Br J Clin Pharmacol 1996; 42: 697–705

Pharmacokinetics of inhaled drugs

BRIAN J. LIPWORTH
Department of Clinical Pharmacology, Ninewells Hospital and Medical School, Dundee DD1 9SY, UK

1 A high therapeutic ratio for the inhaled route of administration is achieved

by delivering doses which achieve a high local concentration in the lung and
relatively low levels of systemic absorption.
2 Pharmacokinetic evaluation of drug absorption from the lungs provides an
accurate and reproducible method for comparing different inhaler delivery
systems, as well as for evaluating bioequivalence of generic drug formulations.
3 The measurement of drug absorption from the lungs may also be applied to
assess the effects of inhalation technique on drug delivery in vivo. For example
with salbutamol delivered via a large volume spacer, lung bioavailability has
been shown to be altered by factors such as the number of actuated puffs,
inhalation-actuation delay and washing procedure.
4 Differences in drug delivery to the lungs between dry powder reservoir and
pressurised metered-dose aerosol devices translate directly into commensurate
differences in clinical efficacy for delivery of both inhaled b -adrenoceptor
2
agonists and corticosteroids.
5 For inhaled corticosteroids, pharmacokinetic evaluation using oral charcoal
to obviate alimentary absorption may be applied to quantify the relative gut
and lung components of systemic bioavailability. In tandem with information
on receptor potency and affinity, drug elimination and distribution, these data
may help in part to explain observed differences between different inhaled
corticosteroids in terms of their systemic bioactivity profiles.
6 Studies are required to evaluate whether pharmacokinetic evaluation of lung
absorption is a suitable way of quantifying delivery of nebulised aminoglycoside
antibiotics, as for example in patients with cystic fibrosis.
7 Pharmacokinetic evaluation appears to have an established role in the
quantification of drug delivery to the lungs and provides important information
which is complimentary to other techniques such as radiolabelled deposition.
The next decade of research into pharmacokinetics of established and novel
drugs and delivery systems is awaited with keen interest, and will hopefully
provide a greater understanding into ways of optimising the benefit-risk ratio
for inhaled drugs.

Keywords pharmacokinetics inhaled drugs lung bioavailability

Introduction adverse effects. This in turn results in a high therapeutic

ratio as compared with systemic delivery by oral or
The inhaled route of administration is the preferred parenteral routes of administration. Over the past
route for delivering preventer and reliever drugs to decade the delivery of inhaled drugs has been continu-
patients with airflow obstructions. The inhaled route ously refined by inhaler delivery devices which optimise
allows the delivery of relatively small doses of drug the deposition of respirable particles.
directly to the airway achieving a high local concen- Why should we be interested in the pharmacokinetics
tration, whilst at the same time minimising systemic of inhaled drugs? The application of pharmacokinetics

Correspondence: B. J. Lipworth, Department of Clinical Pharmacology, Ninewells Hospital and Medical School, Dundee DD1
9SY, UK

© 1996 Blackwell Science Ltd 697

698 B. J. L ipworth

provides a valuable tool for investigating lung deposition enhanced delivery of salbutamol from a low velocity
and bioavailability, and hence ways of optimising drug modified vortex metered-dose actuator (Spacehaler,
delivery. At the same time, pharmacokinetic techniques Evans) resulted in a greater plasma C compared with
max
can be used to predict and explain the systemic adverse an ordinary actuator [2 ]. As one might perhaps predict,
effect profiles of inhaled drugs. The widespread avail- the higher plasma concentration of salbutamol was
ability of sensitive assay techniques, such as high associated with a leftward shift in the dose-response
performance liquid chromatography allows the measure- curve for extrapulmonary b -mediated responses. The
2
ment of plasma and urinary drug levels required for same methodology was subsequently used to compare
pharmacokinetic evaluation. The purpose of this review two nebuliser delivery systems, namely the Hudson
is to appraise the literature on pharmacokinetics of Updraft II and the Ventstream. In vitro studies had
commonly used inhaled drugs and how this may be shown that the Ventstream produced a significant
applied to optimising their use in every-day clinical increase in output of respirable particles compared with
practice. the Hudson [3]. Since the Venstream also increases the
drug delivery by matching the nebuliser output to
inspiratory tidal flow rate, one might expect to find a
difference in lung deposition of salbutamol when
Inhaled b -adrenoceptor agonists comparing the two nebuliser systems. In a study of
2
asthmatic patients (FEV 55% predicted ) [3 ] the
1
Following inhalation of salbutamol, most of the dose Venstream produced an enhanced delivery of salbutamol
(60–80%) is delivered to the oropharynx and hence to across a dose range of 1.25 mg, 2.5 mg and 5 mg,
the gut after swallowing, with a much smaller fraction resulting in an approximate two-fold improvement in
(10–20%) reaching the lungs. The polarity of salbutamol lung absorption as assessed by plasma C or area
max
at salivary pH results in only negligible buccal absorp- under the concentration-time profile (AUC) (Figure 1 ).
tion, and the swallowed fraction undergoes extensive This was associated with a difference between nebulisers
first-pass conjugation in the intestinal wall and liver. in AUC values for bronchodilator and systemic
Since there is no first-pass conjugation in the lung, this responses. Furthermore, each nebuliser exhibited linear
means that initial plasma levels of unchanged salbutamol pharmacokinetics with doubling doses of salbutamol.
will reflect the dose delivered to the lung. Direct The pharmacokinetic approach may also be applied
measurement of plasma salbutamol given by pressurised to evaluate the bioequivalence of different formulations
metered-dose inhaler shows that maximal plasma con- of inhaled salbutamol. In one such study no differences
centration (C ) is achieved within 5–10 min of inha- were found in lung absorption of salbutamol comparing
max
lation, consistent with rapid absorption from the lung three different formulations of inhaled salbutamol given
[ 1,2]. It is however, unclear what proportion of this by pressurised metered dose inhaler [4]. However, in a
initial absorption originates from alveolar or bronchial comparison of a non-CFC metered-dose inhaler formu-
sites. This approach provides a simple method to lation of salbutamol (Airomir, 3M ) with a CFC metered-
compare the relative drug deposition of salbutamol from dose inhaler formulation ( Ventolin, Allen & Hanburys),
different inhaler devices by making a direct measurement a 1.3-fold greater ratio for C was found [5]
max
of lung absorption. In an initial study it was shown that (Figure 2 ). This probably reflects the lower aerosol

