studies on antibiotic aerosols for inhalation in cystic

University of Groningen

Studies on antibiotic aerosols for inhalation in cystic fibrosisWesterman, Elsbeth M.

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:2009

Link to publication in University of Groningen/UMCG research database

Citation for published version (APA):Westerman, E. M. (2009). Studies on antibiotic aerosols for inhalation in cystic fibrosis. [s.n.].

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).

The publication may also be distributed here under the terms of Article 25fa of the Dutch Copyright Act, indicated by the “Taverne” license.More information can be found on the University of Groningen website: https://www.rug.nl/library/open-access/self-archiving-pure/taverne-amendment.

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: 28-04-2022

https://research.rug.nl/en/publications/23b0c38a-ef66-494d-b424-1d1ca986405d

Studies on antibiotic aerosols for inhalation in cystic fibrosis

Elsbeth M. Westerman

E.M. Westerman


Thesis University of Groningen, Groningen, The Netherlands

The work in this thesis was (financially) supported by:Nederlandse Cystic Fibrosis Stichting (NCFS), Baarn, The NetherlandsApotheek Haagse Ziekenhuizen, The Hague, The NetherlandsHaga Teaching Hospital, The Hague, The NetherlandsUniversity of Groningen, Groningen, The Netherlands

© E.M. Westerman, Rotterdam

Printed by: Optima Grafische Communicatie, Rotterdam

Paranimfen:

Mevrouw dr. Ch. van Kesteren

Mevrouw dr. K.D. Schuijt

RIJKSUNIVERSITEIT GRONINGEN


Proefschrift

ter verkrijging van het doctoraat in deWiskunde en Natuurwetenschappen

aan de Rijksuniversiteit Groningenop gezag van de

Rector Magnificus, dr. F. Zwarts,in het openbaar te verdedigen op

maandag 25 mei 2009om 16.15 uur

door

Elisabeth Mechteld Westermangeboren op 3 oktober 1972

te Rotterdam

Promotor: Prof. dr. H.W. Frijlink

Copromotor: Dr. H.G.M. Heijerman

Beoordelingscommissie: Prof. dr. J.R.B.J. Brouwers

Prof. dr. E.J. Duiverman

Prof. dr. C.K. van der Ent

ISBN: 978-90-367-3795-1

978-90-367-3796-8

Aan mijn ouders

Contents

Chapter 1 General introduction 9

Chapter 2 Inhaled medication and inhalation devices for lung disease in patients

with cystic fibrosis: a European Consensus

21

Chapter 3 Aerosolisation of tobramycin (TOBI®) with the PARI LC PLUS® reusable

nebuliser: which compressor to use? -comparison of the CR60® to the

PortaNeb® compressor-

65

Chapter 4 Effect of nebulised colistin sulfate and colistin sulfomethate on lung

function in patients with cystic fibrosis: a pilot study

83

Chapter 5 Design and in vitro performance testing of multiple air classifier

technology in a new disposable inhaler concept (Twincer®) for high

powder doses

93

Chapter 6 Dry powder inhalation of colistin sulfomethate in healthy volunteers: a

pilot study

111

Chapter 7 Dry powder inhalation of colistin in cystic fibrosis patients: a single

dose pilot study

123

Chapter 8 Dry powder inhalation of colistimethate sodium in cystic fibrosis

patients using the Twincer® inhaler: pulmonary deposition after

adapted conditions

139

Chapter 9 Concluding remarks and future perspectives 151

Summary 171

Samenvatting 177

Curriculum vitae 183

List of publications 185

Dankwoord 187

1 General introduction

General introduction 11

Cystic fibrosis: introductionCystic fibrosis (CF) is one of the most common life-shortening, chronic hereditary diseases among

the Caucasian population. Median survival used to lie in childhood or adolescency decades ago,

but has improved considerably to advanced adulthood nowadays. From the 1940’s onward, the

treatment of CF patients has gradually progressed from attempts to cure CF-related symptoms

to organized, multidisciplinary care including early intervention and prevention of CF related

complications. The correction of malabsorption due to exocrine pancreatic insufficiency with

(enteric coated) pancreatic enzyme supplements has been one of the major break-throughs, as

has been the prevention or correction of nutritional failure with calorie-dense oral supplements

or enteral feedings. Relief of airway obstruction, either by physiotherapy and/or inhaled drug

treatment, and suppression of inflammation with oral drugs are two other treatment pillars.

Early and intensive therapy of airway infections with either oral or aerosolized antibiotic drug

treatment is considered to be another major reason for the ongoing rise in life expectancy.

Although drug treatment with aerosolized antibiotics has several therapeutic benefits over

oral treatment, administration of these drugs with currently available nebuliser equipment

takes up a significant amount of time on a daily basis, which negatively influences the quality

of life and adherence to therapy. It is agreed upon that improvements are needed with respect

to both knowledge on optimal aerosolized treatment of pulmonary infections and simplifying

the administration of the drug aerosol. This is the motive for the work described in this thesis.

CF related complications are caused by a disturbed chloride transport in the body. Chloride

channels are missing or malfunctioning which results in failure of the cell to transport chloride

into the extracellular space. As a result, chloride and sodium ions accumulate within cells,

inhibiting water transport across cell membranes and causing dehydration of the mucus that

normally coats these surfaces, thereby affecting the respiratory and digestive systems.

As a result of increased viscosity of the mucus in the respiratory tract, the clearance of

microorganisms is reduced and chronic bacterial infections resulting in inflammation of lung

tissue and fibrosis of the airways are a fact (Tiddens 2002). Staphylococcus aureus (51.5%) and

Pseudomonas aeruginosa (55.0%) infections play the greatest role in morbidity and mortality

(Cystic Fibrosis Foundation 2006). Local application of aerosolized antibiotics in the lung to

combat these microorganisms has been proven successful. The primairy objectives are to pre-

vent the development of or to stabilise a chronic infection with Pseudomonas aeruginosa. Apply-

ing aerosolized tobramycin or colistimethate sodium early in recently acquired Pseudomonas

aeruginosa infections has shown to be effective either as monotherapy or in combination with

ciprofloxacin or intravenously administered tobramycin or colistimethate sodium (Döring and

Hoiby 2004). Once a chronic infection has been established, daily use of aerosolized antibiotics

to stabilise local inflammation and improve pulmonary function (FEV1) is common practice.

This strategy has been shown to result in improvement of lung function, or to slow down

deterioration of lung function and in a reduction in hospital admissions (Kerem et al., 2005).

Chap

ter 1

12

Monotherapy in exacerbation treatment with an aerosolized anti-pseudomonal drug has so

far not been proven effective (Sexauer and Fiel 2003). There is scarce experience with inhaled

antibiotics during exacerbations.

Cystic fibrosis: historyCystic fibrosis has been described as a separate entity for the first time by Dorothy Anderson, a

pathologist from New York, in 1938 (Littlewood 2002). At that time, the focus was on malfunc-

tion of the pancreas and therefore the disease was called cystic fibrosis of the pancreas. At a

later stage it was recognized that the disease is a more generalised secretory defect, influencing

the lung as well as the pancreas. Only in the late 1980’s, the cystic fibrosis transmembrane con-

ductance regulator (CFTR) gene, which encodes a membrane-bound cAMP-regulated chloride

channel, and the mutations causing the chloride transport defect, have been identified caus-

ing this defect. The ΔF508 is the most common mutation, indicating a missing phenylalanine

molecule in position 508 of the 1480 amino acid protein (Koch and Hoiby 1993).

Prospects for CF patients in the early days were bad: most patients died in infancy, predomi-

nantly from lung disease. Fortunately, in the same period the first antibiotics were discovered.

The main pathogen in CF identified at that time was Staphylococcus aureus and penicillin,

active against Staphylococcus aureus, became available for injection in 1944. The first use of an

aerosolized antibiotic in CF dates back from 1946 (Di Sant’Agnese and Andersen 1946). Appar-

ently, patients responded well to a dose of penicillin aerosol 20000 units 7 times a day alone or

in combination with intramuscular penicillin (Littlewood 2002). Whether every CF patient had

access to this therapy in those days could not be found in literature, but aerosol antibiotics have

been used extensively in later years (Hodson et al., 1981, Kuhn 2001). Certainly the penicillin

aerosol had its application in pulmonary diseases like lung abscess (Wolcott and Murphy 1957)

and pneumococcal pneumonia (Kuhn 2001). A dry powder inhaler with penicillin for inhala-

tion became available in the late 1940’s. However, no recording has been found on CF patients

inhaling this dry powder.

Already in the fifties Pseudomonas aeruginosa was cultured but Staphylococcus aureus was

still considered to be the main pathogen. In the early 1970s the first successes on early aggres-

sive (intravenous) antibiotic treatment were documented (Mearns 1972). It was shown that the

mortality rate for children under 5 years of age could be reduced from 14% to 6.5% by early and

vigorous anti-staphylococcal treatment.

In the late seventies it was found that Pseudomonas aeruginosa infection and its severity

were closely associated with the prognosis (Hoiby et al., 1977) and opinion changed from

CF being a lethal disease to CF being a condition that could be controlled. Antibiotics active

against Pseudomonas aeruginosa became available: colistimethate sodium (1959), carbenicillin

(1967), gentamicin (1968), tobramycin (1971). Newer antibiotics became available in later years:


piperacillin (1982), ceftazidime (1983), aztreonam (1986), ciprofloxacin (1986) and meropenem

(1995). Intermittent, three monthly courses of two weeks of intravenous antibiotics were intro-

duced in Denmark with impressive results (Szaff et al., 1983), but these results could not be

confirmed in a subsequent study two decades later (Elborn et al., 2000). Despite good results,

some patients started to relapse between two intravenous courses and aerosolized antibiotics

got renewed interest.

The main objective for this approach was the knowledge that penetration of intravenously

administered aminoglycosides into lung parenchymal tissue and bronchial secretions is low

(Pennington 1981). Inhalation, on the other hand, results in higher sputum concentrations,

direct drug deposition in the lung, a low risk of systemic toxicity and enables long term treat-

ment outside the hospital (Steinkamp 1991). Pulmonary administration of these antibiotics

showed to be effective: long-term twice-daily aerosolized gentamicin (80 mg) and carbenicillin

(1 gram) were effective in patients with a relapsing chronic Pseudomonas aeruginosa infection

(Hodson et al., 1981), as were tobramycin (80 mg) and ticarcillin (1 gram) twice daily (Wall et al.,

1983). Gradually the use of aerosolized antibiotics increased, despite a continuing discussion

on the possible increase in antibiotic resistance (Ashby et al., 1993). Anti-staphylococcal aerosol

treatment has been applied, but it was subsequently shown to have no additional effect over

agents administered orally (Nolan et al., 1982; MacLusky et al., 1986). Furthermore, aerosolized

penicillins were avoided because of concerns regarding development of hypersensitivity,

odour, effect on teeth and greater difficulty in nebulisation (Kerem et al., 2005).

Nowadays, colistimethate sodium and tobramycin are used in anti-pseudomonal aerosol

therapy for prevention or stabilisation of chronic infections. Patients with a chronic Pseudomo-

nas aeruginosa infection benefited from long-term nebulised antibiotics (Jensen et al., 1987),

and early treatment with aerosolized colistimethate sodium or tobramycin to eradicate

Pseudomonas aeruginosa appeared effective too (Littlewood et al., 1985, Valerius et al., 1991,

Frederiksen et al., 1997, Ratjen 2001). Although aerosolized colistimethate sodium has been

used for chronic Pseudomonas aeruginosa infection for decades, proper prospective placebo

controlled trials are missing. A prospective unblinded study on aerosolized colistimethate

sodium and tobramycin showed superiority of tobramycin in improvement of lung function,

but the decrease of Pseudomonas aeruginosa in sputum was similar (Hodson et al., 2002,

Adeboyeku et al., 2006). Some critical remarks should be made however: the subjects had all

been using colistimethate sodium prior to the study so were naïve to tobramycin for inhalation

and a relatively low dose of 80 mg of colistimethate sodium was compared to a regular 300

mg dose of tobramycin. Efficacy of tobramycin solution for inhalation (TSI) has been studied

in a clinical trial with 520 patients, being the first placebo controlled trial of this size with an

anti-pseudomonal antibiotic for inhalation (Ramsey et al., 1999). Treatment with tobramycin

resulted in an improvement in lung function, a decrease of Pseudomonas aeruginosa density in

sputum and a lower risk on hospitalisation.

Chap

ter 1

14

Although inhaled antibiotics appeared to be relatively safe, in some patients airway nar-

rowing was observed during or shortly after nebulisation. Additives, present in intravenous

medications, were designated as one of the possible causes. Although the mechanisms of

drug-induced airway narrowing in CF have not been elucidated, preservative-free preparations

for inhalation are nowadays preferred by many clinicians. Nevertheless, airway narrowing after

colistimethate sodium or tobramycin may occur in susceptible patients (Ramsey et al., 1999,

Hodson et al., 2002).

Cystic fibrosis: factsThe incidence of cystic fibrosis in the Netherlands is 1:4750 and currently 1275 patients have

been identified. There are approximately 36000 CF patients in the 27 countries in the European

Union, corresponding with a prevalence of 0.737 per 10000 patients (Farrell 2008). The Dutch

ratio, 0.781 per 10000 patients, is similar to this. This means that cystic fibrosis meets the require-

ments for an orphan disease (prevalence < 5 per 10000; www.orpha.net). A total of 70000-100000

patients worldwide have been estimated (Hodson et al., 2007, Cystic Fibrosis Worldwide 2005),

of which approximately 30000 patients live in the USA (Cystic Fibrosis Foundation, 2006). There

is a high likelihood of underreporting and underdiagnosing, especially in developing countries.

USA data show that about 30% of patients in the patient registry program were age 18 or older

in 1990 and this percentage has grown to almost 45% in 2006. The predicted median survival

has risen to 36.9 years in 2006 and is expected to increase in future years because of improved

therapeutic interventions (Cystic Fibrosis Foundation, 2006).

Over 90% of CF patient premature deaths can be attributed to loss of pulmonary function

due to infection and inflammation caused by microorganisms (Koch and Hoiby 1993, Geller et

al., 2002, Döring et al., 2000). The presence of Pseudomonas aeruginosa is a significant predic-

tor of mortality. A relationship between chronic respiratory tract infection with Pseudomonas

aeruginosa, decline in lung function and mortality has been found (Henry et al., 1992). The

relative risk of death, adjusted for age and gender, for a CF patient doubles with each 10% loss

of FEV1predicted (Kerem et al., 1992). Effective prevention of pulmonary infection or treatment of

inflammation caused by Pseudomonas aeruginosa is therefore of utmost importance.

Cystic fibrosis: recent developments in aerosolized drug therapyIn recent years, aerosols of other antibiotics (aztreonam, levofloxacin, ciprofloxacin and ami-

kacin) have been studied and are either in a preclinical or clinical phase of development. Next

to these developments, explorations on inhaler devices have attracted attention during the

past two decades. Although aerosolized antibiotics have been used since 1946 and various

inhalers have been available for a long time (Anderson 2005), only in the mid 1980s research

on nebulisers and compressors for use in CF drug aerosolisation was initiated (Newman et al.,


1986a, Newman et al., 1986b). Conventional jet nebulisers were the first nebulisers used, but

soon these devices became known for their inefficiency. An improved pulmonary drug delivery

became possible with the introduction of breath-enhanced, breath-actuated and breath-

controlled devices (Devadason et al., 1997, Asmus et al., 2002, Leung et al., 2004, Kohler et al.,

2005). Ultrasonic nebulisers, which generally have a higher output rate and produce a larger

average particle size than jet nebulisers, were used as well (Rau 2002). However, neither jet nor

ultrasonic nebulisers are ideal for administration of aerosols in CF treatment. These devices are

large, need electricity to operate, are able to deliver only approximately 9-18% of the nominal

drug dose to the lung and require relatively long treatment times (> 10 min.), depending on the

nebuliser used (Le Brun et al., 1999, Touw et al., 1997). Newer wet nebulisers have been designed

to improve drug delivery efficiency and to decrease treatment times. Adaptive aerosol delivery

has been introduced (Leung et al., 2004, Marsden et al., 2002). This technique adapts the delivery

of aerosol to the patient’s breathing pattern, thereby aiming at optimising the delivered dose.

Vibrating mesh technology is a new technology for aerosol generation. Recently, data became

available from clinical studies with aztreonam and a vibrating mesh device (Gibson et al., 2006,

Retsch-Bogart et al., 2008). Although these new developments are promising, currently insuf-

ficient scientific data are available on clinical application in CF patients to fully estimate their

value in CF treatment.

The concept of dry powder inhalation for antibiotics in CF has been explored in recent years as

well. Several initiatives have been taken in the last two decades (Goldman et al., 1990, Crowther

Labiris et al., 1999, Le Brun et al., 2002, Newhouse et al., 2003) and the most recent clinical results

became available lately (Westerman, this thesis, Geller et al., 2007, Pilcer et al., 2008). The use of

a dry powder inhalation system is thought to result in improved lung deposition, in a higher

patient acceptance of and adherence to long term therapy and an improvement of the quality

of life of patients. Furthermore, the reduced treatment time may offer an opportunity for the

introduction of other, new drugs to a patient’s individual treatment.

Aim and outline of this thesisThe focus of this thesis is on investigations in the field of inhaled antibiotics in cystic fibrosis.

This introductory chapter has illustrated that CF treatment has evolved impressively during

the past decades. Inhalation of antibiotic aerosols has been established as chronic treatment

for Pseudomonas aeruginosa positive patients. Those involved in treating CF patients are aware

that there is still much to improve in inhalation treatment. That is why in recent years atten-

tion is growing to optimise aerosol therapy in CF patients, resulting in a gradual increase in

knowledge.

The aim of this thesis is to explore and expand knowledge on inhalation of aerosols in CF

patients. Chapter 2 describes the current knowledge regarding the inhalation of drugs by CF

patients: mechanisms of action, important indications, effects on lung function, exacerbation

Chap

ter 1

16

rates, survival, quality of life and adverse effects. This document serves as an European consen-

sus on inhaled medication and inhalation devices in CF. In chapter 2, the basics of inhalation in

CF and bacteriological safety and performance of nebulisers over time are especially relevant

for the scope of this thesis.

The starting point of the scientific work described in this thesis is the information that has

been obtained from bench studies with an antibiotic drug and a nebuliser, performed by our

group. In chapter 3, the relevance of testing a nebuliser-compressor combination with a specific

drug in a clinical setting is once again proven. The focus is on the influence of a more powerful

compressor on particle size generation and lung deposition of tobramycin for inhalation in

CF patients, compared to an average compressor. A bench study provided the in vitro data,

based upon which the two compressors were selected for application in patients. It was studied

whether the observed size differences in particle size generation between the two compressors

in vitro would result in different lung deposition patterns and adverse effects in vivo.

Chapters 4-8 are dedicated to the development of a dry powder inhaler of colistimethate

sodium for CF patients. The development of the dry powder drug, including the choice of the

antibiotic (chapter 4), and the development of the inhaler (chapter 5) were aspects investigated

prior to testing the drug-inhaler combination in vivo. In addition to testing the clinical feasibility

of the dry powder inhaler for the first time in volunteers (chapter 6) and patients (chapter 7),

initial results have been obtained with dry powder inhalation of colistimethate sodium, includ-

ing lung deposition (pharmacokinetic approach). The effect on lung deposition after adapting

several parameters in the inhalation process is described in chapter 8. Finally, the conclusions,

including strengths and limitations of the studies described in this thesis are discussed and

perspectives and directions for future research are given.


References

Adeboyeku D, Scott S, Hodson ME. Open follow-up study of tobramycin nebuliser solution and colistin in patients with cystic fibrosis. J Cyst Fibros 2006; 5:261-263.

Anderson PJ. History of aerosol therapy: liquid nebulization to MDIs to DPIs. Respir Care 2005; 50:1139-1150.

Ashby BL, Ramsey BW, Smith AL. Aerosolized Tobramycin in Patients with Cystic Fibrosis. N Engl J Med 1993; 329:1659-1660.

Asmus MJ, Stewart BA, Milavetz G, Teresi ME, Han SH, Wang D et al. Tobramycin as a pharmacologic tracer to compare airway deposition from nebulizers. Pharmacotherapy 2002; 22:557-563.

Crowther Labiris NR, Holbrook AM, Chrystyn H, MacLeod SM, Newhouse MT. Dry powder versus intrave-nous and nebulized gentamicin in cystic fibrosis and bronchiectasis. Am J Respir Crit Care Med 1999; 160:1711-1716.

Cystic Fibrosis Foundation. Patient Registry 2006 Annual Report, Bethesda, Maryland, USA.Cystic Fibrosis Worldwide. Annual report. 2005.Devadason SG, Everard ML, Linto JM, Le Souef PN. Comparison of drug delivery from conventional versus

“Venturi” nebulizers. Eur Respir J 1997; 10:2479-2483.Di Sant’Agnese PEA, Andersen DH. Celiac Syndrome IV Chemotherapy I infections of the respiratory tract

associated with cystic fibrosis of the pancreas; observations with penicillin and drugs of the sulphon-amide group, with special reference to penicillin aerosol. Am J Dis Child 1946; 72:17-61.

Döring G, Conway SP, Heijerman HGM, Hodson ME, Hoiby N, Smyth A et al. Antibiotic therapy against Pseudomonas aeruginosa in cystic fibrosis: a European consensus. Eur Respir J 2000; 16:749-767.

Döring G, Hoiby N. Early intervention and prevention of lung disease in cystic fibrosis: a European consen-sus. J Cyst Fibros 2004; 3:67-91.

Elborn JS, Prescott RJ, Stack BH, Goodchild MC, Bates J, Pantin C et al. Elective versus symptomatic antibiotic treatment in cystic fibrosis patients with chronic Pseudomonas infection of the lungs. Thorax 2000; 55:355-358.

Farrell PM. The prevalence of cystic fibrosis in the European Union. J Cyst Fibros 2008; 7:450-453.Frederiksen B, Koch C, Hoiby N. Antibiotic treatment of initial colonization with Pseudomonas aeruginosa

postpones chronic infection and prevents deterioration of pulmonary function in cystic fibrosis. Pediatr Pulmonol 1997; 23: 330-335.

Geller DE, Konstan MW, Smith J, Noonberg SB, Conrad C. Novel tobramycin inhalation powder in cystic fibrosis subjects: Pharmacokinetics and safety. Pediatr Pulmonol 2007; 42:307-13.

Geller DE, Pitlick WH, Nardella PA, Tracewell WG, Ramsey BW. Pharmacokinetics and bioavailability of aero-solized tobramycin in cystic fibrosis. Chest 2002; 122:219-226.

Gibson RL, Retsch-Bogart GZ, Oermann C, Milla C, Pilewski J, Daines C et al. Microbiology, safety, and pharmacokinetics of aztreonam lysinate for inhalation in patients with cystic fibrosis. Pediatr Pulmonol 2006; 41:656-665.

Goldman JM, Bayston SM, O’Connor S, Meigh RE. Inhaled micronised gentamicin powder: a new delivery system. Thorax 1990; 45:939-940.

Henry RL, Mellis CM, Petrovic L. Mucoid Pseudomonas aeruginosa is a marker of poor survival in cystic fibrosis. Pediatr Pulmonol 1992; 12:158-161.

Hodson ME, Gallagher CG, Govan JR. A randomised clinical trial of nebulised tobramycin or colistin in cystic fibrosis. Eur Respir J 2002; 20:658-664.

Hodson ME, Geddes D, Bush A. Cystic Fibrosis. London: Hodder Arnold 2007; 3rd edition.:Hodson ME, Penketh ARL, Batten JC. Aerosol carbenicillin and gentamicin treatment of Pseudomonas

aeruginosa infection in patients with cystic fibrosis. Lancet 1981; 1137-1139.Hoiby N, Flensborg EW, Beck B, Friis B, Jacobsen SV, Jacobsen L. Pseudomonas aeruginosa infection in cystic

fibrosis. Diagnostic and prognostic significance of Pseudomonas aeruginosa precipitins determined by means of crossed immunoelectrophoresis. Scand J Respir Dis 1977; 58:65-79.

Chap

ter 1

18

Jensen T, Pedersen SS, Garne S, Heilmann C, Hoiby N, Koch C. Colistin inhalation therapy in cystic fibrosis patients with chronic Pseudomonas aeruginosa lung infection. J Antimicrob Chemother 1987; 19:831-838.

Kerem E, Conway S, Elborn S, Heijerman H. Standards of care for patients with cystic fibrosis: a European consensus. J Cyst Fibros 2005; 4:7-26.

Kerem E, Reisman J, Corey M, Canny GJ, Levison H. Prediction of mortality in patients with cystic fibrosis. N Engl J Med 1992; 326:1187-1191.

Koch C, Hoiby N. Pathogenesis of cystic fibrosis. Lancet 1993; 341:1065-1069.Kohler E, Sollich V, Schuster-Wonka R, Jorch G. Lung deposition after electronically breath-controlled inha-

lation and manually triggered conventional inhalation in cystic fibrosis patients. J Aerosol Med 2005; 18:386-395.

Kuhn RJ. Formulation of aerosolized therapeutics. Chest 2001; 120:94S-98S.Le Brun PP, de Boer AH, Mannes GP, de Fraiture DM, Brimicombe RW, Touw DJ et al. Dry powder inhalation

of antibiotics in cystic fibrosis therapy: part 2. Inhalation of a novel colistin dry powder formulation: a feasibility study in healthy volunteers and patients. Eur J Pharm Biopharm 2002; 54:25-32.

Le Brun PP, Vinks AA, Touw DJ, Hekelaar N, Mannes GP, Brimicombe RW et al. Can tobramycin inhalation be improved with a jet nebulizer? Ther Drug Monit 1999; 21:618-624.

Leung K, Louca E, Coates AL. Comparison of breath-enhanced to breath-actuated nebulizers for rate, consistency, and efficiency. Chest 2004; 126:1619-1627.

Littlewood J. The history of the development of cystic fibrosis care. www.cysticfibrosismedicine.com 2002.;

Littlewood JM, Miller MG, Ghoneim AT, Ramsden CH. Nebulised colomycin for early pseudomonas colonisa-tion in cystic fibrosis. Lancet 1985; 1:865.

MacLusky I, Levison H, Gold R, McLaughlin FJ. Inhaled antibiotics in cystic fibrosis: is there a therapeutic effect? J Pediatr 1986; 108:861-865.

Marsden RJ, Conway SP, Dodd ME, Edenborough FP, Rigby AS, Taylor CJ et al. Comparison of the HaloLite® adaptive aerosol delivery (AAD™) system with a high output nebuliser system in cystic fibrosis patients. Thorax 2002; 57:48-94 (P114).

Mearns MB. Treatment and prevention of pulmonary complications of cystic fibrosis in infancy and early childhood. Arch Dis Child 1972; 47:5-11.

Newhouse MT, Hirst PH, Duddu SP, Walter YH, Tarara TE, Clark AR et al. Inhalation of a dry powder tobramycin PulmoSphere formulation in healthy volunteers. Chest 2003; 124:360-366.

Newman SP, Pellow PG, Clarke SW. Choice of nebulisers and compressors for delivery of carbenicillin aerosol. Eur J Respir Dis 1986a; 69:160-168.

Newman SP, Pellow PG, Clarke SW. Droplet size distributions of nebulised aerosols for inhalation therapy. Clin Phys Physiol Meas 1986b; 7:139-146.

Nolan G, Moivor P, Levison H, Fleming PC, Corey M, Gold R. Antibiotic prophylaxis in cystic fibrosis: inhaled cephaloridine as an adjunct to oral cloxacillin. J Pediatr 1982; 101:626-630.

Pennington JE. Penetration of antibiotics into respiratory secretions. Rev Infect Dis 1981; 3:67-73.Pilcer G, Goole J, Van Gansbeke B, Blocklet D, Knoop C, Vanderbist F et al. Pharmacoscintigraphic and phar-

macokinetic evaluation of tobramycin DPI formulations in cystic fibrosis patients. Eur J Pharm Biopharm 2008; 68:413-421.

Ramsey BW, Pepe MS, Quan JM, Otto KL, Montgomery AB, Williams-Warren J et al. Intermittent administra-tion of inhaled tobramycin in patients with cystic fibrosis. Cystic Fibrosis Inhaled Tobramycin Study Group. N Engl J Med 1999; 340:23-30.

Rau JL. Design principles of liquid nebulization devices currently in use. Respir Care 2002; 47:1257-75; discussion 1275-1278.

Retsch-Bogart GZ, Burns JL, Otto KL, Liou TG, McCoy K, Oermann C et al. A phase 2 study of aztreonam lysine for inhalation to treat patients with cystic fibrosis and Pseudomonas aeruginosa infection. Pediatr Pulmonol 2008; 43:47-58.


Sexauer WP, Fiel SB. Aerosolized antibiotics in cystic fibrosis. Semin Respir Crit Care Med 2003; 24:717-726.Steinkamp G. [Antibiotic inhalation in cystic fibrosis. A review of the literature]. Monatsschr Kinderheilkd

1991; 139:73-80.Szaff M, Hoiby N, Flensborg EW. Frequent antibiotic therapy improves survival of cystic fibrosis patients

with chronic Pseudomonas aeruginosa infection. Acta Paediatr Scand 1983; 72:651-657.Tiddens HAWM. Detecting early structural lung damage in cystic fibrosis. Pediatr Pulmonol 2002; 34:228-

231.Touw DJ, Jacobs FA, Brimicombe RW, Heijerman HG, Bakker W, BreimerBriemer DD. Pharmacokinetics

of aerosolized tobramycin in adult patients with cystic fibrosis. Antimicrob Agents Chemother 1997; 41:184-187.

Valerius NH, Koch C, Hoiby N. Prevention of chronic Pseudomonas aeruginosa colonisation in cystic fibrosis by early treatment. Lancet 1991; 338:725-726.

Wall MA, Terry AB, Eisenberg J, McNamara M, Cohen R. Inhaled antibiotics in cystic fibrosis. Lancet 1983; 321:13251983.

Wolcott MW, Murphy JD. The changing picture of lung abscess therapy. Dis Chest 1957; 32:62-69.Wolff RK, Niven RW. Generation of aerosolized drugs. J Aerosol Med 1994; 7:89.

2 Inhaled medication and inhalation devices for lung disease in patients with cystic fibrosis: a European consensus

Harry Heijerman, Elsbeth Westerman, Steven Conway, Daan Touw, Gerd Döring for the consensus working group1

1Baroukh Assael, Cystic Fibrosis Center, Verona, Italy; Ian Balfour-Lynn, Royal Brompton & Harefield

NHS Trust, Sydney Street, London, United Kingdom; Gabriel Bellon, Hôpital Debrousse, Lyon,

France; Celeste Barreto, Hospital de Santa Maria, Lisbon, Portugal; Cesare Braggion, Ospedale

Civile Maggiore, Verona, Italy; Steven Conway, St James University Hospital, Leeds, UK; Christiane

De Boeck, University Hospital Gasthuiberg, Leuven, Belgium; Gerd Döring, Institute of Medical

Microbiology and Hygiene, Eberhard-Karls-Universität, Tübingen, Germany; Jean-Christophe

Dubus, Faculté de Médecine, Marseille, France; Irmgard Eichler, EMEA, London; Mark Elkins, Royal

Prince Alfred Hospital, Sydney Australia; Henderik Frijlink, University of Groningen, Groningen,

The Netherlands; Charles Gallagher, St Vincent’s Hospital, Dublin, Ireland; Silvia Gartner, University

Hospital Vall d’Hebron, Barcelona, Spain; David Geller, Nemours Children’s Clinic, Orlando, USA;

Matthias Griese, University of Munich, Munich, Germany; Harry Heijerman, Haga Teaching Hospital,

The Hague, The Netherlands; Lena Hjelte, Karolinska University Hospital Huddinge, Stockholm,

Sweden; Margaret Hodson, Royal Brompton Hospital, London, United Kingdom; Niels Høiby,

Rigshospitalet, Copenhagen, Denmark; James Littlewood, St. James’s University Hospital, Leeds,

UK; Anne Malfroot, Academisch Ziekenhuis, Vrije Universiteit Brussel, Brussels, Belgium; Alexander

Möller, University Children’s Hospital, Zurich, Switzerland; Petr Pohunek, Charles University 2nd

Medical School, Prague, Czech Republic; Tanja Pressler, Rigshospitalet, Copenhagen, Denmark;

Alexandra Quittner, University of Miami, Miami, USA; Felix Ratjen, Hospital for Sick Children,

Toronto, Canada; Martin Schöni, University of Berne, Berne, Switzerland; Giovanni Taccetti,

Ospedale Meyer, Florence, Italy; Harm Tiddens, Erasmus University Medical Center, Rotterdam, The

Netherlands; Daan Touw, Apotheek Haagse Ziekenhuizen, The Hague, The Netherlands; Elsbeth

Westerman, Apotheek Haagse Ziekenhuizen, The Hague, The Netherlands.

Journal of Cystic Fibrosis, submitted for publication.

Chap

ter 2

22

Summary

Inhalation of drugs in cystic fibrosis related lung disease has been proven to be highly effec-

tive. Consequently, an increasing number of drugs and devices have been developed for CF

lung disease or are currently under development. In this European consensus document we

review the current status of inhaled medication in CF, including the mechanisms of action of

the various drugs, their modes of administration and indications, their effects on lung function,

exacerbation rates, survival and quality of life, as well as side effects. Specifically we address

antibiotics, mucolytics/mucous mobilizers, anti-inflammatory drugs, bronchodilators and com-

binations of solutions. Additionally, we review the current knowledge on devices for inhalation

therapy with regard to optimal particle sizes and characteristics of wet nebulisers, dry powder

and metered dose inhalers. Finally, we address the subject of testing new devices before market

introduction.

Inhaled medication and inhalation devices for lung disease in patients with cystic fibrosis: a European consensus 23

Introduction

While the first description of the hereditary disease cystic fibrosis (CF), (Hodson, 2007)

emphazised fatal congenital steatorrhea and pancreatic destruction, lung disease has now

been recognized to have the largest impact on morbidity and mortality in older people with

CF (CF Foundation, 2006). Lung disease develops as a consequence of mutations in the CF

transmembrane conductance regulator (CFTR) gene (Hodson et al., 2007), which encodes a

membrane-bound cAMP-regulated chloride channel: diminished chloride and water secretion

leads to viscous secretions in the affected airways (Matsui et al., 1998; Ratjen & Döring, 2003).

This impairs mucociliary clearance (Matsui et al., 1998), thereby facilitating chronic bacterial

infections, which may start at a very early age (CF Foundation, 2006; Armstrong et al., 1997;

Stern et al., 2002).

Among the bacterial pathogens isolated from airways of CF patients the triad Haemophilus

influenzae, Staphylococcus aureus and Pseudomonas aeruginosa (Govan et al., 1990; van Schilf-

gaarde et al., 1999) has been isolated most frequently. Infections with some members of the B.

cepacia complex are associated with a markedly shortened median survival (Liou et al., 2001).

Other microbial pathogens isolated from CF patients include Stenotrophomonas maltophilia,

Achromobacter xylosoxidans, Mycobacteria ssp., Aspergillus fumigatus and (Döring & Hoiby,

2004) and strict anaerobes (Rogers et al., 2004; Rogers et al., 2006).

Pseudomonas aeruginosa, a Gram-negative bacterium found in many natural and man-made

water sources, is present in approximately 27% of patients aged 2–5 years and approximately

80% of patients aged 25–34 years (CF Foundation, 2006). Thus this opportunistic bacterium

pathogen is regarded as the most important pathogen in CF (Regelmann et al., 1990; Ramsey et

al., 1999; Hoiby & Frederiksen 2000; Kosorok et al., 2001). Respiratory infections with Pseudomo-

nas aeruginosa are difficult to treat due to growth of the pathogen in biofilm-like macrocolonies

(Worlitzsch et al., 2002; Döring & Hoiby, 2004). Nevertheless, various treatment strategies have

been developed during the past few decades that have a significant positive impact on prog-

nosis (Döring & Hoiby, 2004). The predicted median survival age of CF individuals in the USA

increased from 14 years in 1969 to 36.5 years in 2005, and 43% of patients are 18 years of age

or older (CF Foundation, 2005). European registries report similar increases in median survival

ages (Stern et al., 2002). Repeated courses of inhaled antibiotics using high doses for the

treatment of lung disease in CF patients has been applied increasingly in the last two decades

(Döring & Hoiby, 2004). This strategy has circumvented the problem of the poor penetration of

intravenously administered antibiotics into lung parenchymal tissue and bronchial secretions,

and their potential systemic toxicity when given over prolonged periods of time.

One of the most striking characteristics of Pseudomonas aeruginosa is its extraordinary

capacity to develop resistance to virtually all antipseudomonal agents through the selection

of genetic mutations. Repeated and prolonged treatment strategies may therefore increase the

resistance of the pathogen to the applied antibiotics, as demonstrated in trials using tobramycin

Chap

ter 2

24

(Smith et al., 1989), leading to a strategy of intermittently administered of this drug (Ramsey et

al., 1999). Development of antibiotic resistance in Pseudomonas aeruginosa is also facilitated

by the occurrence of hypermutable (or mutator) strains, deficient in the DNA mismatch repair

system (Oliver et al., 2000; Hogart et al., 2007). To avoid the development of resistance and in

an attempt to eradicate non-mucoid Pseudomonas aeruginosa, many European CF centres

started antibiotic treatment early after the first detection of the pathogen with great success

(Valerius et al., 1991; Wiesemann et al., 1998; Ratjen et al., 2001; Gibson et al., 2003; Taccetti

et al., 2005; Gibson et al., 2007). In CF patients initially colonized with mucoid Pseudomonas

aeruginosa strains, or patients in whom initially nonmucoid strains have already switched to

mucoid strains, it may not be possible to eradicate pathogens from their airways,

Chronic airway inflammation is a uniformly observed symptom in patients with CF (Konstan

et al., 1994; Döring & Hoiby, 2004; Döring & Ratjen, 2007). Chronic lung inflammation with

episodes of acute exacerbations initiates several physiologi cal and metabolic changes with del-

eterious effects including weight loss, anorexia, and metabolic breakdown. Thus, as an adjunct

to optimal antibiotic therapy, anti-inflammatory therapy is warranted to avoid a decline in lung

function, tissue remodeling and tissue destruction. Compared to inhaled corticosteroids, the

non-steroidal anti-inflammatory drug ibuprofen gave promising results in children and ado-

lescents with CF (Konstan et al., 1995; Konstan et al., 2007), while a phase III trial in CF patients

with the LTB4-receptor antagonist BIIL 284 (Birke et al., 2001) was terminated due to adverse

effects of the drug. Trials with protease inhibitors including aerosolized recombinant secretory

leukocyte protease inhibitor (SLPI) or α1-proteinase inhibitor (α1-PI) have not been consistently

successful (Döring & Hoiby, 2004; Griese et al., 2008), while antibiotics with anti-inflammatory

effects, such as macrolides, have improved lung function in CF children and adults, infected

with chronic Pseudomonas aeruginosa (Wolter et al., 2002; Equi et al., 2002; Saiman et al., 2003).

One of the open questions in this context is which markers of inflammation and which

diagnostic techniques or molecules should be employed to monitor the success of anti-

inflammatory therapy in people with CF.

Since purulent CF sputum impairs the activity of aerosolized drugs, administration of

aerosolized antibiotics is generally preceded by physiotherapy, and/or bronchodilatators or

mucolytic agents such as recombinant human deoxyribonuclease (rhDNase, dornase alfa)

(Fuchs et al., 1994; Frederiksen et al., 2006). Additionally, drugs improving mucociliary clearance

such as hypertonic saline may be beneficial (Robinson et al., 1997; Elkins et al., 2006).

Since inhalation therapy was discussed as part of a ECFS Consensus Conference in 1999

(Döring et al., 2000) and 2003 (Döring & Hoiby, 2004), several new drug formulations and

new inhalation devices have been developed. Here we review the current status of inhaled

medication in CF, including the mechanisms of action of the various drugs, their optimal

administration and important indications, their effects on lung function, exacerbation rates,

survival and quality of life, as well as side effects. Specifically we address antibiotics, mucolytics/

mucous mobilizers, anti-inflammatory drugs, bronchodilators and combinations of solutions.


Additionally, we review the current knowledge on devices for inhalation therapy with regard

to the characteristics of wet nebulisers, dry powder inhalers (DPIs) and metered dose inhalers

(pMDIs) and their interaction with the drug formulation and patients. Finally, we address the

subject of testing new devices before they are introduced onto the market.

Inhaled medications

AntibioticsTobramycin

The aminoglycoside tobramycin is a bactericidal drug that inhibits protein synthesis by irrevers-

ibly binding to the 30S bacterial ribosome. It is active against most Gram-negative bacilli, but

typically displays no significant activity against BCC strains or Stenotrophomonas maltophilia

while it is active against strains of Enterococcus and Staphylococcus. Tobramycin Solution for

Inhalation (TSI) is registered as TOBI® (300 mg/5 ml) in combination with a PARI LC PLUS® reus-

able jet nebuliser and a suitable compressor resulting in a flow rate of 4-6 L/min and/or a back

pressure of 110-217 kPa. Additionally, tobramycin is present in Bramitob® (300 mg/4 ml) in

combination with a PARI LCPLUS® reusable jet nebuliser and the PARI TURBO BOY® compres-

sor.

Uptake across the bacterial cell wall is energy-dependent and is impaired in anaerobic envi-

ronments (Park et al., 1992). Thus, the low oxygen partial pressure in CF sputum plugs (Worlitz-

sch et al., 2002) may limit the efficacy of this drug. Tobramycin is positively charged and thought

to be bound in CF airways to the negatively charged DNA fibers and Pseudomonas aeruginosa

alginate. Despite these considerations, intermittent (28-day on/28-day off) treatment, using

300 mg of tobramycin twice daily, significantly improved lung function and reduced sputum

Pseudomonas aeruginosa density compared with placebo in CF patients (Ramsey et al., 1993

Footnotes:*: Numbers and letters in brackets refer to the grading of medical scientific publications. For details see Table 1.

Table 1. Grading of medical scientific publications

Level Description1a Systemic review of 2 or more unrelated randomised controlled trials of level 1b.1b Individual randomised controlled trial of good quality and sufficient patient numbers included2a Systemic review cohort studies of level 2b2b Individual cohort study (including low quality RCT; e.g., <80% follow-up)3a Systemic review of case-control studies of level 3b3b Individual Case-Control Study4 Case-series (and poor quality cohort and case-control studies)5 Expert opinion without explicit critical appraisal, or based on physiology, bench research or “first

principles”

Ref: Oxford Centre for Evidence-based Medicine Levels of Evidence (May 2001)

Chap

ter 2

26

[1a]*). Increases in lung function of about 10% at week 20 were most marked in adolescent

patients (aged 13–17 years) and maintained for up to 96 weeks in an open-label extension

study (Ramsey et al., 1999 [1a]). Fewer TSI than placebo recipients required parenteral antip-

seudomonal agents or hospitalisation (Moss et al., 2001 [2b], Moss et al., 2002 [2b]; Murphy et

al., 2004). Two open-label uncontrolled trials have shown that aerosolized tobramycin safely

eradicated Pseudomonas aeruginosa in the majority of CF patients for up to three months

(Ratjen et al., 2001 [2b]; Gibson et al., 2007 [2b]). Pseudomonas aeruginosa eradication was

associated with reduced neutrophilic airway inflammation.

TSI is generally well tolerated. Renal toxicity or hearing loss has not been reported in clinical

trials, although transient mild or moderate tinnitus occurred more frequently in TSI than pla-

cebo recipients (Ramsey et al., 1999 [1a]). Bronchoconstriction following inhalation of TSI has

been reported in both preservative free TSI and tobramycin solutions containing preservatives

such as phenol (Nikolaizik et al., 2002 [2b]). The use of inhaled beta-agonists may prevent the

post-inhalation decline in lung function (Ho et al., 2002 [2b]; Nikolaizik et al., 2002 [2b]).

Colistimethate sodium

Colistimethate sodium (Colomycin®, Promixin®) is a cyclic polypeptide antibiotic, derived

from Bacillus polymyxa varietas colistinus, and belongs to the polymyxin group. Due to their

cationic nature, polymyxin antibiotics can damage cell membranes and are bactericidal for

Gram-negative bacteria. There are no specific requirements concerning inhalation devices for

colistimethate sodium and thus the drug can be administered by ultrasonic or jet nebulisers or

by vibrating mesh devices.

Although colistimethate sodium for inhalation has been prescribed for more than 20 years

in people with CF for the treatment of Pseudomonas aeruginosa infections, controlled trials are

rare. A trial in 40 CF patients showed that inhalation with colistimethate sodium reduces symp-

tom scores and may have a protective effect on lung function (Jensen et al., 1997 [2b]). When

colistimethate sodium was compared with TSI, the latter drug was superior, concerning lung

function improvement, while both treatment regimens decreased Pseudomonas aeruginosa

sputum density (Hodson et al., 2002 [2b]; Adeboyeku et al., 2006 [2b]). The greater improve-

ment in lung function seen with TSI might have resulted from the fact that all patients had

previously used colistimethate sodium but were naïve to TSI. Also the dose of colistimethate

sodium used in the trial (80 mg twice daily) was lower than most physicians would prescribe

in adult CF patients. In combination with oral ciprofloxacin inhaled colistimethate sodium

effectively eradicated Pseudomonas aeruginosa for a period of 24 months in more than 80% of

treated CF patients (Frederiksen et al., 1997 [2b]). A European wide randomised double blinded

phase III study of colistimethate sodium administered by a dry powder inhaler (Colobreathe®)

has been carried out but results have not been reported to date.

Colistimethate sodium is generally well tolerated in CF patients, however. Bronchoconstric-

tion following inhalation is quite common, especially in CF patients, suffering from asthma or


airway hyperresponsiveness (Alothman et al., 2005 [2b]; Cunningham et al., 2001 [2b]). Colis-

timethate sodium must be inhaled promptly after reconstitution, since after prolonged times,

the drug is hydrolyzed into the bases colistin A (polymyxin E1) and colistin B (polymyxin E2).

Polymyxin E1 has been shown in animal studies to cause localized airway inflammation and

eosinophilic infiltration (FDA, 2007). Colistin (sulfate) is not suitable for treating CF patients due

to severe adverse effects (Westerman et al., 2004).

Aztreonam lysine

Aztreonam is a synthetic monobactam (monocyclic beta-lactam) antibiotic, which is active

against Gram-negative aerobic organisms and stable to most ß-lactamases. Aztreonam inhibits

synthesis of bacterial cell walls and has shown to produce clinically significant synergy with

aminoglycosides against Pseudomonas aeruginosa . Aztreonam lysine (AZLI) is a new, currently

unlicensed, formulation for aerosolized treatment of Pseudomonas aeruginosa infection in CF

patients. The AZLI formulation makes this compound safe for inhalation, whereas inhalation

of aztreonam arginine, used for intravenous treatment, can cause airway inflammation after

chronic inhalation therapy in CF patients (McCoy et al., 2008 [1b]). It is delivered by the eFlow®

electronic nebuliser which produces an aerosol with a narrow size distribution allowing periph-

eral lung deposition after 2 min of inhalation (Keller et al., 2003).

A double-blind, placebo-controlled, dose-escalation Phase 1b trial of single daily doses of 75

mg, 150 mg and 225 mg AZLI or placebo, self-administered by clinically stable CF patients >12

years of age, showed retention of anti-pseudomonal activity after nebulisation and no inhibition

by CF sputum (Gibson et al., 2006 [1b]). AZLl was active against multiply resistant Pseudomonas

aeruginosa, and in moderate sputum concentrations showed activity when tested against BCC

complex strains of genomovar I to V. AZLl was well tolerated in CF patients. The most common

adverse events were increased cough particularly in patients with the highest dose. Further

mild to moderate side effects were chest tightness and nasal congestiony. AZLI sputum con-

centrations exceeded the MIC50 for at least four hours post dose. (Gibson et al., 2006 [1b]).

In another double-blind, randomised, placebo-controlled Phase 2 study (Retsch-Bogart et

al., 2008, [1b]), the safety, tolerability and efficacy of 75 mg and 225 mg AZLl, inhaled twice

daily for 14 days was investigated in 105 CF patients with chronic Pseudomonas aeruginosa

infection. The drug significantly reduced Pseudomonas aeruginosa CFU density after seven

and 14 days but did not led to an increased isolation of Staphylococcus aureus, Burkholderia

complex, Stenotrophomonas maltophilia, or Alcaligenes xylosoxidans. FEV1 did not change. AZLI

caused a possible dose-related trend in the incidence and severity of cough in the higher dose.

Therefore, the 75 mg dose trice daily was tested against placebo in a Phase 3 study (Retsch-

Bogart et al., NACFC 2007 [1b]). Patients in the active arm showed a significant improvement in

clinical symptoms, percent change in FEV1, and in Pseudomonas aeruginosa CFU density at 28

days. Adverse events did not differ between the groups.

Chap

ter 2

28

In a further study, 75 mg of AZLI, inhaled twice or trice daily, was tested against placebo

(McCoy et al., 2008 [1b]) in 246 CF patients. At day 28, a significant improvement in clinical symp-

toms, percent change in FEV1 and in Pseudomonas aeruginosa CFU density was noted in both

treatment groups and at the end of the 56 day follow-up period, the treated groups showed a

significantly lesser need for additional inhaled or intravenous antibiotic therapy. Adverse events

did not differ between the groups. In an open-label follow-up study of 75 mg AZLl twice or trice

daily with a alternating 28 days on/off design, improvements in patient reported symptoms,

pulmonary function, and Pseudomonas aeruginosa CFU density were greater in the trice daily

group [1b]. A six-month phase 3 comparator study of 75 mg of AZLl trice daily against 300 mg

of TSI twice daily in a 28days on/28days off design is currently in progress.

Liposomal ciprofloxacin

Ciprofloxacin, a fluoroquinolone which affects gyrase function in bacteria, has been broadly

used by the oral route in patients with CF and other diseases. Aerosolisation of ciprofloxacin as

small particle aerosol or encapsulated in liposomes into guinea pigs, infected with Legionella

pneumophilia (Fitzgeorge et al., 1986), or in mice infected with Francisella tularansis (Conley

et al., 1997), prevented the death of the animals and suggested aerosol delivery to the lower

respiratory tract of CF patients to be effective. When used with appropriate nebuliser devices

liposomal disruption was minimal (Finlay et al., 1998). After a successful Phase 1 safety, toler-

ability and pharmacokinetic trial in healthy volunteers and a preclinical toxicology programme

currently, a Phase 2 safety and efficacy study of inhaled liposomal ciprofloxacin in 24 CF patients

is carried out using Pseudomonas aeruginosa CFU change in sputum as the primary endpoint.

Pharmacokinetic data suggested that once daily dosing may be possible.

Aerosol MP-376

A formulation of the fluoroquinolone levofloxacin for aerosol administration (MP-376) is cur-

rently undergoing clinical evaluation in patients with CF after results in healthy volunteers

have demonstrated that it is well tolerated (Griffith et al., 2007 [1b]). In the single within-subject

ascending dose study of 78, 175 and 260 mg levofloxacin, there were no serious adverse events

or significant changes in respiratory function between treatment groups and placebo. Systemic

absorption appears to be the major route for drug elimination from the lungs.

Amphotericin B

Amphotericin B is a widely used antifungal drug with activity against Cryptococcus neoformans,

Candida albicans, Aspergillus fumigatus and other species. The drug binds to sterols in the

plasma membranes of fungi, thereby interfering with membrane permeability. Its potentially

severe nephrotoxicity and neurotoxicity is a disadvantage of this drug. A liposomal amphot-

ericin B preparation (AmBisome®) reduces drug toxicity whilst maintaining antifungal activity

in murine models of pulmonary aspergillosis (Allen et al., 1994; Gilbert et al., 1996). Although


nebulised liposomal amphotericin B has been studied in different patient populations, data on

clinical efficacy and tolerability are inconclusive, possibility because of the lack of uniformity in

drug doses and administration methods (Knechtel et al., 2007 [1b]; Ruiz et al., 2005; Lowry et al.,

2007 [2b], Mohammed et al., 2006 [2a]).

No controlled trials with nebulised liposomal amphotericin B have been carried out in CF

patients, suffering from Aspergillus fumigatus pulmonary infection. Nebulisation of 50 mg of

liposomal amphotericin B once a week, administered by an adaptive aerosol delivery nebuliser

(HaloLite™) in five CF patients suffering from aggressive bronchopulmonary aspergillosis

(ABPA) once weekly was well tolerated, although the 8 ml drug dose required an up to 150 min

inhalation period (Tiddens et al., 2003). In another study, two persistently infected CF patients

became Aspergillus fumigatus culture negative for one to four months after 10 days of treatment

with 25 mg aerosolized liposomal amphotericin B twice daily (Sanchez-Sousa et al., 1996 [4]).

Mucolytics/Mucous mobilizersDornase alfa

Dornase alfa inhalation solution is a purified solution of recombinant human deoxyribonu-

clease (rhDNase), an enzyme cleaves sputum DNA, thereby reducing sputum viscoelasticity.

Dornase alfa is used in jet nebulisers (and not in ultrasonic nebulisers) connected to a compres-

sor. Multiple short and long term studies have demonstrated significant improvements in FEV1

after dornase alfa treatment compared to placebo in CF patients (Fuchs et al., 1994 [1b]; McCoy

et al., 1996 [1b]) and a good tolerance of the drug. Side effects include voice alteration and rash.

In some studies a significant decrease in the exacerbation rate (Fuchs et al., 1994 [1b]; Quan et

al., 2001 [1b]) and air trapping (Robinson et al., 2005 [2b]) was observed. A Cochrane review

concluded that dornase alfa improves lung function in short as well as long term trials (Jones

& Wallis, 2005 [1a]).

In clinical trials device combinations such as Durable Sidestream® with MOBILAIRE™,

Durable Sidestream® with Porta-Neb®, Hudson T Up-draft II® with Pulmo-Aide®, Respirgard

II Nebulizer® with Pulmo-Aide®, PARI LC PLUS with PARI PRONEB®, PARI BABY™ with PARI

PRONEB® have been used.

Hypertonic Saline

Nebulised hypertonic saline in CF treatment is water for injection (sterile) with a concentra-

tion of 3% to 7% sodium chloride. Increasing salt concentrations on the luminal side of the

respiratory epithelium is thought to hydrate the viscous mucus, thereby improving mucociliary

clearance and hence lung function (Donaldson et al., 2006 [2b]). Several studies have assessed

the efficacy of hypertonic saline in CF patients. A Cochrane review concluded that nebulised

hypertonic saline improves mucociliary clearance in CF patients in short-term clinical trials

and appears to increase lung function compared to control (Wark et al., 2005 [1a]). In a parallel

placebo controlled trial over 48 weeks, FEV1 and FVC increased in 82 patients receiving 7%

Chap

ter 2

30

hypertonic saline to 3.2% and 2.8%, respectively, compared to controls (Elkins et al., 2006 [1b]).

