The Aerosol Drug Delivery Biology Essay


Aerosol drug delivery: developments in device design and clinical use

Myrna B Dolovich, Rajiv Dhand

Lancet 2011; 377: 1032�45

Published Online November 1, 2010 DOI:10.1016/S0140- 6736(10)60926-9

Firestone Institute of Respiratory Health,

St Joseph�s Healthcare, and Department of Medicine, McMaster University,

Hamilton, ON, Canada

(Prof M B Dolovich PEng); and

Division of Pulmonary, Critical

Care and Environmental

Medicine, Department of

Internal Medicine, University

Aerosolised drugs are prescribed for use in a range of inhaler devices and systems. Delivering drugs by inhalation requires a formulation that can be successfully aerosolised and a delivery system that produces a useful aerosol of the drug; the particles or droplets need to be of su?cient size and mass to be carried to the distal lung or deposited on proximal airways to give rise to a therapeutic e?ect. Patients and caregivers must use and maintain these aerosol drug delivery devices correctly. In recent years, several technical innovations have led to aerosol drug delivery devices with e?cient drug delivery and with novel features that take into account factors such as dose tracking, portability, materials of manufacture, breath actuation, the interface with the patient, combination therapies, and systemic delivery. These changes have improved performance in all four categories of devices: metered dose inhalers, spacers and holding chambers, dry powder inhalers, and nebulisers. Additionally, several therapies usually given by injection are now prescribed as aerosols for use in a range of drug delivery devices. In this Review, we discuss recent developments in the design and clinical use of aerosol devices over the past 10�15 years with an emphasis on the treatment of respiratory disorders.

of Missouri, Columbia, MO, USA (Prof R Dhand MD)

Correspondence to: Prof Myrna B Dolovich, Firestone Institute of Respiratory Health, St Joseph�s Healthcare,

50 Charlton Avenue E, Room JT2135, Hamilton, ON L8N 4A6, Canada


In recent years, increased interest in the scienti?c basis of

aerosol therapy has given rise to a growth in technology

that makes use of the inherent advantages of the inhaled

route of drug administration for the treatment of both

pulmonary and non-pulmonary diseases. A key advantage

of this route is that it enables delivery of low doses of an

aerosolised drug to its site of action for a localised e?ect

(ie, directly to airway surfaces), which leads to a rapid

clinical response with few systemic side-e?ects, particularly for aerosolised ?-agonist therapy.1 Drug

delivery to the systemic circulation via the distal lung results in rapid absorption of the drug from this large surface area. However, when inhaled drugs are

administered for e?ects on the airway (eg, inhaled

corticosteroids), systemic absorption of the drug can give

rise to unwanted side-e?ects.

Aerosol deposition in the lung is a?ected by several

factors, including the aerosol-generating system, particle

size distribution of the inhaled aerosol, inhalation pattern

(eg, ?ow rate, volume, breath-holding time), oral or nasal

inhalation, properties of the inhaled carrier gas (eg, carbon

dioxide, heliox [a gas mixture of helium and oxygen]),

air?ow obstruction, and type and severity of lung disease.

The distribution of target sites and local pharmacokinetics

of the drug also a?ect clinical response. The association

between drug deposition and therapeutic response led to

development of aerosol drug delivery devices that have

pulmonary deposition fractions of 40�50% of the nominal

dose compared with the low levels of 10�15% of the

Search strategy and selection criteria

We identi?ed references for this Review by searches of PubMed with the following search terms: �aerosol drug delivery devices�, �aerosol properties/characterization�, �inhalers (MDIs, spacers, dry powder inhalers)�, �aerosol formulations (pressurized, powder, liquid admixtures)�, �HFA and CFC propellants�, �metered-dose inhalers and dose counters�, �generic inhalers�, �nebulizers (pneumatic, vibrating mesh, micropump)�, �breath-actuated inhalers�, �adaptive aerosol delivery�, �aerosol therapy/inhalation therapy (bronchodilators, corticosteroids, anticholinergics)�, �aerosol therapy/vaccines/gene therapy�, �nanoparticles and inhalation�, �inhalers and nanoformulations�, �aerosol therapy and magnetic particles�, �aerosol therapy and lung deposition�, �aerosol therapy and pediatric respiratory disease�, �aerosol therapy and asthma, chronic obstructive pulmonary disease, cystic ?brosis and other respiratory diseases�, �clinical trials (aerosol delivery and clinical response, dose response)�, �aerosol therapy and mechanical ventilation/arti?cial respiration�, �aerosol therapy and non-invasive ventilation�, �Heliox therapy�, and �aerosol therapy and pulmonary hypertension� from January, 2000, to August, 2009. Papers published between 2004 and 2009 were given priority, but we also included papers from the early published works on aerosols that described major ?ndings that are still pertinent today. Relevant review papers and their references were cited on the basis of their relevance. Only papers published in the English language were reviewed. Both authors are actively involved in original research in aerosol drug delivery and clinical use of therapeutic aerosols and have extensive databases for the material covered in this manuscript.

nominal dose that were achieved in the past.2 Particular inhalation patterns of speci?c disease states could be applied to simulate device performance under certain conditions. This simulation would enable adjustments to be made to the device to not only maximise lung aerosol deposition but also to increase the precision and consistency of aerosol drug delivery.3 Compared with previous devices, the increased e?ciency of the newer aerosol drug delivery devices means that similar e?cacy can be achieved with a lower nominal drug dose.

In clinical practice, pressurised metered-dose inhalers (pMDIs) used with or without a spacer device, dry powder inhalers (DPIs), and nebulisers are used for aerosol delivery. In a 2005 systematic review, the authors concluded that these aerosol drug delivery devices were equally e?cacious provided that they were used appropriately.4 In most, but not all the trials reviewed, the investigators tested single dose strengths of ? agonists in di?erent devices. These doses were often designed to approximate the plateau of the dose-response curve, thereby limiting the ability to di?erentiate between devices. Only a few of these studies compared the bronchodilator responses to a

range of ?-agonist doses. Since publication of that systematic review, several new devices have been marketed for clinical use and new clinical uses for inhaled therapies have emerged. Comparative trials now tend to be designed as cumulative dose-response studies or single doses over a therapeutic range.5

New developments in inhaler technology can take 8�10 years, and recent approaches have focused on incorporating the following features: improvement of aerosol dispersion and production of particles within the extra-?ne size range needed for deep lung targeting; development of methods to reduce e?ort required for inhalation; and improvement of delivery e?ciency while maintaining portability and ease of use of the inhaler.

interchangeable in some countries, whereas the label claim dose can be less than that for the nominal dose and equal to that for the emitted dose in other countries. For example, one of the combination therapies (?utica- sone propionate/salmeterol) with a dose strength of 125 ?g/25 ?g in the UK is equivalent to an emitted dose of 115 ?g/21 ?g in the USA.

