The present results may prove useful within their limitations for assessing regional distribution of lung deposition in healthy adults.

In a comparison of the present results with the conventional three-compartment lung deposition, the three lung regions were determined on the basis of the volume of inhaled air measured from the mouth: 50 ml for UA, from 50 to 150 ml for TB, and >150 ml for AL. Previously, the volume of the oropharyngeal cavity, which represents primarily UA, has been reported variably, but a value of 50 and 40 ml is generally accepted as a representative value for men and women, respectively (12). As for the TB region, Hart et al. (8) measured dead space volume (UA + TB) in a large number of adult subjects and found that it was linearly correlated with body height (r² = 0.84) regardless of gender. For a person with a height of 174 cm, which was the average for all of the present male and female subjects, the dead space volume was 150 ml, which was the value used to define the TB region in the present study. When our deposition values in these three compartments are compared with conventional results obtained by various investigators (Fig. 10), there is good agreement in each lung region, indicating that the present results are reasonable representations of deposition in these three lung regions. In Fig. 10 it can be seen that our results of AL deposition of D_p =1 µm tend to be lower than the previous results. This was due partly to the fact that the present results were obtained with the single-breath maneuver, consisting of a maximal exhalation to RV. Because particles that remain airborne in the deep lung regions at the end of inhalation can be exhaled by an extended exhalation from FRC to RV, AL deposition tends to be smaller with the single-breath maneuver, particularly for particles with a low deposition efficiency, e.g., D_p =1 µm. However, the effects are negligible for larger size particles, because particle recovery is essentially complete at the level of FRC during exhalation, as indicated in Fig. 5.

View larger version (20K): [in this window] [in a new window]   Fig. 10.   Comparison of 3-compartment deposition results with existing data. Top, middle, and bottom: upper airway, tracheobronchial, and alveolar deposition, respectively. Open symbols, men; solid symbols, women. Monodisperse aerosols were inhaled via mouth with 500-ml tidal volume and rates of 150 (triangles), 250 (circles), and 500 (squares) ml/s. Dashed lines, mean of conventional deposition data (28).

It is interesting to note that LDE is greater in women than men throughout the entire region of the lung, particularly for coarse particles. It should be noted that lung deposition is governed primarily by three factors: particle size, breathing pattern, and geometry of the lung. However, because both male and female subjects inhaled the same aerosols by using the same breathing patterns, the different results could be attributed to certain differences in lung anatomy between men and women. Because Vp is measured from the mouth, for a given value of Vp the actual anatomic sites to which a bolus aerosol is delivered may vary depending on an individual's lung size, i.e., a deeper penetration in smaller-size lungs and vice versa. As a result, RC values would be lower and deposition would be greater in small-size lungs. In the present study the average lung volume was smaller in women than in men, and this could result in a greater LDE in females at a given Vp. However, according to the studies of Hart et al. (8) showing a good correlation between anatomic dead space and body height, the dead space volume of our female subjects could be smaller than that of our male subjects by ~25 ml. When the volume difference was corrected by shifting the female LDE curve to the right by 25 ml in Fig. 6, LDE was still greater in women in most of the Vp regions. In Fig. 6 it can also be seen that the difference will persist even after a shift of the LDE curve to the right by a volume much larger than 25 ml, particularly for D_p = 5 µm. Although the effects of dead space volume may be assessed to a certain extent by normalizing Vp with thoracic gas volume (TGV; e.g., Vp/TGV), the difference in TGV between our male and female subjects was small (3,378 vs. 2,955 ml), and the normalization would result in only a 13% shift of LDE curve to the right, which does not account for the differences in LDE. These analyses suggest that factors other than lung size or dead space volume might have contributed to a great extent to the difference in LDE of coarse particles between men and women.

Our results show a marked variation in regional deposition within a normal lung, which was characterized by a unimodal peak, the location of which was shifted from the peripheral to proximal regions with an increase in D_p. This finding was consistent with theoretical expectations, because large-size particles deposit primarily by inertial impaction in the proximal airways, in which flow velocity is high before aerosol reaches the peripheral regions, whereas small-size particles penetrate more deeply into the peripheral region and deposit there by sedimentation. However, the actual experimental data for humans have never been reported previously in such detail as shown in the present study. It should be noted that our results show a fairly consistent regional deposition pattern with coarse particles (D_p = 3 and 5 µm), particularly for D_p = 5 µm, regardless of respiratory values used. This could be expected because two primary deposition mechanisms for coarse particles, inertial impaction (which is governed by the parameter D²_p and increases with increasing ) and sedimentation (which is controlled by the parameter D²_p / and therefore increases with decreasing ), could act to compensate for each other as changed. It is also noted that because deposition of coarse particles is controlled by the second order of D_p, particle size is expected to have a greater effect on lung deposition than . Our results showing a marked increase in deposition, with an increase in D_p particularly in the shallow regions of the lung (see Fig. 7), are consistent with these theoretical expectations. In contrast, a deposition pattern of fine particles (D_p = 1 µm) was highly variable with respiratory : a greater deposition in a deeper region with decreasing . This was, in fact, expected, because fine particles deposit mainly by sedimentation, and, with a slow , particles remain longer in the lung and deposit more in deeper lung regions, where sedimentation distance is short. These characteristic deposition patterns were similar between men and women, but they were more pronounced in women, showing enhancement of deposition in the peak regions for both fine and coarse particles. In other words, deposition was more localized within the lung in women. In our results it can also be noted that, for coarse particles, regional deposition is dictated primarily by particle size, whereas deposition distribution is greatly modulated by breathing patterns for fine particles. The implications of these results are that local tissue burdens of coarse particles in the proximal airways are greater in women than men during normal activities, regardless of breathing patterns. This may result in more incidents of respiratory irritation or illness in women than men in an environment with elevated levels of coarse particle.

