Regional Lung Deposition and Bronchodilator Response as a Function of 2 -Agonist Particle Size

Transcription

1 Regional Lung Deposition and Bronchodilator Response as a Function of 2 -Agonist Particle Size Omar S. Usmani, Martyn F. Biddiscombe, and Peter J. Barnes National Heart and Lung Institute, Imperial College London; Royal Brompton Hospital, London, United Kingdom Rationale: Aerosol particle size influences the extent, distribution, and site of inhaled drug deposition within the airways. Objectives: We hypothesized that targeting albuterol to regional airways by altering aerosol particle size could optimize inhaled bronchodilator delivery. Methods: In a randomized, double-blind, placebo-controlled study, 12 subjects with asthma (FEV 1, % predicted) inhaled technetium-99m labeled monodisperse albuterol aerosols (30- g dose) of 1.5-, 3-, and 6- m mass median aerodynamic diameter, at slow (30 60 L/min) and fast ( 60 L/min) inspiratory flows. Lung and extrathoracic radioaerosol deposition were quantified using planar -scintigraphy. Pulmonary function and tolerability measurements were simultaneously assessed. Clinical efficacy was also compared with unlabeled monodisperse albuterol (15- g dose) and 200 g metered-dose inhaler (MDI) albuterol. Results: Smaller particles achieved greater total lung deposition (1.5 m [56%], 3 m [50%], and 6 m [46%]), farther distal airways penetration (0.79, 0.60, and 0.36, respective penetration index), and more peripheral lung deposition (25, 17, and 10%, respectively). However, larger particles (30- g dose) were more efficacious and achieved greater bronchodilation than 200 g MDI albuterol ( FEV 1 [ml]: 6 m [551], 3 m [457], 1.5 m [347], MDI [494]). Small particles were exhaled more (1.5 m [22%], 3 m [8%], 6 m [2%]), whereas greater oropharyngeal deposition occurred with large particles (15, 31, and 43%, respectively). Faster inspiratory flows decreased total lung deposition and increased oropharyngeal deposition for the larger particles, with less bronchodilation. A shift in aerosol distribution to the proximal airways was observed for all particles. Conclusions: Regional targeting of inhaled 2 -agonist to the proximal airways is more important than distal alveolar deposition for bronchodilation. Altering intrapulmonary deposition through aerosol particle size can appreciably enhance inhaled drug therapy and may have implications for developing future inhaled treatments. Keywords: aerosol; asthma; -adrenergic agonists; particle size; radionuclide imaging In many respects, the science of drug delivery to the lungs is still in its infancy. For too long, inefficient device systems have been accepted pragmatically for inhalation therapy in asthma. (Received in original form October 26, 2004; accepted in final form September 23, 2005) This research is supported by an academic grant from GlaxoSmithKline, Research and Development, United Kingdom. An editorial commenting on this article was published in the September 15, 2005 issue of the Journal (Am J Respir Crit Care Med 2005;172: ), ajrccm.atsjournals.org/cgi/content/full/172/6/656. Correspondence and requests for reprints should be addressed to Omar S. Usmani, M.D., Department of Thoracic Medicine, Airways Disease Section, National Heart and Lung Institute, Imperial College London, Dovehouse Street, London SW3 6LY, UK. o.usmani@imperial.ac.uk This article has an online supplement, which is accessible from this issue s table of contents at Am J Respir Crit Care Med Vol 172. pp , 2005 Originally Published in Press as DOI: /rccm OC on September 28, 2005 Internet address: If enough drug reaches the lungs to achieve an adequate clinical response, these devices are considered satisfactory. However, there is a significant wasted portion of the dose that plays no part in the clinical improvement and may give rise to adverse effects. The challenge is to target inhaled medication to its clinically relevant effector cells within the lung, so as to optimize the therapeutic response and minimize potential adverse effects. -Scintigraphy is a powerful tool that has been widely used to visualize and characterize intrapulmonary drug delivery in subjects with asthma (1). Numerous studies have quantified lung deposition patterns, but often in isolation to clinical outcomes (2). The few studies that incorporated efficacy variables, either using indirect radiolabeling techniques involving inert nonpharmacologic monodisperse particles (3 7) or by using direct, physically radiolabeled 2 -agonist aerosols (8 17), have collectively been inconclusive with regard to the relationship of regional airway bronchodilator deposition with therapeutic response. Commercial polydisperse inhaler devices, which lack the specificity of airway targeting, were primarily used within these studies to investigate the advantages of new aerosol delivery systems, rather than explore the science of inhaled drug delivery. Monodisperse pharmacologic aerosols, however, make it possible to undertake translational aerosol research, accurately exploring in vitro concepts of aerosol particle behavior within the human airways in vivo (18). They are composed of uniformly sized particles, with the majority of the aerosol drug mass within a narrow size distribution; therefore, these aerosols have greater discriminative power than polydisperse aerosols to explore differences due to drug particle size. It is well recognized that, of aerosol characteristics, drug particle size is the major determinant governing the site of airway deposition of inhaled medication (19). We previously described our use of a spinning-disk aerosol generator to produce monodisperse albuterol aerosols and reported that 6- and 3- m particles were more potent bronchodilators that 1.5- m particles in subjects with mild moderate asthma (20). We proposed larger particles were better matched to their target site of action. Our hypothesis is that the regional airway distribution of 2 -agonist is important in modulating the bronchodilator response and, by altering intrapulmonary deposition through aerosol particle size, we may be able to optimize inhaled drug delivery. The aim of this study, therefore, was to investigate the dynamic relationship between albuterol particle size, regional lung deposition, inspiratory flow, and bronchodilator response to build up a profile of the optimal aerosol delivery characteristics. We used two-dimensional planar imaging to assess the lung and extrathoracic deposition patterns of three sizes of technetium- 99m labeled monodisperse albuterol aerosols in subjects with mild moderate asthma, with simultaneous measures of efficacy and tolerability. Some of the results of this study have been previously reported in the form of an abstract (21, 22). METHODS Additional details on the research methods are available in the online supplement.