The ?ne particle fraction is obtained from in-vitro particle sizing of the aerosol and indicates the percentage of the aerosol mass contained in particles less than 4�7 ?m. The combination of emitted dose and ?ne particle

300 Corticosteroid and delivery system

Flunisolide pMDI*

Triamcinolone acetonide tube spacer

With generic and subsequent market entry products

becoming increasingly available, in-vitro and in-vivo

studies are needed to establish bioequivalence with

trademarked products.6 Some of the regulatory

requirements for generics have changed in recent years,

particularly for DPI generic products. For example, the

appearance of the generic DPI device could be di?erent to

the originally marketed device while necessarily providing

the same dose of drug to the mouth as the original and

also providing aerosol characteristics that are the same.7

Some generic DPIs have di?erent dose strengths and







Label claim

Beclometasone dipropionate pMDI* Fluticasone propionate pMDI*

Budesonide Turbuhaler (AstraZeneca, USA)

Fluticasone propionate Diskhaler (GlaxoSmithKline, USA)

Emitted dose Fine particle dose

di?erent numbers of doses to the original. These products

might have obtained approval as new drug products or as

subsequent market entry products; the availability of the

same drug in di?erent formats can lead to confusion for

clinicians prescribing and patients adhering to a treatment

plan. In this Review we highlight new developments in

aerosol technology and novel therapeutic uses that have

Figure 1: Measured dose values for di?erent inhaled pMDI and DPI corticosteroids

Measured dose values are shown for label claim (or nominal dose), mean emitted dose, and mean ?ne particle dose

for the inhaled pMDI and DPI corticosteroids used in the Dose of Inhaled Corticosteroids with Equisystemic E?ects

(DICE) trial by the National Institutes of Health and Asthma Clinical Research Network. Di?erences in the mass of

drug available from the various inhalers used in this study led to di?erences in clinical response. Data plotted from

Martin and colleagues;13 ?gure adapted from Dolovich.14 pMDI=pressurised metered-dose inhaler. DPI=dry powder

inhaler. *pMDIs used with Optichamber (Philips Healthcare, Andover, MA, USA), a valved holding chamber.

emerged in recent years to help improve awareness

among clinicians.

on the adequate use of an appropriate administered drug

narrowing owing to oedema, increased secretions, or

smooth muscle constriction, the distribution of inhaled

aerosol is non-uniform, with increased concentrations

deposited in areas of airway narrowing.8 The amount of

drug available for distribution distal to the obstructed

areas is possibly reduced, which can a?ect clinical

outcomes.9,10 By comparing responses with the same drug

from di?erent delivery systems11 or between di?erent





200 400

Corticosteroid and delivery system Beclometasone diproprionate: MDI* Budesonide DPI

Flunisolide pMDI*

Fluticasone propionate DPI�

Fluticasone propionate pMDI*

Triamcinolone acetonide pMDI

plus tube spacer

600 800 1000

drugs within the same device category,12 emitted dose or

?ne particle dose provides a more accurate estimate of the

useful dose available from the inhaler than does the label

claim (?gures 1 and 2). Because of losses within the

inhaler and on the mouthpiece,15 drug delivery as recorded

by emitted dose is less than that for the nominal dose or

label claim (?gure 1). De?ning the unit dose depends on

regulatory practices; nominal dose and label claim are

Fine particle dose (?g <4�7 ?m)

Figure 2: FEV1 response as a function of ?ne particle dose provided by six test corticosteroid inhalers

The FEV1 response (as a percentage of the morning measurement) is shown for the corticosteroids used in the Dose

of Inhaled Corticosteroids with Equisystemic E?ects (DICE) trial by the National Institutes of Health and Asthma

Clinical Research Network.13 The FEV response is plotted against increasing ?ne particle dose�the portion of the

inhaled dose likely to deposit in the lungs and give rise to a response. Reproduced from Parameswaran and

colleagues,12 with permission from Pulsus Group. FEV =forced expiratory volume in 1 s. MDI=metered-dose inhaler.

DPI=dry powder inhaler. *pMDIs used with Optichamber (Philips Healthcare, Andover, MA, USA). �Used with

Optichamber (Philips Healthcare, Andover, MA, USA), a valved holding chamber.



Gas phase

Patient presses

Propellant droplets form at actuator nozzle: �2-phase gas-liquid air-blast�

Metering valve

Expansion chamber

Liquid phase (formulation)

Retaining cup


Metering chamber

High-velocity spray Actuator nozzle

can, which opens

channel between

metering chamber

and atmosphere

Propellants start to boil in expansion chamber

Shearing forces produce ligaments

Initial velocity 30 m/s

Initial droplet size 20�30 ?m

Evaporation and cooling

Figure 3: Key chloro?uorocarbon MDI components and mechanisms of aerosol formation

(A) Key components of a pMDI. (B) When the device is actuated, the drug and propellant mixture exits the metering chamber under pressure; the process by which it

forms an aerosol is shown. Reproduced from Newman and colleagues,23 with permission from the American Association for Respiratory Care.

fraction results in the ?ne particle mass; ?ne particle dose is ?ne particle fraction multiplied by emitted dose (?gure 1), which can be associated with e?cacy.12,13,16,17 Dose metrics obtained in vitro are a useful guide for comparing device performance, assessing the likelihood of depositing drug proximally or distally in the lung, and helping to explain clinical responses. However, in addition to airway diseases, other factors such as mouth-throat geometry and inhalation ?ow pro?les add to the variability in the deposited airway doses in vivo and therefore a?ect the therapeutic response.18

