University of Groningen

The clinical relevance of dry powder inhaler performance for drug deliveryDemoly, Pascal; Hagedoorn, Paul; de Boer, Anne H.; Frijlink, Henderik W.

Published in:Respiratory Medicine

DOI:10.1016/j.rmed.2014.05.009

IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite fromit. Please check the document version below.

Document VersionPublisher's PDF, also known as Version of record

Publication date:2014

Link to publication in University of Groningen/UMCG research database

Citation for published version (APA):Demoly, P., Hagedoorn, P., de Boer, A. H., & Frijlink, H. W. (2014). The clinical relevance of dry powderinhaler performance for drug delivery. Respiratory Medicine, 108(8), 1195-1203.https://doi.org/10.1016/j.rmed.2014.05.009

CopyrightOther than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of theauthor(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).

Take-down policyIf you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediatelyand investigate your claim.

Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons thenumber of authors shown on this cover page is limited to 10 maximum.

Download date: 14-01-2021

Respiratory Medicine (2014) 108, 1195e1203

Available online at www.sciencedirect.com

ScienceDirect

journal homepage: www.elsevier .com/locate/rmed

The clinical relevance of dry powder inhalerperformance for drug delivery

Pascal Demoly a,*, Paul Hagedoorn b, Anne H. de Boer b,Henderik W. Frijlink b

a Departement de Pneumologie et Addictologie, Hopital Arnaud de Villeneuve and University Hospitalof Montpellier, 371 Ave. du Doyen Gaston Giraud, 34295 Montpellier Cedex 5, Franceb Department of Pharmaceutical Technology and Biopharmacy, University of Groningen,Ant. Deusinglaan 1, 9713 AV Groningen, The Netherlands

Received 29 January 2014; accepted 13 May 2014Available online 24 May 2014

KEYWORDSFine particle fraction;Inhaler resistance;Lung deposition;Dry powder inhaler

* Corresponding author. Tel.: þ33 46E-mail address: pascal.demoly@ins

http://dx.doi.org/10.1016/j.rmed.2010954-6111/ª 2014 Published by Elsevi

Summary

Background: Although understanding of the scientific basis of aerosol therapy with dry powderinhalers (DPIs) has increased, some misconceptions still persist. These include the beliefs thathigh resistance inhalers are unsuitable for some patients, that extra fine (<1.0 mm) particlesimprove peripheral lung deposition and that inhalers with flow rate-independent fine particlefractions (FPFs) produce a more consistent delivered dose to the lungs.Objectives: This article aims to clarify the complex inter-relationships between inhaler designand resistance, inspiratory flow rate (IFR), FPF, lung deposition and clinical outcomes, as a bet-ter understanding may result in a better choice of DPI for individual patients.Methods: The various factors that determine the delivery of drug particles into the lungs arereviewed. These include aerodynamic particle size distribution, the inspiratory manoeuvre,airway geometry and the three basic principles that determine the site and extent of deposi-tion: inertial impaction, sedimentation and diffusion. DPIs are classed as either dependent orindependent of inspiratory flow rate and vary in their internal resistance to inspiration. Theeffects of these characteristics on drug deposition in the airways are described using data fromstudies directly comparing currently available inhaler devices.Results: Clinical experience shows that most patients can use a high resistance DPI effectively,even during exacerbations. Particles in the aerodynamic size range from 1.5e5 mm are shownto be optimal, as particles <1.0 mm are very likely to be exhaled again while those >5 mm mayimpact on the oropharynx. For DPIs with a constant FPF at all flow rates, less of the delivereddose reaches the central and peripheral lung when the flow rate increases, risking under-dosing of the required medication. In contrast, flow rate-dependent inhalers increase theirFPF output at higher flow rates, which compensates for the greater impaction on the upperairways as flow rate increases.

7336117; fax: þ33 467042708.erm.fr (P. Demoly).

4.05.009er Ltd.

1196 P. Demoly et al.

Conclusions: The technical characteristics of different inhalers and the delivery and depositionof the fine particle dose to the lungs may be important additional considerations to help thephysician to select the most appropriate device for the individual patient to optimise theirtreatment.ª 2014 Published by Elsevier Ltd.

Introduction

The mainstay of asthma and COPD treatment is via theinhaled route, medication given either to control inflam-mation (inhaled corticosteroids [ICS]) or to prevent orreverse bronchoconstriction (b2-agonist or muscarinicantagonist bronchodilators) [1]. A key advantage of theinhaled route of administration is that lower doses of therequired drug can be delivered directly to the site of ac-tion, resulting in a rapid clinical response with lower po-tential for risk of systemic adverse effects compared withother routes of administration [2,3].

Increased understanding of the scientific basis of aerosoltherapy has led to the development and introduction ofmany new inhaler devices, including nebulisers, pressurisedmetered-dose inhalers (pMDIs) and, particularly, dry pow-der inhalers (DPIs) [3e5]. Approaches including AdaptiveAerosol Delivery (AAD) [6], the AKITA� inhalation system(Activaero, Germany), which controls the breathingmanoeuvre [7] and mist inhalers [8], have utilised andexploited new insights into the relationships between fac-tors such as delivered dose, aerosol particle size, inspira-tory flow rate (IFR), lung deposition and clinical effects.These insights have also led to a greater awareness of theimportance of choice of DPI to suit the individual patient.