2 1498 AMERICAN JOURNAL OF RESPIRATORY AND CRITICAL CARE MEDICINE VOL TABLE 1. SUBJECT DEMOGRAPHIC DATA Subject Sex FVC (L ) FEV 1 (L ) FEV 1 (%p)* BdR (L ) BdR ( %) 101 M F M M M F F F M F F M Mean SD Definition of abbreviations: BdR postbronchodilator FEV 1 compared with baseline values using 200 g metered-dose inhaler spacer; F female; M male. *%p as a percentage of the predicted FEV 1 values (52). % as a percentage of improvement on baseline screening value. Subjects Twelve subjects with stable, mild moderate asthma (FEV 1, % predicted) gave their written, informed consent to participate in this study, which was approved by the local ethics committee (Table 1). Each subject showed 15% or greater bronchodilator improvement in FEV 1 after inhaling 200 g albuterol from a metered-dose inhaler (MDI) and spacer. Subjects were controlled on 500 g or less of inhaled beclomethasone dipropionate daily, but no subjects were receiving longacting 2 -agonist or oral antiasthmatic medication. Seven subjects were steroid-naive. Aerosol Generation, Radiolabeling, and Imaging The generation and radiolabeling of monodisperse (geometric standard deviation 1.22) albuterol aerosols using a spinning-disk aerosol generator, and our image-acquisition procedures of radioaerosol deposition within the lungs and extrathoracic regions, are comprehensively described elsewhere (23, 24). Scintigraphic images of the posterior thorax, anterior thorax, and lateral oropharynx were recorded, in sequence, with the patient repositioned between views (Figure 1). Primary radioactive counts were normalized for image duration and corrected for background radiation, physical radioactive decay, body tissue attenuation, depth of radioactive source, and biological clearance of radioactivity (25). Subjects underwent a krypton-81m ventilation scan to define their lung boundary. Characteristics of the monodisperse aerosol size distributions, drug concentrations, and reproducibility of radioaerosol dosing are detailed further in the online supplement. Radioaerosol Deposition Analysis Total lung and extrathoracic deposition were determined by drawing computer-generated regions of interest around the lung, oropharynx, stomach, and mediastinum. Mass balance of the total delivered radioactive dose was accounted for by summating corrected counts in each anatomic region (lung, oropharynx, stomach, and mediastinum), including that retained in the exhalation filter and mouthpiece. This was taken to be 100% of the radioactivity in the 3 L of inhaled aerosol. Counts totaled from both lungs were used in the analysis. The estimate of lung dose was calculated by dividing the corrected lung counts by the total lung counts for all the components. Regional lung deposition was quantified using a modification of a previously described technique (26). A 5 8 matrix was closely fitted to the krypton ventilation-scan lung boundary, which was superimposed on the posterior thorax aerosol deposition images, and partitioned into central, intermediate, and peripheral airway zones (Figure 2A). Deposition results were expressed as a percentage of the total delivered radioactive dose and, for regional lung deposition, also as a percentage of the total lung dose. Regional lung distribution was expressed as penetration index: the ratio of radioaerosol counts in the peripheral zone to the central lung zones divided by the same count ratio from the ventilation scan (27). Clinical Measurements A medical practitioner performed all physiologic measurements, unaware of the aerosol administered. Initially, heart rate and blood pressure were determined automatically (Critikon, Slough, UK) with the subject supine. FEV 1, FVC, forced expiratory flow between 25 and 75% of FVC (FEF ), and peak expiratory flow (PEF) were then measured using an electronic spirometer (Vitalograph, Buckingham, UK). Finally, indirect potentiometry (Beckman, Buckingham, UK) was used for plasma potassium concentration estimation. Subjects were rescheduled if their pretreatment FEV 1 differed by more than 15% from their screening value. Study Design During eight randomized visits, monodisperse albuterol aerosols of 1.5-, 3-, and 6- m-mass median aerodynamic diameter were separately administered as an unlabeled 15- g dose and technetium-labeled 30- g doses, inhaled at 30 to 60 L/min (slow inhalation). Each dose was delivered as three sequential 1-L bolus breaths followed by a breathhold pause. Subjects also inhaled 200 g MDI albuterol using a spacer and, in an identical manner, a placebo MDI. A double-dummy manner was used with the aerosol delivery and imaging procedures (see Table E4 in the online supplement). Subjects attended three additional visits in which each radiolabeled aerosol was inhaled at more than 60 L/min (fast inhalation). After aerosol delivery and imaging, spirometry, cardiovascular