Pressurised metered-dose inhalers

pMDIs are portable, convenient, multi-dose devices that

use a propellant under pressure to generate a metered

dose of an aerosol through an atomisation nozzle.19

Worldwide, pMDIs are the most widely used inhalation

devices for the treatment of asthma and chronic obstructive

pulmonary disease. Chloro?uorocarbon-propelled pMDIs

were routinely prescribed for several decades, but in

accordance with the Montreal Protocol of 1987,20

chloro?uorocarbon propellants are being replaced by

hydro?uoroalkane propellants that do not have ozone-

depleting properties.21,22 Hydro?uoroalkanes are non-toxic,

non-?ammable, and chemically stable and they are not

carcinogenic or mutagenic. No safety concerns have been

identi?ed with their use in healthy individuals or patients

with asthma. Although hydro?uoroalkane-134a and hydro-

?uoroalkane-227 do not a?ect the atmospheric ozone, they

do marginally contribute to global warming.21

The key components of chloro?uorocarbon pMDIs (ie,

canister, metering valve, actuator, and propellant) are

retained in hydro?uoroalkane pMDIs (?gure 3), but they

have had a redesign. Two approaches were used in the

reformulation of hydro?uoroalkane pMDIs. The ?rst

approach was to show equivalence with the chloro-

?uorocarbon device, which helped regulatory approval,

and was the approach used for salbutamol pMDIs

and some corticosteroid pMDIs. With the Modulite

platform (Chiesi Farmaceutici, Parma, Italy), some

hydro?uoroalkane formulations were matched to their chloro?uorocarbon counterparts on a microgram for microgram basis; therefore, no dosage modi?cation was needed when switching from a chloro?uorocarbon to a hydro?uoroalkane formulation.24 The second approach involved extensive changes, particularly for corticosteroid inhalers containing beclometasone dipropionate, and resulted in solution aerosols with extra-?ne particle size distributions and high lung deposition.25,26 The exact dose equivalence of extra-?ne hydro?uoroalkane beclometasone dipropionate and chloro?uorocarbon beclometasone dipropionate has not been established, but data from most trials have indicated a 2:1 dose ratio in favour of the hydro?uoroalkane pMDI.27 Half the dose of hydro?uoroalkane beclometasone dipropionate Auto- haler (Graceway Pharmaceuticals, Bristol, TN, USA) was as e?ective as twice the dose of budesonide given by Turbuhaler DPI (AstraZeneca, Lund, Sweden).28,29 However, dose equivalence of hydro?uoroalkane beclo- metasone dipropionate Autohaler was noted in com- parison with chloro?uorocarbon ?uticasone propionate.30 Clinicians need to be aware that the Modulite platform also o?ers an extra-?ne formulation of beclometasone dipropionate (as the Fostair inhaler with formoterol fumarate, Chiesi Farmaceutici).

The clinical implications of di?erences in the design and formulation of the new hydro?uoroalkane pMDIs are shown in table 1. The drug output and aerosol characteristics of salbutamol pMDIs are similar to salbutamol chloro?uorocarbon pMDIs, as are broncho- dilator responses and protection against methacholine- induced35 or exercise-induced bronchoconstriction36 in both adults and children with asthma.37

Patients with asthma on regular long-term treatment with a salbutamol chloro?uorocarbon pMDI could safely transition to regular treatment with a hydro?uoroalkane pMDI without any deterioration in pulmonary function, loss of asthma control, increased frequency of hospital admissions, or other adverse e?ects.22 Patients readily accept the use of hydro?uoroalkane pMDIs.38 Salmeterol

CFC pMDI Changes with HFA pMDI Clinical implication

Propellant26 CFCs HFAs HFA-134a is safe, non-toxic, and non-carcinogenic; it is rapidly metabolised and does not accumulate in tissues; it has no ozone- depleting potential and has less greenhouse e?ects than CFCs

Aerosol plume26 High velocity

Cold temperature

Spray emitted as a jet

Reduced velocity Warmer

Rounder cloud con?guration

Decreased oropharyngeal deposition Reduced chances of �cold freon� e?ect Di?erence in feel and taste

Particle size26 Mass median aerodynamic diameter of 3�8 ?m Suspension pMDIs similar to CFCs

Solution pMDIs have lower mass

median aerodynamic diameter

No major change

Lower oropharyngeal deposition, enhanced deposition in the lung,

especially in peripheral lung

Metering chamber25 Volume 50�100 ?L Smaller chamber Less chance of leakage during storage

Less chances of loss of prime (ie, the ?rst actuation after storage contains

a reduced drug dose)

Formulation22 Creaming of suspension

Variable pu?-to-pu? dosing

Tail-o? e?ect*

No ethanol content31

Suspension or solution with ethanol Improved pu? to pu? dosing

Only a few additional doses provided after speci?ed number of doses on label claim

Ethanol used as solvent or co-solvent

No need to shake the aerosol before use for solution pMDIs More consistent clinical e?cacy

Less chance of misuse because spray content decreases substantially when additional actuations are used beyond the speci?ed number of doses on the label claim

Blood ethanol concentrations might lead to failed breath-analyser test within 3 min of inhaling two doses

Priming22 Needs priming before initial use if not used for 4 days Variable priming requirements Check priming instructions according to brand

Actuator ori?ce Ori?ce diameter 0�14�0�6 mm Smaller sized aperture

Finer aerosol particle size

Greater chances of clogging with potential to change aerosol characteristics; recommended to wash actuator once weekly or if spray force decreases

Reduced oropharyngeal deposition; in combination with reduced spray velocity enhances e?ciency of drug deposition in the lung

Dose counter32,33 No dose counter Dose counter on some devices Less chance of underdosing or overdosing as patients can count the number of doses used and establish when canister is nearly empty

Moisture a?nity Moisture leaks into canister Increased moisture a?nity Some HFA pMDIs (eg, Ventolin, GlaxoSmithKline, Ware, UK) have lower

shelf-life after being removed from water-resistant packaging pouch

Temperature dependence

Operates best in warm temperature Less temperature dependence Less chance of losing e?cacy in cold weather

Substantial reduction in dose below 10�C

Cost34 Generic inhalers inexpensive Higher cost of trademarked pMDIs Could change cost-bene?t of using pMDIs

Patients might forego treatment or choose cheaper and less e?ective


CFC=chloro?uorocarbon. pMDI=pressurised metered-dose inhalers. HFA=hydro?uoroalkane. *Variability in quantity of drug in actuations past the number of doses in the canister as speci?ed by the label claim, resulting in less uniform drug doses.