Despite these advances in scientific understanding, anumber of misconceptions about inhalers have proved verypersistent. These include the belief that the optimal flowrate or pressure drop to operate a DPI is 60 L/min or 4 kPa,irrespective of the inhaler design. Other misconceptionsinclude the role of the inhaler’s internal resistance, thebelief that extra fine particles (with mass median aero-dynamic diameter [MMAD] <1 mm) give improved peripherallung deposition and the belief that inhalers with flow rateindependent particle size distribution produce a moreconsistent delivered dose to the lungs. However, it isbeginning to become more widely understood that there isno unique IFR nor pressure drop for optimal use of DPIs. Ithas also become accepted that higher resistance DPIs canbe used effectively, both by COPD patients and by childrenwith an asthma exacerbation, and that they can haveseveral advantages with respect to lung deposition [9e13].Furthermore, there are good theoretical arguments (withexperimental support) to question the benefit of extra fine(<1 mm) particles and flow rate independent aerosol de-livery [14]. And finally, lung deposition is not alwaysimproved by inhaling faster or more forcefully [15].

Despite the many recent advances and innovations,inhalation therapy is also still associated with poor adher-ence and many patients have low levels of ability/dexterityor poor breath-to-hand co-ordination and so use their

inhaler device incorrectly/ineffectively [16,17]. Even withtraining, there is a high potential for handling errors,particularly when patients are prescribed different types ofinhaler concurrently [18]. Patients also develop strongpreferences for specific types of inhalers [19e23], whichcan have an impact on adherence, particularly if they areswitched to a different inhaler type [24]. Switching inhalerdevice, particularly if the decision is made withoutconsultation, is not recommended [25]. Care should also betaken even when switching among the same type of inhalere for example a DPI e because different levels of asthmacontrol are achieved even when delivering the same drug[26].

In an effort to reduce the impact of inevitable handlingerrors, research has been done to produce inhalers that areless dependent on the user for delivery of the optimumdose [27]. In this regard, the performance and specificdesign characteristics of a DPI, such as the delivered par-ticle size distribution and their internal resistance, havebeen shown to be important from a clinical perspective, inthat they can be modified to minimise the influence of thepatient’s IFR on the deposition pattern and dose [3,4,16].

The aerosol particle size distribution is known to be oneof the main factors influencing the deposition pattern of adrug in the lungs. Historically, the optimal aerodynamicparticle size range for deposition in the lung was thought tobe <5 mm. The mass fraction of the dose emitted in this sizerange by a particular inhaler is often referred to as the fineparticle fraction (FPF) [28]. The factors determining theFPF generated by a DPI are complex and interrelated [29].

The aims of this article are to clarify the complexinterplay between inhaler design and resistance, inspira-tory flow manoeuvre, FPF and lung deposition, and tohighlight the clinical relevance of this interplay, in terms ofeffectiveness and outcomes.

Particle size matters

Aerodynamic particle size distribution, in combination withthe inspiratory manoeuvre and airway geometry, de-termines the penetration of drug particles into the airwaysand deposition on the walls of the airways e and this de-termines the dose delivered to the target site[13,16,30e33]. Three basic principles are the main de-terminants of the site and extent of deposition e inertialimpaction, sedimentation and (to a lesser extent) diffusion(Fig. 1) [29,30,34]. Particles that are not deposited areexhaled again.

At higher IFRs, most particles >5 mm in diameter willimpact on the oropharynx and in the first airway bifurca-tion. Particles deposited on the oropharynx will be

Figure 1 Dominant particle deposition mechanisms in the respiratory tract and effect of particle size on the likelihood of particlepenetration and deposition in the airways.

Unravelling misconceptions in DPI use 1197

swallowed, potentially contributing to a reduction in effi-cacy and local oropharyngeal or systemic effects. This wasdemonstrated by Usmani and colleagues in a randomised,double-blind, placebo-controlled clinical efficacy study in12 patients with stable mild-to-moderate asthma quanti-fying deposition with salbutamol aerosols of different par-ticle size (either 1.5, 3 or 6 mm MMAD) at two different flowrates (31 or 67 L/min) [35]. The study demonstrated thatdeposition fractions of smaller particles (1.5 and 3 mm) inthe central and peripheral airways are much higher thanthose of larger (6 mm) particles, particularly at the higherIFR (Fig. 2) [35]. By extrapolation of data from this study, itcan be derived that particles of �1 mm have a greater than40% chance of being exhaled again (both flow rates) due totheir low settling velocity, in spite of a short breath-holdperiod. Bronchodilation (measured as change in forcedexpiratory volume in 1 s [FEV1]) following administration ofsalbutamol significantly decreased with particle size at thehigher IFR [35]. Bronchodilation with the 1.5 mm particleswas unaffected by IFR. Oropharyngeal deposition increased

Figure 2 Effect of different inspiratory flow rates (31 or67 L/min) on lung deposition for three different sizes of par-ticle (1.5, 3.0 or 6.0 mm). Adapted from Usmani et al. (2005)[35].

for all particle sizes at the higher IFR, but this was signifi-cantly greater for the larger particles (p < 0.001). Thesefindings highlight the dynamic interaction between FPF andIFR and show that this translates into clinical outcomes, i.e.decreased lung deposition and increased oropharyngealdeposition with the higher IFR for large particle sizes resultsin a reduced bronchodilator effect (Fig. 3) [35]. The Usmanistudy used monodisperse aerosols (geometric standard de-viation <1.22) and similar effects may be expected foraerosols from DPIs with the same MMAD, although they mayhave slightly wider size distributions.