Hypertonic saline also reduced the percentage of exacerbations (56%) compared to placebo

in this study. In another study, an increase of FEV1 of 15% was observed after 14 days of treat-

ment with hypertonic saline (Eng et al., 1996 [2b]). When hypertonic saline was compared with

dornase alfa once daily and on alternate days in 48 children in an open cross over study, FEV1

increased in the daily dornase alfa group (16%), followed by alternate day dornase alfa (14%)

and only a modest improvement (3%) in patients treated with hypertonic saline (Suri et al., 2001

[2b]). However, large individual differences in response to dornase alfa and hypertonic saline

were found, suggesting that patients should be tested on an individual basis before long term

prescription is started (Ballmann et al., 2002 [2b]). Side effects of nebulised hypertonic saline

include bronchospasm and cough.

Denufosol tetrasodium

Denufosol tetrasodium (denufusol) inhalation solution is a selective P2Y2 receptor agonist

which activates an alternative chloride channel (Kellerman et al., 2008). This activation is

thought to result in an increase in the hydration of the respiratory epithelium, thereby improv-

ing mucociliary clearance and lung function. Compared to UTP and diquafosol, denufusol

shows a prolonged stability (Drutz et al., 1996; Olivier et al., 1996; Yerxa et al., 2002). In a phase

II study in patients with CF, the aerosolized drug improved several lung function parameters

(Deterding et al., 2007 [2b]). Adverse effects, such as cough and immediate decline in lung

function after inhalation, were similar in the placebo and the treatment group. Based on these

findings, a placebo controlled double blind phase III trial was initiated using 60 mg of denufusol

over a period of 6 months.

Lancovutide (Moli1901)

Lancovutide (Moli1901) is thought to activate intracellular calcium in alternative chloride chan-

nels, thereby increasing chloride transport and fluid secretion onto the apical surface of the

airway (Zeitlin et al., 2004). Indeed, in a phase II trial, a significant improvement of FEV1 was

observed by Moli1901 treatment in CF patients (Grasemann et al., 2007). Aerosolized lancovutide

was well tolerated. The most frequent adverse events were non-clinically significant cough and

throat irritation. An exploratory multi-center Phase IIb study is currently ongoing in Europe to

establish the optimal dose of Moli1901 in CF patients. Patients receive either placebo, 2.5 mg

lancovutide daily, every other day, or twice weekly for two months. The primary endpoint is the

change in the percentage of predicted FEV1.

Inhaled anti-inflammatory therapiesInhaled corticosteroids

Inhaled corticosteroids are used to reduce endobronchial inflammation in CF (Konstan et

al., 1993; Khan et al., 1995) and to minimize systemic adverse effects, experienced with oral


prednisolone (Rosenstein & Eigen, 1991). In clinical trials involving CF patients, different doses

of budesonide, beclomethasone or fluticasone propionate (400 -1600 µg/day) were used for

treatments of 3 to 52 weeks (van Haren et al., 1995 [2b]; Nikolaizik & Schoni, 1996 [2b]; Balfour-

Lynn et al., 1997 [1a]; Dauletbaev et al., 1999 [2b]; De Boeck et al., 2007; Bisgaard et al., 1997 [1b];

Wojtczak et al., 2001 [2b]. Decreased bronchial hyperreactivity in non-asthmatic CF patients

was observed in two studies (van Haren et al., 1995 [2b]; Bisgaard et al., 1997 [1b]).

No study has shown a statistically significant increase in lung function, although patients

receiving beclomethasone for 30 days showed a significant change in DLco (diffusing capacity

for carbon monoxide) (Nikolaizik & Schoni 1996). There was no beneficial change in sputum

inflammatory markers (Balfour-Lynn et al., 1997; Bisgaard et al., 1997; Dauletbaev et al., 1999)

but airway markers of inflammation fell markedly in lavage fluid (Wojtczak et al., 2001). Van

Haren et al. demonstrated small but significant improvements in daily symptom scores for

cough and dyspnoea in a small group of 12 patients but no improvement in mean overall

respiratory symptom, wellbeing or appetite scores was seen in a larger study (Balfour-Lynn et

al., 1997).

Inhaled corticosteroid treatment was generally well tolerated and the treatment did not

affect urine and blood cortisol, did not cause any decrease in adrenal reserve or any increase in

airway infection (Wojtczak et al., 2001). However, a recent study showed a significant slowing

in linear growth in pre-pubertal children receiving dry powder fluticasone propionate over 12

months compared to placebo (De Boeck et al., 2007).

The largest study tested the safety of a withdrawal of corticosteroid after switching all study

patients to fluticasone inhalation for a two month run in period. Patients were then randomised

to continue fluticasone or start placebo for the next six months (Balfour-Lynn et al., 2006 [1b]).

There was no difference in the primary outcome measure of time to first exacerbation between

the two groups, nor in lung function changes, oral or intravenous antibiotic use, or rescue

bronchodilator use. This study supports the conclusion from the Cochrane review that there is

neither evidence nor benefit or harm from corticosteroid use in CF (Balfour-Lynn et al., 2000).

The authors suggested that the majority of patients taking inhaled corticosteroids probably do

not need to do so.

Antiproteases

α1-AT and secretory leukoprotease inhibitor (SLPI) are two endogenous serine protease inhibi-

tors which inactivate neutrophil elastase a protease which has been shown to be present in

high concentrations in CF sputum and BALFs (Döring & Ratjen, 2007). Short term aerosol

delivery of α1-AT to 12 CF patients suppressed neutrophil elastase in the epithelial lining fluid

and restored anti-neutrophil elastase capacity (McElvaney et al., 1991). However, a phase II

trial to assess the clinical efficacy and safety of nebulised transgenic α1-AT did not show any

evidence to reduce airway inflammation (Martin et al., 2006 [1b]). In another open short term

study, a decrease in neutropil elastase activity, neutrophils, pro-inflammatory cytokines and

Chap

ter 2

32

Pseudomonas aeruginosa numbers was observed, however, aerosolized α1-AT treatment had

no positive effect on lung function in CF patients (Griese et al., 2007 [2b]). It is generally agreed,

that studies longer than four weeks in young children with moderate lung disease are neces-

sary to show potential drug efficacy of aerosolized α1-AT (Brennan et al., 2007). Aerosolized SLPI

at a dose of 100 mg twice daily for one week reduced epithelial lining fluid neutrophil elastase

in patients with CF, but 50 mg twice daily for two weeks were ineffective (McElvaney et al., 1993

[2b]; Vogelmeier et al., 1996 [2b]). The drug has not been further evaluated in clinical trials.

Bronchodilators

Inhaled bronchodilators are frequently prescribed for CF patients with atopy or those who

develop airway hyperreactivity secondary to bronchial damage (Eggleston et al., 1988 [2b]).

Bronchodilator therapy may increase mucociliary transport, decrease inflammatory damage to

the airways, increase exercise tolerance and decrease dyspnoea (Orenstein et al., 1991). Often

the short acting salbutamol or the long acting salmeterol are used by inhalation.

Most patients show a positive response at some time if repeatedly treated (Hordvik et

al., 1985 [2b]; Pattishall et al., 1990 [2b]). However, there are no long-term controlled trials of

inhaled β2-stimulants. A two month double-blind crossover trial of 90 µg salbutamol four times

daily significantly improved peak expiratory flow rate (PEFR) in patients with bronchial hyper-

responsiveness (Eggleston et al., 1991 [2b]). While lung function was not changed in this study,

treatment with inhaled salbutamol (pMDI, 180 µg b.i.d.) significantly improved respiratory

functions in a 12 month observational study (Konig et al., 1995 [2b]). However, in a subsequent

placebo-controlled double-blinded trial, CF patients, receiving six months of 180 µg inhaled

salbutamol twice daily, did not differ significantly compared to placebo in lung function tests

(Konig et al., 1998). In another study 18% of CF patients, who had salbutamol showed a signifi-

cant increase in FEV1 (Hughes et al., 2006 [2b]). Inhaled short-acting β-agonists did not improve

exercise performance or post exercise dyspnoea in CF patients despite significantly improving

FEV1 (Serisier et al., 2007 [1b]; Dodd et al., 2005 [2b]).

Greater benefits have been reported with the long acting bronchodilator salmeterol. In an

unblinded study (Bargon et al., 1997 [4]), dyspnoea improved even in patients not showing a

positive FEV1 response, when treated with 50 µg salmeterol twice daily for two weeks. In a 24

week treatment period, 100 µg salmeterol given twice daily was well tolerated and associated

with better pulmonary function, fewer interventions, and fewer respiratory symptoms com-

pared to treatment with salbutamol in CF patients with mild to moderate disease (Hordvik et

al., 2002 [2b]). Stable CF patients who responded to day time salbutamol showed significant

increases in nocturnal oxyhaemoglobin saturation, following salmeterol administration before

sleep (Salvatore et al., 2002 [1b]). A Cochrane review concluded that both short and long acting

ß-sympathomimetics can be beneficial in CF patients with bronchodilator responsiveness or

bronchial hyperresponsiveness (Halfide et al., 2005 [2a]).


Bronchial smooth muscle relaxation may increase airway compression and reduce cough

efficiency by inducing large airway collapse (Zach et al., 1985 [2b]) but negative responses are

unusual and collapse is unlikely during normal breathing (Eber et al., 1988 [2b]; Pattishall et al.,

1990 [2b]). No paradoxical responses were found with forced oscillation technique measure-

ments in CF children (Hellinckx et al., 1998 [2b]).

During exacerbations, the efficacy of inhaled bronchodilator therapy may be reduced

(Finnegan et al., 1992 [2b]; Hordvik et al., 1985). However, this concern has not been confirmed

in later studies with inhaled salbutamol (Hordvik et al., 1996) and high dose salmeterol (Hordvik

et al., 1999 [1b]).

Also short term studies of anticholinergic agents have shown benefit in some CF studies

(Weintraub et al., 1989 [2b]; Ziebach et al., 2001 [2b]; Sanchez et al., 1992 [2b]; Sanchez et al.,

1993 [2b]). However, combinations of ß-sympathomimetic and anticholinergic drugs did not

result in synergistic or additive effects in CF patients (Weintraub et al., 1989 [2b]; Ziebach et al.,

2001 [2b]; Sanchez et al., 1992 [2b]; Sanchez et al., 1993 [2b]).

Drug combinations

Inhaled drug combinations have been used in CF patients since nebulisation via a jet-nebuliser

is generally time-consuming. Inhaling a mixed drug solution for inhalation saves time.

Also CF patients sometimes refill their nebuliser with another drug without cleaning in

between courses. Another objective for inhaled drug combinations is to overcome adverse

effects of one drug by another. An example for the latter case is bronchoconstriction caused by

some antibiotics which can be overcome by co-administration of salbutamol.

The following drugs have been studied in various combinations: tobramycin, colistimethate

sodium, salbutamol, budesonide, hypertonic saline, dornase alfa, cromolyn, ipratropium

bromide and N-acetylcysteine. By mixing different drugs, the following questions arise: does

mixing affect the the physico-chemical stability of the drugs, their particle size distribution or

the therapeutic outcome? Which effects have preservatives?

Chemical stability and particle size distribution

The best way to study chemical stability is a visible judgement followed by high-performance

liquid chromatography (HPLC) analysis of the combined solution and a search in relevant

handbooks, e.g., in the Handbook on Injectable Drugs (Trissel 2006), the Drugdex database

(Anonymous, 2007) and The King Guide to Parenteral Admixtures (Anonymous 2007). Table

2 summarizes studies on chemical stability of mixtures of inhalation solutions. Stability has

been proven for combinations of cromolyn with salbutamol, ipratropium, N-acetylcysteine

and budesonide; furthermore, combinations of salbutamol with ipratropium, colistimethate

sodium, tobramycin, N-acetylcysteine and budesonide are stable as well as combinations of

ipratropium with tobramycin, N-acetylcysteine, budesonide and fenoterol or N-acetylcysteine

with fenoterol. N-acetylcysteine is inactivated by oxygen. This is prevented in the commercial

Chap

ter 2

34

product by including EDTA, which has no influence on pulmonary function but is capable of

chelating metal ions. EDTA increases the activity of azithromycin (Imamura et al., 2005) and

colistimethate sodium (Davis et al., 1971) by chelating divalent cations such as calcium. Dor-

nase alfa should not be mixed with any other drug for inhalation due to stability problems of

the protein.

Sometimes the preservative and not the pharmacologically active drug causes an incompat-

ibility. For instance benzalkonium chloride in combination with colistimethate sodium or cro-

molyn forms a hazy cloud (Kamin et al., 2006). Benzalkonium chloride is present in multi dose

formulations of salbutamol and ipratropium. Benzalkonium chloride is a pulmonary irritant and

thus preservative-free solutions are preferred.

Combination of different inhalation fluids may affect particle size distribution due to changes

in surface tension of the aerosol. Only one study addressed the aerosol characteristics after mix-

ing different drugs for inhalation (Berlinski 2006). The authors studied the mean mass aerody-

namic diameter (MMAD), respirable fraction (RF) and respirable mass (RM) of combinations of

salbutamol (albuterol) and cromolyn, ipratropium bromide, tobramycin, N-acetylcysteine and

flunisolide in continuous nebulisation and in breath actuated nebulisation. Most combinations

with salbutamol gave no difference in aerosol characteristics except cromolyn in continuous

nebulisation (MMAD decreases), ipratropium bromide in breath actuated nebulisation (RM

increased), tobramycin in breath actuated nebulisation (RF decreased), and flunisolide in

breath actuated nebulisation (RF and RM decreased). Results make clear that not only chemical

stability must be studied but also aerosol characteristics, such as is shown in Table 3.

Devices for inhaled medication

Basic knowledge on inhalation of drugs will help the prescriber in choosing the right device

for each patient. This relates to the dose of the drug, inhaler specifications and patient charac-

teristics. Table 4 presents a non exhaustive overview of inhaler specifications versus available

inhaler devices.

Physical parametersParticle mass, inhaled mass and respirable mass

The particle mass can be described as the fraction of the nominal dose that leaves the inhaler

during inhalation. The availability of an aerosol is affected by the choice of nebuliser, volume of

fill, residual volume, surface tension of the nebuliser solution, and the nebulizing flow (Coates

et al., 1997).

The inhaled mass is the fraction of a nebuliser charge that is actually inhaled by the patient.

It is a not a fully recognized quality criterion for nebulisers that affects therapeutic efficacy

(Diot et al., 2001). The inhaled mass may differ considerably between nebulisers (Faurisson et


al., 1996). It is therefore possible that the effectiveness of an inhaled drug is dependent on the

delivery system. Part of the inhaled mass may be exhaled again, resulting in a smaller lung dose

(Ilowite et al., 1987). An inhaled mass of the nebuliser charge of approximately 20-40% has

been found in a study with children and adults inhaling isotonic saline (Collis et al., 1990). In CF

children, the inhaled mass ranged from 9% to 14% using a conventional jet nebuliser and 17%

to19% using a Venturi jet nebuliser (Devadason et al., 1997). The inhaled mass can be measured

by putting an inspiration filter on the nebuliser. This estimated deposition probably differs from

real life deposition, as the latter will be influenced by the particle size distribution of the drug,

the age and tidal volume of the patient. The amount of drug on an inspiratory filter may be

comparable, while the mass median aerodynamic diameter (MMAD) of the generated aerosol

may differ significantly, possibly resulting in central or more peripheral pulmonary deposition

(Devadason et al., 1997).

The respirable mass, also called the fine particle fraction, is the portion of the inhaled mass

that is in the particle-size range expected to bypass the upper airways and deposit in the lower

airways. It is generally considered to consist of particles with an aerodynamic diameter between

1-5 μm and these dimensions are thought to result in optimal drug deposition in the peripheral

airways. Smaller particles will be exhaled while larger particles are predominantly lost because

of inertial impaction in the oropharynx. Optimal peripheral deposition has been found to occur

with a MMAD of 2-3 μm, combined with an inhalation flow rate of approximately 15-30 L/min

and the largest inhalation volume convenient for the subject (Newman et al., 1988; Brand et al.,

2005). The range of the respirable mass is related to the desired target area. Relative humidity

appears to influence the respirable mass, depending on the type of nebuliser and the drug

solution (Zhou et al., 2005).

Table 2. Chemical stability of combinations of inhalation solutions.

Drugs CROMO SAL IPRA COLI TOBRA NAC BUDE FENO HTS DORNA BENZA

CROMO - C* [1]** C [1,2] n.d.*** n.d. C [1] C [1,2] n.d. n.d. IC**** IC [1]

SAL - C [1] C [1]***** C [1] C [1] C [1,2] n.d. n.d. IC n.d.

IPRA - n.d. C [1] C [3] C [1] C [4] n.d. IC n.d.

COLI - n.d. C [4] n.d. n.d. n.d. IC IC [1]

TOBRA - n.d. n.d. n.d. n.d. IC n.d.

NAC - n.d. C [3] n.d. IC n.d.

BUDE - C [1] n.d. IC n.d.

FENO - n.d. IC n.d.

HTS - IC n.d.

DORNA - IC [5]

Drugs: CROM: cromolyn; SAL: salbutamol; IPRA: ipratropiumbromide;COLI: colistimethate sodium; TOBRA: tobramycin; NAC:N-acetylcysteine; BUDE: budesonide; FENO: fenoterol; HTS: hypertonic saline; DORN:dornase alfa; BENZ: benzalkonium chloride. *: compatible; ** [References] 1: Kamin et al., 2006; 2: McKenzie et al., 2004; 3: Lee et al., 2005; 4: Drugdex 2007; 5: Kraemer et al. 2007; ***: no data; ****: incompatible, *****: preservative free salbutamol.

Chap

ter 2

36

Tabl

e 3:

Phy

sico

-che

mic

al s

tabi

lity

of c

ombi

natio

ns o

f inh

alat

ion

solu

tions

.

Dru

gCr

omol

ynSa

lbut

amol

Ipra

trop

ium

brom

ide

Colis

timet

hate

so

dium

Tobr

amyc

inN

-ace

tyl

cyst

eine

Bude

soni

deFe

note

rol

Hyp

erto

nic

Salin

eD

orna

seal

fa

Crom

olyn

-A

Bn.

d.n.

d.B

Bn.

d.n.

d.X

Salb

utam

ol-

AB

AA

Bn.

d.n.

d.X

Ipra

trop

ium

br

omid

e-

n.d.

BB

BB

n.d.

X

Colis

timet

hate

so

dium

-n.

d.B

n.d.

n.d.

n.d.

X

Tobr

amyc

in-

n.d.

n.d.

n.d.

n.d.

X

N-a

cety

lcys

tein

e-

n.d.

Bn.

d.X

Bude

soni

de-

Bn.

d.X

Feno

tero

l-

n.d.

X

Hyp

erto

nic

Salin

e-

X

Dor

nase

alfa

-

A: m

isci

ble,

gra

de A

evi

denc

e, B

: mis

cibl

e, g

rade

B e

vide

nce,

n.d

.: no

dat

a, X

: not

mis

cibl

e. T

he h

ighe

st g

rade

of e

vide

nce

for s

afe

com

bina

tion

(A) i

s fo

r com

bina

tions

w

here

che

mic

al s

tabi

lity

has

been

pro

ven

and

aero

sol c

hara

cter

istic

s ar

e no

t alte

red.

The

sec

ond

high

est g

rade

of e

vide

nce

for s

afe

com

bina

tion

(B) i

s fo

r com

bina

tions

w

here

onl

y ch

emic

al s

tabi

lity

has

been

pro

ven

with

out s

tudy

of a

eros

ol c

hara

cter

istic

s. W

hen

chem

ical

or p

hysi

cal i

ncom

patib

ility

has

bee

n pr

oven

, dru

gs s

houl

d no

t be

mix

ed (X

).


Lung dose

The lung dose describes the amount (in mg or fraction of the nominal dose) of the drug that

enters the airways, e.g., passes the vocal cords. A lung dose can be quoted as a percentage of the

nominal dose, but also as a percentage of the particle mass or a percentage of the inhaled mass.

For example, an intrathoracic deposition of ~85% of the emitted aerosol (particle mass) was

measured of which ~77% was deposited in the peripheral lung (Griese et al., 2004). A lung dose

can be estimated by measuring the cumulative excretion of the drug during 24 h in the urine

(Touw et al., 1997; Dequin et al., 2001; Asmus et al., 2002; Aswania et al., 2004) or by radiodeposi-

tion studies (Alderson et al., 1974; Ilowite et al., 1987; Chua et al., 1994; Mukhopadhyay et al.,

1994; Devadason et al., 1997; Diot et al., 1997; Brown et al., 2001; Vanderbist et al., 2001; Byrne et

al., 2003; Pilcer et al., 2007). For newly developed inhalers, theoretical equivalent doses to cur-

rent, standard inhalation treatment have been calculated, based on in vitro testing (Le Brun et

al., 2002; Westerman et al., 2007) or using a dose escalating method (Geller et al., 2007). In older

studies lung doses ranged between ~3% to 8% using conventional jet nebulisers (Alderson et

al., 1974; Ilowite et al., 1987; Collis et al., 1990; Mukhopadhyay et al., 1994). These percentages

improved when newer breath-enhanced and breath-actuated nebulisers were used: estimated

mean lung doses between 9% and 15% were found using a PARI LC® Plus-PARI MASTER® com-

bination (Kohler et al., 2003; Kohler et al., 2004), a Ventstream®-PortaNeb® combination (Le

Brun et al., 1999), a PARI LC® Plus-PulmoAide® combination (Geller et al., 2002) and a PARI LC®

Plus or PARI LL® connected to a PARI BOY® compressor (Newman et al., 1994).

Ultrasonic nebulisers produce a mean lung dose of approximately 14%-18% (Touw et al.,

1997; Kohler et al., 2003). However, also higher mean lung doses were reported. A mean lung

dose of 22% was estimated using the PARI LC® Star-AKITA® system compared to 16% for the

PARI LC® Star-PARI MASTER® combination (Kohler et al., 2005). Comparison of a PARI LC® Plus

– PARI BOY® with an adapted aerosol delivery (AAD) system (HaloLite®) showed a lung uptake

of 20% versus 31% respectively (Byrne et al., 2003). The delivered dose to the lungs with an

Aerodose® breath actuated inhaler was ~35% versus ~9% with the PARI LC® Plus nebuliser

(Newman et al., 2001). A mean lung dose of 32% was measured using the e-Flow® vibrating

mesh device compared to 16% using the PARI LC® Plus - ProNeb® (Coates et al., 2007). A lung

dose of 63% to 73% was found with the I-neb® AAD® System (Nikander et al., 2007), expressed

a fraction of the emitted dose (particle mass).

After inhalation of 25 mg of tobramycin dry powder formulations using an Aerolizer®

capsule inhaler, lung doses of ~53% and ~34% were observed, compared to ~8% using a PARI

LC® Star – PARI TurboBOY® (Pilcer et al., 2007). In a study on lung deposition of budesonide

administered by a dry powder Turbohaler® in CF children, a deposition of 10% to 50% of the

inhaled mass was measured (Devadason, et al., 1997). Deposition data of dry-powder formula-

tions relative to liquid nebulisation have been collected (Le Brun et al., 2002; Geller et al., 2007;

Westerman et al., 2007) but these data do not provide insight into the absolute lung doses

Chap

ter 2

38

Table 4. Inhaler specifications versus available inhaler devices

Inhaler specifications Jet Ultrasonic Soft mist,vibrating mesh

Dry powder inhaler (DPI)

Pressured metered dose inhaler (pMDI)

Evidence based/based on clinical efficacy studies in CF patients.

yes yes no no yes

For all ages yes yes yes no (for age > 6y)

yes (holding chamber for infants)

General/generic use; useful for many drugs and/or disease states

yes yes yes no no

Fast (nebulisation time) no no yes, intermediate yes yes

Small size, easy to carry/portability

no no yes yes yes

Noise yes yes no no no

External power source (electricity, battery) needed

yes yes yes no no

Durability yes yes no data not applicable

not applicable

Price (initial expense) low-inter-mediate

low-inter-mediate

high low low

Large fraction of the output of the inhaler has a particle size of 1-5 micron

yes/nno* yes/no* yes yes yes

Multiple dose capacity no no no depends on design

yes

Variable payload possible yes yes yes no no

Breathing coordination required

no no no yes yes

Useful in tidal breathing/low velocity of the aerosol

yes yes yes yes and no* no; yes (holding chamber)

Dead volume yes yes yes, but generally smaller than jet/ultrasonic devices

not applicable

not applicable

High risk on bacterial contamination

yes yes yes no data/no** no data**

Preparation and cleaning is easy

no no no yes yes

Cleaning after each use (bacterial contamination and maintenance)

yes yes yes no*** no***

Periodical maintenance and/or replacement to keep up efficiency

yes yes yes no no

Adapted from Wolff &Niven, 1994 and Rau, 2002* depends on device(s) used** no data/no: unknown risk in case of a multiple use design / no risk on bacterial contamination in case of disposable design;*** manufacturers’ instructions vary


due to the chosen study design. An advantage of dry powder inhalation is a reduced loss of

aerosolized drug due to exhalation (Pilcer et al., 2007) and leakage.

Output and Output rate

Jet and ultrasonic nebulisers

Drug output from a jet or ultrasonic nebuliser is characterized by the generated particle size

distribution and the particle mass. Drug output also depends on a number of different variables

including the nebuliser type and the flow rate or (ultrasonic) frequency. A higher flow results

in a higher output rate and a larger drug output and (central) deposition (Laube et al., 2000).

Increasing the nebulising flow also results in a smaller particle size distribution (MMAD) and

a higher respirable fraction (Coates et al., 1997; de Boer et al., 2003; Westerman et al., 2008),

which may also improve peripheral lung deposition. However, the total drug output levels off

at a certain point: the optimal inspiratory flow rate has been found to be ~15 L/min to 30 L/min,

depending on the jet nebuliser used (Phipps et al., 1990; Le Brun et al., 1999; Ho et al., 2001;

Leung et al., 2004; Brand et al., 2005). Further variables are the humidity of the generating gas,

temperature, concentration, viscosity, density, the physical state (solution versus suspension)

and surface tension of the fluid during aerosolization (Phipps et al., 1990; Everard et al., 1992;

Langford et al., 1993; McCallion et al., 1995; McCallion et al., 1996; LeBrun et al., 1999; Weber

et al., 1997; Rau et al., 2002). Importantly the nebuliser configuration can affect the particle

size and the amount of aerosol inhaled (O’Riordan et al., 1997). It has been suggested that the

nebuliser configuration should be precisely specified in treatment protocols.

Each type of jet nebuliser has its own resistance, and it is therefore mandatory to test a

nebuliser-compressor combination for output and flow rate prior to starting therapy. Data on

output rates of compressors, provided by manufacturers, are often based on a configuration

without a nebuliser connected to it. With an identical driving airflow, the resulting output rate

will differ between various jet nebulisers, as each device has its own internal resistance. Simi-

larly, ultrasonic devices had a comparable or greater output than jet devices when comparing

nebuliser output with normal saline (Weber et al., 1997). Different methods have been applied

to assess drug output rate, which can be expressed in volume (ml/min) and amount of drug

(mg/min). Methods include weighing a nebuliser (Weber et al., 1997) or measuring the change

in osmolarity or concentration before and after nebulisation (Touw et al., 1996, Leung et al.,

2004). Additionally, direct measurement of the aerosol on filters can be used (Tandon et al.,

1997; Vecellio et al., 2004). Results from estimating the aerosolized volume may be misleading

when the increase in drug concentration within the nebuliser, caused by evaporation of the

solvent, is not taken into account (Touw et al., 1996). Therefore, drug output rate in mg/min is a

better parameter to describe nebuliser output than ml/min. (Le Brun et al., 1999).

Chap

ter 2

40

Other inhalers

Inhaling from a pressurized metered dose inhaler (pMDI) without a spacer/holding chamber

has a similar principle to jet nebulisation. A high external driving flow is generated which is

responsible for drug output and output rate. The patient inhales the drug using its own inspira-

tory flow, preferably with good hand-mouth coordination. Output and output rate from vibrat-

ing mesh devices also depend to some extent on the inhalation technique of the patient. If

using a passive DPI, the inspiratory flow of the patient is the only driving source for drug release

and dispersion from the inhaler. A DPI has an internal resistance which has to be overcome

by the inspiratory flow. However, this interplay may be used to guide effective regional drug

deposition in the lung.

Residual volume


The residual volume, or dead volume, is defined as the volume of solution remaining in the

nebuliser at the endpoint of nebulisation (Weber et al., 1997), which is typically in the range of

1 ml to 3 ml (Hess et al., 2000) or 38% to 61% of a drug dose (Touw et al., 1997; Kradjan et al.,

1985; Ilowite et al., 1987; Devadason et al., 1997), depending on the nebuliser used. Small doses

are especially affected. The residual volume depends on the design of the nebuliser, particularly

the extent of the internal surface, the surface tension, the viscosity of the drug solution and

the wetness of the nebuliser (McCallion et al., 1995; Ho et al., 1999; Hess et al., 2000). A higher

fill volume may reduce the relative extent of the residual volume (Hess 1996). The drug loss

may also be decreased and the output improved by tapping the nebuliser (Hess et al., 2000).

Residual volumes of ultrasonic nebulisers are often larger than for jet nebulisers (McCallion et

al., 1995).

Other inhalers

Residual volumes in novel electronically operated vibrating mesh devices are generally lower

than traditional nebulisers. The residual volume of the I-neb® AAD® system is approximately

0.1 ml (Denyer et al., 2004). The residual volumes in percentage of the nominal dose depend

on the fill volume (0.25-1.4 ml) of the devices. No clinical observations on residual volume are

available. The eFlow® Rapid has a residual volume of ~1 ml, while ~28% of the drug dose was

found in a clinical study with the eFlow® (Coates et al., 2007; Li et al., 2008). In general, powder

retention in dry powder inhalers is low, minimising drug wastage.


Particle size distribution


The particle size distribution of an aerosol is to a great extent defined by the design and

operating principle (f.e. jet or ultrasonic technique) of the nebuliser. Additionally, it depends

on the applied driving air flow or ultrasonic frequency, the inspiratory flow generated by the

patient, the temperature of the solution and the physical characteristics of the nebulised drug

(O’Riordan et al., 1997; Le Brun et al., 1999). Manufacturer’s data on particle size distribution are

frequently based on normal saline solution, and nebulising drug solutions may result in altered

particle size distribution (Clay et al., 1983; Newman et al., 1985; Newman et al., 1994; Nikander

et al., 1994; et al., Devadason 1997) and variable administration times. Due to a lower surface

tension, colistimethate sodium tends to foam during nebulisation, resulting in smaller droplets

with uncompromised biological activity (Weber et al., 1997; Diot et al., 1997; Diot et al., 2001).

Due to this foaming, administration of colistimethate sodium with an ultrasonic nebuliser is

problematic (Weber et al., 1997).

Other inhalers

Some important parameters that affect drug dispersion from a DPI are the inhaled flow rate (de

Boer et al., 2006; Pilcer et al., 2007), the inhaled volume and the internal resistance of the device

(Tiddens et al., 2006). Large fine particle fractions with a peripheral deposition of approximately

10-20% have been found using a newly developed DPI (Pilcer et al., 2007). No similar data have

been published to date on the newer electronically operated devices, like the I-neb AAD® and

eFlow®.

Polydisperse and monodisperse aerosols

The majority of drug aerosols have a polydisperse, asymmetrical distribution, according to the

log-normal law (Diot et al., 2001). Monodisperse aerosols have uniformly sized particles with a

narrow size distribution, which can be imitated with a polydisperse aerosol, as both distribu-

tions are comparable provided the width of the distribution is not too large (σg <2) (Brand et

al., 2005). Hygroscopic changes in particle size appear to be negligible if the concentration of

the monodisperse aerosol is high (Finlay et al., 1998). Generated at an optimal drug particle size

for a target region in the airways, a monodisperse aerosol might result in the most effective

treatment (Usmani et al., 2005). Clinical data on the use of monodisperse aerosols are scarce

and these aerosols are currently not used in the treatment of of CF lung disease. However, they

are studied in aerosol research to gain knowledge on aerosol particle behaviour in vivo (Heyder

et al., 2004; Brand et al., 2005).

Chap

ter 2

42

Administration time


To define the administration time of a nebulised drug, an endpoint has to be defined when

aerosolization is finished. In clinical studies, definitions of endpoints vary including the absence

of mist (Tonnesen et al., 1984; Kradjan et al., 1985; McCallion et al., 1995; Geller et al., 2003;

Kohler et al., 2003), sputtering from the nebuliser (Eisenberg et al., 1997; Weber et al., 1997), and

the absence of mist for 10 to 30 sec (Coates et al., 1997; Standaert et al., 1998; Shah et al., 1997;

Devadason et al., 1997; Kradjan et al., 1985; Newman et al., 1985; Hess et al., 1996; Coates et al.,

1998; Ho et al., 2001).

Others defined three possible end points for nebulisation: sputtering time, total time and

clinical time (Kradjan et al., 1985). Sputtering is the point when aerosolisation becomes erratic.

This point in time corresponds with an 8-fold drop in the total number of particles, read by a

laser diffraction analyzer (Reisner et al., 2001). Total time is when production of aerosol ceases

and clinical time is somewhere between sputtering and total time and approximates the point

when a patient or therapist typically stops a treatment. Delivery time by a jet nebuliser may vary

when connected to a compressor compared to hospital dry compressed air (Leung et al., 2004).

Furthermore, tapping of the nebuliser may introduce greater subjectivity in the measurement

(Kradjan et al., 1985). Delivery time by an ultrasonic nebuliser may be negatively influenced by

a higher drug concentration or higher viscosity (Weber et al., 1997). Determination of aerosol

output and residual volume depends on the definition of the end of the nebulisation and it is

therefore important that this parameter is clearly defined.

Other inhalers

The newer electronically operated nebulisers generally switch off at a point when the dose in

the reservoir is aerosolized or at a set time (i.e., after 10 min). However, it is not clear whether

a dose is always completely nebulised within this time frame, which has been shown with 6

month old e-Flow® Rapid devices (Rottier et al., 2009). A mean nebulisation time of 5 minutes

with the I-neb® AAD® System was found in a 3-month observation period (Dyche et al., 2007).

DPIs require one or several inhalation manoeuvres which generally take about 1 min to 2 min,

just as pMDIs.

Drug waste during aerosolization


Using a constant output jet nebuliser, a substantial part of a nominal drug dose may be lost

because of aerosol, generated during the non-inspiratory part of the respiratory cycle (Collis et

al., 1990; Everard et al., 1992). Since the introduction of the Venturi-nebulisers / breath-enhanced

and breath-actuated nebulisers, drug delivery has improved considerably (Knoch et al., 1994;


Newman et al., 1994; Newnham et al., 1994; Nikander et al., 1994; Devadason et al., 1997; Leung

et al., 2004). Additionally, drug loss due to exhalation contributes to drug wastage.

Other inhalers

Data on drug wastage in CF patients using pMDIs, vibrating mesh devices and DPIs are sparse.

Using the eFlow®, ~34% of the nominal dose charge was found on the expiratory filter (Coates

et al., 2007) and ~1% of the emitted drug dose (particle mass) was wasted using the I-neb AAD

System (Nikander et al., 2007).

General purpose nebuliser

Only drug-device combinations tested in clinical studies for efficacy and safety, particulary

concerning drugs with a small therapeutic window, should be used by CF patients. This is espe-

cially relevant for jet nebulisers, which often are used with various compressors, each with its

own specification. For new drugs, characterization of the drug-device combination in a clinical

study is essential for making in vitro bridging studies possible. Bridging studies may be useful in

finding an alternative nebulising device when the preferred device is unavailable.

There are many general purpose nebuliser devices available worldwide but availability dif-

fers from country to country. Therefore, pharmaceutical companies, marketing drugs for inhala-

tion, are responsible to provide evidence-based recommendations for their aerosol device in

each country in which the drug is marketed. The use of inhaled drugs in children should receive

specific attention. A pediatric investigation plan (PIP), already in use by the EMEA, on drugs

for inhalation in CF should include testing appropriate inhaler devices in combination with a

specific drug in children. Regulatory authorities should promote the use of specific drug-device

combinations in CF patients and set guidelines for bridging studies (CHMP 2006).

Bacteriological safety and performance of nebulisers over time

Cleaning of inhaler devices that are used for aerosolisation of liquids is important for bacterio-

logical safety and to ensure that the performance is not compromised. As DPIs and pMDIs are

far less influenced by hygienic threats, this paragraph focuses on wet nebulisers.

Bacteriological safety

Wet nebulisers may become a source of bacterial infection of the respiratory tract and con-

tamination of home equipment with bacterial pathogens after suboptimal cleaning proce-

dures has been documented ((Barnes et al., 1987; Pitchford et al., 1987; Hutchinson et al., 1996;

Jakobsson et al., 1997; Rosenfeld et al., 1998; Rosenfeld et al., 2001; Vassal et al., 2000). Reports

that CF patients would have acquired bacterial infections from respiratory therapy equipment

during home use however, are lacking (Saiman et al., 2004) and bacterial organisms grown

from patients’ sputum specimens and the respective devices did not correlate (Jakobsson et al.,

1997; Jakobsson et al., 2000) in contrast to another study (Rosenfeld et al., 1998). Nevertheless,

Chap

ter 2

44

early Pseudomonas aeruginosa acquisition in young children was associated with the use of

aerosolized drugs and clinic exposures (Kosorok et al., 1998). A low risk of microbial contami-

nation of CF inpatients with CF pathogens from the interior of a disposable nebuliser over a

24 h period was reported (O’Malley et al., 2007). Also, the bacterial flora from environmental

sources, for example from tap water (Kurtz et al., 1995; Lavallee et al., 1995; Saiman et al., 2004)

may contaminate a nebuliser, as well as the colonising flora of the oropharynx (Hutchinson

et al., 1996). Importantly, nebuliser devices should not be shared within CF patients as this

has been associated with the acquisition of Burkholderia cepacia complex strains (Tablan et

al., 1985). Cleaning and drying of nebulising equipment between uses decreases the risk of

acquiring pathogens, including Burkholderia cepacia complex (Hutchinson et al., 1996; Jakobs-

son et al., 1997; Jakobsson et al., 2000; Walsh et al., 2002).

Performance over time

Without cleaning or proper maintenance between runs, some nebulisers may require a lon-

ger time to complete aerosolisation, although particle size distribution and output are not

necessarily affected. Unwashed devices fail to produce an optimal aerosol after long term use

(Standaert et al., 1998). Patients may assess the functionality of their devices by visual inspec-

tion for mist production, cracks or leaks, and checking the nebulisation time (Standaert et al.,

1998). In a study using vibrating mesh-nebulisers, no modification of the membrane function

could be detected (Bakuridze et al., 2007). After daily use for six to twelve months of an eFlow®

Rapid nebuliser and a PARI LC® Plus – PARI TurboBOY® combination, changes in droplet size

distribution and a decrease in output rate were reported (Rottier et al., 2009).

Cleaning of nebuliser equipment

Soaking and rinsing with tap water (Rosenfeld et al., 2001), warm soapy water (Rosenfeld et al.,

2001; Standaert et al., 1998; Bakuridze et al., 2007) and boiling water (Saiman et al., 2004) as well

as using a dish washer (Standaert et al., 1998; Saiman et al., 2004) have been proposed to clean

nebuliser devices. However, Pseudomonas aeruginosa is only killed at temperatures of ~70°C.

Sodium hypochlorite, isopropylalcohol and ethanol (70% to 90%) are effective (Saiman et al.,

2003; Saiman et al., 2004), in contrast with the suboptimal results obtained with acetic acid and

quaternary ammonium salts (Reychler et al., 2005 ; Rutala et al., 2000; Chatburn et al., 1988).

Drying of the equipment after disinfection is important (Hutchinson et al., 1996; Jakobsson et

al., 1997; Jakobsson et al., 2000; Walsh et al., 2002). Patients should receive clear, written and oral

instructions, how to keep the bacterial contamination risk as low as possible (Jakobsson et al.,

1997; Kosorok et al., 1998; Jakobsson et al., 2000; Lester et al., 2004) to ensure patient adherence

to the cleaning of the nebulising equipment several times daily. Also, home visits by nurses

have been advocated to improve compliance (Jakobsson et al., 1997; Jakobsson et al., 2000).

As a result of improved compliance with cleaning protocols, bacterial contamination may be

prevented or decreased (Wexler et al., 1991; Hutchinson et al., 1996).


Disposable nebulisers are frequently re-used to reduce expenses and for convenience

(Pankhurst et al., 1996). But the consequences of long term use of disposable nebulisers are

poorly understood (Rosenfeld et al., 1998) and suggestions in this context differ widely. For

instance, the nebuliser should be changed every 24 h to reduce the risk of infection (Simmons et

al., 1982); old plastic tubing and atomizing chambers should be replaced at six month intervals

(Hutchinson et al., 1996), or at longer intervals up to four years (Lester et al., 2004). The interior

tubes should be dried with the aid of a compressor, attached to the nebuliser (Pankhurst et al.,

1996) and the service of equipment once a year is advised (Jakobsson et al., 1997).

Patient parametersDeposition pattern and breathing pattern

Causal relationships between the sites of drug deposition and the patients’ response have been

established in diseases of the respiratory tract (Marshall et al., 2000), but no data exist for cystic

fibrosis. Effective targeting of a given region in the lung, for instance the peripheral airways,

has been defined when >50% of the total drug deposition occurs in that region (Heyder et al.,

2004; Brand et al., 2005). Studies with radiolabelled aerosols in CF have measured regional dis-

tribution of deposited aerosol throughout the lung, but often without a direct relationship to

efficacy (Brown et al., 2001; Ilowite et al., 1987; Laube et al., 2000). Ideally, the radiolabel should

be tied to the drug in a 1:1 ratio, to visualize pulmonary drug dispersion and drug efficacy.

Many factors affect total and regional deposition. The underlying disease process is a major

determinant of the final deposition pattern (Smaldone et al., 1994). The deposition pattern is

influenced by age and by the lung function of the individual patient (Devadason et al., 1997;

Diot et al., 1997). Patients with high FEV1 values tend to have a more homogeneous distribu-

tion in the lung including peripheral airways than those with a low FEV1 (Ilowite et al., 1987).

However, the total amount of drug in the lung can be equivalent in both groups of patients. The

Table 5. Grades of recommendations

A Good scientific evidence suggests that the benefits of the clinical service substantially outweighs the potential risks. Clinicians should discuss the service with eligible patients (consistent level 1 studies).

B At least fair scientific evidence suggests that the benefits of the clinical service outweighs the potential risks. Clinicians should discuss the service with eligible patients (consistent level 2 or 3 studies or extrapolations from level 1 studies).

C At least fair scientific evidence suggests that there are benefits provided by the clinical service, but the balance between benefits and risks are too close for making general recommendations. Clinicians need not offer it unless there are individual considerations (level 4 studies or extrapolations from level 2 or 3 studies).

D At least fair scientific evidence suggests that the risks of the clinical service outweighs potential benefits. Clinicians should not routinely offer the service to asymptomatic patients (level 5 evidence or troublingly inconsistent or inconclusive studies of any level).

E Scientific evidence is lacking, of poor quality, or conflicting, such that the risk versus benefit balance cannot be assessed. Clinicians should help patients understand the uncertainty surrounding the clinical service.

Ref: Oxford Centre for Evidence-based Medicine Levels of Evidence (May 2001)

Chap

ter 2

46

wide variation of aerosol deposition in CF patients may, among other parameters, be caused by

the different breathing patterns of patients (Bennett et al., 1987; Ilowite et al., 1987; Standaert

et al., 1998; Diot et al., 1997). Indeed, controlled breathing has resulted in a narrower range of

deposition (Ilowite et al., 1987).

An optimal inspiratory flow rate (L/min) results in an optimal deposited dose. The deposition

rate is related to the patient’s respiratory rate and the type of nebuliser used. An increased

inspiratory flow rate led to an increased deposition rate with a breath-actuated nebuliser

(Halolite®), while such a relation was not seen with a breath-enhanced (PARI LC® Plus) or a

breath-enhanced/breath-actuated device (AeroEclipse®) (Leung et al., 2004). Slow and deep

inhalation using adaptive aerosol delivery devices such as the Akita® or the I-neb® AAD®

System may improve the lung deposition (Kohler et al., 2005; Denyer et al., 2004).

The optimal particle size

Drug particles are characterized by size, shape and density. Deposition properties of drug

particles are described by the aerodynamic behaviour and geometric standard deviation of the

particles (Brand et al., 2005). Sedimentation, inertial impaction and diffusion are the mechanisms

by which inhaled particles from an inhaled aerosol deposit upon airway surfaces. The overall

deposition is a result of the interaction of these mechanisms. During slow deep breathing sedi-

mentation is efficient in the peripheral airways and during rapid breathing inertial impaction is

efficient in the large tracheobronchial passages. Major determinants of the deposited fraction

and distribution of the aerosol in the lung include the inspiratory flow rate, the particle size and

the inhaled volume. By varying these factors, drug delivery to specific regions in the lung may

be accomplished, while minimizing losses in the oropharynx (Laube et al., 2000). By controlling

the inhalation flow rate and the inhaled volume, a higher peripheral deposition and reduced

dose variability can be obtained (Brand et al., 2000).

Because of these variables, the optimal particle size for peripheral deposition in CF airways

is not known and probably does not exist, as the particle size alone does not determine the

deposition result. Furthermore, penetration of particles in the CF lung is influenced by a

decrease in airway caliber, caused by airway infection, airway inflammation, by an increased

mucus layer, mucus plugging or a combination of these factors. As a possible consequence, the

optimal particle size may differ between patients according to the disease state. The impact

of infection and inflammation upon airway caliber has not been studied systematically in CF

patients, (Martonen et al., 1995). Nevertheless, two clinical trials have given insight concerning

the particle size of drugs in CF patients. Using dornase alfa, both studies showed a trend for

more improvement in pulmonary function tests with smaller particles versus larger particles,

suggesting that targeting peripheral airways with this drug may be advantagious (Shah et al.,

1997, Geller et al., 1998).


Infants and small children

It is generally agreed that many therapy strategies for CF would have the greatest benefit in

infants and young children, before the onset of irreversible lung disease. However, this patient

group is the most challenging to treat with an aerosol (Geller et al., 1997, Tiddens et al., 2007),

since infants and young children have smaller upper and lower airways, faster respiratory

rates and lower inhaled volumes. Also, infants tend to prefer nasal breathing, which may filter

the aerosol and reduce the lung dose of the drug. Some children become fussy or cry during

aerosol administration, which dramatically decreases the lung dose of the drug (Geller et al.,

1997). Even though the lung dose in infants is several-fold smaller than in older children and

adults, their lungs are also much smaller. Therefore lung deposition is proportionate to size

(Chua 1994), so adjusting the nominal dose for children is not always necessary. As an example,

Rosenberg reported serum tobramycin levels in young CF children that were similar to those

of older subjects, using the same nominal dose and delivery system (Rosenfeld 2001 J Pediatr).

Pulmonary scintigraphy is probably the most valuable method for studying deposition of drugs

in lungs, also in young children. In scintigraphic lung deposition studies in CF infants a ~10-fold

lower lung deposition has been observed compared to adults (Mallol et al., 1996). Deposition

can be reduced by bronchial obstruction and inhalation without a correct facemask. A mouth-

piece must be used as early as possible when children get older (Clavel et al., 2007).

Questions and answers

Q1 What are the advantages and disadvantages of inhalation medication in CF patients

compared to other drug administrations?

A Advantages:

• GenerationofhighdruglevelsinCFairways

• Limitedsystemictoxicityduetolowsystemicdrugabsorption

• Fastonsetofaction

• Nodruginactivationbeforereachingthetargetorgan

• Directdrugactionontargetsite

• Suitableforhometherapy

Potential disadvantages:

• Uncertaintyaboutdrugdoseatthetargetsite

• Severelyaffectedlungareasmaynotbereached

• Drugdeliverydependsoninhalationtechniqueanddeviceperformance

• Localsideeffects(e.g.,cough,airwaynarrowing,hoarseness)

• Variablesystemicdrugabsorption

• Timeconsumingdrugadminstration

• Needforeducationandtraining

• Limitedinformationondruginteractionsinthelung

Chap

ter 2

48

• Specificdrugsmayneedspecificdeliverydevices

• Pooradherence

• Potentialpollutionoftheenvironment

• Potentialdevicecontaminationandpatientinfection

• Needforhygienecontrolandmaintenanceoftheequipment

• Limitssocialfunctioning

Q2 What are the current indications for inhalation medication in CF?

A Current indications, based on level A or B clinical evidence, include:

• MaintenancetherapyforchronicPseudomonas aeruginosa infection

• EarlyeradicationtherapyforPseudomonas aeruginosa

• Improvementofairwayhydration

• Improvementofmucusclearance

• Documentedbronchialhyperreactivity

Q3 What are optimal endpoints in studies testing the efficacy of inhaled medication in

CF?

A The optimal endpoint is survival which is difficult to test in clinical trials in CF patients.

Established surrogate endpoints are:

• FEV1

• Pulmonaryexacerbations

• Qualityoflife

Potential surrogate endpoints are:

• lungfunctionparametersotherthanFEV1

• ImagingtechniquessuchasHRCT

• Exercisetolerance

• Markersoflunginflammation

• Preventionoflunginfection

Surrogate endpoints have serious limitations. They may depend on age and disease sever-

ity of the included patients as well as on the drug tested. Only few exacerbations occur

in many CF patients, especially in those with mild lung disease. The terms pulmonary

exacerbations and stable lung function lack a consistent definition. Limited information

for quality of life can be obtained in young patients. In general, clinical endpoints in

children below the age of 6 years are difficult to achieve. Independent of the endpoints

chosen, novel drugs for inhalation should be tested in CF patients, treated according to

the best standards of care.


Q4 Should the effect of inhaled medication be evaluated in individual patients?

A Drugs for inhalation which have obtained market authorization should be repeatedly

monitored in eligible CF patients with regard to potential side effects and the need for

continuous administration. The treatment with a given drug should be prescribed for the

patient in the way its efficacy has been determined.

Q5 How should the priority of different inhaled medications be established?

A Inhaled medications are difficult to compare, since there are only few comparative trials.

For drug prioritization, the level of clinical evidence and the potential benefit for the

patients should be taken into consideration.

Q6 How should the sequence of different inhaled medications during a treatment ses-

sion be determined?

A Information is limited. Mucus clearance and bronchodilator therapies should precede

antimicrobial treatment by inhalation. Drug interactions should always be considered.

Q7 Which inhaled antibiotics can be recommended?

A Tobramycin [A] and colistimethate sodium [B] preparations for inhalation are recom-

mended. Microbiological breakpoints for systemic infections (susceptible, intermediate,

resistant) do not predict the clinical efficacy of inhaled antibiotics [B].

Q8 Should inhaled antibiotics be used on alternate months or continuously?

A The decision between a continuous or alternate month antibiotic therapy strategy

depends on the drug and the clinical status of the patients. Alternate month therapy

reduces the selective antibiotic pressure and may thus reduce the development of anti-

biotic resistance, observed during continuous therapy [B]. Comparative trials between

both strategies are lacking

Q9 Can microorganisms be eradicated using aerosolized antibiotics, and if so, is mono-

therapy as effective as combination therapy with other aerosolized antibiotics or

antibiotics administered by other routes?

A Eradication of early Pseudomonas aeruginosa infection and prevention of chronic

Pseudomonas aeruginosa infection can be achieved in the majority of patients and should

be attempted with aerosolized antibiotics or aerosolized antibiotics combined with oral

antibiotics in patients with CF [A]. Different antibiotic regimens have been successful.

Due to lack of comparator studies, it is unclear whether monotherapy or aerosol/systemic

combination therapy is more effective. Re-occurrence of Pseudomonas aeruginosa in CF

airways after successful eradication should lead to new attempts to eradicate the patho-

gen by using the same or more intensive therapy strategies [B].

Chap

ter 2

50

Q10 Should inhaled antibiotic therapy be continued upon eradication of Pseudomonas

aeruginosa, and if so, for which time period?

A Clinical data supporting continuous use of antibiotics after eradication of Pseudomonas

aeruginosa in CF airways are lacking. Regardless of undetectable Pseudomonas aeruginosa

in throat swaps and negative serum antibody titers against Pseudomonas aeruginosa, the

administration of inhaled antibiotics may be continued for longer period of time, in case

Pseudomonas aeruginosa is suspected to be present in the sinuses or the small airways

[D].

Q11 Is aerosol and intravenous administration of a given antibiotic drug at the same

time superior to the administration of the drug by either route?

A For the treatment of Pseudomonas aeruginosa infections, antibiotics could be adminis-

tered by aerosol and intravenously at the same time to reach high drug concentrations in

the lung [D]. However, there is no scientific evidence to answer this question.

Q12 Are novel inhaled antibiotics for the treatment of Pseudomonas aeruginosa and other

CF-related bacterial pathogens needed?

A For the treatment of Pseudomonas aeruginosa and other CF-related bacterial patho-

gens new developments of inhaled antibiotics are urgently needed, since tobramycin

and colistimethate sodium are not sufficiently effective and are not tolerated by all CF

patients.

Q13 Should inhaled corticosteroidal drugs be used for the treatment of CF lung inflam-

mation and how effective are they?

A Inhaled corticosteroidal drugs should be considered in CF patients with clinical diagnosis

of concomitant asthma, not controlled by short-acting bronchodilators. A 2-3 month

treatment is recommended [D]. Regular anti-inflammatory therapy, regardless of symp-

toms, is not recommended [A]. Inhaled corticosteroidal drugs can be safely withdrawn

even in CF patients who have been treated for years, and who are not symptomatic. It is

recommended to reduce the dose of inhaled corticosteroidal drugs or withdraw the drug

whenever possible, particularly when clinical benefit has not been demonstrated [A].

There is no clinical evidence for the benefit of the use of inhaled corticosteroidal drugs in

aggressive bronchopulmonary aspergillosis in CF patients.

Q14 Should the use of recombinant human DNase be recommended for CF patients

regardless of age and if not, which criteria should be used for implementing this

treatment strategy?


A CF patients ≥ 6 years with mild, moderate and severe lung disease should be treated with

recombinant human DNase [A]. Evidence for efficacy is lacking in patients < 6 years of

age.

Q15 Should the use of hypertonic saline be recommended for CF patients regardless

of age and if not, which criteria should be used for implementing this treatment

strategy?

A CF patients ≥ 6 years should be treated with hypertonic saline for short-term use to

improve lung function [A] and for long-term treatment to improve lung function, reduce

exacerbations [B] and improve the quality of life. Evidence for efficacy is lacking in patients

< 6 years of age. Cinical trials comparing the established dose of 7% saline in 4 mL, twice

daily, to lower concentrations or less frequent dosing have not been performed. The

therapeutic effect of hypertonic saline differs from the mode of action of recombinant

human DNase and therefore the two drugs cannot replace each other.

Q16 Which CF patients should be treated with bronchodilators?

A For CF patients with persistent wheeze or exercise-induced bronchospasm, potentially

suffering from CF asthma, who experience symptomatic relief from this treatment,

short-acting bronchodilators should be used [D]. A significant response to treatment

may support the use of bronchodilators, but responses may be quite variable. Evidence

for benefit from regular use of bronchodilators prior to physiotherapy is lacking.

Evidence for a sustained benefit on mucociliary clearance is lacking. Bronchodilators

may be necessary before inhaled antibiotics and hypertonic saline are administered

[B]. Long-acting bronchodilators should be used in CF patients with asthma who can-

not be controlled with short-acting bronchodilators and inhaled corticosteroids alone

[A]. There is insufficient evidence to support the use of short-acting anticholinergic

agents.

Q17 Should systemic absorption of marketed inhaled antibiotics be routinely measured

and if so, when and how often should drug levels be measured?

