8 2001 OPA (Overseas Publiskm Aarociation) N.V.

Journal nf Drug Targering.2001. Vol. 9. No.4. pp. 295 302

Published by license under
Reprints available directly from the publisher
the Hawood Academic Publishers imprint
Photocopying permitted by license only pan of The Gordan and Bnach Publishing Gmup.
Journal of Drug Targeting Downloaded from informahealthcare.com by Virginia Commonwealth University on 06/23/14

Novel Nanoparticles for Pulmonary Drug Administration

PAUL A. DICKINSONab*,S. WYN HOWELLSa and IAN W. KELLAWAYaC

'The Welsh School of Pharmacy, Card@ University, Cathays Park Card& CFI 3XE UK. bPreformulationand Biopharmaceutics, Pharma-
ceutical and Analyrical R&D. AstraZeneca, Mereside, Alderley Park, Macclesfield, Cheshire SKlO 4TG and 'The School of Phurmacy. Vni-
versity of London. 29/39 Brunswick Square, London, WClN IAY

(Received August 22, 2000; In final form January 25, 2001)

A novel one-step, low energy method, which avoids harsh processing conditions including
potentially toxic and chemically reactive cross-linking agents, for the production of
hydrophilic drug nanoparticles suitable for dispersion in the hydrofluoroalkane propellants
was investigated. Reverse-phase microemulsions were used as the template for the production
For personal use only.

of nanoparticles. Two microemulsion systems were investigated: water/sodium bis(2-ethyl-

hexyl) sulphosuccinate (A0T)liso-octane and water/ lecithinlpropan-2-oYiso-octane. Nano-
particles were captured by snap freezing with subsequent freeze-drying. Nanoparticles were
dispersed in 1.1,1,2,3.3.3-heptafluoropropane (HFA-227) and the aerosol performance of the
pressurised metered dose inhaler (pMDI) assessed by cascade impaction. Spherical nanoparti-
cles less than 300 nm in size were produced. Nanoparticles produced using AOT as the sur-
factant could not be dispersed in HFA-227. However lecithin based nanoparticles could be
dispersed in co-solvent modified HFA-227 and produced fine aerosols (Mass Median Aero-
dynamic Diameter S 1.5 pm. fine particle fraction > 58 %). This data suggests that a high
fraction of the nanoparticles would be deposited (targeted) within the lung with the deposition
being mainly alveolar. That is the ideal deposition profile for the systemic delivery of drugs
via the lungs.

Keywords: PMDI. nanoparticle, propellant, salbutamol. pulmonary deposition, ~ r o ~ ~ l

INTRODUCTION It should be noted that if the lungs are to be consid-

ered for the systemic delivery of biotechnology com-
The lungs are a promising non-invasive route for the pounds a high percentage of the dose must be
delivery of peptidergic drugs as they demonstrate a delivered to the lungs and the site of deposition
relatively high permeability to hydrophilic macromol- should be as peripheral as possible (Colthorpe et al.,
ecules and express relatively low peptidase/protease 1992). Efficient delivery and peripheral deposition is
activity (Wall, 1995). A major obstacle, however, to important to ensure high bioavailability and as a cor-
the eventual clinical pulmonary administration of bio- ollary reproducible dosing of potent biotechnology
technoloay products is the formulation of these drugs products which may possess small therapeutic indi-
into a convenient, portable and effective dosage form. ces.
* Corresponding author: Tel: +44 (0)1625 513391, Fax: +44 (0)1625 517436, E-mail: paul.dickinson@astneca.com Preformulation
and Biopharmaceutics. Pharmaceutical and Analytical R&D. AstraZeneca, Mereside, Alderley Park, Macclesfield, Cheshire SK I0 4TG.