1.25 mg 2.5 mg 5.0 mg

5 5 5 *
*
*
Plasma salbutamol (ng ml )
–1

4 4 4 *
*
3 3 3 *
*
* *
* * V
2 2 V 2
* *
* H
1 V 1 1
H

H
0 0 0
5 10 20 40 5 10 20 40 5 10 20 40 (60) (120) (240)
Time (min) Time (min) Time (min)

Figure 1 Pharmacokinetic time profiles for plasma salbutamol after inhalation of 1.25 mg, 2.5 mg (1.25 mg+1.25 mg) or 5.0mg
(2.5 mg+2.5 mg) as sequential cumulative doses administered via a Ventstream (V ) or Hudraft Updraft II ( H) nebulisers to
eight asthmatic patients (FEV 55±2% predicted). Values shown are means and s.e.mean. Asterisk denotes a significant
1
difference between the two nebulisers. (T aken from: Newnham & L ipworth, T horax 1994; 49: 762–770, with permission of the
BMJ publishing group [3 ]).

© 1996 Blackwell Science Ltd British Journal of Clinical Pharmacology 42, 697–705
 Pharmacokinetics of inhaled drugs 699

4.5 * 50 6

Urinary salbutamol (µg 30 min –1)

Plasma salbutamol (ng ml–1)
* 5
4.0 45 **

Cmax (ng ml–1)

3.5 40
*** ***
3
3.0 35
2

2.5 // 30
Cmax Cav Urine 1

Figure 2 Lung absorption of salbutamol, shown as plasma

0
salbutamol concentration: for peak (C ) and average (C )
max av SP MP SP delay AS
levels measured over 20 min after inhalation of 1200 mg of
salbutamol given via chlorofluorocarbon containing metered-
0.7
dose inhaler, Ventolin (CFC, %), CFC-free metered-dose
inhaler, Airomir (CFC-free, &) and Diskhaler dry powder
0.6
inhaler ( DPI, b). Also shown are values for relative lung
absorption as 30 min urinary salbutamol excretion. Asterisk
0.5

∆log TR (log units)

denotes significant difference between CFC-free and either of
the other inhalers. There were no differences in urinary
0.4 *
salbutamol excretion. (Adapted from data in: Clark &
L ipworth, Br J Clin Pharmacol 1996; 41: 247–249 [ 5]).
0.3 ***

0.2

particle plume velocity with the non-CFC formulation, 0.1

as has been demonstrated in vitro [6 ]. This observa-
tion is in keeping with the similar finding of enhanced 0
lung absorption with a modified actuator device SP MP SP delay AS
(Spacehaler), which also had a reduced aerosol plume
Figure 3 Peak concentration (C ) of plasma salbutamol
velocity [2 ]. Salbutamol delivered by CFC containing max
(top) and finger tremor ( TR) response ( bottom) after
metered-dose aerosol ( Ventolin) and dry powder inhaler inhalation of 1200 mg salbutamol: via single puffs with
( Ventodisk, Allen & Hanburys) results in comparable volumatic spacer (SP), multiple puffs with spacer (MP),
lung bioavailability, suggesting that both types of inhaler single puffs with 20 s delay before inhalation from spacer
devices can be prescribed on a microgram equivalent (SPdelay), and single puffs with an antistatic treated spacer
basis [5] (Figure 2). (AS). All spacers were pre-washed prior to each study day.
In a study of salbutamol delivered via a large volume Asterisk denotes significant difference between SP and any
spacer device observed values for C were 1.9-fold other inhalation sequence (*P<0.05, **P<0.01,
max ***P<0.001). Values are means and s.e.mean. (Modified
greater using single puffs without delay compared with
after: Clark & L ipworth, T horax 1996, in press, with
the same dose as multiple puffs (4 puffs at a time), and
permission of the BMJ Publishing Group [7]).
1.8-fold greater compared with the same dose using an
inhalation delay of 20 s [7]. These effects were mirrored
by commensurate differences in extrapulmonary b -
2
adrenoceptor activity (Figure 3 ). The effects of multiple percentage for lung deposition if a standard intravenous
actuations and inhalation delay are similar to those injection of the same drug is used as a pharmacokinetic
observed in vitro for salbutamol, corticosteroids and internal standard for comparison. There appears to be
cromoglycate delivered by large volume spacer [8–10 ]. good agreement between estimates of lung deposition of
In vitro studies have also shown that reduction in terbutaline using pharmacokinetic and technetium
electrostatic charge within the spacer using an antistatic radiolabel aerosol methods.
lining is associated with a significant improvement in In a pharmacokinetic study inhaled terbutaline in
delivery of respirable particles [11 ]. The pharmaco- a dose of 250 mg produced a lung deposition of 19%
kinetic technique was applied to show that washing the when administered by dry powder reservoir device
spacer is an effective way of eliminating such electrostatic (Turbuhaler, Astra) compared with 8% when delivered
charge and hence optimising drug delivery [ 7]. by metered-dose inhaler [13]. At the same time a dose
An alternative, albeit indirect, method of evaluating ratio of 251 was demonstrated in terms of equivalent
pharmacokinetics of inhaled b -adrenoceptor agonists bronchodilator response comparing 250 mg of terbutaline
2
is to measure urinary excretion. One application of this via a turbuhaler and 500 mg via a metered-dose inhaler.
technique involved the measurement of urinary excretion However, such dose ratios are better evaluated by
of inhaled terbutaline, using oral activated charcoal to performing proper dose-response studies providing this
eliminate the gastrointestinal component of systemic is done on the steep part of the curve. In one such
bioavailability [12 ]. This enables calculation of a study, the bronchodilator-dose response (measured as

© 1996 Blackwell Science Ltd British Journal of Clinical Pharmacology 42, 697–705
700 B. J. L ipworth