Although, historically, the optimal aerodynamic particlesize range for deposition in the lung was thought to be<5 mm, basic physics, deposition modelling and also in vivodeposition studies (e.g. Usmani et al. [35]) indicate thatparticles smaller than 1.0 mm are undesirable because oftheir extremely low settling velocity [29]. Because thesettling velocity increases with the square of the particle

Figure 3 Clinical effects of relationship between inspiratoryflow rate (slow [31 L/min] and fast [67 L/min]) and particle sizefor monodisperse salbutamol aerosols in 12 patients with stablemild-to-moderate asthma [35]. FEV1, forced expiratory volumein 1 s; ns, not significant.

1198 P. Demoly et al.

diameter, the probability of deposition by sedimentationincreases rapidly when the particle diameter is increased.For 1.5 mm particles, the time to travel the same distanceby falling is 2.25 times shorter than that for 1 mm particles.Also, the chance of inertial deposition in the larger airwaysincreases with increasing diameter, but for particles ofapproximately 1.5 mm, the oropharyngeal deposition is stilllow and widely flow rate independent [35]. Therefore,particles of 1.5 mm are more appropriate than <1.0 mmparticles and the optimal size range for inhalation seems tobe 1.5e5 mm rather than <5 mm.

Figure 4 (A) FPF <5 mm by pressure drop across differentinhalers and (B) flow rates of various inhalers at which the FPF<5 mm reaches 25% of the label claim. FPF, fine particlefraction.

Drug distribution and concentration in thewhole lung

For aerosols from DPIs a reasonably equal distribution(1:1:1) of the total lung dose has been found between theupper (conducting), intermediate (transitional) and lower(peripheral) airways [33,36]. Usmani et al. reported com-parable lung distributions for 1.5e3 mm particles inhaled at30e67 L/min: about two-thirds of total lung dose wasdeposited in the central plus intermediate and one-third inthe peripheral airways (Fig. 2) [35]. This seems to be in fairagreement with the lung distribution from DPIs, althoughthe definitions for central, intermediate and peripherallung may have been different. However, due to the expo-nential increase in airway surface area towards the alveoli,so that over 95% of lung surface area is within peripheralairways, the actual concentration for a given dose of drug issignificantly higher in the conducting airways than therespiratory airways [33,37]. Differences in average con-centration between the conducting and peripheral airwaysfor a 1:1:1 deposition ratio may approach a factor of 100.Deviations from the 1:1:1 ratio in the deposition pattern donot really level out this extreme concentration difference.Even for a ratio of 1:2:3 in favour of a much higher pe-ripheral deposition, which is difficult to achieve in practicewith pMDIs and DPIs, the average concentration differencebetween conducting and peripheral airways is still a factor>30. At the clinical level, this may be highly relevant fordrugs that need to particularly target the smaller airways,like corticosteroids, or that need to be more evenlydistributed over the entire lung, like antibiotics. Only forbronchodilators has this concentration gradient beenconsidered less relevant so far. Although b2-agonists andanticholinergics target autonomic receptors on the airwaysmooth muscle of large and small airways [38], cholinergicactivity of the lung is most pronounced in the large airways[39]. Additionally, the amount of smooth muscle graduallydecreases from the bronchioles towards the alveoli.Therefore, targeting the larger airways with bronchodila-tors has been the objective until now. However, severalstudies are currently underway to investigate the contri-bution of small airways to asthma and COPD. Recently, ithas been proposed that muscarinic receptors may have amuch greater role in the pathophysiology of obstructiveairway diseases than previously thought [40]. Drugs liketiotropium may potentially inhibit airway inflammation andremodelling [41], whereas for aclidinium an important rolein inhibiting fibroblastemyofibroblast transition has beenreported, which is a key step in peribronchiolar fibrosis

formation [42]. For these reasons, and also to achieve thedesired synergistic effect with ICS, achieving high drugconcentrations in the central and peripheral lung maybecome a future challenge for anticholinergic bronchodi-lators too. However, a balance must be achieved on parti-cle size, because any drug capable of reaching the alveoli ismore likely to be absorbed into the circulation and increasethe potential for adverse systemic effects [29].

Flow rate: dependence or independence?

DPIs fall into two main classes e according to whether theirFPF output is dependent on, or independent of, the user’sIFR [13,16,29,30]. Fig. 4 shows that the percentage of ICSfine particles <5 mm from the (Pulmicort� and Symbicort�)Turbuhaler� and the (Budelin�) Novolizer� inhaler in-creases at higher IFR, while for the (Seretide�) Diskus�,budesonide Cyclohaler� and (Rolenium�) Elpenhaler�, thepercentage is relatively unchanged at different IFRs. It hasto be borne in mind that pharmaceutical companies mayuse different definitions for their label claims, which blursany performance comparison between devices. Somecompanies take the weighed drug amount as reference fortheir label claim whereas others base their label claiminstead on the average drug mass released from themouthpiece.

