In vitro deposition of fluticasone aerosol from a metered-dose inhaler with and without two common valved holding chambers Michael J. Asmus, PharmD; Judy Liang, PharmD; Intira Coowanitwong, MS; Ramin Vafadari; and Guenther Hochhaus, PhD
Background: Previous in vitro aerosol deposition experiments indicate that the corticosteroid respirable dose from a metered-dose inhaler (MDI) can vary by threefold depending on the specific valved holding chamber (VHC) MDI combination. Objective: We compared in vitro aerosol deposition from a fluticasone propionate MDI (Flovent, GlaxoSmithKline, Research Triangle Park, NC) to that of the same MDI used in combination with two VHCs (EasiVent, Dey, Napa, Ca; and AeroChamber-Plus, Monaghan Medical Corp, Plattsburgh, NY) to evaluate how these VHCs affect the respirable dose of fluticasone. Methods: The respirable dose (aerosol particles 1 to 5 m in size) of fluticasone was determined by sampling 5 ⫻ 110-g actuations from each configuration (MDI alone, MDI plus AeroChamber-Plus, and MDI plus EasiVent) in multiples of ten using a well established, in vitro cascade impactor method. Fluticasone aerosol was washed from individual impactor stages with 50% methanol and quantified via ultraviolet high-pressure liquid chromatography. Differences among outcomes were determined using analysis of variance. Results: Mean respirable dose from AeroChamber-Plus (47.9 ⫾ 6.9 g/actuation) was not different (P ⬎ 0.05) from that produced by the MDI alone (50.3 ⫾ 2.2 g/actuation). EasiVent respirable dose (27.0 ⫾ 3.6 g/actuation) was less than that produced by either the AeroChamber-Plus or the MDI alone (P ⬍ 0.001). Conclusions: VHCs do not display equivalent in vitro performance with a fluticasone MDI. If a patient needs a VHC, clinicians should use available in vitro performance information to aid in selecting the best device. Ann Allergy Asthma Immunol 2002;88:204–208.
INTRODUCTION Several metered-dose inhaler (MDI) valved holding chambers (VHCs) are available today in the United States. The primary purpose of any VHC is to briefly hold the MDI puff before inhalation to minimize the detrimental effects of poor timing between MDI actuation and patient inhalation. Another important function of all VHCs is to allow the MDI propellant time to evaporate after actuation. Evaporation of
The College of Pharmacy, University of Florida, Gainesville, Florida. This study was sponsored by Forest Laboratories, Inc, New York, NY. Received for publication July 26, 2001. Accepted for publication in revised form October 15, 2001.
propellant promotes the formation of smaller aerosol particles that are likely to be carried into the small airways. Further, all VHCs function to collect “large” aerosol particles that would otherwise deposit into the oropharynx. The National Institutes of Health National Asthma Education and Prevention Program recommends a VHC for all patients who can not effectively use a MDI alone and a VHC for all asthma patients prescribed an inhaled corticosteroid to reduce the risk of topical corticosteroid-induced adverse effects such as thrush and dysphonia.1 In vitro aerosol deposition data indicate that VHC choice has no impact on the aerosol characteristics of bronchodilators from an MDI.2 Further, clinical studies in asthmatic subjects
have been unable to discern measurable differences in bronchodilator effect among various albuterol MDIVHC combinations.3,4 Therefore, many clinicians and third-party payers conclude that all VHCs available in the United States are equivalent. However, this assumption may not be true for inhaled corticosteroids, as several aerosol deposition experiments indicate drug delivery characteristics are unique for each inhaled corticosteroidVHC combination.2,5–7 These in vitro studies indicate that the quantity of inhaled corticosteroids available for delivery to the human lung can vary by as much as threefold, depending on the VHC.2 Therefore, interchanging one VHC with another may result in wide fluctuations in the aerosol