295
 296 PAUL A. DICKINSON eta/.
The most popular device for the delivery of tradi-
tional drugs to the lungs is the pressurized metered
dose inhaler (pMDI) (Boyd, 1935) which has substan-
tial advantages in portability and robustness over
other devices. Currently most pMDI are formulated as
suspensions in which micronised drug is suspended in
the propellant. This inevitably leads to central and
oro-pharyngeal deposition as the drug particle size
Journal of Drug Targeting Downloaded from informahealthcare.com by Virginia Commonwealth University on 06/23/14
can not be reduced below approximately 2 to 3 pm.
An approach to improve the pulmonary delivery of
drugs (both traditional and biotechnologicals) would
be to produce much smaller drug particles. That is,
instead of producing micronised drug particles in the
2-3 p n range to produce nanoparticles 30-200 nm in
size. Indeed radiolabelled aerosols with aerodynamic
diameters less than 100 nm are routinely used for lung
ventilation scanning due to the high penetration and
deposition of the aerosol (Burch et al., 1986). An
approach which has similarities with this strategy has
been shown to be successful for pMD1 delivery. In
For personal use only.
this instance solution pMDIs have been formulated
which resulted in much smaller drug particles on pro-
pellant evaporation. This produced better in vitm and
in vivo aerosol performance and drug deposition
(Ashworth et al., 1991; Harnor et al., 1993; Warren
and Farr, 1995). The formulation of pMDIs as solu-
tions suffers, however, from some disadvantages.
These include the difficulty in producing a molecular
dispersion of many drugs in the propellants. If a
molecular dispersion is produced it is probable that
the chemical stability of the drug is reduced. The pro-
duction of nanoparticles which can be dispersed in the
propellants should maintain the integrity of the drug
while providing the substantial advantages of the
solution system.
The aim of the work described in this paper was to
develop a technology capable of providing efficient
pulmonary delivery of drugs from pMDIs. The
approach identified for the production of nanoparti-
cles was to use a reverse microemulsion as a template.
If the swollen micelles containing the drug f a matrix
excipient could be captured and reduced to a solid this
would produce a nanoparticle. An added attraction
was that judicious choice of surfactant for the forma-
tion of the microemulsion should also allow resuspen-
sion of the nanoparticles in the HFAs. That is the
surfactant may remain associated at the nanoparticle
surface and facilitate HFA dispersion.
MATERIALS AND METHODS
Chemicals were of analytical grade or HPLC grade as
appropriate and were used as received. Sodium
bis(2-ethylhexyl) sulphosuccinate (AOT), polyacry lic
acid (Mw = 2000) (PAA) and iso-octane were pur-
chased from Aldrich Chemical Co Ltd (Poole, UK).
Lecithin (egg, -9O%), n-hexane and propan-2-01 were
purchased from Fisher Scientific UK Ltd (Loughbor-
ough, UK). Bromothymol blue (BTB) was procured
from BDH (Poole, UK).
Salbutamol Sulphate and 1, I , 1,2,3,3,3-heptafluoro-
propane (HFA-227) were pharmaceutical grade and a
generous gift from ML Laboratories (St Albans, UK)
and Solvay Fluor (Hannover, Germany) respectively.
Preparation of microemulsions ternary phase
diagrams
Microemulsions were produced by adding filtered
distilled water to a surfactant iso-octane solution. The
mixture was gently agitated with a bench vortex for
several seconds and allowed to stand for 15 minutes
to ensure complete equilibrium of the mixture. Water
addition was successively repeated to determine the
phase boundary between the crystal clear micellar
phase (isotropic) and opaque multiphase phase of the
system (non-isotropic). Two surfactant systems were
investigated: sodium bis(2-ethylhexyl) sulphosucci-
nats (AOT) and 1ecitihin:propan-2-01( 1 :3 w/w).
Production of nanoparticles from microemulsions
The drug with or without PAA or lactose were dis-
solved in water and added to the surfactantliso-octane
mixture to give a final aqueous phase of 1-30 % w/w
(Table I). The microemulsion was then snap frozen by
submersion in liquid nitrogen. The iso-octane and
water were removed by freeze-drying at -55 OC.
 NOVEL NANOPARTICLES
TABLE I Formulae of microemulsions used for nanoparticle production
Su~acranr(%w/w)
Formulation
Iso-octane (%w/w) AOT Lecithin: Pmpan-2-01(1:3)
A 78 6 -