measurements, and samples for plasma potassium were taken at 10 min from the beginning of inhalation and at 10-min intervals for 60 min, then every 15 min for 90 min. For efficacy measurements, the best value (maximal) out of two was taken at each time point, whereas only single measurements were undertaken for tolerability markers. Statistical Analysis For the efficacy and tolerability variables, the change in each measurement from baseline (pretreatment) values was plotted at each time point for the duration of the study (time, 0 90 min) to derive a time response curve for each subject. The area under the time response curve (AUC) was calculated using trapezoidal integration, and this value was weighted (wauc) for duration of time (28). For each biological variable, the mean of all the waucs from the 12 subjects was calculated and this value was used in the statistical analysis. Comparisons between the clinical effects of each treatment and between the radiolabeled deposition data of each treatment were undertaken using parametric analysis of covariance, including factors for subject, period, treatment, and multiple comparisons. On the basis of our previous data

3 Usmani, Biddiscombe, and Barnes: 2 -Agonist Regional Lung Deposition 1499 Figure 1. (A ) Posterior thorax, (B) anterior thorax, and (C) lateral oropharyneal -camera images of aerosol deposition using technetium-99m labeled albuterol particles of 1.5- m mass median aerodynamic diameter (MMAD). Red areas indicate regions of highest radioactivity, and black of least radioactivity. (20), with 12 subjects there was 80% power to detect a clinically relevant difference of 0.25 L in FEV 1 between the treatments, assuming a significance level of 0.05 and an estimate of variability of 0.25 L. RESULTS Radioaerosol Deposition Images Typical deposition patterns for the three monodisperse aerosols are shown in Figures 1 and 2. Although the subjects with asthma were stable and asymptomatic, regional ventilatory inhomogeneity was evident as patchy aerosol deposition within the lung images, which suggested native airway obstruction. Total Lung and Extrathoracic Deposition Total lung deposition (TLD) was greater with the 1.5- m aerosols compared with the 6- (p 0.01) and 3- m aerosols (p 0.40; Table 2). Oropharyngeal deposition increased with increasing particle size, whereas the exhaled aerosol fraction was greatest with the 1.5- m aerosols. No significant difference in mediastinal deposition and mouthpiece retention was noted between the three particle sizes. Lung Distribution and Regional Deposition Penetration index values increased with decreasing particle size, which indicated that 1.5- m particles penetrated further into the lung periphery, whereas 6- m particles were more proximally distributed within the airways (Table 2). Differences were highly significant between all particle size comparisons (p 0.001). Regional deposition analysis expressed as a percentage of TLD confirmed peripheral region deposition was significantly greater with smaller particles in the following order: m (p 0.001; Table 2). Conversely, 6- m particles achieved significantly more combined central and intermediate (C I) region deposition compared with the other particles (p 0.001). Similarly, when expressed as a percentage of the delivered dose, peripheral region deposition increased significantly with decreasing particle size (all comparisons, p 0.001). However, although C I as a percentage of delivered dose was greatest for the 6- m particles, differences between particle sizes were not significant. Clinical Assessment Pretreatment baseline FEV 1 values were similar between the eight study groups and the screening value (see Table E5; p 0.05). Larger particles produced greater FEV 1 (ml, means SD) bronchodilation in the following order: m, at both 15- g ( , , and , respectively) and 30- g doses ( , , and , respectively). Likewise, particle size effects with FEF and PEF mirrored those of FEV 1 in that greater airway responses were achieved with increasing particle size (see Table E6). These measurements showed significant difference between 6- and 1.5- m aerosols at both doses, but not between other particle size comparisons. Figure 2. (A ) Regions of interest, where a 5 8 matrix was closely fitted to the krypton ventilation-scan lung boundary and superimposed on the posterior thorax aerosol deposition images, which was partitioned into a central zone, 2 3 matrix centered on the lung hilum; an intermediate zone, 3 5 matrix excluding the central zone; and a peripheral zone defined as the remaining lung field within the lung outline. (B) 1.5 m, (C) 3 m, and (D )6 m MMAD posterior thorax images of technetium-99m labeled albuterol aerosol deposition in the lungs. Red areas indicate regions of highest radioactivity, and black of least radioactivity.