dose reaching the lungs and potentially result in therapeutic failures. Fluticasone propionate MDI (Flovent, GlaxoWellcome, Research Triangle Park, NC) is the most often prescribed inhaled glucocorticoid in the United States for patients with asthma.8 The objective of the present study was to compare the in vitro aerosol deposition characteristics of a fluticasone propionate MDI alone to the aerosol deposition of the same MDI when used in combination with two common VHCs to determine whether or not these VHCs are equivalent. METHODS Sample Collection The deposition characteristics of fluticasone aerosol delivered from a fluticasone propionate 110 g MDI alone, a fluticasone propionate MDI (Flovent) attached to the AeroChamber-Plus VHC
ANNALS OF ALLERGY, ASTHMA, & IMMUNOLOGY
(Monaghan Medical Corp, Plattsburgh, NY), and a fluticasone propionate MDI (Flovent) attached to the EasiVent VHC (Dey, Napa, CA) were examined using an established in vitro cascade impactor method.9,10 Aerosols particles that enter the cascade impactor are sorted by aerodynamic diameter onto eight different stages as shown in Figure 1. Aerosol deposition onto the “throat” intake port is reflective of in vivo oropharyngeal deposition.11 Aerosol deposition on stages 0, 1, and 2 of the cascade impactor correlate to particles 5 to 10 m in size. These particles are too large for therapeutic use in asthma and will deposit into the large upper airways, eg, the trachea.12 Aerosols that deposit on stages 3, 4, and 5 of the cascade impactor correlate to particles 1 to 5 m in size, a range considered to be ideal for deposition and retention within the small human airways (ie, bronchi and bronchioles).12 Aerosols that collect on stages 6, 7, and in the final filter (F) of the cascade impactor correlate to particles ⬍1 m in size, a range considered to be too small for therapeutic use in asthma and one that favors deposition into the terminal bronchioles and alveoli.12 Fluticasone aerosol from all three configurations was sampled into the cascade impactor (Mark II-ACFM eight-stage nonviable ambient sampler, Andersen Instruments Inc, Atlanta, GA) through an artificial throat manufactured according to United States Pharmacopoeia (USP) standards.10 Aerosol was drawn into the impactor at 28.3 L/minute ⫾ 5% by a vacuum pump (Graseby-Andersen, Model 20 –709, Atlanta, GA). Resultant airflow through the impactor was calibrated to conform to the manufacturer’s requirements and simulated a tidal inspiratory flow of approximately 30 L/minute. Before each sampling run, airflow through the impactor was allowed to equilibrate for at least 5 minutes. Before each run, each fluticasone propionate MDI canister was primed five times (in a separate room) to assure uniformity among actuations. Each MDI canister was shaken for 10 seconds before each actuation. Before
VOLUME 88, FEBRUARY, 2002
Figure 1. Schematic diagram of the cascade impactor sampling method for all three configurations tested. The aerodynamic size range of aerosol collection at each impactor stage is listed on the righthand side of the schematic diagram.
each sampling run, the MDI-valve was washed twice with methanol and dried to remove any trace drug that resulted from priming. Each configuration was tested 10 times and, in each trial, five actuations were sampled through the cascade impactor. After each actuation into the impactor, airflow was maintained for 30 seconds before the next actuation. After the final actuation, flow through the impactor was maintained for an additional 30 seconds. Laboratory personnel wore antistatic outer clothing containing carbon fibers during sample collection to minimize transfer of static charges to VHCs. All VHCs were washed in a liquid detergent solution and allowed to air-dry before use to minimize static attraction of fluticasone aerosol during sample collection.13 Ambient temperature (20° ⫾ 2° C) and humidity (64 ⫾ 1%) in the aerosol laboratory were strictly controlled to minimize interday variability in sample collection.