B 25 - 45
C 35 - 45
D 40 - 40
Journal of Drug Targeting Downloaded from informahealthcare.com by Virginia Commonwealth University on 06/23/14
a. BTB = bromothymol blue, PAA = Polyacrylic acid (M,2000).
Measurement of nanoparticle size by photon
correlation spectroscopy (PCS)
The nanoparticles were dispersed in filtered
iso-octane and sonicated for 3 min, transferred to a
quartz cuvette and sealed to prevent iso-octane evapo-
ration. The cuvette was then inserted in a Coulter N4
plus and multimodal analysis performed using the
Coulter N4 plus standard distribution processor tech-
For personal use only.
nology (SDP). Analysis was repeated for n>7.
Determination of drug release kinetics
from nanoparticles
The release rate of bromothymol blue from
BTB/PAA/AOT nanoparticles and salbutamol sul-
phate from salbutamol SulphateAecithin nanoparti-
cles at 37 "C was determined. Briefly, nanoparticles
were accurately weighed ( 2 0 4 0 mg) and 5 mL of
distilled water added, the mixture stirred and 100 pL
samples periodically removed for up to 20 minutes.
BTB and salbutamol sulphate were assayed by direct
UV-vis (6 15 nm) and HPLC-UV analysis respectively
and dissolution plots generated.
HPLC assay of salbutamol sulphate
Salbutamol sulphate concentrations were determined
using a reverse-phase (C18) column (HiChrom
S50DS2, HiChrom Ltd. UK) and UV detection
(SpectroMonitor 3200, LDC, UK) at 278 nm. Sam-
ples were introduced onto the column using a Promis
(LDC, UK) autoinjector (100 pl injection volume).
Chromatograms were stored on a computer system
297
Aqueous Phase (%w/w) /composition
16 I BTB' (0.06%w/w). PAAa( 17.7% w/w)
30 / salbutamol sulphate (17% w/v)
20 / salbutamol sulphate (17% wlv)
20 I salbutamol su1phate:lactose (70:30)(17% w/v)
and analysed using Axxiom software (Model 747,
LCD, UK). The mobile phase (flow rate = 1 mumin)
consisted of methanol - water (containing heptane
sulphonic acid 1.1013 gA) 5545 v/v, adjusted to pH
3.0 with glacial acetic acid. The flow rate was main-
tained using a Constametric 3200 pump (LDC,UK).
Samples were prepared in internal standard solution
(ethanol 600 mL, water to lo00 mL, bamethane
7 pgmL-').
Scanning electron microsccrpy (SEM)
of nanoparticles
Excess surfactant was removed by washing the nano-
particles with iso-octane using centrifugation. The
nanoparticles were dried under a nitrogen stream and
splutter layered with gold. Scanning electron micro-
graphs of the nanoparticles were taken with a Philips
XL 20 SEM.
Production of HFA nanoparticle formulations
Nanoparticles (-30 mg) were either added directly to
a plastic pMDI vial or dispersed in a suitable organic
liquid (0.5-1 g) and then added to a plastic pMDI
vial. A Bespak BK357 100 pL metering valve was
crimped in place and 14 g of HFA-227 added using a
pressure burette. The vial was sonicated for 3 Inin.
Visual evaluation of the formulation was performed
prior and subsequent to sonication and at later time
points to assess formulation stability and homogene-
ity.
 298 PAUL A. DICKINSON er at.
Assessment of aerosol performance
The aerosol performance of the nanoparticle HFA for-
mulation was assessed by cascade impaction. Briefly,
the plates of an Andersen cascade impactor were
coated with polyethylene glycol (Mw 300) to reduce
particle bounce and reentrainment. The nanoparticle
pMDI was primed by firing five shots to waste and
Journal of Drug Targeting Downloaded from informahealthcare.com by Virginia Commonwealth University on 06/23/14
then 20 actuations were introduced into the Andersen
cascade impactor, operating at 28.3 Lmin-’, via a BP
two-stage liquid impinger inlet. The actuator, inlet,
impactor stages and filter were then washed with 10
mL internal standard solution by sonication in poly-
thene bags. The washings were assayed for salbuta-
mol sulphate concentration by HPLC-UV. A
preliminary assessment of aerosol performance as a
function of actuator geometry was performed.
RESULTS AND DISCUSSION
For personal use only.
Initially an AOT based microemulsion was investi-
gated as the template for nanoparticle production.
AOT microemulsions are well documented in the lit-
erature (e.g. Novaki and El Seoud, 1998), do not
require co-surfactants for formation and are relatively
robust. They, therefore, offer an ideal microemulsion
for investigation. The poor solubility of AOT in the
HFA propellant would probably mean that any nano-
particles produced would not however be capable of