Whilst an inhaler that produces a consistent percentageof FPF <5 mm independent of the IFR of the patient mayappear at first sight to be the best way to ensure repro-ducible lung deposition, the reality is quite the opposite. AnIFR-dependent inhaler produces an increased FPF <5 mm at

Figure 5 (A) Flow-rate independent inhalers such as Diskus�

provide a decreasing dose to the central and peripheral airwaysas air flow rates (pressure drops) increase, whereas (B) flow-rate dependent inhalers such as Novolizer� maintain theirlung deposited dose as air flow rates (and pressure drops) in-crease [29,30,47].

Figure 6 Scanning electron micrographs of (A) adhesivemixture and (B) soft spherical agglomerates/pellets.

Unravelling misconceptions in DPI use 1199

higher IFR e thus compensating for the shift in depositiontowards larger airways due to increased inertial impactionin the oropharynx and first airway bifurcation[29,30,33,36,43e48], which otherwise reduces depositionin the peripheral respiratory airways of the lung at higherIFRs. The compensation should ensure that a relativelyconstant dose is delivered to where it is needed in the lungregardless of IFR. An example of the difference in lungdeposition between constant and IFR-dependent FPF <3 mmis given in Fig. 5 for a comparison between the Diskus� andNovolizer�. Based on the data from an in vitro depositionstudy and by making use of deposition probability equationsfor particles in the delivered size range, the proportions ofthis FPF that are deposited in the upper airways and centralplus peripheral airways, respectively, can be assessed, as afunction of the inspiratory flow rate [30].

IFR-independent inhalers, such as the Diskus�, producea constant percentage of FPF over a wide range of IFRs,but, at higher IFRs, significantly reduced amounts of theconstant fine particle dose actually reach the peripheralairways. The decrease in computed dose reaching the pe-ripheral airways with this IFR-independent inhaler is from6.7% to 2.2% (Fig. 5A) [30]. Clinically, this means that

patients with higher or variable IFRs using IFR-independentinhalers risk under-dosing, particularly with ICS therapies,which, unlike the fast-acting bronchodilators, do not giveimmediate response feedback to the patient so that theycan tell whether their inhalation was insufficient andrepeat the procedure if necessary [49]. In contrast, withincreasing percentage FPF at higher IFR, the computedperipheral lung dose from the Novolizer� varies from 6.3%to 10.8% (Fig. 5B). These computed values are in goodagreement with the data from a lung deposition study inhealthy volunteers with radiolabeled budesonide aerosolsfrom the Novolizer� yielding peripheral lung depositions of6.5% (at 45 L/min), 7.8% (60 L/min) and 8.5% (90 L/min),respectively [36].

Meeting resistance

Drugs for inhalation by DPI in the treatment of asthma andCOPD therapy are supplied in one of two forms, either as aso-called adhesive mixture or as soft spherical agglomer-ates, referred to as pellet formulation (Fig. 6). In adhesivemixture formulations, the micronised drug particles aredistributed over the surface of coarse carrier particles bynatural adhesion. Pellet formulations may consist of pure

Table 1A Review of some corticosteroid (ICS) delivering DPIs.

Inhaler ICS Manufacturer % FPF at 4 kPa Resistancea FPF as Fu(IFR) Multi-/single-unit doseDefined as % FPF

Flixotide Diskus FLU GSK <5 mm 20e25 Medium/low Constant Multi-unitSeretide Diskus FLU GSK <5 mm 20e25 Medium/low Constant Multi-unitFlixotide Diskhaler FLU GSK <5 mm 25e30 Medium/low Constant Multi-unitRolenium Elpenhaler FLU Elpen <5 mm 15e20 Medium/low Constant Single-unitPulmicort Turbuhaler BUD AstraZeneca <5 mm 30e35 Medium/high Increasing Multi-unitSymbicort Turbuhaler BUD AstraZeneca <5 mm 45e50 Medium/high Increasing Multi-unitBudelin Novolizer BUD Meda <5 mm 40e45 Medium/low Increasing Multi-unitBudesonide Easyhaler BUD Sandoz <5 mm 20e25 Medium/low Increasing Multi-unitBudesonide Clickhaler BUD Merck Generics <5 mm 10e15 Medium/high Slightly increasing Multi-unitBudesonide Cyclohaler BUD Teva Pharma <5 mm 25e30 Low Slightly increasing Single-unitBudesonide Jethaler BUD Ratiopharm <5.1 mm 35e40 Medium/high Increasing Multi-unitFoster NEXThaler BDP Chiesi <5 mm 40e45 Medium/high Constant Multi-unitAsmanex Twisthaler MOM Merck & Co <5 mm 30e35 [51,52] High Slightly increasing Multi-unit

FPF as percent of label claim, unless stated otherwise.a Defined as: high (IFR at 4 kPa < 45 L/min), medium/high (IFR at 4 kPa between 45 and 60 L/min); medium/low (IFR at 4 kPa between

60 and 80 L/min) and low (IFR at 4 kPa > 80 L/min). BUD, budesonide; BDP, beclomethasone (dipropionate); DPI, dry powder inhaler;FLU, fluticasone (propionate); FPF, fine particle fraction; ICS, inhaled corticosteroid, IFR, inspiratory flow rate; MOM, mometasone(furoate).