Fluticasone Assay After each trial, fluticasone was washed from the MDI-valve, actuator, VHC, USP throat, and each of eight collection plates within the cascade impactor with 5 to 25 mL of methanol and an equal volume of distilled water. The volume used to wash each section of the apparatus was varied to eliminate concentrations of fluticasone below the assay limit of detection. Three 1-mL aliquots of the resulting solutions were collected for high-pressure liquid chromatography (HPLC) analysis. Budesonide was added to each aliquot as an internal standard (final concentration ⫽ 5 g/mL) during the HPLC analysis. Fluticasone calibration standard and quality controls curves were prepared in a similar methanol and water mixture (50:50 vol/vol) from two different 100-g/mL fluticasone stock solutions. Fluticasone content from all washings was determined by injecting 200-L samples into a re-
verse-phase HPLC system with a 254-nm ultraviolet detector. Correlation coefficients for the standard curves (ie, peak area ratio of fluticasone propionate and budesonide against concentration) were between 0.998 and 1.000. The lower limit for the quantitation of fluticasone in methanol and water was 0.5 g/mL ⫾ 7.5%. Data Analysis The primary outcome measures were the quantity of fluticasone aerosol emitted within the ideal size range (1 to 5 m) for deposition and retention in the small human airways (ie, respirable dose), and the quantity collected in the throat intake port (ie, oropharyngeal dose). The respirable dose and oropharyngeal dose were normalized to reflect the fact that a fluticasone propionate 110 g MDI (Flovent) actually dispenses 125 g from the canister during each actuation.14 Labeling for this product in the United States is based upon the quantity of fluticasone delivered at the MDI mouthpiece.14 Differences among outcomes were determined using one-way analysis of variance. A Tukey’s multiple comparison procedure was used to compare individual means when the overall analysis of variance results were statistically significant (P ⬍ 0.05). All data analysis was performed using Minitab release-12 statistical software (Minitab Inc, State College, PA). RESULTS The mean (⫾ SD) normalized respirable dose (ie, aerosol size 1 to 5 m) of fluticasone from AeroChamber-Plus VHC (47.9 ⫾ 6.9 g/actuation) was not different (P ⬎ 0.05) compared with that produced by the fluticasone propionate-110 MDI alone (Flovent; 50.3 ⫾ 2.2 g/actuation). However, EasiVent VHC produced a mean normalized respirable dose (27.0 ⫾ 3.6 g/actuation) that was less than that produced by either the AeroChamberPlus VHC or the fluticasone propionate-110 MDI alone (P ⬍ 0.001). The mean normalized oropharyngeal dose from the fluticasone propionate-110
MDI alone (48.0 ⫾ 7.2 g/actuation) was greater (P ⬍ 0.001) than the oropharyngeal dose from either the AeroChamber-Plus (1.0 ⫾ 1.8 g/actuation) or EasiVent (1.7 ⫾ 3.9 g/ actuation) VHCs. The difference in oropharyngeal dose between the two VHCs was not significant (P ⬎ 0.05). A graphical representation of these results is provided in Figure 2. DISCUSSION The highest respirable dose we observed was from the fluticasone propionate MDI alone, although this result was not significantly larger than the respirable dose observed from the MDI attached to the AeroChamber-Plus. These data suggest that using the AeroChamber-Plus VHC in combination with a fluticasone propionate-110 MDI does not alter the respirable dose available to an asthmatic patient. The respirable dose of fluticasone from the EasiVent VHC was 45% less com-
pared with that from the MDI alone or the AeroChamber-Plus. Although the dose-response relationship for inhaled corticosteroids in the treatment of asthma has been difficult to establish, a 45% decrease in respirable dose of inhaled fluticasone is likely to be a clinically relevant difference.15 Both VHCs tested decreased significantly the dose of fluticasone aerosol trapped in the oropharynx compared with the MDI alone. This reduction in oropharyngeal dose is reflective of the fact that large, unrespirable aerosol particles are collected in the VHC rather than in the impactor throat intake port when either VHC is added. These data suggest that either VHC tested would be an acceptable method to decrease the oropharyngeal dose and thereby diminish the risk of topical fluticasoneinduced adverse effects such as thrush and dysphonia. Few comparisons of VHC performance have been published to date.