4 1500 AMERICAN JOURNAL OF RESPIRATORY AND CRITICAL CARE MEDICINE VOL TABLE 2. MASS BALANCE AND REGIONAL LUNG DISTRIBUTION Particle Size ( m) TLD OpS Ex M Mp PI C I (%TLD) P(%TLD) C I P * Definition of abbreviations: C I central and intermediate; Ex exhaled; M mediastinum; Mp mouthpiece; OpS oropharynx and stomach; P peripheral; PI penetration index; TLD total lung deposition. Data are as a percentage of the delivered radionuclide dose except for PI (ratio) and where indicated. Data are presented as means SD. *p 0.01 compared with 6- m particle size. p compared with other particle sizes. All other particle size pairwise comparisons were not significant. %TLD as a percentage of TLD. Particle size differences in FVC were not significant. All treatments (monodisperse and MDI) demonstrated highly significant increases in all lung function measurements compared with placebo (p 0.001). The 6- m monodisperse aerosol at the 30- g dose was a more potent bronchodilator than 200- g MDI albuterol (FEV 1, ml), and achieved a comparable FEV 1 response to the MDI dose at a 15- g dose. The study was not powered to test equivalence with the MDI. Bronchodilation, judged by FEV 1, was greatest at the 30- g dose compared with the 15- g dose for each particle size, but no significant difference was noted between doses. In addition, no significant difference was observed in the tolerability measurements for each particle size/ dose combination compared with placebo, or between doses for the same particle size comparison (see Table E6). Time response profile curves for FEV 1 showed there was a distinct and sustained separation of bronchodilator particle size effects throughout the duration of the study, whereby the 1.5- m particles were unable to achieve the response obtained with the 6- and 3- m particles (Figure 3). response in asthma, and both drug particle size and inspiratory flow are important factors in achieving this aim. Although 2 -receptor density is greatest in the alveolar region (29), airway smooth muscle is relatively sparse, being predominantly located in the conducting airway region, which is where 2 - agonist should be deposited to achieve effective bronchodilation (30). Despite greater TLD and farther distal airway penetration with smaller 1.5- m albuterol particles during slow inhalation, we found larger particles achieved significantly greater bronchodilation. We reasoned the different innate physical deposition characteristics of the particle sizes allowed selective regional airway targeting in that albuterol was favorably directed to the conducting airway smooth muscle using larger particles, whereas smaller particles were preferentially directed to peripheral alveoli (31, 32). In addition, the marked increase in airway surface area toward the lung periphery will have diluted the topical Inspiratory Flow Effects on Deposition and Clinical Response Mean ( SD) inspiratory flows for slow and fast inhalation were and L/min, respectively. Fast inhalation increased TLD for 1.5- m particles ( 4.6%, p 0.05), but decreased for 3- ( 1.4%, p 0.42) and 6- m ( 24.7%, p 0.001) particles (Figure 4). Oropharyngeal deposition increased for all particle sizes, but this was significantly greater for the larger particles (p 0.001). Particle distribution was shifted more centrally (Figures 5 and 6A), and these differences were highly significant compared with slow inhalation (p 0.001). There was no significant difference in FEV 1 for 1.5- m aerosols, but a significant decrease was observed for 3- and 6- m aerosols (p 0.001; Figure 6B). No correlation was observed of either total lung dose or regional deposition with particle size and lung function measures at either slow or fast inhalation flows (see Tables E7 and E8, respectively). DISCUSSION This is the first study to combine the assessment of regional airway drug deposition using radiolabeled monodisperse albuterol aerosols, with the simultaneous measure of clinical response, to investigate the science of bronchodilator particle size effects in vivo. We have shown that, by using different aerosol particle sizes to deliver inhaled 2 -agonist to lung regions containing the target effector cells and by reducing aerosol losses in regions where albuterol has little effect, we may optimize inhaled bronchodilator delivery to the lungs. Our data show new evidence that clearly demonstrates the importance of the regional lung deposition of albuterol in determining the bronchodilator Figure 3. FEV 1 time response profile curves. Placebo (dashed line, open diamonds) and albuterol metered-dose inhaler 200 g(dotted line, closed diamonds) are shown on both graphs. Monodisperse albuterol aerosols are shown as follows: 1.5 m (circles), 3 m (squares), and 6 m (triangles) at(a ) 15- g (open symbols) and (B) 30- g (closed symbols) doses. Data are presented as means (of 12 patients) of the maximal change in each patient s individual FEV 1 response (from baseline value) at each time point SEM.

5 Usmani, Biddiscombe, and Barnes: 2 -Agonist Regional Lung Deposition 1501 Figure 4. Effect of fast inhalation (shaded bars) compared with slow inhalation (white bars) on aerosol deposition in lung (L), oropharynx (O), and exhaled fraction (E) for each particle size. Data shown are for 30- g albuterol doses. *p 0.001; p All other flow rate comparisons not significant. mucosal drug concentration within the alveoli compared with the proximal conducting region (33). The successful delivery of inhaled medication to the lungs also requires minimal upper airway aerosol losses. Smaller particles largely bypass the filtering mechanisms and abrupt airway geometry of the upper airways, which accounts for their less oropharyngeal deposition, whereas larger particles, which deposit chiefly by impaction due to their greater inertia, are more likely to leave the inspired air stream during sudden changes in airflow direction, particularly in the oropharynx and at airway bifurcations. It is not entirely surprising that, despite delivering the largest lung dose, the smaller particles were also exhaled the Figure 5. Effect of fast inhalation (F) compared with slow inhalation (S) on aerosol deposition in central and intermediate (striped bars) and peripheral (shaded bars) regions expressed as (A ) percentage of radionuclide dose delivered for each particle size and (B) percentage of total lung deposition. Data shown are for 30- g albuterol doses. Figure 6. Effect of fast inhalation (shaded bars) compared with slow inhalation (white bars) on(a ) aerosol penetration into the lung and (B ) FEV 1.*p 0.001; p ns not significant. most, as our previous in vitro data have shown that 1.5- m particles can remain airborne for a considerable time, even with the breath-hold pause maximizing the effect of gravitational sedimentation (23). Particles that remain airborne in the larger airways are likely to be exhaled due to a greater settling distance before coming into contact with the airway walls. Hence, it would be expected that much less drug is deposited in the conducting airway region for smaller particles. On reaching the lungs, by virtue of their particle size deposition characteristics, C I region aerosol deposition, as a proportion of the lung dose, was significantly greater with 6- m particles, whereas 1.5- m particles favored peripheral region deposition (Table 2). The C I region of interest encompasses the majority of the conducting airways (generations 0 16), where 2 -agonists should be directed (34). Yet, C I deposition quantified as a percentage of the delivered dose revealed similar values between the three particles. So, how can this lead to a significant difference in clinical efficacy? Herein lies a limitation of twodimensional planar imaging of the three-dimensional lung structure, as overlying alveolar deposition may have obscured the true conducting airway dose in the C I region. However, this study is unique because our planar images are distinct representations of monodisperse aerosol distributions within the airways, whose specificity allows us to delineate the very nature of the airways contributing to the C I region. Together with experimental data that predict that 1.5- m particles deposit mainly in the alveolar region (31), we can infer that the C I region of interest for the 1.5- m particles is mainly drug deposition in overlying peripheral alveoli rather than in conducting airways, whereas for the 6- m particles, our imaging data support the C I region as deposition predominantly in the conducting airways. Although imaging residual radioactivity in the lungs at 24 h may have given another estimate of alveolar deposition (35), this was not a feasible option in our study due to the relatively quick dissociation of radiolabel from drug (see Figure E1).