Table 1: pMDIs: problems with CFC-propelled pMDIs, changes made with HFA-propelled pMDIs, and clinical implications of modi?cation

hydro?uoroalkane pMDIs (Serevent, GlaxoSmithKline, Ware, UK) and hydro?uoroalkane combinations of long-acting ? agonists and corticosteroids (Advair, GlaxoSmithKline, Ware, UK; Symbicort, AstraZeneca, Lund, Sweden) have similar e?cacies as the chloro- ?uorocarbon formulations.39 Coordinated e?orts by device manufacturers, pharmaceutical companies, regulatory agencies, and health-care providers have resulted in minimum disruption in the transition from chloro- ?uorocarbon to hydro?uoroalkane pMDIs.

Breath-actuated MDIs

Problems in precisely coordinating device actuation

with inhalation lead to poor drug delivery, sub-optimum

asthma control, and increased inhaler use. Breath-

actuated pMDIs, such as the Maxair Autohaler

(Graceway Pharmaceuticals, Bristol, TN, USA) and

Easibreathe (IVAX, Miami, FL, USA), were developed to

overcome the problem of poor coordination between

pMDI actuation and inhalation. The devices consistently

actuate early in inspiration at an inspiratory ?ow rate of

about 30 L/min and are uniformly well accepted by patients,40 with fewer than 5% of patients unable to achieve the threshold inspiratory ?ow rate required for actuation.

Patients who used the Maxair Autohaler achieved higher pulmonary deposition (21%) than did patients who had poor coordination while using a conventional chloro?uorocarbon pMDI (7%), but the clinical e?ects for both groups were similar.41 Some investigators reported improved outcomes with breath-actuated pMDIs,29 but changes in formulations, particle size, and ?ne particle dose could account for the di?erences reported. Increased use of breath-actuated inhalers might improve asthma control42 and reduce overall cost of asthma therapy compared with conventional pMDIs. However, oropharyngeal deposition with breath-actuated pMDIs is as high as that with chloro?uorocarbon pMDIs. As breath-actuated devices cannot be used with valved holding chambers, the oropharyngeal side-e?ects from corticosteroids could be a problem for some patients. Moreover, gastrointestinal absorption of some

inhaled corticosteroids, such as beclometasone dipropionate, could lead to an increased frequency of systemic side-e?ects.

Other pMDI technologies that provide more precise targeting of the respiratory tract include the Vortex Nozzle Actuator (Kos Pharmaceuticals, Morrisville, NC, USA), Synchro-Breathe (Vortran Medical Technology, Sacramento, CA, USA), and Tempo Inhaler (MAP Pharmaceuticals, Mountain View, CA, USA).

Dose counters

Dose counters provide a reliable method for patients to

monitor their use of drugs. As the over?ll is typically

10%, pMDIs can continue to function after the labelled

number of doses has been given, but the amount of

drug in each spray can be inconsistent, especially for

chloro?uorocarbon products. Mechanical dose counters

are accurate and reliable,32 whereas add-on dose

counters, such as the Doser device (MediTrack Products,

Hudson, MA, USA) might lose accuracy over time.43 The

MD Turbo (Teamm Pharmaceuticals, Morrisville, NC,

USA), or other electronic devices, are not widely used in

clinical practice.43

Spacers and holding chambers

Spacer devices are categorised as add-on devices, extension

devices, or holding chambers and they improve e?cacy

by providing more reliable delivery of pMDI drugs to

patients who have di?culty in coordinating inhalation

with pMDI actuation.

Spacer devices have three basic designs�the open

tube, the reservoir or holding chamber, and the reverse-

?ow design, in which the pMDI, placed close to the

mouth, is ?red in the direction away from the patient.

Adding a one-way valve creates a holding chamber,

enabling retention of aerosol within the chamber for a

?nite time after pMDI actuation. Holding chambers

produce a ?ne aerosol because of the high level of

impaction of larger drug particles and partial

evaporation of propellant within the chamber.44 As

substantial di?erences exist between these three

categories of spacer design, the most appropriate spacer

for the patient�s age and ability to self-treat should be

carefully considered.

Device-related factors contribute to variability in drug

delivery.45 For example, larger-volume spacers and holding

chambers capture and retain more of the aerosol cloud,

whereas smaller-volume spacers and holding chambers

reduce the amount of available aerosol generated from the

impaction of the formulation on their walls. The

characteristics of various spacers and e?ects on delivery,

lung deposition, and clinical e?cacy of inhaled drugs are

well described elsewhere.21

Electrostatic charge

Drug deposits can build up on walls of plastic spacers

and holding chambers, mostly because of electrostatic

charge. Aerosols remain suspended for longer periods within holding chambers that are manufactured from non-electrostatic materials than other materials (?gure 4A). Thus, an inhalation might be delayed for 2�5 s without a substantial loss of drug to the walls of metal or non-conducting spacers.46,47 The electrostatic charge in plastic spacers can be substantially reduced by washing the spacer in mild detergent followed by a water rinse to prevent inhalation of dried detergent particles.

In children with asthma, salbutamol delivered through plastic spacers has a similar e?cacy to that delivered through non-electrostatic or metal spacers.48 In patients with chronic obstructive pulmonary disease, tiotropium delivered from a pMDI through a non-static spacer provided a similar clinical bene?t to that given by the trademarked DPI.49 An increased ?ne particle dose available from antistatic spacers could lead to an increased number of systemic adverse e?ects with long-term inhaled corticosteroids use. For example, more adrenal suppression was reported after the hydro?uoroalkane ?uticasone propionate was delivered through two antistatic plastic spacers and one metal spacer than that reported with the pMDI alone.50

Facemask interface

A valved holding chamber ?tted with an appropriate

facemask is used to give pMDI drugs to neonates,

young children, and elderly patients.51 The two key

factors for optimum aerosol delivery are a tight but

comfortable facemask ?t and reduced facemask dead

space.52�54 Because children have low tidal volumes

and inspiratory ?ow rates, comfortable breathing

through a facemask requires low resistance inspiratory

or expiratory valves.