FEV ) showed a 1.5-fold dose-ratio comparing terbuta- systemic bioavailability

1
line given by turbuhaler and metered dose inhaler.
Furthermore, the 1.5-fold greater ratio for broncho- 1000 µg
dilator efficacy was exactly mirrored by the ratio for
hypokalaemic response, inferring that lung deposition Gut 50% Lung 20%
determines both airway and systemic effects [14 ].
Another indirect urinary pharmacokinetic method
involves using the measurement of 30 min urinary 500 µg Deposition 200 µg
excretion of salbutamol and its conjugates in order to
differentiate between the components of lung and
gastrointestinal absorption [15 ]. This premise depends 90% 99% first-pass No first-pass
on the assumption that most of the urinary excretion of
unchanged salbutamol in the first 30 min after inhalation Systemic
50 µg 5 µg 200 µg
is due predominantly to absorption from the lungs. circulation
However, it is pertinent to note that in a study where it BUD FP
was possible to detect significant differences in plasma
Figure 4 Schematic representation of the relative lung and
salbutamol between non-CFC and CFC containing
gut components comprising total systemic bioavailability for
metered-dose aerosol formulations, the same was not inhaled budesonide ( BUD) and fluticasone propionate (FP),
the case when using 30 min urinary excretion (Figure 2) both given via the same inhaler device. Because of extensive
[ 5]. Furthermore, the interindividual coefficient of hepatic first-pass but no lung first-pass inactivation for both
variability is approximately two-fold greater with the drugs, the main determinant of systemic bioavailability is
urinary compared with the plasma method [4 ]. This that derived from the lung. Mouth rinsing, to reduce local
suggests that given the availability of a sufficiently adverse effects, will act to attenuate further the small
sensitive assay, the direct measurement of plasma component of gut bioavailability.
salbutamol is the preferred method to quantify absorp-
tion from lungs. Inhaled corticosteroids
On first principles, it might be predicted that lung
deposition and hence, bioavailability from the lung, Surprisingly little has been published on the pharmaco-
would be altered in patients with airflow obstruction as kinetics of inhaled corticosteroids. This probably reflects
a consequence of reduced small airway calibre and the lack of widely available sensitive assays for measuring
hence reduced peripheral lung absorption. In this respect plasma levels. Earlier studies with tritium labelled
there are some data to show that asthmatics with a inhaled budesonide showed a plasma elimination half-
50% reduction in predicted normal FEV have an life of 2.0 h [19]. Calculated values for oral budesonide
1
approximate two-fold reduction in fenoterol C com- showed a systemic bioavailability of 11% indicating a
max
pared with normals given the same dose (1.6 ng ml−1 high degree of first-pass hepatic metabolism (89%) of
in asthmatics vs 3.1 ng ml−1 in normals) [1,16]. the swallowed dose. For the inhaled route the systemic
Furthermore the difference in plasma fenoterol concen- bioavailability was calculated at 73% indicating a
tration between the two groups was associated with a considerable absorption from the lung of the unchanged
marked difference in the chronotropic response to drug. This is in keeping with in vitro studies showing
fenoterol, being greater in the group of normal subjects. extensive biotransformation in the liver but not at all in
This is in keeping with the observation that enhanced the lung [20]. For beclomethasone diproprionate partial
absorption from the lung is associated with increased biotransformation to beclomethasone-17-monoprop-
systemic b -adrenoceptor mediated effects [2 ]. ionate, an active metabolite, occurs in both liver and
2
The use of plasma pharmacokinetics has also been lung. Precise pharmacokinetic evaluation of the degree
employed to evaluate the effect of inhaler technique on of liver and lung first-pass metabolism of beclome-
delivery of terbutaline by the dry powder reservoir thasone dipropionate in humans is not available.
inhaler system [17 ]. These results show that lung However, in terms of hepatic first-pass metabolism of
volume prior to inhalation had no effect on lung the swallowed dose, data from a recent study using oral
absorption when used with a peak inspiratory flow rate activated charcoal, revealed approximately 60–70%
in excess of 60 l min−1. At an inspiratory flow rate of first-pass hepatic inactivation [21 ]. Fluticasone pro-
30 l min−1 drug absorption was reduced to a small prionate like budesonide has no lung first-pass metab-
degree, and the bronchodilator dose-response curve was olism, but has 99% hepatic first-pass metabolism of the
not significantly altered. When subjects first exhaled swallowed dose [22 ].
into the turbuhaler prior to inhalation, there was a Therefore it is the lung rather than the gut component
marked reduction in plasma levels and this was of absorption which will be the main determinant of the
accompanied by a commensurate right shift in the overall systemic bioavailability, at least for budesonide
bronchodilator dose-response curve as well as systemic and fluticasone propionate where there is extensive
b -mediated effects. Improvements in drug delivery to hepatic first-pass but no lung first-pass metabolism
2
the lung using 30 min urinary salbutamol excretion have (Figure 4 ). For beclomethasone dipropionate the lower
been shown with a metered dose inhaler in association degree of hepatic first-pass inactivation results in the
with exhaling to residual volume, slow inhalation and swallowed fraction of the inhaled dose assuming greater
breath holding [18 ]. importance in determining the systemic activity profile.

© 1996 Blackwell Science Ltd British Journal of Clinical Pharmacology 42, 697–705
 Pharmacokinetics of inhaled drugs 701