6 1502 AMERICAN JOURNAL OF RESPIRATORY AND CRITICAL CARE MEDICINE VOL Physiologic pulmonary function assessment including spirometry is unable to accurately differentiate between the distinct conducting and alveolar airway regions or small ( 2 mm) and large intrapulmonary airways, but newly established techniques are more promising (36, 37). Mathematical modeling of our data may allow an estimate of airway composition in the C I and peripheral regions (38). Such models range from simple empirical, which do not incorporate lung geometry, to complex multiple-path mathematical models based on lung structure constructed from actual measurements (39, 40). Our data are in general agreement with mathematical and experimental predictions when single-breath inhalation, rather than tidal breathing, is considered (41). Single photon emission computed tomography may have achieved greater accuracy in airway definition but is limited by high doses of radioactivity and its much longer image acquisition times (1). Our aim was to investigate the science of how asthmatic airways handle and respond to 2 -agonist aerosols of different particle size in a clinically stable human model rather than to explore symptom response to treatment, or relate mucosal airway dose to the pharmacodynamic response as others have done (42). We found 6- and 3- m particles were more efficacious in all lung function variables compared with 1.5- m particles, at both 15- and 30- g doses. Although the study was not powered to test equivalence with the 200- g MDI albuterol dose, a 30- g delivered dose of 6- m particles gave a better FEV 1 improvement, suggesting that monodisperse aerosols allow a marked reduction in drug dose without compromising efficacy. The estimated lung dose was 16.9, 15.3, and 13.8 g for the 1.5-, 3-, and 6- m particles, respectively. A further advantage of this study was that by using low 2 -agonist doses on the steep part of the dose response curve, our model has good sensitivity for detecting therapeutic improvements due to optimal deposition patterns. Our data are in agreement with others in that lower doses of 2 -agonist than are currently delivered may be used to achieve similar degrees of bronchodilation (43). However, although the large particles inhaled very slowly produced a greater acute bronchodilator effect, it may be that they were cleared more rapidly from the proximal airways by the mucociliary escalator, as opposed to the 1.5- m particles preferentially depositing in the nonciliated alveolar regions of the lung periphery. The improvement in FEV 1 after inhalation of a 2 -agonist mainly represents smooth muscle relaxation, as well as a balance between that of increased mucociliary clearance, through a combination of increased airway ciliary beat and mucus hydration (44, 45), and of increased mucus secretion, although -agonists only weakly stimulate mucus output in human airways (46). Thus, the greater peripherally deposited and slowly cleared fine-particle albuterol aerosol may have had a more sustained bronchodilator response than the larger particles. We, however, were unable to assess this effect, as lung function was not measured beyond 90 min after inhalation of the doses. In contrast, the peripherally deposited 1.5- m particles have the potential to be absorbed more quickly into the systemic circulation and, in effect, clear faster from the airways compared with the actions of the proximal mucociliary apparatus on the larger particles. Furthermore, as a result of this peripheral deposition systemic absorption relationship, we would have expected a greater effect on tolerability markers with the 1.5- m particles compared with the larger particles. The lack of effect on tolerability is most likely a consequence from the extremely low monodisperse aerosol drug doses used. Although particle size effects in lung function between 15- and 30- g doses were not statistically significant for all paired same-size comparisons, it was observed that doubling the drug dose to 30 g had a minimal effect on the clinical response with the 1.5- m particles ( FEV 1, 10 ml), indicating a flat dose response relationship and reinforcing the notion that these particles were largely delivered to the incorrect lung region with regard to bronchodilation. In contrast, for the 3- m particles, and to a greater extent with the 6- m particles, doubling the drug dose had a greater therapeutic effect ( FEV 1,37and66ml, respectively), suggesting the larger particles were delivered to their target site of 2 -agonist action to achieve effective bronchodilation. The primary clinical endpoint used to assess the bronchodilator dose response effects and comparative efficacy between particle sizes was FEV 1. Although the 6- and 3- m salbutamol particles at both 15- and 30- g doses were more efficacious in FEF and PEF measures compared with the 1.5- m particles, no clear trend was seen with FVC (see Table E6). FVC and PEF are not only associated with a greater degree of variability than FEV 1 but are also not generally regarded as the most sensitive indicator of airflow