Thorax 1991;46:712-716

Improvement of drug delivery with a breathactuated pressurised aerosol for patients with poor

inhaler technique

Stephen P Newman, Alex W B Weisz, Neda Talaee, StewartW Clarke

Department ofThoracic Medicine,Royal Free Hospitaland School ofMedicine, LondonNW3 2QGS P NewmanA W B WeiszN TalaeeS W ClarkeReprint requests to:Dr S P Newman,Pharmaceutical Profiles Ltd,2 Faraday Buildings,Highfields Science Park,University Boulevard,Nottingham NG7 2QP.Accepted 7 July 1991

AbstractBackground The metered dose inhaler isdifficult to use correctly, synchronisingactuation with inhalation being the mostimportant problem. A breath actuatedpressurised inhaler, designed to helppatients with poor inhaler technique,was compared with a conventionalmetered dose inhaler in terms of aerosoldeposition and bronchodilator response.Methods Radioaerosol deposition andbronchodilator response to 100 pg sal-butamol were measured in 18 asthmaticpatients, who inhaled from a conven-tional metered dose inhaler by their ownchosen metered dose inhaler technique,from a conventional metered doseinhaler by a taught metered dose inhalertechnique, and from a breath actuatedpressured inhaler (Autohaler).Results In the 10 patients who couldcoordinate actuation and inhalation ofthe inhaler on their own deposition ofaerosol in the lungs and bronchodilatorresponse were equivalent on the threestudy days. By contrast, in the eightpatients who could not coordinate themean (SEM) percentage of the dosedeposited in the lungs with their owninhaler technique (7-2% (3-4%)) was sub-stantial lower than those attained by thetaught metered dose inhaler technique(22-8% (2-5%)) and by Autohaler (20-8%(1P7%)).Conclusion Although of little additionalbenefit to asthmatic patients with goodcoordination, the Autohaler is poten-tially a valuable aid to those with poorcoordination, and should be consideredin preference to a conventional metereddose inhaler in any patient whose inhalertechnique is not known to be satisfac-tory.

The pressurised metered dose inhaler wasintroduced in 19561 and has become the mostpopular inhalation device in the treatment andprophylaxis of asthma. It is compact, port-able, and convenient, containing at least 200metered doses; these features explain its wide-spread acceptance by patients and its con-tinued use after several decades. It suffers froma major disadvantage, however, in that it isdifficult to use correctly." It has beenestimated that as many as half of adults with

asthma and an even greater proportion ofchildren derive little benefit from theirmetered dose inhalers because of inefficientinhaler technique.6 Further, some patientswhose inhaler technique is initially satisfac-tory develop a poor technique with time.7

Probably the most important error con-founding the use of a metered dose inhaler isfailure to coordinate or synchronise the actua-tion of the inhaler with inhalation.3 A breathactuated pressurised inhaler (Aerolin Auto-haler, 3M Riker) has been devised to helpsuch patients.89 This device (fig 1) uses aconventional pressurised canister, which istriggered by a spring when the patient inhalesthrough the mouthpiece at a low inhalationflow rate of 22-36 1/min.8 We have comparedaerosol deposition and bronchodilator res-ponse following inhalation from the Autohalerand from a conventional metered dose inhalerin patients with good and bad coordination.

MethodsPATIENTSWe studied 19 patients (10 male) with asthmaaged 18-78 years (baseline forced expiratoryvolume in one second (FEVy) 24-96% predic-ted). One patient was subsequently withdrawnbecause he failed repeatedly to keep his

,_ _LEVER

SPRING

CANISTER

TRIGGERING__,MECHANISM-

MOUTHPIECE

Figure I Breath actuated pressurised inhaler (AerolinAutohaler, 3M Riker).