Inhalation technique

All young children should be given a holding chamber-

type spacer with their pMDI, otherwise inhalation of

pMDI aerosols is likely to be ine?cient in more than 50%

of patients.55 Tidal breathing from a holding chamber and

facemask should be encouraged in patients who are

unable to use pMDIs appropriately. In preschool children

who were less than 5 or 6 years of age, two to six tidal

breaths seem to be su?cient to inhale the aerosol. In

infants and young children, the tidal volume (based on

the child�s weight if not possible to measure directly) to

spacer volume ratio should be taken into account when

selecting a spacer device.56

Dry powder inhalers

Several new, innovative DPIs are available for the

treatment of asthma and chronic obstructive pulmonary

disease57 (?gure 4B) and for delivery of a range of other

drugs such as proteins, peptides, and vaccines.58 The

challenge is to combine suitable powder formulations

with DPI designs that generate small particle aerosols.59,60


ACE spacer AeroChamber Plus Flow-Vu Vortex LiteAire



Aerolizer Turbuhaler HandiHaler Diskus



MicroAir NE-U22 Aeroneb GO eFlow I-neb Respimat

Figure 4: Examples of marketed spacers and holding chambers, dry-powder inhalers available by prescription or in development, and nebulisers that incorporate new-generation technology

(A) The ACE spacer (Smiths Medical, Rockland, MA, USA), the EZ-Spacer (FSC Laboratories, Charlotte, NC, USA), and the Inspirease spacer (not shown) are examples of reverse-?ow designs;

AeroChamber Plus Flow-Vu (Trudell Medical International, London, ON, Canada), Vortex (PARI Respiratory Equipment, Midlothian, VA, USA), and Nebuchamber (AstraZeneca, Lund, Sweden; not

shown) are examples of metal or non-conducting valved holding chambers. The LiteAire (Thayer Medical, Tucson, AZ, USA) is a collapsible, disposable, valved paper spacer. (B) The Aerolizer (Schering

Plough, Kenilworth, NJ, USA) and Handihaler (Boehringer-Ingelhein, Ingelheim, Germany) dry-powder inhalers are capsule devices; the Turbuhaler (AstraZeneca, Lund, Sweden) is a reservoir

dry-powder inhaler; the Diskus (GlaxoSmithKline, Ware, UK) is a multi-unit dose dry-powder inhaler with single doses of drug encapsulated in foil blisters; the Manta single-dose dry-powder inhaler

(Manta Devices, Boston, MA, USA) is a disposable, low-cost inhaler that uses a foil blister for drug storage with a unique internal opening technology. (C) The MicroAir NE-U22 (Omron, Vernon Hills, IL,

USA), Aeroneb GO (Aerogen, Galway, Ireland), eFlow (PARI, Midlothian, VA, USA ), and I-neb (Respironics, Murrysville, PA, USA) incorporate vibrating mesh or vibrating plate aerosol generators. I-neb

and Prodose (Pro?le Therapeutics, Bognor Regis, UK; not shown) use adaptive aerosol delivery technology for drug delivery. The Respimat inhaler (Boehringer-Ingelheim, Ingelheim, Germany) is the

?rst of a new class of hand-held inhalers called soft mist inhalers. Both the Respimat and the AERx (Aradigm, Hayward, CA, USA; not shown) are high e?ciency devices that use precise dosimetric

systems. The Respimat inhaler has a multi-dose capability.

Use of DPIs is expected to increase with the phasing out of chloro?uorocarbon production along with increased availability of drug powders and development of novel powder devices.22,57

Powder storage

DPI doses can be pre-metered in the form of single

capsules or foil blisters or as multi-single unit dose

disks; alternatively, device metering of bulk powder can

be done with reservoir devices. As drug delivered from a

DPI mainly depends on the ability of the patient to

generate a su?cient pressure drop across the device on

inhalation, inconsistent e?orts by the patient could

result in substantial variability between doses. With a capsule-based DPI, the patient can take a second inhalation if powder clearly remains in the capsule after the initial breath.

Form and function

Breath actuation is a major advantage of DPIs over

pMDIs. However, exhalation into a DPI could result in

the loss of the dose positioned in the inhalation channel.

For reservoir DPIs, the powder remaining in the

reservoir can, over time, be a?ected by added humidity

in the exhaled breath. DPIs that rely on the inspiratory

e?ort of the patient to dispense a dose (passive or

patient-driven devices) ensure delivery on inhalation, but a su?cient inspiratory ?ow rate is needed to aerosolise the drug powder. Other DPI designs (active or power-assisted designs) incorporate battery-driven impellers and vibrating piezoelectric crystals that reduce the need for the patient to generate a high inspiratory ?ow rate, an advantage for many patients. In power- assisted DPI designs, the powder is released from storage by external means, such as directing compressed air through the DPI, and is then held in a storage or valved holding chamber. Enhanced sedimentation of drug particles in the chamber reduces the dose of drug released and decreases the particle size of the powder dispensed.

Resistance and performance

Drug delivery to the lung ranges between 10% and 37%

of the emitted dose for several marketed DPIs.61 Recent

improvements in DPI design enable the dose to be

dispensed independent of inspiratory ?ow rate between

30 L/min and 90 L/min. DPIs with medium resistance

to air?ow are designed to operate at an optimum rate of

60 L/min, but even this ?ow rate might be di?cult to

achieve for some patients, especially elderly patients

with severe chronic obstructive pulmonary disease.62

Although ?ow independence is advantageous for

consistent drug delivery from a DPI, this independence

could be a disadvantage when adult doses are given to

children. The risk of overmedicating children with

these DPIs could be partly o?set by the low inspiratory

volumes of children. Dose titration should be done to

avoid overdosing.

The physical design of the inhaler establishes its

speci?c resistance to air?ow (measured as the square

root of the pressure drop across the device divided by

the ?ow rate through the device), with current designs

having speci?c resistance values ranging from about

0�02�0�2 (cmH2O� �/(L/min). With high-resistance

devices, breathing at the optimum inspiratory ?ow

rate for the particular DPI selected helps to produce a

?ne powder aerosol with increased delivery to the

lung. Children younger than 6 years cannot

consistently inhale from a DPI with the proper

inspiratory ?ow rate and pMDIs with valved holding

chambers are preferable.63 Children older than 6 years

can successfully use a DPI even during acute

asthma exacerbations.64

Other factors for device use

Because of variations in the design and performance of

DPIs, patients might not use all DPIs equally well.