The differences in the hepatic first-pass inactivation for lung deposition appears to translate into a commen-
therefore explain the observation that for budesonide surate dose-ratio for clinical efficacy. In a double-blind
the addition of a large volume spacer to a metered dose randomized double-dummy cross over study of 241
inhaler will increase systemic bioactivity ( by increasing stable asthmatic children, halving their dose of budeson-
lung absorption), whilst for beclomethasone diproprion- ide via the nebuhaler resulted in a clinical relapse in
ate the addition of a spacer reduces systemic activity 126 cases [26 ]. Of these 126 patients, 64 were ran-
( by reducing gut absorption). In other words, for domized to continue with their usual budesonide dose
beclomethasone dipropionate the reduction in gastro- via the nebuhaler, with the other 62 being randomized
intestinal absorption with the spacer outweighs the to use half their usual dose via the turbuhaler. After 9
concomitant increase in lung absorption, because of the weeks of evaluation, there were no differences between
low degree of hepatic first-pass inactivation. Thus, for the two groups in terms of symptom control, spirometry,
fluticasone propionate like budesonide, the use of a peak flow rate or exercise challenge. A similar degree of
large volume spacer would increase systemic bioactivity, corticosteroid dose-reduction for turbuhaler vs nebu-
because lung absorption is the major determinant, and haler has been reported during step-down therapy to
this would be increased with the spacer. In any event, identify the lowest maintenance dose of inhaled budeson-
mouth-rinsing should be routinely employed when using ide over a prospective 4 year period of follow-up in
inhaled corticosteroids to reduce local adverse effects asthmatic children [27]. This suggests that at least with
such as oropharyngeal candidiasis. This in turn acts to inhaled budesonide, greater lung deposition with the
reduce further the component of gut bioavailability. turbuhaler allows lower maintenance doses during the
Detailed studies have evaluated the plasma pharmaco- step-down phase of the management guidelines.
kinetics of inhaled budesonide in children and adults. The pharmacokinetics of inhaled corticosteroids may
In a study of adult healthy volunteers a single 1000 mg also determine the systemic adverse effect profile,
dose of budesonide was given via a turbuhaler or particularly for drugs such as budesonide and fluticasone
metered dose inhaler along with intravenous adminis- propionate which have high hepatic first-pass, but no
tration of a 500 mg dose of budesonide as a reference lung first-pass metabolism. Fluticasone propionate has
[ 23]. The same treatments were also administered with approximately 2–3 times greater glucocorticoid potency
concomitant oral charcoal to obviate gastrointestinal than budesonide [28 ]. Thus if lung absorption is the
absorption and hence evaluate the bioavailability from major determinant of systemic bioactivity, one might
the lung. The pharmacokinetic profile for both inhaler expect to find commensurate differences between the
devices showed rapid absorption with a t for two drugs in terms of their respective systemic bioactiv-
max
budesonide of 0.3 h and elimination half-life of 2.3 h. In ity. This hypothesis was investigated in a comparison of
the presence of charcoal-block the lung bioavailability single inhaled doses of budesonide and fluticasone
was calculated at 32% for the turbuhaler vs 18% for ranging from 400–2000 mg, both given by metered-dose
the metered dose inhaler. Comparative values for total aerosol with mouth rinsing to asthmatic adults in a
bioavailability in the absence of charcoal (i.e. lung plus placebo controlled double-blind randomized cross-over
gut) were 38% for the turbuhaler against 26% for the study [29]. Suppression of overnight urinary cortisol
metered dose inhaler, giving a clear indication of the excretion, a sensitive marker of adrenal activity, showed
relative lung and gut components. approximately two-fold greater suppression with flut-
In asthmatic children given inhaled (1000 mg) and icasone 500 mg than with budesonide 400 mg. For 08.00 h
intravenous (500 mg ) doses of budesonide without serum cortisol and ACTH, at doses above 1000 mg on
charcoal, total systemic bioavailability was calculated at the steep part of the dose-response curve there was
approximately 30% of the nominal inhaled dose from a three-fold greater adrenal suppression with fluticasone
tube spacer (Inhalet) and 15% from a nebuliser (Pari- than with budesonide on a mg equivalent basis. Similar
inhalerboy) [24 ]. Data from the same study revealed a differences in systemic activity have been found compar-
value for clearance of budesonide which was 40% higher ing single inhaled doses of fluticasone and budesonide
than comparable values in adults, along with a shorter (400–1250 mg) given by spacer to asthmatic children,
elimination half-life of 1.5 h. The higher clearance and with three-fold greater suppression of urinary cortisol
shorter half-life of budesonide in children would seem excretion [30 ].
to be advantageous in terms of reducing the burden of Thus it is evident that lung absorption of a cortico-
systemic adverse effects. steroid with enhanced potency produces greater systemic
The charcoal-block method has also been used to activity on the steep part of the dose-response curve.
investigate the pharmacokinetics of plasma budesonide The plasma elimination half-life of inhaled budesonide
given as a 1000 mg dose via a turbuhaler to asthmatic is 2.3 h compared with 14.4 h for inhaled fluticasone
children. It was found that charcoal reduced the systemic [23,31]. The longer elimination half-life for fluticasone
absorption (as AUC (0,4 h)) by approximately 20%. The should result in greater steady-state accumulation during
absorption of the same dose of budesonide was higher repeated dosing. This is supported by a study where
comparing turbuhaler with nebuhaler ( both without there was a large step-up in adrenal suppression between
charcoal ) being two-fold greater [ 25]. Allowing for single and repeated doses of fluticasone 1000 mg twice
greater oropharangeal deposition of budesonide with a daily (25% vs 55% suppression), whereas with budeson-
turbuhaler than with a nebuhaler, these data suggest an ide 800 mg twice daily this effect was much less
almost two-fold greater lung delivery of budesonide pronounced (26% vs 34% suppression) [32 ]. Other
when comparing the two devices. Interestingly, this ratio factors which may contribute to greater steady-state

© 1996 Blackwell Science Ltd British Journal of Clinical Pharmacology 42, 697–705
702 B. J. L ipworth