712

on Novem

ber 22, 2020 by guest. Protected by copyright.

http://thorax.bmj.com

/T

horax: first published as 10.1136/thx.46.10.712 on 1 October 1991. D

ownloaded from

Improvement of drug delivery with a breath actuated pressurised aerosolfor patients with poor inhaler technique

appointments. Immediately before the studall patients were shown to have at least 15°'reversibility in FEV, and in maximaexpiratory flow at 75% forced vital capacit(Vmax25) after inhaling 200 jig salbutamofrom a pressurised metered dose inhalerPatients withheld treatment with an inhale(beta agonist for 12 hours before each hospitavisit, and inhaled ipratropium bromide wawithdrawn for 24 hours and methylxanthinefor 48 hours. Patients were issued with a shoracting beta2 agonist (rimeterol) that they coulcuse if required up to three hours before eacitest. Each patient gave informed consent irwriting and the studies were approved bothlocally by the hospital's ethical practice sub-committee and by the Administration olRadioactive Substances Advisory Committee.STUDY DESIGNEach patient performed three radioaerosolstudies on three different days separated by atleast 48 hours. On the first study day (calledthe own metered dose inhaler day) a conven-tional radiolabelled metered dose inhaler wasused by the patient's own chosen technique.The patient was asked to take a single metereddose using the inhalation technique that he orshe normally used. No instructions on use ofthe inhaler were given; the mouthpiece coverwas removed and the inhaler was shaken andthen handed to the patient in the uprightposition. The second and third study dayswere randomised, and consisted of inhalationfrom a conventional metered dose inhaler by ataught technique (the taught metered doseinhaler day) and from an Autohaler. On thesetwo days patients practised the inhalationmanoeuvre with a placebo aerosol until theirperformance was judged to be competent.Patients were instructed to take a slow, deepinhalation, followed by 10 seconds' breathholding and exhalation via a filter. With thetaught technique the metered dose inhaler wasactuated by an observer during the inhalationmanoeuvre. After each inhalation patientsexhaled via a low resistance filter (Inspiron002290) to trap any exhaled radioaerosol.They wore a respiratory inductance plethys-mography band (Respitrace Corporation)around the chest, which enabled inhaledvolume, inhaled flow rate, and breath holdingtime to be registered.'0 We performed thestudy with patients' own technique beforethose with the taught technique or Autohalerso that their own technique would not beinfluenced by recent instruction.

RADIOAEROSOL STUDIESPressurised metered dose inhalers werelabelled by the direct addition of the radio-nuclide technetium-99m to canisters contain-ing salbutamol (Salbulin, 3M Riker) asdescribed." Briefly, `mTc is extracted inchloroform, which subsequently evaporates inan empty aerosol canister. Two millilitres of asuspension comprising propellants (chloro-fluorocarbons 11, 12, and 114), surfactant (sor-bitan trioleate), and salbutamol were added atbelow - 60°C, and a 25 pl metering valve was

y added by a crimper. Subsequently, eacho metered dose delivered about 100,ig sal-d butamol and 20 MBq 99"Tc. The techniquey was assessed by extracting metered doses via a)I multistage liquid impinger," 12 which classifiesr. aerosol into different size bands; aerosol pene-d trating to stages 3 and 4 of the impingerl consists mainly of droplets smaller than 5 jums diameter. The distributions of untreateds drug, drug treated with the chloroform andt 9'Tc mixture, and radiolabel on differentI parts of the impinger were compared. Theseh data (table 1) showed that the distribution ofi radiolabel among the different parts of thei impinger closely mirrored that of drug, and

that the distribution of drug within the impin-f ger was changed little by the addition of theradiolabel in chloroform. We concluded thatthe radiolabel would act as a marker for thepresence of drug during a deposition study.After inhalation of a single labelled dose of100 jg salbutamol a posterioanterior view ofthe chest and lateral view of the oropharynxwere taken immediately by a gamma camera(Ohio Nuclear 110) coupled to a Nodecrestcomputer. Counts were corrected for attenu-ation during the passage of radiation to thegamma camera. Deposition on the actuatorand exhaled air filter was compared withthat from a metered dose collected on afurther filter; activity not recovered on theactuator or on the exhaled air filter wasassumed to be in the body and was fractionedaccording to the corrected counts from lungsand oropharynx. Activity recorded in theoesophagus and stomach was assumed to havearisen from aerosol deposited in the oro-pharynx. A posteroanterior ventilation scanusing krypton-81m was also performed oneach patient, and was used to define the lungborders on the aerosol scans; the lung fieldswere subsequently divided into central, inter-mediate, and peripheral lung zones."