1200 P. Demoly et al.

drug particles or a mixture of drug particles with amicronised excipient. In both examples, the excipient isusually lactose, which is used to dilute the drug andimprove the flow properties to assist reproducible dosing tothe patient. Adhesive mixtures and pellet formulationsconsist initially of particles that are too large to penetrateto the target area. They have to be de-agglomerated ordispersed during inhalation to release the drug particles inthe appropriate aerodynamic size range.

The energy for de-agglomeration is derived from theinhaled air stream. Well-designed inhalers transfer the ki-netic energy of the airflow effectively into drag, inertial or

Table 1B Review of some bronchodilator (BD) delivering DPIs.

Inhaler BD Manufacturer % FPF at 4 kPa

Defined as % F

Serevent Diskus SAL GSK <5 mm 20eSeretide Diskus SAL GSK <5 mm 20eSalbutamol Cyclohaler SAL Teva <5 mm 25eOxis Turbuhaler FOR AstraZeneca <5.8 mm 35eSymbicort Turbuhaler FOR AstraZeneca <5 mm 40eFormatris Novolizer FOR Meda <5 mm 40eRolenium Elpenhaler FOR Elpen <5 mm 15eFoster NEXThaler FOR Chiesi <5 mm 35eForadil Aerolizer FOR Merck & Co. <5.8 mm 25eOnbrez Breezhaler IND Novartis <5 mm 35eSeebri Breezhaler GB Novartis <5 mm 32eEklira Genuair ACC Almirall <5 mm 35eSpiriva HandiHaler TIO Boehringer I <5 mm 15e

FPF as percent of label claim unless stated otherwise.a Defined as: high (IFR at 4 kPa < 45 L/min), medium high (IFR at 4 k

60 and 80 L/min) and low (IFR at 4 kPa > 80 L/min). ACC, aclidiniuformoterol (fumarate dihydrate); FPF, fine particle fraction; GB, glyco(maleate); SAL, salmeterol (xinafoate); TIO, tiotropium (bromide).

frictional forces, which break down the pellet formulationsinto primary particles or detach drug particles from thecarrier surface [30]. Optimal utilisation of the kinetic en-ergy to generate de-agglomeration forces requires designfeatures that largely determine the inhaler resistance to airflow. Such features can, for instance, be turbulent shearzones (creating drag and lift forces) or whirl and circulationchambers containing impact bodies. Inhalers with highlyeffective de-agglomeration principles are more likely tohave a higher air flow resistance and provide greater lungdeposition than those with a low internal resistance[16,37,50].

Resistancea FPF as Fu(IFR) Multi-/single-unit dosePF

25 Medium/low Slightly increasing Multi-unit25 Medium/low Constant Multi-unit30 Low Increasing Single-unit40 Medium/high Increasing Multi-unit45 Medium/high Increasing Multi-unit45 Medium/low Increasing Multi-unit20 Medium/low Constant Single-unit40 Medium/high Constant Multi-unit30 [53] Low Increasing Single-unit40 Low Constant Single-unit52 [54] Low Constant Single-unit40 Medium/low Slightly increasing Multi-unit20 High Constant Single-unit

Pa between 45 and 60 L/min); medium low (IFR at 4 kPa betweenm (bromide); BD, bronchodilator; DPI, dry powder inhaler; FOR,pyrronium (bromide); IFR, inspiratory flow rate; IND: indacaterol

Unravelling misconceptions in DPI use 1201

Generally, effective utilisation of the available energyfor de-agglomeration results in an increasing FPF withincreasing IFR. The clinical relevance of the internal resis-tance of an inhaler refers not only to a higher proportion ofFPF dose achieved at the same inspiratory effort and acompensating increase in FPF at higher IFR. Resistance alsohas an impact on the site of deposition of drug particles inthe respiratory tract. High-resistance inhalers reduce themaximal attainable flow rate, which equals the quotient ofthe square root of the pressure drop achieved (given inOkPa) and the inhaler resistance (in kPa0.5 min L�1), thusreducing oropharyngeal deposition. Inhaling against a highresistance furthermore opens up the oropharynx and vocalcords, providing a wider passageway for the aerosol andincreasing total lung dose [37].