suppression with fluticasone include a more prolonged three-fold difference in AUC (0,240 min), and was
receptor residency time [33 ] and greater lipophilicity mirrored by similar differences in C . There were
max
[ 34] compared with either budesonide or beclome- however, no differences in plasma concentration-time
thasone dipropionate. The higher degree of lipophilicity curves when comparing inhalation from a spinhaler
for fluticasone may result in enhanced systemic tissue with and without a 10 s breath hold at the end of
retention in fat stores, in effect acting as a reservoir in inspiration. In the same study pharmacokinetics of
the body at steady-state. sodium cromoglycate delivered as a solution directly
into the airways was investigated in patients undergoing
diagnostic bronchoscopy. The plasma concentration-
time profile was found to be similar to those obtained
Inhaled sodium cromoglycate and nedocromil sodium after normal inhalation and values for C and
max
AUC(0,240 min) were intermediate between the higher
The pharmacokinetics of inhaled sodium cromoglycate two inspiratory flow rates for the spinhaler. Comparison
and nedocromil sodium has been extensively evaluated of the AUC values after bronchoscopic administration
both in normal subjects and in patients with obstructive with those by spinhaler indicates that only 10% of the
airways disease. After inhalation of sodium cromoglycate nominal dose reaches the airways. Furthermore compari-
there is rapid absorption from the alveolar vascular bed son of the AUC after bronchoscopic administration
with peak plasma levels being reached within 20 min with that after intravenous infusion suggests that
[ 35–38]. approximately 70% of the drug was bioavailable in
However absorption appears to occur at two different terms of absorption from the lungs.
rates with an initial fast phase from the alveoli and a The effects of bronchoconstrictor challenge with
later slow phase from the bronchial epithelium. This methacholine, histamine and adenosine monophosphate
slower phase of absorption is probably due to the high (AMP) on the pharmacokinetics of inhaled sodium
degree of hydrophilicity of sodium cromoglycate which cromoglycate have been evaluated in a series of studies
results in a lesser propensity for absorption from the from the same laboratory. Methacholine challenge
bronchial epithelial barrier compared with the endo- resulted in a 23% lower FEV and a 2.8-fold higher
1
thelium of the alveolar vascular bed. Thus, inhaled central5peripheral lung deposition ratio (with
sodium cromoglycate exhibits absorption rate-limited technetium-99m) in comparison with saline [ 41].
or ‘flip-flop’ kinetics, as has been shown by comparison Inhalation of methacholine was associated with signifi-
of elimination after intravenous and inhaled dosing cant increase in C but not in AUC for plasma
max
[ 35–37]. cromoglycate. The greater C after methacholine is
max
The urinary excretion of sodium cromoglycate given difficult to explain if initial rapid absorption occurs
by dry powder inhaler (Spinhaler, Fisons) has been used from the peripheral alveolar vascular bed, particularly
to evaluate the effect of airway calibre on lung since technetium studies suggested increased central
bioavailability in patients with obstructive airways deposition after methacholine inhalation.
disease compared with normal controls [ 39]. Urinary The effects of inspiratory flow rate on pharmaco-
excretion as a measure lung bioavailability is possible, kinetics of sodium cromoglycate given by spinhaler and
because when sodium cromoglycate is given orally less its protection against AMP challenge were evaluated in
than 2% is absorbed, and when given by the intravenous asthmatic subjects [42 ]. Values for C and AUC
max
or inhaled route the drug is also excreted unmetabolized. showed proportional attenuation with associated
It was found that urinary excretion of sodium cromog- reduction in inspiratory flow rate. Furthermore, both
lycate was markedly reduced in patients with chronic inspiratory flow rate and AUC correlated significantly
bronchitis but not in asthmatics as compared with with the degree of protection afforded against AMP
normal controls. The explanation for this may be the induced bronchoconstriction. This shows that the
combined effects of a lower FEV and a lower inspiratory inspiratory flow rate used to inhale sodium cromoglycate
1
flow rate in the chronic bronchitic compared with dry powder is an important determinant of protection
asthmatic group, resulting in reduced lung bioavail- against bronchial challenge. The smaller degree of lung
ability. However in another study looking at lung absorption at lower inspiratory flow rates probably
bioavailability of sodium cromoglycate given via a reflects a reduction in fragmentation of particles in the
spinhaler, there was a reduction in plasma C and spinhaler device and hence a lower delivered mass of
max
AUC comparing asthmatic patients vs normal volun- respirable particles.
teers, and this was also mirrored by urinary excretion In another study following histamine inhalation which
[ 40]. There was a large inter-subject variability in produced a 20% fall in FEV , initial absorption of
1
plasma concentration of sodium cromoglycate which sodium cromoglycate from lung was significantly
was attributed to differences in inhalation technique, increased as evidenced by lower values for t and
max
particularly with respect to inspiratory flow rate and absorption half-life [43]. This phenomenon was
breath holding. observed in both histamine non-responsive (control
In a study of normal subjects using the spinhaler normal subjects) and hyperresponsive subjects and was
there were clear differences in plasma concentration therefore independent of bronchoconstriction. The sug-
time curves for three different peak inspiratory flow gested mechanism for enhanced initial absorption of
rates [37]. For example, mean inspiratory flow rates of sodium cromoglycate was an increase in permeability of
57 l min−1 and 184 l min−1 resulted in an approximate the bronchial epithelium due to histamine. This in turns

© 1996 Blackwell Science Ltd British Journal of Clinical Pharmacology 42, 697–705
 Pharmacokinetics of inhaled drugs 703