Therefore, DPIs that dispense the same drug might not

be readily interchangeable.65 Dose counters in new-

generation DPIs provide patients with either a

numerical display of the number of doses remaining or

a colour indicator as a reminder to renew their

prescription in time.


Nebulisers are devices that convert a liquid in solution or

suspension into small droplets.

Pneumatic or jet nebulisers

Jet nebulisers use compressed gas ?ow to break up the

liquid into a ?ne mist�the protruding surfaces of

primary and/or secondary ba?es within the nebuliser

are positioned in the path of the aerosol created so that

the large liquid droplets impinge upon them, leading to

a reduced and more useful particle size of the exiting

aerosol.66 Substantial variances in nebuliser performance

are caused by di?erences in their design, the source of

energy (compressed gas or electrical compressor), gas

?ow and pressure, connecting tubing, interface used

(spacer, and mouthpiece or mask), and the breathing

pattern of the patient.

Unlike pMDIs and DPIs, no special inhalation

techniques are needed for optimum delivery with

nebulisers. However, conventional nebulisers, which

need compressed gas or a compressor to operate, are

generally not portable; they have poor delivery e?ciency

and treatment times are much longer than that for

pMDIs and DPIs.

Substantial aerosol wastage with continuously operated

jet nebulisers could be reduced by attaching a T-piece

and corrugated tubing or a reservoir bag to collect aerosol

generated during exhalation (Circulaire, Westmed,

Tucson, AZ, USA)�drug aerosol is then inhaled from

the reservoir with the next inspiratory breath.67 Breath-

enhanced and dosimetric nebulisers reduce drug loss

during exhalation by incorporating design features

such as one-way valves.68 These features have been

used for delivery of pentamidine, with ?lters placed

in the expiratory tubing to prevent environmental

contamination with pentamidine after exhalation.

Ultrasonic nebulisers

In these devices, sound waves generated by vibrating a

piezoelectric crystal at high frequency (>1 MHz) are

transmitted to the surface of the drug solution, resulting

in the formation of standing waves. The crests of these

waves are then broken up into droplets. The precise

mechanism of aerosol generation by ultrasonic nebulisers

is not yet fully understood.69 Older models of ultrasonic

nebulisers are costly and bulky and have a tendency to

malfunction. Moreover, compared with newer ultrasonic

designs, their relative ine?ciency in nebulising drug

suspensions, liposomes, or more viscous solutions are

major limitations to their use.

E?ect of formulation

The presence of a preservative in a drug solution and

admixture with other drugs a?ect nebuliser output and

aerosol characteristics.70,71 Drug mixtures need to be

physically and chemically compatible.72,73 Since July, 2007,

the US Centers for Medicare and Medicaid Services

stopped reimbursement for pharmacy-compounded nebuliser drugs.

Delivery by mouthpiece versus facemask

Aerosol deposition in the nasal passages substantially

reduces pulmonary drug delivery and bronchodilator

e?cacy;74 however, facemasks might be necessary for

the treatment of acutely dyspnoeic or uncooperative

patients. For optimum e?cacy, the facemask should

produce a tight seal75,76 to avoid aerosol leakage and

aerosol deposition around the eyes. The orientation of

the nebuliser with regard to the facemask a?ects the

pattern of aerosol deposition. Although �front-loaded�

masks (ie, in which the nebuliser is inserted directly

into the facemask in front of the mouth) provide more

aerosolised drug, they also produce greater facial and

ocular deposition than do �bottom-loaded� masks

(ie, in which the aerosol enters the mask from below

the mouth).77 Aerosol deposition on the face and eyes

could be reduced by use of a prototype mask that

incorporates vents in the mask and has cut-outs in the

eye region.3,77

Continuous aerosol delivery

In patients with acute severe asthma, short-acting

bronchodilators (eg, salbutamol 5�15 mg/h) are

commonly given continuously78 with large-volume

nebulisers or the high-output extended aerosol respiratory

therapy nebuliser, which can provide consistent drug

output for 4 h70 to 8 h,79 respectively. Patients with acute

asthma have some bene?ts from continuous broncho-

dilator therapy in the emergency department.80

Nebuliser and compressor combinations

Nebuliser performance for use at home depends on the

Vernon Hills, IL, USA), eFlow (PARI, Midlothian, VA, USA), and I-neb (Respironics, Murrysville, PA, USA)� and these are compared with conventional jet and ultrasonic nebulisers in table 2. The aerosol characteristics depend on the physicochemical properties of the solution.87,88 Vibrating mesh or vibrating plate nebulisers have a higher lung deposition,85 negligible residual volumes, a faster rate of nebulisation than do jet nebulisers, and they e?ectively nebulise solutions and suspensions, as well as liposomal formulations,89 proteins, such as ?-1 antiprotease90 and dornase alfa.91 Denaturation of non-complexed, supercoiled DNA occurs during nebulisation, which is similar to jet nebulisers.69 In patients with cystic ?brosis, vibrating mesh nebulisers e?ciently deliver tobramycin,92 and escalating doses of aztreonam lysinate.93 The residual volume varies with the design of the eFlow device by PARI. One design of the eFlow device has a low residual volume to minimise drug wastage, whereas another design has a larger residual volume, which is comparable to that in jet nebulisers. Although the e?ciency of drug delivery in the latter design is comparable to breath- enhanced jet nebulisers, treatment times are shorter with the eFlow.

The cost of these vibrating mesh and vibrating plate devices is comparable to that of ultrasonic nebulisers, but is much higher than that of conventional jet nebulisers. All vibrating mesh and vibrating plate nebulisers must be cleaned regularly to prevent build-up of deposit and blockage of the apertures, especially when suspensions are aerosolised.

Jet Ultrasonic Vibrating mesh Features

choice of an appropriate compressor,81 and some nebuliser

manufacturers specify the compatible compressors for

Power source Compressed gas or

electrical mains

Electrical mains Batteries or electrical


optimum performance (eg, PARI LC Plus Reusable Nebuliser and DeVilbiss Pulmo-Aide compressor [Somerset, PA, USA] for inhalation of tobramycin).