Most inhalers deliver FPFs between 20% and 30% of thelabel claim at pressure drops between 2 and 4 kPa (Fig. 4A).For inhalers that deliver more or less the same amount ofFPF over the entire range of attainable pressure drops(>2 kPa), higher pressure drops (corresponding with higherflow rates) are not desirable, because this is at the cost oftotal (in particular central and peripheral) lung depositiondue to increased losses in the oropharynx at the higher flowrates. DPIs with IFR-independent FPF generation arefrequently low-resistance devices, which facilitate the gen-eration of higher IFRs. For instance, the increase in flow ratebetween 2 and 4 kPa through a high-resistance inhaler likethe HandiHaler� is only from 28 to 39 L/min versus57e106 L/min for the low-resistance Aerolizer�. In contrast,DPIs delivering a higher FPF with increasing IFR arefrequently high-resistance devices. Fig. 4B compares theflow rates at which FPF <5 mm reaches the value of 25% oflabel claim for the inhalers presented in Fig. 4A. The dif-ference is between 36.4 L/min for the medium/high-resis-tance Symbicort� Turbuhaler� and 82.2 L/min for the low-resistance Cyclohaler�. On the basis of the impactionparameter (based on flow rate) as predictor for inertialimpaction, this might give a 2.25 times higher chance oforopharyngeal impaction for the same particle size for thelow-resistance device. For the Diskus�, Clickhaler� andElpenhaler� FPF values of 25% of label claim were not ob-tained at any flow rate with the devices used to produceFig. 4A. Attaining a pressure drop of 2e4 kPa e which issufficient to operate most inhalers successfully and results inclinically effective drug delivery e is easier when using ahigh-resistance inhaler than when using a low-resistancedevice, independent of the type and severity of the dis-ease [13,31,50]. Even during exacerbations of asthma andCOPD, patients are able to achieve the required pressuredrop and the performance of an inhaler with a higher resis-tance is less influenced by the underlying degree of bron-choconstriction than with a low-resistance inhaler [31]. Also,young children and adult patients with severely impairedlung function generate sufficiently high pressure drops toenable them to use IFR-dependent higher resistance inhalerssuch as the Turbuhaler� and Novolizer� effectively [13,31].

The ideal dry powder inhaler?

Determining the best inhaler is dependent on the circum-stances and characteristics of each patient, so may vary in

different clinical settings. However, based on the consid-erations discussed above and balancing the requirements ofa broad range of patients, the ideal DPI device would havethe following characteristics: a high and consistent IFR-dependent FPF (contributing to consistent central and pe-ripheral lung deposition regardless of IFR) and a medium tohigh internal resistance to limit the IFR and minimise loss ofthe FPF dose in the oropharynx. In addition, the inhalershould be easy for patients to learn and maintain correctusage and small enough to be convenient to carry around orstore, to minimise incorrect use and maximise adherence tothe therapy [16]. Table 1 shows how some of the widelyused inhalers measure up to these requirements. Originatordevices like the Turbuhaler� and Novolizer� deliver a highFPF (1e3 mm) to the central and peripheral airways, wherethe medication being delivered (either bronchodilator oranti-inflammatory ICS) is generally most needed, up to 40%or more of the label claim at the flow rate correspondingwith 6 kPa [29,30,33,43e46]. In contrast, some generic in-halers like the Elpenhaler� deliver more or less the samelow FPF of <20% at all flow rates.

Conclusions

There are a growing number of dry powder inhalers avail-able and these differ greatly in their design, technicalcharacteristics and other individual features. Some inhalershave characteristics that mean they are likely to work for avariety of patients, which may provide a certain level ofconvenience for physicians. However, there are many fac-tors to consider when selecting the optimal inhaler forpatients with asthma or COPD. Apart from the type of druginside the inhaler, the level of clinical evidence for its ef-ficacy and safety, doctor and patient preferences, thetechnical characteristics of the different inhalers and thedelivery and deposition of the fine particle dose to thelungs may be important additional considerations to helpthe physician to select the most appropriate device for theindividual patient to optimise their treatment.

Conflict of interest statements

Pascal Demoly is a consultant and a speaker for Staller-genes, ALK, Circassia and Chiesi and was a speaker forMerck, AstraZeneca, Menarini and GlaxoSmithKline.

Paul Hagedoorn receives royalties from Almirall andMeda (Genuair and Novolizer sales) and is a speaker forAstraZeneca.

Anne de Boer receives royalties from Almirall and Meda(Genuair and Novolizer sales) and is a speaker forAstraZeneca.

Henderik W. Frijlink’s employer has a royalty agreementregarding the Novolizer and the Genuair and he has been aspeaker for AstraZeneca and Meda.

Acknowledgements

Medical writing support for this manuscript was provided byMary Hines and David Candlish, inScience Communications,

1202 P. Demoly et al.

Springer Healthcare. This support was funded byAstraZeneca.

References

[1] Global Initiative for Asthma (GINA). Global strategy forasthma prevention and management [cited 2013 Aug 20];Available from: http://www.ginasthma.org/local/uploads/files/GINA_Report_March13.pdf; 2012.

[2] Dolovich MB, Ahrens RC, Hess DR, Anderson P, Dhand R,Rau JL, et al. Device selection and outcomes of aerosoltherapy: evidence-based guidelines: American College ofChest Physicians/American College of Asthma, Allergy, andImmunology. Chest 2005;127:335e71.

[3] Dolovich MB, Dhand R. Aerosol drug delivery: developmentsin device design and clinical use. Lancet 2011;377:1032e45.

[4] Aerosol Drug Management Improvement Team (ADMIT). TheADMIT webpage [cited 2013 Aug 20]; Available from: http://www.admit-online.info/en/; 2013.

[5] Vincken W, Dekhuijzen PR, Barnes P. The ADMIT series e is-sues in inhalation therapy. 4) How to choose inhaler devicesfor the treatment of COPD. Prim Care Respir J e J Gen PractAirw Group 2010;19:10e20.