suggests that the pharmacokinetics of inhaled sodium attempted to evaluate pharmacokinetics for inhaled
cromoglycate may be altered by inflammatory mediators delivery of aminoglycoside antibiotics. The polarity of
present at the site of drug absorption from the airways, gentamicin is such that it is poorly absorbed from the
although the effect appears to be relatively short-lived gastrointestinal tract. Hence plasma levels of gentamicin
and not clinically relevant. The plasma pharmacokinetics following inhalation will directly reflect absorption from
of cromoglycate have also been investigated in tandem the lungs. However, as with cromoglycate, a high degree
with its effects on exercise induced bronchoconstriction, of hydrophilicity may result in a greater propensity for
with a ceiling in response being observed at inhaled absorption across the endothelium of the alveolar
doses which resulted in plasma levels of about 4 ng ml−1 vascular bed compared with the bronchial mucosa. In
[ 44]. This illustrates the point that the plasma concen- one study, serum levels of gentamicin 1 h post-dosing
tration is not directly related to the local dose- ranged from 3.0 to 12.0 mg ml−1 after intramuscular
response effect. administration and 1.3 to 6.8 mg ml−1 after intratracheal
The pharmacokinetics of inhaled nedocromil sodium administration, both given in a daily dose of 240 mg
are similar to those of sodium cromoglycate in that it [50]. In a study of four patients, 40 mg of nebulised
exhibits two absorption components consistent with a gentamicin achieved mean levels of 22.2 mg ml−1 in
‘flip-flop’ model [ 45]. As is the case with cromoglycate, tracheal aspirate and 0.2 mg ml−1 in serum [ 51],
the terminal half-life of nedocromil represents the although no details of nebuliser apparatus or sampling
absorption half-life, with absorption from the lungs times are available. Thirty minutes after intratracheal
becoming rate limiting. After inhalation of 4 mg dose administration, mean peak serum concentration of
by a pressurised metered dose inhaler, differences were gentamicin was 1.04 mg ml−1, with levels in bronchial
observed between normal volunteers and asthmatic secretions at 4 h (43 mg ml−1) exceeding minimum
patients in terms of a prolonged t and lower values bacterial inhibitory concentrations. Serum gentamicin
max
for C and AUC in the asthmatics [45 ]. The calculated levels have also been measured 1 h after an 80 mg dose
max
bioavailability for inhaled nedocromil was 9.2% of the given via a nebuliser, via oral aerosolisation or via a
nominal dose in normals vs 5.7% in asthmatics. As with tracheotomy tube, both with intermittent positive press-
cromoglycate absorption of an oral dose is less than ure breathing [ 52]. Serum gentamicin levels were 0.1 to
2%. With inhalation of both cromoglycate and nedocro- 0.94 mg ml−1 after tracheal aerosolisation and 0.1 to
mil, forced expiration and deep inspiration results in 0.16 mg ml−1 after normal inhalation. Moreover, 1 h
enhanced drug absorption [46,47]. It has also been after nasotracheal administration of gentamicin levels
found that significant increases in plasma nedocromil ranged from 1.2 to 4.0 mg ml−1. Serum concentration of
concentration occurs following exercise but not after gentamicin, were evaluated 1 h after intramuscular
valsalva manoeuvres or hyperventilation, suggesting (1.5 mg kg−1), intratracheal (40 mg) and oral nebulised
that enhanced bioavailability occurs as a consequence (40 mg) administration to children with cystic fibrosis
of an increase in lung volume with exercise [48]. In a [53]. Mean serum concentration of gentamicin was
comparison of a intravenous and inhaled nedocromil much lower after aerosolised (0.20 mg ml−1) than intra-
during AMP challenge no relationship was found tracheal (0.53 mg ml−1) routes. Both aerosolised and
between either C or AUC and protection against intratracheal routes gave adequate bronchial levels
max
AMP bronchoconstriction [49]. Since inhaled but not above minimum bacterial inhibitory concentration.
systemic administration of nedocromil produced protec- Nebulised delivery of tobramycin in a single-dose of
tion, there is no direct relationship between plasma 120 mg achieved sputum levels in excess of 100 mg ml−1
levels and degree of protection afforded against in patients with cystic fibrosis [54 ].
challenge. Taken together, these studies suggest that therapeutic
nebulised doses of gentamicin achieve adequate levels
in the bronchial tree with only minimal systemic
absorption. It is unclear whether it is possible to
Other inhaled drugs evaluate lung delivery of gentamicin with different
nebulisers using pharmacokinetic methods, since this
The nebulised route of delivery now has an established would require a proper concentration-time profile to
role for delivering antibiotics to the lung in patients quantify accurately lung absorption.
with the chronic bronchial sepsis as well as to delivery Delivery of nebulised morphine to the lung has been
pentamidine prophylaxis for patients with HIV infection. suggested as a way of avoiding hepatic first-pass
The type of nebuliser is important in determining the metabolism and an alternative method for achieving
size of particles and preferred site of distribution for rapid analgesia compared with the oral or prarenteral
delivery of the drug. For example, in delivering nebulised route of administration. It has also been suggested that
pentamidine it is more preferable to produce peripheral it may relieve breathlessness in patients with associated
deposition to the alveolar compartment of the lung. It bronchial carcinoma by a direct action on airway
would therefore seem appropriate in these situations to nociceptive receptors. In a recent study the pharmaco-
apply the pharmacokinetic technique for evaluating kinetics of morphine were compared when given by the
drug delivery in order to compare different nebuliser intravenous, oral and nebulised routes of administration
systems, as has successfully been done with nebulised to 10 healthy subjects [55]. Nebulised delivery of
salbutamol [3 ]. morphine was associated with an initial rapid absorption
There have been a number of studies which have from the lung with a t of 10 min and a time profile
max
© 1996 Blackwell Science Ltd British Journal of Clinical Pharmacology 42, 697–705
704 B. J. L ipworth