Multi-dose liquid inhalers

The Respimat inhaler (Boehringer-Ingelheim, Ingelheim,

Germany)82 is a novel aerosol drug delivery device that

uses the energy from a compressed spring to force a

metered dose of the liquid drug formulation through

a narrow nozzle system created using microchip

technology. The aerosol produced has a high ?ne particle

fraction and a high e?ciency of pulmonary drug delivery,

up to 50% for some formulations.83 This inhaler is

available for clinical use in Europe but has not yet been

approved in North America.

Vibrating mesh or aperture plate nebulisers

Figure 4C shows the characteristics of nebulisers that

use a vibrating mesh or plate with several apertures84�86�

Aeroneb (Aerogen, Galway, Ireland), MicroAir (Omron,

Portability Restricted Restricted Portable

Treatment time Long Intermediate Short

Output rate Low Higher Highest

Residual volume 0�8�2�0 mL Variable but low ?0�2 mL

Environmental contamination

Formulation characteristics

Temperature Decreases* Increases� Minimum change

Concentration Increases Variable Minimum change

Suspensions Low e?ciency Poor e?ciency Variable e?ciency

Denaturation Possible� Probable� Possible�

Cleaning Required, after single use Required, after multiple use Required, after single use

Cost Very low High High

*For jet nebulisers, the temperature of the reservoir ?uid decreases about 15�C during nebulisation because of evaporation. �For ultrasonic neubulisers, vibration of the reservoir ?uid causes a temperature increase during aerosol generation, which can be as high as 10�15�C. �Denaturation of DNA occurs with all the nebulisers.

Table 2: Comparison of di?erent nebulisers

Aerosol droplet containing magnetic nanoparticles

nebuliser systems include the AKITA system (Activaero, Gemuenden, Germany),95 the Small Particle Aerosol Generator (ICN Pharmaceuticals, Costa Mesa, CA, USA),96 and humidi?ed high-?ow nasal cannulae.97

Targeting aerosol delivery in the lung

The ability to target drugs to speci?c sites of disease is a

major unmet need of aerosol therapy.



Passive targeting

The �passive targeting� approach directs deposition

mainly to the airways or preferentially to the more

peripheral airways and alveolar compartment by

modi?cation of aerosol droplet size,2 breathing pattern,

depth and duration of holding a breath, timing of the

aerosol bolus in relation to inspiratory air?ow, drug-

aerosol dosage, and density of the inhaled gas.2,4 Similarly,

a substantial fraction of the inhaled aerosol can be

deposited at areas of airway narrowing during

exhalation, especially when ?ow-limited segments are

present.3 Airway targeting can also reduce oropharyngeal

drug deposition, thereby reducing the risk of

local98 and systemic99 side-e?ects resulting from the

swallowed dose.

Figure 5: Mechanism of action of nanomagnetosols

Magnetic nanoparticles are mixed with the drug solution but the drug is not actually bound to the particles, thus

magnetic nanoparticles do not have to be formulated speci?cally for each drug. Because each aerosol droplet

contains many magnetic particles, they appear as a large magnetic particle to an external magnetic ?eld. The

increased size of the aerosol droplets improves the ability to guide the nanoparticles to the desired region(s) of the

lung by use of a strong external magnetic ?eld. Reproduced from Plank,102 with permission from Elsevier.

Adaptive aerosol delivery

These devices use software-driven monitoring and control

systems that monitor inspiratory ?ow, breathing

frequency, and inspiratory time, providing aerosol delivery

only during inspiration. The I-neb and Prodose system

(Pro?le Therapeutics, Bognor Regis, UK) use an adaptive

aerosol delivery disc�a plastic disc containing a microchip

and antenna�to control drug delivery.94 The I-neb is a

vibrating mesh nebuliser, whereas the Prodose is powered

by a compressor. In addition to delivering a precise drug

dose, other useful features of the I-neb are the provision

of feedback to the patient on dose completion along with

details of each treatment. These data can be transmitted

via a modem to a remote location, which enables

continuing assessment of adherence of the patient to the

drug regimen. The Pulmonary Drug Delivery System

Clinical (Nektar Therapeutics, San Carlos, CA, USA),

another breath-synchronised, high-e?ciency vibrating

plate nebuliser, can be used both during mechanical

ventilation and spontaneous breathing. Other novel

Active targeting

The �active targeting� approach localises drug deposition

by directing the aerosol to the diseased area of lung or,

alternatively, by using molecular or biological recognition,

providing a more controlled and reproducible delivery to

predetermined targets in the lung than by passive

targeting. For example, the AeroProbe intracorporeal

nebulising catheter (TMI, London, ON, Canada) could be

inserted into the working channel of a ?bre-optic

bronchoscope to deliver genes100 or chemotherapeutic

drugs101 directly to a lung lobe.

Recently, inert superparamagnetic iron oxide

nanoparticles added to the nebuliser solution were used

to guide aerosol to the a?ected region of the lung by

means of a strong external magnetic ?eld (?gure 5).103 A

range of therapeutic agents, including genes, could be

packaged for delivery by this technique.102


Heliox (a gas mixture of 80% helium and 20% oxygen),

which has one-third the density of air, results in more

peripheral deposition of inhaled aerosol particles than

does air, especially in the presence of airway constriction.