[6] Denyer J, Nikander K, Smith NJ. Adaptive aerosol de-livery (AAD) technology. Expert Opin Drug Deliv 2004;1:165e76.

[7] Fischer A, Stegemann J, Scheuch G, Siekmeier R. Novel de-vices for individualized controlled inhalation can optimizeaerosol therapy in efficacy, patient care and power of clin-ical trials. Eur J Med Res 2009;14(Suppl. 4):71e7.

[8] Anderson P. Use of Respimat Soft Mist inhaler in COPD pa-tients. Int J Chron Obstruct Pulmon Dis 2006;1:251e9.

[9] Adcock IM, Caramori G, Kirkham PA. Strategies for improvingthe efficacy and therapeutic ratio of glucocorticoids. CurrOpin Pharmacol 2012;12:246e51.

[10] Chrystyn H, Niederlaender C. The Genuair(R) inhaler: anovel, multidose dry powder inhaler. Int J Clin Pract 2012;66:309e17.

[11] Andrzej E, Malgorzata BE. Easyhaler e ideal dry powderinhaler. Pol Merkur Lekarski e Organ Pol Tow Lek 2010;28:307e10.

[12] Abdelrahim ME. Emitted dose and lung deposition of inhaledterbutaline from Turbuhaler at different conditions. RespirMed 2010;104:682e9.

[13] Selroos O, Borgstrom L, Ingelf J. Performance of Turbuha-ler(R) in patients with acute airway obstruction and COPD,and in children with asthma : understanding the clinicalimportance of adequate peak inspiratory flow, high lungdeposition, and low in vivo dose variability. Treat Respir Med2006;5:305e15.

[14] Hoppentocht M, Hagedoorn P, Frijlink HW, de Boer AH.Technological and practical challenges of dry powder in-halers and formulations. Adv Drug Deliv Rev; 2014. http://dx.doi.org/10.1016/j.addr.2014.04.004.

[15] Azouz W, Chrystyn H. Clarifying the dilemmas about inhala-tion techniques for dry powder inhalers: integrating sciencewith clinical practice. Prim Care Respir J e J Gen Pract AirwGroup 2012;21:208e13.

[16] Chrystyn H, Price D. Not all asthma inhalers are the same:factors to consider when prescribing an inhaler. Prim CareRespir J e J Gen Pract Airw Group 2009;18:243e9.

[17] Lavorini F, Magnan A, Dubus JC, Voshaar T, Corbetta L,Broeders M, et al. Effect of incorrect use of dry powder in-halers on management of patients with asthma and COPD.Respir Med 2008;102:593e604.

[18] Self TH, Wallace JL, George CM, Howard-Thompson A,Schrock SD. Inhalation therapy: help patients avoid thesemistakes. J Fam Pract 2011;60:714e21.

[19] van der Palen J, van der Valk P, Goosens M, Groothuis-Oudshoorn K, Brusse-Keizer M. A randomised cross-over trialinvestigating the ease of use and preference of two drypowder inhalers in patients with asthma or chronic obstruc-tive pulmonary disease. Expert Opin Drug Deliv 2013;10:1171e8.

[20] Hodder R, Reese PR, Slaton T. Asthma patients preferRespimat Soft Mist inhaler to Turbuhaler. Int J Chron ObstructPulmon Dis 2009;4:225e32.

[21] Welch MJ, Nelson HS, Shapiro G, Bensch GW, Sokol WN,Smith JA, et al. Comparison of patient preference and easeof teaching inhaler technique for Pulmicort Turbuhalerversus pressurized metered-dose inhalers. J Aerosol Med eOff J Int Soc Aerosols Med 2004;17:129e39.

[22] Sheth K, Bernstein JA, Lincourt WR, Merchant KK,Edwards LD, Crim CC, et al. Patient perceptions of an inhaledasthma medication administered as an inhalation powder viathe Diskus or as an inhalation aerosol via a metered-doseinhaler. Ann Allergy Asthma Immunol e Off Publ Am CollAllergy Asthma Immunol 2003;91:55e60.

[23] Price D, Lee AJ, Sims EJ, Kemp L, Hillyer EV, Chisholm A,et al. Characteristics of patients preferring once-dailycontroller therapy for asthma and COPD: a retrospectivecohort study. Prim Care Respir J e J Gen Pract Airw Group2013;22:161e8.

[24] Chrystyn H. Do patients show the same level of adherencewith all dry powder inhalers? Int J Clin Pract Suppl; 2005:19e25.

[25] Thomas M, Price D, Chrystyn H, Lloyd A, Williams AE, vonZiegenweidt J. Inhaled corticosteroids for asthma: impact ofpractice level device switching on asthma control. BMC PulmMed 2009;9:1.

[26] Thomas M, Williams AE. Are outcomes the same with all drypowder inhalers? Int J Clin Pract Suppl; 2005:33e5.

[27] Brocklebank D, Ram F, Wright J, Barry P, Cates C, Davies L,et al. Comparison of the effectiveness of inhaler devices inasthma and chronic obstructive airways disease: a systematicreview of the literature. Health Technol Assess 2001;5:1e149.

[28] Hickey AJ, Martonen TB, Yang Y. Theoretical relationship oflung deposition to the fine particle fraction of inhalationaerosols. Pharm Acta Helv 1996;71:185e90.