similar to that of the intravenous route. The systemic salbutamol metered dose inhalers. T horax 1996; 51:
bioavailability of morphine calculated from the AUC 325–326.
was 5% of the nebulised dose and 24% of the oral dose. 5 Clark DJ, Lipworth BJ. Lung bioavailability of chlorofluo-
The calculated value for bioavailability of the nebulised rocarbon free, dry-powder and chloroflorocarbon contain-
ing formulations of salbutamol. Br J Clin Pharmacol 1996;
dose was similar to that from radiolabelled aerosol
41: 247–249.
studies. However it is conceivable with more efficient
6 Barry PW, O’Callaghan C. The use of the chlorofluorocar-
nebulisers which boost inspiratory phase delivery, or bon free salbutamol preparation, Airomir, with different
with dosimetry, that the bioavailability would be spacer devices. T horax 1995; 50(Suppl 2 ): A78.
significantly enhanced. Whether the nebulised route 7 Clark DJ, Lipworth BJ. Effect of multiple actuations,
provides more rapid analgesia compared with the oral delayed inhalation and antistatic treatment on the lung
route requires confirmation from clinical studies. bioavailability of salbutamol via a spacer device. T horax
1996; (in press).
8 Barry PW, O’Callaghan C. Multiple actuations of salbuta-
mol MDI into a spacer device reduce the amount of drug
recovered in the respirable range. Eur Respir J 1994; 71:
Conclusions 1707–1709.
9 O’Callaghan C, Lynch J, Cant M, Robertson C.
Delivery of drugs by the inhaled route results in a high Improvement in sodium cromoglycate delivery from a
therapeutic ratio with delivery of relatively low doses spacer device by use of an antistatic lining, immediate
achieving a high local concentration. The application of inhalation, and avoiding multiple actuations of the drug.
T horax 1993; 48: 603–606.
pharmacokinetic techniques allows accurate and repro-
10 O’Callaghan C, Cant M, Robertson C. Delivery of
ducible quantification of drug delivery by measuring
beclomethasone dipropionate from a spacer device: what
plasma levels to reflect lung absorption, although this is dose is available for inhalation. T horax 1994; 49: 961–964.
dependent on sufficiently sensitive assays. This technique 11 Barry PW, O’Callaghan C. The effect of delay, multiple
may be applied to compare different inhaler devices actuations and spacer static charge on the in vitro delivery
such as dry powder, pressurised aerosol and nebulised of budesonide from the nebuhaler. Br J Clin Pharmacol
formulations. It may also be used to evaluate the 1995; 40: 76–78.
bioequivalence of different drug formulations, as for 12 Borgstrom L, Newman S, Wheis ZA, Moren F. Pulmonary
example with salbutamol. Other applications include deposition of inhaled terbutaline: comparison of scanning
the pharmacokinetic measurement of drug delivery to gamma camera and urinary excretion methods. J Pharm
Sci 1992; 81: 753–755.
the lung to assess inhalation technique in patients, and
13 Borgstrom L, Derom E, Stahl E, Wahlinboll E, Pauwels
this has resulted in providing important information on
R. The inhalation device influences lung deposition and
the use of large volume spacers. Pharmacokinetic bronchodilating effect of terbutaline. Am J Respir Crit
evaluation with inhaled corticosteroids using techniques Care Med 1996; 153: 1636–1640.
such as the charcoal-block method may be used to 14 Erkstrom T, Andersson AC, Skedinger M, Lindblah C,
assess the relative gut and lung components of systemic Stahl E. Dose potency relationship of terbutaline inhaled
bioavailability. This along with information on drug via Turbuhaler or via pressurised metered dose inhaler.
elimination and distribution provides a rational expla- Ann Allergy Asthma Immunol 1995; A: 228–332.
nation for observed differences at steady-state in the 15 Hindle M, Chrystyn H. Determination of relative bioavail-
systemic bioactivity profile between inhaled corticostero- ability of salbutamol to the lung following inhalation. Br
ids. Further possible applications of the pharmacokinetic J Clin Pharmacol 1992; 34: 311–315.
16 Lipworth BJ, Newnham DM, Clark RA, Dhillon DP,
technique include the assessment of delivery of nebulised
Winter JH, McDevitt DG. Comparison of the relative
antibiotics via different nebuliser systems, as for example,
airways and systemic potencies of inhaled fenoterol and
in patients with cystic fibrosis. salbutamol in asthmatic patients. T horax 1995; 50: 54–61.
17 Engel T, Scharaling B, Skovsted B, Heinig JH. Effects,
side-effects and plasma concentrations of terbutaline in
adult asthmatics after inhaling from a dry-powder inhaler
References device at different inhalation flows and volumes. Br J Clin
Pharmacol 1992; 33: 439–444.
1 Newnham DM, Wheeldon NM, Lipworth BJ, McDevitt 18 Hindle M, Newton DAG, Chrystyn H. Investigations of
DG. A single-dosing comparison of the relative cardiac an optimal inhaler technique with the use of urinary
beta 1/beta 2 activity of inhaled fenoterol and salbutamol salbutamol excretion as a measure of relative bioavailability
in normal subjects. T horax 1993; 48: 656–658. to the lung. T horax 1993; 48: 607–610.
2 Newnham DM, McDevitt DG, Lipworth BJ. Comparison 19 Ryrfeldt A, Andersson P, Edsbacker S, Tonnesson M,
of the extrapulmonary b -adrenoceptor responses and Davis P, Pauwels R. Pharmacokinetics and metabolism of
2
pharmacokinetics of salbutamol given by standard metered budesonide, a selective glucocorticoid. Eur J Respir Dis
dose inhaler and modified actuator device. Br J Clin 1982; 63(Suppl 122): 89–95.
Pharmacol 1993; 36: 445–450. 20 Andersson P, Ryrfeldt A. Biotransformation of the topical
3 Newnham DM, Lipworth BJ. Nebuliser performance, glucocorticoids, budesonide and beclomethasone,
pharmacokinetics, airways and systemic effects of sal- 17,21-diapropionate in human liver and lung homogenate.
butamol given via a novel nebuliser delivery system J Pharm Pharmacol 1984; 36: 763–765.
( Ventstream). T horax 1994; 49: 762–770. 21 Trescoli-Serrano C, Ward MJ, Rajput R, Garcia-Zorcaco
4 Clark DJ, Gordon-Smith J, McPhate G, Clark GA, M. Does swallowing charcoal affect gastrointestinal absorp-
Lipworth BJ. Lung bioavailability of generic and innovator tion of inhaled beclamethasone. Eur Respir J 1995; 8: S304.

© 1996 Blackwell Science Ltd British Journal of Clinical Pharmacology 42, 697–705
 Pharmacokinetics of inhaled drugs 705