LUNG FUNCTIONFEVy (Vitalograph Spirometer) and Vmax25(Ohio Spirometer plus XY plotter) weremeasured immediately before inhalation of thelabelled dose and 15, 30, 60, and 90 minuteslater. The highest value from three attemptswas taken at each time point. Studies with thetaught metered dose inhaler technique andwith the Autohaler were started only ifbaseline FEVy was within 15% of its value onthe "own metered dose inhaler" day.

Table I Evaluation of radioaerosol technique: mean(SEM) percentage distribution of untreated drug, drugtreated with chloroform and technetium-99m, andradiolabel on the actuator, "throat," andfour stages of amiultistage liquid impinger

Untreated Treateddrug drug Radiolabel

Actuator 8 6 (3 0) 12 6 (2-8) 13 9 (2 6)Throat 38.0 (4.5) 43 5 (8-7) 41 5 (3 7)Stage 1 7 7 (1-7) 4 0 (1 6) 4 0 (2-7)Stage 2 6-7 (1 8) 4-2 (2 0) 6 4 (0 5)Stage3 86(1 8) 65(41) 65(22)Stage4 30 5 (1 6) 29 2 (2 1) 27 7 (0-8)

713

on Novem

ber 22, 2020 by guest. Protected by copyright.

http://thorax.bmj.com

/T

horax: first published as 10.1136/thx.46.10.712 on 1 October 1991. D

ownloaded from

Newman, Weisz, Talaee, Clarke

Figure 2 Depositionpatterns of radiolabelledaerosol in a goodcoordinator (top) and in abad coordinator (bottom)for studies with their ownmetered dose inhaler(MDI) technique, thetaught MDI technique,and the Autohaler. Whenthe bad coordinatorinhaled with his own MDItechnique the inhaler wasactuated after ;ompletionof inhalation.

OWN MDI TAUGHT MDI W'I OHAL. F. PF

t.'W4

O)WN MD' Ji..QH-T MD!

1

*.;;s .r.

',...ti:

'.K._

;'r

o; Mi-

.,\SylX { .' ; 0. .l ,=|1/. r;

:

':

.. 'ii*,leS

_;

STATISTICAL ANALYSIS

Analysis of variance was used to assess theeffects on deposition and bronchodilator res-

ponse due to inhaler, period, and carryover

from the previous inhaler. Subsequent mul-tiple pairwise comparisons between inhalerswere carried out using Bonferroni t tests.'3 Ap value below 0-05 was taken to indicatestatistical significance. From inspection of theinhalation manoeuvre when the patientinhaled according to his or her own technique,.patients were divided into two groups, withgood and bad coordination respectively; datafrom the two groups were tested separately.

ResultsGOOD COORDINATORSIn the study using their own metered doseinhaler technique, 10 of the 18 patients (meanbaseline FEV, 51% predicted, range 24-72%)actuated the inhaler during the course ofinhalation, and were hence classified as goodcoordinators. Most of these patients used aninhalation technique that was close to optimal;six of the 10 patients inhaled at less than 60 1/min, and seven of the 10 patients held theirbreath for more than six seconds.

For the ten good coordinators there were nosignificant differences between the three study

714

.il ., I"

li,.1,.:,,a p. -I.-Aftw'..-7gli

on Novem

ber 22, 2020 by guest. Protected by copyright.

http://thorax.bmj.com

/T

horax: first published as 10.1136/thx.46.10.712 on 1 October 1991. D

ownloaded from

715Improvement of drug delivery with a breath actuated pressurised aerosolfor patients with poor inhaler technique

Figure 3 Meanpercentage changes inFEV, (with standarderrors) and in maximalexpiratoryflow at 75%forced vital capacity( Pmax25) in 10 goodcoordinators (upperfigures) and in eight badcoordinators (lowerfigures),for studies withtheir own metered doseinhaler (MDI) technique(*), the taught MDItechnique (* ), and theAutohaler (0).