A Aminoglycosides: measurement of serum trough levels may be performed on a regular

basis in CF patients with reduced renal function, and in CF patients with normal renal

function but at risk for nephrotoxicity (e.g., potential nephrotoxic co-medication such

as non-steroidal anti-inflammatory drugs and immunosuppressants). Colistimethate

sodium cannot be measured in routine laboratories.

Q18 What tests are required before a new inhalation device is introduced for use in CF

patients?

Chap

ter 2

52

A The introduction of a new inhalation device must be accompanied by clinical compara-

tive studies, including the medications recommended by the manufacturer. The design

of such studies is dependent on the drug.

Q19 Can drugs for inhalation be mixed in one device?

A It is not recommended to mix medications for inhalation prior to their use. If mixing is

needed, the mixture should have been tested for chemical and physical compatibility.

Q20 How should a new medication for inhalation be tested before its use in CF patients.

A Safety studies in animals and phase I, II and III studies should be performed according

to regulatory requirements. New medications for inhalation should be tested for phar-

macokinetics, including systemic absorption, safety and efficacy, in young CF children,

in adequate numbers, and in CF adults. Studies should always be performed using

combination(s) with predefined device(s). As drug concentrations are highly variable

in sputum specimens, they are likely to be the wrong parameter for pharmacokinetics,

peripheral drug deposition and efficacy.


References

Adeboyeku D, Scott S, Hodson ME. Open follow-up study of tobramycin nebuliser solution and colistin in patients with cystic fibrosis. J Cyst Fibros 2006; 5:261-263.

Alderson PO, Secker-Walker RH, Stominger DB, Markham J, Hill RL. Pulmonary deposition of aerosols in children with cystic fibrosis. J Pediatr 1974; 84:479-484.

Allen SD, Sorenson KN, Nejdl MJ, Durrant C, Proffit RT. Prophylactic efficacy of aerosolized liposomal (AmBi-sone) and non-liposomal (Fungizone) amphotericin B in murine pulmonary aspergillosis. J Antimicrob Chemother 1994; 34:1001-1013.

Alothman GA, Ho B, Alsaadi MM, et al. Bronchial constriction and inhaled colistin in cystic fibrosis. Chest 2005; 127:522-529.

Anonymous. Drugdex database: http://www.thomsonhc.com, consulted on 10th December 2007a.Anonymous. The King guide to parenteral admixtures, Loose leaf version, 2007b.Armstrong DS, Grimwood K, Carlin JB et al. Lower airway inflammation in infants and young children with

cystic fibrosis. Am J Respir Crit Care Med 1997; 156:1197-1204.Asmus MJ, Stewart BA, Milavetz G, et al. Tobramycin as a pharmacologic tracer to compare airway deposi-

tion from nebulizers. Pharmacotherapy 2002; 22:557-563.Aswania O, Ritson, S., Iqbal, SM., Bhatt, J., Rigby AS, A.S., Everard, M.L. Intra-subject variability in lung dose in

healthy volunteers using five conventional portable inhalers. J Aerosol Med. 2004; 17:231-238.Bakuridze L, Andrieu V, Dupont C, Dubus JC. Does repeated disinfection of the e-Flow rapid nebulizer affect

in vitro performance? J Cyst Fibros 2007; 6:309-310.Balfour-Lynn IM, Klein NJ, Dinwiddie R. Randomised controlled trial of inhaled corticosteroids (fluticasone

propionate) in cystic fibrosis. Arch Dis Child 1997; 77:124-130.Balfour-Lynn IM, Walters S, Dezateux C. Inhaled corticosteroids for cystic fibrosis. Cochrane Database Syst

Rev 2000; 1:Art. No: CD001915. DOI: 10.1002/14651858.CD001915Balfour-Lynn IM, Multicenter randomised controlled trial of withdrawal of inhaled corticosteroids in cystic

fibrosis. Am J Respir Crit Care Med 2006; 173:1356-1362.Ballmann M, von der Hardt H. Hypertonic saline and recombinant human DNase: a randomised crossover

pilot study in patients with cystic fibrosis. J Cyst Fibros 2002; 1:35–37.Bargon J, Viel K, Dauletbaev N, Wiewrodt R, Buhl R. Short-term effects of regular salmeterol treatment on

adult cystic fibrosis patients. Eur Respir J 1997; 10:2307-2311.Barnes KL, Clifford R, Holgate ST, Murphy D, Comber P, Bell E. Bacterial contamination of home nebuliser. Br

Med J (Clin Res Ed) 1987; 295:812.Bennett WD, Smaldone GC. Human variation in the peripheral air-space deposition of inhaled particles. J

Appl Physiol 1987; 62:1603-1610.Berlinski A, Waldrep JC. Nebulized drug admixtures: effect on aerosol characteristics and albuterol output.

J Aerosol Med 2006; 19: 484-490.Birke FW, Meade CJ, Anderskewitz R, Speck GA, Jennewein HM. In vitro and in vivo pharmacological char-

acterization of BIIL 284, a novel and potent leukotriene B(4) receptor antagonist. J Pharmacol Exp Ther 2001; 297:458-466.

Bisgaard H, Pedersen SS, Nielsen KG et al. Controlled trial of inhaled budesonide in patients with cystic fibrosis and chronic bronchopulmonary Pseudomonas aeruginosa infection. Am J Respir Crit Care Med 1997; 156:1190-1196.

Brand P, Friemel I, Meyer T, Schulz H, Heyder J, Haussinger K. Total deposition of therapeutic particles during spontaneous and controlled inhalations. J Pharm Sci 2000; 89:724-731.

Brand P, Meyer T, Haussermann S, et al. Optimum peripheral drug deposition in patients with cystic fibrosis. J Aerosol Med 2005; 18:45-54.

Brennan S. Revisiting α1-antitrypsin therapy in cystic fibrosis: can it still offer promise? Eur Respir J 2007; 29:229-230.

Chap

ter 2

54

Brown JS, Zeman KL, Bennett WD. Regional deposition of coarse particles and ventilation distribution in healthy subjects and patients with cystic fibrosis. J Aerosol Med 2001; 14:443-454.

Byrne NM, Keavey PM, Perry JD, Gould FK, Spencer DA. Comparison of lung deposition of colomycin using the HaloLite and the Pari LC Plus nebulisers in patients with cystic fibrosis. Arch Dis Child 2003; 88:715-718.

Chatburn RL, Kallstrom TJ, Bajaksouzian S. A comparison of acetic acid with a quaternary ammonium compound for disinfection of hand-held nebulizers. Respir Care 1988; 33:179-187.

Chua HL, Collis GG, Newbury AM et al. The influence of age on aerosol deposition in children with cystic fibrosis. Eur Respir J 1994; 7:2185-2191.

Clavel A, Boulaméry A, Bosdure E et al. Nebulisers comparison with inhaled tobramycin in young children with cystic fibrosis. J Cyst Fibros. 2007; 6:137-143.

Clay MM, Pavia D, Newman SP, Lennard-Jones T, Clarke SW. Assessment of jet nebulisers for lung aerosol therapy. Lancet 1983; 2:592-594.

Collis GG, Cole CH, Le Souef PN. Dilution of nebulised aerosols by air entrainment in children. Lancet 1990; 336:341-343.

Committee for medicinal products for human use (CHMP). Guideline on the pharmaceutical quality of inhalation and nasal products. EMEA/CHMP/QWP/49313/2005 Corr, 2006.

Conley J, Yang HM, Wilson T, et al. Ciprofloxacin: aerosol characterisation and efficacy against Francisella tularensis infection in mice. Antimicrob Agents Chemother 1997; 41:1288-1292.

Coates AL, Allen PD, MacNeish CF, Ho SL, Lands LC. Effect of size and disease on estimated deposition of drugs administered using jet nebulization in children with cystic fibrosis. Chest 2001; 119:1123-1130.

Coates AL, Green M, Leung K et al. The challenges of quantitative measurement of lung deposition using 99mTc-DTPA from delivery systems with very different delivery times. J Aerosol Med 2007; 20:320-330.

Coates AL, MacNeish CF, Lands LC, Meisner D, Kelemen S, Vadas EB. A comparison of the availability of tobramycin for inhalation from vented vs unvented nebulizers. Chest 1998; 113:951-956.

Coates AL, MacNeish CF, Meisner D, Kelemen S, Thibert R, MacDonald J. The choice of jet nebulizer, nebulizing flow, and addition of albuterol affects the output of tobramycin aerosols. Chest 1997; 111:1206- 1212.

Cunningham S, Prasad A, Collyer L, Carr S, Balfour-Lynn IM, Wallis C. Bronchoconstriction following nebu-lised colistin in cystic fibrosis. Arch Dis Child. 2001; 84:432-433.

Cystic Fibrosis Foundation. Cystic Fibrosis Foundation Patient Registry 2005 Annual Data Report. Bethesda: Cystic Fibrosis Foundation, 2006.

Dauletbaev N, Viel K, Behr J, et al. Effects of short-term inhaled fluticasone on oxidative burst of sputum cells in cystic fibrosis. Eur Respir J 1999;14:1150-1155.

Davis SD, Iannetta A, Wedgwood RJ. Activity of colistin against Pseudomonas aeruginosa . Inhibition by calcium. J Infect Dis 1971; 124:610-612.

De Boeck K, De Baets F, Malfroot A, Desager K, Mouchet F, Proesmans M. Do inhaled corticosteroids impair long-term growth in prepubertal cystic fibrosis patients? Eur J Pediatr 2007; 166:23-28.

de Boer AH, Hagedoorn P, Frijlink HW. The choice of a compressor for the aerosolisation of tobramycin (TOBI) with the PARI LC PLUS reusable nebuliser. Int J Pharm 2003; 268:59-69.

de Boer AH, Hagedoorn P, Westerman EM, Le Brun PP, Heijerman HG, Frijlink HW. Design and in vitro per-formance testing of multiple air classifier technology in a new disposable inhaler concept (Twincer) for high powder doses. Eur J Pharm Sci 2006; 28:171-178.

Denyer J, Nikander K, Smith NJ. Adaptive aerosol delivery (AAD®) technology. Expert Opin Drug Deliv 2004; 1:165-176.

Dequin PF, Faurisson F, Lemarie E et al. Urinary excretion reflects lung deposition of aminoglycoside aerosols in cystic fibrosis. Eur Respir J 2001; 18:316-322.

Deterding R, Retsch-Bogart G, Milgram L et al. Safety and tolerability of denufosol tetrasodium inhalation solution, a novel P2Y2 receptor agonist: results of a phase 1/phase 2 multicenter study in mild to moder-ate cystic fibrosis Pediatr Pulmonol 2005; 39:339-348.


Devadason SG, Everard ML, Linto JM, Le Souef PN. Comparison of drug delivery from conventional versus “Venturi” nebulizers. Eur Respir J 1997; 10:2479-2483.

Devadason SG, Everard ML, MacEarlan C etEt al. Lung deposition from the Turbuhaler in children with cystic fibrosis. Eur Respir J 1997; 10:2023-2028.

Diot P, Palmer LB, Smaldone A, DeCelie-Germana J, Grimson R, Smaldone GC. RhDNase I aerosol deposition and related factors in cystic fibrosis. Am J Respir Crit Care Med 1997a; 156:1662-1668.

Diot P, Gagnadoux F, Martin C et al.. Nebulization and anti-Pseudomonas aeruginosa activity of colistin. Eur Respir J 1997b; 10:1995-1998.

Diot P, Dequin PF, Rivoire B et al. Aerosols and anti-infectious agents. J Aerosol Med 2001; 14:55-64.Dodd JD, Barry SC, Daly LE, Gallagher CG. Inhaled beta-agonists improve lung function but not maximal

exercise capacity in cystic fibrosis. J Cystic Fibros 2005; 4:101-105.Donaldson SH, Bennett WD, Zeman KL, Knowles MR, Tarran R, Boucher RC. Mucus clearance and lung func-

tion in cystic fibrosis with hypertonic saline. N Engl J Med 2006; 354:241–250.Döring G, Conway SP, Heijerman HGM et al. for the Consensus Committee. Antibiotic therapy against

Pseudomonas aeruginosa in cystic fibrosis: a European consensus. Eur Respir J 2000; 16:749-767.Döring G, Hoiby N, for the Consensus Study Group. Early intervention and prevention of lung disease in

cystic fibrosis: a European consensus. J Cyst Fibros 2004; 3:67-91.Döring G, Ratjen F. Immunology of cystic fibrosis. In: Cystic Fibrosis, ME Hodson, D Geddes, A Bush (eds).

Arnold Hammer, London 2007; 69-80.Drutz D, Shaffer C, LaCroix K, et al. Uridine 50-triphosphate (UTP) regulates mucociliary clearance via puri-

nergic receptor activation. Drug Dev Res 1996; 37:185Dyche T, Prince IR, Nikander K. Use of I-neb® AAD® patient logging system data to identify aerosol treat-

ment issues in patients with cystic fibrosis. ATS Conference AJRCCM 2007; A782.Eber E, Oberwaldner B, Zach M. Airway obstruction and airway wall instability in cystic fibrosis: The isolated

and combined effects of theophylline and sympathomimetics. Pediatr Pulmonol 1988; 4:205-212.Eggleston PA, Rosenstein BJ, Stackhouse CM, Alexander MF. Airway hyperreactivity in cystic fibrosis. Chest

1988; 94:360-365.Eggleston PA, Rosenstein BJ, Stackhouse CM, Mellits D, Baumgardner RA. A controlled trial of long-term

bronchodilator therapy in cystic fibrosis. Chest 1991; 99:1088-1092.Eisenberg J, Pepe M, Williams-Warren J et al. A comparison of peak sputum tobramycin concentration in

patients with cystic fibrosis using jet and ultrasonic nebulizer systems. Aerosolized Tobramycin Study Group. Chest 1997; 111: 955-962.

Elkins MR, Robinson M, Rose BR et al. A controlled trial of long-term inhaled hypertonic saline in patients with cystic fibrosis. N Engl J Med. 2006; 354:229-240.

Eng PA, Morton J, Douglass JA, Riedler J, Wilson J, Robertson CF. Short-term efficacy of ultrasonically nebu-lized hypertonic saline in cystic fibrosis. Pediatr Pulmonol 1996; 21:77–83.

Equi A, Balfour-Lynn IM, Bush A, Rosenthal M. Long term azithromycin in children with cystic fibrosis: a randomized, placebo-controlled crossover trial. Lancet 2002; 360:978-84.

Etherington C, Bosomworth M, Clifton I, et al. Measurement of N-acetyl-b-D-glucosaminidase in adult patients with cystic fibrosis: before, during and after treatment with intravenous antibiotics. J Cyst Fibros 2007; 6:67-73.

Everard ML, Clark AR, Milner AD. Drug delivery from jet nebulisers. Arch Dis Child 1992; 67:586-591.Faurisson F, Dessanges JF, Grimfeld A et al. Comparative study of the performance and ergonomics of

nebulizers in cystic fibrosis. Rev Mal Respir 1996; 13:155-162.FinlayWH, Wong JP. Regional lung deposition of nebulised liposome-encapsulated ciprofloxacin. Int J

Pharm 1998;167:121-127.Finlay WH. Estimating the type of hygroscopic behavior exhibited by aqueous droplets. J Aerosol Med 1998;

11:221-229.Finnegan MJ, Hughes DV, Hodson ME. Comparison of nebulised and intravenous terbutaline during exacer-

bations of pulmonary infection in patients with cystic fibrosis. Eur Respir J 1992; 5:1089-1091.

Chap

ter 2

56

Fitzgeorge RB, Baskerville A, Featherstone AS,. Treatment of experimental Legionnaires disease by aerosol administration of rifampicin, ciprofloxacin, and erythromycin. Lancet 1986; 8479:502-503.

Frederiksen B, Koch C, Høiby N. Antibiotic treatment of initial colonization with Pseudomonas aeruginosa postpones chronic infection and prevents deterioration of pulmonary function in cystic fibrosis. Pediatr Pulmonol. 1997; 23:330-335.

Frederiksen B, Pressler T, Hansen A, Koch C, Høiby N. Effect of aerosolized rhDNase (Pulmozyme®) on pul-monary colonization in patients with cystic fibrosis. Acta Paediatr 2006; 95:1070-1074.

Fuchs HJ, Borowitz DS, Christiansen DH Et al. Effect of aerosolized recombinant human DNase on exacerba-tions of respiratory symptoms and on pulmonary function in patients with cystic fibrosis. N Engl J Med 1994; 331:637-642.


Geller DE, Rosenfeld M, Waltz DA, Wilmott RW. Efficiency of pulmonary administration of tobramycin solu-tion for inhalation in cystic fibrosis using an improved drug delivery system. Chest 2003; 123:28-36.


Gibson RL, Emerson J, McNamara S et al. Significant microbiological effect of inhaled tobramycin in young children with cystic fibrosis. Am J Respir Crit Care Med 2003; 167:841-849.

Gibson RL, Retsch-Bogart GZ, Oermann C et al. Microbiology, safety, and pharmacokinetics of aztreonam lysinate for inhalation in patients with cystic fibrosis. Pediatr Pulmonol 2006; 41:656-665.

Gibson RL, Emerson J, Mayer-Hamblett N et al. Duration of treatment effect after tobramycin solution for inhalation in young children with cystic fibrosis. Pediatr Pulmonol 2007; 42:610-623.

Gilbert BE. Liposomal aerosols in the management of pulmonary infections. J Aerosol Med 1996; 1:111-122.

Gilbert BE, Proffitt RT. Aerosolized AmBisone treatment of pulmonary Cryptococcus neoformans infection in mice. J Aerosol Med 1996;9:263-276.

Govan JRW, Glass S. The microbiology and therapy of cystic fibrosis lung infections. Rev Med Microbiol 1990; 1:19-28.

Grasemann H, Stehling F, Brunar H et al. Inhalation of Moli1901 in patients with cystic fibrosis. Chest. 2007; 131:1461-1466.

Griese M, Ramakers J, Krasselt A Et al. Improvement of alveolar glutathione and lung function but not oxida-tive state in cystic fibrosis. Am J Respir Crit Care Med 2004; 169:822-888.

Griese M, Latzin P, Kappler M et al. α1-antitrypsin inhalation reduces airway inflammation in cystic fibrosis patients. Eur Respir J 2007; 29:240-250.

Griese M, Kappler M, Gaggar A, Hartl D. Inhibition of airway proteases in cystic fibrosis lung disease. Eur Respir J 2008; 32:783–795.

Griffith DC, Rock J, Hansen C et al. Pharmacokinetics and safety of MP-376 (levofloxacin solution for inhala-tion) in normal healthy volunteers and cystic fibrosis patients. Pediatr Pulmonol 2007; 42 (S30): 303

Halfhide C, Evans HJ, Couriel J. Inhaled bronchodilators for cystic fibrosis. Cochrane Database Syst Rev. 2005; 4:CD003428

Hellinckx J, De Boeck K, Demedts M. No paradoxical bronchodilator response with forced oscillation tech-nique in children with cystic fibrosis. Chest 1998; 113:55-59.

van Haren EHJ, Lammers JWJ, Festen J, Heijerman HGM, Groot CAR, van Herwaarden CLA. The effects of the inhaled corticosteroid budesonide on lung function and bronchial hyperresponsiveness in adult patients with cystic fibrosis. Resp Med 1995; 89:209-214.

Hess D, Fisher D, Williams P, Pooler S, Kacmarek RM. Medication nebulizer performance. Effects of diluent volume, nebulizer flow, and nebulizer brand. Chest 1996; 110:498-505.

Hess DR. Nebulizers: principles and performance. Respir Care 2000; 45:609-622.Heyder J. Deposition of inhaled particles in the human respiratory tract and consequences for regional

targeting in respiratory drug delivery. Proc Am Thorac Soc 2004; 1:315-320.


Ho BL, Ho SL, Dupuis A, Corey M, Coates AL. Evaluation of bronchial constriction in children with cystic fibrosis after inhaling two different preparations of tobramycin. Chest 2002; 122:930-934.

Ho SL, Coates AL. Effect of dead volume on the efficiency and the cost to deliver medications in cystic fibrosis with four disposable nebulizers. Can Respir J 1999; 6:253-260.

Ho SL, Kwong WTJ, O’Drowsky L, Coates AL. Evaluation of four breath-enhanced nebulizers for home use. J Aerosol Med 2001; 7:467-475.

Hodson ME, Gallagher CG, Govan JRW. A randomised clinical trial of nebulised tobramycin or colistin in cystic fibrosis. Eur Respir J 2002; 20:658–664.

Hodson ME, Geddes D, Bush A, eds. Cystic Fibrosis 3rd ed. Part 2. London: Hodder Arnold, 2007.Hogardt, M., Schmoldt S, Henke C, Bader L, Heesemann J. Stage-specific adaptation of hypermutable

Pseudomonas aeruginosa isolates during chronic pulmonary infection in patients with cystic fibrosis. J Infect Dis 2007; 195:70-80.

Hoiby N, Frederiksen B. Microbiology of cystic fibrosis. In: Hodson ME, Geddes D, eds. Cystic Fibrosis. Lon-don: Arnold, 2000; 83-108.

Hordvik NL, Konig P, Morris D, Kreutz C, Barbero GJ. A longitudinal study of bronchodilator responsiveness in cystic fibrosis. Am Rev Respir Dis 1985; 131:889-893.

Hordvik NL, Sammut PH, Judy CG, Strizek SJ, Colombo JL. The effects of albuterol on the lung function of hospitalised patients with cystic fibrosis. Am J Respir Crit Care Med 1996; 154:156-160.

Hordvik NL, Sammut PH, Judy CG, Colombo JL. Effects of standard and high doses of salmeterol on lung function of hospitalised patients with cystic fibrosis. Pediatr Pulmonol 1999; 27:43-53.

Hordvik NL, Sammut PH, Judy CG, Colombo JL. Effectiveness and tolerability of high-dose salmeterol in cystic fibrosis. Pediatr Pulmonol 2002; 34:287-296.

Hughes T, Clifton I, Etherington C, et al. A retrospective study to assess the response to nebulised salbutamol during a period of stability in adult patients with cystic fibrosis. J Cyst Fibros 2006; 5(Suppl1):S44.

Hutchinson GR, Parker S, Pryor JA, et al. Home-use nebulizers: a potential primary source of Burkholderia cepacia and other colistin-resistant, gram-negative bacteria in patients with cystic fibrosis. J Clin Micro-biol 1996; 34:584-587.

Imamura Y, Higashiyama Y, Tomono K, et al. Azithromycin exhibits bactericidal effects on Pseudomonas aeruginosa through interaction with the outer membrane Antimicrob Ag Chemother 2005; 49:1377-1380.

Ilowite JS, Gorvoy JD, Smaldone GC. Quantitative deposition of aerosolized gentamicin in cystic fibrosis. Am Rev Respir Dis 1987; 136:1445-1449.

Jakobsson B, Hjelte L, Nystrom B. Low level of bacterial contamination of mist tents used in home treatment of cystic fibrosis patients. J Hosp Infect 2000; 44:37-41.

Jakobsson BM, Onnered AB, Hjelte L, Nystrom B. Low bacterial contamination of nebulizers in home treat-ment of cystic fibrosis patients. J Hosp Infect 1997; 36:201-207.

Jensen T, Pedersen SS, Garne S, Heilmann C, Høiby N, Koch C. Colistin inhalation therapy in cystic fibrosis patients with chronic Pseudomonas aeruginosa lung infection. J Antimicrob Chemother 1987; 19:831-838.

Jones AP, Wallis CE. Recombinant human deoxyribonuclease for cystic fibrosis (Cochrane Review). In::The Cochrane Library, Issue 3, 2003. Oxford: Update Software.

Kamin W, Schwabe A, Krämer I. Inhalation solutions: which ones are allowed to be mixed? Physico-chemical compatibility of drug solutions in nebulizers. J Cyst Fibros 2006; 5:205-213.

Keller M, Stempfle P, Reul K, Waldner R, Lintz FC, Opolski S. High efficiency pulmonary drug delivery of a novel formulation of aztreonam by the eFlow™. Poster 030. Congress of the International Society for Aerosols in Medicine. Baltimore, 2003

Kellerman D, Rossi Mospan A, Engels J, Schaberg A, Gorden J, Smiley L. Denufosol: a review of studies with inhaled P2Y2 agonists that led to Phase 3. Pulm Pharmacol Ther 2008; 21: 600-607.

Keir GJ, Serisier DS, Bowler SD. Bronchodilator effects of tiotropium in stable cystic fibrosis – a placebo-controlled study. www.thoracic.org.au/asm2007 pa07.pdf

Chap

ter 2

58

Khan TZ, Wagener JS, Bost T, Martinez J, Accurso FJ, Riches DWH. Early pulmonary inflammation in infants with cystic fibrosis. Am J Respir Crit Care Med 1995; 151:1075-1082.

Knechtel SA, Klepser ME. Safety of aerosolized amphotericin B. Expert Opin Drug Saf 2007; 6:523-532.Knoch M, Wunderlich E, Geldner S. A nebulizer system for highly reproducible aerosol delivery. J Aerosol

Med 1994; 7:229-237.Kohler E, Sollich V, Schuster-Wonka R, Huhnerbein J. Lung deposition in cystic fibrosis patients using an

ultrasonic or a jet nebulizer. J Aerosol Med 2003; 16:37-46.Kohler E, Sollich V, Schuster-Wonka R, Huhnerbein J, Jorch G. Does wearing a nose clip during inhalation

improve lung deposition? J Aerosol Med 2004; 17:116-122.Kohler E, Sollich V, Schuster-Wonka R, Jorch G. Lung deposition after electronically breath-controlled inha-

lation and manually triggered conventional inhalation in cystic fibrosis patients. J Aerosol Med 2005; 18:386-95.

Konig P, Gayer D, Barbero GJ, Shaffer J. Short-term and long-term effects of albuterol aerosol therapy in cystic fibrosis: a preliminary report. Pediatr Pulmonol 1995; 20:205-214.

Konig P, Poehler J, Barbero GJ. A placebo-controlled, double-blind trial of the long-term effects of albuterol administration in patients with cystic fibrosis. Pediatr Pulmonol 1998; 25:32-36.

Konstan MW, Berger M. Infection and inflammation of the lung in cystic fibrosis. In: Davis PB, ed. Cystic Fibrosis. New York: Marcel Dekker Inc, 1993

Konstan MW, Hilliard KA, Norvell TM, Berger M. Bronchoalveolar lavage findings in cystic fibrosis patients with stable, clinically mild lung disease suggest ongoing infection and inflammation Am J Respir Crit Care Med 1994; 150:448-454.

Konstan MW, Byard PJ, Hoppel CL, Davis PB. Effect of high-dose ibuprofen in patients with cystic fibrosis. N Engl J Med, 1995; 332:848-854.

Konstan MW, Schluchter MD, Xue W, Davis PB. Clinical use of Ibuprofen is associated with slower FEV1 decline in children with cystic fibrosis. Am J Respir Crit Care Med. 2007; 176:1084-1089.

Kosorok MR, Jalaluddin M, Farrell PM et al. Comprehensive analysis of risk factors for acquisition of Pseudomonas aeruginosa in young children with cystic fibrosis. Pediatr Pulmonol 1998; 26:81-88.

Kosorok MR, Zeng L, West SE, Rock MJ, Splaingard ML, Collins J, Farrell PM. Acceleration of lung disease in children with cystic fibrosis after Pseudomonas aeruginosa acquisition. Pediatr Pulmonol 2001; 32:277-287.

Kradjan WA, Lakshminarayan S. Efficiency of air compressor-driven nebulizers. Chest 1985; 87:512-516.Kraemer I, Schwabe A, Lichtinghagen R, Kamin W. Physiochemical compatibility of nebulizible drug mix-

tures containing dornase alfa and ipatropium and/or albuterol. Pharmazie 2007; 62:760-766.Kurtz MJ, Van Zandt K, Burns JL. Comparison study of home catheter cleaning methods. Rehabil Nurs 1995;

20:212-214.Langford SA, Allen MB. Salbutamol output from two jet nebulizers. Respir Med 1993; 87:99-103.Laube BL, Jashnani R, Dalby RN, Zeitlin PL. Targeting aerosol deposition in patients with cystic fibrosis:

effects of alterations in particle size and inspiratory flow rate. Chest 2000; 118:1069-1076.Lavallee DJ, Lapierre NM, Henwood PK et al. Catheter cleaning for re-use in intermittent catheterization:

new light on an old problem. SCI Nurs 1995; 12:10-12.Le Brun PP, de Boer AH, Gjaltema D, Hagedoorn P, Heijerman HG, Frijlink HW. Inhalation of tobramycin in

cystic fibrosis. Part 2: optimization of the tobramycin solution for a jet and an ultrasonic nebulizer. Int J Pharm 1999; 189:215-225.

Le Brun PP, de Boer AH, Mannes GP et al. Dry powder inhalation of antibiotics in cystic fibrosis therapy: part 2. Inhalation of a novel colistin dry powder formulation: a feasibility study in healthy volunteers and patients. Eur J Pharm Biopharm 2002; 54:25-32.

Le Brun PP, Vinks AA, Touw DJ et al. Can tobramycin inhalation be improved with a jet nebulizer? Ther Drug Monit 1999; 21:618-624.

Lee TY, Chen CM, Lee CN, Chiang YC, Chen HY. Compatibility and osmolality of inhaled N-acetylcysteine nebulizing solution with fenoterol and ipratropium. Am J Health Syst Pharm 2005; 62:828-833.


Lester MK, Flume PA, Gray SL, Anderson D, Bowman CM. Nebulizer use and maintenance by cystic fibrosis patients: a survey study. Respir Care 2004; 49:1504-1508.

Leung K, Louca E, Coates AL. Comparison of breath-enhanced to breath-actuated nebulizers for rate, consistency, and efficiency. Chest 2004; 126:1619-1627.

Li Z, Zhang Y, Wurtz BS et al Characterization of nebulized liposomal amikacin (Arikace™) as a function of droplet size. J Aerosol Med ahead of print. doi:10.1089/jamp.2008.0686.

Liou TG, Adler FR, FitzSimmons SC, Cahill BC, Hibbs JR, Marshall BC: Predictive 5-year survivorship model of cystic fibrosis. Am J Epidemiol 2001; 153:345-352.

Lowry CM, Marty FM, Vargas SO et al. Safety of aerosolized liposomal versus deoxycholate amphotericin B formulations for prevention of invasive fungal infections following lung transplantation: a retrospective study. Transpl Infect Dis 2007; 9:121-125.

Mallol J, Rattray S, Walker G, Cook D, Robertson CF. Aerosol deposition in infants with cystic fibrosis. Pediatr Pulmonol. 1996; 21:276-281.

Marshall BG, Wangoo A, Harrison LI, Young DB, Shaw RJ. Tumour necrosis factor-alpha production in human alveolar macrophages: modulation by inhaled corticosteroid. Eur Respir J 2000; 15:764-770.

Marshall LM, Francis PW, Khafagi FA. Aerosol deposition in cystic fibrosis using an aerosol conservation device and a conventional jet nebulizer. J Paediatr Child Health 1994; 30:65-67.

Martin SL, Downey D, Bilton D et al. Safety and efficacy of recombinant alpha1-antitrypsin therapy in cystic fibrosis. Pediatr Pulmonol 2006; 41:177-183.

Martonen T, Katz I, Cress W. Aerosol deposition as a function of airway disease: cystic fibrosis. Pharm Res 1995; 12:96-102.

Matsui H, Grubb BR, Tarran R et al. Evidence for periciliary liquid layer depletion, not abnormal ion composi-tion, in the pathogenesis of cystic fibrosis airways disease. Cell 1998; 95:1005-1010.

McCallion ON, Taylor KM, Thomas M, Taylor AJ. Nebulization of fluids of different physicochemical proper-ties with air-jet and ultrasonic nebulizers. Pharm Res 1995; 12:1682-1688.

McCallion ONM, Taylor KMG, Bridges PA, Thomas M, Taylor AJ. Jet nebulisers for pulmonary drug delivery. Int J Pharm 1996; 130:1-14.

McCoy K, Hamilton S, Johnson C. Effects of 12-week administration of dornase alfa in patients with advanced cystic fibrosis lung disease. Chest 1996; 110:889–895.

McCoy KS, Quittner AL, Oermann CM, Gibson RL, Retsch-Bogart GZ, Montgomery AB. Inhaled aztreonam lysine for chronic airway Pseudomonas aeruginosa in cystic fibrosis. Am J Respir Crit Care Med 2008; 178:921-8.

McElvaney NG, Hubbard RC, Birrer P et al. Aerosol α1-antitrypsin treatment for cystic fibrosis. Lancet 1991; 337:392-394.

McElvaney NG, Doujaiji B, Moan MJ, Burnham MR, Wu MC, Crystal RG. Pharmacokinetics of recombinant secretory leukoprotease inhibitor aerosolized to normals and individuals with cystic fibrosis. Am Rev Respir Dis 1993; 148:1056-1060.

McKenzie JE, Cruz-Rivera M. Compatibility of budesonide inhalation suspension with four nebulizing solu-tions. Ann Pharmacother 2004; 38:967-972.

Miall LS, McGinley NT, Brownlee KG, Conway SP. Methicillin resistant Staphylococcus aureus (MRSA) infection in cystic fibrosis. Arch Dis Child 2001; 84:160-162.

Mohammed RA, Klein KC. Inhaled amphotericin B for prophylaxis against invasive Aspergillus infections. Ann Pharmacother 2006; 40:2148-2154.

Moss RB. Administration of aerosolized antibiotics in cystic fibrosis patients. Chest 2001; 120 (3 Suppl): 107S-113S.

Moss RB. Long-term benefits of inhaled tobramycin in adolescents with cystic fibrosis. Chest 2002; 121:55-63.

Mukhopadhyay S, Staddon GE, Eastman C, Palmer M, Davies ER, Carswell F. The quantitative distribution of nebulized antibiotic in the lung in cystic fibrosis. Respir Med 1994; 88:203-211.

Chap

ter 2

60

Murphy TD, Anbar RD, Lester LA et al.Treatment with tobramycin solution for inhalation reduces hospital-izations in young CF subjects with mild lung disease Pediatr Pulmonol. 2004; 38:314-320.

Newman SP, Flora M, Hirst PH et al. Pharmacoscintigraphy of TOBI® in the Pari LC Plus™ and the Aerodose™ inhaler. J Aerosol Med 2001; 14:388.

Newman SP, Pellow PG, Clay MM, Clarke SW. Evaluation of jet nebulisers for use with gentamicin solution. Thorax 1985; 40:671-676.

Newman SP, Pitcairn GR, Hooper G, Knoch M. Efficient drug delivery to the lungs from a continuously oper-ated open-vent nebulizer and low pressure compressor system. Eur Respir J 1994; 7:1177-1181.

Newman SP, Woodman G, Clarke SW. Deposition of carbenicillin aerosols in cystic fibrosis: effects of nebu-liser system and breathing pattern. Thorax 1988; 43:318-322.

Newnham DM, Lipworth BJ. Nebuliser performance, pharmacokinetics, airways and systemic effects of salbutamol given via a novel nebuliser delivery system (“Ventstream”). Thorax 1994; 49:762-770.

Nikander K. Drug delivery systems. J Aerosol Med 1994; 7(Suppl 1): S19-24.Nikander K, Prince I, Coughlin S, Warren S, Taylor G. Lung deposition of 99mTc-DTPA delivered with an adap-

tive aerosol delivery system. Abstract 3264. European Respiratory Society Annual Congress, Stockholm, 2007.?

Nikolaizik WH, Schoni M. Pilot study to assess the effect of inhaled corticosteroids on lung function in patients with cystic fibrosis. J Pediatr 1996; 128:271-274.

Nikolaizik WH, Trociewicz K, Ratjen F. Bronchial reactions to the inhalation of high-dose tobramycin in cystic fibrosis. Eur Respir J 2002; 20:122-126.

Oliver A, Canton R, Campo P, Baquero F, Blazquez J. High frequency of hypermutable Pseudomonas aerugi-nosa in cystic fibrosis lung infection. Science 2000; 288:1251-1254.

Olivier KN, Bennett WD, Hohneker KW, et al. Acute safety and effects on mucociliary clearance of aero-solized uridine 50-triphosphate7amiloride in normal human adults. Am J Respir Crit Care Med 1996; 154:217–223.

O’Malley CA, VandenBranden SL, Zheng XT, Polito AM, McColley SA. A day in the life of a nebulizer: surveil-lance for bacterial growth in nebulizer equipment of children with cystic fibrosis in the hospital setting. Respir Care 2007; 52:258-262.

O’Riordan TG, Amram JC. Effect of nebulizer configuration on delivery of aerosolized tobramycin. J Aerosol Med 1997; 10:13-23.

Orenstein DM. Long-term inhaled bronchodilator therapy in cystic fibrosis. Chest 1991; 99:1061Pankhurst CL, Philpott-Howard J. The environmental risk factors associated with medical and dental equip-

ment in the transmission of Burkholderia (Pseudomonas) cepacia in cystic fibrosis patients. J Hosp Infect 1996; 32:249-255.

Park MK, Myers RA, Marzella L. Oxygen tensions and infections: modulation of microbial growth, activity of antimicrobial agents, and immunologic responses. Clin Infect Dis 1992; 14:720-740.

Pattishall EN. Longitudinal response of pulmonary function to bronchodilators in cystic fibrosis. Pediatr Pulmonol 1990; 9:80-85.

Phipps PR, Gonda I. Droplets produced by medical nebulizers. Some factors affecting their size and solute concentration. Chest 1990; 97:1327-1332.

Pilcer G, Goole J, Van Gansbeke B, et al. Pharmacoscintigraphic and pharmacokinetic evaluation of tobramy-cin DPI formulations in cystic fibrosis patients. Eur J Pharm Biopharm 2008; 68:413-421.

Pitchford KC, Corey M, Highsmith AK et al. Pseudomonas species contamination of cystic fibrosis patients’ home inhalation equipment. J Pediatr 1987; 111:212-216.

Quan JM, Tiddens HAWM, Sy JP et al. A two-year randomized, placebo-controlled trial of dornase alfa in young patients with cystic fibrosis with mild lung function abnormalities. J Pediatr 2001; 139:813–820.

Ramsey BW, Dorkin HL, Eisenberg JD et al. Efficacy of aerosolized tobramycin in patients with cystic fibrosis. N Engl J Med 1993; 328:1740-1746.

Ramsey BW, Pepe MS, Quan JM et al. Intermittent administration of inhaled tobramycin in patients with cystic fibrosis. Cystic Fibrosis Inhaled Tobramycin Study Group. N Engl J Med 1999; 340:23-30.


Ratjen F, Döring G, Nikolaizik W. Eradication of Pseudomonas aeruginosa with inhaled tobramycin in patients with cystic fibrosis. Lancet 2001; 358:983-984.

Ratjen F, Döring G. Cystic fibrosis. Lancet 2003; 361:681-689.Rau JL. Design principles of liquid nebulization devices currently in use. Respir Care 2002; 47:1257-1275;

discussion 1275-1258.Regelmann WE, Elliott GR, Warwick WJ, Clawson CC. Reduction of sputum Pseudomonas aeruginosa density

by antibiotics improves lung function in cystic fibrosis more than do bronchodilators and chest physio-therapy alone. Am Rev Respir Dis 1990; 141:914-921.

Reychler G, Aarab K, Van Ossel C, Gigi J, Simon A, Leal T. In vitro evaluation of efficacy of 5 methods of disinfection on mouthpieces and facemasks contaminated by strains of cystic fibrosis patients. J Cyst Fibros 2005; 4:183-187.

Reisner C, Katial RK, Bartelson BB, Buchmeir A, Rosenwasser LJ, Nelson HS. Characterization of aerosol out-put from various nebulizer/compressor combinations. Ann Allergy Asthma Immunol 2001; 86:566-574.

Retsch-Bogart GZ, Quittner A, Montgomery AB et al. Phase 3 trial (AIR-CF1) measuring improvement in respiratory symptoms in patients with cystic fibrosis (CF) following treatment with aztreonam lysine for inhalation (AZLI). Pediatr Pulmonol 2007; 42 (S30): 310

Retsch-Bogart GZ, Burns JL, Otto KL et al. A phase 2 study of aztreonam lysine for inhalation to treat patients with cystic fibrosis and Pseudomonas aeruginosa infection. Pediatr Pulmonol 2008; 43:47-58.

Robinson M, Hemming AL, Regnis JA et al. Effect of increasing doses of hypertonic saline on mucociliary clearance in patients with cystic fibrosis. Thorax 1997; 52:900-903.

Robinson TE, Goris ML, Zhu HJ et al. Dornase alfa reduces air trapping in children with mild cystic fibrosis. Chest 2005; 128:2327–2335.

Rogers GB, G.B., Carroll MP, M.P., Serisier DJ, D.J., Hockey PM, P.M., Jones, G., Bruce, K.D. Characterization of bacterial community diversity in cystic fibrosis lung infections by use of 16s ribosomal DNA terminal restriction fragment length polymorphism profiling. J Clin Microbiol 2004; 42:5176-5183.

Rogers GB, G.B., Carroll MP, M.P., Serisier DJ, D.J.et al. Use of 16S rRNA gene profiling by terminal restriction fragment length polymorphism analysis to compare bacterial communities in sputum and mouthwash samples from patients with cystic fibrosis. J Clin Microbiol 2006; 44:2601-2604.

Rosenfeld M. Home nebulizer use among patients with cystic fibrosis. J Pediatr 1998; 132:125-131.Rosenfeld M, Gibson R, McNamara S, Emerson J, McCoyd KS, Shell R., Serum and lower respiratory tract

drug concentrations after tobramycin inhalation in young children with cystic fibrosis. J Pediatr 2001; 139:572-577.

Rosenfeld M, Joy P, Nguyen CD, Krzewinski J, Burns JL. Cleaning home nebulizers used by patients with cystic fibrosis: is rinsing with tap water enough? J Hosp Infect 2001; 49:229-30.

Rosenstein BJ, Eigen H. Risks of alternate day prednisolone in patients with cystic fibrosis. Pediatrics 1991; 87:245-246.

Rottier BL, van Erp CJP, Sluyter TS, Heijerman HGM, Frijlink HW, de Boer AH. Changes in performance of the Pari eFlow® rapid and Pari LC Plus™ during 6 months use by CF patients. J Aerosol Med Pulm Drug Deliv 2009; 22: 1-7 (DOI: 10.1089/jamp.2008.0712).

Ruiz I, Molina I, Gavalda J et al. Safety and tolerability of nebulised liposomal Amphotericin B in haemato-logic patients. 45th Interscience Conference on antimicrobial agents and chemotherapy, 2005. Poster M-977.

Rutala WA, Barbee SL, Aguiar NC, Sobsey MD, Weber DJ. Antimicrobial activity of home disinfectants and natural products against potential human pathogens. Infect Control Hosp Epidemiol 2000; 21:33-38.

Sabet M, Saechao B, King P et al. In vivo antibacterial activity of aerosol MP-376 in a mouse model of pulmo-nary infection. Pediatr Pulmonol 2007; 42 (S30): 304

Saiman L, Marshall BC, Mayer-Hamblett N. et al. Azithromycin in patients with cystic fibrosis chronically infected with Pseudomonas aeruginosa : a randomized controlled trial. JAMA. 2003; 290:1749-1756.

Chap

ter 2

62

Saiman L, Siegel J. Infection control recommendations for patients with cystic fibrosis: microbiology, impor-tant pathogens, and infection control practices to prevent patient-to-patient transmission. Am J Infect Control 2003b; 31:S1-62.

Saiman L, Siegel J. Infection control in cystic fibrosis. Clin Microbiol Rev 2004; 17:57-71.Salvatore D. D’Andria M. Effects of salmeterol on arterial oxyhemoglobin saturations in patients with cystic

fibrosis. Pediatr Pulmonol 2002; 34:11-15.Sanchez I, Holbrow J, Chernick V. Acute bronchodilator response to a combination of beta-adrenergic and

anticholinergic agents in patients with cystic fibrosis. J Pediatr. 1992; 120:486-488.Sanchez I, De Koster J, Holbrow J, Chernick V. The effect of high doses of inhaled salbutamol and ipratropium

bromide in patients with stable cystic fibrosis. Chest. 1993; 104:842-846.Sanchez-Sousa A, Alvarez ME, Maiz L et al. Control of Aspergillus bronchial colonization in cystic fibrosis

patients: preliminary data using AmBisone aerosol therapy. XIIIth International Cystic Fibrosis Congress, 1996, S256.

Serisier DJ. Coates AD. Bowler SD. Effect of albuterol on maximal exercise capacity in cystic fibrosis. Chest 2007; 131:1181-1187.

Shah PL, Scott SF, Geddes DM et al. An evaluation of two aerosol delivery systems for rhDNase. Eur Respir J 1997; 10:1261-1266.

Simmons BP, Wong ES. Guideline for prevention of nosocomial pneumonia. Infect Control 1982; 3:327-333.Smaldone GC. Deposition patterns of nebulized drugs: is the pattern important? J Aerosol Med 1994;

7(Suppl 1):S25-S32.Smith AL, Ramsey BW, Hedges DL et al. Safely of aerosol tobramycin administration for 3 months to patients

with cystic fibrosis. Pediatr Pulmonol 1989; 7:265-271.Standaert TA, Morlin GL, Williams-Warren J, et al. Effects of repetitive use and cleaning techniques of dispos-

able jet nebulizers on aerosol generation. Chest 1998; 114:577-586.Stern M, Sens B, Wiedemann B, Busse O. Qualitätssicherung Mukoviszidose – Überblick über den Gesund-

heitszustand der Patienten in Deutschland 2001. Eds, Zentrum für Qualitätsmanagement im Gesund-heitswesen. Ärztekammer Hannover, 2002.

Suri R, Metcalfe C, Lees B et al.Comparison of hypertonic saline and alternate-day or daily recombinant human deoxyribonuclease in children with cystic fibrosis: a randomised trial. Lancet. 2001; 358:1316-1321.

Tablan OC, Chorba TL, Schidlow DV et al. Pseudomonas cepacia colonization in patients with cystic fibrosis: risk factors and clinical outcome. J Pediatr 1985; 107:382-387.

Taccetti G, Campana S, Festini F, Mascherini M, Döring G. Early eradication therapy against Pseudomonas aeruginosa in cystic fibrosis patients. Eur Respir J 2005; 26:1-4.

Tiddens HA, Pfaff SJ, van der Zanden T, Ruijgrok EJ. Weekly nebulisation of liposomal amphotericine B for the treatment of prednisone dependent ABPA. Pediatr Pulmonol 2003; 36 (S25): 301

Tiddens HA, Geller DE, Challoner P et al. Effect of dry powder inhaler resistance on the inspiratory flow rates and volumes of cystic fibrosis patients of six years and older. J Aerosol Med 2006; 19:456-465.

Tonnesen P, Stovring S. Positive expiratory pressure (PEP) as lung physiotherapy in cystic fibrosis: a pilot study. Eur J Respir Dis 1984; 65:419-422.

Touw DJ. Optimization of tobramycin treatment in cystic fibrosis: a pharmacokinetic approach. Thesis Rijksuniversiteit Leiden 1996; (Chapter 10): 179-188.

Touw DJ, Jacobs FA, Brimicombe RW, Heijerman HG, Bakker W, Breimer DD. Pharmacokinetics of aerosolized tobramycin in adult patients with cystic fibrosis. Antimicrob Agents Chemother 1997; 41:184-187.

Trissel LA. Handbook on injectable drugs 14th ed, ASHP, 2006Usmani OS, Biddiscombe MF, Barnes PJ. Regional lung deposition and bronchodilator response as a func-

tion of beta2-agonist particle size. Am J Respir Crit Care Med 2005; 172:1497-1504.Valerius NH, Koch C, Høiby N. Prevention of chronic Pseudomonas aeruginosa colonization in cystic fibrosis

by early treatment. Lancet 1991; 338:725-726.


Vanderbist F, Wery B, Baran D, Van Gansbeke B, Schoutens A, Moes AJ. Deposition of nacystelyn from a dry powder inhaler in healthy volunteers and cystic fibrosis patients. Drug Dev Ind Pharm 2001; 27:205-212.

van Schilfgaarde M, Eijk P, Regelink A et al. Haemophilus influenzae localized in epithelial cell layers is shielded from antibiotics and antibody-mediated bactericidal activity. Microb Pathog 1999; 26:249-262.

Vassal S, Taamma R, Marty N et al. Microbiologic contamination study of nebulizers after aerosol therapy in patients with cystic fibrosis. Am J Infect Control 2000; 28:347-351.

Vogelmeier C, Gillissen A, Buhl R. Use of secretory leukoprotease inhibitor to augment lung antineutrophil elastase activity. Chest 1996; 110:261S-266S.

Walsh NM, Casano AA, Manangan LP, Sinkowitz-Cochran RL, Jarvis WR. Risk factors for Burkholderia cepacia complex colonization and infection among patients with cystic fibrosis. J Pediatr 2002; 141:512-517.

Wark PA, McDonald V, Jones AP. Nebulised hypertonic saline for cystic fibrosis. Cochrane Database Syst Rev 2005, Issue 3. Art No: CD001506.pub2. DOI: 10.1002/14651858.CD001506.pub2.

Weber A, Morlin G, Cohen M, Williams-Warren J, Ramsey B, Smith A. Effect of nebulizer type and antibiotic concentration on device performance. Pediatr Pulmonol 1997; 23:249-260.

Weintraub SJ, Eschenbacher WL.The inhaled bronchodilators ipratropium bromide and metaproterenol in adults with CF. Chest. 1989; 95:861-864.

Westerman EM, Le Brun PPH, Touw DJ et al. Effect of nebulized colistin sulphate and colistin sulphomethate on lung function in patients with cystic fibrosis: a pilot study. J Cyst Fibros 2004; 3:23-28.

Westerman EM, De Boer AH, Le Brun PP et al. Dry powder inhalation of colistin in cystic fibrosis patients: A single dose pilot study. J Cyst Fibros 2007; 6:284-292.

Westerman, EM, De Boer AH, Le Brun PP et al.. Aerosolization of Tobramycin (TOBI®) with the PARI LC PLUS® Reusable Nebulizer: Which Compressor to Use? Comparison of the CR60® to the PortaNeb® Compres-sor. J Aerosol Med Pulm Drug Deliv 2008; 21:269-280.

Wexler MR, Rhame FS, Blumenthal MN, Cameron SB, Juni BA, Fish LA. Transmission of gram-negative bacilli to asthmatic children via home nebulizers. Ann Allergy 1991; 66:267-271.

Wiesemann HG, Steinkamp G, Ratjen F et al. Placebo controlled, double blind, randomized study of aero-solized tobramycin for early treatment of Pseudomonas aeruginosa colonization in patients with cystic fibrosis. Pediatr Pulmonol 1998; 25:88-92.

Wojtczak HA, Kerby GS, Wagener JS et al. Beclomethasone diproprionate reduced airway inflammation without adrenal suppression in young children with cystic fibrosis: A pilot study. Pediatr Pulmonol 2001; 32:293-302.

Wolff RK, Niven RW. Generation of aerosolized drugs. J Aerosol Med 1994; 7:89:106.Wolter J, Seeney S, Bell S, Bowler S, Masel P, McCormack J. Effect of long term treatment with azithromycin

on disease parameters in cystic fibrosis: a randomised trial. Thorax 2002; 57:212-216.Worlitzsch D, Tarran R, Ulrich M et al. Effects of reduced mucus oxygen concentration in airway Pseudomo-

nas infections of cystic fibrosis patients. J Clin Invest 2002; 109:317-325.Yerxa BR, Sabater JR, Davis CW, et al. Pharmacology of INS37217 [P1-(uridine 50)-P4-(20-deoxycytidine 50)

tetraphosphate, tetrasodium salt], a next-generation P2Y2 receptor agonist for the treatment ofcystic fibrosis. J Pharmacol Exp Ther 2002; 302:871–880.

Zach MS, Oberwaldner B, Forche G, Polgar G. Bronchodilators increase airway instability in cystic fibrosis. Am Rev Respir Dis. 1985; 131:537-543.

Zeitlin PL, Boyle MP, Guggino WB, Molina L. A phase I trial of intranasal Moli1901 for cystic fibrosis. Chest. 2004; 125:143-149.

Ziebach R,. Pietsch-Breitfeld B,. Bichler M,. Busch A,. Riethmuller J,. Stern M. Bronchodilatory effects of salbutamol, ipratropium bromide, and their combination: double-blind, placebo-controlled crossover study in cystic fibrosis. Pediatr Pulmonol 2001; 31:431-435.

Zhou Y, Ahuja A, Irvin CM, Kracko D, McDonald JD, Cheng YS. Evaluation of nebulizer performance under various humidity conditions. J Aerosol Med 2005; 18:283-293.

3 Aerosolisation of tobramycin (TOBI®) with the PARI LC PLUS® reusable nebuliser: which compressor to use?-comparison of the CR60® to the PortaNeb® compressor-

Elsbeth M Westerman, Anne H de Boer, Daan J Touw, Paul PH Le Brun, Albert C Roldaan, Henderik W Frijlink, Harry GM Heijerman

J Aerosol Med Pulm Drug Deliv 2008; 21:269-280.

Chap

ter 3

66

Summary

Aerosol output, aerosol output rate, and aerosol size distribution are influenced by the com-

pressed air flow rate through the nebuliser cup. Testing a nebuliser–compressor with a drug for

inhalation in cystic fibrosis (CF) patients is mandatory prior to starting therapy.

Tobramycin solution for inhalation (TSI), TOBI®, is licensed in Europe with a recommendation

for a “suitable” compressor connected to the PARI LC PLUS® nebuliser. To select a compressor,

five devices were tested in a previous in vitro study and this resulted in a subsequent in vivo

study. Two compressors [CR60® and PortaNeb® (PN)] were compared in an open, randomised,

crossover single dose pilot study in 10 CF patients to assess the most suitable device for inhala-

tion of TSI with the PARI LC PLUS® nebuliser. Lung function (FEV1 and FVC), pharmacokinetics

[PK; safety (Cmax, Ctrough)], lung deposition (indirect method AUC0–6), nebulisation time, and

patients’ experiences (questionnaire) were determined and compared. It was found that values

of Cmax and AUC0–6 were higher with the CR60® than with the PortaNeb®: 0.70 versus 0.54

mg/L, p = 0.005, and 2.54 versus 2.01 h.mg/L, p = 0.017, respectively. Tmax after use of the CR60®

appeared earlier (0.64 vs. 0.85 h, p = 0.005). Transient airway narrowing was measured in three

patients (2 x PN;1 x CR60®) versus subjective chest tightness in seven patients (CR60® > PN).

A shorter nebulisation time for CR60® of 13.2 min compared to PN 16.1 min (p = 0.022) was

observed, which was the main reason why patients preferred the CR60® (n = 7). No toxic serum

levels were reached after inhalation of TSI.

The CR60® compressor may seem advantageous based on a higher lung deposition and

a shorter nebulisation time, but a study in a large CF population to provide information on a

possible higher risk of toxicity of TSI is called for.

