Predictions of inter- and intra-lobar deposition patterns of inhaled particles in a five-lobe lung model.

OBJECTIVE: To develop a stochastic five-lobe lung model and to compute particle deposition fractions in the five lobes, considering anatomical as well as ventilatory asymmetry. MATERIALS AND METHODS: The stochastic five-lobe lung model was derived from an existing stochastic model for the whole lung, which implicitly contains information on the lobar airway structure. Differences in lobar ventilation and sequential filling of individual lobes were simulated by a stochastic lobar ventilation model. Deposition fractions of inhaled unit density particles in the five lobes were calculated by an updated version of the Monte Carlo deposition code Inhalation, Deposition, and Exhalation of Aerosols in the Lung (IDEAL). RESULTS: Simulations for defined exposure and breathing conditions revealed that the two lower lobes receive higher deposition and the two upper lobes lower deposition, compared to the average deposition for the whole lung. The resulting inter-lobar distribution of deposition fractions indicated that the non-uniform lung morphometry is the dominating effect, while non-uniform ventilation only slightly enhances the lobar differences. The relation between average lobe-specific deposition fractions and corresponding average values for the whole lung allowed the calculation of lobe-specific deposition weighting factors. DISCUSSION: Comparison with limited deposition measurements for upper vs. lower (U/L) and left vs. right (L/R) lobes revealed overall agreement between experimental and theoretical data. Calculations of the L/R deposition ratio for inhaled aerosol boli confirmed the hypothesis of Möller et al. that the right lung is less able to expand at the end of a breath because of the restrictive position of the liver.

CELLULAR DOSE DISTRIBUTIONS OF INHALED RADON PROGENY AMONG DIFFERENT LOBES OF THE HUMAN LUNG.

Basal and secretory cell doses in the different lobes of the human lung following inhalation of short-lived radon progeny were calculated for a five-lobe asymmetric, stochastic lung model, considering the non-uniform ventilation of the lobes. Dose calculations for defined exposure conditions revealed that the upper lobes receive higher doses than the average bronchial dose for the whole lung, with the right upper lobe receiving the highest dose. The resulting inter-lobar distribution of cellular bronchial doses indicated that the non-uniform lung morphometry is the dominating factor, while non-uniform ventilation only slightly enhances the lobar differences. The comparison of average lobe-specific bronchial doses with the average bronchial dose for the whole lung allows the calculation of lobe-specific dose weighting factors, which can be used to convert average bronchial doses based on symmetric airway generation or bronchial compartment models to lobar bronchial doses.

The degree of inhomogeneity of the absorbed cell nucleus doses in the bronchial region of the human respiratory tract.

Inhalation of short-lived radon progeny is an important cause of lung cancer. To characterize the absorbed doses in the bronchial region of the airways due to inhaled radon progeny, mostly regional lung deposition models, like the Human Respiratory Tract Model (HRTM) of the International Commission on Radiological Protection, are used. However, in this model the site specificity of radiation burden in the airways due to deposition and fast airway clearance of radon progeny is not described. Therefore, in the present study, the Radact version of the stochastic lung model was used to quantify the cellular radiation dose distribution at airway generation level and to simulate the kinetics of the deposited radon progeny resulting from the moving mucus layer. All simulations were performed assuming an isotope ratio typical for an average dwelling, and breathing mode characteristic of a healthy adult sitting man. The study demonstrates that the cell nuclei receiving high doses are non-uniformly distributed within the bronchial airway generations. The results revealed that the maximum of the radiation burden is at the first few bronchial airway generations of the respiratory tract, where most of the lung carcinomas of former uranium miners were found. Based on the results of the present simulations, it can be stated that regional lung models may not be fully adequate to describe the radiation burden due to radon progeny. A more realistic and precise calculation of the absorbed doses from the decay of radon progeny to the lung requires deposition and clearance to be simulated by realistic models of airway generations.

Radon in the Exhaled Air of Patients in Radon Therapy.

In the Gastein valley, numerous facilities use radon for the treatment of various diseases either by exposure to radon in air or in radon rich thermal water. In this study, six test persons were exposed to radon thermal water in a bathtub and the time-dependent radon activity concentration in the exhaled air was recorded. At temperatures between 38°C and 40°C, the radon activity concentration in the water was about 900 kBq/m3 in a total volume of 600 l, where the patients were exposed for 20 min, while continuously sampling the exhaled air during the bathing and 20 min thereafter. After entering the bath, the exhaled radon activity concentration rapidly increased, reaching some kind of saturation after 20 min exposure. The radon activity concentration in the exhaled air was about 8000 Bq/m3 at the maximum, with higher concentrations for male test persons. The total radon transfer from water to the exhaled air was between 480 and 1000 Bq, which is equivalent to 0.08% and 0.2% of the radon in the water.

Modeling particle deposition in the Balb/c mouse respiratory tract.