30

20

10

% Change from baseline FEV,: good coordinators % Change from baseline VmaX25: good coordinators

b /t- - I~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~I

15 30 60 90 Min

% Change from baseline FEV,: bad coordinators % Change from baseline Vmax26: bad coordinators

90Min 90 Min

days in the deposition fractions in lungs, oro-

pharyx, actuator, and exhaled air (table 2).Typical deposition patterns are shown in figure2. There was a trend towards less deposition inthe lungs for inhalation by the taught metereddose inhaler technique, but this did not reachstatistical significance. Regional depositionwithin the lungs was also similar for the threestudy days: the mean (SE) percentage of thedose in the peripheral lung zone was 6 5 (1 -0)for "own metered dose inhaler," 5-4 (1 0) for"taught metered dose inhaler," and 7-1 (1-2)for the Autohaler.

Baseline values of FEV, (own metered doseinhaler 1-92 (029) 1, taught metered doseinhaler 1-87 (0-29) 1, Autohaler 1-97 (030) 1)and Vmax25 were similar on the three studydays. Areas under the FEVy and Vmax25 bron-chodilator response curves did not vary sig-nificantly between the three studies (fig 3).

BAD COORDINATORSEight patients (mean baseline FEV1 57%predicted, range 31-96%) were observed tohave bad coordination with their own metereddose inhaler technique; of these, four actuatedthe inhaler before the beginning of inhalationand four actuated it after the completion ofinhalation.The fractionation of deposition for these

eight patients is shown in table 2; comparedwith the other two studies, there was a sig-nificant reduction in deposition in the lungs(p < 001) during inhalation by their ownmetered dose inhaler technique. The percen-

tage of the dose deposited in the lungs withtheir own metered dose inhaler technique was

less than 1%, 11-8%, 8-6%, and 28-3% in thefour patients who actuated the metered doseinhaler before inhalation; these patients haddelay times before starting to inhale of about 3,1-5, 1 0, and 0 5 seconds, suggesting an inverse

relation between lung deposition and delaytime. In three of the four patients who actuatedthe metered dose inhaler after inhaling, lessthan 2% of the dose reached the lungs (fig 2); inthe fourth patient lung deposition was 5% ofthe dose, but this was confined to a single area inthe large bronchi ofone lung. Deposition in theperipheral lung zone was 8-6% (0 9%) of thedose for "taught metered dose inhaler" and8-4% (11-%) for the Autohaler, but we did notattempt to quantify this measure for "ownmetered dose inhaler" as it was immeasurablysmall in five of the eight patients.There was little variation in either baseline

FEVy (own metered dose inhaler 1-36 (0 15)1,taught metered dose inhaler 1-39 (0-16) 1,Autohaler 1-41 (0 15) 1) or baseline Vmax25 onthe three study days. The area under the FEVybronchodilator response curve was significantly(p < 005) greater with taught metered doseinhaler and Autohaler than with "own metereddose inhaler" (fig 3). The area under the Vmax25bronchodilator response curve was significantly(p < 005) greater for the Autohaler than for"own metered dose inhaler."

Table 2 Mean (SEM) percentage of the aerosol doselocated at various sites after inhalation by goodcoordinators (n = 10) and by bad coordinators (n = 8)

Site Own MDI Taught MDI Autohaler

GOOD COORDINATORSLungs 18 6 (2-9) 12 8 (1-8) 17 5 (2-8)Oropharynx 64.4 (3 8) 71 1 (2-5) 61 2 (4-5)Actuator 16 1(20) 15 8(1 2) 210(2-8)Exhaled 0 7 (0-4) 0 3 (0-1) 0-3 (0-1)BAD COORDINATORSLungs 7 2 (3-4) 22 8 (2 5)* 20 8 (1-7)*Oropharynx 67-7 (4-7) 59-3 (2 3) 60 7 (2-2)Actuator 23-5 (3 8) 17 6 (0 9) 18 2 (1-3)Exhaled 1-7 (0-9) 0 3 (0-1) 0-2 (0-1)

*p < 0-01 in the comparison with own metered doseinhaler (MDI) technique.

on Novem

ber 22, 2020 by guest. Protected by copyright.