Aerosolisation of tobramycin (TOBI®) with the PARI LC PLUS® reusable nebuliser: which compressor to use? 67

Introduction

Inhaling the anti-pseudomonal drug tobramycin, an aminoglycoside antibiotic, interferes with

cystic fibrosis (CF) disease progression by reducing pulmonary deterioration and improving

lung function (Ramsey et al., 1999; Hodson et al., 2002). The goal of therapy with inhaled

antibiotics in CF is to reach the entire bronchial tree from the upper tract, which may contain

reservoirs of Pseudomonas aeruginosa from where the lower airways can be infected (Taylor

et al., 1992), to the small peripheral airways where disease progression starts (Tiddens 2002;

Worlitzsch et al., 2002). Considering the fact that the respiratory airways (generations 17–23)

comprise nearly 95% of the total surface area, it may be obvious that targeting of antibiotics in

CF therapy should aim primarily at the peripheral small airways to reach sufficiently high drug

concentrations throughout the respiratory system. Recently, good experimental support for

theoretical calculations has been obtained showing that effective peripheral deposition under

normal breathing conditions requires particles ≤ 2 μm (Usmani et al., 2005), whereas total lung

deposition is higher for these small particles compared to particles with larger diameters.

Jet nebulisers are the most commonly used devices for administration of tobramycin solu-

tion for inhalation (TSI). TOBI® is a licensed TSI for CF accompanied with recommended use

of a nebuliser and compressor (Ramsey et al., 1999). In Europe, compressor specifications are

a flow rate of 4–6 L/min and/or a pressure difference of 110–217 kPa when connected to the

PARI LC PLUS® nebuliser (Novartis Pharma B.V. 2006). In a previous in vitro study five different

compressors connected to the PARI LC PLUS® nebuliser were compared (De Boer et al., 2003).

It was found that the median droplet diameters from these combinations for TOBI® ranged

from nearly 4 to less than 2 μm, depending on the compressor used and the flow rate by

which the drug solution was aerosolized into the measuring device. In this study, the Freeway

Freedom® (jet flow of 3.3 L/min) produced the largest and the CR60® (jet flow of 6.9 L/min)

the smallest aerosol particles. The CR60® also yielded the narrowest size distribution and

exhibited the highest output rate (shortest nebulisation time), whereas the influence of the

inspiratory flow rate on the aerosol particle size distribution appeared to be almost negligible.

Based on these specific advantages, in particular the fineness of the aerosol, the CR60® was

presented as the most appropriate compressor for the administration of TOBI®. However, the

jet flow through the PARI LC PLUS®–CR60® combination is higher than specified in the product

information (Novartis Pharma B.V. 2006) and an improved peripheral deposition might result

in a rise in tobramycin serum concentrations with possibly increased systemic toxicity as as a

consequence (Hoffmann et al., 2002; Edson et al., 2004; Ahya et al., 2005; Scheenstra et al., 2007).

Therefore, an in vivo study was recommended to investigate these effects. This study fulfils that

recommendation.

The aim of this study was to investigate whether using the CR60® compressor in combina-

tion with the PARI LC PLUS® nebuliser results in an improved lung deposition and bioavailability

Chap

ter 3

68

for TOBI® in CF patients as expected based on previous studies (Geller et al., 2003; Usmani et al.,

2005), and if so, whether this affects safety with respect to systemic exposure.

The PARI LC PLUS®–CR60® nebuliser–compressor combination was compared to the Porta-

Neb® (PN) compressor attached to the same jet nebuliser, which is a frequently used combina-

tion for the administration of TOBI® (jet flow of 4.7 L/min). Additionally, it was tested whether

an improved lung deposition has an effect on postdose lung function and adverse events.

Finally, patients were asked for their preference to either one of the nebuliser–compressor

combinations.

Materials and methods

Study population and study design

Ten Pseudomonas aeruginosa positive cystic fibrosis outpatients, on maintenance treatment

with either TOBI® or colistimethate sodium for inhalation were asked to participate in an open,

randomised, crossover study. Inclusion criteria were age ≥ 18 years, clinical diagnosis of CF and

a positive sweat test or two CF related mutations, informed consent, and routine use of nebu-

lised tobramycin of colistimethate sodium. Exclusion criteria were exacerbation of pulmonary

function, intravenous use of tobramycin, tobramycin hypersensitivity, impaired renal function

(serum creatinin >176.8 μmol/L corresponding with an estimated creatinin clearance <60 mL/

min.), treatment with an investigational drug within a month prior to enrolment, pregnancy,

potentially pregnant or nursing women, and any other condition, which, in the opinion of

the clinician would make the subject unsuitable for enrolment. Demographic data are shown

in Table 1. Patients were asked to miss their regular evening dose of TOBI® or colistimethate

sodium prior to each study day and all nebulised drugs, including β-agonists, in the morning

Table 1: Demographic data

patient sex age(year)

height(cm)

weight(kg)

BMI(kg/m2)

Basal FEV1a

(%)FEV1/FVC

ratio

P1 m 29 168 61 21.6 66 64

P2 f 25 162 58 22.1 57 51

P3 m 22 174 60 19.8 57 65

P4 m 29 170 70 24.2 28 42

P5 f 36 162 51 19.4 57 74

P6 f 29 164 55 20.4 58 64

P7 f 28 160 50 19.5 39 55

P8 m 26 171 51 17.4 84 75

P9 m 52 183 67 20.0 35 30

P10 m 23 174 54 17.8 37 38

median 29 169 57 19.9 57 59

range 22-52 160-183 50-70 17.4-24.2 28-84 30-75

a average value of baseline FEV1 predicted on study days 1 and 2.


on the study day. On 2 separate days, the patients inhaled a single dose of TOBI® using the PARI

LC PLUS® nebuliser connected to either the PortaNeb® compressor or the CR60® compressor.

A maximum of 10 days was allowed between visit 1 and 2. The study was approved of by the

ethical review board of the hospital and was performed according to the Helsinki declaration.

Patients were fully informed by the investigators and gave their written consent before partici-

pating in the study.

Nebuliser, compressors, and administration of TOBI®TOBI® was obtained from Novartis, Arnhem, the Netherlands. TOBI® is an aqueous solution

of 300 mg tobramycin and 11.25 mg sodium chloride in single-dose ampoules of 5 mL (60

mg/mL), adjusted to pH 6. PARI LC PLUS® reusable nebulisers (PARI GmbH, Munich, Germany),

CR60® and PortaNeb® compressors (Medic-Aid Ltd. West Sussex, UK) were supplied by

Romedic B.V., Meerssen, the Netherlands. The aerodynamic characteristics of both nebuliser-

compressor combinations have been tested previously in an in vitro study (De Boer et al., 2003),

see Table 2.

Table 2: Particle size distribution of both nebuliser-compressor combinations (μm)a

Pari LC Plus® – CR60® Pari LC Plus® - PortaNeb®20 L/min 30 L/min 40 L/min 20 L/min 30 L/min 40 L/min

X10 0.70 0.70 0.71 0.77 0.78 0.79

X50b 1.82 1.83 1.83 2.49 2.53 2.30

X90 6.45 5.98 5.25 7.56 7.15 6.04

a Based on Fig. 5 from De Boer et al. (2003)b May be considered as mass median aerodynamic diameter (MMAD); De Boer et al. (2003)

To prevent bacterial cross contamination, each patient was supplied with a personal nebuliser.

All nebulisers were tested for reproducibility of performance in advance and had been cleansed

before first-time use (De Boer et al., 2003). After each use, the nebulisers were cleaned according

to a protocol, using warm soapy water and alcohol 70% after which the nebulisers were dried in

ambient air. The drug was administered at room temperature. The nebuliser cup was filled and

the compressor operated by the patient. Patients were instructed to inhale the drug according

to manufacturer’s instructions (Novartis Pharma B.V. 2006), wear noseclips, and to stop the

compressor temporarily in case of cough or other discomforts. They received no additional

instructions in order to influence daily nebulising routine as little as possible. By decision of the

patient, inhalation treatment with salbutamol was allowed after inhalation of TOBI®, in case

of severe discomfort due to airway narrowing and/or chest tightness. The nebulisation time,

defined by the time from turning on the compressor until the first observation of sputtering or

on indication by the patient, was measured using a stopwatch. Osmolality (Osmomat 030-D-3P,

Gonotec GmbH, Berlin, Germany) was measured in the drug residue in the nebuliser cup after

inhalation and compared to the osmolality of a TOBI® ampoule (175 mOsmol/kg) to determine

the effect of jet nebulisation on physicochemical properties of the tobramycin solution. The

Chap

ter 3

70

delivered dose, released from the inhalation device, was calculated by subtracting the amount

of drug residue in the nebuliser cup and in the ampoule from the drug label claim of 300 mg.

The drug residue in the nebuliser and ampoule was collected by rinsing these objects with 500

mL of water, according to protocol.

Pulmonary function

Pulmonary function tests were performed before, 5, and 30 min after inhalation or nebulisa-

tion. FEV1, FVC, and respiratory frequency were measured using a calibrated Masterlab pneu-

motachograph (Jaeger, Würzburg, Germany). The patients received instructions prior to the

pulmonary function tests. Postdose lung function test results were related to expected baseline

values. Measured lung function parameters were normalized to the reference values proposed

by the European Community for Coal and Steel (Quanjer et al., 1993). Percentage changes are

relative to baseline, and not a percentage fall of predicted. Airway narrowing (bronchospasm)

was defined by a reduction in FEV1 of 10% or more 5 min or 30 min after nebulisation of the study

drug. Patients were observed and asked for adverse effects during this period of observation.

Blood sampling

Venous blood samples were collected from patients prior to the TOBI® dose (in case of TOBI®

use at home) and 1, 3, and 5 h after nebulisation, on both study days.

Tobramycin concentration analysis

Tobramycin concentrations in serum and rinsing water were determined by fluorescence

polarization immunoassay (TdxFLx-system, Abbot Diagnostics, North Chicago, IL). The lower

limit of quantitation (LLQ) was 0.1 mg/L. To determine low serum concentrations of tobramycin

after inhalation a modified assay, with an lower limit of quantitation of 0.01 mg/L, was applied

(Touw et al., 1996).

Pharmacokinetic analysis

Serum concentrations of tobramycin were used as an indirect method to compare lung deposi-

tion in patients [Cmax, area under the time–concentration curve (AUC)]. This method has been

applied to compare inhalation devices previously (Asmus et al., 2002; Geller et al., 2003; Geller et

al., 2007). Cmax and Ctrough were also registered as a safety parameter. Based on experience with

inhaled and intravenous administration of aminoglycosides, safe serum levels are considered

to be a peak (Cmax) of <4 mg/L (Steinkamp et al., 1989; Cooney et al., 1994; Eisenberg et al., 1997;

Ramsey et al., 1999; Geller et al., 2002; Hodson et al., 2002; Chuchalin et al., 2007; Lenoir et al.,

2007) and a trough of <1 mg/L (Smyth et al., 2005).

Individual pharmacokinetic parameter values were calculated by iterative two-stage Bayes-

ian regression using nebuliser specific pharmacokinetic models for inhaled tobramycin. At

first, two-compartment open population pharmacokinetic models were made for each type of


nebuliser. Population modelling was performed using the MW/Pharm pharmacokinetic software

package version 3.60 (MediWare, Groningen, The Netherlands, Proost and Meijer 1992). In brief,

data on the patients’ gender, age, weight, height, tobramycin doses, and tobramycin serum

concentrations were recorded in the program. Creatinine clearance values were estimated

using the formula proposed by Jelliffe (Jelliffe 1971) for adult patients. A model previously

derived from inhalation data obtained from CF patients in our hospital (Touw et al., 1997) was

used as a starting (or prior) model. Model parameters were: Kel,m (1/h) 0.01 (fixed), Kel,r (1/h/(mL/

min) 0.00133 ± 0.01, V1 (L/kg) 0.422 ± 0.2, K12 (1/h) 0.07 ± 0.07, K21 (1/h) 0.21 ± 0.21, Kapo (1/h)

1.5 ± 1 and Fpo 0.1 (fixed). Because there were no patients included in the study with a poor

renal function, population analysis would provide no information on non renal or metabolic

elimination rate constant (Kelm). Therefore, Kelm was fixed at a value of 0.01 h-1, corresponding

with a tobramycin terminal half-life of 70 h, a value obtained from literature (Ng 1980). Second,

because only aerosolized tobramycin was administered, population analysis would provide no

information on absolute bioavailability (F). Therefore F was fixed at a value of 0.1 and total body

clearance (CL) and volume of distribution (Vd) were expressed as CL/F and Vd/F respectively

(Le Brun et al., 1999). After building the nebuliser specific pharmacokinetic population models

each patient was analyzed and the nebuliser specific model was fitted to the corresponding

concentration–time data points. Thus, obtained individual values for AUC, total body clearance

(CL/F), volume of distribution (Vd/F), maximum concentration (Cmax), trough concentration

(Ctrough), and time of maximum concentration (tmax) were used to compare both nebulisers.

Bioavailability (F) of the CR60® relative to the PortaNeb® was calculated with the following

formula: AUC(CR60®)/AUC(PortaNeb®) x Dose(PortaNeb®)/Dose(CR60®).

Questionnaire

The patients were asked for their experience using the two compressors on both study days in

relation to the daily use of their nebuliser–compressor combination at home (see Table 5).

Statistical analysis

Statistical analysis of the data was performed using the Wilcoxon signed rank test (SPSS 14.0,

SPSS Inc. Chicago, IL). Data are expressed by median and range. A p-value < 0.05 was considered

to be statistically significant.

Results

Bioavailability and safety

Table 3 summarizes the pharmacokinetic results, showing that the AUC and Cmax after CR60®

use are higher and the pharmacokinetic safety parameters Cmax and Ctrough were below their

toxic limits. Median Cmax after using the CR60® compressor was 1.3 times higher compared

Chap

ter 3

72

Tabl

e 3:

Pha

rmac

okin

etic

resu

lts

CR60

®Po

rtaN

eb®

V1/

FKa

AU

C 0-6

t max

C max

C trou

ghCl

/FV

1/F

KaA

UC 0-

6t m

axC m

axC tr

ough

Cl/F

p10.

0149

0.24

361.

980.

630.

530.

071.

260.

0394

0.30

221.

830.

810.

490.

072.

44

p20.

0141

0.25

932.

700.

660.

730.

091.

520.

0387

0.29

691.

980.

780.

520.

082.

14

p30.

0147

0.24

452.

260.

640.

600.

081.

640.

0340

0.29

802.

150.

940.

550.

101.

69

p40.

0145

0.23

233.

050.

660.

880.

121.

630.

0330

0.26

282.

531.

130.

600.

141.

56

p50.

0159

0.22

822.

090.

590.

550.

091.

160.

0457

0.28

251.

550.

780.

410.

072.

86

p60.

0141

0.30

672.

400.

630.

700.

061.

510.

0363

0.32

832.

040.

860.

560.

071.

73

p70.

0155

0.24

532.

620.

610.

700.

090.

720.

0397

0.29

141.

710.

810.

450.

072.

07

p80.

0118

0.54

655.

800.

642.

320.

082.

320.

0242

0.49

186.

180.

881.

950.

130.

85

p90.

0138

0.22

882.

450.

720.

620.

101.

380.

0325

0.28

041.

870.

960.

450.

111.

22

p10

0.01

160.

5130

6.65

0.64

2.49

0.06

0.75

0.03

000.

4518

3.64

0.83

1.14

0.05

1.32

med

ian

0.01

440.

2450

2.54

0.64

0.70

0.09

1.44

0.03

520.

300

2.01

0.85

0.54

0.08

1.71

rang

e0.

0116

–

0.01

590.

2282

–

0.54

651.

98 –

6.6

50.

59 –

0.7

20.

53 –

2.4

90.

06 –

0.1

20.

71 –

2.3

20.

0242

–

0.04

570.

2628

–

0.49

181.

55 –

6.1

80.

78 –

1.1

30.

41 –

1.9

50.

05 –

0.1

40.

85 –

2.8

6

p0.

333

0.01

7*0.

005*

0.00

5*0.

587

0.16

9

F CR60

/FPO

RTA

NEB

1.26

V1/F

(L/k

g): d

istr

ibut

ion

volu

me

Ka (h

-1):

abso

rptio

n ra

te c

onst

ant

AUC 0–

6 (h

.mg/

L): a

rea

unde

r the

cur

ve fr

om 0

to 6

ht m

ax (h

): tim

e to

max

imum

con

cent

ratio

nC m

ax (m

g/L)

: max

imum

ser

um c

once

ntra

tion.

C trou

gh (m

g/L)

: tro

ugh

seru

m c

once

ntra

tion,

12

h af

ter d

ose

Cl/F

(L/h

/kg)

: cle

aran

ceF CR

60®/

F PORT

AN

EB®:

rela

tive

bioa

vaila

bilit

y, c

alcu

late

d by

AU

C CR60

®/AU

C Port

aNeb

® x D

ose Po

rtaN

eb®/

Dos

e CR60

)*

stat

istic

ally

sig

nific

ant


to the PortaNeb® compressor (p = 0.005); when calculating the mean Cmax, this difference is

approximately 1.4 times higher. However, because of the extreme distribution of data, we have

decided to use median instead of mean values. The AUC’s were significantly different, resulting

in a relative bioavailability for the CR60® being 126% of that for the PortaNeb®, indicating a

26% higher absorption of tobramycin from the lung into the circulation, suggestive for a higher

exposure after the use of the CR60® (see Table 3 and Fig. 1). This higher Cmax and a larger AUC

correspond with a delivered dose from the PARI LC PLUS®–CR60® combination of only 84% of

the delivered dose by the PortaNeb® compressor. Cl/F and Ka were comparable between the

two groups. No correlation was observed between pre-treatment lung function parameters

(FEV1, FVC) and Cmax or AUC.

Lung function and adverse effects

Inhalation of TSI was performed with no major pulmonary adverse effects (see Table 4). A fall

in FEV1 of 10.4% and 14.4% was observed in patient 2 and 3, respectively, shortly after using

the PortaNeb® compressor, but not after using the CR60® compressor. Patient 9 showed a fall

in FEV1 of 8.6% shortly after and 12.9% 30 min after inhalation using the CR60®, but not after

PortaNeb® use. Seven patients experienced subjective symptoms of chest tightness after the

CR60® and five patients after the PortaNeb® (questionnaire).

Patient 2 experienced a fall in FVC of 19.1% (PortaNeb®) and 10.7% (CR60®), respectively.

This patient ranked the adverse effects (predominantly chest tightness) due to inhalation at

score 3 on both study days, see Table 5. No correlation between pre-treatment lung function

test results and fall in FEV1 or FVC after inhalation was observed. None of the patients needed

treatment with salbutamol. The tobramycin Cmax and Ctrough values of all patients were less than

4 mg/L and 1 mg/L, respectively.

Fig. 1. Serum concentration-time curves for individual patients after nebulisation of tobramycin with two compressors.

Chap

ter 3

74

Delivered dose and nebulisation time

The delivered tobramycin dose after use of the CR60® was 130.3 mg (range 95.5–215.7), 43.3%

of the drug label claim. The PortaNeb® delivered a nonsignificant higher dose of 155.1 mg

(range 117.5–170.0) or 51.7% of the drug label claim (p = 0.285). A nonsignificant difference

in the increase in osmolality after inhalation was observed of, respectively, 123% for the Por-

taNeb® and 129% for the CR60® compressor. It is expected that this increase in concentration

will not have influenced the particle size distribution of the aerosol output during nebulisation,

as shown in vitro (De Boer et al., 2003).

The PARI LC PLUS®–CR60® combination resulted in a significantly faster drug delivery of

TOBI® (median 13.2 min, range 11.1–15.8) than the PARI LC PLUS®–PortaNeb® combination

(median 16.1 min, range 11.8–19.4), p = 0.022. All patients had a consistent respiratory fre-

quency on both study days, but with a wide interindividual range of 10 to 22 inhalations per

minute. No correlation was found between respiratory rate and Cmax or AUC.

Patients’ impression

No differences were observed regarding ease of administration and noise. Seven patients pre-

ferred the CR60® over the PortaNeb® compressor, mainly because of the shorter nebulisation

Table 4: Influence of study drug on lung function

CR60® PORTANEB®FEV1

predose, baseline(% predicted)

at 5 min, relative to baseline


predose, baseline(% predicted)



Patient 1 2.56 (65.7) -1.2% -5.9% 2.57 (66.0) -5.4% -2.7%

Patient 2 1.74 (55.2) -8.0% -5.7% 1.83 (58.1) -10.4% -2.7%

Patient 3 2.22 (52.1) -1.4% 1.4% 2.63 (61.7) -14.4% -7.6%

Patient 4 1.08 (27.1) -2.8% 1.9% 1.15 (28.8) -6.1% -4.3%

Patient 5 1.67 (57.9) -4.8% 0.0% 1.59 (55.2) 0.0% 2.5%

Patient 6 1.85 (58.7) -1.1% 1.1% 1.79 (56.8) -1.1% -1.1%

Patient 7 1.20 (39.9) -4.2% -1.7% 1.15 (38.5) -6.1% -3.5%

Patient 8 3.29 (80.1) 3.3% -0.6% 3.60 (87.7) -1.7% -2.5%

Patient 9 1.39 (35.8) -8.6% -12.9% 1.34 (34.7) -5.2% 0.0%

Patient 10 1.61 (37.8) -1.2% 9.3% 1.58 (36.9) 1.3% 1.9%

FVCPatient 1 3.91 (85.4) -1.0% -1.8% 4.11 (89.7) -6.3% -4.9%

Patient 2 3.45 (95.5) -19.1% -1.4% 3.54 (98.0) -10.7% -1.1%

Patient 3 3.53 (70.1) 0.0% 1.1% 3.92 (77.9) -8.2% -4.1%

Patient 4 2.69 (57.2) -5.6% 2.6% 2.60 (55.5) -1.5% -1.5%

Patient 5 2.25 (67.5) -4.9% -3.1% 2.18 (65.5) -1.8% -0.5%

Patient 6 2.90 (80.2) 0.0% 0.0% 2.80 (77.4) -1.1% -1.1%

Patient 7 2.15 (62.5) 1.9% 0.0% 2.16 (62.8) -7.4% -1.9%

Patient 8 4.50 (93.1) 0.2% -1.8% 4.70 (97.1) -2.1% -1.9%

Patient 9 4.55 (93.8) 1.3% 2.6% 4.61 (95.1) -1.1% 2.0%

Patient 10 4.29 (85.2) -4.9% 2.3% 4.12 (81.9) -0.2% 4.1%


time. Five of them used the CR60® already for home treatment with TSI or colistimethate

sodium, and therefore this result may have been biased. Among the three patients using the

PortaNeb® at home, two preferred the CR60® after having used this compressor in the study.

Three patients preferred the PortaNeb® for various reasons; two of them were used to the Free-

way Freedom® at home. Three patients indicated to experience chest tightness as an adverse

effect during or shortly after drug delivery at home (score ≥ 2, see Table 5); one of them used a

bronchodilator prophylactically before inhaling the antibiotic. None of the three patients were

hindered in their daily life activities. Seven patients did not experience chest tightness; five

of them used a bronchodilator prophylactically at home before inhaling the antibiotic, one

patient mentioned nausea (score 2), and one patient did not have any adverse effect at all. All

patients were on a twice daily dosing scheme of TSI or colistimethate sodium for inhalation. The

time needed for aerosolization of a dose at home varied between 10 and 30 min.

Table 5: Questionnaire

Question ScoreConcerning drug administration on each study day1. Did you experience adverse effects during or after nebulisation today?2. When did these adverse effects occur?

3. How do you rate today’s nebuliser-compressor combination with respect to:– ease of operation– noise– time needed for administration– overall judgement

Concerning antibiotic for inhalation use at home1. Do you experience adverse effects during or after nebulisation at home?2. When do these adverse effects occur?

3. If you experience chest tightness because of tobramycin, do you use other (inhalation)drugs to reduce this effect?

4. Do these adverse effects influence your daily life?5. How many times a day do you nebulise an antibiotic?6. How long does the administration of a dose take (on average)?7. Which nebuliser-compressor combination do you use at home?

Conclusion: which of the two compressors do you prefer?

1-2-3-4-5-60-10 min after start inhalation10-20 min20-30 min> 30 min

1-2-3-4-5-61-2-3-4-5-61-2-3-4-5-61-2-3-4-5-6

1-2-3-4-5-60-10 min after start inhalation10-20 min20-30 min> 30 min

Yes / No

1-2-3-4-5-6…. times…. minutes…………………

PortaNeb® / CR60®1= none, 2 = minor, 3 = moderate, 4 = tolerable, 5 = serious, 6 = severe

Chap

ter 3

76

Discussion

The aim of this study was to investigate whether inhalation of a single dose of TOBI® (300 mg)

with the CR60® compressor in combination with the PARI LC PLUS® nebuliser results in an

improved lung deposition in CF patients compared to less powerful compressors, as expected

on the basis of previous in vitro results with this combination (De Boer et al., 2003). We have

shown that both AUC and Cmax are significantly higher with the CR60® (jet flow rate 6.9 L/min)

than when using the PortaNeb® compressor (4.7 L/min). We have also found that Cmax and

Ctrough from the CR60®–PARI LC PLUS® combination stay well below the limits for safety derived

from other studies. Finally, we have observed that none of the patients experienced serious

adverse effects, whereas post dose changes in lung function were not significantly different

between the two compressors.

Pharmacokinetics of tobramycin for inhalation, applied with various nebulisers, have been

subject of a great number of clinical studies (Steinkamp et al., 1989; Weber et al., 1989; Muk-

hopadhyay et al., 1993; Cooney et al., 1994; Eisenberg et al., 1997; Touw et al., 1997; Le Brun et

al., 1999; Ramsey et al., 1999; Le Brun et al., 2001; Geller et al., 2002; Geller et al., 2003; Poli et

al., 2007). Doses of 80-600 mg in tobramycin solutions (in drug concentrations of 4%–10%)

have been studied, resulting in serum levels (Cmax) ranging from not detectable to 3.6 mg/L

at approximately 1–1.5 h after administration of a dose and a bioavailability of 1.0%–16.6%.

Particularly, serum levels have been used for (toxicological) monitoring during inhalation treat-

ment (Le Brun et al., 1999; Ramsey et al., 1999; Le Brun et al., 2001; Geller et al., 2003; Newhouse

et al., 2003; Geller et al., 2007; Poli et al., 2007), but detailed information on the correlation

between lung exposure and pharmacokinetic data is lacking. Therefore, neither Cmax nor AUC

provides solid information on the amount of drug in the peripheral small airways. However,

from scintigraphic studies it is known that a greater total lung deposition and farther distal

airway penetration can be achieved with smaller particles. Usmani et al., (2005) have shown

for monodisperse salbutamol particles of 1.5, 3, and 6 μm, administered at a slow inhalation

(30–60 L/min), that 1.5 μm particles achieve a 12% higher total lung deposition and a 32%

higher peripheral deposition than 3 μm particles.

These findings, albeit for a different drug, and the knowledge that tobramycin is not effec-

tively absorbed by the gastrointestinal tract (Asmus et al., 2002) nor readily crosses epithelial

membranes (Neu 1976), combined with the results in our study of a higher tobramycin serum

peak concentration achieved earlier if finer droplets are produced after CR60® use, are at

least indicative for a higher exposure to peripheral airways where the conditions for absorp-

tion (large surface area and thin barrier) are optimal. Considering furthermore that the same

patients inhaled the study drug with both compressors within a period of only 10 days, and the

fact that the baseline lung function parameters differed only slightly between both days, it may

be expected that intraindividual variability did not dominate the outcome of the study.


Despite the low intraindividual variability in this study, extrapolation of the results to other

patients should be done with great care, as interpatient variability in lung deposition is known

to be large for CF patients. It is not well understood why the interpatient variability in Cmax

values after inhalation of a drug can be as large as shown in Table 3. Judged by the distribution

of baseline FEV1, our small patient group reflected the general CF population. Perhaps this

heterogeneity, among other parameters (Bennett and Smaldone, 1987) has contributed to the

observed interpatient variability. If so, this would justify studying subgroups for instance by

dividing a general CF population into three to four groups according to their FEV1.

Inhalation of the study drug by patients 8 and 10 resulted in a relatively high Cmax on both

occasions compared to the other patients in this group. Patient 8 had a baseline FEV1 of >75%

predicted, which is higher than the official specifications in the registration summary leaflet of

TOBI® (Novartis Pharma B.V. 2006). However, in clinical practice inhalation treatment with TSI

may start as soon as Pseudomonas aeruginosa has been detected in patients with minimal lung

damage (Wiesemann et al., 1998), and we therefore did not exclude this patient from the study.

To study the influence of patient 10 on the outcome of the study we reanalyzed the data after

exclusion of this patient, but this had no effect on the final conclusions. Similar interpatient

variability has been observed in other studies (Touw et al., 1997; Le Brun et al., 1999).

Early signs of nephrotoxicity (reduced creatinine clearance, serum creatinine within normal

range) after repeated intravenous doses of tobramycin have been described in literature (Ped-

ersen et al., 1987) and may be suggested as a possible cause of this interpatient variability. As a

result, tobramycin serum levels may rise, especially after repeated doses, caused by a reduced

drug clearance. No data on the number of intravenous courses or creatinine clearance values

were available for the subjects in this study. We were therefore unfortunately not able to per-

form an analysis on the possible influence of this phenomenon on the results in this study.

We furthermore suggest that differences in inhalation maneuvers on both study days are at

least partly responsible for the variation in Cmax observed in patient 10. However, considering

ignorance of many variables (e.g., relationship between lung deposition and serum concentra-

tions), interpretation of these data should be performed carefully. Although the Cmax is most

frequently used as an indirect parameter in investigating pulmonary drug dose delivery, other

studies have also presented the AUC as a relevant parameter (Touw et al., 1997; Le Brun et al.,

1999; Geller et al., 2003; Newhouse et al., 2003; Geller et al., 2007; Poli et al., 2007). The AUC may

be a better descriptor for total lung bioavailability, as this parameter comprises several effects

that depend on the type of inhalation device used, including the time needed for inhalation

and the absorption rate of a drug dose.

In several studies serum concentrations have been collected at 1 h after the end of inhala-

tion of TSI, assuming that the Cmax is reached at approximately 1 h after inhalation (Ramsey et

al., 1999; Prescribing information PrTOBI* 2006). However, Cmax and tmax are influenced by the

absorption rate and elimination rate of a drug dose and this study shows that tmax between

inhalation devices may be significantly different. This observation is supported by other

Chap

ter 3

78

pharmacokinetic studies (Touw et al., 1997; Le Brun et al., 1999; Le Brun et al., 2001; Geller et

al., 2003; Lenoir et al., 2007), in which mean tmax in patient groups fluctuated between 0.98 and

1.93 h, depending on the dose administered and/or the inhalation device used. It is therefore

recommended to perform pharmacokinetic calculations on inhalation data so Cmax, AUC, and

tmax values between studies can be compared.

Safety and efficacy of inhaled tobramycin were demonstrated in various previous studies

(Steinkamp et al., 1989; Cooney et al., 1994; Eisenberg et al., 1997; Ramsey et al., 1999; Geller et

al., 2002; Hodson et al., 2002; Chuchalin et al., 2007; Lenoir et al., 2007). The limit of safety for

Cmax in our study, 4 mg/L, has been derived from the observed Cmax values and the absence of

toxicity in CF patients described in these references and by Eisenberg et al., 1997. To the best

of our knowledge, in the literature only Cmax values have been related to safety in inhalation

studies with tobramycin for CF patients. However, Ctrough values are generally predictors of

possible aminoglycoside toxicity, and we therefore introduce this second limit as a parameter

for safety. The Ctrough value of 1 mg/L was chosen based on the current view on risk of toxicity

in intravenous treatment once daily (Smyth et al., 2005) accompanied with the knowledge that

serum levels after inhalation of tobramycin are lower and a 12-hourly dosing interval provides

sufficient time for renal clearance of the drug, provided renal function is normal. Next to the

extent of systemic exposure after absorption from the lung into the blood circulation (safety

parameters Cmax and Ctrough), a possible influence on auditory acuity, on vestibular effects (tin-

nitus) and on renal clearance may contribute to the overall safety profile of TSI (Ramsey et al.,

1999; Hoffmann et al., 2002; De Boer et al., 2003; Edson et al., 2004; Ahya et al., 2005; Lenoir et al.,

2007; Scheenstra et al., 2007). Except for Cmax and Ctrough, these safety parameters can only be

assessed after multiple doses.

Transient airway narrowing was measured in two patients after using the PortaNeb® at

t = 5 min and in one patient after using the CR60® at t = 30 min. This is in line with Hodson

et al., (2002) and Geller et al., (2007), who recorded airway narrowing in 11.3% and 10% of

their patients, respectively, after administration of a TSI dose. The origin of airway narrowing

caused by inhalation of tobramycin in particular is unknown (Chua et al., 1990; Ramagopal and

Lands, 2000; Alothman et al., 2002; Nikolaizik et al., 2002). We therefore use “airway narrowing”

to describe the reaction of a decline in FEV1 to an inhaled drug, as we do not have conclusive

evidence for the underlying mechanism. We suggest that airway narrowing may be caused by

the presence of a high concentration of drug particles in the bronchiolar area in susceptible CF

patients, resulting in contraction of the bronchiolar muscles as a result. This might happen if a

high aerosol concentration is formed in the larger airways because of an unfavourable particle

size-flow rate combination and/or reduced peripheral airway lumen caused by disease progres-

sion. This hypothesis may be studied in future.

Summarizing, the results from this study are in good agreement with literature data and the

results from the in vitro study in which the same compressors were compared (De Boer et al.,


2003). The median delivered dose of TSI when using the CR60® compressor was 16% lower

compared to the PortaNeb®, but resulted in a 26% higher median bioavailability and a 30%

higher Cmax that was attained 27% earlier (tmax). Furthermore, the median nebulisation time

for the CR60® was 18% shorter, albeit with a lower delivered dose. Although pharmacokinetic

data do not provide solid proof for the site of deposition, the higher AUC and Cmax in combina-

tion with a shorter tmax for a patient group showing low intrapatient variation in lung function

during the study, are indicative for improved peripheral deposition from the CR60®–PARI LC

PLUS® combination. A higher peripheral lung deposition is likely to result in a higher exposure

of Pseudomonas aeruginosa to tobramycin and an improved anti-pseudomonal effect: it would

be informative to learn more on this relationship between a higher peripheral deposition and

clinical effects, as has been found for inhaled beclometasone (Marshall et al., 2000). Based

on our results, it is expected that CF therapy with TOBI® can be further optimised by using a

more powerful compressor for the PARI LC PLUS® nebuliser. However, further investigations

are required to find more evidence for this expectation and to establish the long-term clinical

effects, including a possible higher risk of toxicity, of tobramycin inhalation treatment with the

CR60® in a large population of CF patients.

Acknowledgements

The authors wish to thank Mr. G. van der Meijden (Adult cystic fibrosis center, Haga Teaching

Hospital) for performing the pulmonary function tests, Mrs. A.I. de Graaf and Mr. J.J.M. de Clip-

peleir (Clinical, Pharmaceutical and Toxicological Laboratory, Apotheek Haagse Ziekenhuizen)

for the tobramycin assays, Mr. C.J.P. van Erp for his support in performing this study and Dr. F.M.

Hull who carefully commented on the manuscript.

Chap

ter 3

80

References

Ahya VN, Doyle AM, Mendez JD, Lipson DA, Christie JD, Blumberg EA et al. Renal and vestibular toxicity due to inhaled tobramycin in a lung transplant recipient. J Heart Lung Transplant 2005; 24:932-935.

Alothman GA, Alsaadi MM, Ho BL, Ho SL, Dupuis A, Corey M et al. Evaluation of bronchial constriction in children with cystic fibrosis after inhaling two different preparations of tobramycin. Chest 2002; 122:930-934.

Asmus MJ, Stewart BA, Milavetz G, Teresi ME, Han SH, Wang D et al. Tobramycin as a pharmacologic tracer to compare airway deposition from nebulizers. Pharmacotherapy 2002; 22:557-563.

Bennett WD, Smaldone GC. Human variation in the peripheral air-space deposition of inhaled particles. J Appl Physiol 1987; 62:1603-1610.

Chua HL, Collis GG, Le Souef PN. Bronchial response to nebulized antibiotics in children with cystic fibrosis. Eur Respir J 1990; 3:1114-1116.

Chuchalin A, Csiszer E, Gyurkovics K, Bartnicka MT, Sands D, Kapranov N et al. A formulation of aerosolized tobramycin (Bramitob) in the treatment of patients with cystic fibrosis and Pseudomonas aeruginosa infection: a double-blind, placebo-controlled, multicenter study. Paediatr Drugs 2007; 9 Suppl 1:21-31.

Cooney GF, Lum BL, Tomaselli M, Fiel SB. Absolute bioavailability and absorption characteristics of aero-solized tobramycin in adults with cystic fibrosis. J Clin Pharmacol 1994; 34:255-259.

De Boer AH, Hagedoorn P, Frijlink HW. The choice of a compressor for the aerosolisation of tobramycin (TOBI) with the PARI LC PLUS reusable nebuliser. Int J Pharm 2003; 268:59-69.

Edson RS, Brey RH, McDonald TJ, Terrell CL, McCarthy JT, Thibert JM. Vestibular toxicity due to inhaled tobramycin in a patient with renal insufficiency. Mayo Clin Proc 2004; 79:1185-1191.

Eisenberg J, Pepe M, Williams-Warren J, Vasiliev M, Montgomery AB, Smith AL et al. A comparison of peak sputum tobramycin concentration in patients with cystic fibrosis using jet and ultrasonic nebulizer systems. Aerosolized Tobramycin Study Group. Chest 1997; 111:955-962.

Geller DE, Konstan MW, Smith J, Noonberg SB, Conrad C. Novel tobramycin inhalation powder in cystic fibrosis subjects: pharmacokinetics and safety. Pediatr Pulmonol 2007; 42:307-313.


Geller DE, Rosenfeld M, Waltz DA, Wilmott RW. Efficiency of pulmonary administration of tobramycin solu-tion for inhalation in cystic fibrosis using an improved drug delivery system. Chest 2003; 123:28-36.


Hoffmann IM, Rubin BK, Iskandar SS, Schechter MS, Nagaraj SK, Bitzan MM. Acute renal failure in cystic fibrosis: association with inhaled tobramycin therapy. Pediatr Pulmonol 2002; 34:375-377.

Jelliffe RW. Estimation of creatinine clearance when urine cannot be collected. Lancet 1971; i:975-976.Le Brun PP, Vinks AA, Touw DJ, Hekelaar N, Mannes GP, Brimicombe RW et al. Can tobramycin inhalation be

improved with a jet nebulizer? Ther Drug Monit 1999; 21:618-624.Le Brun PPH, Mannes GPM, Brimicombe RW, Frijlink HW, Heijerman HGM. Pharmacokinetic modelling of

tobramycin after high dose inhalation in patients with cystic fibrosis. In: Optimization of antibiotic inhalation therapy in cystic fibrosis. Studies on nebulized tobramycin. Development of a colistin dry powder inhaler system. Thesis Rijksuniversiteit Groningen, The Netherlands 2001.;

Lenoir G, Antypkin YG, Miano A, Moretti P, Zanda M, Varoli G et al. Efficacy, safety, and local pharmacokinetics of highly concentrated nebulized tobramycin in patients with cystic fibrosis colonized with Pseudomo-nas aeruginosa . Paediatr Drugs 2007; 9 Suppl 1:11-20.

Marshall BG, Wangoo A, Harrison LI, Young DB, Shaw RJ. Tumour necrosis factor-alpha production in human alveolar macrophages: modulation by inhaled corticosteroid. Eur Respir J 2000; 15:764-770.

Mukhopadhyay S, Baer S, Blanshard J, Coleman M, Carswell F. Assessment of potential ototoxicity following high-dose nebulized tobramycin in patients with cystic fibrosis. J Antimicrob Chemother 1993; 31:429-436.


Neu HC. Tobramycin: an overview. J Infect Dis 1976; 134 Suppl:S3-S19.Newhouse MT, Hirst PH, Duddu SP, Walter YH, Tarara TE, Clark AR et al. Inhalation of a dry powder tobramycin

PulmoSphere formulation in healthy volunteers. Chest 2003; 124:360-366.Ng PK. Determining aminoglycoside dosage and blood levels using a programmable calculator. Am J Hosp

Pharm 1980; 37:225-231.Nikolaizik WH, Trociewicz K, Ratjen F. Bronchial reactions to the inhalation of high-dose tobramycin in cystic

fibrosis. Eur Respir J 2002; 20:122-126.Prescribing information PrTOBI* 2006. Novartis Pharmaceuticals Canada Inc., Quebec, Canada.Product information TOBI® (Samenvatting van de produktkenmerken) 2006. Novartis Pharma B.V., Arnhem,

The NetherlandsPoli G, Acerbi D, Pennini R, Soliani Raschini A, Corrado ME, Eichler HG et al. Clinical pharmacology study of

Bramitob, a tobramycin solution for nebulization, in comparison with Tobi. Paediatr Drugs 2007; 9 Suppl 1:3-9.

Proost JH, Meijer DKF. MW\Pharm, an integrated software package for drug dosage regimen calculation and therapeutic drug monitoring. Comput Biol Med 1992; 22:155-163.

Quanjer PH, Tammeling GJ, Cotes JE, Pedersen OF, Peslin R, Yernault JC. Lung volumes and forced ventila-tory flows. Report Working Party Standardization of Lung Function Tests, European Community for Steel and Coal. Official Statement of the European Respiratory Society. Eur Respir J Suppl 1993; 16:5-40.

Ramagopal M, Lands LC. Inhaled tobramycin and bronchial hyperactivity in cystic fibrosis. Pediatr Pulmonol 2000; 29:366-370.

Ramsey BW, Pepe MS, Quan JM, Otto KL, Montgomery AB, Williams-Warren J et al. Intermittent administra-tion of inhaled tobramycin in patients with cystic fibrosis. Cystic Fibrosis Inhaled Tobramycin Study Group. N Engl J Med 1999; 340:23-30.

Scheenstra RJ, Heijerman HG, Touw DJ, Rijntjes E. Cochleotoxicity of systemically administered tobramycin in Cystic Fibrosis patients. J Cyst Fibros 2007; 6:S94.

Smyth A, Tan KH, Hyman-Taylor P, Mulheran M, Lewis S, Stableforth D et al. Once versus three-times daily regimens of tobramycin treatment for pulmonary exacerbations of cystic fibrosis--the TOPIC study: a randomised controlled trial. Lancet 2005; 365:573-578.

Steinkamp G, Tummler B, Gappa M, Albus A, Potel J, Doring G et al. Long-term tobramycin aerosol therapy in cystic fibrosis. Pediatr Pulmonol 1989; 6:91-98.

Taylor RF, Morgan DW, Nicholson PS, Mackay IS, Hodson ME, Pitt TL. Extrapulmonary sites of Pseudomonas aeruginosa in adults with cystic fibrosis. Thorax 1992; 47:426-428.

Tiddens HAWM. Detecting early structural lung damage in cystic fibrosis. Pediatr Pulmonol 2002; 34:228-231.

Touw DJ, de Graaf AI, de Goede P. Evaluation of a fluorescence polarographic immunoassay with increased sensitivity for measurement of low concentrations of tobramycin in serum. Ther Drug Monit 1996; 18:189-193.

Touw DJ, Jacobs FA, Brimicombe RW, Heijerman HG, Bakker W, BreimerBriemer DD. Pharmacokinetics of aerosolized tobramycin in adult patients with cystic fibrosis. Antimicrob Agents Chemother 1997; 41:184-187.

Usmani OS, Biddiscombe MF, Barnes PJ. Regional lung deposition and bronchodilator response as a func-tion of beta2-agonist particle size. Am J Respir Crit Care Med 2005; 172:1497-1504.

Weber A, Smith A, Hack B, Williams-Warren J, Ramsey B, Unadkat J. Tobramycin serum pharmacokinetics after aerosol administration. Pediatr Pulmonol 1989; s4:136.

Wiesemann HG, Steinkamp G, Ratjen F, Bauernfeind A, Przyklenk B, Doring G et al. Placebo-controlled, double-blind, randomized study of aerosolized tobramycin for early treatment of Pseudomonas aerugi-nosa colonization in cystic fibrosis. Pediatr Pulmonol 1998; 25:88-92.

Worlitzsch D, Tarran R, Ulrich M, Schwab U, Cekici A, Meyer KC et al. Effects of reduced mucus oxygen con-centration in airway Pseudomonas infections of cystic fibrosis patients. J Clin Invest 2002; 109:317-325.

4 Effect of nebulised colistin sulfate and colistin sulfomethate on lung function in patients with cystic fibrosis: a pilot study

Elsbeth M Westerman, Paul PH Le Brun, Daan J Touw, Henderik W Frijlink, Harry GM Heijerman.

J Cyst Fibros 2004; 3:23-28.

Chap

ter 4

84

Summary

Pulmonary administration of colistin is one of the antimicrobial treatments used in cystic

fibrosis (CF) patients chronically infected with Pseudomonas aeruginosa . Dry powder inhala-

tion of colistin may be an attractive alternative to nebulisation of colistin. However, nebulised

colistin can cause airway narrowing in CF patients. Therefore, in the progress of developing a

dry powder formula, the choice of the inhaler and its contents should be guided by optimal

efficacy and the least possible side effects. To investigate the side effects, a study was initiated

to compare the tolerability of colistin sulfate to colistimethate sodium (colistin sulfomethate)

per nebulisation in CF-patients.

Nine CF-patients chronically infected with Pseudomonas aeruginosa participated in a double

blind, randomised cross over study. On two visits to the outpatient clinic, patients were submit-

ted to either nebulised colistin sulfate or colistimethate sodium solution. Lung function tests

were performed immediately before and 15 and 30 min after nebulisation.

Nebulisation of colistin sulfate caused a significant larger mean decrease in lung function

compared to nebulised colistimethate sodium. A significant decrease in mean changes (SD)

in FEV1 at 30 min and FVC at 15 and 30 min after nebulisation compared to baseline of -7.3%

(8.6%), -5.7% (7.3%) and -8.4% (7.5%) respectively was seen after colistin sulfate nebulisation

compared to colistimethate sodium (P<0.05). Seven patients were not able to complete the

nebulisation of colistin sulfate because of throat irritation and severe cough.

Based on these results it was concluded that inhalation with nebulised colistin sulfate is

not suitable for treatment of CF patients chronically infected with Pseudomonas aeruginosa .

Colistimethate sodium is the drug of choice for pulmonary administration of colistin.

Effect of nebulised colistin sulfate and colistin sulfomethate on lung function in patients with cystic fibrosis: a pilot study 85

Introduction

Nebulised colistin is one of the antimicrobial agents recommended for use in patients with cys-

tic fibrosis (CF) chronically infected with Pseudomonas aeruginosa (Döring et al., 2000). For this

therapy, commercially available vials for intravenous administration, containing colistimethate

sodium (colistin sulfomethate), are generally used. As nebulisation of drugs in general is a

time consuming activity, influencing daily life of patients, an alternative method of pulmonary

delivery of colistin would be welcome. Therefore, dry powder inhalation of colistin may be an

attractive alternative to nebulisation of colistin in CF-patients (Le Brun et al., 2002).

In vivo, colistimethate sodium is transformed into colistin sulfate, which is thought to have

a more potent antibacterial effect than the parent compound (Shawar 1999). Therefore, we

initially considered colistin sulfate as the compound of choice in the development of a colistin

dry powder for inhalation. In a previous pilot study, the feasibility of colistin sulfate as a dry

powder inhalation was investigated both in healthy volunteers and in patients (Le Brun 2002).

This newly developed dry powder inhalation system was highly appreciated by the patients

and provided a pulmonary deposition comparable to the deposition observed after nebulisa-

tion of colistimethate sodium solution. However, as reported, a decrease in pulmonary function

and the occurrence of non productive cough after dry powder inhalation of colistin sulfate was

observed in a number of patients, whereas no such effects were seen in the volunteer group.

Furthermore, no such side effects were observed in CF-patients after nebulisation of colistin as

sulfomethate. The origin of the side effects after colistin sulfate administration was not clear

and it was concluded that further research was necessary. Either a suboptimal particle size dis-

tribution of the dry powder or the chemical properties of colistin sulfate were held responsible

for these effects.

Improvement of particle size distribution is within reach. However, if the side effects were

provoked by the physical chemical properties of colistin sulfate, colistimethate sodium would

be the appropriate chemical form for further development of a dry powder inhalation system.

To investigate the latter hypothesis while excluding the first, both colistin salts should be tested

in a dissolved form. Therefore, the aim of this study was to compare the tolerability of colistin

sulfate to colistimethate sodium per nebulisation in cystic fibrosis patients.


Patients

Nine patients (five females) with CF, diagnosed by clinical history and confirmed by pathologi-

cal sweat tests or DNA analysis, volunteered to participate in the study. Patients were clinically

stable over the last 3 months, as determined by pulmonary function tests. All patients were

Chap

ter 4

86

chronically infected with Pseudomonas aeruginosa and therefore on maintenance treatment

with nebulised colistimethate sodium. Exclusion criteria were exacerbation of pulmonary

infection, colistin hypersensitivity, treatment with an investigational drug within a month prior

to enrolment, pregnancy, potentially pregnant or nursing women. The study was performed

according to the Helsinki declaration and was approved by the medical ethical review board

of the hospital. Patients were fully informed by the investigator and a written consent was

obtained from every patient.

Study protocol

Patients were asked to visit the outpatient clinic two times, with an interval of at least 5 days.

Patients were instructed not to nebulise colistimethate sodium or any other inhalation medica-

tion on the morning of the day of visit to the outpatient clinic. On the first visit, the patient

nebulised a solution of either colistin sulfate or colistimethate sodium and on the second visit

vice versa. This was done in a randomly assigned double-blind order. Blinding and randomisa-

tion was performed by the hospital pharmacy. Lung function tests were performed just before

and 15 and 30 min after nebulisation was completed. The patients were asked five questions

concerning their daily use of colistimethate sodium. The questionnaire and different scales

used for scoring are given in Fig. 1.

Materials

Colistimethate sodium (Colistin parenteral®, Grünenthal GmbH, Aachen, Germany) and colistin

sulfate (Ph. Eur. 1997, Duchefa, Haarlem, The Netherlands) were supplied by the hospital phar-

macy. An amount of 160 mg colistimethate sodium or 100 mg colistin sulfate was dissolved in

6 ml 0.9% aqueous sodium chloride solution by the hospital pharmacy prior to the test. The

solutions contained an equivalent amount of colistin (67 mg/6 ml). The pH of the colistimethate

sodium solution was approximately 7.4; the osmolality 366 mOsm/kg. The pH and osmolality of

the colistin sulfate solution were 5 and 306, respectively (pH meter Metrohm 713, Herisau, Swit-

zerland; osmolality meter Knauer A 0300, Berlin, Germany). Nebulisation of the colistin solutions

was done using a combination of a Porta-Neb® compressor and a Ventstream® jet nebuliser

Figure 1

Questionnaire

Question Score

1-2-3-4-5-6

0-10 min / 10-20 min / 20-30

min after nebulization

1-2-3-4-5-6

1. Do you experience adverse effects during or after nebulisation?

2. When do these adverse effects occur?

3. Do these adverse effects influence your daily life?

4. Do you use other (inhalation)drugs to decrease chest tightness after

nebulisation?

yes / no

1 = none, 2 = minor, 3 = moderate, 4 = tolerable, 5 = serious, 6 = severe

Fig. 1. Questionnaire.


(Medic Aid, Romedic, Meerssen, The Netherlands). The patients were instructed to operate

the device until the complete dose was released. In case of adverse effects of any kind during

nebulisation, participants were allowed to stop the inhalation of the aerosol temporarily.

Pulmonary function

Forced expiratory volume in one second (FEV1) and forced vital capacity (FVC) were measured

using a calibrated Masterlab pneumotachograph (Jaeger, Würzburg, Germany). Lung function

tests were performed following the guidelines of the European Respiratory Society (Quanjer

et al., 1993). A fall in FEV1 of 10% or more was considered as clinically significant. Percentage

changes are relative to baseline, and not a percentage fall of predicted.


To compare the effects on lung function of the two colistin forms, the changes from baseline

in the parameters FEV1 and FVC, found after administration of colistin sulfate, were compared

to the changes found after administration of colistimethate sodium using the paired Student’s

t-test. A p<0.05 was considered to be significant.

Results

Nine patients participated in the study (five females). Mean age was 29 years (range 20–41).

Mean baseline values (SD) of FEV1 and FVC were 57.8% (11.9%) and 80.2% (9.5%), respectively

(% predicted). After seven patients had been tested the severity of the adverse effects caused

us to perform an interim analysis. However, the subsequent two patients were again studied

under blinded conditions. The pulmonary function test results of the patients before and after

nebulisation of either one of the colistin solutions are presented in Fig. 2.

Colistin sulfate

Two patients completed nebulisation without subjective adverse effects, despite a decrease

in FEV1,15-0 of 31.3% and 10.0%, respectively. The remaining seven patients were not able to

complete nebulisation of colistin sulfate because of throat irritation and severe cough. In one of

these seven patients a clinically significant fall in FEV1 of 31.5% was seen 15 min after nebulisa-

tion. Five patients showed a fall in FEV1 15 and 30 min after nebulisation and in one patient

no effect on FEV1 was observed. FEV1 values 30 min after nebulisation were not significantly

altered compared to t=15 min. Chest tightness was noticed by those patients who were able

to continue nebulisation for a longer period of time. Next to the two patients that completed

nebulisation, two patients were able to nebulise at least 80% of the colistin sulfate solution. The

chest tightness reported by these four patients lasted 2–3 days. In the seven patients that par-

tially nebulised the solution, severe coughing was accompanied by perspiration and a sensation

Chap

ter 4

88

of heat. The irritating effect of the solution was most pronounced in the throat. Some patients

noticed an increased mucus production. All patients noted an unpleasant taste. One patient

needed treatment with a bronchodilator drug after lung function tests were completed.

Colistimethate sodium

All patients completed nebulisation of colistimethate sodium. A clinically significant fall in

FEV1 15 min after nebulisation was observed in two patients. This effect ameliorated after 30

min. In one of these two patients, the fall in FEV1 was accompanied by a decrease in FVC. This

Fig. 2. Individual FEV1 and FVC-values (deviation from baseline in percentage), for nine patients, after nebulisation of colistin sulfate and colistimethate sodium respectively.


patient indicated that she also experienced chest tightness during daily colistimethate sodium

nebulisation. No effects other than in daily use were noticed by the patients.

Comparison of colistin sulfate to colistimethate sodium

The results in Table 1 and Fig. 2 show that the decrease in lung function is more severe after

administration of colistin sulfate than after administration of colistimethate sodium. A statisti-

cally significant difference in mean changes in FEV1,30-0, FVC15-0 and FVC30-0 after colistin sulfate

nebulisation compared to colistimethate sodium was observed. However, the difference in

decrease in FEV1 between the two colistin salts, observed 15 min after nebulisation, was not

statistically significant.

Questionnaire

The questionnaire was intended to give insight in the daily use of colistin (as sulfomethate)

by the patients. Four patients experienced adverse effects during or after nebulisation (score

3), three patients scored 2 and two patients did not experience adverse effects at all (score

1). Adverse effects occurred at 0–10 min after starting nebulisation in eight patients and after

10–20 min in one patient. None of the patients indicated that the adverse effects influenced

their daily life. All patients used inhalation medication daily. Six patients sometimes used a

short acting β2-sympathicomimetic inhalation drug.