22 Harding SM. The human pharmacology of fluticasone Holgate ST. Effect of methacholine induced bronchocon-
propionate. Resp Med 1990; 84 ((Suppl A)): 25–29. striction on the pulmonary distribution and plasma
23 Thorsson L, Edsbacker S, Conradson TB. Lung deposition pharmacokinetics of inhaled sodium cromoglycate in
of budesonide from turbuhaler is twice that from a subjects with normal and hyperreactive airways. T horax
pressurised metered dose inhaler. Eur Respir J 1994; 7: 1988; 43: 611–616.
1839–1844. 42 Richards R, Simpson SF, Renwick AG, Holgate ST.
24 Pedersen S, Steffensen G, Ekman I, Tonnesson M, Borga Inhalation rate of sodium cromoglycate determines plasma
O. Pharmacokinetics of budesonide in children with pharmacokinetics and protection against AMP-induced
asthma. Eur J Clin Pharmacol 1987; 31: 579–582. bronchoconstriction in asthma. Eur Respir J 1988; 1:
25 Pedersen S, Steffensen G, Ohlsson SV. The influence of 896–901.
orally deposited budesonide on the systemic availability of 43 Richards R, Fowler C, Simpson S, Renwick AG, Holgate
budesonide after inhalation from a turbuhaler. Br J Clin ST. Inhaled histamine increases the rate of absorption of
Pharmacol 1993; 36: 211–214. sodium cromoglycate from the lung. Br J Clin Pharmacol
26 Agertoft L, Pedersen S. Importance of the inhalaton device 1992; 33: 337–341.
on the effect of budesonide. Arch Dis of Child 1993; 44 Patel KR, Tullett WM. Neale MG, Wall RT, Tan KM.
69: 130–133. Plasma concentrations of sodium cromoglycate given by
27 Agertoft L, Pedersen S. Effects of long-term treatment with nebulisation and metered dose inhalers in patients with
an inhaled corticosteroid on growth and pulmonary exercise-induced asthma: relationship to protective effect.
function in asthmatic children. Respir Med 1994; 88: Br J Clin Pharmacol 1986; 21: 231–233.
373–381. 45 Neale MG, Brown K, Foulds RA, Lal S, Morris DA,
28 English AF, Neate MS, Quint DJ, Sareen M. Biological Thomas D. The pharmacokinetics of nedocromil sodium a
activities of some corticosteroids used in asthma. Am new drug for the treatment of reversible obstructive airways
J Respir Crit Care Med 1994; 149 (Suppl): A212. disease in human volunteers and patients with reversible
29 Clark DJ, Grove A, Cargill RI, Lipworth BJ. Comparative obstructive airways disease. Br J Clin Pharmacol 1987;
adrenal suppression with inhaled budesonide and flutica- 24: 493–501.
sone propionate in adult asthmatic patients. T horax 1996; 46 Richards R, Fowler C, Simpson SF, Renwick AG, Holgate
51: 262–266. ST. Deep inspiration increases the absorption of inhaled
30 Clark DJ, Clark RA, Lipworth BJ. Comparative systemic sodium cromoglycate. Br J Clin Pharmacol 1989; 27:
bioactivity of inhaled budesonide and fluticasone propi- 861–865.
onate in asthmatic children. T horax 1996; (in press). 47 Summers QA, Singh S, Honeywell RG, Renwick AG,
31 Thorsson L, Dahlstrom K, Edsbacker S, Kullen N, Paulson Holgate ST. The effect of respiratory manoeuvres and
J, Wiren JE. Pharmacokinetics and systemic effects of pharmacological agents on the pharmacokinetics of nedoc-
inhaled fluticasone propionate in healthy subjects. Br romil sodium after inhalation. Br J Clin Pharmacol 1992;
J Clin Pharmacol 1996, (in press). 33: 431–438.
32 Lonnebo A, Grahnen A, Jansson B, Brundin M, 48 Ghosh SK, Neale MG, Patel KR. The effects of physiologi-
Ling-Andersson A, Eckernass SA. An assessment of the cal manoeuvres on the absorption of inhaled nedocromil
systemic effects of single and repeated doses of inhaled sodium. Br J Clin Pharmacol 1994; 37: 305–308.
fluticasone propionate and inhaled budesonide in healthy 49 Summers QA, Honeywell R, Renwick AG, Holgate ST.
volunteers. Eur J Clin Pharmacol 1996; 49: 456–463. The protective efficacy of inhaled oral and intravenous
33 Hogger P, Rohdewald P. Binding kinetics of fluticasone nedocromil sodium against adenosine-5-monophosphate
propionate to the human glucocorticoid receptor. Steroids induced bronchoconstriction in asthmatic volunteers. Pulm
1994; 59: 597–602. Pharmacol 1990; 3: 190–197.
34 Wurthwein G, Rehder S, Rohdewald P. Lipophilicity and 50 Klastersky J, Geuning C, Movawad E, Daneau D.
receptor affinity of glucocorticoids. Pharm Ztg W issl 1992; Endotracheal gentamicin in bronchial infections in patients
4: 161–167. with tracheostomy. Chest 1972; 161: 117–120.
35 Fuller RW, Collier JG. The pharmacokinetic assessment 51 Odio W, Van Laer E, Klastersky J. Concentrations of
of sodium cromogylcate. J Pharm Pharmacol 1983; 35: gentamicin in bronchial secretions after intramuscular and
289–292. endotracheal administration. J Clin Pharmacol 1975;
36 Neale MG, Brown K, Hodder RW, Auty RM. The 15: 518–524.
pharmacokinetics of sodium cromoglycate in man after 52 Lake KB, Van Dyke JJ, Roumsfeld JE. Combined topical
intravenous and inhalation administration. Br J Clin and pulmonary systemic gentamicin: the question of safety.
Pharmacol 1986; 22: 373–382. Chest 1975; 68: 62–64.
37 Richards R, Dickson CR, Renwick AG, Lewis RA, Holgate 53 Baran D, Dachy A, Klastersky J. Concentration of
ST. Absorption and disposition kinetics of cromolyn gentamicin in bronchial secretions of children with cystic
sodium and the influence of inhalation technique. fibrosis or tracheostomy. Int J Clin Pharmacol 1975;
J Pharmacol Exp T her 1987; 241: 1028–1032. 12: 336–341.
38 Moss GF, Jones KM, Ritchie JT, Cox JSG. Plasma levels 54 Mukhopadhyay S, Staddon GE, Eastman C, Palmer M,
and urinary excretion of disodium cromoglycate after Rhysdavies E, Carswell F. The quantitative distribution of
inhalation by human volunteers. T oxicol Appl Pharmacol nebulised antibiotic in the lung and cystic fibrosis. Respir
1971; 147–156. Med 1994; 88: 203–211.
39 Benson MK, Curry SH, Mills GGD’A. Hughes DTD. 55 Masood AR, Thomas SHL. Systemic absorption and
Uptake of disodium cromoglycate in obstructive airways nebulised morphine compared with oral morphine in
disease. Clin Allergy 1973; 3: 389–394. healthy subjects. Br J Clin Pharmacol 1996; 41: 250–252.
40 Auty RM, Brown K, Neale MG, Snashawl PD. Respiratory
tract deposition of sodium cromoglycate is highly depen-
(Received 2 April 1996,
dent upon technique of inhalation using the spinhaler. Br
J Dis Chest 1987; 81: 371–380. accepted 23 July 1996 )
41 Richards R, Has A, Simpson SF, Britten A, Renwick AG.

© 1996 Blackwell Science Ltd British Journal of Clinical Pharmacology 42, 697–705