http://thorax.bmj.com

/T

horax: first published as 10.1136/thx.46.10.712 on 1 October 1991. D

ownloaded from

Newman, Weisz; Talaee, Clarke

DiscussionA breath actuated pressurised aerosol was firstdescribed almost 20 years ago,'4 but the earlymodel was relatively unsuccessful for severalreasons. The inhaler was rather bulky, itsoperation was somewhat noisy and violent, andsome patients had difficulty in generating asufficient flow rate to trigger the mechanism.Additionally, the original breath actuatedinhaler contained the non-selective beta agon-ist isoprenaline. The new breath actuatedinhaler (Aerolin Autohaler) is compact andquiet in operation, functions at a low inhaledflow rate that is easily achieved by patients withobstructive airways disease'5 and is optimal fordrug delivery,'1'8 and contains the selectivebeta2 receptor agonist salbutamol.Although not necessarily reflecting the

population at large, the asthmatic patientsentering our study fell into two approximatelyequal groups with good and bad coordinationwhen using a metered dose inhaler according totheir own chosen technique. In the 10 goodcoordinators the Autohaler gave no additionalbenefit over a conventional metered doseinhaler. Surprisingly, however, there was atrend towards less lung deposition with thetaught technique, which resulted primarilyfrom data from two individuals. Overall, thisdifference was not significant, but we cannotexclude the possibility that while instructingpatients in the "correct" technique we wereintroducing some additional and undetectederror that caused this technique to be inferior totheir own chosen method. The use of a metereddose inhaler with the mouth held open resultsin greater deposition than when the mouth isclosed,'9 but this cannot explain the observedresult because every patient used a metereddose inhaler with the lips firmly closed roundthe mouthpiece.The data were substantially different in the

bad coordinators. In four patients who chose toactuate the metered dose inhaler after inhalingboth lung deposition and bronchodilator re-sponse were substantially reduced, and wefailed to observe a characteristic depositionpattern of aerosol in the lungs on the gammacamera scans. This arose because there was noinhaled airstream available to conduct aerosoldeep into the lungs. The observation of agreatly reduced bronchodilator response whenthe dyscoordination occurs at the end of inhala-tion is in agreement with the results of earlierstudies. 620 The data from the four patients whodelayed between actuating the metered doseinhaler and starting to inhale suggests that drugdelivery and delay time are inversely related,and that a minimal delay of less than one secondmay not compromise drug delivery; this againis in agreement with the findings of earlierstudies.'6 20

In our patients the Autohaler thus "correc-ted" for poor coordination, both drug deposi-tion and the bronchodilator response beingequivalent to those obtained with a correctlyused conventional metered dose inhaler.Subjectively, we were impressed by the easewith which several patients with a very poormetered dose inhaler technique learnt to usethe Autohaler successfully. It would be interes-ting to knowhow effectively such patients retaintheir ability to use the Autohaler, but this was

outside the scope of the present study.Evidence from a clinical study2' suggests thatpatients prefer the Autohaler to a conventionalmetered dose inhaler and that they find theformer easier to use. The Autohaler will not,however, help patients who stop inhaling at themoment of actuation,22 and the device must beformulated with "ozone friendly" propellants2'in the near future. The Autohaler should,however, be a valuable alternative to dry pow-der inhalers'2 24 and spacer devices25 26 forpatients unable to use a conventional pressur-ised metered dose inhaler because of coordina-tion difficulties.

We wish to thank 3M Health Care for their financial support ofthis study and Mr J Tunnicliffe of Sheffield Statistical Servicesfor performing the statistical analysis.

1 Freedman T. Medihaler therapy for bronchial asthma: a newtype of aerosol therapy. Postgrad Med 1956;20:667-73.

2 Epstein SW, Manning CPR, Ashley MJ, Corey PN. Surveyofthe clinical use ofpressurised aerosol inhalers. Can MedAss J 1979;120:813-6.

3 Crompton GK. Problems patients have using pressurisedaerosol inhalers. Eur J Respir Dis 1982;63(suppl 119):57-65.

4 Pedersen S, Frost L, Arnfred T. Errors in inhalationtechnique and efficiency in inhaler use in asthmaticchildren. Allergy 1986;41:118-24.