Figure 2. Mean dose of fluticasone ⫾ standard deviation within the ideal size range (1 to 5 m) for retention in small human airways (ie, respirable dose), normalized per actuation, displayed as light bars. Mean normalized respirable dose of fluticasone from AeroChamber-Plus (AC-Plus; 47.9 g/actuation) was not different from that produced by fluticasone propionate-110 MDI alone (50.3 g/actuation; P ⬎ 0.05). EasiVent produced a mean normalized respirable dose (27.0 g/actuation) that was less than that produced by either the AeroChamber-Plus or fluticasone propionate MDI alone (P ⬍ 0.001). Mean dose of fluticasone aerosol ⫾ standard deviation collected in the apparatus USP throat (ie, oropharyngeal dose), normalized per actuation, displayed as the dark bars. Mean normalized oropharyngeal dose of fluticasone from the Flovent MDI alone (MDI; 48.0 g) was larger than oropharyngeal dose from either VHC (P ⬍ 0.001). Mean normalized oropharyngeal dose from the two VHCs was not different (P ⬎ 0.05).
ANNALS OF ALLERGY, ASTHMA, & IMMUNOLOGY
Those that do exist highlight performance characteristics of VHCs available outside the United States. A review of articles indexed by Medline, abstracts presented at American Thoracic Society; American Academy of Allergy, Asthma, and Immunology; American College of Chest Physicians; as well as the American College of Asthma, Allergy and Immunology annual meetings over the past 3 years revealed no information about the performance of the EasiVent VHC. Because AeroChamber-Plus is newly available, little information, all of it in the form of scientific abstracts, has been published regarding its performance.16 –18 Our results are the first to compare the performance of the AeroChamber-Plus VHC to any MDI alone as well as to the EasiVent VHC. The biggest limitation of the present study is that in vitro results do not necessarily correlate with in vivo outcomes. Unfortunately, it is impossible to duplicate exact human physiology using a machine model. We chose our method because it is the one specified by the USP.10 Patient specific variables such as breathing pattern, lung volume, and lung function are not accounted for in this study. Future work should include measurement of aerosol deposition under more physiologic conditions, especially those physiologic conditions likely to be encountered in patients with asthma. Another limitation of the present study is that our results are only reflective of performance characteristics of these two VHCs with the current fluticasone propionate MDI (Flovent) formulation. Our results may not be reflective of performance with other inhaled formulations of the same medication or of different inhaled glucocorticoids. Although our study was conducted at only one flow rate (28.3 L/minute), other investigators have show that flow rates of 60 and 90 L/minute have little impact on the respirable dose from several corticosteroid MDIs when compared with that achieved at 30 L/minute.19,20
VOLUME 88, FEBRUARY, 2002
Differences exist in the in vitro performance characteristics of the two VHCs tested with a fluticasone MDI. These data suggest that an asthmatic patient would receive the same dose of fluticasone to the lungs when using a MDI alone with good inhaler technique or the same MDI attached to an AeroChamber-Plus VHC. Our data further suggest that the same fluticasone MDI when mated to an EasiVent VHC, would result in 45% less fluticasone reaching the lungs of an asthmatic patient compared with the MDI alone. This difference is likely to be clinically relevant. Finally, our data suggest that either VHC tested would be an effective means to decrease the quantity of fluticasone deposited into the oropharynx and thereby diminish the risk of topical adverse effects such as thrush and dysphonia.
ACKNOWLEDGMENTS The authors thank Yaning Wang, MS, for his technical assistance and Jeff Stark, MS, for assistance with preparation of this manuscript.