In children with airway obstruction, the rate of aerosol

deposition is enhanced while breathing heliox compared

with breathing oxygen.104

When heliox, rather than air or comparable mixtures of

oxygen and air, is the driving gas in a ventilator circuit,

aerosolised drug delivered from a pMDI is increased.105

By contrast, drug output from a nebuliser decreases

when it is operated with heliox instead of air.106 To ensure

adequate nebuliser output with heliox, the ?ow of heliox

has to be increased from the conventional 6�8 L/min to 15 L/min.105 Similar changes occur when vibrating mesh nebulisers use heliox rather than air.107

Aerosol delivery during mechanical ventilation Drug delivery to patients on mechanical ventilation is complicated by the presence of an arti?cial airway. The major factors that a?ect the e?ciency of drug delivery during mechanical ventilation include: the position of the patient, the aerosol generator and its con?guration in the ventilator circuit, aerosol particle size, synchronisation of aerosol generation with inspiratory air?ow from the ventilator, conditions in the ventilator circuit, and ventilatory measurements. Dhand and Guntur108 provide further discussion on the methods to optimise aerosol therapy in this setting and the use of inhaled therapies in adult, paediatric, and neonatal patients. Nebulisers and pMDIs, but not DPIs, are routinely used for broncho- dilator therapy in mechanically ventilated patients. With optimum techniques of administration, the e?ciency of aerosol drug delivery achieved with these devices is comparable to that in ambulatory, non-intubated patients. Similarly, as with ambulatory patients with chronic obstructive pulmonary disease, combination therapy with

short-acting ? agonists and anticholinergic drugs produces

additive bronchodilation in ventilator-supported patients.

Aerosol delivery in patients receiving noninvasive

positive pressure ventilation is less e?cient than that in

patients receiving invasive mechanical ventilation.109

Non-conventional therapeutic uses


Flumist (MedImmune, Gaithersburg, MD, USA), a live

attenuated in?uenza vaccine given by nasal spray,110 and

other inhaled spray-dried formulations containing whole

inactivated virus or split subunit vaccine, could be used

for in?uenza prevention.111 In the early 1990s, about

4 million children were immunised against measles with

the Classical Mexican Device�a home-built system that

incorporated a jet nebuliser from IPI Medical Products

(Chicago, IL, USA). Aerosolised vaccine against measles

provided a stronger and more durable boosting response

than did vaccination by injection in school-age children112

and is now being tested by WHO in mass immunisation

campaigns.113 Similarly, an inhaled measles and rubella

vaccine,114 a triple vaccine (measles, mumps, and

rubella),115 a dry powder formulation of live attenuated

measles vaccine,116 and inhaled vaccines for protection

against inhaled bioterrorism agents such as anthrax and

tularaemia are under development.117,118

Inhaled prostanoids

Epoprostenol (an intravenous prostacyclin) improves

survival in patients with pulmonary arterial hypertension,

but, compared with intravenous administration,

aerosolised prostacyclins have a higher selectivity for

intrapulmonary e?ects with few systemic e?ects.119

Iloprost, a stable analogue of prostacyclin, has a longer half-life than prostacyclin (20�30 min vs ~3 min), producing pulmonary vasodilation for 30�90 min. Six to nine inhalations of iloprost daily improved exercise capacity, functional capacity, and pulmonary haemo- dynamics in patients with pulmonary arterial hypertension with few side-e?ects.120 The combination of inhaled iloprost with oral sildena?l121 or oral bosentan122 further enhanced and prolonged the pulmonary vasodilator e?ects. Inhaled treprostinil, another prostacyclin analogue, had a more prolonged pulmonary vasodilator e?ect than did inhaled iloprost.123

Inhaled ciclosporin

Aerosolised ciclosporin prevents or delays post-lung

transplant rejection and improves survival compared

with an immunosuppressive regimen without aerosolised


Gene therapy

Aerosolised gene therapy could be used to correct speci?c

genetic abnormalities in patients with cystic ?brosis and

?-1 antitrypsin de?ciency125 and possibly for the treatment

of lung cancer126 and other non-genetic diseases, such as

pulmonary hypertension and acute lung injury.

Nebulisation of liquid-suspended gene particles,

although ine?cient, remains the mainstay for inhaled

gene therapy. Because of its viscosity, the concentration of

DNA that can be readily nebulised is less than 5 mg/mL.

Fragmentation owing to shear stresses, preferential

nebulisation of solute, and adhesion of DNA to plastic

surfaces results in less than 10% of the DNA in the

nebuliser cup being emitted from the nebuliser.69,127

Device selection

The appropriateness of a device for a patient in a given

clinical situation depends on several factors. The

following questions should be asked before making a

selection. In what devices is the drug being prescribed

available and how do these di?erent devices compare in

terms of ease of use, performance, clinical e?cacy, and

safety? Is the device likely to be available for several

years? Do the published works support the advertised in-

vitro performance information of reliable and

reproducible aerosolised drug delivery and its clinical

e?cacy with a minimum or no side-e?ect pro?le? Is the

device patient-friendly with regard to operation and

maintenance? Is the device clinically useful on a broad

scale (ie, can it be used to treat di?erent patient

populations in various clinical settings and patients in

di?erent age-groups)? Is the device cost e?ective in terms

of purchase price, price to maintain, and cost to train

caregivers in use and to teach patients? Is the device

reusable and can it be used with many drugs? And is

reimbursement available for the device?

Correct use of aerosol drug delivery devices is

important for successful therapy. Patients, physicians,

and other healthcare workers must be adequately instructed in the proper use of aerosol devices prescribed.128 Additionally, adherence to the therapeutic regimen must be emphasised to the patient or caregiver.129 Reviewing the patient�s inhaler technique on subsequent o?ce or clinic visits is important for good disease management and to maintain adherence of the patient to therapy.16 If the selected delivery device does not provide satisfactory treatment or results in unacceptable side-e?ects, other equally e?ective options are available.4


In the past 10�15 years, several innovative developments

have advanced the ?eld of inhaler design. There are

many choices in all device categories that incorporate

features providing e?cient aerosol delivery to treat

various lung and systemic diseases. Attempts to improve

topical delivery to selective areas of the lung or new

approaches to access the distal lung for systemic therapy

are continually being investigated and they have the

potential to provide more advanced aerosol drug delivery

technologies than those currently available.


Both authors contributed equally to the preparation of this paper.

Con?icts of interest

Within the past 5 years, MBD has received research grants from P?zer,

GlaxoSmithKline, and AstraZeneca, and has received consultancy fees

from Cogentus Pharmaceuticals, Novartis Imaging, AstraZeneca, and

Boehringer Ingelheim. MBD has consulted for Medicines in Need but did

not receive any fees. Within the past 5 years, RD has received speaker fees

from GlaxoSmithKline, Boehringer-Ingelheim, P?zer, and Bayer,

consultancy fees from Novartis and Bayer, and research support from

Sepracor and Trudell Medical International. RD was a principal site

investigator for a clinical trial sponsored by Novartis.