5 Lindgren S, Bake B, Larsson S. Clinical consequences ofinadequate inhalation technique in asthma therapy. Eur JRespir Dis 1987;70:93-8.

6 Crompton GK. Adult patients' difficulties with inhalers. In:Inspiration: developments in inhalation therapy. Oxford:Medicine Publishing Foundation, 1989:1-6. (Symposiumseries No 26.)

7 Paterson IC, Crompton GK. Use of pressurised aerosols byasthmatic patients. BMJ 1976;i:76-7.

8 Baum EA, Bryant AM. The development and laboratorytesting of a novel breath-actuated pressurised inhaler.J Aerosol Med 1988;1:219-20.

9 Baum EA. Design, development and testing of a new breathactuated inhaler. In: Inspiration: developments in inhala-tion therapy. Oxford: Medicine Publishing Foundation,1989:21-30.

10 Stradling JR, Chadwick GA, Quirk C, Phillips T.Respiratory inductance plethysmography: calibrationtechniques, their validation and effects ofposture. BullEurPhysiopathol Respir 1985;21:317-24.

11 Newman SP, Clark AR, Talaee N, Clarke SW. Pressurisedaerosol deposition in the human lung with and without an"open" spacer. Thorax 1989;44:706-10.

12 Newman SP, Moren F, Trofast E, Talaee N, Clarke SW.Deposition and clinical efficacy of terbutaline sulphatefrom Turbuhaler, a new multi-dose powder inhaler. EurRespir J 1989;2:247-52.

13 Glantz SA. Primer of biostatistics. New York: McGraw Hill,1987:276-7.

14 Crompton GK. Breath-activated aerosol. BMJ 1971;ii:652-3.

15 Wallace WAH, Lenny J, Cooksey E, Greening AP, Cromp-ton GK. Ability of patients with severe airflow limitationto trigger a new breath-actuated inhaler. Thorax 1989;44:318P.

16 Newman SP,-Ravia D, Clarke SW. How should a pressurisedbeta-adrenergic bronchodilator be inhaled? Eur JRespirDis 1981;62:3-20.

17 Newman SP, Pavia D, Garland N, Clarke SW. Effects ofvarious inhalation modes on the deposition of radioactivepressurised aerosols. Eur J Respir Dis 1982;63(suppl1 19):57-65.

18 Lawford P, McKenzie D. Pressurised aerosol inhalertechnique: how important are inhalation from residualvolume, inspiratory flow rate and time interval betweenpuffs? Br J Dis Chest 1983;77:276-81.

19 Dolovich MB, Ruffin RE, Roberts R, Newhouse MT.Optimal delivery of aerosols from metered dose inhalers.Chest 1981;80(suppl):911-5.

20 RivlinJ, MindorffC, Levison H, KazimF, Reilly P, WorsleyG. Effect of administration technique on bronchodilatorresponse to fenoterol in a metered dose inhaler. JPediatrics 1983;102:470-2.

21 Crompton GK, Duncan J. Clinical assessment of a newbreath-actuated inhaler. Practitioner 1989;233:268-9.

22 Pedersen S, Mortensen S. Use ofdifferent inhalation devicesin children. Lung 1990;168(suppl):653-7.

23 Newman SP. Metered dose pressurised aerosols and theozone layer. Eur Respir J 1990;3:495-7.

24 Crompton GK. Clinical use of dry powder systems. Eur JRespir Dis 1982;63(suppl 122):96-8.

25 Konig P. Spacer devices used with metered dose inhalers.Breakthrough or gimmick? Chest 1985;88:276-84.

26 Dolovich MB, RuffinRE, Corr D, Newhouse MT. Clinicalevaluation of a simple demand-inhalation MDI aerosoldelivery device. Chest 1983;84:36-41.

716

on Novem

ber 22, 2020 by guest. Protected by copyright.

http://thorax.bmj.com

/T

horax: first published as 10.1136/thx.46.10.712 on 1 October 1991. D

ownloaded from