REFERENCES 1. National Asthma Education and Prevention Program. Expert Panel Report 2. Guidelines for the diagnosis and management of asthma. US Department of Health and Human Services, National Institutes of Health: Bethesda, MD. 1997. NIH Publication no 97– 4051;10. 2. Ahrens R, Lux C, Bahl T, Han S. Choosing the metered-dose inhaler spacer or holding chamber that matches the patient’s need: evidence that specific drug being delivered is an important consideration. J Allergy Clin Immunol 1995;96:288 –294. 3. Gunawardena KA, Sohal T, Jones JI, et al. The spacehaler for delivery of salbutamol: a comparison with the standard metered-dose inhaler plus volumatic spacer device. Respir Med 1997;91:311–316. 4. Santati S, Preuthipan A. Clinical comparison of a homemade fruit juice
spacer and two standard spacers in children. Am J Respir Crit Care Med 2000;161:A343. Barry PW, O’Callaghan C. The output of budesonide from spacer devices assessed under simulated breathing conditions. J Allergy Clin Immunol 1999; 104:1205–1210. Barry PW, O’Callaghan C. A comparative analysis of the particle size output of beclomethasone diproprionate, salmeterol xinafoate and fluticasone propionate metered dose inhalers used with the Babyhaler, Volumatic and Aerochamber spacer devices. Br J Clin Pharmacol 1999;47:357–360. Dalby RN, Somaraju S, Chavan VS, Jarvis D. Evaluation of aerosol drug output from the OptiChamber and AeroChamber spacers in a model system. J Asthma 1998;35:173–177. Latner AW. Top 200 drugs of 1999. Pharm Times 2000;66:16 –32. Olson B. Can in vitro measurements predict the lung deposition of drypowder inhalers? J Aerosol Med 1997; 10:S13–S16. Physical tests and determinations, aerosols. The Pharmacopeia of the United States of America, 24th ed. Rockville, MD: USP Convention, 2000. 1895– 1912. Phelan RF, Cuddihy RG, Fisher GL, et al. Main features of the proposed NCRP respiratory tract model. Rad Protein Dos 1991;38:179 –184. Task Group on Lung Dynamics. Deposition and retention models for internal dosimetry of the human respiratory tract. Health Physics 1966;12:173–208. Pierart F, Wildhaber JH, Vrancken I, et al. Washing plastic spacers in household detergent reduces electrostatic charge and greatly improves delivery. Eur Respir J 1999;13:673– 678. Flovent (fluticasone propionate) 110 g inhalation aerosol [package insert]. Research Triangle Park, NC: GlaxoWellcome Inc; 1999. Hofstra WB, Neijens HJ, Duiverman EJ, et al. Dose-responses over time to inhaled fluticasone propionate treatment of exercise- and methacholineinduced bronchoconstriction in children with asthma. Pediatr Pulmonol 2000;29:415– 423. Mitchell JP, Nagel MW, Schmidt JN, Morton RW. Performance of an improved small volume valved holding
chamber. Eur Respir J 2000;16 (Suppl):99. 17. Mitchell J, Nagel M, Coppolo D. In vitro performance of valved holding chamber (VHCs) at flow rates appropriate for low flow rate users. J Allergy Clin Immunol 2000;105:S10. 18. Mitchell JP, Nagel MW, Wiersema KJ, et al. Comparison of a large and a small volume holding chamber (VHC) for the delivery of salmeterol
xinafoate. Ann Allergy Asthma Immunol 2001;86:101. 19. Smith KJ, Chan HK, Brown KF. Influence of flow rate on aerosol particle size distributions from pressurized and breath-actuated inhalers. J Aerosol Med 1998;11:231– 245. 20. Davies N, Feddah M, Brown K, Gipps E. In-vitro characterisation of metered dose inhaler versus dry pow-
der inhaler glucocorticoid products: influence ofinspiratory flow rates. J Pharm Pharm Sci 2000;3:318 –324. Requests for reprints should be addressed to: Michael J. Asmus, PharmD University of Florida Health Sciences Center Box 100486 Gainesville, FL 32610-0486 E-mail: [email protected]
Answers to CME examination—Annals of Allergy, Asthma, and Immunology, February 2002 Berger WE: Monoclonal anti-IgE antibody: a novel therapy for allergic airways disease. Ann Allergy Asthma Immunol 2002;88: 152-161. 1. a 2. c 3. d 4. d 5. c
ANNALS OF ALLERGY, ASTHMA, & IMMUNOLOGY