dispersion in the HFA. Also the known membrane
toxicity of AOT probably makes it unsuitable for
inclusion in pulmonary drug delivery systems. A ter-
nary phase diagram (Fig. 1) was produced to identify
the microemulsion formulae that gave the best aque-
ous to surfactant ratio.
Traditionally nanoparticles have been produced by
the use of reactive crosslinking agents and/or mono-
mers (Alltmann et al., 1998; Munshi et a!., 1995).
This approach is unattractive due to concerns about
the chemical stability of any drug (especially biotech-
nologicals) and also the potential toxicity of any
residual monomer or crosslinker remaining in the
nanoparticle. For biodegradable polymers such as
poly(1actic acid) and ploy(1actide-co-glycoldide) vari-
0
’ h a
FIGURE 1 Ternary phase diagram of AOTliso-octanelwater system
at 2OoC
ous methods have been proposed for the preparation
of nanoparticles (Fessi et al., 1989; Bodmeier and
Cohen, 1989; Alltmann et al., 1992; Murakami et al.,
1997). Nanoparticles can also be produced where the
matrix consists of the drug itself e.g. cyclosporin A
(Ford et al., 1999). We have directed our research
towards identifying a low energy universal process
which would capture the microemulsion template,
dehydrate and form the nanoparticle in 1 step. Snap
freezing by submersion in liquid nitrogen and
freeze-drying met all these criteria and produced nan-
oparticles.
Polyacrylic acid was used as the matrix (Table 1)
for 3 reasons. Firstly, the inclusion of a polymer
offered the potential of producing “generic” nanopar-
ticles into which most hydrophilic drugs could be
incorporated, secondly a polymer matrix offered
potential for controlled release, finally the production
of particles containing a higher molecular weight
hydrophilic polymer would suggest that the approach
should be applicable to biotechnologicals.
A SEM of representative BTB:PAA nanoparticles
formed using this technique is shown in Figure 2. The
nanoparticles appear spherical and approximately 250
nm in size which confirmed the photon correlation
spectroscopy sizing data (232 f 58 nm, n = 7 mean f
sd).
The nanoparticles formed from the
BTB:PAA/AOT/iso-octane microemulsion were
 NOVEL NANOPARTICLES
Journal of Drug Targeting Downloaded from informahealthcare.com by Virginia Commonwealth University on 06/23/14
FIGURE 2 Scanning electron micrograph of BTB:PAA nanoparticles (fomwlation A)
For personal use only.
4'. I . 1 . I . 1 . 1 .
0 2 0 0 Q ) 8 0 8 m 1 Q l o 1 2 0 0
Time (seoonds)
FIGURE 3 BTB release from BTB:PAA nanoparticles (formula A)
versus time (mean i sd, n=3)
somewhat larger than the swollen micelles of the
microemulsion (<45 nm). This is probably due to
some phase changes and/or aggregation occurring
during the snap-freezing step or a decrease in the den-
sity of the nanoparticle core during lyophilisation.
Nevertheless the particles were within the size range
expected to produce good aerosol characteristics
(Gonda, 1981 ).
The release of bromothymol blue from BTB:PAA
nanoparticles (formula A, Table I) is shown in
Figure 3. The nanoparticles showed some ability to
retard the release of bromothymol blue with complete
release occurring by approximately 10 min. It may
299
have been expected that with massive specific surface
area of the particles and high water solubility of BTB
and PAA, release would have been very rapid. PCS
analysis suggested that nanoparticles approximately
800 nm in size existed 24 hours after the start of the
BTB release study. These particles were probably
PAA ghosts formed by ester bond generation, which
lead to crosslinking of the PAA, during the drying
(dehydration) stage of production. This observation
together with the slower hydration that would occur if
the particles were deposited on the alveolar epithe-
1 ' lium suggests that PAA nanoparticles may have
potential as a controlled release delivery system, par-
ticularly for high molecular weight solutes.
Unsurprisingly it proved impossible to disperse the
BTB:PAA nanoparticles in the HFA propellants. Even
with the inclusion of a co-solvent for AOT
(iso-octane) up to 10 96 w/w the nanoparticles aggre-
gated and adsorbed on to the wall of the pMDI vial.
This is probably as a consequence of the very low sol-
ubility of AOT in HFA-134a and HFA-227 (Blondino
and Byron, 1998).
Having developed a method for the production of
nanoparticles a more pharmaceutically acceptable