Discussion

The aim of this study was to assess the tolerability, defined as a possible clinically relevant

difference in pulmonary function and adverse effects, during and after nebulisation of two

different chemical entities of colistin. The results of this study show that nebulisation of colistin

sulfate is not tolerated by CF-patients in contrast to nebulisation of colistimethate sodium. A

Table 1 Differences in effect on lung function between colistin sulfate and colistimethate sodium:mean changes in FEV1 and FVC from baseline values after administration of colistin sulfate and after colistimethate sodium have been compared at t=15 and t=30 minutes.

change in FEV1 (S.D.)(15 min vs baseline)

change in FEV1 (S.D.)(30 min vs baseline)

change in FVC (S.D.)(15 min vs baseline)

change in FVC (S.D.)(30 min vs baseline)

colistin sulfate -11,3% (12,6%) -10,2% (9,4%) -10,4% (9,9%) -10,5% (9,5%)

colistimethate sodium -6,0% (7,0%) -2,9% (3,3%) -4,7% (8,1%) -2,1% (4,9%)

results (difference) -5,3% (12,2%) -7,3% (8,6%) -5,7% (7,3%) -8,4% (7,5%)

p = 0,225 p = 0,034 p = 0,047 p = 0,010

Change in ∆FEV1 (S.D.): mean change in FEV1 from baseline after colistin sulfate nebulisation minus mean change in FEV1 from baseline after colistimethate sodium nebulisation. S.D. = standard deviationChange in ∆FVC (S.D.): mean change in FVC from baseline after colistin sulfate nebulisation minus mean change in FVC from baseline after colistimethate sodium nebulisation. S.D. = standard deviation

Chap

ter 4

90

relationship between adverse effects and adherence to treatment by CF-patients has been

established; airway narrowing may be a reason for poor and discontinuing the therapy (Abbott

et al., 1994; Maddison et al., 1994). Therefore, colistin sulfate should not be considered as the

chemical entity of choice in treatment of CF patients chronically infected with Pseudomonas

aeruginosa .

The patients in this study were used to daily nebulisation of colistimethate sodium. Although

some decrease in lung function was observed in all patients and in two patients a clinically

significant decrease in FEV1 was observed 15 min after nebulisation, only one patient noticed

airway narrowing. In contrast, serious side effects (cough and irritation) and deterioration of

lung function were observed after administration of colistin sulfate. In three patients, a reduc-

tion in FEV1 of 10% or more was observed.

The adverse effects found after administration of the nebulised colistin sulfate solution

appeared to be more serious compared to the adverse effects after dry powder inhalation,

observed by Le Brun et al., 2002. Airway narrowing after inhalation of antibiotics in CF patients

is quite common. Airway narrowing and chest tightness after nebulisation of colistimethate

sodium has been reported in literature (Maddison et al., 1994; Cunningham et al., 2001). Cun-

ningham et al. reported a decrease in FEV1 of more than 10% in 20 out of 58 children (34%)

immediately and 15 min after nebulisation. In 9% of these children, this decrease still persisted

at 30 min after nebulisation. Maximal airway narrowing was measured immediately after

nebulisation in 13 patients, after 15 min in five patients and after 30 min in two patients. Thirty-

five of 46 patients (76%) in a study by Maddison et al. developed bronchoconstriction after

nebulisation of colistin. No definition of clinically significant bronchoconstriction was reported

by the authors. Maximal bronchoconstriction was observed immediately after nebulisation in

30 patients, after 15 min in three patients and after 30 min in two patients. No change in FEV1

was reported in seven patients.

Recent published data concerning bronchial reactions to the inhalation of several tobramycin

preparations, including high-dose tobramycin, in 12 CF-patients with moderate disease show a

significant bronchial obstruction (10% decrease in mean FEV1) shortly after nebulisation. Bron-

choconstriction was most severe after nebulisation of high-dose tobramycin. After 10 min of

inhalation lung function tests had normalised. These results support the suggestion that lung

function tests generally normalise within 10 min after nebulisation (Nikolaizik et al., 2002). Our

results show a decrease in FEV1 similar as described by Maddison et al., 1994 and Cunningham

et al., 2001. A statistically significant change in mean FEV1, 30 min after nebulisation compared

to baseline, between colistin sulfate and colistimethate sodium was observed. However, no

significant difference in FEV1 at t=15 min between the two colistin salts was seen. These obser-

vations show the lasting effect of colistin sulfate compared to the relatively rapid recovery of

lung function after colistimethate sodium nebulisation. If the results of Cunningham et al. and

Maddison et al. would be applicable to our patient group, maximal airway narrowing could

have occurred immediately after cessation of nebulisation. Additionally, the decrease in forced


vital capacity at 15 and 30 min compared to baseline, seen after colistin sulfate nebulisation,

was significantly lower compared to nebulised colistimethate sodium.

To our knowledge no earlier data concerning the effect of nebulised colistin sulfate in CF-

patients have been published. The observed effects during and after colistin sulfate nebulisation

are caused by a yet unknown mechanism. Whether the tonicity or pH of both solutions were of

any influence on the results, remains unclear. Although adverse effects, related to nebulisation

of colistimethate sodium, were reported in the questionnaire, none of the patients indicated

that it influences their daily life. Apparently, lung function parameters improve within 30 min

after completing nebulisation in most patients, as proven by our data.

Conclusion

This study showed that airway narrowing after nebulisation of colistin sulfate was significantly

more severe than after nebulisation of colistimethate sodium. Most patients were forced to stop

nebulising colistin sulfate because of throat irritation and (severe) cough. The mechanism caus-

ing the observed effects is not elucidated. As most patients experienced serious side effects,

it is concluded that nebulised colistin as sulfate is not suitable for treatment of CF patients

chronically infected with Pseudomonas. aeruginosa. Future research towards the development

of a colistin dry powder inhalation will focus on the use of colistimethate sodium.

Acknowledgments

The authors wish to thank the pharmacy technicians (Apotheek Haagse Ziekenhuizen, The

Hague, The Netherlands) for preparing the study medication and Mr G. van der Meyden (Adult

Cystic Fibrosis Center, Haga Teaching Hospital, The Hague, The Netherlands) for performing the

pulmonary function tests.

Chap

ter 4

92

References

Abbott J, Dodd M, Bilton D, Webb AK. Treatment compliance in young adults with cystic fibrosis. Thorax 1994; 49:115-120.

Cunningham S, Prasad A, Collyer L, Carr S, Lynn IB, Wallis C. Bronchoconstriction following nebulised colistin in cystic fibrosis. Arch Dis Child 2001; 84:432-433.


Le Brun PP, de Boer AH, Mannes GP, de Fraiture DM, Brimicombe RW, Touw DJ et al. Dry powder inhalation of antibiotics in cystic fibrosis therapy: part 2. Inhalation of a novel colistin dry powder formulation: a feasibility study in healthy volunteers and patients. Eur J Pharm Biopharm 2002; 54:25-32.

Maddison J, Dodd M, Webb AK. Nebulized colistin causes chest tightness in adults with cystic fibrosis. Respir Med 1994; 88:145-147.

Nikolaizik WH, Trociewicz K, Ratjen F. Bronchial reactions to the inhalation of high-dose tobramycin in cystic fibrosis. Eur Respir J 2002; 20:122-126.


Shawar R, Anderson, S., Barker, L, Nguyen, L., VanDevanter DR, D.R., Tanaka, S.K. In vitro assessment of bactericidal activity and relevance to pharmacodynamics of aerosolized antibiotics. Neth J Med 1999; 54:S38-S39.

5 Design and in vitro performance testing of multiple air classifier technology in a new disposable inhaler concept (Twincer®) for high powder doses

Anne H de Boer, Paul Hagedoorn, Elsbeth M Westerman, Paul PH Le Brun, Harry GM Heijerman, Henderik W Frijlink

Eur J Pharm Sci 2006; 28:171-178.

Chap

ter 5

94

Summary

Dry powder inhalation of antibiotics in cystic fibrosis (CF) therapy may be a valuable alternative

for wet nebulisation, because it saves time and it improves lung deposition.

In this study, it is shown that the use of multiple air classifier technology enables effective

dispersion of large amounts of micronised powder (up to 25 mg). X50-values of the aerosol from

laser diffraction analysis obtained with the Twincer® disposable inhaler concept (containing

multiple air classifier technology) are practically the same as that for the pure drug in the range

of dose weights between 0 and 25 mg. Only for the highest dose weights, a minor fraction

(5–7.5%) of small agglomerates (5–15 μm) is released from the inhaler. Moreover, the size distri-

bution of the aerosol is practically the same at 1 and 4 kPa. Cascade impactor results confirm the

good performance of the multiple classifier concept. Unprocessed micronised particles or soft

spherical agglomerates can be used, and special particle engineering processes are not neces-

sary. Only a minor fraction of coarse sweeper crystals in the formulation is desired to reduce the

total inhaler losses for colistimethate sodium to less than 5–6% at 4 kPa. The classifiers can be

designed to retain these crystals with more than 95% efficiency.

Design and in vitro performance testing of multiple air classifier technology in the Twincer® 95

Introduction

In the past decade, a great interest has been developed in the pulmonary administration of

high drug doses, e.g. for systemically active substances that cannot be given via the oral route

because of a poor bioavailability. Such drugs have to be delivered to the peripheral airways

where the permeability is high and the surface area for absorption is large (Patton, 1996; Kim

and Follinsbee, 1997; Groneberg et al., 2003). The preferable drug particle size for deposition in

the small airways is between 1 and 5 μm under the conditions of quiet breathing and a certain

period of breath hold (Gerrity, 1990; Martonen and Katz, 1993; Schultz, 1998).

Many systemically active drugs via the pulmonary route are peptides and proteins which

have to be stabilised with sugar glass technology, using spray drying or spray-freeze drying

techniques (Gribbon et al., 1996; van Drooge et al., 2004). Sugar glass technology increases

the amount of powder to be dispersed and mostly, such powders are highly hygroscopic and

cohesive. All these factors contribute to a low (and highly variable) bioavailability, as reported

for instance for insulin until now. They result in substantial drug losses in the device (including

the dose system) and upper airway deposition of insufficiently dispersed powders, or particles

released with too high velocity (Newhouse et al., 2003; Patton et al., 2004).

A special group of high dose drugs administered via the pulmonary route are the antibiotics,

given for instance in cystic fibrosis (CF) therapy. CF is a hereditary disease, characterised by

secretions of extremely high viscosity from exocrine glands in the airways (Smith et al., 1996).

The increased viscosity of the mucus hinders clearance of micro-organisms from the respira-

tory tract (Geddes, 1997). The inflammatory response to infected sites mainly with Haemophilus

influenza and Staphylococcus aureus during childhood, followed by Pseudomonas aeruginosa in

later years (Touw et al., 1995; Wood, 1996) gradually causes airway damage, which is irreversible

and eventually leads to death (Ferrari et al., 2002). Different target areas for antibiotic drugs

in CF have been mentioned, including the bronchial lumen (e.g. Ramsey, 1996; Van Devanter

and Montgommery, 1998), the smaller bronchioles (Touw et al., 1995; Geddes, 1997) and more

recently the peripheral airways (Tiddens, 2002). Different antibiotics are used for inhalation by

nebulisation, such as gentamicin (Newman et al., 1985), colistimethate sodium (Li et al., 2001)

and tobramycin (Le Brun et al., 1999). Conventional nebulisation of the high doses (e.g. 160

mg for gentamicin and colistimethate sodium and 300 mg for tobramycin), may require up to

30 minutes. This influences the quality of life and it is disadvantageous from the viewpoint of

patient adherence to treatment.

An alternative for wet nebulisation is dry powder inhalation. Dry powder inhalers for

gentamicin (Goldman et al., 1990; Crowther Labiris et al., 1999), colistin sulfate (Le Brun et al.,

2002), colistimethate sodium (Flynn et al., 2000) and tobramycin sulfate (Newhouse et al., 2003)

have been presented with varying success. These studies have shown that sputum (or plasma)

concentrations and lung depositions may be comparable to those achieved with nebulisation,

but the inhaled doses from dry powder inhalers were rather high so far. They varied from 160

Chap

ter 5

96

to 180 mg for gentamicin, 150 mg for tobramycin sulfate (PulmoSphere formulation), 30–150

mg for colistimethate sodium to 25 mg for colistin sulfate. The reasons for these high doses

are ineffective powder dispersion which leads to poor lung deposition. The dose weights

mentioned were for the pure drug, except for the PulmoSphere formulation, which contained

10% excipient. As a result, inhalation of a relatively large number of powder quantities for a

single dose was necessary, varying from 3 (for colistin sulfate) to 32 (for gentamicin). Not even

particle engineering (like processing of tobramycin sulfate into PulmoSphere) appeared to be

sufficiently effective to reduce the number of inhalations (Newhouse et al., 2003).

The aim of this study was to develop a highly effective, so-called passive (i.e. breath driven)

powder de-agglomeration principle for a new disposable inhaler concept which enables

the pulmonary administration of high drug doses in one or two inhalation manoeuvres. The

application is primarily for antibiotics and for sugar glass formulations containing therapeutic

proteins (or lipophilic drugs), but the inhaler may also be used for other drugs, like for instance

medication that has to be given only once (e.g. vaccines). The de-agglomeration principle

(multiple air classifier technology) has been designed to produce a high fine particle fraction

already at a relatively low inspiratory effort without using particle-engineered technologies.

This will make it possible to inhale slowly and to minimise throat and upper tract deposition.

Another objective was to minimise the amount of inert excipient in the inhalation powder,

which (in combination with efficient excipient retention by the inhaler) reduces the amount of

powder to be inhaled.


The disposable Twincer®, multiple classifier dry powder inhaler

The design of the disposable multi classifier dry powder inhaler is shown in Fig. 1. Basically,

the inhaler consists of three plate-like parts and a blister strip for the powder formulation with

the micronised drug. The plate-like parts have various projections and depressions which

constitute the air flow passages and the blister chamber when the parts are assembled, as

shown and explained more in detail by de Boer et al. (2004). The powder formulation is stored

in a blister which has a long cover foil. This cover foil is folded and sticks out from the rear end

of the inhaler. By pulling the cover foil, the blister is opened and connected to the powder

channel and the inlet to this channel (Fig. 1). Air passing through the powder channel during

inhalation entrains the powder from the blister, and the powder flow is divided between two

(or more) parallel classifiers, which are circular depressions in the bottom plate (classifier plate)

of the inhaler. The classifiers shown in Fig. 1 have two additional air channels each (to maintain

a stable tangential flow pattern inside the classifiers), but there may be different numbers of

such channels. Instead of having basically a cylindrical wall, the classifier may also be polygonal,

depending on the type of powder formulation to be processed (de Boer et al., 2004). For the


Twincer® concept depicted in Fig. 1, the classifier discharge holes (in the discharge plate) are

circular. Around each hole (on the bottom side of the discharge plate facing the classifier plate)

there is a circular rim projecting into the classifier chamber. The ratio of the diameter of the

discharge hole to that of the classifier chamber and the height of the rims around the discharge

holes (in relation to the depth of the classifier chamber) can be varied to control the residence

time for the powder in the classifier and the efficiency with which carrier or sweeper particles

are retained. The discharge holes are connected with two discharge channels ending in the

mouthpiece channel of the inhaler which ends (for the prototype tested) as a straight narrow

opening. Bypass channels around the discharge holes are used to reduce the inhaler accumula-

tions and to control the total inhaler resistance.

In this study, only the de-agglomeration efficiency of the multiple classifier principle has

been tested. Optimisation of the entire inhaler design (including the mouthpiece, which con-

trols the aerosol discharge flow in the oral cavity) was not (yet) the objective.

The example shown in Fig. 1 is the basic design of the Twincer® multiple classifier inhaler.

Various concepts with different classifier designs for different drug formulations have been

constructed and tested. Fig. 2A shows the machined concept used for a proof of principle with

colistimethate sodium in CF patients and healthy volunteers, of which the results have been

presented by Westerman et al., 2007a, 2007b). This concept has no blister, but two different

4

Figure 1

2. Blister strip

1. Classifier plate

3. Discharge plate

4. Cover plate

Inhaler parts:

Powder channelClassifiers

Classifier air channels

Bypass channels

Classifier discharge hole with circular rim projecting

into the classifier

Mouthpiece channel

Inlet powder channel

2. Blister strip

1. Classifier plate

3. Discharge plate

4. Cover plate

Inhaler parts:

Powder channelClassifiers

Classifier air channels

Bypass channels

Classifier discharge hole with circular rim projecting

into the classifier

Mouthpiece channel

Inlet powder channel

Figure 1 Presentation of the Twincer® as a disposable inhaler for high doses of moisture sensitive materials stored in a blister. The drawing shows the basic concept which consists of three plate-like inhaler parts and the blister with a long pull-off strip. Relevant inhaler parts (and functions) discussed in the text are indicated in the drawing.

Chap

ter 5

98

dose compartments (filled manually) which can be connected with the powder channel by

means of a slide. Fig. 2B shows a first moulded prototype containing a single blister (300mm3)

to be used for a patient study with cyclosporine.

Materials

Micronised colistimethate sodium was purchased from Alpharma (Copenhagen, Denmark).

Characteristic values for the size distribution (obtained from laser diffraction analysis using

RODOS dispersion at 3 bar) after micronisation (see Section 2.3) are X10 = 0.88 μm; X50 = 1.99

5

Figure 2

A

B

Slide connecting the air inlet to the powderchannel with the dose compartments

Air inlet to the powder channel

Dose compartments(hidden under slide)

Mouthpiece channel

A

B

A

B

Slide connecting the air inlet to the powderchannel with the dose compartments

Air inlet to the powder channel

Dose compartments(hidden under slide)

Mouthpiece channel

Figure 2 A. Presentation of the (machined) concept used for a proof of principle to administer 2 x 12.5 mg colistimethate sodium to 7 healthy volunteers and 10 CF patients. The concept has no blister, but two dose compartments for the drug formulation that can be connected with the air inlet (hole) and the powder channel by means of a slide.B. Presentation of a first moulded prototype of the Twincer® (with blister; not visible) to be used for a patient study with cyclosporine.


μm; X90 = 3.53 μm and X100 = 5.00 μm. Pharmatose 110M and 80M (starting materials for the

preparation of special size fractions of sweeper crystals) were supplied by DMV International

(Veghel, The Netherlands). A Pulmicort® 200 Turbuhaler® (reference device) was obtained from

the local pharmacist. All Twincer® devices used in this study were machined devices, similar

to that shown in Fig. 2A. They were constructed by the research workshop of the Faculty of

Medicine (University of Groningen).

Methods

For the micronisation of the colistimethate sodium to the desired size distribution for inhalation,

a spiral jet mill with 0.8mm nozzle (50 AS, Alpine, Augsburg, Germany) was used. Lactose size

fractions (sweeper crystals) of 63–100; 150–200 and 250–355 μm were derived from Pharmatose

110M and 80M, respectively, by 20 min vibratory sieving (Analysette 3, Fritsch, Idar-Oberstein,

Germany) followed by 20 min air jet sieving (A200, Alpine). Sweeper crystals and colistimethate

sodium were either mixed (10 min) in the indicated quantities, using a tumbling mixer with a

(160 ml) stainless steel mixing container (Turbula 2TC, WA Bachofen AG, Basel, Switzerland), or

weighed separately into the dose compartment(s) of the Twincer®.

Size distributions of the starting materials (drugs and sweeper fractions) were measured

with laser diffraction technique, using a HELOS BF MAGIC (Sympatec, Clausthal-Zellerfeld,

Germany) with standard Windox software. Powders were dispersed with a RODOS dry powder

disperser (Sympatec) at 0.5; 3 or 5 bar. Computations of diffraction data (obtained with a 100

mm lens for the drug) into particle size distributions were made with the Fraunhofer theory. For

the sweeper fractions 200 and 500 mm lenses were used. The size distributions of the aerosols

from the Twincer® and the Turbuhaler® were measured with the same laser diffraction appara-

tus (100 mm lens), and the inhalers were connected to a previously described inhaler adapter

(INHALER 2000, Sympatec; de Boer et al., 2002a). Start of the measurements was triggered on

the optical concentration in the aerosol cloud (0.2% on channel 30), and the measurements

were stopped either after the signal decreased to a value lower than 0.2% on the same channel,

or after 3 s of real measurement time.

For the in vitro deposition of the aerosol from two different Twincer® concepts, a four stage

impactor with glass constructed induction port was used (Hallworth and Andrews, 1976). Frac-

tions deposited on the impactor stages were allowed to dissolve for at least 1 h in 20 ml of

demineralised water (per stage) and the solutions were analysed with a slightly modified folin

phenol method as described by Lowry (Lowry et al., 1951). Cascade impactor results given are

the mean of two series of three inhalations each. For the calculation of the theoretical cut-off

diameters of the impactor stages, a density of 1400 kg/m3 for the colistimethate sodium was

used.

Chap

ter 5

100

Results and discussion

The de-agglomeration efficiency of the Twincer®The mechanisms of powder deagglomeration in an air classifier have been described before

(De Boer et al., 2003). Fig. 3 shows the size distribution of the aerosol from the Twincer®

concept presented in Fig. 2A for budesonide spherical pellets (taken from the Pulmicort® 200

Turbuhaler®) in comparison with that from the Turbuhaler® at 1 and 4 kPa. The dose from the

Twincer® was 2 mg with 2 mg sweeper crystals (size fraction 150–200 μm). This is 10 times

higher than the nominal dose from the Turbuhaler®. Nevertheless, the deagglomeration

efficiency of the Twincer® for this relatively high dose at 1 kpa is already as good as that of the

Turbuhaler® at 4 kPa. For comparison, the primary size distribution of the drug obtained from

RODOS dispersion at 5 bar is also shown in Fig. 3. The role of sweeper crystals in a classifier as

powder de-agglomeration principle has been described before (De Boer et al., 2002b). They

remove adhering drug particles from the inner walls of the classifier chambers during inhala-

tion and reduce the inhaler losses.

The effect of dose weight on the size distribution of the aerosol

The effect of the dose weight on the size distribution of the aerosol from different Twincer®

concepts at 4 kPa is shown in Fig. 4A for micronised colistimethate sodium. All concepts used

for the experiments had two parallel classifiers with a circular shape and each classifier had one

0

20

40

60

80

100

0.1 1 10 100 1000upper class limit (µm)

cum

ulat

ive

volu

me

perc

ent (

%)

RODOS 5 bar

Twincer 1 kPa

Twincer 4 kPa

Turbuhaler 1 kPa

Turbuhaler 4 kPa

Figure 3 Size distribution of the aerosol (from laser diffraction analysis) produced by the Twincer® concept in Fig. 2A for budesonide soft spherical pellets (used in the Turbuhaler®) at 1 and 4 kPa in comparison with the size distributions from the Turbuhaler® and the RODOS dry powder disperser. The dose from the Turbuhaler® was 0.2 mg (pure drug); the total dose from the Twincer® was 4 mg (mixture with 2 mg drug and 2 mg sweeper crystals 150-200 µm). Mean of five doses.


powder channel and two additional air channels, as shown in Fig. 1. The differences between

the concepts were confined to different diameters for the classifier chambers (11 mm for con-

cept 1 and 15 mm for the concepts 2 and 3), different heights of the rims around the discharge

holes (1.5 mm for concept 2, and 2 mm for concepts 1 and 3) and slightly different amounts of

bypass air entering the mouthpiece channel along the discharge holes. For comparison, the

7

Figure 4

A.

0

2

4

6

8

10

0 5 10 15 20 25 30

dose weight (mg pure drug)

diam

eter

(µm

)

concept 1

concept 1 + sweeper

concept 2 + sweeper

concept 3 + sweeper

B.

0

4

8

12

16

20

0.1 1 10 100

upper class limit (µm)

perc

ent i

n cl

ass

(%)

dose is 8 mg

dose is 25 mg

Figure 4 A. Size distributions of the aerosols for colistimethate sodium at 4 kPa from three different Twincer® concepts as function of the dose weight (pure drug). Closed symbols are for X50- and open symbols for X90-values from laser diffraction analysis; similar symbols (open and closed) refer to the same concept. Each data point is the mean of two experiments; the spread is too small to be indicated.B. Volume frequency distribution at 4 kPa for concept 2 in Fig. 4A for two different dose weights of 8 and 25 mg of pure drug.

Chap

ter 5

102

X50- (closed symbol) and X90-value (open symbol) of the primary drug particles (obtained with

RODOS dispersion at 3 bar) are shown on the abscissa. The dose weights on the ordinate are for

the pure drug without sweeper crystals. For concept 1, the additional amount of sweeper (frac-

tion 250–300 μm) was 15% (mixed with the drug); for the concepts 2 and 3 the same sweeper

weight of 2 mg (fraction 150–200 μm) was used for all dose weights (drug and sweeper crystals

weighed separately into the dose compartments). Fig. 4A shows that the size distribution of

the aerosol cloud from the concepts 2 and 3 is hardly influenced by the dose weight up to

(and including) 25 mg of pure micronised drug. The X50-value is constant and, which is also

proof for the excellent dispersion reproducibility of the inhaler, only 10% higher than the X50

of the primary particles from RODOS dispersion. For concept 1, only the X90-value increases

slightly with increasing dose weight and the magnitude of the increase appears to depend on

whether sweeper crystals are used or not. The reason for the increase in X90 for concept 1 is the

relatively short residence time of the powder in the classifier. For this concept, the diameter of

the classifier chamber is relatively small compared to the diameter of the discharge hole. As a

result, part of the larger drug agglomerates may already be discharged from the classifier before

de-agglomeration is complete. The effect becomes first noticeable at higher classifier payloads

when agglomerates crowd each other out, particularly when sweeper crystals are also present

in the classifier. When the residence time is slightly increased and large particle retention is

improved (as in concepts 2 and 3), the fraction of the drug released as small agglomerates

becomes almost negligible, also at higher dose weights. This is shown in Fig. 4B for concept 2.

The particle size distributions from 8 to 25 mg doses are exactly the same, which confirms that

the de-agglomeration is good at all payloads, but a minor secondary peak of small agglomer-

ates (with a peak around 9 μm) occurs at the highest payload. These agglomerates represent

only approximately 5% of the total dose.

Reduction of the inhaler losses and retention of sweeper crystals

One of the problems to solve when high drug doses are dispersed with a high efficiency is the

inhaler accumulation. Particularly when a classifier type of de-agglomeration principle is used,

fine particle adhesion onto the cylindrical classifier wall may be substantial, unless the surface

area of this wall can be reduced (e.g. by increasing the number of additional air channels: De

Boer et al., 2004) or sweeper crystals are added to the drug formulation. Surrounding a classifier

with a large number of tangential air channels has the consequence that more space must be

provided for the total classifier arrangement. With two parallel classifiers side by side in the

same plane, this increases the dimensions of the inhaler quite substantially. For that reason,

the use of sweeper crystals seems to be a better option, particularly for a single use device for

which retained crystals do not have to be removed from the classifiers after inhalation, since

the complete inhaler is disposed. The mass fraction of sweeper crystals in the formulation can

be kept relatively low to obtain the desired effect, whereas the crystals may be retained in the

classifier during inhalation to avoid deposition in the upper respiratory tract.


The effect of sweeper action on total inhaler accumulation at 4 kPa for micronised colis-

timethate sodium is shown in Fig. 5 for one particular Twincer® concept as function of the

dose weight. Without sweeper crystals the total inhaler losses are around 30% of the total drug

dose. With 15% (w/w) sweeper crystals (size fraction 250–355 μm) in the formulation, the losses

are reduced to approximately 20% of the drug dose at all dose weights. The reduction is the

same for this type of drug when other sweeper fractions in the same weight percentages are

used. The remaining fine particle losses for this concept were found in the mouthpiece channel,

particularly against the top plate above the discharge holes where the sweeper crystals are not

effective. Reduction of these losses therefore requires other means, such as the arrangement of

bypass flows directing the discharge flow from the classifiers.

The addition of sweeper crystals to the formulation could be disadvantageous, as deposition

of these crystals in the upper respiratory tract may cause irritation in the patient. In contrast

with a single air classifier having the same longitudinal axis as the inhaler mouthpiece, large

(carrier or sweeper) particles from the Twincer® are not deposited in the front of the mouth by

centrifugal action, but rather in the oropharynx. Therefore, sweeper retention may be desired.

Fig. 6 shows the retention efficiency of Twincer® concept 1 (from Fig. 4A) with different classifier

discharge holes (with diameters varying between 4 and 5 mm) and for two different sweeper

size fractions at 4 kPa. For all discharge holes, the retention of a coarse sweeper fraction (250–355

μm) is rather complete, but the retention efficiency for a smaller fraction (63–100 μm) depends

particularly on the distance between the rim around the discharge holes and the bottom of the

8

Figure 5

0

10

20

30

40

50

0 5 10 15 20 25

dose weight (mg)

inha

ler l

osse

s (%

of d

ose) without sweeper

with sweeper

Figure 5 Inhaler accumulation for colistimethate sodium at 4 kPa of concept 1 in Fig. 4A with and without (15% w/w) sweeper lactose (250-355 µm from Pharmatose 80M) as function of the dose weight (weight is for drug plus sweeper).

Chap

ter 5

104classifier chamber (varied between 0.5 and 2.5 mm), which is determined by the height of the

rim. This enables control of the degree of passage for the sweeper crystals without influencing

the deagglomeration efficiency, as shown in Fig. 4A where the rim height between concepts 2

and 3 is varied over the same distance as between the concepts D5/2 and D5/1.5 in Fig. 6. There

is neither a noticeable influence on the total inhaler resistance from the height of the rim, which

is exactly the same 0.034 kPa0.5 min lN−1 for all configurations shown in Fig. 6.

Effect of the inspiratory effort on the fine particle fraction (FPF)

The effect of the inspiratory effort (1 or 4 kPa) on the in vitro deposition of colistimethate

sodium from the Twincer® concepts 2 and 3 in Fig. 4A in a multi stage impactor is shown in Fig.

7. For these concepts (referred to as C2 and C3), 1 and 4 kPa correspond with 30 and 60 lN/min,

respectively. In Fig. 7 cumulative subfractions of particles are indicated within the size fraction

<5 μm. The total fine particle fractions (<5 μm) are of the same order of magnitude for both

concepts, but the distributions of particles within these fractions are different, particularly at 4

kPa. The difference is highly reproducible, as can be concluded from the spread bars which are

approximately the same for each of the cumulative subfractions up to 5 μm. The difference in

the subfractions may be the result of a difference in the (theoretical) cut-off diameters of the

classifiers, which are 13.6 and 9.6 μm for concept 2 and 19.2 and 13.6 μm for concept 3 at 30 and

60 l/min, respectively (for colistimethate sodium).

9

Figure 6

0

20

40

60

80

100

120

D5 D4.5 D4 D5/2.5 D5/1.5 D5/1 D5/0.5

discharge channel

swee

per r

eten

tion

(%)

sweeper 250-355 µm

sweeper 63-100 µm

Figure 6 Sweeper retention (as percent of dose weight) at 4 kPa for two different lactose size fractions by concept 1 in Fig. 4A with different configurations for the discharge channel. Numbers (on the ordinate) in combination with the letter D refer to the diameter of the discharge hole (mm). Numbers following the slash mark refer to the height of the slit between the rim around the discharge hole and the bottom of the classifier chamber. Mean of two experiments; dose weight (sweeper only) is 25 mg.


Fig. 7 shows that the fine particle fraction <5 μm is considerably higher at 4 kPa than at

1 kPa (on average 55.8% and 40.0% of the real dose for both concepts, respectively). This is

not primarily the result of an improved de-agglomeration efficiency with increasing flow rate

through the inhaler. The main reason is a much higher inhaler accumulation at 1 kPa. As shown

in Fig. 5, the inhaler losses of the concept depicted in Fig. 2 can be reduced to 20% of the real

dose by adding a small fraction of sweeper crystals. By adding also a minor bypass flow to the

discharge channels, the inhaler losses can be further reduced to <10% of the real dose at 4 kPa

(5–7% for the experiments presented in Fig. 7). The bypass flow is directed over the discharge

holes, which deflects the powder flow from these holes and reduces particle collision with the

top plate. At 1 kPa, corresponding with 30 l/min, the air velocity of the bypass flow is insuf-

ficiently high however, which makes further concept improvement necessary (inhaler losses

are on average 25.3 and 5.8% for the experiments in Fig. 7 at 1 and 4 kPa, respectively). The

difference in inhaler losses between 1 and 4 kPa (19.5%) explains fairly well the difference in fine

particle fraction <5 μm (15.8% in Fig. 7). Less than 5% of the total dose has been deposited on

the stages 1 and 2 of the impactor and less than 8% in the induction port to the impactor (the

same for both pressure drops). The ‘missing’ 25% of the dose (also the same for both pressure

drops) has been released from the inhaler within the size fraction 5–8.5 μm (at 4 kPa) and 5–12

μm (at 1 kPa), respectively. In this comparison, the upper sizes of the ‘missing’ fractions equal

the cut points of the second impactor stage at 1 and 4 kPa.

Fig. 8 confirms that the size distributions of the aerosol (from laser diffraction analysis) from

concept 2 in Fig. 7 differ only slightly between the different pressure drops. The peaks of the

10

Figure 7

0

10

20

30

40

50

60

1(C2) 1(C3) 4(C2) 4(C3)

pressure drop (kPa)

perc

ent o

f rea

l dos

e (%

) < 5.0 µm

< 3.0 µm

< 1.5 µm

Figure 7 Fine particle fractions from the Twincer® concepts 2 and 3 (in Fig. 4A, referred to as C2 and C3) collected in a four stage impactor at 1 and 4 kPa. The dose is 8 mg colistimethate sodium with 2 mg sweeper crystals (size fraction 150-200 µm). Data given are the mean of two series of three inhalations.

Chap

ter 5

106volume frequency distribution curves appear at exactly the same diameter for the aerosols

generated at 1, 2 and 4 kPa. The difference between the curves is confined to a small difference

in the volume of the larger particles. For comparison, the size distributions from RODOS disper-

sion at 0.5 bar are also given in Fig. 8. A difference between the RODOS and the Twincer® exists

only for the finest particles (<1.5 μm), which cannot be dispersed completely into primary enti-

ties by the Twincer®. In contrast, the Twincer® disintegrates larger drug agglomerates more

effectively, particularly after a slight pressure has been applied to the powder (RODOS 0.5 bar

indicated with asterisks). A slight compression of the powder (into a coherent cake) may be

necessary to fill large powder weights in the dose compartments of the Twincer® having a fixed

volume. In spite of a 12.5 times lower dispersion pressure for the Twincer® (4 kPa, which equals

0.04 bar) compared to the RODOS disperser (0.5 bar), the Twincer® de-agglomerates such a

powder cake more effectively.

Conclusions

Although the machined copies of the Twincer® used for this in vitro study with colistimethate

sodium were not fully optimised yet with respect to air flow resistance, powder entrainment

from the dose system, inhaler accumulations and discharge flow from the mouthpiece, it has

11

Figure 8

0

4

8

12

16

20

0.1 1 10 100

upper class limit (µm)

volu

me

perc

ent i

n cl

ass

(%)

RODOS 0.5 bar

RODOS 0.5 bar*

Twincer 1 kPa

Twincer 2 kPa

Twincer 4 kPa

Figure 8 Volume frequency distribution curves for the aerosol from Twincer-concept 2 (in Fig. 4A) at 1, 2 and 4 kPa (8 mg colistimethate sodium + 2 mg sweeper) in comparison with the size distributions from RODOS dispersion at 0.5 bar. Mean of two series per experiment. RODOS 0.5 bar*: drug mixture is slightly compressed (a similar compression as to fit in the dose compartment of the Twincer®: see text).


been shown that high fine particle fractions can be obtained as the result of a high de-agglom-

eration efficiency. Special particle engineering processes to reduce the interparticulate forces

are not necessary. A 2 mg dose of budesonide can be dispersed at 1 kPa with the Twincer®

with the same efficiency as a 0.2 mg dose at 4 kPa with the Turbuhaler®. For all prototypes

used in this study, the size distribution in the aerosol (at 4 kPa) is largely independent of the

dose weight between 0 and 25 mg. It is inherent in the dispersion of micronised powders in

a classifier type of de-agglomeration principle that a certain fraction of the particles is lost by

adhesion to the classifier walls. However, the addition of 10–15% sweeper crystals to the formu-

lation (in a size fraction larger than 50 μm) appears to be effective in reducing this accumulation

inside the classifiers to less than 5% of the total dose at all flow rates. The remaining inhaler

losses are found in the discharge channels but by directing bypass flows around the discharge

holes, these losses can be further reduced to a total inhaler accumulation of only 5–6% (for

colistimethate sodium) at 4 kPa. The Twincer® classifiers can be modified to retain the sweeper

crystals with high efficiency (>95%) in order to minimise excipient particle deposition in the

(upper) respiratory tract.

With the current Twincer® design, powder doses up to 25 mg of pure drug can effectively

be de-agglomerated. Further optimisation of the design could raise this to a dose of 50 mg.

The high de-agglomeration efficacy makes the inhaler suitable for highly cohesive formula-

tions, such as solid dispersions of drugs in sugar glasses. Even for these formulations, battery

powered dispersion systems or pressurised gas canisters, as for instance described by Young

et al. (2004) are not necessary. The good dispersion of high drug dose and the good moisture

protection of the drug formulation in a blister make the Twincer® also suitable for other

applications, like the administration of rhDNAse in CF. Its simple design reduces the production

costs of the Twincer®, as the three plate-like parts (with the blister) can simply be stacked and

clicked together. This makes the inhaler suitable for single use, which has several advantages.

For instance, it prevents contamination and the development of antibiotic resistant bacteria, or

inhaler pollution (e.g. for hygroscopic drug formulations). Moreover, it makes the inhaler suit-

able for single use therapy, as for instance pulmonary vaccination. It has already been shown

that the pulmonary route may be effective for vaccines against measles (LiCalsi et al., 1999;

Dilraj et al., 2000) and influenza (Jemski and Walker, 1976).

Acknowledgements

The authors would like to thank Mrs. J. Beekhuis for carefully screening the manuscript and Mr.

W. de Goede and Mr. E. Schut of the research workshop of the Faculty of Medicine (University of

Groningen) for preparing test systems and machined prototypes of the Twincer®.

Chap

ter 5

108

References

Crowther Labiris NR, Holbrook AM, Crystyn H, Macleod SM, Newhouse MT. Dry powder versus intravenous and nebulized gentamycin in cystic fibrosis and bronchiectasis. Am J Respir Crit Care Med 1999; 160: 1711–1716.

de Boer AH, Gjaltema D, Hagedoorn P, Schaller M, Witt W, Frijlink HW. Design and application of a new modular adapter for laser diffraction characterization of inhalation aerosols. Int J Pharm 2002a; 249: 233–245.

de Boer AH, Le Brun PPH, van der Woude HG, Hagedoorn P, Heijerman HGM, Frijlink HW. Dry powder inhala-tion of antibiotics in cystic fibrosis therapy, part. 1: development of a powder formulation with colistin sulphate for a special test inhaler with an air classifier as de-agglomeration principle. Eur. J. Pharm. Biopharm. 2002b; 54: 17–24.

de Boer AH, Hagedoorn P, Gjaltema D, Goede J, Frijlink HW. Air classifier technology (ACT) in dry powder inhalation part 1. Introduction of a novel force distribution concept (FDC) explaining the performance of a basic air classifier on adhesive mixtures. Int J Pharm 2003; 260: 187–200.

de Boer AH, Hagedoorn P, Frijlink HW. Dry powder inhaler and method for pulmonary inhalation of dry powder. Eur. Patent Application EP 1488819, 2004.

Dilraj A, Cutts FT, de Castro JF, et al.Wheeler JG, Brown D, Roth C, Coovadia HM, Bennet JV. Response to different measles vaccine strains given by aerosol and subcutaneous routes to schoolchildren: a ran-domised trial. Lancet 2000; 355: 798–803.

Ferrari S, Geddes DM, Alton EWFW. Barriers to and new approaches for gene therapy and gene delivery in cystic fibrosis. Adv Drug Del Rev 2002; 54: 1373–1393.

Flynn RA, Goldman MH, Lovely JR, 2000. Micronised pharmaceutical compositions. WO 00/16745 A3, 2000.Geddes DM. Nebilized therapy and cystic fibrosis. Eur Respir Rev 1997; 7: 173–176.Gerrity TR. Pathophysiological and disease constraints on aerosol delivery. In: Byron PR (Ed.), Respiratory

drug delivery. CRC Press, Boca Raton, USA, pp. 1–38, 1990.Goldman JM, Mayston SM, O’Connor S, Meigh RE. Inhaled micronised gentamicin powder: a new delivery

system. Thorax 1990; 45: 939–940.Gribbon EM, Hatley RHM, Gard T, Blair JA, Kampinga J, Roser B. Trehalose and novel hydrophobic sugar

glasses in drug stabilisation and delivery. In: Karsa DR, Stephenson RA (Eds.), Chemical Aspects of Drug Delivery Systems. RSC, London, pp. 138–145, 1996.

Groneberg DA, Witt C, Wagner U, Chung KF, Fischer W. Fundamentals of pulmonary drug delivery. Resp Med 2003; 97: 382–387.

Hallworth GW, Andrews UG. Size analysis of suspension inhalation aerosols by inertial separation methods. J Pharm Pharmacol 1976; 28: 898–907.

Jemski JV, Walker JS, 1976. Aerosol vaccination of mice with a live, temperature sensitive recombinant influenza virus. Infect Immumnol 1976; 13: 818–824.

Kim CS, Follinsbee LJ. Physiological and biomechanical factors relevant to inhaled drug delivery. In: Adjei AW, Gupta PK (Eds.), Inhalation delivery of therapeutic peptides and proteins. Marcel Dekker, New York, pp. 3–25, 1997.

Le Brun PPH, de Boer AH, Gjaltema D, Hagedoorn P, Heijerman HGM, Frijlink HW. Inhalation of tobramycin in cystic fibrosis part 1: the choice of a nebulizer. Int J Pharm 1999; 189: 205–214.

Le Brun PPH, de Boer AH, Mannes GPM, et al.de Fraîture DMI, Brimicombe RW, Touw DJ, Vinks AA, Frijlink HW, Heijerman HGM.. Dry powder inhalation of antibiotics in cystic fibrosis therapy: part 2. Inhalation of a novel colistin dry powder formulation: a feasibility study in healthy volunteers and patients. Eur J Pharm Biopharm 2002; 54: 25–32.

Li J, Turnridge J, Milne R, Nation RL, Coulthard K. In vitro pharmacodynamic properties of colistin and colistin methanesulfonate against Pseudomonas aeruginosa isolates from patients with cystic fibrosis. Antimicrob Agents Chemother 2001; 45: 781–785.


LiCalsi C, Christensen T, Bennett JV, Phillips E, Witham C. Dry powder inhalation as a potential delivery method for vaccines. Vaccine 1999; 17: 1796–1803.

Lowry OH, Rosebrough NJ, Farr AL, Randall RJ. Protein measurement with the folin phenol reagent. J Biol Chem 1951; 193: 265–275.

Martonen TB, Katz IM. Deposition patterns of aerosolized drugs within human lungs: effects of ventilatory parameters. Pharm Res 1993; 10: 871–878.

Newhouse MT, Hirst PH, Duddu SP, et al.Walter YH, Tarara TE, Clark AR, Weers JG. Inhalation of a dry powder tobramycin PulmoSphere formulation in healthy volunteers. Chest 2003; 124: 360–366.

Newman SP, Pellow PG, Clay MM, Clarke SW.. Evaluation of jet nebulisers for use with gentamicin solution. Thorax 1985; 49: 671–676.

Patton JS. Mechanisms of macromolecule absorption by the lungs. Adv Drug Del Rev 1996; 19: 3–36.Patton JS, Bukar JG, Eldon MA. Clinical pharmacokinetics and pharmacodynamics of inhaled insulin. Clin

Pharmacokinet 2004; 43:, 781–801.Ramsey BW. Management of pulmonary diseases in patients with cystic fibrosis. N Engl J Med 1996; 335:

179–188.Schultz H. Mechanisms and factors affecting intrapulmonary particle deposition: implications for efficient

inhalation therapies. Pharm Sci Technol Today 1998; 8: 336–344.Smith JJ, Travis SM, Greenberg EP, Welsh MJ. Cystic fibrosis airway epithelia fail to kill bacteria because of

abnormal surface fluid. Cell 1996; 85: 229–236.Tiddens HAWM. Detecting early structural lung damage in cystic fibrosis. Pediatr Pulmonol 2002; 34:

228–231.Touw DJ, Brimicombe RW, Hodson ME, Heijerman HGM, Bakker W., 1995. Inhalation of antibiotics in cystic

fibrosis. Eur Respir J 1995; 8: 1594–1604.Van Devanter DR, Montgommery AB. Pure biologically active colistin, its components and a colistin formu-

lation for treatment of pulmonary infections. WO 9826827, 1998.van Drooge DJ, Hinrichs WLJ, Frijlink HW. Incorporation of lipophilic drugs in sugar glasses by lyophilization

using a mixture of water and tertiary butyl alcohol as solvent. J Pharm Sci 2004; 93: 713–725.Westerman EM, De Boer AH, Le Brun PP, Touw DJ, Frijlink HW, Heijerman HG. Dry powder inhalation of

colistin sulphomethate in healthy volunteers: a pilot study. Int J Pharm 2007a; 335: 41-45.Westerman EM, De Boer AH, Le Brun PP, Touw DJ, Roldaan AC, Frijlink HW et al. Dry powder inhalation of

colistin in cystic fibrosis patients: a single dose pilot study. J Cyst Fibros 2007b; 6:284-292.Wood AJJ. Management of pulmonary disease in patients with cystic fibrosis. N Engl J Med 1996; 335:

179–188.Young PM, Thompson J, Woodcock D, Aydin M, Price R. The development of a novel high-dose pressurized

aerosol dry-powder device (PADD) for the delivery of pumactant for inhalation therapy. J Aerosol Med 2004; 17: 123–128.

6 Dry powder inhalation of colistin sulfomethate in healthy volunteers: a pilot study

Elsbeth M Westerman, Anne H de Boer, Paul PH Le Brun, Daan J Touw, Henderik W Frijlink, Harry GM Heijerman.

Int J Pharm 2007; 335:41-45.

Chap

ter 6

112

Summary

Pulmonary administration of the antimicrobial drugs colistimethate sodium (colistin sul-

fomethate) and tobramycin has been shown to be effective in slowing down pulmonary

deterioration in cystic fibrosis (CF) patients. Both drugs are administered by liquid nebulisation,

a technique known to have disadvantages. Dry powder inhalation may be an attractive alterna-

tive. We investigated inhalation of colistimethate sodium dry powder using a newly developed

Twincer® device in healthy volunteers.

Eight healthy volunteers inhaled a single dose of 25 mg colistimethate sodium dry powder

each, using the Twincer® inhaler. The median diameter (X50) of the dry powder was 1.6 μm (X10

= 0.7 μm, X90 = 3.1 μm), measured by laser diffraction technique. Pulmonary function tests were

performed before, 5 and 30 min after inhalation. Serum samples were drawn at t = 15 min, 45

min, 1.5 h, 2.5 h, 3.5 h, 5.5 h, 7.5 h and 24 h after inhalation.

The colistimethate sodium dry powder inhaler was well tolerated: no clinically relevant

effect on FEV1 was observed nor did the volunteers experience adverse effects.

Dry powder inhalation of colistimethate sodium using the Twincer® inhaler is well tolerated

by healthy volunteers. A pilot study in cystic fibrosis patients is therefore considered safe in

developing a dry powder inhalation of colistimethate sodium for everyday CF treatment.

Dry powder inhalation of colistin sulfomethate in healthy volunteers: a pilot study 113

Introduction

Cystic fibrosis (CF) is a hereditary disease affecting multiple organs. Anomalies in the respiratory

system are prominent, resulting in main disease symptoms. Mortality due to respiratory failure

is common. Due to a genetic defect of the transmembrane conductance regulator (CFTR)

gene on chromosome 7, chloride secretion in the lung is disturbed resulting in highly viscous

sputum. As a consequence mucoid clearance is reduced, allowing foreign material (bacteria)

to remain behind. Chronic infection with bacteria, especially Pseudomonas aeruginosa, induces

inflammation and fibrosis of the lung, resulting in deterioration of lung capacity.

Pulmonary administration of antimicrobial drugs, such as the anti-pseudomonal drugs colis-

timethate sodium (colistin sulfomethate) and tobramycin, has been shown to be effective in

interfering with and slowing down of pulmonary deterioration (Touw et al., 1995; Mukhopadhyay

et al., 1996; Hodson et al., 2002). Both drugs are administered by liquid nebulisation, using either

an ultrasonic or jet nebuliser technique. Although nebulisation of drugs may have advantages

for specific patient groups, routine home use of nebulisers in CF patients has some drawbacks.

Most prominent is the amount of time necessary for complete nebulisation of a dose. Including

drug preparation and cleaning afterwards, nebulisation of a dose takes approximately 15–30

min, depending on the nebuliser used. Adherence of CF patients to nebulised therapy has been

shown to be poor (Abbott et al., 1994; Conway et al., 1996; Burrows et al., 2002). Furthermore,

individual breathing pattern, physico-chemical properties of (combinations of ) inhalation

liquids and nebuliser performance play a role in the nebulisation process, all of influence on

the treatment effect.

The best approach to reduce the influence of the breathing pattern of an individual patient

on drug deposition of an inhalation dose is to design inhalation devices that depend less

strongly on patient characteristics. Adaptive aerosol delivery (AAD) is an example, as is the

AKITA® inhalation system. Dry powder inhalers (DPI’s) have a main advantage in time gain in

relation to drug administration, compared to liquid nebulisation. Furthermore, the size of a

DPI is more practical and convenient in everyday use, there is no need for an electrical power

source to administer a dose and a DPI device is generally lower in costs because of a relatively

simple design in which there is no need for electronic parts. Because of these advantages, an

increase in patient adherence is to be expected. However, patient characteristics too play a

role in successfully inhaling a dry powder drug. The currently commercially available ‘breath

controlled’ dry powder inhalers all require a minimal inspiratory flow rate in order to reach a

sufficient deagglomeration of the dry powder dose. Few inhalers have been designed in order

to diminish the influence of the patient’s ability to produce an adequate inspiratory flow rate

on the inhalation manoeuvre. The dependency on inspiratory flow rate may be reduced by the

use of additional energy for dispersion of the dose (Exubera® inhaler) or by a larger fine particle

fraction (Novolizer® inhaler) to compensate for a shift in deposition of the dose to the upper

airways at higher inspiratory flow rate. Another option is to design an inhaler which combines

Chap

ter 6

114

a high dispersion effectiveness with a relatively high internal resistance with which a maximal

fine particle fraction can be generated even at a low inspiratory effort. This technique is applied

in the current study by using the Twincer® inhaler. In CF patients, inspiratory flow is generally

not influenced by the pulmonary disease process (Sarinas et al., 1998) and therefore dry powder

inhalation may be useful.

The optimal dose of nebulised colistimethate sodium in CF is not known; most adult patients

are treated with 2 million units (approximately 160 mg) of colistimethate sodium twice daily

based on empiric experience. Administration of colistimethate sodium as a dry powder requires

a calculation of an equivalent dose to liquid nebulisation. The efficiency of the dry powder

inhaler is therefore of main importance. The aim is to administer an equivalent pulmonary

dose using the least possible number of inhalations. However, to achieve this, an inhalation

of colistimethate sodium dry powder will have to consist of milligrams instead of micrograms.

Commercially available dry powder inhalers are suited for doses in the microgram to 1 mg

range. De Boer et al. (2006) described the development of a newly dry powder inhalation device

(Twincer®), intended for disposable use, which can deliver doses effectively in the 1–25 mg

range. In vitro results of this new DPI device are promising. We investigated the inhalation of

colistimethate sodium dry powder using the Twincer® device in healthy volunteers.


Study population

Eight healthy volunteers were recruited by advertisement. Inclusion criteria were age 18–40

years, no history of chronic pulmonary disease, not pregnant or breast-feeding, non-smoking

and an informed consent. The volunteers’ demographic data are listed in Table 1. The study was

performed according to the Helsinki declaration and was approved of by the medical ethical

review board of the hospital. Volunteers were fully informed by the investigators and a written

informed consent was obtained from every volunteer.

Table 1 Volunteer demographic data

Subject Sex (M/F) Age (years) Height (cm) Weight (kg)

V1 F 33 163 64

V2 M 31 175 85

V3 F 22 176 72

V4 F 20 171 71

V5 F 21 155 54

V6 F 31 158 91

V7 F 23 173 64

V8 M 22 180 75

Mean (S.D.) 25.4 (5.3) 168.9 (9.1) 72.0 (11.9)


Study drug

Colistimethate sodium (Ph. Eur. quality, sterile) was obtained from Alpharma, Copenhagen,

Denmark. One milligram contained approximately 13,200 units. The particle size of the raw

material had to be reduced in order to obtain particles in the desired (aerodynamic) size

range of 1–5 μm. Particle size reduction was performed in the hospital pharmacy under Good

Manufacturing Practice conditions using a jet mill (50 AS, Alpine, Augsburg, Germany). A mass

median diameter (X50) of 1.6 μm was obtained (X10 = 0.7 μm, X90 = 3.1 μm), determined by laser

diffraction technique. The fine particle fraction (<5 μm) was 43.8% at 34 l/min (1 kPa), 48.0% at

48 l/min (2 kPa) and 50.6% at 67 l/min (4 kPa) from a dose of 2×12.5 mg colistimethate sodium.

The dry powder container was protected from air humidity as much as possible, in order to

prevent agglomeration of hygroscopic colistimethate sodium powder particles between mill-

ing and use. Each volunteer inhaled a colistimethate sodium dry powder mixture containing

colistimethate sodium 83.3% (25 mg) and lactose 16.7% with a machined prototype of the

Twincer® (University of Groningen, Groningen, The Netherlands; De Boer et al., 2006) on day 1.

The Twincer® inhaler devices were manually filled with 12.5 mg of micronised colistimethate

sodium in each of the two dose compartments. Approximately 2.5 mg of lactose (150–200 μm,

Pharmatose 100 M, DMV International, Veghel, The Netherlands) was added to each compart-

ment, to act as a sweeper. Preparation of the inhaler was done shortly before each administra-

tion.

Prior to inhalation the volunteers received inhalation instructions. This was done using an

empty inhaler connected to an electronic inspiratory flow measurement device (University of

Groningen, Groningen, The Netherlands). Inspiratory flow rate (l/min) and inhalation time (s)

were registered for each volunteer, after the inhalation manoeuvre could be performed consis-

tently. Participants were instructed to perform a deep and powerful inhalation during 2–3 s and

to hold their breath during a few seconds. Inhalation of the colistimethate sodium dry powder

dose was done immediately after instruction. After inhalation, the volunteers were observed

and asked for adverse effects. After inhalation, the inhaler was tested for the amount of drug

remaining in the device. The inhaler was rinsed with water and the remaining colistimethate

sodium was determined using a modification of the protein assay described by Lowry (Lowry et

al., 1951). The actual inhaled dose of colistimethate sodium was calculated by subtracting the

amount of drug remaining in the device after inhalation from the total dose weighed into the

inhaler. The next day (day 2), each volunteer swallowed 80 mg colistimethate sodium dissolved

in 3 ml NaCl 0.9% on an empty stomach, in order to investigate gastro-intestinal absorption of

colistin.