obstruction. Our particle size data may be usefully related to the main categories of commercial devices (see Table E9). The aerosol generation system and mode of delivery of our apparatus is analogous to that of an MDI with spacer, where the drug particles are already airborne at the point of inhalation and there is no requirement for breath actuation or coordination. Particle size generation within our system was not flow-dependent as our apparatus is a low-resistance system, with no carrier particles and no requirement to aerosolize the drug with the patient s inspiratory breath. In contrast, devices such as the Turbuhaler and the Diskus, both dry-powder inhalers, are dependent on the energy of the patient s inspiratory force to aerosolize the drug particles from within the device by separating the lactose carrierdrug particle complexes (47, 48). Nebulizers, on the other hand, are usually dependent on tidal-breathing delivery methods as opposed to our bolus single-breath delivery. Although commercial devices do not generate monodisperse aerosols, much of the output aerosol size distribution and drug dose from these systems is between 3- and 6- m mass median aerodynamic diameter as assessed by in vitro techniques, which is the range of sizes we have shown in vivo in our study to be most potent, at least for bronchodilators. Yet, with such devices, the drug dose is distributed across a wide polydisperse particle size distribution. We believe for albuterol that monodisperse aerosols with their narrow distribution, generated at an optimal drug particle size, are essentially encompassing the best fraction of the polydisperse aerosol distribution; that is, the drug dose contained within the most effective part of the aerosol distribution, while avoiding the ineffective or wasted particle-size dose fraction of the polydisperse distribution that may give rise to adverse effects. Our data support the rationale for delivering aerosols to the lungs with a narrower geometric standard deviation than those devices in current use. It should be appreciated, though, that airway deposition results from an intricate relationship between aerosol particle size and other factors such as inspiratory flow and airway geometry, and in vivo studies and determinations will always be necessary. We have shown that inspiratory flow plays an important role in influencing deposition efficiency and that the altered airway distribution patterns account for the observed clinical response. At higher inspiratory flows, penetration index decreased for all particle sizes, implying a proximal airway shift of aerosol distribution compared with slow inhalation. Oropharyngeal deposition increased particularly for 3- and 6- m particles more prone to inertial impaction, with a corresponding decrease in their TLD and a consequent fall in FEV 1. During fast inhalation, the 1.5- m particles achieved better bronchodilation compared with the 6- and 3- m particles, which may be due to a combination of greater lung deposition and relatively more proximal

7 Usmani, Biddiscombe, and Barnes: 2 -Agonist Regional Lung Deposition 1503 airway distribution. Indeed, compared with slow inhalation, the 1.5- m particles achieved a minor improvement in FEV 1 (22 ml), which was not significant and could not be entirely accounted for by the small, albeit statistically significant, increase in TLD (4.6%). This suggested that the proximal intrathoracic airway shift of drug deposition from alveolar to conducting regions at higher inspiratory flows must be of importance, and would imply better matching of the 1.5- m drug particles to airway smooth muscle and 2 -receptors compared with slow inhalation conditions. It is tempting to conclude that inhaled 2 -agonist aerosols should be delivered by smaller particles because they are least affected by flow rate and particle inertia and, subsequently, may be more forgiving of poor inhaler technique. However, the bronchodilator response achieved by 1.5- m particles at higher inspiratory flows was substantially lower compared with 3- and 6- m particles during slow inhalation, and our data highlight the dynamic interaction between particle size and inspiratory flow. The mode of aerosol inhalation has also been shown to determine drug deposition and distribution within the intrapulmonary airways (19). Experimental data and mathematical models predict lung deposition increases with increasing particle size from 0.5 to 10 m under tidal-breathing conditions (39). However, using single-breath inhalations augmented with a breath-hold pause, the deeper breaths and increased lung residence time significantly favors the smaller particles, which have lower inertial losses, allowing their deposition and lung penetration to exceed that of the larger particles (49). In conclusion, the data support our hypothesis that targeting drug to regional airways is important in maximizing the therapeutic response and, for 2 -agonists, conducting airway deposition is more significant than alveolar deposition, or indeed, TLD. Larger albuterol particles, when correctly delivered, achieve