microemulsion system was investigated. Investigating
another system also allowed some measure of the ver-
satility of the production method to be assessed. A
lecithin-based microemulsion system was utilised as
the template for nanoparticle production, primarily
because lecithin is approved for pulmonary adminis-
 300 PAUL A. DICKINSON eta!.
Journal of Drug Targeting Downloaded from informahealthcare.com by Virginia Commonwealth University on 06/23/14
FIGURE 4 Ternary phase diagram of 1ecithin:propan-2-01
(1 :3)/iso-octane/water at 20 "C
tration. Production of nanoparticles with either salb-
utamol sulphate or salbutamol sulphate and lactose
cores were investigated. Salbutamol sulphate was
For personal use only.
chosen as a model drug due to it's hydrophilicity,very
low solubility in the HFA propellants and easy assay.
Lactose was added as a matrix or diluent. Lactose is
currently used in pulmonary drug delivery and there-
fore presents no toxicity concerns. It has the potential
to act as a cryoprotectant should this approach be uti-
lised for peptides and proteins and also as a diluent
for highly potent drugs.
The ternary phase diagram for 1ecithin:propan-2-01
(1:3)liso-octane/water system is shown in Figure 4
and was referenced for the selection of the systems
used for the production of nanoparticles.
A SEM of representative salbutamol sulphate nan-
oparticles formed is shown in Figure 5, the relatively
poor resolution is related to the difficulty in manipu-
lating such small particles and the fact that the scan-
ning electron microscope is operating towards it's
lower limit. In addition there is the effect of the sur-
factant coat. Again the nanoparticles appear spherical
and generally less than I 0 0 nm in size. This was con-
firmed by the photon correlation spectroscopy sizing
data (Table 11). The data is presented as the mean size
as determined by scattering intensity, certainly if the
size by weight was reported, it would be smaller and
for formula B nanoparticle population 2 would
become insignificant. Weight distribution can be pro-
duced using Mie Theory however due to problems
associated with estimating the refraction of the nano-
particles these are not reported. Appropriate nanopar-
ticles for aerosolisation were formed regardless of the
microemulsion formulae (Table I and 11). Again the
nanoparticles are greater in size than the swollen
micelles from which they were formed, they are how-
ever smaller than the BTB:PAA nanoparticles proba-
bly due to improvements in the snap freezing process
which reduced the freezing time and presented less
opportunity for phase changes and aggregation to
occur.
The release of salbutamol sulphate from salbuta-
mol sulphate nanoparticles (formula B, Table I) is
shown in Figure6 The nanoparticles showed rela-
tively fast release of salbutamol sulphate with com-
plete release occurring by approximately 4 min. It
may have been expected that with the massive spe-
cific surface area of the particles and high water solu-
bility of salbutamol sulphate release would have been
instantaneous. The slightly delayed dissolution is
probably a consequence of the hydrophobic lecithin
layer coating the nanoparticles.
The nanoparticles produced from lecithin-based
microemulsions could not be dispersed in pure
HFA- 134a or HFA-227 to any great extent. However
if they were first dispersed in a co-solvent (n-hexane)
they formed a stable dispersion in the HFA:hexane
blend (955 w/w). The formulation appeared as a
homogeneous very fine suspension. This suggests
some flocculation of the nanoparticles within the hex-
ane/HFA blend (when dispersed in hexane an opti-
cally clear system was produced) however these flocs
must have been less than I pm in size or continually
deflocculated and reformed as no sedimentation or
creaming occurred even over several months.
As expected, actuator geometry was shown to
affect aerosol performance with small orifice diame-
ters producing higher fine particle fractions. The aero-
sol performance of hexane 5 % w/w, HFA-227 95 I
w/w formulation was measured. For the investigation
a 0.25 mm orifice diameter was utilised. The fine par-
ticle fraction, mass median aerodynamic diameter
(MMAD)and geometric standard deviations (GSD)
of the aerosol produced are shown in Table 111.
 NOVEL NANOPARTICLES 30 1
Journal of Drug Targeting Downloaded from informahealthcare.com by Virginia Commonwealth University on 06/23/14