Pulmonary function tests

Pulmonary function tests were performed before, 5 and 30 min after inhalation. Forced expira-

tory volume in 1 s (FEV1) and forced vital capacity (FVC) were measured using a calibrated Mas-

terlab pneumotachograph (Jaeger, Würzburg, Germany). The volunteers received lung function

Chap

ter 6

116

test instructions. Measured pulmonary function parameters were normalized to the reference

values proposed by the European Community for Coal and Steel (Quanjer et al., 1993). Change

(percentage) in lung function is relative to baseline, and not a percentage fall of predicted. A

reduction in FEV1 of 10% or more was considered to be clinically significant.

Blood sampling and serum analysis

Venous blood samples were collected after inhalation of colistimethate sodium. Samples were

taken at 15 min, 45 min, 1.5 h, 2.5 h, 3.5 h, 5.5 h, 7.5 h and 24 h after inhalation. Furthermore,

after ingestion of colistimethate sodium on day 2, blood was collected from the volunteers at

t=1 and 3 h. Analysis of serum samples was performed using a modification of the method

described by Le Brun (Le Brun et al., 2000). A calibration curve in the concentration range of

24–120 μg/l was used. The method of analysis was shown to be linear over a concentration

range of 24–724 μg/l. A lower limit of quantification (LLQ) of 11 μg/l was calculated by the

method described by Kucharczyk (1993).


Serum concentration results were studied using the MW\Pharm software (version 3.58, Medi-

Ware, Groningen, tThe Netherlands; Proost and Meijer 1992). A population pharmacokinetic

model was made from data of all volunteers using the Bayesian iterative two stage method

(KINPOP module). Subsequently individual pharmacokinetic parameters were calculated, using

the obtained population model.

Results

Pulmonary function

No clinically relevant effect on lung function was observed from inhalation of 2×12.5 mg colis-

timethate sodium. Pulmonary function test data are listed in Table 2. None of the volunteers

experienced objective or subjective adverse effects due to inhalation of the colistin dry pow-

der. Mean peak inspiratory flow rate through the empty inhaler device was 76.9 l/min (range

68.9–86.9); mean inhalation time was 2.3 s (range 1.7–2.6). All participants were able to hold

their breath for at least 3–4 s.

Pharmacokinetic data

Colistimethate sodium inhalation pharmacokinetics could be described using a one-

compartment open model with drug input from a peripheral compartment without lag time

and drug elimination from the central compartment. An overview of individual fitted serum

concentration time curves of the volunteers is displayed in Fig. 1. Pharmacokinetic parameters

are listed in Table 3. The actual dose of colistimethate sodium inhaled by each subject was used


in the pharmacokinetic calculations. Colistin serum concentrations after oral ingestion of 80

mg of liquid colistimethate sodium on day 2 remained below the LLQ of the analytical assay

method used and were therefore not detectable. It may be concluded that no contribution of

gastro-intestinally absorbed colistimethate sodium to serum concentrations after inhalation is

to be expected.

Table 2 Volunteer pulmonary function data

Subject FEV1,t=0min ΔFEV1,t=5min ΔFEV1,t=30min FVC t=0min ΔFVC t=5min ΔFVC t=30min

V1 3.72 (124) -4.8 -1.1 4.60 (133) -2.6 -2.8

V2 4.15 (100) -1.7 -1.0 5.37 (109) +1.3 -2.2

V3 4.15 (111) -1.9 -1.0 4.26 (100) -0.7 -1.4

V4 4.00 (113) +2.5 -1.0 4.97 (123) -0.4 -3.2

V5 3.43 (118) +2.3 +3.8 3.92 (118) +2.8 +1.5

V6 3.07 (107) -1.3 -2.9 3.42 (104) +0.9 -1.5

V7 3.71 (103) -3.2 -0.8 3.96 (96) -4.6 -0.5

V8 5.46 (121) -1.1 -0.2 6.44 (120) +0.9 +0.2

Mean 3.96 (112) 4.62 (113)

SD 0.71 (9) 0.96 (13)

FEV1: forced expiratory volume in 1 s (l), % predicted. ΔFEV1: change in FEV1 compared to baseline, in %. FVC: forced vital capacity (l), % predicted. ΔFVC: change in FVC compared to baseline, in %. t = 0 min: baseline; prior to inhalation of colistin. t = 5 min: at t = 5 min after inhalation of colistin. t = 30 min: at t=30 min after inhalation of colistin. S.D.: standard deviation.

Fig. 1. Colistin dry powder inhalation (volunteers)

Chap

ter 6

118

Discussion

This pilot study was intended to investigate the feasibility, including tolerability and potential

clinical use, of a single dose of a dry powder formulation of colistimethate sodium in a newly

developed inhaler in healthy volunteers. The results are promising and justify a subsequent

study in cystic fibrosis patients.

To our knowledge, no study on dry powder inhalation of colistimethate sodium in healthy

volunteers has been published before. Le Brun et al., (2002) investigated a dry powder for-

mulation of colistin sulfate, using an inhaler prototype based on single classifier technology.

Colistimethate sodium, used in the present study, is converted in vivo into a large number of

derivatives, of which a substantial amount is turned into colistin and colistin sulfate. It has been

shown that colistin sulfate shows a stronger antimicrobial effect than the parent compound

(Schwartz et al., 1959-1960; Barnett et al., 1964; Beveridge and Martin 1967). Volunteers and

patients experienced cough as adverse effect of inhalation of colistin sulfate, but the dry pow-

der device was appreciated by the patients despite these adverse effects. In a subsequent study,

Westerman et al. (2004) found that the sulfate salt of colistin was responsible for these adverse

effects. This resulted in the use of the sulfomethate salt of colistin in the current study. The

colistimethate sodium dry powder inhalation was well tolerated by all volunteers. No clinically

relevant influence on lung function was measured, nor did any of the volunteers experience

adverse effects during or after inhalation.

Pharmacokinetic profiles, based on colistin serum concentrations, give insight into pulmo-

nary deposition of the drug and can be used for comparison of different methods of inhalation

of the same drug (Auty et al., 1987; Lipworth 1996). Serum concentration data of colistimethate

sodium and derived pharmacokinetic parameters in the current study should be compared

cautiously with data from Le Brun et al., (2002), retrieved from a different dry powder inhaler

and from both colistin sulfate (DPI) and colistimethate sodium (nebulisation). Comparison of

Table 3 Colistimethate sodium dry powder inhalation in volunteers: pharmacokinetic results

mean 95% C.I.

Actual dose (mg) 21.5 18.6-24.4

AUC0-4 (h.μg/L) 274.8 185.2-364.3

Cmax (μg/L) 89.9 59.4-120.4

tmax (h) 1.1 0.9-1.2

Ka (h-1) 2.4 1.9-2.8

t1/2,el (h) 2.75 2.68-2.82

Cl/F (L/h/kg) 51.2 36.3-66.0

Actual dose: nominal dose minus remainder of colistimethate sodium in inhaler after inhalation. AUC0–4: area under the curve from 0 to 4 h. Cmax: maximumplasma concentration. tmax: time to maximum plasma concentration.t1/2,el: terminal half-life. Cl/F: clearance following pulmonary administration;F: bioavailability (unknown). C.I.: confidence interval.


our values with pharmacokinetic data from intravenously used colistin (Reed et al., 2001; Li et

al., 2003) is difficult because of the different methods of serum sample analysis and pharma-

cokinetic analysis used.

Using serum concentrations as surrogate parameters of lung deposition implies that no

drug should be absorbed in the gastro-intestinal tract. It is generally accepted that colistin is

not absorbed after oral administration. However, no supporting data in literature were known

to us. To determine if gastro-intestinal absorption of colistimethate sodium, expressed in serum

concentration, occurs, we subjected the volunteers to an oral intake of colistimethate sodium.

The results confirm that oral intake of colistin indeed does not result in a significant contribu-

tion to serum concentration.

Successful drug inhalation depends on the individual patient and specifications of the

drug and inhaler device (Brand et al., 2000). In general, an aerodynamic particle size of 1–5

μm is required to obtain peripheral pulmonary deposition. In dry powder inhalation, a proper

balancing between the obtained particle size distribution in the aerosol and the inspiratory

flow rate is of main importance. The Twincer® inhaler has been subjected to extensive in vitro

testing in combination with an optimised formulation of colistimethate sodium. This resulted

in a relatively simple inhaler design intended for single use that is low in production costs, and

a micronised colistin dry powder formulation that, with the use of lactose sweeper crystals, can

be administered with high efficiency due to the properties of the inhaler (De Boer et al., 2006).

To prevent small particles from being exhaled and to improve pulmonary deposition (by

sedimentation and diffusion), the participants in this study were instructed to perform a

deep and powerful inhalation, lasting about 2–3 s, and subsequently to hold their breath for

a few seconds. This instruction was based upon in vitro results obtained with the same the

dry powder formulation and inhaler device (De Boer et al., 2006). The maximal inspiratory flow

rates generated by our subjects and the inhalation times were within the predefined range.

The volunteers inhaled the study drug immediately after instruction with an empty inhaler in

order to be able to compare the measured maximum inspiratory flow rate and inhalation time

to those achieved during inhalation of the drug.

The dose was calculated based upon the in vitro efficiency of the inhaler: at a constant inspira-

tory flow rate of 67 l/min (4 kPa) approximately 60% of the powder mixture is in size range of

1–5 μm (fine particle fraction, De Boer et al., 2006). Subsequently, the dose was adjusted to an

average pulmonary deposition of 10% reached after nebulisation of a 160 mg drug dose with

a Ventstream® nebuliser and a PortaNeb® compressor (Le Brun et al., 1999). The amount of

colistimethate sodium lost during inhalation due to inhaler retention (approximately 7%) in our

study is in line with the in vitro experiments by De Boer et al. (2006).

In conclusion, dry powder inhalation of 25 mg colistimethate sodium, using the newly

developed Twincer® inhaler, is well tolerated by healthy volunteers. A subsequent pilot study in

cystic fibrosis patients is therefore considered to be safe in developing a dry powder inhalation

of colistin for everyday treatment of patients with CF.

Chap

ter 6

120

Acknowledgements

The authors wish to thank Mr. P. Hagedoorn (Department of Pharmaceutical Technology and

Biopharmacy, University of Groningen) for his technical assistance in preparing the study,

Mr. G. van der Meyden (Adult Cystic Fibrosis Center, Haga Teaching Hospital) for performing

the pulmonary function tests and Mr. H. Trumpie (Clinical, Pharmaceutical and Toxicological

Laboratory, Apotheek Haagse Ziekenhuizen) for the optimisation of the colistin analysis and

assay of the serum concentrations.


References

Abbott J, Dodd M, Bilton D, Webb AK. Treatment compliance in young adults with cystic fibrosis. Thorax 1994; 49:115-120.

Auty R, Brown K, Neale M, Snashall P. Respiratory tract deposition of sodium cromoglycate is highly depen-dent upon technique of inhalation using the Spinhaler. Br J Dis Chest 1987; 81:371-380.

Barnett M, Bushby SRM, Wilkinson S. Sodium sulphomethyl derivatives of polymyxins. Brit J Pharmacol 1964; 23:552-574.

Beveridge EG, Martin AJ. Sodium sulphomethyl derivatives of polymyxins. Br J Pharmac Chemother 1967; 29:125-135.

Brand P, Friemel I, Meyer T, Schulz H, Heyder J, Haussinger K. Total deposition of therapeutic particles during spontaneous and controlled inhalations. J Pharm Sci 2000; 89:724-731.

Burrows JA, Bunting JP, Masel PJ, Bell SC. Nebulised dornase alpha: adherence in adults with cystic fibrosis. J Cyst Fibros 2002; 1:255-259.

Conway SP, Pond MN, Hamnett T, Watson A. Compliance with treatment in adult patients with cystic fibrosis. Thorax 1996; 51:29-33.


Kucharczyk N. Estimation of the lower limit of quantitation, a method detection performance parameter for biomedical assays, from calibration curves. J Chromatogr 1993; 612:71-76.

Le Brun PP, de Boer AH, Mannes GP, de Fraiture DM, Brimicombe RW, Touw DJ et al.,. Dry powder inhalation of antibiotics in cystic fibrosis therapy: part 2. Inhalation of a novel colistin dry powder formulation: a feasibility study in healthy volunteers and patients. Eur J Pharm Biopharm 2002; 54:25-32.

Le Brun PP, de Graaf AI, Vinks AA. High-performance liquid chromatographic method for the determination of colistin in serum. Ther Drug Monit 2000; 22:589-593.

Le Brun PP, Vinks AA, Touw DJ, Hekelaar N, Mannes GP, Brimicombe RW et al.,. Can tobramycin inhalation be improved with a jet nebulizer? Ther Drug Monit 1999; 21:618-624.

Li J, Coulthard K, Milne R, Nation RL, Conway S, Peckham D et al.,. Steady-state pharmacokinetics of intravenous colistin methanesulphonate in patients with cystic fibrosis. J Antimicrob Chemother 2003; 52:987-992.

Lipworth BJ. Pharmacokinetics of inhaled drugs. Br J Clin Pharmacol 1996; 42:697-705.Lowry OH, Rosebrough NJ, Farr AL, Randall RJ. Protein measurement with the folin phenol reagent. J Biol

Chem 1951; 193:265-275.Mukhopadhyay S, Singh M, Cater JI, Ogston S, Franklin M, Olver RE. Nebulised antipseudomonal antibiotic

therapy in cystic fibrosis: a meta-analysis of benefits and risks. Thorax 1996; 51:364-368.Proost JH, Meijer DKF. MW\Pharm, an integrated software package for drug dosage regimen calculation

and therapeutic drug monitoring. Comput Biol Med 1992; 22:155-163.Quanjer PH, Tammeling GJ, Cotes JE, Pedersen OF, Peslin R, Yernault JC. Lung volumes and forced ventila-

tory flows. Report Working Party Standardization of Lung Function Tests, European Community for Steel and Coal. Official Statement of the European Respiratory Society. Eur Respir J Suppl 1993; 16:5-40.

Reed MD, Stern RC, O’Riordan MA, Blumer JL. The pharmacokinetics of colistin in patients with cystic fibrosis. J Clin Pharmacol 2001; 41:645-54.

Sarinas PS, Robinson TE, Clark AR, Canfield J, Chitkara RK, Fick RB. Inspiratory flow rate and dynamic lung function in cystic fibrosis and chronic obstructive lung diseases. Chest 1998; 114:988-992.

Schwartz BS, Warren MR, Barkley FA, Landis L. Microbiological and pharmacological studies of colistin sulfate and sodium colistinmethanesulfonate. Antibiot Annu 1959-1960; 7:41-60.

Touw DJ, Brimicombe RW, Hodson ME, Heijerman HG, Bakker W. Inhalation of antibiotics in cystic fibrosis. Eur Respir J 1995; 8:1594-1604.

7 Dry powder inhalation of colistin in cystic fibrosis patients: a single dose pilot study

Elsbeth M Westerman, Anne H de Boer, Paul PH Le Brun, Daan J Touw, Albert C Roldaan, Henderik W Frijlink, Harry GM Heijerman.

J Cyst Fibros 2007; 6:284-292.

Chap

ter 7

124

Summary

Dry powder inhalation (DPI) may be an alternative to nebulisation of drugs in the treatment of

chest infections in cystic fibrosis (CF) patients. In a pilot study the feasibility of a colistimethate

sodium dry powder inhaler (prototype Twincer®) by a single dose in CF-patients was assessed

and compared to nebulised colistimethate sodium.

Ten CF-patients, chronically infected with Pseudomonas aeruginosa, participated in a ran-

domised cross over study. On two visits to the outpatient clinic, patients inhaled colistimethate

sodium as 25 mg dry powder (Twincer®) or as 158 mg nebulised solution (Ventstream® nebu-

liser, PortaNeb® compressor). Pulmonary function tests were performed before, 5 and 30 min

after inhalation. Serum samples were drawn prior to each dose and at 15, 45 min, 1.5; 2.5; 3.5

and 5.5 h after inhalation.

The DPI was well tolerated by the patients: no significant reduction in FEV1 was observed.

Relative bioavailability of DPI to nebulisation was approximately 140% based on actual dose

and approximately 270% based on drug dose label claim.

The colistimethate sodium DPI (Twincer® inhaler) is well tolerated and appreciated by

CF-patients. Optimisation with respect to particle size and internal resistance of the inhaler is

necessary to attain equivalent pulmonary deposition to liquid nebulisation.

Dry powder inhalation of colistin in cystic fibrosis patients: a single dose pilot study 125

Introduction

Inhalation of (antimicrobial) drugs is a cornerstone in the treatment of patients with cystic

fibrosis (Döring et al., 2000). The currently available antimicrobial drugs for inhalation,

tobramycin and colistimethate sodium, are commercially available as solution for nebulisation

only. Although both drugs are widely used, the lack of convenience and efficiency regarding

pulmonary administration of tobramycin and colistimethate sodium is a drawback for cystic

fibrosis patients (Le Brun et al., 2000a). Dry powder inhalation of these drugs may therefore be

an attractive alternative to nebulisation.

Nebulisation of liquids is a complex procedure that is affected by many different factors,

which may furthermore show various interactions and mutual dependencies. Major factors

influencing the outcome of nebulisation treatment with jet nebulisers are the breathing pat-

tern and adherence to therapy (Laube et al., 2000; Kettler et al., 2002; Brand et al., 2005), the

physico-chemical properties of the drug solution, the design of the nebuliser in relation to the

jet flow pressure, the cleaning of the apparatus (Hutchinson et al., 1996; Standaert et al., 1998),

the nebulisation time and the complex ‘interactions’ that may exist between (some of ) these

parameters (Hurley 1994; Ferron 1996; McCallion et al., 1996; Coates et al., 1997; Le Brun et al.,

1999a; Ho 2001; Katz et al., 2001; Standaert 2001; De Boer et al., 2003).

In contrast, dry powder inhalation is a relatively simple mode of drug administration. Major

advantages of a dry powder inhaler over nebuliser–compressor combinations are the time-

efficient way of drug administration and ease of portability. An efficient inhaler device, a dry

powder formulation with an optimal particle size in relation to the target area and an adequate

inspiratory flow rate are the basic requirements for an efficacious dry powder inhalation.

Information on dry powder inhalers has recently been reviewed (Frijlink and De Boer 2004; Rau

2005; Telko 2005). De Boer et al. (2006) described the properties of a new dry powder inhaler

concept for high drug doses in development (Twincer®), that is capable to deliver relatively

large amounts of dry powder into the lungs.

In cystic fibrosis, the main target area for pulmonary administered drugs is the peripheral

part of the airways. Therefore, dry powder particle size should be within the 1–5 μm range

(=fine particle fraction) (Bates 1966; Zanen 1996). The inspiratory flow rate should be relatively

low for dry powder particles in order to reach the small airways and to avoid impaction (due to

high flow rates) in the oropharynx as much as possible. Because the inspiratory muscles in CF

are normal and maximal inspiratory flow in most CF patients is comparable to that in healthy

volunteers, CF patients are capable of generating high flow rates through a DPI (Sarinas et al.,

1998). For these reasons, the air flow resistance of the Twincer® used for this study was kept

relatively high (0.040 kPa0.5 min l−1) in order to reduce the flow rate through the DPI.

Several clinical studies have been published on dry powder inhalation in cystic fibrosis.

Tobramycin was tested by Newhouse et al. (2003) in healthy volunteers; dry powder inhalation

of gentamicin has been described by Crowther Labiris et al. (1999) (3 CF patients and 7 patients

Chap

ter 7

126

with bronchiectasis) and Goldman et al. (1990) (forty patients undergoing routine bronchos-

copy); colistin sulfate has been tested by Le Brun et al. (2002) (6 healthy volunteers and 5 CF

patients) and colistimethate sodium in 8 healthy volunteers by Westerman et al. (2007).

The Twincer® inhaler device, used in the volunteer study by Westerman et al. (2007) and in

the current study, is a multi air classifier based inhaler under development, having two parallel

classifiers in the same plane. It has the approximate size of a credit card (Fig. 1). This dispos-

able inhaler is designed for single use and for delivery of large powder doses. The powder

dose is kept in a blister, preventing possible adverse effects of moisture on the dry powder

and subsequently on inhaler efficiency. Several inhaler experiments have been performed in

vitro. The powder formulation of colistimethate sodium is a physical mixture of the micronized

drug with sweeper crystals (16.7%). The function of these crystals, as described more in detail

previously (De Boer et al., 2006), is to sweep micronised drug particles from the walls of both

classifiers during inhalation. The crystalline particles are retained by the two classifiers because

of their large diameters (150–200 μm): only particles <10 μm are released. The micronized

colistimethate sodium in this formulation has a (volume) median diameter, measured with

laser diffraction apparatus, of 1.6 μm (X10=0.7 μm, X90=3.1 μm). The median diameter of the

colistimethate sodium in the aerosol from the Twincer® inhaler is comparable at flow rates of

30 l/min and 60 l/min. The fine particle fraction (<5 μm) was 43.8% at 34 l/min (1 kPa), 48.0% at

48 l/min (2 kPa) and 50.6% at 67 l/min (4 kPa) after a dose of 2×12.5 mg colistimethate sodium,

indicating that emission of the dose from the inhalation device is to a great extent independent

of the inspiratory flow. The in vitro tests have been described by De Boer (De Boer et al., 2006).

The objectives of this pilot study were to assess the feasibility, including tolerability and

potential clinical use of the colistimethate sodium dry powder Twincer® inhaler by a single dose

Fig. 1. Twincer® dry powder inhaler, a multi air classifier based inhaler having two parallel classifiers in the same plane.


in individual cystic fibrosis patients and to compare these data to data obtained after a single

nebulised dose of colistimethate sodium. Based on equivalence and on the characteristics of

the dry powder inhaler, a dose of 25 mg colistimethate sodium as dry powder was calculated

to be approximately equivalent to 2 million units of colistimethate sodium as solution for

nebulisation.

Design

Feasibility assessment

Tolerability was assessed by monitoring adverse effects after inhalation, by measuring pulmo-

nary function at standardised time-points and to compare these results to the data obtained

after a single nebulised dose of colistimethate sodium. Pulmonary deposition of the dry powder

was assessed by using serum concentrations of colistin components as a surrogate parameter

for peripheral pulmonary deposition and, again, compared to the serum concentrations mea-

sured after an single nebulised dose of colistin in each individual patient.

Study population

Ten patients on maintenance treatment with colistimethate sodium, visiting the outpatient

clinic, were asked to participate in an open, randomised, crossover study. Inclusion and

exclusion criteria are listed in Table 1. As this study was intended as a pilot study, applying an

inhaler under development, no special requirements were made on the inclusion (e.g. disease

state, FEV1) of the participants. Each participant was its own control. Subjects’ characteristics

are shown in Table 2. Patients were asked to stop their regular nebulisation of colistimethate

Table 1: In- and exclusion criteriaInclusion criteria:- Age ≥ 18 years- Clinical diagnosis of CF and a positive sweat test or two CF gen mutations- FEV1 > 30% predicted values- Routine use of nebulised colistin- Normal kidney function (estimated creatinin clearance > 50 ml/min)- Normal liver function (liver enzymes within normal range)- Informed consentExclusion criteria:- Exacerbation of pulmonary infection according to criteria by Fuchs et al.a

- Intravenous use of colistin- Colistin hypersensitivity- Pregnant, potentially pregnant or nursing women- Any other condition which in the opinion of the clinician would make the subject unsuitable for enrolment - Treatment with an investigational drug within a month prior to enrolmenta Fuchs HJ et al. N Engl J Med 1994;331:637–42.

Chap

ter 7

128

sodium 3 days prior to each study visit. The study was approved of by the ethical review board

of the hospital and was performed according to the Helsinki declaration.

Written informed consent to participate in the study was obtained from all patients

before commencing any protocol-specified procedure. On two separate days, the patients

inhaled a single dose of colistimethate sodium in a dry powder formulation or nebulised the

colistimethate sodium solution. The order of treatment was determined by randomisation. A

minimum of 3 days and a maximum of 10 days were allowed between visits 1 and 2.

Study drugDry powder inhalation (DPI)

One dose of the dry powder consisted of two inhalations of 12.5 mg colistimethate sodium

each. The powder mixture was composed of colistimethate sodium 83.3% and lactose sweeper

crystals 16.7%, resulting in a total dose of 25 mg colistimethate sodium and 5mg lactose

(fraction 150–200 μm from Pharmatose 100M, DMV International, Veghel, The Netherlands).

Colistimethate sodium (Ph. Eur quality, sterile, Alpharma, Copenhagen, Denmark; 13,200 units/

mg) was micronised to obtain a particle size in the 1–5 μm range, using a jet mill (50 AS, Alpine,

Augsburg, Germany). Particle size reduction and filling of the inhaler with the colistin-lactose

mixture was performed in the hospital pharmacy according to good manufacturing practice

(GMP) guidelines. Prior to inhalation the patients were trained for the inhalation manoeuvre

using an empty inhaler connected to an electronic inspiratory flow measurement device

(University of Groningen, Groningen, The Netherlands). Maximal inspiratory flow rate (l/min)

and inhalation time (s) were measured when the inhalation manoeuvre could be performed

consistently. Inhalation of the colistin dry powder dose was done immediately after instruction.

Table 2: Patient characteristics

Patient sex (M/F)

age (y)

height (cm)

weight (kg)

Body Mass Index (BMI)

FEV1 (% pred.) baseline*

FVC (% pred. baseline*)

P1 M 38 186 65 18.8 37 66

P2 F 20 168 68 24.1 61 72

P3 F 28 171 68 23.3 74 82

P4 F 24 162 58 22.1 57 98

P5 F 29 169 54 18.9 89 96

P6 F 46 166 53 19.2 46 87

P7 F 32 159 49 19.4 46 74

P8 F 35 181 83 25.3 32 66

P9 M 24 183 73 21.8 71 94

P10 M 39 184 95 28.1 104 112

Median(range)

31 (20-46)

170(159-186)

67 (49-95)

22.0 (18.1-28.1)

59 (32-104)

85 (66-112)

*Average of baseline values on day 1 and 2


Participants were instructed to perform a deep and powerful inhalation during 2–3 s and to

hold their breath during a few seconds. After inhalation, the patients were observed and asked

for adverse effects. An interval of 2 min between the two dry powder inhalations was offered in

case of adverse effects (e.g. cough). However, none of the patients used this facility.

Nebulisation

One dose of drug solution for nebulisation consisted of 2 million units colistimethate sodium.

The exact dose of 158 mg was calculated using the activity of the colistimethate sodium batch

(Ph. Eur quality, sterile, Alpharma, Copenhagen, Denmark; approx. 13,200 units/mg) and dis-

solved in 6 ml of sterile NaCl 0,9% solution by the hospital pharmacy. To prevent spill of colistin

solution due to foaming during nebulisation, the dose was transferred to the nebuliser cup

in two portions of 3 ml. The drug was nebulised using a Ventstream® jet nebuliser and a Por-

taneb® compressor (Medic Aid, Romedic, Meerssen, The Netherlands). Patients were instructed

to continue nebulisation until the inhaler started sputtering.

Actual dose

The dry powder inhaler contained a nominal dose of 25 mg (drug dose label claim), the nebuliser

cup a nominal dose of 158 mg (drug dose label claim) of colistimethate sodium. After inhalation

or nebulisation, a fraction of the dose remains in the inhalation device. For determination of

the fraction of the drug remaining in the inhalation device a modification of the Lowry protein

assay was used (Lowry et al., 1951). The actual dose is the amount of drug that has been emitted

from the inhalation device and can be calculated by subtracting the amount of drug retained in

the device from the nominal dose.

Pulmonary function

Pulmonary function tests were performed before, 5 and 30 min after inhalation or nebulisation.

Forced expiratory volume in one second (FEV1) and forced vital capacity (FVC) were measured

using a calibrated Masterlab pneumotachograph (Jaeger, Würzburg, Germany). The patients

received instructions to carry out lung function tests. Pulmonary function test results were

related to expected baseline values. Measured lung function parameters were normalized to

the reference values proposed by the European Community for Coal and Steel (Quanjer et al.,

1993). Change (percentage) in lung function is relative to baseline, and not a percentage fall of

predicted. A reduction in FEV1 of 10% or more was considered to be clinically relevant. Patients

were observed and asked for adverse effects during this period.

Blood sampling and analysis

Serum concentrations have been used in this study to reflect peripheral pulmonary deposition

in an indirect way. Venous blood samples were collected from patients prior to the colistin dose

and at 15, 45 min, 1.5; 2.5; 3.5 and 5.5 h after inhalation or nebulisation, on both study days.

Chap

ter 7

130

Serum samples were stored at −70 °C prior to analysis. Sample analysis was performed using a

modified High Pressure Liquid Chromatographic (HPLC) assay (Le Brun et al., 2000b) after forced

hydrolysation (2 h at 60 °C) with a lower limit of quantification of 11 μg/l. Serum concentrations

were determined by relating the total peak area of the E1 and E2 components of colistin to a

calibration curve. No interference of the assay by co-medication of the patients was observed.


The pharmacokinetic profile (AUC, Cmax, tmax) of colistin was used as an indirect method

to compare pulmonary deposition of the two ways of administration of colistin in patients.

Serum concentrations and patient characteristics were imported into MW\Pharm software

(version 3.58, MediWare, Groningen, The Netherlands; Proost and Meijer 1992). This computer

program offers the possibility to calculate pharmacokinetic parameters (see Table 4) for each

individual patient, based on pharmacokinetic properties of a group of patients (population),

using a Bayesian algorithm. A population pharmacokinetic model was made from all patient

data using the Bayesian iterative two stage method (KINPOP module). This resulted in an open

one compartment model with first-order elimination and extravascular absorption without

a lag time. Subsequently individual pharmacokinetic parameters were calculated, using this

population model.

Questionnaire

The patients were asked questions concerning their daily use of nebulised colistimethate

sodium in relation to the new dry powder inhalation method.


The Kolmogorov–Smirnov test was used to test the data for normal distribution. The Student’s

paired t-test for changes within groups was applied to compare both inhalation methods. Data

are expressed by means and standard deviation or confidence intervals. A p-value of 0.05 was

considered to be statistically significant.

Results

Pulmonary function test

The colistimethate sodium dry powder was well tolerated by the patients. No clinically relevant

reduction in FEV1 was seen after inhalation of the dry powder at t=5 and t=30 min. After liquid

nebulisation, patients 2 and 7 had a transient reduction in FEV1 at t=5 min of 11.4% and 12.8%

respectively, which had improved at t=30 min. Patients 4 and 7 showed a reduction in FVC at

t=5 min after dry powder inhalation of 13.9% and 16.3% and patients 2 and 7 had a reduction

of 15.1% and 12.7% after liquid nebulisation respectively. The FVC improved for all patients


after 30 min except for patient 4 after DPI. The FVC of this patient remained at approx. 95% of

predicted at t=1 and t=4 h (data not shown). See Table 3.

Mean peak inspiratory flow rate through the empty inhaler device was 67.9 L/min (range

56.5–83.8); mean inhalation time was 2.7 s (range 1.9–4.3).

Colistin pharmacokinetics

An overview of the pharmacokinetic results is displayed in Table 4. A statistically significant

difference in area under the curve (AUC0–4), maximum concentration (Cmax) and time to

maximum concentration (tmax) between DPI and nebulisation was observed. A mean relative

bioavailability (FDPI /Fneb) of 1.4 (95% C.I. −0.2 to 3.0) was calculated using the equation FDPI /

Fneb=(AUCDPI /AUCneb)×(Dneb /DDPI), based on the mean actual doses (D) and mean AUC’s. This

implies a 140% efficiency of the DPI compared to nebulisation. Based on the drug dose label

claim, FDPI /Fneb was approximately 270%. Individual fitted serum concentration time curves of

the patients after inhalation and nebulisation are displayed in Fig. 2. For each patient, the actual

dose of colistimethate sodium was used in the pharmacokinetic analysis.

Table 3: Individual FEV1 and FVC-values after dry powder inhalation and nebulisation of colistin

FEV1

DPI at t=0 min, (% predicted)

DPI at t=5 min, relative to baseline (%)


Neb. at t=0 min, (% predicted)

Neb. at t=5 min, relative to baseline (%)


patient 1 1.71 (39) -4.68 -5.26 1.55 (35) 0.00 5.16

patient 2 2.15 (63) 0.01 -0.47 2.01 (59) -11.44 -5.97

patient 3 2.55 (74) 0.39 0.38 2.56 (74) -6.25 -5.08

patient 4 1.84 (58) -8.15 -7.07 1.73 (55) -4.05 5.78

patient 5 2.84 (85) -2.11 1.41 3.07 (92) -5.86 -3.58

patient 6 1.35 (48) -3.70 -5.19 1.23 (44) 0.81 -6.5

patient 7 1.43 (50) -6.99 -5.59 1.17 (41) -12.82 -7.69

patient 8 1.20 (33) -1.67 3.33 1.13 (31) -4.42 4.42

patient 9 3.21 (69) 4.36 2.49 3.41 (73) -2.05 -3.23

patient 10 4.59 (107) -3.27 -3.7 4.31 (100) -5.8 -4.41

FVC

patient 1 3.72 (69) -4.03 -6.45 3.39 (63) -2.36 6.19

patient 2 2.81 (72) -1.07 0.00 2.79 (71) -15.05 -7.53

patient 3 3.28 (83) 1.22 -0.91 3.22 (81) -3.42 -3.11

patient 4 3.68 (102) -13.86 -11.68 3.36 (93) -4.46 8.33

patient 5 3.65 (95) -1.64 0.00 3.73 (97) -1.07 -1.07

patient 6 2.92 (89) -7.53 -4.45 2.74 (84) -5.11 -2.92

patient 7 2.57 (77) -16.34 -3.50 2.37 (97) -12.66 -1.69

patient 8 2.88 (68) -0.35 -0.35 2.69 (64) -1.49 4.83

patient 9 5.14 (93) -0.58 -0.39 5.29 (95) -3.78 -3.78

patient 10 6.10 (116) -2.95 -1.80 5.90 (112) -2.54 -3.56

Chap

ter 7

132

Questionnaire

All patients were accustomed to nebulisation of colistimethate sodium twice daily. Four patients

indicated that they experienced tolerable adverse effects during or shortly after nebulisation of

colistimethate sodium at home (Fig. 3). Daily life activities were not influenced by these effects.

Three patients used a bronchodilator in combination with colistimethate sodium nebulisation.

Nine participants rated their experience with the dry powder inhaler as ‘positive’ or ‘excellent’.

One participant did not appreciate the mouthpiece of the inhaler and scored ‘good’. Four

patients indicated the adverse effects during or after DPI as minor. Most patients experienced

transient mild cough immediately after inhalation which was classified as ‘seen with the use of

other DPIs’ and was rated as ‘minor’ or not rated at all (‘none’).

Discussion

The objectives of this pilot study were to assess the feasibility, including tolerability and potential

clinical use of the colistimethate sodium dry powder Twincer® inhaler by a single dose in cystic

fibrosis patients and to compare the collected data by this inhalation method in each individual

patient to data obtained after nebulisation of colistimethate sodium. The study was performed

in 10 CF patients and the results can therefore not be extrapolated to the CF population in

general. The inhaler was well tolerated by the patients: no clinically relevant reduction in FEV1

was observed and all patients were enthusiastic about using the inhaler. Two patients showed

a decrease in FVC following both dry powder inhalation and liquid nebulisation. Patient 4,

known for bronchial hyperreactivity, reacted on both inhalation methods. However, this patient

Table 4: Colistin dry powder inhalation in patients: pharmacokinetic results (n=10)

Mean DPI 95% C.I. Mean Neb 95% C.I. p-value

actual dose (mg) 23.2 22.0-24.3 72.5 60.8-84.3

AUC0-4 (h.μg/L) 200.7 129.0-272.4 459.5 259.0-660.0 0.02*

Cmax (μg/L) 66.3 41.4-91.1 144.3 83.7-205.0 0.02*

tmax (h) 0.86 0.80-0.93 1.34 1.23-1.46 <0.001*

t1/2,el (h) 3.2 3.1-3.3 3.0 2.7-3.4 0.23

Cl/F (L/h/kg) 1.2 0.8-1.6 1.6 1.1-2.1 0.17

FDPI/Fneb 1.4 -0.2-3.0

Actual dose: nominal dose minus remainder of colistin in inhalation device after inhalation.AUC0-4: area under the curve from 0 to 4 hrs.Cmax: maximum plasma concentration.tmax: time to maximum concentrationt1/2,el: terminal half-lifeCl/F: clearance following pulmonary administrationFDPI/Fneb: relative bioavailabilityC.I.: confidence interval* statistically significant


noticed airway narrowing after nebulisation only and not after dry powder inhalation. As the

FVC remained at approx. 95% of predicted at t=1 and t=4 h after DPI, it may possibly be that the

baseline FVC value of this patient on that day (102% of predicted, compared to baseline 93% of

predicted on the second trial day; 7 days apart) should be considered as an outlier.

The AUC0–4, Cmax and tmax after DPI (25 mg colistimethate sodium) were significantly lower

than after nebulisation (158 mg colistimethate sodium), therefore no equivalence between

both colistin administrations could be demonstrated. To our knowledge, no data on colistin

Fig. 2. Serum concentration-time curves for individual patients after DPI and nebulisation of colistin

Chap

ter 7

134

bioavailability in relation to optimal inhalation treatment are available in literature. Therefore,

no judgement can be given on the possible clinical consequence of a lower AUC0–4, Cmax and

tmax seen with the current DPI. Our research on colistin dry powder inhalation focuses on

equivalency with the current standard of colistin nebulisation. The relative bioavailability of

colistimethate sodium DPI to colistimethate sodium nebulisation was approximately 140%

(FDPI /Fneb=1.4) based on actual dose, and approximately 270% based on drug dose label

claim, showing improved deposition with the DPI. These data are indicative for the performance

of the DPI relative to the used nebuliser–compressor combination. In literature a lung dose of

nebulisers between 1% and 32% has been reported, depending on nebulising principle (f.e. jet

or ultrasonic) and the patient’s breathing pattern (Newman 1983; O’Callaghan 1997). In a previ-

ous study on aerosolized tobramycin a deposition efficiency of approximately 10% was found

(Touw et al., 1997; Le Brun et al., 1999b). Using an identical nebuliser–compressor combination,

the estimated dose of colistimethate sodium reaching the target area will therefore be approxi-

mately 16 mg. Based on a dosing efficiency of colistimethate sodium in the Twincer® inhaler

of approximately 60% (De Boer et al., 2006), the amount of colistimethate sodium dry powder

to be inhaled would be approximately 25 mg colistimethate sodium (excluding excipient) to

obtain an equivalent dose. This dose corresponds with 2 inhalations. Based on this calculation

and the narrow size distribution for the fine particles in the aerosol, at least an equivalent Cmax

or AUC0–4 between both methods of administration was expected. However, we were not able

to prove equivalence of a colistimethate sodium dry powder dose of 25 mg administered by

the Twincer® inhaler to 158 mg colistimethate sodium administered by nebulisation.

The pharmacokinetic parameters Cmax, tmax and AUC provide information concerning the

bioavailability of a drug. In a study in which dry powder inhalation is compared to nebulisation,

AUC is probably the best parameter for comparison, as a difference in administration time exists

Question Score

Concerning colistin use at home1a. How many times a day do you nebulise colistin?1b. Do you experience adverse effects during or after nebulisation?2. When do these adverse effects occur?

3. Do these adverse effects influence your daily life?

4. Do you use other (inhalation)drugs to decrease chest tightness after nebulisation?

Concerning colistin dry powder inhaler (study)1. What is your experience with the colistin dry powder inhaler? (1 = very bad, 6 = excellent)2. Did you experience adverse effects during or after nebulisation?3. When did these effects occur?

……1 – 2 – 3 – 4 – 5 – 6

0-10 min / 10-20 min / 20-30 min after nebulisation

1 – 2 – 3 – 4 – 5 – 6

yes / no

1 – 2 – 3 – 4 – 5 – 6

1 – 2 – 3 – 4 – 5 – 60-10 min / 10-20 min / 20-30 min after nebulisation

Figure 3: Questionnaire 1 = none, 2 = minor, 3 = moderate, 4 = tolerable, 5 = serious, 6 = severe


between dry powder inhalation (approx. 1 min) and nebulisation (approx. 15–20 min) that will

be of influence on the Cmax and tmax. Theoretically, many factors will have contributed to this

result. Most prominent is the relation between particle size distribution, inspiratory flow rate

and pulmonary deposition. Although a large fine particle fraction is supposed to correspond

with a high peripheral deposition and subsequently a relatively large amount of absorption

into the blood stream, this process can be suboptimal if the inspiratory flow rate is too high

(causing impaction in the oropharynx) or the particle size is too small (leading to exhalation

of the drug) or too large (when the drug does not reach the small airways). In vitro deposition

experiments do not reflect in vivo administration, and we cannot rule out the possibility that

a relatively high fraction of the dose has been deposited in the oropharynx due to the high

inspiratory flow (range 56–84 l/min) generated by the patients. Furthermore, we hypothesize

that part of the dose could have been exhaled from the periphery of the lung due to very

small particles in the aerosol despite the fact that patients were instructed to hold their breath

during a few seconds after inhalation. The absolute lower deposition after DPI compared to

nebulisation may be an explanation for the difference of observed airway narrowing seen in

some patients after nebulisation but not after DPI. A lower deposition stands for less exposure

to colistin particles and therefore a lower risk of airway narrowing.

Dry powder inhalation of the sulfate salt of colistin has been described by Le Brun et al.,

(2002). In this pilot study, colistin sulfate was administered using an inhaler based on the cyclone

principle by de Boer et al., (2002). Although the dry powder inhalation was appreciated by the

patients, some experienced serious cough as adverse effect. This appeared to be a result of the

chemical structure of the colistin sulfate salt and it was concluded that colistimethate sodium,

which is currently used by many CF patients by nebulisation, is the colistin salt of choice for

further development of a dry powder inhaler (Westerman et al., 2004). Serum concentration

results from the study by Le Brun (Le Brun et al., 2002) and the current study should be com-

pared with caution, as data obtained in the Le Brun study are retrieved from a different dry

powder inhaler and from both colistin sulfate (DPI) and colistimethate sodium (nebulisation)

administration, while only colistimethate sodium was used in the current study.

Interindividual variation in serum concentrations was remarkable and corresponds with

other data on dry powder inhalation (Westerman et al., 2007, Cochrane et al., 2000; Le Brun et

al., 2002; Aswania et al., 2004). The population in this study consisted of CF patients in different

conditions regarding FEV1 (range 32–104%) and this may have contributed to the observed

range of serum concentrations. Little is known about the absorption mechanism of colistin

from the (peripheral) airways into the bloodstream. Several mechanisms have been described

by which drug molecules can enter the bloodstream from the alveoli (Patton 1996; Tronde et al.,

2003) but no data are available on absorption of colistin, nor is there information on comparing

absorption of fluid particles to dry particles. Next to physical and chemical properties of the

drug in question, lung morphology (for example extent of fibrosis and amount of sputum)

probably plays an important role. No contribution to serum concentrations is to be expected as

Chap

ter 7

136

a result of gastrointestinal absorption following ingested colistimethate sodium during inhala-

tion (Westerman et al., 2007).

A future study will focus on using the Twincer® inhaler with a dry powder formulation with

a slightly larger mass median diameter, an intermediate inspiratory flow rate (40–50 l/min) and

a breath hold of approximately 10 s. The inspiratory flow rate is inversely proportional to the

internal resistance of an inhaler (Sarinas et al., 1998; Kettler et al., 2002). Applying an intermedi-

ate inspiratory flow rate when using an identical Twincer® inhaler will result in adjustment of

the aerodynamic deposition behaviour of the dry particles and an optimal deposition in the

target area of the lung. Secondly, it is expected that the larger particle size will reduce the frac-

tion of the dry powder that is exhaled, resulting in a higher (peripheral) deposition, which will

be consolidated by breath holding.

In conclusion, this first clinical study on dry powder inhalation of colistimethate sodium

using the Twincer® inhaler has shown that the drug is well tolerated and appreciated by cystic

fibrosis patients. Optimisation with respect to particle size and inspiratory flow rate is necessary

to attain comparable pulmonary deposition after liquid nebulisation. This will be incorporated

in a future study.

Acknowledgements

The authors wish to thank Mr. P. Hagedoorn (Department of Pharmaceutical Technology and

Biopharmacy, University of Groningen) for his technical assistance in this study, Mr. G. van der

Meyden (Adult Cystic Fibrosis Center Haga Teaching Hospital) for performing the pulmonary

function tests and Mr. H. Trumpie (Clinical, Pharmaceutical and Toxicological Laboratory, Apo-

theek Haagse Ziekenhuizen) for the optimisation of the colistin analysis and assay of the serum

concentrations.


References

Aswania O, Ritson, S., Iqbal, SM, ., Bhatt, J., Rigby AS, A.S., Everard, M.L. Intra-subject variability in lung dose in healthy volunteers using five conventional portable inhalers. J Aerosol Med. 2004; 17:231-238.

Bates DV, Fish BR, B.R., Hatch TF, T.F., Mercer TT, T.T., Morrow, P.E. Deposition and retention models for internal dosimetry of the human respiratory tract. Task group on lung dynamics. Health Phys 1966; 12:173-207.

Brand P, Meyer T, Haussermann S, Schulte M, Scheuch G, Bernhard T et al. Optimum peripheral drug deposi-tion in patients with cystic fibrosis. J Aerosol Med 2005; 18:45-54.

Coates AL, MacNeish CF, Meisner D, Kelemen S, Thibert R, MacDonald J et al. The choice of jet nebulizer, nebulizing flow, and addition of albuterol affects the output of tobramycin aerosols. Chest 1997; 111:1206-1212.

Cochrane MG, Bala MV, M.V., Downs KE, K.E., Mauskopf, J., Ben-Joseph RH., R.H.. Inhaled corticosteroids for asthma therapy: patient compliance, devices, and inhalation technique. Chest 2000; 117:542-550.

Crowther Labiris NR, Holbrook AM, Chrystyn H, MacLeod SM, Newhouse MT. Dry powder versus intrave-nous and nebulized gentamicin in cystic fibrosis and bronchiectasis. Am J Respir Crit Care Med 1999; 160:1711-1716.

De Boer AH, Hagedoorn P, Frijlink HW. The choice of a compressor for the aerosolisation of tobramycin (TOBI) with the PARI LC PLUS reusable nebuliser. Int J Pharm 2003; 268:59-69.

De Boer AH, Le Brun PP, van der Woude HG, Hagedoorn P, Heijerman HG, Frijlink HW. Dry powder inhalation of antibiotics in cystic fibrosis therapy, part 1: development of a powder formulation with colistin sulfate for a special test inhaler with an air classifier as de-agglomeration principle. Eur J Pharm Biopharm 2002; 54:17-24.

de Boer AH, Hagedoorn P, Westerman EM, Le Brun PPH, Heijerman HGM, Frijlink HW. Design and in vitro performance testing of multiple air classifier technology in a new disposable inhaler concept (Twincer®) for high powder doses. Eur J Pharm Sci 2006;28:171–178.


Ferron GA, Busch, B., Gebhart, J., Roth, C. Evaporation of aerosols produced with eleven jet nebulizers. J Aerosol Sci 1996; 27:S381-S382.

Frijlink HW, De Boer AH. Dry powder inhalers for pulmonary drug delivery. Expert Opin Drug Deliv 2004; 1:67-86.

Goldman JM, Bayston SM, O’Connor S, Meigh RE. Inhaled micronised gentamicin powder: a new delivery system. Thorax 1990; 45:939-940.

Ho SL, Kwong WTJ, W.T.J., O’Drowsky, L., Coates, A.L. Evaluation of four breath-enhanced nebulizers for home use. J Aerosol Med 2001; 7:467-475.

Hurley PK, Smye SW, S.W., Cunliffe, H. Assessment of antibiotic aerosol generation using commercial jet nebulizers. J Aerosol Med 1994; 7:217-228.

Hutchinson GR, Parker S, Pryor JA, Duncan-Skingle F, Hoffman PN, Hodson ME et al. Home-use nebulizers: a potential primary source of Burkholderia cepacia and other colistin-resistant, gram-negative bacteria in patients with cystic fibrosis. J Clin Microbiol 1996; 34:584-587.

Katz SL, Ho SL, Coates AL. Nebulizer choice for inhaled colistin treatment in cystic fibrosis. Chest 2001; 119:250-255.

Kettler LJ, Sawyer SM, Winefield HR, Greville HW. Determinants of adherence in adults with cystic fibrosis. BMJ 2002; 57:459-464.

Laube BL, Jashnani R, Dalby RN, Zeitlin PL. Targeting aerosol deposition in patients with cystic fibrosis: effects of alterations in particle size and inspiratory flow rate. Chest 2000; 118:1069-1076.

Le Brun PP, de Boer AH, Gjaltema D, Hagedoorn P, Heijerman HG, Frijlink HW. Inhalation of tobramycin in cystic fibrosis. Part 1: the choice of a nebulizer. Int J Pharm 1999a; 189:205-214.

Le Brun PP, Vinks AA, Touw DJ, Hekelaar N, Mannes GP, Brimicombe RW et al. Can tobramycin inhalation be improved with a jet nebulizer? Ther Drug Monit 1999b; 21:618-624.

Chap

ter 7

138

Le Brun PPH, de Boer AH, Heijerman HGM, Frijlink HW. A review of the technical aspects of drug nebuliza-tion. Pharm World Sci 2000a; 22:75-81.

Le Brun PPH, de Graaf AI, Vinks AATMM. High-performance liquid chromatographic method for the deter-mination of colistin in serum. Ther Drug Monit 2000b;5:589–593.


Lowry OH, Rosebrough NJ, Farr AL, Randall RJ. Protein measurement with the folin phenol reagent. J Biol Chem 1951; 193:265-275.

McCallion ONM, Taylor KMG, Bridges PA, Thomas M, Taylor AJ. Jet nebulisers for pulmonary drug delivery. Int J Pharm 1996; 130:1-11.

Newhouse MT, Hirst PH, Duddu SP, Walter YH, Tarara TE, Clark AR et al. Inhalation of a dry powder tobramycin PulmoSphere formulation in healthy volunteers. Chest 2003; 124:360-366.

Newman SP. Deposition and effects of inhalation aerosols. AB DRACO, Lund, Sweden; 1983: 13-22;O’Callaghan C, Barry, P.W. The science of nebulised drug delivery. Thorax 1997; 52 (Suppl 2):S31-S44.Patton JS. Mechanisms of macromolecule absorption by the lungs. Adv Drug Deliv Rev 1996; 19:3-36.Proost JH, Meijer DKF. MW\Pharm, an integrated software package for drug dosage regimen calculation

and therapeutic drug monitoring. Comput Biol Med 1992; 22:155-163.Quanjer PH, Tammeling GJ, Cotes JE, Pedersen OF, Peslin R, Yernault JC. Lung volumes and forced ventila-

tory flows. Report Working Party Standardization of Lung Function Tests, European Community for Steel and Coal. Official Statement of the European Respiratory Society. Eur Respir J Suppl 1993; 16:5-40.

Rau JL. The inhalation of drugs: advantages and problems. Respir Care 2005; 50:367-382.Sarinas PS, Robinson TE, Clark AR, Canfield J, Chitkara RK, Fick RB. Inspiratory flow rate and dynamic lung

function in cystic fibrosis and chronic obstructive lung diseases. Chest 1998; 114:988-992.Standaert TA, Bohn SE, S.E., Aitken ML, M.L., Ramsey, B. The equivalence of compressor pressure-flow rela-

tionships with respect to jet nebulizer aerosolization characteristics. J Aerosol Med 2001; 14:31-41.Standaert TA, Morlin GL, Williams-Warren J, Joy P, Pepe MS, Weber A et al. Effects of repetitive use and clean-

ing techniques of disposable jet nebulizers on aerosol generation. Chest 1998; 114:577-586.Telko MJ, Hickey, A.J. Dry powder inhaler formulation. Respir Care. 2005; 50:1209-1227.Touw DJ, Jacobs FA, Brimicombe RW, Heijerman HG, Bakker W, BreimerBriemer DD. Pharmacokinetics

of aerosolized tobramycin in adult patients with cystic fibrosis. Antimicrob Agents Chemother 1997; 41:184-187.

Tronde A, Norden, B., Marchner, H., Wendel AK, A.K., Lennernas, H., Bengtsson, U.H. Pulmonary absorption rate and bioavailability of drugs in vivo in rats: structure-absorption relationships and physicochemical profiling of inhaled drugs. J Pharm Sci 2003; 92:1216-1233.

Westerman EM, Le Brun PPH, Touw DJ, Frijlink HW, Heijerman HGM. Effect of nebulized colistin sulphate and colistin sulphomethate on lung function in patients with cystic fibrosis: a pilot study. J Cyst Fibros 2004; 3:23-28.

Westerman EM, de Boer AH, Le Brun PPH, Touw DJ, Frijlink HW, Heijerman HGM. Dry powder inhalation of colistin sulfomethate in healthy volunteers: A pilot study. Int J Pharm 2007; 335:41-45.

Zanen P, Go LT, L.T., Lammers, J.W. Optimal particle size for beta 2 agonist and anticholinergic aerosols in patients with severe airflow obstruction. Thorax 1996; 51:977-980.

8 Dry powder inhalation of colistimethate sodium in cystic fibrosis patients using the Twincer® inhaler: pulmonary deposition after adapted conditions

Elsbeth M Westerman, Anne H de Boer, Paul PH Le Brun, Daan J Touw, Albert C Roldaan, Henderik W Frijlink, Harry GM Heijerman

to be submitted

Chap

ter 8

140

Summary

The aim of this study was to investigate the effect of an adapted particle size distribution of

a colistimethate sodium (colistin sulfomethate) dry powder dose inhaled at an intermediate

inspiratory flow rate with an injection moulded, ready-to-use Twincer® inhaler in order to

improve lung deposition.

Seven CF patients participated in an open, randomised, cross over, single dose pilot study.

A 25 mg dry colistimethate sodium dry powder dose was compared with a 158 mg liquid dose

administered with a Ventstream® nebuliser connected to a PortaNeb® compressor. The dry

powder dose was well tolerated by the patients. Cmax- and AUC- values after dry powder inhala-

tion were significantly lower compared to jet nebulisation. However, relative bioavailability

(Fdpi/Fneb) was 3.0 (nominal dose) and 1.4 (actual dose) respectively. The effect of the adapted

inhalation conditions (larger median particle diameter of 2.1 μm and an inspiratory flow rate

of 40-45 l/min) on lung deposition could not be established with the pharmacokinetic method

used and gamma scintigraphy may be of additional value in future studies. The results support

further research on dry powder inhalation of drugs in CF using the Twincer® inhaler.