Compared with men, LDF in women is markedly enhanced in the shallow regions of the lung, particularly with large-size particles. This may result from certain structural or geometrical differences in the upper and large conducting airways between men and women. By using an acoustic reflectance method, Martin et al. (24) measured tracheal dimensions in a large number of male and female subjects (age = 20-35 yr) and found that the cross-sectional area of the trachea was smaller by ~32% in women than men at the same level of lung volume. These results were consistent with other findings obtained by different methods, including roentgenogram (3) and computed tomography (6). Because a sudden change in dimensions between adjacent airways is unlikely, one may also expect smaller dimensions in the large conducting airways subsequent to the trachea. An increase in inertial impaction and flow turbulence in the small-size airways could, in turn, result in enhancement of particle deposition in the shallow regions of the lung. In fact, in our earlier theoretical studies utilizing Weibel's symmetrical lung model, we found a 50% increase in deposition of particles 3 µm in diameter in the large conducting airway when the diameters of the airways were uniformly reduced by 25% (16). In recent experimental studies using various bifurcating airway models (17), it was found that airway deposition increased with a power function of an impaction parameter, Stokes number, the value of which is inversely proportional to the third power of airway diameter for a given . The results suggest that a 30% reduction in airway cross-sectional area could result in a deposition increase in the bifurcating airways by >100%. On the basis of the measurement of radioactivity from the lung after inhalation of radiolabeled aerosols, Pritchard et al. (25) also reported an increase in deposition in the shallow TB regions of the lung in women compared with men. It should also be noted that the size of the larynx is usually smaller in women than men (7). A smaller opening in the larynx could generate a stronger turbulent jet into the trachea and could result in deposition enhancement in the trachea and airways downstream (5). The small size, together with certain differences in geometry and dynamic motion of the larynx, could also lead to deposition enhancement in the larynx itself. The present results are in line with expectations from the earlier studies. However, differences in the structure and function of the upper and large airways between men and women, and the precise role of these differences in particle deposition needs to be investigated further.

Although different methods were used in measuring lung deposition, previous studies showed consistently small or no differences in total lung deposition between men and women. By analyzing the differences between radioactivity collected on the inspiratory and expiratory filters, Pritchard et al. (25) found that total lung deposition was in good agreement between men and women for particles in the size range of 2.5 to 7.5 µm in diameter. In the present study the breathing pattern was not controlled, and the subjects inhaled aerosols spontaneously with their own breathing pattern. However, women usually breathe with smaller VT than men during spontaneous breathing (2). Therefore, if female subjects inhale the aerosols with the same VT as that of men, lung deposition may increase in women. Recently, Bennett et al. (2) investigated total lung deposition of particles (D_p = 2.0 µm) under spontaneous as well as a selected fixed breathing condition. They found a greater deposition (~17%) in women than men with a fixed breathing pattern. However, the difference was not significant under the spontaneous breathing condition. In the case of smaller-size particles, a deposition index of 1-µm-diameter particles was measured for a group of young and old normal subjects by using a rebreathing technique, and no significant differences between men and women were found in deposition in both age groups with a controlled breathing pattern (21). The present results, showing no difference in TDF for D_p = 1 µm between men and women, but a greater deposition in women than men for D_p = 3 and 5 µm, are consistent with the earlier studies. This, together with regional deposition enhancement in women, may lead to a greater health risk in women than men. However, the ventilation rate is smaller in women compared with men, e.g., 6.0 vs. 8.6 l/min during normal breathing (2), and daily activity patterns vary widely among individuals in real life. Therefore, these exposure factors need to be incorporated into the present results, which were obtained under carefully controlled breathing conditions, to assess accurately the potential health risk of particulate matter.

In conclusion, we measured detailed regional lung deposition patterns of inhaled particles in young healthy adults and investigated the gender difference. We found marked differences in regional deposition patterns between men and women and enhancement of regional dose in women in varying regions depending on particle size. These findings may have significant implications in health risk assessment concerning inhaled particles on one hand and will be useful for targeted aerosol delivery in pharmaceutical and clinical studies on the other.