The mouse lung has become increasingly important as a surrogate of the human lung for inhalation risk assessment. The main structural difference between the two lungs is that the airway branching of the human lung is relatively symmetric, while that of the mouse lung is distinctly asymmetric or monopodial. The objectives of this study were to develop a stochastic, asymmetric particle deposition model for the Balb/c mouse and to compare predicted deposition patterns with those in the human lung. The asymmetric bronchial airway geometry of the Balb/c mouse was based on a statistical analysis of several lung casts, while, in the absence of pertinent data, the asymmetric acinar airway geometry was represented by an allometrically scaled-down version of the rat acinar region, assuming structural similarity. Deposition of inhaled particles in nasal, bronchial and acinar airways for mouse-specific breathing conditions was computed with the Monte Carlo deposition model IDEAL-mouse. While total deposition for submicron particles decreases with increasing diameter in a fashion similar to that in the human lung, the effect of inhalability and nasal pre-filtration significantly reduces total deposition in the mouse lung for particles with diameters greater than about 3 µm. The most notable difference between submicron particle deposition in the mouse and human airways is the shift of the deposition distribution from distal airway generations in the human lung to upper airway generations in the mouse lung. However, if plotted as a function of airway diameter, both deposition distributions are quite similar, indicating that airway diameter may be a more appropriate morphometric parameter for extrapolation purposes than airway generation.

Calculation of hygroscopic particle deposition in the human lung.

CONTEXT: Inhaled hygroscopic aerosols will absorb water vapor from the warm and humid air of the human lung, thus growing in size and consequently changing their deposition properties. OBJECTIVE: The objectives of the present study are to study the effect of a stochastic lung structure on individual particle growth and related deposition patterns and to predict local deposition patterns for different hygroscopic aerosols. MATERIALS AND METHODS: The hygroscopic particle growth model proposed by Ferron et al. has been implemented into the stochastic asymmetric lung deposition model IDEAL. Deposition patterns were calculated for sodium chloride (NaCl), cobalt chloride (CoCl2 · 6H2O), and zinc sulfate (ZnSO4 · 7H2O) aerosols, representing high, medium and low hygroscopic growth factors. RESULTS: Hygroscopic growth decreases deposition of submicron particles compared to hydrophobic particles with equivalent diameters due to a less efficient diffusion mechanism, while the more efficient impaction and sedimentation mechanisms increase total deposition for micron-sized particles. Due to the variability and asymmetry of the human airway system, individual trajectories of inhaled particles are associated with individual growth factors, thereby enhancing the variability of the resulting deposition patterns. DISCUSSION AND CONCLUSIONS: Comparisons of model predictions with several experimental data for ultrafine and micrometer-sized particles indicate good agreement, considering intersubject variations of morphometric parameters as well as differences between experimental conditions and modeling assumptions.

Comparison of stochastic lung deposition fractions with experimental data.

Deposition fractions of inhaled particles predicted by different computational models vary with respect to physical and biological factors and mathematical modeling techniques. These models must be validated by comparison with available experimental data. Experimental data supplied by different deposition studies with surrogate airway models or lung casts were used in this study to evaluate the stochastic deposition model Inhalation, Deposition and Exhalation of Aerosols in the Lung at the airway generation level. Furthermore, different analytical equations derived for the three major deposition mechanisms, diffusion, impaction, and sedimentation, were applied to different cast or airway models to quantify their effect on calculated particle deposition fractions. The experimental results for ultrafine particles (0.00175 and 0.01) were found to be in close agreement with the stochastic model predictions; however, for coarse particles (3 and 8 µm), experimental deposition fractions became higher with increasing flow rate. An overall fair agreement among the calculated deposition fractions for the different cast geometries was found. However, alternative deposition equations resulted in up to 300% variation in predicted deposition fractions, although all equations predicted the same trends as functions of particle diameter and breathing conditions. From this comparative study, it can be concluded that structural differences in lung morphologies among different individuals are responsible for the apparent variability in particle deposition in each generation. The use of different deposition equations yields varying deposition results caused primarily by (i) different lung morphometries employed in their derivation and the choice of the central bifurcation zone geometry, (ii) the assumption of specific flow profiles, and (iii) different methods used in the derivation of these equations.

Modeling intersubject variability of bronchial doses for inhaled radon progeny.

The main sources of intersubject variations considered in the present study were: (1) size and structure of nasal and oral passages, affecting extrathoracic deposition and, in further consequence, the fraction of the inhaled activity reaching the bronchial region; (2) size and asymmetric branching of the human bronchial airway system, leading to variations of diameters, lengths, branching angles, etc.; (3) respiratory parameters, such as tidal volume, and breathing frequency; (4) mucociliary clearance rates; and (5) thickness of the bronchial epithelium and depth of target cells, related to airway diameters. For the calculation of deposition fractions, retained surface activities, and bronchial doses, parameter values were randomly selected from their corresponding probability density functions, derived from experimental data, by applying Monte Carlo methods. Bronchial doses, expressed in mGy WLM-1, were computed for specific mining conditions, i.e., for defined size distributions, unattached fractions, and physical activities. Resulting bronchial dose distributions could be approximated by lognormal distributions. Geometric standard deviations illustrating intersubject variations ranged from about 2 in the trachea to about 7 in peripheral bronchiolar airways. The major sources of the intersubject variability of bronchial doses for inhaled radon progeny are the asymmetry and variability of the linear airway dimensions, the filtering efficiency of the nasal passages, and the thickness of the bronchial epithelium, while fluctuations of the respiratory parameters and mucociliary clearance rates seem to compensate each other.