Dry powder inhalation of colistimethate sodium using the Twincer®: pulmonary deposition after adapted conditions 141

Introduction

Drug targeting to the lungs of CF patients, especially relevant to antibiotics, receives increasing

attention nowadays. Administration of antibiotics by inhalation improves lung function and

reduces the number of exacerbations and hospital admissions (Murphy et al., 2004; Flume et

al., 2007). A majority of CF patients inhale anti-pseudomonal drugs on a daily basis. Accord-

ing to the US Cystic Fibrosis Foundation patient registry, 61.8% of the patients registered in

2006 were eligible for treatment with an anti-pseudomonal drug for inhalation (tobramycin;

no data provided for colistimethate sodium (colistin sulfomethate)), corresponding with over

15000 patients who inhale an anti-pseudomonal drug in the USA every day (Cystic Fibrosis

Foundation, 2006). The median survival of CF patients has risen from approximately 25 years

in 1985 (Davis 2006) up to 36.9 years in 2006 (Cystic Fibrosis Foundation, 2006). It is likely that

inhaled antibiotics will have contributed considerably to this (Frederiksen et al., 1996), given

the fact that survival is related to (relative) lung function (FEV1) (Kerem et al., 1992) and anti-

pseudomonal drugs have proven to slow down pulmonary deterioration (Ryan et al., 2003).

However, little is known on the optimal dose (Brochet et al., 2007), optimal particle size and the

optimal inhalation method in relation to patient characteristics (e.g. age, disease progression)

for these antibiotics. Furthermore, patient adherence to inhalation therapy is known to be

poor (Arias Llorente et al., 2008), mainly because of the time and effort needed, twice daily, to

aerosolize a drug dose and to clean the nebuliser. Therefore, optimisation of, and particularly

time saving in inhalation therapy for CF patients is likely to further improve current treatment

results.

Dry powder inhalation of anti-pseudomonal drugs is expected to have several advantages

over conventional jet nebulisation techniques: patient adherence to inhalation therapy is

expected to improve whereas the risk of contamination can be eliminated when using a dis-

posable inhaler (this study). Several publications on dry powder inhalation of antibiotics in CF

have been released recently (Geller et al., 2007; Pilcer et al., 2008). Tobramycin dry powder has

been inhaled by 90 (Geller et al., 2007) and 9 CF patients (Pilcer et al., 2008) with promising

results. Dry powder inhalation of the anti-pseudomonal drug colistimethate sodium has been

described in a study with healthy volunteers and CF patients (Westerman et al., 2007a, 2007b).

The dry powder drug configuration was well tolerated and appreciated by the CF patients but

the pharmacokinetic parameters Cmax and AUC after powder inhalation of the 25 mg dose of

colistimethate sodium from the Twincer® were lower than observed after a 158 mg dose of

colistimethate sodium using the Ventstream® nebuliser and PortaNeb compressor. However,

the relative bioavailability of the DPI was 2.7 times higher with reference to the nominal dose.

The clinical pilot study described here is a sequel to the previous patient study in which

a prototype (machined) inhaler was used under suboptimal conditions for (peripheral) lung

deposition with respect to powder formulation and inhalation manoeuvre (Westerman et al.,

2007b). The goal of the present study was to investigate lung deposition of colistimethate

Chap

ter 8

142

sodium with an adapted particle size distribution administered with a ready-to-use Twincer®

at an intermediate inspiratory flow rate. The patients were asked for a breath hold of approxi-

mately 10 seconds following inhalation and lung deposition from dry powder inhalation was

compared to that from liquid nebulisation (Darquenne et al., 2000; Usmani et al., 2005).

Materials and MethodsStudy population

After having given written informed consent, seven CF patients visiting the outpatient clinic

and on long term treatment with aerosolized colistimethate sodium or tobramycin participated

in an open, randomised, cross over, single dose pilot study. Demographics of the patients stud-

ied are presented in Table 1. As this study was intended as a pilot study, applying an inhaler

under development, no special requirements were made on the inclusion (e.g. disease state,

FEV1) of the participants. Each participant was its own control. Patients were asked to stop their

regular nebulisation of colistimethate sodium 3 days prior to each study visit. On two separate

days, the patients inhaled a single dose of colistimethate sodium in a dry powder formulation

or nebulised the colistimethate sodium solution. The order of treatment was determined by

randomisation. A minimum of 3 days and a maximum of 10 days were allowed between visits

1 and 2.

The study was approved of by the regional ethical review board and was performed accord-

ing to the Helsinki declaration. Inclusion criteria were age ≥18 years, clinical diagnosis of CF

and a positive sweat test or two CF gen mutations, FEV1 >25% of predicted values, routine

use of nebulised colistimethate sodium, normal kidney function (estimated creatinin clearance

>50 ml/min), normal liver function (liver enzymes within normal range) and informed consent.

Exclusion criteria were exacerbation of pulmonary infection (according to criteria by Fuchs et

al., 1994), intravenous use of colistimethate sodium, colistimethate sodium hypersensitivity,

Table 1: Patient characteristics

Patient sex (M/F)

age (y)

height (cm)

weight (kg)

Body Mass Index (BMI)

FEV1 (% pred. baseline)*

FVC (% pred. baseline)*

P1 F 29 164 55 20.4 58.1 78.9

P2 M 29 170 70 24.2 28.2 58.9

P3 F 37 181 83 25.3 34.2 69.5

P4 F 26 162 58 22.1 47.1 82.1

P5 F 20 154 59 24.8 99.1 104.6

P6 F 29 160 55 21.4 59.8 72.9

P7 M 24 173 58 19.3 56.8 78.7

Median (range)

29 (20-37)

164(154-181)

58(55-83)

22.1 (19.3-25.3)

56.8 (28.8-99.1)

78.7 (58.9-104.6)

*Average of baseline values on day 1 and 2


pregnancy, suspected pregnancy, breast feeding or any other condition which in the opinion

of the clinician would make the subject unsuitable for enrolment and treatment with an

investigational drug within a month prior to enrolment. Each patient inhaled a dry powder

dose of colistimethate sodium with the Twincer® (University of Groningen, Groningen, The

Netherlands) and a wet aerosol of colistimethate sodium with the Ventstream® inhaler driven

by the PortaNeb® compressor (Medic Aid, Romedic, Meerssen, The Netherlands).

Drug and devices

One dry powder dose consisted of 25 mg colistimethate sodium (Ph. Eur quality, sterile, Alp-

harma, Copenhagen, Denmark; 13,200 units/mg), micronised in a jet mill (50 AS, Alpine, Augs-

burg, Germany), plus 4 mg of lactose sweeper crystals (fraction 150–200 μm from Pharmatose

100 M, DMV International, Veghel, The Netherlands). One dose was administered by a single

inhalation. The (volume) median diameter (X50) of the colistimethate sodium used in this study

was 2.1 µm (X10 = 0.9 µm, X90 = 3.8 µm), which is 1.3 times larger than that of the powder used in

the first colistin DPI study (X10= 0.7 μm, X50 = 1.6 μm, X90 = 3.1 μm)(Westerman et al., 2007b).

Injection moulded Twincer® inhalers were used, slightly adapted compared to the devices

used in the previous study in the sense that they were designed with one dose compartment

in which a blister was placed. PVC-coated aluminium blisters were supplied by Tommy Nielsen,

Esbjerg, Denmark and sealed with a peelable lid after filling one dose for inhalation with a Uni-

versal 301 FS blister machine (Tommy Nielsen). The dry powder doses were weighed in by hand

into the blisters according to good manufacturing practice (GMP) guidelines and subsequently

the inhaler parts were assembled using an ultrasonic welding machine (Rinco Ultrasonics,

Switzerland).

Patients were instructed to inhale at a moderate flow rate (40-45 L/min) and to hold their

breath subsequently for 10 seconds with a minimum of 7 seconds. Breath hold periods

were timed by the investigator using a stopwatch. Prior to inhalation, patients were given

the opportunity to practise the inhalation manoeuvre by inhaling though an instrumented

dummy-Twincer®. All inhalation flow curves were recorded and as soon as a series of consis-

tent manoeuvres was obtained, the colistimethate sodium dry powder inhaler was given to

the patient for administration of the drug. In case of a visually incomplete discharge of the

dose from the blister, the patient was asked to repeat the inhalation procedure. Patients were

observed and asked for adverse effects after inhalation and during the study day. From the

consistent flow recordings the (mean) peak, average flow rate and the inhaled volume were

computed.

A dose of 2 million units or 158 mg of colistimethate sodium (Ph. Eur quality, sterile, Alpharma,

Copenhagen, Denmark; approx. 13,200 units/mg) was dissolved in 6 ml of sterile isotonic saline

by the hospital pharmacy shortly before nebulisation and was transferred into the Ventstream®

nebuliser cup in two portions. An expiration filter was connected to the nebuliser. Patients were

instructed to continue nebulisation until the inhaler started sputtering.

Chap

ter 8

144

The drug output (actual dose) of the inhalation device was calculated by subtracting the

retained mass of drug in the device from the nominal dose originally put in the device.

Drug retention was determined by weighing the Twincer® before and after inhalation (ana-

lytical balance, Mettler-Toledo, The Netherlands) and by using a modified Lowry protein assay

for the Ventstream® nebuliser (Lowry et al., 1951).

Study parameters

Pulmonary function tests (FEV1, FVC) were performed before, 5 and 30 min after inhalation or

nebulisation, using a calibrated Masterlab pneumotachograph (Jaeger, Würzburg, Germany).

The patients received instructions to carry out lung function tests. Measured lung function

parameters were normalized to the reference values proposed by the European Community

for Coal and Steel (Quanjer et al., 1993) and the results were related to the predicted baseline

values for each patient. Any change (%) in lung function after drug inhalation was expressed

relative to the patient’s actual baseline values, and is therefore not related to the predicted

baseline values. A reduction in FEV1 of 10% or more was considered to be a clinically relevant

proof of airway reactivity.

Venous blood sampling, storage and sample analysis of serum sampling was performed

identically to the former study (Westerman et al., 2007b). Similarly, no interference of the assay

by co-medication of the patients was observed. Individual pharmacokinetic profiles were cal-

culated based on the output/actual dose for each patient, using the method described earlier

(Westerman et al., 2007b) while a new population pharmacokinetic model was made based

upon data of the 7 patients. The results have been used as an indirect reflection of peripheral

pulmonary deposition.

The same questionnaire that has been used previously (Westerman et al., 2007b) was applied

to get an impression of the patient’s experience with the dry powder inhaler compared to wet

nebulisation at home. Answers were given on a 6 scale scoring system (none (1) - severe (6) or

very bad (1) – excellent (6)).


The data were tested for a normal distribution using the Kolmogorov-Smirnov test. The Student’s

paired t-test for changes within groups was applied to compare both inhalation methods. Data

are expressed by means and confidence intervals. The significance level was set at p=0.05.

ResultsPulmonary function tests

Both colistimethate sodium doses (either from Twincer® or from Ventstream®) were well toler-

ated by all patients. Patient 1 suppressed a cough reflex during inhalation of the dry powder

which is likely to have negatively influenced the output of the inhaler (15.2 mg).


Lung function tests after dry powder inhalation showed no abnormalities. See Table 2. Patient

2 and 4 experienced chest tightness both after dry powder inhalation and nebulisation (see

results Questionnaire). No airway narrowing (fall in FEV1 > 10%) was observed after dry powder

inhalation. Nebulisation caused a fall in FEV1 of 13.5% in patient 7 shortly after nebulisation, which

improved to a decrease of 9.7% at t = 30 min. Patients 3 and 4 had a post-nebulisation fall in FVC

of 10.5% and 14.8% respectively, which after 30 minutes was unchanged for patient 3 (12.5%)

but improved for patient 4. Mean inspiratory flow rate was 35.3 L/min (range 33.7-37.3 L/min, C.V.

4%) and peak inspiratory flow rate 43.9 L/min (range 40.6-48.8 L/min, C.V. 6%), corresponding

with a calculated inhaled volume of 1.8 L (range 0.9-2.4 L, C.V. 28%). No correlations were found

between mean inspiratory flow rate, peak inspiratory flow rate, inhaled volume and AUC or Cmax.

All patients were able to hold their breath for at least 7 seconds after inhaling the dry powder.

Colistimethate sodium output and pharmacokinetics

The retained drug mass in the DPI, calculated with reference to the nominal dose, was 25.5%

(range 20.2-38.0%) which is a substantial increase compared to the 7.6% (range 3.0-22.0%) in

the previous study. The residual volume in the Ventstream® after nebulisation was 33.2% (range

20.3-47.5%) which is lower than the approximately 40% with the same nebuliser-compressor

combination observed in other studies (Le Brun et al., 2002; Westerman et al., 2007. However,

Table 2: Individual FEV1 and FVC-values after dry powder inhalation and nebulisation of colistimethate sodium

FEV1

DPI at t=0 min, (% predicted)



Neb. at t=0 min, (% predicted)



patient 1 1.84 (58) -2.7% -0.5% 1.83 (58) -7.1% -5.5%

patient 2 1.11 (28) -8.1% -9.0% 1.14 (29) -6.1% -2.6%

patient 3 1.18 (33) 3.4% 5.9% 1.30 (36) -6.2% -5.4%

patient 4 1.47 (47) -6.8% -6.8% 1.48 (47) -0.7% 4.7%

patient 5 2.85 (100) 3.2% 3.9% 2.81 (98) -6.1% 2.1%

patient 6 1.74 (58) -3.5% -5.2% 1.84 (61) -2.2% -2.2%

patient 7 2.42 (57) -3.3% -1.7% 2.38 (56) -13.5% -9.7%

FVC

patient 1 2.87 (79) -1.4% 1.1% 2.84 (79) -0.70% 0.0%

patient 2 2.79 (59) -5.7% -6.5% 2.74 (58) -9.9% -7.7%

patient 3 2.92 (70) 1.4% -1.0% 2.87 (69) -10.5% -12.5%

patient 4 2.79 (78) -4.7% 1.8% 3.10 (86) -14.8% -4.5%

patient 5 3.44 (105) 2.9% 1.2% 3.42 (104) -6.1% -1.5%

patient 6 2.49 (72) -2.4% -2.0% 2.53 (73) -2.8% -0.8%

patient 7 3.92 (79) -0.5% -2.3% 3.91 (79) -9.5% -7.7%

Baseline values in liters (% predicted)Percentages are given relative to baseline

Chap

ter 8

146

data from colistimethate sodium measurements in the expiration filter (data not shown)

indicated a higher collected amount in this study compared to the previous study. As a result,

the estimated inhaled fraction will have been lower which is an explanation for the lower Cmax

and AUC found. Despite this, the relative bioavailability based on the output (actual dose) is

comparable to that calculated in the previous study (Fdpi/Fneb is 1.4), whereas based on the

nominal dose it is even higher (Fdpi/Fneb is 3.0). These and other pharmacokinetic parameters

are displayed in Table 3. Also in this study the differences between the AUC0-4, Cmax and tmax

between the two inhalations were statistically significant, while t1/2 and CL/F, obviously, were

not. Variation in PK parameters, expressed in C.V., was large. Figure 1 displays the individual

fitted serum concentration time curves after inhalation.

Questionnaire

Four patients were very positive about the inhaler concept (score 6), because of the ease of

use, including not having to prepare the drug prior to administration and to clean the inhaler

afterwards, better hygiene and an expected improvement in adherence to inhalation treatment.

One patient was positive but with a lower score (score 5) because she thought it difficult to

inhale at the prescribed inhalation flow and two patients were only moderately content due to

powder particle retention in the mouth and a strong taste or a cough reflex during inhalation.

These patients therefore scored 3 with respect to adverse effects. One patient scored 2 as an

adverse effect because of a dry, itching throat and two other patients experienced some chest

Table 3: Colistimethate sodium dry powder inhalation in patients: pharmacokinetic results (n=7)

Mean DPI 95% C.I. C.V. (%) Mean Neb 95% C.I. C.V. (%) p-value

Actual dose (mg) 18.6 18.2-19.1 9 55.7 51.3-60.1 28

AUC0-4 (h.μg/L) 182.5 117.7-247.3 48 380.3 334.5-426.1 43 0.02*

Cmax (μg/L) 62.7 38.4-86.9 52 122.5 107.7-137.3 43 0.02*

tmax (h) 0.74 0.70-0.79 9 1.16 0.74-1.59 13 <0.001*

t1/2,el (h) 2.9 2.8-3.0 10 2.8 2.4-3.3 6 0.50

Cl/F (L/h/kg) 1.1 1.0-1.2 34 1.5 1.4-1.7 42 0.09

FDPI/Fneb 1.4 0.4-2.4 30

Actual dose nominal dose minus remainder of colistin in inhalation device after inhalation.AUC0-4 area under the curve from 0 to 4 h.Cmax maximum plasma concentrationtmax time to maximum concentrationt1/2,el terminal half-lifeCl/F clearance following pulmonary administrationFDPI/Fneb relative bioavailability, calculated on actual doseC.I. confidence intervalC.V. coefficient of variation* statistically significant


tightness for which a rating 2 was given. Two patients did not experience any adverse effect at

all. The adverse effects appeared within 10 minutes after inhalation with the DPI or nebulisation

with the jet nebuliser. Patient 2 and 4 experienced chest tightness after dry powder inhalation,

but their fall in FEV1 was within 10% from the baseline. Five patients went through a period of

chest tightness after jet nebulisation, of which patient 2 and 4 indicated that this effect was more

severe than after dry powder inhalation. Four patients felt a cough reflex, three of them coughed

during breath hold, which may possibly have had a negative influence on the lung dose. Two

patients thought the cough reflex was generated by the air flow through the nebuliser. Four

patients mentioned the taste of the dry powder, of which two patients disliked this taste.

Discussion

This study was conducted with the objective to improve lung deposition of inhaled colis-

timethate sodium with the Twincer® dry powder inhaler in CF patients. We have investigated

the influence of a slightly larger median particle diameter in combination with a lower inspira-

tory flow rate compared to a previous study, expecting that this would yield a higher (periph-

eral) deposition. We indeed obtained a slightly higher relative bioavailability (Fdpi/Fneb) of 3.0

(compared to 2.7 in the previous study) based on the nominal dose and an equivalent relative

bioavailability of 1.4 based on the actual dose (Westerman et al., 2007). Obviously, the small

number of participants makes it impossible to draw distinct conclusions from our results, but

the data are useful in further dry powder development of drugs in CF treatment.

In this study four modifications to the previous study have been introduced: a dry powder

formulation with a slightly larger median particle diameter, an intermediate inhalation peak

flow rate, a longer breath-hold and an injection moulded Twincer® inhaler.

The median particle diameter in this study was 2.1 μm compared to 1.6 μm in the previous

study. Although these proportions are indicative for the two powder formulations that have been

used, it is actually the combination of the median particle diameter and particle size distribution

Fig. 1. Serum concentration-time curves after inhalation of colistimethate sodium.

Chap

ter 8

148

that should be considered when estimating and calculating powder behaviour in the lung. How-

ever, these calculations are beyond the scope of this study. The larger median particle diameter

was introduced to enlarge lung deposition primarily by improving sedimentation. The inhalation

peak flow rate in this study was 43.9 l/min (was: 67.9 l/min). An intermediate flow rate is expected

to reduce impaction of the aerosolized dry particles in the oropharynx and to result in an effective

lung deposition at the same time. The longer breath hold was introduced to give the aerosol

more time for deposition by sedimentation in the bronchiolal and alveolar regions of the lung.

Finally, the injection moulded Twincer® inhaler is a ready-to-use device which is able to effectively

disperse 25 mg in one inhalation manoeuvre. The Twincer® used in the previous study was a pro-

totype which was filled and assembled by hand on each study day. This model was constructed

with two dose compartments which contained 12.5 mg of colistimethate sodium each.

The results, expressed as relative bioavailabilities, were similar to the results in the previous

study and have been obtained by a pharmacokinetic analysis. Another, theoretical method for

estimating the obtained lung dose from inhalers is to calculate the impaction parameter, that

incorporates the particle aerodynamic diameter or geometric diameter and the average airflow

rate and, depending on the equation used, the particle density (DeHaan and Finlay 2004). The

impaction parameter is often used to predict drug loss in the oral cavity and oropharynx.

Applying this equation on the results obtained in the previous and the current study results

in comparable impaction parameters of 4346 and 4840 μm2cm3s-1, corresponding with a loss

of approximately 9.1% and 10.1% in the oral cavity and oropharynx respectively. Therefore,

the fraction of the 25 mg colistimethate sodium that was able to enter the lungs has been

equal in both dry powder inhalation studies, and it is likely that inertial deposition in the upper

respiratory tract will have been comparable as well. Theoretically, the advantage of the inhaler

in the current study is therefore to be found in a higher sedimentation velocity in central and

peripheral airways. It is to be expected that a higher lung deposition has been obtained in

this study, as terminal velocity increases with a larger particle diameter and more time has

been created for sedimentation in lung generations 11-23 because of an extended breath hold

(Usmani et al., 2005). Unfortunately, no convincing evidence for this theoretical advantage has

been found with the pharmacokinetic method used for determining lung deposition (relative

bioavailability). Gamma scintigraphic methods can support further studies, as regional differ-

ences in deposition cannot be measured with PK method (Chrystyn 2001).

Among other (unknown) factors, interpatient variability will have influenced the results in

this pilot study. This variability is well known in lung deposition studies in CF patients and is

especially relevant in studies with a small number of subjects. This variability is made up by

variation in inhalation technique and variation in lung deposition due to different stages of

disease progression. In this study the coefficient of variation of Cmax and AUC after dry powder

inhalation (52% and 48% respectively) is larger compared to jet nebulisation (43% and 43%

respectively). However, this variation is predominantly caused by patient 5. Excluding this patient

results in a C.V. of 24% and 23% for Cmax and AUC after dry powder inhalation respectively,


and this suggests that nebulisation is less reproducible than dry powder inhalation. Without

the data of patient 5, the relative bioavailability changes slightly in favour of wet nebulisation,

without changing the observed trends and significance in pharmacokinetic parameters.

The interpatient variability is a weak point in comparative study designs in CF as is jet nebu-

lisation as reference treatment which is known for its variable results. It is a paradox, as clinically

administration of colistimethate sodium by jet nebulisation is still the gold standard. The value

of comparing relative bioavailabilities within this study or between the previous study and this

study is therefore questionable, as the results are guided by patient variability.

The new inhaler was appreciated by the patients and the dry powder drug dose was well toler-

ated. Airway reactivity (FEV1) was measured in one patient after wet nebulisation but not after

dry powder inhalation. Post-nebulisation decrease in FVC values was observed in two patients,

but not after dry powder inhalation. Similar results were obtained in the previous study, and

further research is needed to elucidate whether these effects are a result of the physical form of

the inhaled drug (aerosol in droplets versus powder particles), local high drug concentrations

in the proximal lung and/or the absolute amount of drug entering the target area in the lung,

in relation to the disease state of the particular patient.

The residual mass in the DPI was higher than previously observed, which can be attributed

to the introduction of a large blister to the design of the injection moulded Twincer®. The blister

used for this colistimethate sodium study was designed to contain 12.5 mg of spray-freeze dried

powder which has a large volume due to its high porosity. The same blister may contain 60 mg

of colistimethate sodium and as a result of only partially filling the blister (25 mg), considerable

losses by adhesion onto the blister wall and blister seal occurred during transport and handling

of the Twincer® devices. Without the blister seal in vitro data indicate only marginal drug loss

within 1% of the nominal dose for payloads of 5-50 mg (data not shown).

In conclusion, colistimethate sodium dry powder inhalation with the Twincer® inhaler is

efficient and well tolerated by CF patients. The influence of altered inhalation conditions on

lung deposition could not be established with the pharmacokinetic method used. Gamma

scintigraphy may be of additional value in future studies. The results support further research

on dry powder inhalation of drugs in CF using the Twincer® inhaler.

Acknowledgements

The authors wish to thank Mr P. Hagedoorn (Department of Pharmaceutical Technology and

Biopharmacy, University of Groningen) for his technical assistance in this study, Mr G. van der

Meyden (Adult Cystic Fibrosis Center Haga Teaching Hospital) for performing the pulmonary

function tests and Mr. H. Trumpie (Clinical, Pharmaceutical and Toxicological Laboratory, Apo-

theek Haagse Ziekenhuizen) for the assay of the serum concentrations.

Chap

ter 8

150

References

Arias Llorente RP, Bousono Garcia C, Diaz Martin JJ. Treatment compliance in children and adults with Cystic Fibrosis. J Cyst Fibros 2008; 7: 359-367.

Brochet MS, McDuff AC, Bussieres JF, Caron E, Fortin G, Lebel D et al. Comparative efficacy of two doses of nebulized colistimethate in the eradication of Pseudomonas aeruginosa in children with cystic fibrosis. Can Respir J 2007; 14:473-479.

Chrystyn H. Methods to identify drug deposition in the lungs following inhalation. Br J Clin Pharmacol 2001; 51:289-299.

Cystic Fibrosis Foundation. Patient Registry 2006 Annual Report, Bethesda, Maryland, USA. 2006;Darquenne C, Paiva M, Prisk GK. Effect of gravity on aerosol dispersion and deposition in the human lung

after periods of breath holding. J Appl Physiol 2000; 89:1787-1792.Davis PB. Cystic fibrosis since 1938. Am J Respir Crit Care Med 2006; 173:475-482.DeHaan WH, Finlay WH. Predicting extrathoracic deposition from dry powder inhalers. Aerosol Science

2004; 35:309-331.Flume PA, O’Sullivan BP, Robinson KA, Goss CH, Mogayzel PJ, Jr., Willey-Courand DB et al. Cystic fibrosis

pulmonary guidelines: chronic medications for maintenance of lung health. Am J Respir Crit Care Med 2007; 176:957-969.

Frederiksen B, Lanng S, Koch C, Hoiby N. Improved survival in the Danish center-treated cystic fibrosis patients: results of aggressive treatment. Pediatr Pulmonol 1996; 21:153-158.

Fuchs HJ, Borowitz DS, Christiansen DH, Morris EM, Nash ML, Ramsey BW et al. Effect of aerosolized recom-binant human DNase on exacerbations of respiratory symptoms and on pulmonary function in patients with cystic fibrosis. The Pulmozyme Study Group. N Engl J Med 1994; 331:637-642.


Kerem E, Reisman J, Corey M, Canny GJ, Levison H. Prediction of mortality in patients with cystic fibrosis. N Engl J Med 1992; 326:1187-1191.


Lowry OH, Rosebrough NJ, Farr AL, Randall RJ. Protein measurement with the folin phenol reagent. J Biol Chem 1951; 193:265-275.

Murphy TD, Anbar RD, Lester LA, Nasr SZ, Nickerson B, VanDevanter DR et al. Treatment with tobramycin solution for inhalation reduces hospitalizations in young CF subjects with mild lung disease. Pediatr Pulmonol 2004; 38:314-320.

Pilcer G, Goole J, Van Gansbeke B, Blocklet D, Knoop C, Vanderbist F et al. Pharmacoscintigraphic and phar-macokinetic evaluation of tobramycin DPI formulations in cystic fibrosis patients. Eur J Pharm Biopharm 2008; 68:413-421.


Ryan G, Mukhopadhyay S, Singh M. Nebulised anti-pseudomonal antibiotics for cystic fibrosis. Cochrane Database Syst Rev 2003; CD001021.

Usmani OS, Biddiscombe MF, Barnes PJ. Regional lung deposition and bronchodilator response as a func-tion of beta2-agonist particle size. Am J Respir Crit Care Med 2005; 172:1497-1504.

Westerman EM, Le Brun PPH, Touw DJ, Frijlink HW, Heijerman HGM. Effect of nebulized colistin sulphate and colistin sulphomethate on lung function in patients with cystic fibrosis: a pilot study. J Cyst Fibros 2004; 3:23-28.

Westerman EM, De Boer AH, Le Brun PP, Touw DJ, Frijlink HW, Heijerman HG. Dry powder inhalation of colistin sulphomethate in healthy volunteers: a pilot study. Int J Pharm 2007a; 335: 41-45.

Westerman EM, De Boer AH, Le Brun PP, Touw DJ, Roldaan AC, Frijlink HW et al. Dry powder inhalation of colistin in cystic fibrosis patients: a single dose pilot study. J Cyst Fibros 2007b; 6:284-292.

9 Concluding remarks and future perspectives

Concluding remarks and future perspectives 153

Introduction

Inhalation of antibiotics has become a cornerstone in the treatment of pulmonary Pseudomonas

aeruginosa infections in cystic fibrosis (CF) patients, either in preventing or stabilising chronic

infections, sometimes combined with intravenous therapy. As this approach has proven to be

successful in contributing to an improved life expectancy, extensive attention is now given to

the technical aspects of inhalation. The focus is on innovating drug delivery devices, sometimes

combined with specific drug formulations, which allow for the administration of large doses in

a short time frame and in a reproducible way. The majority of CF patients are confronted with

extensive and time-consuming drug administration, several times a day. Nebulised drugs are

a major contributor to this and have a major impact on the daily life of a cystic fibrosis patient.

Therefore it is worthwhile to investigate every possible option to improve this life-long routine.

The ultimate aim is to prevent, limit and treat lung damage in order to maintain and improve

clinical well-being and quality of life.

In general, drugs are administered by inhalation to exert a clinical response, either by direct

action on lung tissue or after systemic absorption. Pulmonary administration of drugs has been

investigated in treating several diseases in recent years. Next to COPD and asthma, for which

inhalation therapy has been the golden standard for years, insulin, antibiotic, antimycotic,

immunosuppressant and chemotherapeutic agents have been subject of clinical trials. The

objectives for inhalation of the drug are either to make administration easier (insulin), to reduce

systemic adverse effects and to obtain higher local concentrations which result in increased

efficacy (antibiotics, antimycotics, immunosuppressants, chemotherapeutic agents). Inhaling

an antibiotic agent in CF treatment aims at all three objectives, and offers the patient the

opportunity of preventive anti-pseudomonal therapy at home, with intravenous treatment as

the only alternative route for administration of aminoglycosides or colistimethate sodium.

High local concentrations are not only favourable with respect to efficacy, but will also reduce

the risk of the development of bacterial resistance due to suboptimal local concentrations of

antibiotics. Furthermore, a possible role is reserved for drug-device combinations which are

able to produce high local concentrations in treating CF patients who suffer from antibiotic-

resistant strains of bacteria. In extreme circumstances, it is imaginable that high local drug

concentrations will be the last treatment option for CF patients with extremely resistant strains

of bacteria who do not respond sufficiently to various intravenously administered antibiotics.

Furthermore, treatment of acute pulmonary exacerbations may be possible if local antibiotic

concentration can reach sufficiently high levels; this hypothesis deserves further study.

The current challenge in inhalation therapy in CF is to reach the most peripheral parts of the lung

to combat Pseudomonas aeruginosa and other pathogens with antibiotics like colistimethate

sodium and tobramycin, as CF disease starts from there (Tiddens 2002). This area of the lung is

characterised by a large airway surface area and a thin barrier to the systemic circulation and is

Chap

ter 9

154

the most difficult part of the lung to be reached with aerosols. While targeting at the periphery,

other, more central parts of the lung which are colonised with the pathogens, will be in contact

with the antibiotic also. Unfortunately, the relation between the deposition of a high dose in the

peripheral parts of the lung and the alleged improved clinical response has not been proven

yet, nor is information available on how high this dose should be. The lack of knowledge on the

optimal lung dose of antibiotics, and the inter-device and inter-individual variability between

patients makes it impossible to predict successful inhalation treatment beforehand. Several

factors add to the variability found in inhalation treatment:

• Nebulisersandinhalershavevariableefficiencieswhichmayresult insignificantvaria-

tions in the lung dose.

• Physicalpropertiesofthedrugmayaffectaerosolizationresults;withtheobtainedpar-

ticle size of the aerosol and the output rate as most prominent variables.

• Eachpatient introduces several variabilities to treatmentoutcome: their disease state

and inhalation technique are only two examples.

All these factors contribute to a highly variable pulmonary drug deposition.

As at present inhalation of antibiotics in CF has become mainstay, it is a challenge to investi-

gate whether individualised inhalation regimens can be established. This implies that the basic

mechanisms of the inhalation process in CF patients and the influence of disease progression,

airway anatomy and the ‘compatibility’ of a patient with the various inhalers should be further

explored. If this approach turns out to be successful, perhaps in future inhalation therapy will

be initiated based on the optimal match between the patient’s and aerosol generation device

profiles.

Considerations on study design and inhaler development

Pharmacokinetic and scintigraphic studies

The clinical research described in this thesis applied a pharmacokinetic methodology to assess

lung deposition. Two or three dimensional imaging technologies are also used to study pulmo-

nary deposition in inhalation treatment. Pharmacokinetic methods are useful for determining

the total lung dose after inhalation, as has been proven for some drugs or drug-device combi-

nations (Newnham et al., 1993). Urinary concentration measurements after inhaling tobramycin

can be used as well (Asmus et al., 2002). These methods work well for comparing relative lung

deposition, but do not provide information on regional deposition in the lung, whereas two

dimensional (gamma scintigraphy) or three dimensional (single photon emission computed

tomography (SPECT) or positron emission tomography (PET)) (Newman et al., 2003, Eberl et al.,

2006) methods provide both. Furthermore, these techniques may be useful in studies regarding

dose-response relationships. However, these scintigraphic techniques require a reformulation

of the original dry powder drug because of the radiolabeling process which can influence the


particle size distribution and thereby the clinical results. This study approach is obviously more

complex and more expensive to perform.

Therefore it was decided that the first clinical study in the development of the Twincer®

colistin dry powder inhaler, described in this thesis, would start with a pharmacokinetic analy-

sis, with feasibility as a major topic. Now that satisfactory phase I results have been obtained,

two or three dimensional imaging studies can possibly provide new, valuable information on

regional deposition of the drug. Moreover, these studies could deepen our understanding on

the relationship between physical characteristics of the generated aerosol and lung deposition.

Obtaining data on regional deposition of the dry powder particles is desirable, especially if

these data can be related to clinical efficacy parameters as local bioavailability is pertinent to

reflect efficacy of drugs that act directly in the lung (Pilcer et al., 2008). Furthermore it may

be worthwhile to study the relation between a clinical efficacious lung dose in the peripheral

lung and pharmacokinetic parameters (AUC, Cmax, tmax), as this might be used as a less invasive

method for therapeutic drug monitoring in future.

It is, however, a challenge to radiolabel the colistimethate sodium dry powder particles

without changing their aerodynamic behaviour and to use this material in a clinical inhalation

study within the period of activity of the radionuclide. A method for labelling a tobramycin dry

powder for inhalation has been described recently (Pilcer et al., 2008) but to our best knowledge

no such experiment has been performed with colistimethate sodium yet. In the Pilcer study

lung deposition, measured by gamma scintigraphy, was lower than the fine particle fraction

measured in vitro, which was attributed to the nature of CF and its severity (Pilcer et al., 2008).

This is in agreement with other data (Newman and Chan, 2008). A relationship between the

data on lung deposition obtained by gamma scintigraphy and the pharmacokinetic method

was described.

Influences on lung pharmacokinetics

In general, serum concentration levels obtained after drug inhalation are considered to reflect

pulmonary drug deposition in the lung, provided that no gastro-intestinal absorption of the

drug in question occurs, as is the case with colistimethate sodium and aminoglycosides. This is

an indirect or surrogate method to determine pulmonary drug deposition. Drug concentrations

in sputum are used as a measure of pulmonary drug deposition too, but the highly variable

results, likely to be caused by patient- and disease specific factors, make sputum PK analysis

less valuable compared to serum PK analysis (Geller et al., 2007; Laube et al., 1989). Moreover,

sputum concentrations are merely a reflection of drug deposition in the upper airways and

provide little information on the deposition in the deeper airways or alveoli.

Research on lung deposition using pharmacokinetic methods is complicated by the fact

that little is known of fundamental pharmacokinetics and pharmacodynamics of inhaled drugs

in CF patients.

Chap

ter 9

156

Only a few studies have incorporated pharmacokinetic methods in their data analysis,

whereas most clinical (efficacy) studies have either omitted to perform a proper pharmacoki-

netic analysis on their data or have not measured serum levels at all.

Serum levels measured shortly after inhalation possibly reflect alveolar deposition as the

alveolar surface area is by far the largest of all lung parts (Patton, 1996), but at the same time

the amount of drug in the alveoli might be redistributed to the large airways by the pulmonary

circulation (Chrystyn, 2001). The total lung dose is not influenced by this, but the concentration

at the peripheral target site in CF probably is and may possibly influence therapeutic efficacy.

Similarly, an identical total lung dose may be deposited more centrally for instance because

severe disease progression prevents peripheral deposition (Mukhopadhyay et al., 1994). It is

not clear if or how such differences in local lung distribution are reflected by pharmacokinetic

data.

Furthermore, it has been argumented for tobramycin that the physical state (solution versus

dry powder) of the drug influences the extent of systemic absorption: in the study by Pilcer

(Pilcer et al., 2008) a dry powder lung dose resulted in a lower systemic absorption, including a

delay in tmax, and the authors suggested that this would be beneficial for the local antimicrobial

treatment in the CF lung. These findings cannot be compared to the results described in Chap-

ters 7 and 8, as no data are available on the obtained lung doses in both studies, albeit that the

tmax-values after dry powder administration were shorter compared to wet nebulisation, indi-

cating a faster absorption process. Further studies are required in order to be able to estimate

the influence of the physical state of a drug on systemic absorption as well as on therapeutic

efficacy. Colistimethate sodium has surface tension lowering properties, which may influence

absorption characteristics as well.

Lung deposition data in pharmacokinetic inhalation studies are expressed as the maximum

serum concentration (Cmax) level and the area under the curve (AUC). Of these two, the AUC

is a better descriptor in trials in which inhalation devices with different inhalation times and/

or different absorption rates are being compared (chapter 3, Chrystyn, 2001). If the supposed

relationship between (peripheral) lung deposition, clinical efficacy and AUC can be established

in the future, this parameter might be used to optimise treatment on an individual basis and

facilitate in vivo bridging studies. The absorption rate of a drug, reflected by the tmax, might

be a descriptor in lung deposition studies as well. Different mechanisms by which (systemic)

absorption takes place, have been described (Patton, 1996). The objective in antibiotic aero-

sol treatment is obviously a high local peripheral concentration in the lung, but perhaps the

pharmacokinetic parameters, based on serum levels, remain useful for describing lung deposi-

tion in future. For example it would be interesting to learn whether a shorter tmax, found in

the Chapters 3, 7 and 8, indeed corresponds with a larger fine particle fraction, as has been

suggested in Chapter 3. In these three clinical pilot-studies, drug aerosol inhalation using the

PARI-LC Plus®-CR60® nebuliser-compressor combination (tobramycin) and the Twincer® dry


powder inhaler (colistimethate sodium) resulted in shorter tmax-values while in vitro data with

these devices showed a larger fine particle fraction, compared to the reference devices.

Inhaled antibiotics and safety

Although the AUC might be the preferred parameter to reflect lung deposition, it should be

noted that, especially for the aminoglycosides (tobramycin, gentamicin, amikacin), knowledge

on Cmax and Ctrough is valuable too. For this class of antibiotics, efficacy with the lowest risk of

toxicity (nephro- and ototoxicity) in intravenous therapy is related to high peak and low trough

serum levels, preferably in an extended dose interval regime (Touw et al., 2007). In contrast with

the extensive knowledge available on the risk of toxicity with intravenous use of aminoglyco-

sides, only sparse, anecdotal information is available on the toxicity risk of inhaled tobramycin

on a daily basis for longer periods of time, as is the case in CF treatment. In long term term trials,

no nephro- or ototoxicity were observed, but these studies were performed with a nominal dose

of 80 mg tobramycin (Steinkamp et al., 1989, MacLusky et al., 1989). A shorter, 12-week study,

with 600 mg nebulised tobramycin three times daily, revealed no toxicity. Generally, a compro-

mised renal function can result in increased toxicity in aminoglycoside therapy, for which high

trough serum levels (> 1 mg/L) are a marker (Hoffmann et al., 2002, Edson et al., 2004, Kahler

et al., 2003). Extrapolation of toxic serum levels in intravenous therapy to estimate toxicity for

aerosolized aminoglycosides cannot be done without due consideration however, as first of

all no data are available on possible toxic effects of high concentrations of aminoglycosides

within the lung and secondly no relationship has been established between intrapulmonary

concentrations and serum drug levels as an indirect parameter of deposition. Therefore it is

difficult to estimate aminoglycoside toxicity in inhalation treatment. Until more data are avail-

able, we are confined to extrapolating knowledge from i.v. treatment to aerosolisation in CF for

estimation of toxicity. Therefore, pragmatic toxic limits (Ctrough > 1 mg/L, Cmax > 4 mg/L) have

been introduced in chapter 3. Fortunately, although only sparse information is available on risk

for toxicity on the inner ear during inhalation treatment, short term nebulisation up to 4 weeks

and 24 weeks appears to have no negative effects on auditory function (Mukhopadhyay et al.,

1993, Ramsey et al., 1999, Lenoir et al., 2007, Chuchalin et al., 2007). It is nevertheless advisable

to collect information on possible toxicity by performing audiometric (especially high tone)

measurements during long term inhalation treatment (> 6 months continuously or monthly

on/off) with aminoglycosides, as these data are lacking for the CF population.

Colistimethate sodium toxicity is generally limited to reports of airway narrowing and chest

tightness, whereas nephrotoxicity and neurotoxicity have been observed after intravenous

treatment (Falagas and Kasiakou, 2005). Colistin is the most potent (Bergen et al., 2006) but

also more toxic compound (Chapter 4) to which a fatality has been attributed recently (McCoy,

2007). Apparently hydrolysis of colistimethate sodium into colistin prior to intravenous or

inhaled administration exerts toxic effects, while this process appears to be safe once it occurs

within the human body.

Chap

ter 9

158

Next to intrinsic drug toxicity, clinicians are responsible for monitoring patients that are

susceptible for toxicity, e.g. reduced renal function, nephrotoxic co-administered drugs and

auditory effects, among which high tone loss (Scheenstra et al., 2007). Obviously, this includes

extra awareness during concomitant treatment of inhaled and intravenous aminoglycosides.

How to obtain and study clinical efficacy?

The lack of knowledge on the relationship between lung dose and drug serum levels, as

discussed earlier, is a major drawback for the ability to estimate a clinical response after drug

inhalation. It is expected that a high concentration of antibiotic in the peripheral lung will have

the best antimicrobial effect and will therefore result in the best clinical response. Research on

the relation between peripheral deposition and clinical effects, as has been done for inhaled

beclometasone (Marshall et al., 2000), is needed. Once such a relationship has been established,

further steps in exploring inhalation therapy in CF can be made, especially with respect to

therapeutic drug monitoring.

Fundamental research on the interaction of deposited aerosolized particles, both liquid

and dry powder particles, on lung tissue infected with bacterial colonies, will support our

understanding of the mechanisms by which a clinical effect occurs. Furthermore, radiolabeled

deposition studies, e.g. with polydisperse (droplets covering a range of sizes) and monodisperse

(all droplets approximately the same size) particles will broaden our knowledge on particle

size behaviour in relation to disease progression in the lung (Usmani et al., 2005). In studies

with heterodisperse radioisotope-labeled particles, it has been shown that in patients with CF,

compared to normal subjects, there is a marked heterogeneity in the pattern of deposition of

particles in the lung (Alderson et al., 1974), due to e.g. obstructed and/or distorted airways and

decreased regional ventilation.

In the meantime, comparing pharmacokinetic drug data after using two or more different

inhalation systems is one of the simplest way of estimating (relative) lung deposition, albeit in

an indirect manner. The clinical efficacy and safety of a nominal dose of 300 mg tobramycin

(Ramsey et al., 1999, Bowman, 2002, Moss, 2002) and 160 mg colistimethate sodium (Freder-

iksen et al., 1997), administered with a jet nebuliser has been established over the years. This

knowledge is the basis of current anti-pseudomonal inhalation treatment and is the gold

standard to which new (including ‘me too’) preparations are being compared (Poli et al., 2007).

Given the fact that only TOBI® comes with a recommendation on the nebuliser to be used,

these doses are a remarkable ‘standard’, as it is well known that the ultimate lung dose is always

highly variable due to various causes, as discussed previously.

A consequence of the comparative study design method is that an equivalent dose (and

not an optimal dose) is searched for, which is most likely to result in a comparable lung deposi-

tion. This dose is expected to result in a comparable therapeutic efficacy and can be obtained

for example by performing a dose escalation study. It is believed that, by direct comparison,

patient bound parameters that may influence the study results will be present in all study


situations and may therefore be considered not to influence relative comparisons. The results

can therefore directly be attributed to the device or drug presentations studied. This approach

needs to be put in perspective however, as intra-patient variability may affect results, as well as

inter-patient variability (Chrystyn et al., 1998). Mucus plugging and airway inflammation result

in a reduction of the tidal volume, trapping of the drug in obstructed areas, a minimal airflow

and thus a reduced peripheral deposition. It is likely that study outcomes will be influenced by

heterogeneously composed patient groups, as peripheral deposition is inversely proportional

to disease severity and correlated to the FEV1 (Mukhopadhyay et al., 1994). This is especially

relevant in studies with a small number of subjects. It is therefore important to employ well-

designed cross-over designs and to search for new approaches (Simon and Chinchilli, 2007) in

order to try to reduce the influence of the aforementioned variability.

Nevertheless, pharmacokinetic inhalation studies with a comparative design are feasible

in CF patients, once one is aware of this variability. Ideally, the pharmacokinetic method is

accompanied by a scintigraphic analysis and performed in a large number of study subjects, to

compensate for the large variability described earlier.

Inhaler development

It is a challenge to aerosolize a dry powder dose in the over 10 milligram mass range, as this

requires an inhaler which is capable to release and disperse large amounts of powder particles

at a flow rate that is optimal for peripheral deposition in the lung. The urge for high dose

administration of 20 mg sodium cromoglycate has led to the development of the first dry

powder capsule inhaler in 1967 (Smith and Parry-Billings, 2003). Therefore, it is not surprising

that several dry powder inhaler studies with antibiotics in CF have used capsule inhalers for

administration of the drug as they can contain large amounts of dry powder. Although these

inhalers are relatively easy to use for testing a new antibiotic dry powder principle in CF treat-

ment, capsule inhalers have some major disadvantages. Poor efficiency of lung delivery (lung

dose approximately 10%), patient related issues concerning reloading of a capsule prior to each

dose (Smith and Parry-Billins, 2003) and drug/device related issues (brittleness of the capsule

after long storage time, inhaler or capsule retention; Vidgren, 1988) have led to development

of more efficient (multidose) dry powder inhalers for COPD and asthma treatment. Although

understandable with respect to speed up commercial availability of a the dry powder device

for antibiotics, it is unfortunate that much effort is put into the development of capsule inhalers

for treating CF patients nowadays, while technically superior inhalers have been designed and

are available.

The Twincer® inhaler is, because of its technical features, able to release and disperse high

powder doses in a range of 25-50 mg in one single inhalation. The 25 mg dose has been tested

in vivo (this thesis), whereas in vitro results clearly indicate the feasibility of higher doses. An

additional advantage of the Twincer® design is the single use (disposable) concept, which

reduces the risk of (re-)contamination of the user with (resistant) bacteria via the inhaler and of

Chap

ter 9

160

a possible decrease in stability of the drug because of humid exhaled air into the device. Fur-

thermore the small size, the redundancy of an external power source and low production costs

of the four-component device make this inhaler a promising design for future applications.

Future perspectives and recommendations

Fundamental science

Future research should continue to focus on fundamental aspects related to drug inhalation in

CF. The influence of various factors like disease progression, aerosol behaviour in the lung and

patient related factors on the deposited dose should be mapped. The use of monodisperse par-

ticles for inhalation in a research environment with lung deposition imaged using scintigraphy,

is expected to increase our understanding of critical parameters regarding inhaler and inhala-

tion parameters in CF (Usmani et al., 2005). More knowledge is needed on the characteristics

of the target region in the lung in order to be able to define the optimal clinical response in

antibiotic inhalation treatment. This will correspond with an optimal lung dose, which equals

an effective but safe drug concentration level both in the lung and in the systemic circulation

with a minimal risk on adverse drug reactions.

Knowing the target area, the drug dose including particle size specifications, and the type

of inhaler suitable for the individual patient will help in aiming at an optimal drug concen-

tration at the target site and an optimal therapeutic effect. These data may help to establish

a therapeutic index, which is the ratio between clinical efficacy and adverse drug reactions

(including toxicity). The influence of the aerosol particle size (fine particle fraction (FPF), in vitro)

and the particle size distribution within the FPF, on the therapeutic index, has been reviewed

recently (Weda et al., 2008). The therapeutic index can be established for each antibiotic drug in

combination with a specific inhaler, using information on the dose-efficacy response curve, the

absorption pharmacokinetic profile of the drug from the lung into the systemic circulation and

the dose-adverse drug reaction curve. Possibly this will support individual therapeutic drug

monitoring, provided a sound dose-response relationship can be established in the future. If so,

perhaps more information can be obtained on why some patients can be treated with inhaled

antibiotics successfully and others cannot. If the use of a specific drug-inhaler combination

for a patient results in a lung dose that is too low for effective treatment, perhaps a different

inhaler will give improved results. If, however, the patient is not adherent to therapy, intensified

guidance from CF healthcare professionals may be an alternative intervention.

Adherence to therapy

Adherence of CF patients to therapy with aerosolized drugs is poor, predominantly due to the

time consuming drug administration and cleaning procedures (Arias Llorente et al., 2008). At the

same time, adherence to treatment influences the outcome of aerosolized treatment (Kettler


et al., 2002). It is therefore of major importance that devices for administration of aerosolized

drugs will be further developed, aiming at a short(er) administration time, a reduction in risk of

device contamination, an user friendly size and preferably no dependence on an external bat-

tery or electricity, aiming at improving adherence to therapy and quality of life of CF patients.

The Twincer® dry powder inhaler has all these features: inhalation can be performed within

one minute; the inhaler is designed for single use, avoiding the risk of device contamination; it

is the size of a credit card and uses the patient’s inhalation flow to release the drug dose (see

also Chapter 2). These advantages over regular jet- and ultrasonic nebulisers, combined with

the Twincer®’s potential for effective regional drug deposition in the lung makes this inhaler a

promising new development in CF treatment.

A specific topic that deserves attention is the potential difference between inhalation of dry

and wet particles on the subjective well-being of patients. Several patients who took part in

the studies described in this thesis, indicated to have missed their regular nebulisation sessions

which they had to stop at least 24 hours prior to study days. While dry powder inhalation seems

to have several advantages over wet nebulisation, including an improved adherence to therapy,

it might well be that for certain patients wet nebulisation will remain daily routine in their drug

administration to improve their subjective well-being. If so, it is imaginable that dose-critical

drugs, like antibiotics, will be administered by dry powder inhalation, while ‘wetting’ of the

lower airways is done by nebulisation of isotonic of hypertonic saline.

Airway narrowing

A complication of drug inhalation that is present in a substantial number of CF patients, is tran-

sient airway narrowing. This description is chosen deliberately, as the exact cause of a transient

fall in lung function parameters (FEV1, FVC) shortly after inhalation of drug in CF patients is

not clear. Although constriction of the airways (bronchoconstriction) due to the tightening of

smooth muscle is a likely cause, swelling of the airway lining or increased mucus amounts in the

airways may also cause narrowing of the airways in CF patients.

Several mechanisms have been proposed to cause a post-inhalation decline in FEV1 dur-

ing or after drug inhalation. Some CF patients have a coexisting asthma or an inherent airway

hyperreactivity, which may explain these post-inhalation observations. Drug related causes

have been suggested, such as mast cell degeneration (hypertonic solutions (Alothman et al.,

2005). Airway narrowing has furthermore been attributed to excipients, such as antioxidants

and preservatives, and pH and osmolality. However, it has been found that solutions for inhala-

tion both with or without excipients may cause airway responsiveness in susceptible patients

(Alothman et al., 2002). Similarly, hypo-, iso- and hypertonic solutions provoked airway nar-

rowing symptoms in CF patients (Dodd et al., 1997), suggesting that tonicity alone, within a

reasonable range from isotonicity, is probably not responsible for airway narrowing. The small

differences between pH-values of aerosolized antibiotics and normal saline makes the acidity a

less likely cause of airway narrowing (Chua et al., 1990).

Chap

ter 9

162

Generally, approximately 10-20% of CF patients react on a nebulised antibiotic dose with a

fall in FEV1,predicted >10% (Hodson et al, 2002a, 200b; Geller et al., 2007), although higher numbers

for inhaled colistimethate sodium have been described (Alothman et al., 2005, Cunningham et

al. 2001). However, this clinical observation does not necessarily mean that the patient experi-

ences chest tightness. It is worthwhile investigating the underlying mechanisms of this airway

reactivity, as this might elucidate why subjective chest tightness is not always correlated with

a fall in FEV1,predicted > 10% (chapter 8), which is considered to be clinically relevant (MacLusky

et al., 1986).

Furthermore, perhaps a consensus can be reached on which points in time lung function

tests should be performed after drug inhalation in order to estimate airway reactivity. Most

susceptible patients react during or shortly after the end of the nebulisation session and FEV1

recovers within 30 minutes for the majority of patients. This is why in this thesis timepoints of

5 min and 30 min after inhalation were chosen to perform pulmonary function tests. Further-

more, the presence of airway narrowing after dry powder inhalation and wet nebulisation was

shown to be different for each patient: no fall in FEV1,predicted > 10% was observed after dry

powder inhalation, but several patients reacted on wet nebulisation (this thesis). Whether dry

powder inhalation has a lower potential for causing airway reactions or these observations are

merely caused by a relatively lower inhaled dose, is not known at present and deserves further

investigation.

Airway narrowing caused by a high concentration of drug particles in the bronchiolar area

resulting in contraction of the bronchiolar muscles and a fall in FEV1 in susceptible CF patients

has been hypothesized (Chapter 3). This centrally deposited drug aerosol can be a result of an

unfavorable particle size-flow rate combination and/or a reduced peripheral airway patency

caused by disease progression. In contrast, a drug aerosol with optimal properties for periph-

eral airway deposition that is inhaled with a suitable flow rate and inhalation volume, may show

a lower potential for airway narrowing due to a lower drug concentration in the central airways.

This hypothesis may be studied in future.

Dosing frequency and treatment of exacerbations

While once daily intravenous dosing of aminoglycosides has been studied and applied in daily

routine (Smyth et al., 2005), no data are available on once daily versus multiple-times daily dos-

ing of aerosolized aminoglycosides in CF. Theoretically, once daily dosing might reduce the risk

of toxicity and be advantageous with respect to adherence to inhalation treatment which may

in turn increase efficacy. On the other hand, new treatment strategies with a higher daily dosing

frequency than the current twice-a-day dosing schemes may be worthwhile exploring as well.

Because of the availability of the new inhaler devices (mesh technology, dry powder inhalers

e.g. the Twincer®) treatment times can drop considerably. This offers a perspective for a more

frequent aerosol administration (up to 4 or 5 times daily), especially for dry powder inhalers,

once clinical proof has been obtained that these drug-device combinations are effective in


anti-pseudomonal treatment. It is worthwhile to investigate whether such dosing schemes will

be superior to the current schemes.