[29] Frijlink HW, De Boer AH. Dry powder inhalers for pulmonarydrug delivery. Expert Opin Drug Deliv 2004;1:67e86.

[30] de Boer AH, Hagedoorn P, Gjaltema D, Goede J, Frijlink HW.Air classifier technology (ACT) in dry powder inhalation part4. Performance of air classifier technology in the Novolizermulti-dose dry powder inhaler. Int J Pharm 2006;310:81e9.

[31] Borgstrom L. On the use of dry powder inhalers in situationsperceived as constrained. J Aerosol Med e Off J Int SocAerosols Med 2001;14:281e7.

[32] Agertoft L, Pedersen S. Lung deposition and systemic avail-ability of fluticasone Diskus and budesonide Turbuhaler inchildren. Am J Respir Crit Care Med 2003;168:779e82.

[33] Newman SP. Aerosol deposition considerations in inhalationtherapy. Chest 1985;88:152Se60S.

[34] Frijlink HW, de Boer AH. Trends in the technology-drivendevelopment of new inhalation devices. Drug Discov TodayTechnol 2005;2:47e57.

[35] Usmani OS, Biddiscombe MF, Barnes PJ. Regional lungdeposition and bronchodilator response as a function ofbeta2-agonist particle size. Am J Respir Crit Care Med 2005;172:1497e504.

[36] Newman SP, Pitcairn GR, Hirst PH, Bacon RE, O’Keefe E,Reiners M, et al. Scintigraphic comparison of budesonide

Unravelling misconceptions in DPI use 1203

deposition from two dry powder inhalers. Eur Respir J 2000;16:178e83.

[37] Ehtezazi T, Horsfield MA, Barry PW, O’Callaghan C. Dynamicchange of the upper airway during inhalation via aerosoldelivery devices. J Aerosol Med e Off J Int Soc Aerosols Med2004;17:325e34.

[38] Barnes PJ. Distribution of receptor targets in the lung. ProcAm Thorac Soc 2004;1:345e51.

[39] White MV. Muscarinic receptors in human airways. J AllergyClin Immunol 1995;95:1065e8.

[40] Meurs H, Dekkers BG, Maarsingh H, Halayko AJ, Zaagsma J,Gosens R. Muscarinic receptors on airway mesenchymal cells:novel findings for an ancient target. Pulm Pharmacol Ther2013;26:145e55.

[41] Pera T, Zuidhof A, Valadas J, Smit M, Schoemaker RG,Gosens R, et al. Tiotropium inhibits pulmonary inflammationand remodelling in a guinea pig model of COPD. Eur Respir J2011;38:789e96.

[42] Milara J, Serrano A, Peiro T, Gavalda A, Miralpeix M,Morcillo EJ, et al. Aclidinium inhibits human lung fibroblastto myofibroblast transition. Thorax 2012;67:229e37.

[43] Asking L, Maayan C, Petterson G. Flutide Diskus less consis-tent than Pulmicort Turbuhaler with respect to in vitro fineparticle dose proportionality. Am J Respir Crit Care Med2009;163:A441.

[44] Thorsson L, Edsbacker S, Kallen A, Lofdahl CG. Pharmacoki-netics and systemic activity of fluticasone via Diskus andpMDI, and of budesonide via Turbuhaler. Br J Clin Pharmacol2001;52:529e38.

[45] Bisgaard H, Klug B, Sumby BS, Burnell PK. Fine particle massfrom the Diskus inhaler and Turbuhaler inhaler in childrenwith asthma. Eur Respir J 1998;11:1111e5.

[46] Borgstrom L, Asking L, Thorsson L. Idealhalers or realhalers?A comparison of Diskus and Turbuhaler. Int J Clin Pract 2005;59:1488e95.

[47] DeHaan WH, Finlay WH. Predicting extrathoracic depositionfrom dry powder inhalers. J Aerosol Sci 2004;35:309e31.

[48] Kohler D. The Novolizer: overcoming inherent problems ofdry powder inhalers. Respir Med 2004;98(Suppl. A):S17e21.

[49] Laube BL, Janssens HM, de Jongh FH, Devadason SG,Dhand R, Diot P, et al. What the pulmonary specialist shouldknow about the new inhalation therapies. Eur Respir J 2011;37:1308e31.

[50] de Koning JP. Dry powder inhalation: technical and physio-logical aspects, prescribing and use: chapter 2: effect ofresistance to airflow on the inspiratory flow curve. Gronin-gen, The Netherlands: University of Groningen; 2001.

[51] Kondo T, Tanigaki T, Tazaki G, Wanatabe H, Ishii H,Tsunoda T, et al. Areguri 2010;59:950e5.

[52] Yang TT, Li S, Wyka B, Kenyon D. Drug delivery performanceof the mometasone furoate dry powder inhaler. J AerosolMed 2001;14:487e94.

[53] Weuthen T, Roeder S, Brand P, Mullinger B, Scheuch G.In vitro testing of two formoterol dry powder inhalers atdifferent flow rates. J Aerosol Med 2002;15:292e303.

[54] Therapeutic Goods Administration. In: Department of Health& Ageing, editor. Australian public assessment report forglycopyrronium (as bromide); 2013.