Lung dosimetry for inhaled radon progeny in smokers.

Cigarette smoking may change the morphological and physiological parameters of the lung. Thus the primary objective of the present study was to investigate to what extent these smoke-induced changes can modify deposition, clearance and resulting doses of inhaled radon progeny relative to healthy non-smokers (NSs). Doses to sensitive bronchial target cells were computed for four categories of smokers: (1) Light, short-term (LST) smokers, (2) light, long-term (LLT) smokers, (3) heavy, short-term (HST) smokers and (4) heavy, long-term (HLT) smokers. Because of only small changes of morphological and physiological parameters, doses for the LST smokers hardly differed from those for NSs. For LLT and HST smokers, even a protective effect could be observed, caused by a thicker mucus layer and increased mucus velocities. Only in the case of HLT smokers were doses higher by about a factor of 2 than those for NSs, caused primarily by impaired mucociliary clearance, higher breathing frequency, reduced lung volume and airway obstructions. These higher doses suggest that the contribution of inhaled radon progeny to the risk of lung cancer in smokers may be higher than currently assumed on the basis of NS doses.

Comparison of radon lung dosimetry models for the estimation of dose uncertainties.

In order to investigate the degree of dose uncertainty produced by different models, three dosimetry models were compared with each other, representing different classes of models: (i) The RADEP/IMBA model based on the ICRP Human Respiratory Tract Model, a deterministic regional compartment model, (ii) the RADOS model, a deterministic airway generation model and (iii) the IDEAL dosimetry model, a stochastic airway generation model. The outputs of the three models for defined mining exposure conditions were compared at three different levels: deposition fractions for attached and unattached radon progeny; nuclear transformations, reflecting the combined effect of deposition and clearance; and resulting cellular doses. Resulting dose exposure conversion factors ranged from 7.8 (median) mSv/WLM (IDEAL) to 11.8 mSv/WLM (RADEP/IMBA), with 8.3 mSv/WLM (RADOS) as an intermediate value. Despite methodological and computational differences between the three models, resulting dose conversion factors do not appreciably differ from each other, although predictions by the two generation models are consistently smaller than that for the RADEP/IMBA model.

The effect of morphological variability on surface deposition densities of inhaled particles in human bronchial and acinar airways.

Deposition fractions in human airway generations were computed with a stochastic deposition model, which is based on a randomly, asymmetrically dividing lung morphology, applying Monte Carlo techniques. Corresponding uncorrelated surface deposition densities were obtained by dividing the average deposition fraction in a given generation by the average total surface area of that generation. In order to consider the statistical correlation between deposition probability and linear airway dimensions in each airway, correlated surface deposition densities were calculated by dividing the deposition fraction in a randomly selected bronchial or acinar airway by the surface area of that airway and by the total number of bronchial or acinar airways in that generation. Average surface deposition densities are relatively constant throughout bronchial airway generations, while average acinar surface deposition densities exhibit a distinct decrease with rising penetration into the acinar region. Due to the correlation between deposition fraction and surface area in a given airway generation, average correlated surface deposition densities are consistently higher than average uncorrelated densities, particularly in the acinar region, where differences can be as high as a few orders of magnitude. Already significant statistical fluctuations of the deposition fractions in individual airway generations are even exacerbated for surface deposition densities, with coefficients of variation about twice as high as for correlated deposition fractions.

Mucociliary and long-term particle clearance in the airways of healthy nonsmoker subjects.

Spherical monodisperse ferromagnetic iron oxide particles of 1.9-microm geometric and 4.2-microm aerodynamic diameter were inhaled by 13 healthy nonsmoking subjects using the shallow bolus technique. The bolus width was 100 ml, and the penetration front depth was 150 +/- 27 ml. The mean flow rate during inhalation and exhalation was 250 ml/s. The Fowler dead space and the phase 1 dead space of the airways were 282 +/- 49 and 164 +/- 34 ml, respectively. Deposition was below 20% without breath holding and 51 +/- 8% after an 8-s breath-holding time. We attempted to confine the bolus deposition to the bronchial airways by limiting the bolus front depth to the phase 1 dead space volume. Particle retention was measured by the magnetopneumographic method over a period of 9 mo. Particle clearance from the airways showed a fast and a slow phase; 49 +/- 9% followed the fast phase with a mean half-time of 3.0 +/- 1.6 h and characterized the mucociliary clearance. The remaining fraction was cleared slowly with a half-time of 109 +/- 78 days. The slow clearance phase was comparable to clearance measurements from the lung periphery of healthy nonsmokers, which allowed macrophage-dependent clearance mechanisms of the slow cleared fraction to be taken into account. Despite the fact that part of the slowly cleared particles may originate from peripheral deposition, the data demonstrate that mucociliary clearance does not remove all particles deposited in the airways and that a significant fraction undergoes long-term retention mechanisms, the origin of which is still under discussion.