FIGURE 5 Scanning electron micrograph of salbutamol sulphate nanoparticles (formulation B)

The aerosol performance of the nanoparticle pMDI

+-I
was very good with a high fine particle fraction and
T low MMAD. This data suggests that a high fraction of
the nanoparticles would be deposited within the lung
with the deposition being mainly alveolzr. That is the
ideal deposition profile for the systemic delivery of
drugs via the lungs. This deposition compares favour-
ably to that of a model solution system with the same
actuator, solids load and hexane concentration which
produced an aerosol with a fine particle fraction of 88
For personal use only.

f 8 %, MMAD of 1,14 f 0.03 pm and GSD of 2.12 f

@ loo 400 (60 yo 1wo 1100 0.05 pm (mean f sd, n=3). This represents the best
Tine (semds) aerosol that can be produced. The increased fine parti-
cle fraction is as a consequence of reduced inlet depo-
FIGURE 6 Salbutamol sulphate release from salbutamol sulphate
nanoparticles (formula B. Table I ) versus time (mean i sd, n=3) sition. Inlet deposition is probably greater for the
nanoparticle systems due to the surfactant retarding
TABLE I1 Size of salbutamol sulphate i lactose nanoparticles
propellant evaporation although the possibility of the
produced from various lecithin based microemulsions (mean i sd, co-localisation of more than 1 nanoparticle in a pro-
n=7) pellant aerosol droplet or aggregated nanoparticles
Formulation Nonopartick? size by PCS cannot be excluded.
(see Table I ) (peak intensity)
B Population I (50.4 %): 216 i 43 nm
Population 2 (49.6 %): 40 i I5 nm
C 3 4 i 17nm CONCLSIONS
D 69 i 39 nm
In conclusion we have developed a novel low-energy,
TABLE 111 Fine particle fractions, mass median aerodynamic one step process that is capable of producing nanopar-
diameter (MMAD) and geometric standard deviation (GSD)of ticles of pure drug or drug encapsulated in a polymer
aerosols generated by salbutamol sulphate i lactose
nanoparticle/hexane/HFA/HFA-227pMDl (mean i sd. n = 5 from 3 or non-polymer matrix. For the first time, the process
batches of nanoparticles each) avoids harsh processing conditions including poten-
Formulation Fine Particle Fraction MMAD GsD (m) tially toxic and chemically reactive cross-linking
(see Table I ) (exdevice, d . 8 pm)(%) (pm) agents. The process is capable of producing nanopar-
B 58.3 i 6.8 1.2i0.4 2 . 3 i 1.1 ticles from physicochemically diverse compounds
C 65.5 f 5.1 1.3 i 0.2 2.3 i 0.3 and different microemulsions systems. The process
D 59.0 i 4.6 1.5i0.6 1 . 9 i 0 . 8 should be applicable to the production of peptide and
 302 PAUL A. DICKINSON er a/.
protein nanoparticles. Further, the nanoparticles once
produced would be capable of dispersion in the HFA
propellants and the pMDI systems created, produced
aerosols which suggest extensive and peripheral pul-
monary deposition.
Acknowledgements
The authors are grateful to the Engineering and Phys-
Journal of Drug Targeting Downloaded from informahealthcare.com by Virginia Commonwealth University on 06/23/14
ical Sciences Research Council who supported this