greater bronchodilation than smaller particles, as there is better matching between their intrapulmonary distribution and airway smooth muscle. We believe the optimal physical aerosol characteristics and patient delivery profiles need to be definitively researched for each inhaled drug class and aerosol therapy tailored to the type and severity of respiratory disease, the predominant site of airway pathology and age of patient. For example, inhaled corticosteroids act on the glucocorticoid receptor, which is present throughout the bronchial tree (50), whereas the pathology in emphysema is largely in the peripheral airways (51). Of concern, the outdated inefficient polydisperse aerosol device technology of the last 50 yr is being applied to the development of a new generation of inhalers, without precise exploration of the basic aerosol science to guide the progress of efficient purposespecific devices that should be developed. Conflict of Interest Statement : O.S.U. has previously served on an advisory board for GlaxoSmithKline (GSK) and has received lecture fees from GSK and Boehringer Ingelheim. M.F.B. has received a research grant and been reimbursed for attending several conferences by GSK. P.J.B. has previously served as a consultant to GSK; is a member of scientific advisory boards for GSK, Boehringer Ingelheim, Altana, Pfizer, and Duska; has received lecture fees from GSK, AstraZeneca, Boehringer Ingelheim, and unrestricted grants from GSK, AstraZeneca, Boehringer Ingelheim, Novartis, Millenium, and Scios. Acknowledgment : The authors thank Sally Meah for assistance with the study, and acknowledge Alison Moore and Sally Stone for their support, and Mike Williams and Deborah Tandy for help with the statistical analysis. References 1. Newman SP, Pitcairn GR, Hirst PH, Rankin L. Radionuclide imaging technologies and their use in evaluating asthma drug deposition in the lungs. Adv Drug Deliv Rev 2003;55: Pauwels R, Newman S, Borgstrom L. Airway deposition and airway effects of antiasthma drugs delivered from metered-dose inhalers. Eur Respir J 1997;10: Mitchell DM, Solomon MA, Tolfree SE, Short M, Spiro SG. Effect of particle size of bronchodilator aerosols on lung distribution and pulmonary function in patients with chronic asthma. Thorax 1987;42: Nieminen MM, Holli H, Lahdensuo A, Muittari A, Karvonen J. Aerosol deposition in automatic dosimeter nebulization. Eur J Respir Dis 1987; 71: Zainudin BM, Tolfree SE, Short M, Spiro SG. Influence of breathing pattern on lung deposition and bronchodilator response to nebulised salbutamol in patients with stable asthma. Thorax 1988;43: Johnson MA, Newman SP, Bloom R, Talaee N, Clarke SW. Delivery of albuterol and ipratropium bromide from two nebulizer systems in chronic stable asthma: efficacy and pulmonary deposition. Chest 1989; 96: Zainudin BM, Biddiscombe M, Tolfree SE, Short M, Spiro SG. Comparison of bronchodilator responses and deposition patterns of salbutamol inhaled from a pressurised metered dose inhaler, as a dry powder, and as a nebulised solution. Thorax 1990;45: Ruffin RE, Montgomery JM, Newhouse MT. Site of beta-adrenergic receptors in the respiratory tract: use of fenoterol administered by two methods. Chest 1978;74: Ruffin RE, Dolovich MB, Oldenburg FA, Newhouse MT. The preferential deposition of inhaled isoprotenerol and propanolol in asthmatic patients. Chest 1981;80: Newman SP, Moren F, Trofast E, Talaee N, Clarke SW. Deposition and clinical efficacy of terbutaline sulphate from Turbuhaler, a new multidose powder inhaler. Eur Respir J 1989;2: Newman SP, Weisz AW, Talaee N, Clarke SW. Improvement of drug delivery with a breath actuated pressurised aerosol for patients with poor inhaler technique. Thorax 1991;46: Newman SP, Talaee N, Clarke SW. Salbutamol aerosol delivery in man with the Rondo spacer. Acta Ther 1991;17: Newman SP, Moren F, Trofast E, Talaee N, Clarke SW. Terbutaline sulphate Turbuhaler: effect of inhaled flow rate on drug deposition and efficacy. Int J Pharm 1991;74: Hultquist C, Wollmer P, Eklundh G, Jonson B. Effect of inhaled terbutaline sulphate in relation to its deposition in the lungs. Pulm Pharmacol 1992;5: Melchor R, Biddiscombe MF, Mak VH, Short MD, Spiro SG. Lung deposition patterns of directly labelled salbutamol in normal subjects and in patients with reversible airflow obstruction. Thorax 1993;48: Newman SP, Clarke SW. Bronchodilator delivery from Gentlehaler, a new low-velocity pressurized aerosol inhaler. Chest 1993;103: Vidgren M, Arppe J, Vidgren P, Hyvarinen L, Vainio P, Silvasti M, Tukiainen H. Pulmonary deposition and clinical response of 99mTclabelled salbutamol delivered from a novel multiple dose powder inhaler. Pharm Res 1994;11: Morrow PE. An evaluation of the physical properties of monodisperse and heterodisperse aerosols used in the assessment of bronchial function. Chest 1981;80: Dolovich M. Influence of inspiratory flow rate, particle size, and airway calibre on aerosolised drug delivery to the lung. Respir Care 2000; 45: Usmani OS, Biddiscombe MF, Nightingale JA, Underwood SR, Barnes PJ. Effects of bronchodilator particle