Extrapolating intravenous therapy knowledge on aminoglycosides to inhalation treatment

would mean that higher peak serum levels correspond with a higher lung deposition. High

tobramycin doses of 600 mg and 1000 mg, administered with jet nebulisers, have been inves-

tigated (Le Brun 1999, Le Brun, 2001), corresponding with an estimated lung dose of 44 and 67

mg respectively, which, compared to an estimated lung dose of 27 mg after 300 mg tobramycin

(Newman et al., 2001) is approximately 1.5-2.5 times higher. A dose of 1000 mg was considered

at that time to be the highest possible dose that can be administered with a jet nebuliser within

an acceptable amount of time (30 min.). However, more efficient nebulisers, including vibrating

mesh nebulisers, are available nowadays which may deliver higher amounts of drug within

a shorter time frame. Dry powder inhalers, like the Twincer® inhaler, may enable high dose

pulmonary drug administration to an even further extent. This justifies renewed attention for

the application of high dose inhaled antibiotics for treating pulmonary exacerbations (Le Brun

et al., 2001), as a large dose may be administered within a reasonable time period, which may

result in improved therapeutic results. Further investigations on the highest lung dose that can

be reached with the newer devices within an acceptable time frame, are needed to estimate

the potential added value in treating pulmonary exacerbations. Le Brun et al. (2001) estimated

that a lung dose of 200 mg will result in a peak serum concentration of 9.2 mg/L, which is an

acceptable result in case of intravenous treatment with tobramycin. In general, a local drug

concentration of at least 4-5 times (beta-lactam antibiotics) to 10 times (aminoglycoside antibi-

otics) the minimal inhibitory concentration (MIC) of (susceptible) bacteria is needed for effective

antibiotic treatment. In case of resistant bacteria and/or treatment of exacerbations in CF, the

multiplication sum needed will be even higher. Obviously, the deposition pattern of the drug

in the lung will have an additional influence on the ultimate effect. Furthermore, the optimal

dosing frequency should be studied, as a prolonged elimination half life has been observed

after inhalation of a high 1000 mg dose of tobramycin (Le Brun et al. 2001). Obviously thorough

investigations, including therapeutic drug monitoring, risk on toxicity and collection of clini-

cal data, are required before the value of this hypothesis can be estimated. Unfortunately no

results from large clinical studies with aminoglycosides or colistimethate sodium administered

with these new nebulisers and DPI’s are available to date, not to mention the limited amount of

safety data (risk of toxicity) for normal and high dose administration of antibiotics.

Bridging studies

An inhaled drug should only be prescribed in combination with the inhaler device after (posi-

tive) results obtained from clinical studies with this combination in CF patients have become

available (Chapter 2, CHMP, 2006). This is valid for new drugs and new inhalers but should also be

applicable for older drugs that have been on the market for years. With respect to the numerous

inhaler devices that are commercially available, it is impossible to study every existing device

Chap

ter 9

164

with every existing drug. Therefore, bridging studies may be an answer to close the gap. In a

bridging study the tested drug-device combination in CF patients serves as a reference to a

drug-device combination not tested in this population. The comparison is made based on data

from in vitro (bench) studies. Currently sparse information is available on in vitro - in vivo correla-

tions in inhalation therapy, but knowledge on this subject is evolving (Weda et al. 2008, CHMP,

2006). The work described in this thesis is an example of combined in vitro – in vivo research:

after in vitro development studies on the drug and the devices, drug-device combinations have

been tested in vivo. The lack of information on local drug deposition within the lung in this work

however makes it difficult to connect in vitro and in vivo data. Incorporating in vivo scintigraphic

studies in future development plans will improve information on this subject.

Clinical efficacy from a different perspective

As discussed previously, clinical efficacy of treatment with aerosolized antibiotics in CF is guided

by a large variability in patients and drug-device combinations. Up till now, most research

initiatives on this subject have focused either on clinical efficacy or drug-device efficacy. The

need for combining these strategies has been explained in this thesis.

Another, new approach may prove to be useful in estimating the likelikhood of a successful,

effective aerosol treatment for an individual patient using experience from a large group of

patients (population) in the past. This goal might be reached by using existing data for pattern

recognition in the individual patient and to continuously improve and re-analyse this data set

by adding information obtained from that particular patient. Support from statistical method-

ology is necessary to accomplish this, for example with Bayesian methodology. This methodol-

ogy is frequently applied in pharmacokinetic calculations, also in CF research (chapters 6,7 and

8 of this thesis). Furthermore, the methodology has been used in risk calculations in prenatal,

carrier and neonatal screening in CF.

Bayesian methodology is a method of statistical inference which combines prior informa-

tion (including a probability distribution) about a population parameter with the data of, for

example, an individual patient. With the help of Bayes’s theorem, a posterior probability distri-

bution for the parameter is obtained, which provides the basis for statistical inference concern-

ing the parameter. To use this methodology in estimating clinical efficacy of aerosolized drugs,

several variables of interest have to be defined and coded first. After a substantial amount of

information per variable per patient has been collected, a population model can be obtained,

which can in turn be optimised by the data of each patient that is collected in future. If proven

successful, this approach can support clinical decision making by predicting with a certain prob-

ability that, for example, a male patient, 25 years old, FEV1 of 54% with a chronic Pseudomonas

aeruginosa infection for 2 years, might have the best clinical result using an inhaler X with drug

Y and dose Z twice daily. This hypothesis certainly deserves further exploration.


To conclude

Inhaler development

Rapid administration of an antibiotic aerosol with a small device, which is easy to use and has

an effective, reproducible and reliable performance, reflects the ideal situation for CF patients.

Present developments, such as the Twincer® inhaler, are promising. In addition to its simple

design and therefore low production costs, the concept of a single use disposable inhaler is an

advantage in preventing cross contamination of (resistant) bacteria and possible deterioration

of the performance of the nebuliser and stability of the drug because of accidental exhalation

into the inhaler. Additionally, the risk of environmental contamination during inhalation of a

drug dose, associated with conventional (jet) nebulisation, is negligible with the use of a dry

powder inhaler.

However, the position of dry powder inhalation of antibiotics in cystic fibrosis treatment is

still confined to pilot studies (this thesis, Pilcer et al. 2008, Geller et al., 2007). Future perspec-

tives are promising, as many potential improvements have not yet been explored. Long-term

studies are needed to assess the true value of the potential benefits of dry powder antibiotic

inhalers in cystic fibrosis.

Innovations and pharmaco-economics

The cystic fibrosis population is estimated at ~ 100,000 patients worldwide, with a high

likelihood of underreporting and underdiagnosing, especially in developing countries (Cystic

Fibrosis Worldwide, 2005). In terms of cost-effective drug development for manufacturers, this

population is small, making this field unprofitable for larger innovative pharmaceutical com-

panies. Therefore, an opportunity has been created to develop drugs in the framework of an

orphan drug programme both in Europe and the USA.

Well-designed, long-term studies with a large number of CF patients are required to assess

the clinical value of all newly developed drug-device combinations. This is a challenge by itself,

as the number of eligible CF patients per (specialised) CF centre is generally small. This thesis

includes three pilot-studies with a small number of subjects. These studies may form the basis

for the larger clinical trials. The limited number of patients in these three pilot-studies is partly

a result of the difficulty to find eligible patients.

To be able to include large numbers of patients, well-designed randomised multicentre

studies are required under professional supervision and coordination. Great expenses are

generally involved in this type of research, which is beyond the scope of many clinicians and

(academic) institutes. To gain new results from scientific work in an efficient way, multidisci-

plinary cooperation between scientists, clinicians, pharmacists and other professionals, under

which condition the research described in this thesis has been performed and as advocated by

the EuroCareCF (www.eurocarecf.eu) initiative, is essential for research in cystic fibrosis. Con-

sensus on e.g. outcomes, end-points and definitions of parameters in research and performing

Chap

ter 9

166

studies accordingly will facilitate the comparison of results from various bench- and clinical

studies. Furthermore, this will hopefully enable and facilitate the participation of CF patients

in an efficient way.

Next to technical innovations, the search for effective antibiotics to combat Pseudomonas

aeruginosa, Burkholderia cepacia and other Gram-negative bacteria will remain an ongoing

quest. Besides colistimethate sodium and tobramycin, other anti-pseudomonal antibiotics for

inhalation, such as amikacin, ciprofloxacin and aztreonam are also under investigation, either

for dry powder inhalation as for wet nebulisation. It is likely that these approaches will result in

further improvement in the daily treatment and life expectancy of cystic fibrosis patients.

In terms of the diversity in technical developments and drug formulation requirements, it

is mandatory to keep the total costs of inhalation treatment within a realistic range and to

perform pharmaco-economic research, as state-of-the-art treatment should be within reach for

every patient with cystic fibrosis. This is a shared responsibility of scientists, the drug and device

industry, national and international regulatory authorities and prescribers.

For cystic fibrosis patients, the last two decades have been characterised by an improved life

expectancy, better treatment options and rapidly developing new therapies. CF has become a

chronic disease with multiple treatment targets, resulting in increasingly complex treatment

strategies. More than ever it is important that multidisciplinary teams of physicians, nursing

staff, physiotherapists, dieticians, social workers and specialised pharmacists collaborate to

optimise treatment now and in the future. In addition, a joint effort with scientists and pharma-

ceutical companies to continuously search for new developments in drug aerosol therapy will

hopefully further improve life expectancy and quality of life.


References

Alderson PO, Secker-Walker RH, Stominger DB, Markham J and Hill RL. Pulmonary deposition of aerosols in children with cystic fibrosis. J Pediatr 1974; 4: 479-84.

Alothman GA, Alsaadi MM, Ho BL, Ho SL, Dupuis A, Corey M, et al. Evaluation of bronchial constriction in children with cystic fibrosis after inhaling two different preparations of tobramycin. Chest 2002; 3: 930-4.

Alothman GA, Ho B, Alsaadi MM, Ho SL, O’Drowsky L, Louca E, et al. Bronchial constriction and inhaled colistin in cystic fibrosis. Chest 2005; 2: 522-9.

Arias Llorente RP, Bousono Garcia C, Diaz Martin JJ. Treatment compliance in children and adults with Cystic Fibrosis. J Cyst Fibros 2008; 7: 359-67.

Asmus MJ, Stewart BA, Milavetz G, Teresi ME, Han SH, Wang D, et al. Tobramycin as a pharmacologic tracer to compare airway deposition from nebulizers. Pharmacotherapy 2002; 5: 557-63.

Bergen PJ, Li J, Rayner CR and Nation RL. Colistin methanesulfonate is an inactive prodrug of colistin against Pseudomonas aeruginosa . Antimicrob Agents Chemother 2006; 6: 1953-8.

Bowman CM. The long-term use of inhaled tobramycin in patients with cystic fibrosis. J Cyst Fibros 2002; Suppl 2: 194-8.

Chrystyn H, Allen MD, Corlett SA and Tomlinson HS. Simultaneous measurement of pharmacodynamic and pharmacokinetic parameters which can be used to evaluate the equivalence of inhaled salbutamol. Am Rev Respir Crit Care Med 1998; A636.

Chrystyn H. Methods to identify drug deposition in the lungs following inhalation. Br J Clin Pharmacol 2001; 4: 289-99.

Chua HL, Collis GG, Le Souef PN. Bronchial response to nebulized antibiotics in children with cystic fibrosis. Eur Respir J 1990; 3:1114-16.

Chuchalin A, Csiszer E, Gyurkovics K, Bartnicka MT, Sands D, Kapranov N, et al. A formulation of aerosolized tobramycin (Bramitob) in the treatment of patients with cystic fibrosis and Pseudomonas aeruginosa infection: a double-blind, placebo-controlled, multicenter study. Paediatr Drugs 2007; 21-31.

Committee for medicinal products for human use (CHMP). Guideline on the pharmaceutical quality of inhalation and nasal products. EMEA 2006.

Cunningham S, Prasad A, Collyer L, Carr S, Lynn IB and Wallis C. Bronchoconstriction following nebulised colistin in cystic fibrosis. Arch Dis Child 2001; 5: 432-3.

Cystic Fibrosis Worldwide. Annual report, 2005.Dodd ME, Abbott J, Maddison J, Moorcroft AJ, Webb AK. Effect of tonicity of nebulised colistin on chest

tightness and pulmonary function in adults with cystic fibrosis. Thorax 1997; 52:656-58.Eberl S, Chan HK and Daviskas E. SPECT Imaging for Radioaerosol Deposition and Clearance Studies. J

Aerosol Med 2006; 1: 8-20.Edson RS, Brey RH, McDonald TJ, Terrell CL, McCarthy JT and Thibert JM. Vestibular toxicity due to inhaled

tobramycin in a patient with renal insufficiency. Mayo Clin Proc 2004; 9: 1185-91.Falagas ME and Kasiakou SK. Colistin: the revival of polymyxins for the management of multidrug-resistant

gram-negative bacterial infections. Clin Infect Dis 2005; 9: 1333-41.Frederiksen B, Koch C and Hoiby N. Antibiotic treatment of initial colonization with Pseudomonas aeruginosa

postpones chronic infection and prevents deterioration of pulmonary function in cystic fibrosis. Pediatr Pulmonol 1997; 5: 330-5.

Geller DE, Konstan MW, Smith J, Noonberg SB and Conrad C. Novel tobramycin inhalation powder in cystic fibrosis subjects: pharmacokinetics and safety. Pediatr Pulmonol 2007; 4: 307-13.

Hodson ME and Gallagher CG. New clinical evidence from the European tobramycin trial in cystic fibrosis. J Cyst Fibros 2002a; Suppl 2: 199-202.

Hodson ME, Gallagher CG and Govan JR. A randomised clinical trial of nebulised tobramycin or colistin in cystic fibrosis. Eur Respir J 2002b; 3: 658-64.

Chap

ter 9

168

Hoffmann IM, Rubin BK, Iskandar SS, Schechter MS, Nagaraj SK and Bitzan MM. Acute renal failure in cystic fibrosis: association with inhaled tobramycin therapy. Pediatr Pulmonol 2002; 5: 375-7.

Kahler DA, Schowengerdt KO, Fricker FJ, Mansfield M, Visner GA and Faro A. Toxic serum trough concentra-tions after administration of nebulized tobramycin. Pharmacotherapy 2003; 4: 543-5.

Kettler LJ, Sawyer SM, Winefield HR, Greville HW. Determinants of adherence in adults with cystic fibrosis. Thorax 2002; 57:459-64.

Laube BL, Links JM, LaFrance ND, Wagner HN, Jr., Rosenstein BJ. Homogeneity of bronchopulmonary distri-bution of 99mTc aerosol in normal subjects and in cystic fibrosis patients. Chest 1989; 95:822-30.

Le Brun PP, Vinks AA, Touw DJ, Hekelaar N, Mannes GP, Brimicombe RW, et al. Can tobramycin inhalation be improved with a jet nebulizer? Ther Drug Monit 1999; 6: 618-24.

Le Brun PPH, Mannes GPM, Brimicombe RW, Frijlink HW and Heijerman HGM. Pharmacokinetic modelling of tobramycin after high dose inhalation in patients with cystic fibrosis. In: Optimization of antibiotic inhalation therapy in cystic fibrosis. Studies on nebulized tobramycin. Development of a colistin dry powder inhaler system. Thesis Rijksuniversiteit Groningen, The Netherlands 2001.

Lenoir G, Antypkin YG, Miano A, Moretti P, Zanda M, Varoli G, et al. Efficacy, safety, and local pharmacoki-netics of highly concentrated nebulized tobramycin in patients with cystic fibrosis colonized with Pseudomonas aeruginosa . Paediatr Drugs 2007; 11-20.

MacLusky I, Levison H, Gold R and McLaughlin FJ. Inhaled antibiotics in cystic fibrosis: is there a therapeutic effect? J Pediatr 1986; 5 Pt 2: 861-5.

MacLusky IB, Gold R, Corey M and Levison H. Long-term effects of inhaled tobramycin in patients with cystic fibrosis colonized with Pseudomonas aeruginosa . Pediatr Pulmonol 1989; 1: 42-8.

Marshall BG, Wangoo A, Harrison LI, Young DB and Shaw RJ. Tumour necrosis factor-alpha production in human alveolar macrophages: modulation by inhaled corticosteroid. Eur Respir J 2000; 4: 764-70.

McCoy KS. Compounded colistimethate as possible cause of fatal acute respiratory distress syndrome. N Engl J Med 2007; 22: 2310-1.

Moss RB. Long-term benefits of inhaled tobramycin in adolescent patients with cystic fibrosis. Chest 2002; 1: 55-63.

Mukhopadhyay S, Baer S, Blanshard J, Coleman M and Carswell F. Assessment of potential ototoxicity fol-lowing high-dose nebulized tobramycin in patients with cystic fibrosis. J Antimicrob Chemother 1993; 3: 429-36.

Mukhopadhyay S, Staddon GE, Eastman C, Palmer M, Davies ER and Carswell F. The quantitative distribution of nebulized antibiotic in the lung in cystic fibrosis. Respir Med 1994; 3: 203-11.

Newman SP, Flora M, Hirst PH, Klimowicz M, Schaeffler B, Speirs R, et al. Pharmacoscintigraphy of TOBI® in the Pari LC Plus™ and the Aerodose™ inhaler. J Aerosol Med 2001; 3: 388.

Newman SP, Pitcairn GR, Hirst PH and Rankin L. Radionuclide imaging technologies and their use in evaluat-ing asthma drug deposition in the lungs. Adv Drug Deliv Rev 2003; 7: 851-67.

Newman SP and Chan HK. In Vitro/In Vivo Comparisons in Pulmonary Drug Delivery. J Aerosol Med 2008; 21: 1-8.

Newnham DM, McDevitt, D.G., Lipworth, B.J. Comparison of the extrapulmonary b2-adrenoceptor responses and pharmacokinetics of salbutamol given by standard metered dose inhaler and modified activator device. Br J Clin Pharmacol 1993; 445-50.

Patton JS. Mechanisms of macromolecule absorption by the lungs. Adv Drug Deliv Rev 1996; 3-36.Pilcer G, Goole J, Van Gansbeke B, Blocklet D, Knoop C, Vanderbist F, et al. Pharmacoscintigraphic and phar-

macokinetic evaluation of tobramycin DPI formulations in cystic fibrosis patients. Eur J Pharm Biopharm 2008; 2: 413-21.

Poli G, Acerbi D, Pennini R, Soliani Raschini A, Corrado ME, Eichler HG, et al. Clinical pharmacology study of Bramitob, a tobramycin solution for nebulization, in comparison with Tobi. Paediatr Drugs 2007; 3-9.

Ramsey BW, Pepe MS, Quan JM, Otto KL, Montgomery AB, Williams-Warren J, et al. Intermittent adminis-tration of inhaled tobramycin in patients with cystic fibrosis. Cystic Fibrosis Inhaled Tobramycin Study Group. N Engl J Med 1999; 1: 23-30.


Scheenstra RJ, Heijerman HG, Touw DJ and Rijntjes E. Cochleotoxicity of systemically administered tobramy-cin in Cystic Fibrosis patients. J Cyst Fibros 2007; Suppl 1: S94.

Simon LJ, Chinchilli VM. A matched crossover design for clinical trials. Contemp Clin Trials 2007; 28:638-646.

Smith IJ and Parry-Billings M. The inhalers of the future? A review of dry powder devices on the market today. Pulm Pharmacol Ther 2003; 2: 79-95.

Smyth A, Tan KH, Hyman-Taylor P, Mulheran M, Lewis S, Stableforth D, et al. Once versus three-times daily regimens of tobramycin treatment for pulmonary exacerbations of cystic fibrosis--the TOPIC study: a randomised controlled trial. Lancet 2005; 9459: 573-8.

Steinkamp G, Tummler B, Gappa M, Albus A, Potel J, Doring G, et al. Long-term tobramycin aerosol therapy in cystic fibrosis. Pediatr Pulmonol 1989; 2: 91-8.

Tiddens HAWM. Detecting early structural lung damage in cystic fibrosis. Pediatr Pulmonol 2002; 228-231.Touw DJ, Knox AJ, Smyth A. Population pharmacokinetics of tobramycin administered thrice daily and once

daily in children and adults with cystic fibrosis. J Cyst Fibros 2007;6:327-333.Usmani OS, Biddiscombe MF and Barnes PJ. Regional lung deposition and bronchodilator response as a

function of beta2-agonist particle size. Am J Respir Crit Care Med 2005; 12: 1497-504.Vidgren. In vitro and in vivo deposition of drug particles inhaled from pressurized aerosol and dry powder

inhaler. Drug Dev Ind Pharm 1988; 2649-2665.Weda M, Zanen P, de Boer AH, Barends DM and Frijlink HW. The therapeutic index of locally acting inhaled

drugs as a function of their fine particle mass and particle size distribution: a literature review. Curr Drug Deliv 2008; 2: 142-7.

Summary

Summary 173

In this thesis, studies on the treatment of CF patients with antibiotic containing aerosols are

described. After an overview of the history of cystic fibrosis related to inhalation therapy in

chapter 1, chapter 2 deals with the first European consensus on inhaled medication and

inhalation devices in CF. This document describes the current knowledge on treatment with

inhaled drugs (among which are antimicrobial agents). Next to the drugs currently in use and

new drug therapies in development, the fundamentals of inhalation in CF treatment and practi-

cal aspects of inhalation of drugs (e.g. compatibility of mixtures and cleaning) are described.

This comprehensive review incorporates the fact that several new drugs, drug formulations and

new inhalation devices have been developed and have been marketed recently. The consensus

is intended as a reference document which may form the basis for an improved understanding

of inhalation treatment in CF by clinicians, pharmacists and other health care professionals and

for further research activities.

In chapter 3, the focus is on the administration of tobramycin for inhalation (TOBI®), using

an identical nebuliser with two different compressors in CF patients. Although the manufac-

turer of TOBI® has recommended a specific nebuliser and compressor for drug administration,

many patients use another combination for inhalation. The choice for a nebulising device is

often determined by the device that is already present at home for use with other drugs or by

the reimbursement policy of the health insurance company. This may result in variation of the

delivered dose and lung deposition compared to the reference combination and is likely to

influence treatment effects. For example, using a compressor with higher flow results in smaller

particles and a shorter nebulisation time. This may not only increase therapeutic efficacy but

also induce a higher risk on adverse effects due to a higher (peripheral) lung deposition. In

chapter 3 it was concluded that both compressors, the PortaNeb® and the CR60®, attached

to the PARI LC PLUS® nebuliser, can be used to nebulise tobramycin solution for inhalation,

but with the restriction that a higher risk on adverse effects or toxicity with the CR60® has not

been fully explored. On the other hand, an increased therapeutic efficacy would be a major

advantage. Both clinical effects should be subject for a future study.

This example of differences between compressors underlines the importance of studying

each nebulising device with a drug formulation for inhalation in a clinical situation in CF.

This requirement has been adopted by the European consensus on inhaled medication and

inhalation devices in CF (chapter 2) and is also incorporated in the EMEA Guideline on the

pharmaceutical quality of products for inhalation and nasal products, which states that ‘for

products for nebulisation the nebuliser system(s) and settings that were proven to be effective

and safe in vivo must be indicated, including information on the droplet size distribution, drug

delivery rate and total drug delivered’.

Colistin is a polypeptide antibiotic that is most potent in its colistin free base form. How-

ever, currently the sulfomethate salt, which needs hydrolisation to become active, is routinely

administered. More activity per milligram of drug is an advantage in dry powder inhalation,

174

Sum

mar

y

as, theoretically, less powder mass per inhalation is needed to obtain a comparable clinical

effect to a reference treatment (in this case wet nebulisation). This favours colistin sulfate over

colistimethate sodium, which is a prodrug that is converted in vivo into several metabolites and

colistin. The first explorations on a colistin dry powder inhaler were done with colistin sulfate,

but a subsequent study, described in chapter 4, demonstrated that colistin sulfate is not suit-

able for inhalation in CF patients due to its irritating effects. As a consequence, colistimethate

sodium was designated as the chemical form of choice, resulting in a larger dry powder mass

per dose, based on colistin activity per mg of dry powder mass.

In chapter 5, the development of a disposable dry powder inhaler (Twincer®) for high pow-

der doses is described. The development process of a dry powder inhaler is influenced by drug-,

device- and patient-bound parameters. It is the combination of these parameters that deter-

mines the success of the inhaler. This chapter focuses on the design and in vitro results of this

inhaler loaded with a colistimethate sodium dry powder mixture. It was shown that the inhaler

is able to produce high fine particle fractions as a result of an extremely high de-agglomeration

efficiency and with a total inhaler accumulation of only 5-6% at 4kPa (approx. 67 L/min). Its

simple design, the fact that it is for single use (disposable), the good moisture protection of the

drug formulation in a blister and low production costs make this inhaler suitable for applica-

tions in CF treatment but also for other aerosolized drugs for local or systemic use.

After having determined that colistimethate sodium is the colistin-salt of choice for inhala-

tion in CF-patients, a bench study showed that effective dispersion of colistimethate sodium

in a dose up to 25 mg in the newly designed Twincer® dry powder inhaler can be obtained,

resulting in particles (volume median diameter) with a X10, X50 and X90 of 0.7 μm, 1.6 μm, 3.1 μm

(chapter 7) and 0.9 μm, 2.1 μm and 3.8 μm (chapter 8) respectively, using dry powder mixtures

with different primary size distributions, measured by laser diffraction analysis and favourable

for peripheral deposition in the lung. The obtained fine particle fraction varied from 43.8% at

34 l/min (1 kPa) to 50.6% at 67 l/min (4 kPa) which indicates that the size distribution of the

emitted aerosol is comparable at different inspiratory flows (chapter 7).

These results show that the multiple air classifier technique, applied in the Twincer® inhaler,

is a highly effective de-agglomeration principle that enables patients to inhale a high drug

dose as a dry powder aerosol with a high fine particle fraction at a relatively low inspiratory

effort. A lower inspiratory flow reduces oropharyngeal and upper respiratory tract deposition

and increases deposition deeper in the lung.

Inhalation with this colistimethate sodium dry powder inhaler with the Twincer® inhaler in

pilot studies by small groups of healthy volunteers and cystic fibrosis patients (chapters 6, 7,

8) resulted in highly variable Cmax and AUC-values, which is not surprising as inter-patient vari-

ability in CF patients is well known. Comparison of inhalation with the colistimethate sodium

dry powder inhaler to nebulisation with the nebuliser-compressor in CF patients resulted in a

140% and 270% (chapter 7) or 140% and 300% (chapter 8) higher relative lung dose based

on actual dose and nominal dose respectively. However, absolute values of Cmax and AUC after

Summary 175

25 mg colistimethate sodium dry powder were lower compared to the 158 mg colistimethate

sodium solution. Based on these results, a dry powder dose of approximately 50 mg has been

estimated to result in a similar bioavailability (serum levels) compared to (jet) nebulisation

(data not shown). The volunteer study (chapter 6) showed that no detectable serum levels

were found after oral ingestion of 80 mg of colistimethate sodium and therefore all colistin

serum level measurements can be attributed to systemic absorption following inhalation.

These consistent results show that dry powder inhalation of colistimethate sodium with the

Twincer® inhaler is more efficient than wet nebulisation of the same drug using a Ventstream®-

PortaNeb® inhaler/compressor combination. Moreover, treatment with the Twincer® is less

time consuming compared to (jet) nebulisation (approximately less than one minute per dose

versus 30 minutes). Finally, the use of a disposable dry powder inhaler is safer from a microbio-

logical perspective.

Based on these findings, clinical studies on optimising the amount of drug that can be

inhaled in one manoeuvre with the Twincer® accompanied with studying peripheral deposi-

tion after inhalation are advocated.

Furthermore, although the optimal (lung) dose of colistimethate sodium or tobramycin is

not known, peripheral deposition studies with a focus on clinical efficacy will hopefully help to

find an answer to the question what dose should be put in which inhaler used by which patient

to obtain a maximal clinical response with minimal toxicity.

The results from chapters 6, 7 and 8 also show that differences between nominal and actual

inhaled doses (particle mass) in jet nebulisation are large: a large fraction of the dose stays

behind in the nebuliser cup. This observation stresses the importance of determining the actual

dose (output) in inhalation studies, as it provides information on the amount of drug that can

be made available for inhalation. This information is especially informative in jet nebulisation

studies, as jet nebulisers are known for their dose release inefficiency.

From the above it may be obvious that many parameters determine research outcomes in

CF inhalation research, predominantly guided by limitations because of lack of fundamental

knowledge of the inhalation of drugs in CF. With clinical efficacy as a starting point and central

objective at the same time, continuing scientific efforts are needed to improve inhalation

therapy for CF patients in future.

Samenvatting

Samenvatting 179

De laatste twee decennia is er grote vooruitgang geboekt in de behandeling van taaislijmziekte

(cystic fibrosis, CF). Zo is de mediane overleving gestegen van circa 25 jaar in 1985 naar circa 37

jaar in 2006. Toenemende aandacht voor en verbeterd inzicht in het ziekteproces heeft er onder

andere toe geleid dat het inhaleren van antibiotica een vaste pijler is geworden in de behande-

ling van CF-patiënten. Deze antibiotica zijn er allereerst op gericht om de Pseudomonas aeru-

ginosa bacterie te bestrijden, de bacterie die een grote rol speelt in de ontstekingsprocessen,

inclusief verbindweefseling (fibrose), in de long. Patiënten krijgen antibiotica voor inhalatie

voorgeschreven om te voorkomen dat een chronische ontsteking met deze bacterie ontstaat

of om een eenmaal bestaande chronische ontsteking te beteugelen. Daarnaast worden antibi-

otica per inhalatie en per intraveneuze infusie wel gecombineerd tijdens de behandeling van

een acute verergering van de ontstekingen in de long (acute exacerbatie).

Het inhaleren van deze antibiotica met de huidige mogelijkheden is een belastende bezig-

heid voor een CF-patiënt. Het klaarmaken en inhaleren van een dosis en het schoonmaken

van de apparatuur neemt ongeveer een half uur in beslag, veelal 2 maal per dag. Naast deze

antibiotica gebruikt een CF-patiënt nog meer geneesmiddelen, waardoor een aanzienlijk

gedeelte van de dag besteed wordt aan geneesmiddelengebruik. Het is niet verwonderlijk dat

de therapietrouw van CF-patiënten daar onder te lijden heeft. Het is daarom zinvol om elke

mogelijke verbetering van deze levenslange situatie te onderzoeken.

Dit proefschrift beschrijft een aantal onderzoeken dat uitgevoerd is om meer te weten te komen

over de behandeling met antibiotica per inhalatie in CF-patiënten.

Na een historisch overzicht van de taaislijmziekte in relatie tot behandeling per inhalatie

in hoofdstuk 1, wordt in hoofdstuk 2 aandacht besteed aan de huidige kennis over het met

inhalatie-geneesmiddelen (inclusief antibiotica) behandelen van patiënten met deze ziekte.

Naast al in gebruik zijnde geneesmiddelen passeren nieuwe middelen de revue, alsmede de

grondbeginselen van inhalatie in de behandeling van taaislijmziekte en praktische aspecten

van het inhaleren van medicatie (bijvoorbeeld het mengen van inhalatievloeistoffen en het

schoonmaken van apparatuur). Dit uitgebreide overzicht fungeert als de eerste Europese

consensus over inhalatiemedicatie en inhalatieapparatuur voor de behandeling van taai-

slijmziekte. Recente nieuwe ontwikkelingen op het gebied van inhalatie-apparatuur zijn in

de beschouwing meegenomen. Deze consensus is bedoeld als naslagwerk voor medische

beroepsbeoefenaren, (ziekenhuis)apothekers en andere werkers in de gezondheidszorg en

voor toekomstig vervolgonderzoek.

In hoofdstuk 3 wordt de toediening van tobramycine voor inhalatie (TOBI®) in CF-patiënten

onderzocht, gebruik makend van een identieke vernevelaar die beurtelings verbonden is

aan twee verschillende compressoren. Hoewel de fabrikant van TOBI® een specifieke verne-

velaar en compressor adviseert, is de praktijk van alledag dat patiënten meestal een andere

vernevelaar-compressor-combinatie gebruiken. Deze keuze wordt vaak bepaald doordat een

vernevelaar-compressor-combinatie reeds thuis aanwezig is voor de toepassing met andere

180

Sam

enva

ttin

g

inhalatiemedicatie of door het vergoedingsbeleid van de zorgverzekeraar. Dit kan tot gevolg

hebben dat de afgegeven geneesmiddeldosis uit de vernevelaar en de hoeveelheid genees-

middel die in de long terecht komt (longdepositie) variabel is ten opzichte van de vernevelaar-

compressor-combinatie die als referentie dient. Dit kan de effectiviteit van de behandeling

beïnvloeden. Zo kan bijvoorbeeld een compressor die een krachtiger luchtstroming tot stand

brengt, kleinere deeltjes voor inhalatie produceren en een kortere vernevelduur bewerkstelli-

gen. Dit resulteert mogelijk in een grotere therapeutische effectiviteit maar kan gelijktijdig ook

zorgen voor een groter risico op bijwerkingen (toxiciteit) doordat de longdepositie groter wordt

en deeltjes mogelijk dieper (perifeer) in de long terechtkomen. De conclusie in hoofdstuk 3 is

dat zowel de PortaNeb® als de CR60®-compressor gebruikt kunnen worden voor het verneve-

len van tobramycine voor inhalatie met de PARI LC PLUS® vernevelaar, met de beperking dat

een mogelijke grotere kans op bijwerkingen (toxiciteit) tijdens gebruik van de CR60® nog niet

volledig onderzocht is. Aan de andere kant zou een groter farmacotherapeutisch effect ook

zeer welkom zijn. Beide fenomenen dienen in vervolgonderzoek nader onderzocht te worden.

Dit voorbeeld van variabele uitkomsten tussen twee compressoren toont het belang van

het toetsen van elk inhalatieapparaat met een geneesmiddelformulering onder de klinische

omstandigheden van taaislijmziekte. Deze vereiste is overgenomen in de Europese consen-

sus voor inhalatiemedicatie en inhalatieapparatuur voor de behandeling van taaislijmziekte

(hoofdstuk 2) en is ook opgenomen in de EMEA Guideline on the pharmaceutical quality of

products for inhalation and nasal products. Daarin staat vermeld dat voor produkten, bedoeld

voor verneveling, de vernevelapparatuur en omstandigheden aangegeven dienen te worden

waarin effectiviteit en veiligheid in patiënten (in vivo) is aangetoond, inclusief informatie over

de deeltjesgrootteverdeling, snelheid van geneesmiddelafgifte en over de totale hoeveelheid

van het geneesmiddel die is afgegeven.

Colistine is een antibioticum dat uit veel aminozuren bestaat (polypeptide) en heeft als vrije

base het meest krachtige antibiotisch effect. Het colistinesulfomethaat-zout, dat gehydroly-

seerd dient te worden voordat het activiteit vertoont, wordt echter in de dagelijkse routine

toegepast. Meer activiteit per milligram geneesmiddel is een voordeel in droogpoederinhalatie,

omdat theoretisch gezien minder poedermassa nodig is om een vergelijkbaar behandelingsef-

fect te zien in vergelijking met een referentiebehandeling (in dit geval vloeistofverneveling).

Op basis hiervan verdient colistinesulfaat de voorkeur boven colistinesulfomethaat, dat in het

lichaam in meerdere tussenprodukten (metabolieten) en colistine omgezet wordt.

De eerste verkenningen voor een colistine droogpoederinhalator zijn uitgevoerd met colis-

tinesulfaat, maar in een vervolgonderzoek (hoofdstuk 4) is aangetoond dat colistinesulfaat

niet geschikt is voor inhalatie door CF-patiënten omdat het irriterend werkt in de long. Daarom

is colistinesulfomethaat als eerste keus aangewezen, met als gevolg een grotere droogpoeder-

massa per dosis, gebaseerd op de activiteit van colistine per milligram van de grondstof.

Samenvatting 181

In hoofdstuk 5 wordt de ontwikkeling van een droogpoeder-inhalator (Twincer®) voor een-

malige inhalatie van hoge geneesmiddeldoseringen beschreven. Het ontwikkelproces van een

droogpoederinhalator wordt beïnvloed door geneesmiddel-, apparatuur- en patiëntgebonden

parameters. De combinatie van deze factoren bepaalt uiteindelijk het succes van de inhalator.

Dit hoofdstuk gaat in op het ontwerp en de resultaten van laboratoriumonderzoek (in vitro

onderzoek) met deze inhalator gevuld met een droogpoedermengsel van het antibioticum

colistinesulfomethaat (colistimethate sodium). Aangetoond is dat de inhalator in staat is om een

grote fractie inhaleerbare deeltjes (fine particle fraction, FPF) te produceren doordat samenge-

klonterde of samengeplakte poederdeeltjes op een zeer efficiente wijze uit elkaar getrokken

worden (de-agglomeratie). De hoeveelheid poedermengsel die achterblijft in de inhalator is

dan ook slechts 5-6% bij een inhalatiesnelheid van circa 60 L/min. Het eenvoudige ontwerp

voor eenmalig gebruik, de goede bescherming van de geneesmiddelformulering tegen vocht

door de verpakking in een blister en de lage produktiekosten maken deze inhalator geschikt

voor toepassing bij de behandeling van patiënten met taaislijmziekte maar ook voor andere

geneesmiddelen die per inhalatie toegediend worden, voor zowel een lokaal effect (in de long)

als een systemisch effect (in het lichaam).

Nadat de keus voor deze zoutvorm van colistine gemaakt was, heeft laboratoriumonderzoek

aangetoond dat colistinesulfomethaat in een dosis tot 25 mg effectief fijn verdeeld (gedisper-

geerd) kan worden met de nieuw ontworpen Twincer® droogpoederinhalator, resulterend in

deeltjes (volume mediane diameter) met een X10, X50 en X90 van achtereenvolgens 0.7 μm, 1.6

μm, 3.1 μm (hoofdstuk 7) en 0.9 μm, 2.1 μm and 3.8 μm (hoofdstuk 8), gebruik makend van

droogpoedermengsels met een verschillende primaire deeltjesgrootte, gemeten met laser

diffractie en geschikt voor depositie diep in de long. De verkregen fijne deeltjesgrootte (FPF)

liep uiteen van 43.8% bij een inhalatiesnelheid van 34 l/min (1 kPa) tot 50.6% bij 67 l/min (4

kPa) respectievelijk, wat aantoont dat de deeltjesgrootteverdeling van de afgegeven aerosol

vergelijkbaar is bij verschillende inhalatiesnelheden (hoofdstuk 7).

Deze resultaten laten zien dat de ´multiple air classifier´ techniek, die toegepast is in de Twin-

cer® inhalator, een zeer effectief de-agglomeratie-principe heeft. Het biedt mogelijkheden aan

patiënten om hoge droogpoederdoseringen van een geneesmiddel met een grote fractie fijne

deeltjes (FPF) te inhaleren met een relatieve geringe inhalatiesnelheid. Een lagere inhalatiesnel-

heid vermindert de kans op het neerslaan van poederdeeltjes in de mond-keelholte en het

bovenste gedeelte van de luchtwegen en vergroot de geneesmiddeldepositie diep in de long.

De inhalatie met de colistinesulfomethaat droogpoederinhalator in proef-onderzoeken met

kleine groepen gezonde vrijwilligers en CF-patiënten (hoofdstukken 6,7,8) laat een grote vari-

atie zien in de maximale serumspiegel (Cmax) en biologische beschikbaarheid (AUC), hetgeen

niet verbazingwekkend is gezien de reeds bekend grote variabiliteit tussen CF-patiënten.

Vergelijking van de inhalatie van colistinesulfomethaat met de droogpoederinhalator en de

vernevelaar-compressor-combinatie in CF-patiënten geeft een relatief grotere longdosis van

140% en 270% (hoofdstuk 7) en van 140% en 300% (hoofdstuk 8), respectievelijk gebaseerd

182

Sam

enva

ttin

g

op de actuele dosis (particle mass) en nominale dosis. De absolute waarden van de Cmax en

de AUC na inhalatie van 25 mg colistinesulfomethaat-droogpoeder zijn lager dan de 158 mg

colistinesulfomethaat vloeistofdosis. Op basis van deze resultaten is berekend dat een droog-

poederdosis van ongeveer 50 mg vermoedelijk overeenkomt met een vergelijkbare biologische

beschikbaarheid (serumconcentraties) in vergelijking met (jet-)verneveling (hoofdstuk 8). Het

vrijwilligersonderzoek (hoofdstuk 6) heeft laten zien dat er geen meetbare serumconcentraties

gevonden worden na het opdrinken van 80 mg colistinesulfomethaatoplossing. Dat betekent

dat alle gemeten serumconcentraties in deze onderzoeken tot stand komen door opname van

het geneesmiddel vanuit de long het lichaam in.

Deze onderling vergelijkbare resultaten laten zien dat droogpoederinhalatie van colistine-

sulfomethaat met de Twincer® inhalator efficiënter is dan vloeistofverneveling van hetzelfde

antibioticum met de Ventstream®-PortaNeb® vernevelaar-compressor combinatie. Bovendien

kost de behandeling met de Twincer® minder tijd in vergelijking met jet-verneveling (minder

dan een minuut ten opzichte van ongeveer 30 minuten). Tot slot is een inhalator voor eenmalig

gebruik veiliger vanuit microbiologisch oogpunt.

Gebaseerd op deze bevindingen wordt geadviseerd om klinisch onderzoek uit te voeren naar

het optimaliseren van de hoeveelheid geneesmiddel dat, met de Twincer®, in één inhalatieteug

toe te dienen is, gepaard gaande met het bestuderen van de bereikte perifere longdepositie.

Hoewel de optimale (long)dosis van colistinesulfomethaat of tobramycin niet bekend is, zullen

longdepositie-onderzoeken, gericht op klinische effectiviteit, hopelijk ondersteuning bieden

bij het vinden van een antwoord op de vraag welke dosis in welk inhalatieapparaat voor welke

patiënt nodig is om een maximaal klinisch resultaat met een minimale toxiciteit te bereiken.

De resultaten in hoofdstuk 6,7 en 8 laten ook zien dat de verschillen tussen nominale en

actueel geïnhaleerde doseringen (particle mass) bij jetverneveling groot zijn: een groot gedeelte

van de dosis blijft achter in de vernevelaar. Deze waarneming onderstreept het belang van het

bepalen van de actuele dosis (output) in vernevelonderzoeken, omdat het informatie verschaft

over de hoeveelheid geneesmiddel die beschikbaar is voor inhalatie. Deze informatie is met

name nuttig in onderzoeken met jetvernevelaars, omdat jetvernevelaars bekend staan om een

niet-efficiënte dosisafgifte.

Het mag duidelijk zijn dat vele factoren de uitkomsten bepalen van het inhaleren van genees-

middelen door CF-patiënten. Dit onderzoeksveld wordt voornamelijk geleid door beperkingen,

gezien het gebrek aan kennis over de grondbeginselen van inhalatie van geneesmiddelen bij

taaislijmziekte. Met een effectieve behandeling van patiënten als tegelijkertijd uitgangspunt

en hoofddoel voor ogen, zijn voortdurende wetenschappelijke initiatieven nodig om inhalatie-

therapie voor patiënten met taaislijmziekte in de (nabije) toekomst te verbeteren. Hopelijk kan

het onderzoek dat beschreven is in dit proefschrift daaraan bijdragen.

Curriculum vitae 183

Curriculum vitae

Elsbeth Westerman werd geboren op 3 oktober 1972 te Rotterdam. In 1991 behaalde zij het

diploma gymnasium β aan het Rijnlands Lyceum in Wassenaar. In 1991 werd gestart met de

studie Farmacie aan de Universiteit Utrecht. De doctoraalopleiding werd afgerond in 1996

met een onderzoek in de ziekenhuisapotheek van het Universitair Medisch Centrum Utrecht

naar de populatiefarmacokinetiek van gentamicine bij pasgeborenen, onder begeleiding

van drs A. van Dijk, mevrouw dr. C.M.A. Rademaker en prof. dr. J.H. Glerum. In 1998 behaalde

zij het apothekersdiploma, waarna zij als apotheker werkzaam was in de Apotheek Haagse

Ziekenhuizen te Den Haag, het Laboratorium Nederlandse Apothekers (KNMP) te Den Haag

en het Albert Schweitzer Ziekenhuis te Dordrecht. In 1999 werd gestart met de opleiding tot

ziekenhuisapotheker in de Apotheek Haagse Ziekenhuizen. Het registratieonderzoek tijdens

de opleiding vormde de start van het promotieonderzoek dat is beschreven in dit proefschrift,

onder begeleiding van en in samenwerking met dr. H.G.M. Heijerman, prof. dr. H.W. Frijlink,

dr. D.J. Touw, dr. A.H. de Boer en dr. P.P.H. Le Brun. Voor dit onderzoek ontving zij de Christina

Onderzoekssubsidie 2002 van de Nederlandse Cystic Fibrosis Stichting (NCFS) en de Chanfleury

van IJsselsteynprijs 2007 van het HagaZiekenhuis te Den Haag.

Elsbeth Westerman is thans werkzaam als ziekenhuisapotheker in het HagaZiekenhuis /

Apotheek Haagse Ziekenhuizen te Den Haag.

List of publications 185

List of publications

Westerman EM, Le Brun PPH, Frijlink HW, Heijerman HGM. Sulfaat voldoet niet. Pharm Weekbl

2004; 138: 703-707.

Westerman EM, Le Brun PPH, Touw DJ, Frijlink HW, Heijerman HGM. Effect of nebulized colistin

sulphate and colistin sulphomethate on lung function in patients with cystic fibrosis: a pilot

study. J Cyst Fibros 2004; 3:23-28.

de Boer AH, Hagedoorn P, Westerman EM, Le Brun PPH, Heijerman HGM, Frijlink HW. Design

and in vitro performance testing of multiple air classifier technology in a new disposable inhaler

concept (Twincer) for high powder doses. Eur J Pharm Sci 2006; 28:171-178.

Hunfeld NG, Westerman EM, Boswijk DJ, de Haas JA, van Putten MJ, Touw DJ. Quetiapine in

overdosage: a clinical and pharmacokinetic analysis of 14 cases. Ther Drug Monit 2006; 28:

185-189.

Westerman EM, de Boer AH, Le Brun PPH, Touw DJ, Frijlink HW, Heijerman HGM. Dry powder

inhalation of colistin sulphomethate in healthy volunteers: A pilot study. Int J Pharm 2007;

335:41-45.

Westerman EM, Heijerman HGM, Frijlink HW. Dry powder inhalation versus wet nebulisation

delivery of antibiotics in cystic fibrosis patients. Expert Opin Drug Deliv 2007;4:91-94.

Westerman EM, De Boer AH, Le Brun PPH, Touw DJ, Roldaan AC, Frijlink HW, Heijerman HGMl.

Dry powder inhalation of colistin in cystic fibrosis patients: A single dose pilot study. J Cyst

Fibros 2007; 6:284-292.

Veltkamp SA, Westerman EM, Sprij AJ, Sum BS, Touw DJ. Gentamicin in preterm neonates: an

extended interval dosing schedule. EJHP-S 2007;13:92-97.

Sum BS, Veltkamp SA, Westerman EM, Sprij AJ, Touw DJ. An extended-interval gentamicin

dosage regimen in newborns: a prospective study. EJHP-S 2007;13:98-104.

Westerman EM, de Boer AH, Touw DJ, Le Brun PPH, Roldaan AC, Frijlink HW, Heijerman HGM.

Aerosolisation of tobramycin (TOBI®) with the PARI LC PLUS® reusable nebuliser: which com-

pressor to use? -comparison of the CR60® to the PortaNeb® compressor-. J Aerosol Med Pulm

Drug Deliv 2008; 21:269-280.

186

List

of p

ublic

atio

ns

Touw DJ, Westerman EM, Sprij AJ. Therapeutic drug monitoring of aminoglycosides in neo-

nates. Clin Pharmacokinet 2009; 48:71-88.

Heijerman HGM, Westerman EM, Conway SP, Touw DJ, Döring G. Inhaled medication and

inhalation devices for lung disease in patients with cystic fibrosis: a European Consensus. J Cyst

Fibros, submitted for publication

Dankwoord 187

Dankwoord

Veel mensen hebben bijgedragen aan de totstandkoming van dit proefschrift. Terwijl ik mij

realiseer dat de kans groot is dat ik onbedoeld iemand vergeet, wil ik graag een aantal van hen

via deze weg bedanken.

Op de eerste plaats wil ik de patiënten en vrijwilligers bedanken die hebben meegedaan aan

een of meerdere klinische onderzoeken die zijn beschreven in dit proefschrift. Elk met een eigen

motivering, elk met een eigen verhaal. De gemeenschappelijke drijfveer van de CF-patiënten,

namelijk om bij te dragen aan een verbetering van het inhaleren van medicatie, voor jezelf en

voor andere CF-patiënten, stond voorop. En sommigen reisden daarvoor zelfs een aantal malen

het halve land door.

Dr. H.G.M. Heijerman, beste Harry. Mijn co-promotor, maar vooral inspirator en motor achter dit

onderzoek. De omstandigheden waaronder ik de verschillende onderzoeken op de polikliniek

longziekten heb kunnen uitvoeren, de gastvrijheid en, vooral niet te vergeten, de door jou

georganiseerde ECFS-congresreizen met het multidisciplinaire CF-team van het HagaZieken-

huis laten goede herinneringen aan mijn onderzoeksperiode achter. Onze vrijdagmiddagbe-

sprekingen, bedoeld om de voortgang van het onderzoek door te nemen, resulteerden altijd

vrij snel in interessante discussies over totaal andere onderwerpen. Hoe dat nou kon?!

Prof. dr. H.W. Frijlink, beste Erik. Mijn promotor. Jouw enthousiasme en tomeloze energie heb-

ben steeds positieve impulsen gegeven aan het onderzoek, dank je wel. De nu al jarenlang

bestaande onderzoekslijn met de drie partners Apotheek Haagse Ziekenhuizen, het CF-centrum

van het HagaZiekenhuis en jouw afdeling werpt gestaag vruchten af. Het is voor mij een voor-

recht geweest om met de Twincer® droogpoederinhalator de eerste klinische onderzoeken bij

patiënten uit te mogen voeren.

Dr. D.J. Touw, beste Daan. ‘Mag ik wat vragen? Altijd.’ Dank je wel voor de goede samenwerking.

Het samen brainstormen over de farmacokinetiek van geïnhaleerde colistine en tobramycine

en jouw snelle en kritische antwoorden op mijn vragen hebben me reuze geholpen.

Dr. A.H. de Boer, beste Anne. Door jouw omvangrijke kennis, je kritische blik en nauwkeurig-

heid heb ik veel geleerd op het gebied van inhalatie en techniek. Het was altijd boeiend om

met jou te discussiëren, of het onderwerp nou inhalatieonderzoek, recente maatschappelijke

gebeurtenissen of -gewoon- Friesland betrof. Ik denk er met veel plezier aan terug.

188

Dan

kwoo

rd

Dr. P.P.H. Le Brun, beste Paul. Het estafette-stokje heb ik van jou overgenomen. Dank je wel voor

het helpen bij het opstarten van mijn onderzoek.

Corrie Haring, Henk Trumpie, Hans de Clippeleir, Sandra de Graaf en de andere medewerkers

van het Klinisch farmaceutisch en toxicologisch laboratorium van de Apotheek Haagse Zie-

kenhuizen: jullie werk vormt een van de pijlers van dit onderzoek. Waar zouden we zijn zonder

betrouwbare (serum)analyses van colistine en tobramycine? Ondanks drukte was het altijd

mogelijk om analyses ten behoeve van mijn onderzoek in te plannen. Hartelijk dank hiervoor.

Ik heb bewondering voor jullie vasthoudendheid als het met de analyse van colistine weer eens

tegen zat.

Medewerkers van de afdeling Receptuur van de Apotheek Haagse Ziekenhuizen: hartelijk

dank voor de ondersteuning bij het bereiden van de onderzoeksmedicatie.

Drs. B.H. Graatsma, beste Hayo. Dank voor de mogelijkheid om dit onderzoek vanuit de Apo-

theek Haagse Ziekenhuizen uit te voeren en het in mij gestelde vertrouwen dat de combinatie

met een volledige baan uitvoerbaar is. Het is gelukt!

Guus van der Meijden, het was een plezier om als team samen met jou de longfunctietesten

af te nemen bij de deelnemers aan de verschillende onderzoeken. Jouw kennis, kunde en

didactisch vermogen hebben zeker geleid tot betrouwbare uitkomsten.

Dr. A.C. Roldaan en Jane de Vries, beste Bert en Jane. Dank voor de hulp bij het recruteren van

deelnemers aan de verschillende onderzoeken. Dat het niet makkelijk was, weten we. Desal-

niettemin zijn we er toch goed in geslaagd!

Paul Hagedoorn en andere medewerkers van vakgroep Farmaceutische technologie en biofar-

macie van de RUG: hartelijk dank voor de gastvrijheid waarmee ik elke keer weer ontvangen

werd. Paul, met Anne als duo staan jullie aan de wieg van de Twincer®. Het was prettig om

met jou samen te werken. (Oud) promovendi Hans de Koning en Gerrit Zijlstra, dank voor de

adviezen en gesprekken. Cyril van Erp, dank je wel voor jouw ondersteuning bij mijn onderzoek

tijdens jouw bijvakperiode in Den Haag.

Sommige mensen hebben dit onderzoek mogelijk gemaakt door er te zijn als ik er niet kon

zijn: mijn collega-apothekers van het HagaZiekenhuis wil ik bedanken voor hun begrip en

collegialiteit waardoor ik het onderzoek met een gerust hart kon uitvoeren. Met name Judith

Schornagel-van Kemenade bedank ik voor de opvang tijdens de eerste onderzoeksjaren.

Het apotheekteam van het HagaZiekenhuis en andere medewerkers van de Apotheek

Haagse Ziekenhuizen en het HagaZiekenhuis bedank ik voor hun voortdurende belangstel-

ling.

Dankwoord 189

Erik Wilms en Nicole Hunfeld: wij zijn zo ongeveer gelijk opgetrokken met onze onderzoeken

in de Apotheek Haagse Ziekenhuizen. Ook voor jullie is de eindstreep in zicht. Succes met de

laatste loodjes!

Vrienden en vriendinnen, dank voor jullie belangstelling de afgelopen jaren. Het was altijd leuk

om vragen over het onderzoek te beantwoorden. Maar veel belangrijker: jullie waren er altijd

voor de broodnodige afleiding.

Charlotte en Kirsten, ik ben trots dat jullie mijn paranimfen willen zijn! Dank jullie wel voor jullie

niet aflatende interesse in mijn onderzoekstraject en de stimulans om het werk af te maken.

Als ervaringsdeskundigen weten jullie als geen ander hoe taai het soms kan zijn. Ambitieus

als jullie en ik zijn, liggen er vast nog meer mooie ontwikkelingen in het verschiet! Maar eerst

maken we er op 25 mei een bijzondere dag van.

Mijn (schoon)familie wil ik graag bedanken voor alle steun en interesse. Mijn zus, Anne Marie:

veel succes bij het afronden van jouw proefschrift. Soms lopen zaken anders dan voorzien: het

alsnog bereiken van het einddoel is een extra bijzondere prestatie!

Mijn ouders: ik ben jullie bijzonder dankbaar voor alles wat jullie mij meegegeven hebben.

Ik draag dit proefschrift graag aan jullie op. Lieve papa, dank je wel voor het meeleven, het

meelezen van (concept)versies en jouw wijze adviezen. Ik weet zeker dat mama net zo trots op

dit resultaat zou zijn als jij dat bent!

Lieve Antal, deze klus zit er op. Dank je wel voor je begrip, humor en relativeringsvermogen.

Het heeft mij absoluut geholpen bij het afronden van het onderzoek en manuscript. Als erva-

ringsdeskundige was je een goede raadgever. Geen berg te hoog! Welke bergwandeling zullen

we nu gaan maken?

Elsbeth

Rotterdam, maart 2009

studies on antibiotic aerosols for inhalation in cystic

Documents