work through the ROPA scheme (Ref: GR/L 72961).
Parts of this work are described in patent application
P.A. Dickinson, S.W.Howells and I.W. Kellaway
(2000) Particulate Composition. UK Patent Applica-
tion 0009773.3 with priority.date 19 April 2000.
References
Allemann, E., Gurny, R and Doelker. E. (1992) Preparation of
aqueous polymeric nanodispersions by a reverse salting-out
process. influence of process parameters on particle size. Inr.
J. Pharm. 87: 247-253,
Allemann, E., Leroux, 1.-C. and Gunny. R. (1998) Polymeric nano-
For personal use only.
and micropa&les for the oral delivery of peptides and peptid-
omimetics.Adv. Drug Def.Rev. 34:171-189.
Ashworth. H.L., Wilson, C.G.. Sims, E.E., Wotton, P.K. and Hardy,
J.G. (1991) Delivery of propellant soluble drug from a
metered dose inhaler. Thorax 46.245-247.
Blondino, F.E. and Byron, P.R (1998) Surfactant dissolution and
water solubilization in chlorine-free liquefied propellants.
Drug Develop. Ind. Pharm. 24: 935-945.
Bodmeier, R and Cohen. H. (1989) Evaluation of biodegradable
poly(lactide) pellets prepared by direct compression. J.
Pharm. Sci. 78: 819-822.
Boyd, G. (1995) The continued need for metered dose inhalers J.
Aemsol Med. 8: S9-S I I.
Burch, W.M., Sullivan, P.J. and McLaren. C.J. (1986) Technegas -
A new ventilation agent for lung scanning. Nuclear Med.
Comm. 7 :865-871.
Colthorpe. P.. Farr, S.J.. Taylor, G.. Smith, I.J. and Wyatt, D.
(1992). The pharmacokinetics of pulmonary-delivered insulin:
A comparison of intratracheal and aerosol administration to
the rabbit. Pharm. Res. 9 764-768.
Fessi. H.. Puisieux, F., Devissaguet, J.P., Ammoury, N.and Benita,
S. (1989) Nanocapsule formation by interfacial polymer depo-
sition following solvent displacement. Inr. J. Pharm. 55:
RI-R4.
Ford. J., Woolfe, J. and Florence. A.T. (1999) Nanospheres of
cyclosporin A: poor oral absorption in dogs. Inr. J. Pharm.
183: 3-6.
Gonda, 1. (1981) A semiempirical model of aerosol deposition in
the human respiratory tract for mouth inhalation. J. Phann.
Phannacol. 33: 692-696.
Harnor, K.J.. Perkins. A.C., Wastie. M..Wilson, C.G., Sims. E.E.,
Feely, L.C. and Farr, S.J. (1993). Effect of vapour pressure on
the deposition pattern from solution phase metered dose inhal-
ers. h r . J. Pharm. 95: 11 1-1 16.
Murakami. H.. Kawashima. Y.,Niwa, T., Hino, T.. Takeuchi, H.
and Kobayashi, M. (1997) Influence of the degrees of hydro-
lyzation and polymerization of poly(vinylalcoho1)on the prep-
aration and properties of poly(D1-lactide-co-glycolide)
nanoparticle. lnt. J. Pharm. 149: 43-449.
Novaki. L.P. and El Seoud, O.A. (1998) A fourier transform infm-
red study on the StNCtUre of water solubilized by reverse
aggregates of sodium and magnesium bis(2-ethylhexyl) sulfo-
succinates in organic solvents. J. Colloid Inreflace Sci. 202:
391-398.
Wall, D.A. (1995) Pulmonary absorption of peptides and proteins.
Drug Del. 2: 1-20.
Warren. S.J. and Pam. S.J. (1995) Formulation of solution metered
. dose inhalers and comparison with aerosols emitted from con-
ventional suspension systems. Inr. J. Phann. 124: 195-203.