size in asthmatic patients using monodisperse aerosols. J Appl Physiol 2003;95: Usmani OS, Biddiscombe MF, Meah S, Barnes PJ. Bronchodilator aerosol particle size effects and regional lung deposition in asthmatics [abstract]. Fourth World Asthma Meeting. Bangkok: Asthma Foundation of Thailand; p Usmani OS, Biddiscombe MF, Barnes PJ. Effects of varying particle size and inhalation flow rate on lung deposition and clinical response using monodisperse albuterol aerosols [abstract]. J Aerosol Med 2003; 200: Biddiscombe MF, Usmani OS, Barnes PJ. A system for the production and delivery of monodisperse salbutamol aerosols to the lungs. Int J Pharm 2003;254: Usmani OS, Biddiscombe MF, Underwood SR, Barnes PJ. Characterization of the generation of radiolabeled monodisperse albuterol particles using the spinning-top aerosol generator. J Nucl Med 2004;45: Snell NJ, Ganderton D. Assessing lung deposition of inhaled medications. Consensus statement: British Association for Lung Research. Respir Med 1999;93: Agnew JE, Bateman JR, Pavia D, Clarke SW. A model for assessing bronchial mucus transport. J Nucl Med 1984;25:

8 1504 AMERICAN JOURNAL OF RESPIRATORY AND CRITICAL CARE MEDICINE VOL Agnew JE. Characterizing lung aerosol penetration. J Aerosol Med 1991; 3: Altman DG. Practical statistics for medical research. London: Chapman & Hall; Barnes PJ, Basbaum CB, Nadel JA, Roberts JM. Localization of betaadrenoreceptors in mammalian lung by light microscopic autoradiography. Nature 1982;299: Ebina M, Yaegashi H, Chiba R, Takahashi T, Motomiya M, Tanemura M. Hyperreactive site in the airway tree of asthmatic patients revealed by thickening of bronchial muscles: a morphometric study. Am Rev Respir Dis 1990;141: Gerrity TR, Lee PS, Hass FJ, Marinelli A, Werner P, Lourenco RV. Calculated deposition of inhaled particles in the airway generations of normal subjects. J Appl Physiol 1979;47: Heyder J, Gebhart J, Rudolf G, Schiller CF, Standlhofen W. Deposition of particles in the human respiratory tract in the size range um. J Aerosol Sci 1986;17: Clarke SW. Aerosols as a way of treating patients. Eur J Respir Dis Suppl 1986;146: Wiebel ER. Morphometry of the human lung. Berlin: Springer-Verlag; Bennett WD, Chapman WF, Lay JC, Gerrity TR. Pulmonary clearance of inhaled particles 24 to 48 hours post deposition: effect of betaadrenergic stimulation. J Aerosol Med 1993;6: Verbanck S, Schuermans D, Meysman M, Paiva M, Vincken W. Noninvasive assessment of airway alterations in smokers: the small airways revisited. Am J Respir Crit Care Med 2004;170: Verbanck S, Schuermans D, Noppen M, Van Muylem A, Paiva M, Vincken W. Evidence of acinar airway involvement in asthma. Am J Respir Crit Care Med 1999;159: Martonen TB, Yang Y, Dolovich M. Definition of airway composition within gamma camera images. J Thorac Imaging 1994;9: Asgharian B, Hofmann W, Bergmann R. Particle deposition in a multiplepath model of the human lung. Aerosol Sci Technol 2001;34: Martonen T, Zhang Z, Yu G, Musante C. Three-dimensional computer modelling of the human upper respiratory tract. Cell Biochem Biophys 2001;35: Rudolf G, Kobrich R, Stahlhofen W. Modelling and algebraic formulation of regional aerosol deposition in man. J Aerosol Sci 1990;21:s403 s Ruffin RE, Kenworthy MC, Newhouse MT. Response of asthmatic patients to fenoterol inhalation: a method of quantifying the airway bronchodilator dose. Clin Pharmacol Ther 1978;23: Fishwick D, Bradshaw L, Macdonald C, Beasley R, Gash D, Bengtsson T, Bondesson E, Borgstrom L. Cumulative and single-dose design to assess the bronchodilator effects of beta2-agonists in individuals with asthma. Am J Respir Crit Care Med 2001;163: Devalia JL, Sapsford RJ, Rusznak C, Toumbis MJ, Davies RJ. The effects of salmeterol and salbutamol on ciliary beat frequency of cultured human bronchial epithelial cells, in vitro. Pulm Pharmacol 1992;5: Salathe M. Effects of beta-agonists on airway epithelial cells. J Allergy Clin Immunol 2002;110:S275 S Phipps RJ, Williams IP, Richardson PS, Pell J, Pack RJ, Wright N. Sympathomimetic drugs stimulate the output of secretory glycoproteins from human bronchi in vitro. Clin Sci (Lond) 1982;63: Kamin W, Genz T, Roeder S, Scheuch G, Trammer T, Juenemann R, Cloes R. Mass output and particle size distribution of glucocorticosteroids emitted from different inhalation devices depending on various inspiratory parameters. J Aerosol Med 2002;15: Tarsin W, Assi KH, Chrystyn H. In-vitro intra- and inter-inhaler flow rate-dependent dosage emission from a combination of budesonide and eformoterol in a dry powder inhaler. J Aerosol Med 2004;17: Kim CS. Methods of calculating lung delivery and deposition of aerosol particles. Respir Care 2000;45: Adcock IM, Gilbey T, Gelder CM, Chung KF, Barnes PJ. Glucocorticoid receptor localization in normal and asthmatic lung. Am J Respir Crit Care Med 1996;154: Jeffery PK. Remodeling in asthma and chronic obstructive lung disease. Am J Respir Crit Care Med 2001;164:S28 S Working Party of the European Community for Coal and Steel. Standardization of lung function tests. Clin Respir Physiol 1983;19:7 10.