-
KONA No.23 (2005) 109
INTRODUCTION
Dry powder inhalers (DPIs) represent a significantadvance in
pulmonary drug delivery, mainly by over-coming patient related
issues of co-ordination withconventional pressurised metered dose
inhaler sys-tems. The f luidisation, de-aggregation and
dispersionof a dry powder formulation are achieved via thepatients
inspiratory action. Arguably more so than
P. Begat and R. Price1
Pharmaceutical Technology Research Group Department of Pharmacy
& Pharmacology University of Bath*H. Harris, D.A.V. Morton and
J.N. StaniforthVectura Ltd.**
The Influence of Force Control Agents on the Cohesive-Adhesive
Balance in Dry Powder Inhaler Formulations
Abstract
The aim of the study was to investigate the specific inf luence
of force control agents (FCAs)(leucine, lecithin and magnesium
stearate) on the interfacial properties of a salbutamol
sulphate-lactose dry powder inhaler formulation. The inf luence of
FCAs on the cohesive and adhesiveforce balance was directly
assessed via an atomic force microscopy (AFM) colloid probe
technique,with a recently developed cohesive-adhesive balance (CAB)
graphical analysis procedure. Co-processing of constituent
particles was conducted by a novel dry mechanical fusion
method(Mechanofusion). The in vitro deposition profile of the model
salbutamol sulphate formulationswas investigated using a Monohaler
DPI device with a next generation impactor (NGI) apparatus.The
CAB-graph analysis of a salbutamol sulphate-lactose binary system
suggested a predispositionfor an interactive mixture. However, the
reduced intermixing coefficient (Fdrug-lactose/Fdrug-drug)
sug-gested that a significant amount of energy would be required to
overcome the strong adhesive inter-action for ef ficient dispersion
of the drug from a lactose surface. The processing of lactose
withleucine, lecithin or magnesium stearate, prior to formulating
with the drug, significantly reducedthe adhesive interactions of
the salbutamol with modified lactose samples. The CAB analyses
indi-cated that the reduced intermixing coefficients shifted to
such an extent that cohesive drug interac-tions dominated. These
dramatic shifts in the balance of forces were shown to lead to poor
blendhomogeneity and potential for significant segregation between
drug and carrier particles. Con-versely, the conditioning of
salbutamol sulphate with leucine, lecithin and magnesium
stearate,which modified both the adhesive and cohesive
interactions, formed homogenous interactive blendswith
advantageously weaker drug-lactose interactions. Formulations with
pre-conditioned drug, incontrast to conditioned lactose, of fered
the best drug delivery performances. The use of the colloidAFM
technique in combination with the cohesive-adhesive balance (CAB)
approach provided a veryaccurate means of predicting dry powder
formulation behaviour and the specific inf luence of partic-ulate
interactions on aerosol performance.
Key words: Inter-particulate, AFM, Magnesium stearate, DPI,
Inhalation, Aerosols, Mechanofusion, Nanotech-nology
Accepted: August 12, 2005* Bath, BA2 7AY, UK** 1 Prospect West,
Chippenham, SN14 6FH, UK1 Corresponding author
TEL: +44 (0) 1225 383644FAX: +44 (0) 1225 386114E-mail:
[email protected]
-
other drug delivery platforms, the characteristic prop-erties of
the dry powder formulation are criticallyimportant to the effective
performance of a DPI sys-tem1, 2).
The formulation should exhibit good f low proper-ties to aid
metering, f luidisation and avoid excessivedevice retention.
Meanwhile, the f luidised powdermust disperse into a fine aerosol
(5 m) for effi-cient drug delivery3, 4). This leads to the
well-knownparadox that respirable sized particles tend to behighly
cohesive, which causes entrainment problemsdue to their poor f
lowability and limits the dispersibil-ity into an aerosol cloud5,
6). In addition, strong cohe-sion forces hinder the handling of the
powder duringmanufacture.
To overcome the highly cohesive nature of res-pirable powders,
the drug is commonly co-processed(blended) with larger carrier
particles of an inertexcipient to aid f lowability and drug
re-dispersion.This carrier based formulation approach is limited
bythe restrictive availability of excipient materials. Theonly
widely approved excipients for use as carriersare lactose and
glucose. Thus, the development of adry powder formulation is a
highly specialised, com-plex and unpredictable operation. A
formulation istypically required to go through several iterative
andoptimisation steps before a product specification canbe
achieved, and even then variability over time andbetween batches is
common.
By blending a micronised drug with a carrier, theshear forces
generated may be sufficient to overcomethe cohesive (drug-drug)
interactions in forming aninteractive mixture. Drug particles need
to be suf-ficiently attracted to the carrier during mixing
tosupport blend homogeneity, device filling and formu-lation
stability. Yet the active ingredient must be read-ily detached from
the carrier upon activation to forma fine particle cloud. Thus, the
balance of inter-partic-ulate forces within the carrier-based
formulation iscritically important.
Numerous techniques have been applied to modifyparticulate
interactions in dry powder formulations.The majority have targeted
the physical propertiesof the carrier. These include modifying the
shape7),size8), or surface features such as rugosity 9-12) of
theexcipient. Other methods involve the manufacture ofmore uniform
respirable drug particulates by particleengineering technologies
such as spray drying13) orsupercritical f luid precipitation14).
One of the mostsimple and popular advances is via the addition of
aternary agent, such as fine particles of lactose. Forthese complex
blends, it is proposed that the ternary
agent occupies high energy sites on the carrier parti-cles, such
as clefts and areas of increased moleculardisorder15, 16). As a
consequence, only low energysites remain available for drug-carrier
adhesive inter-action. The possibility of a marked reduction in
parti-cle adhesion would facilitate more effective drugdetachment
upon device actuation. Extensive workon the use of fines in dry
powder formulations andtheir inf luence on delivery performances
have beenreported17-19).
In addition to passifying active sites, researchershave shown
that the addition of fine ternary particlesmay lead to the
formation of fine particle multiplets ormetastable agglomerates20).
The critical formation ofagglomerate particles, which remain
adhered to thecoarse carrier lactose during processing and
handling,may dramatically reduce the inspirational energy
re-quirements in elutriating and de-aggregating drugparticles upon
aerosolisation.
The co-processing of carrier particles with low sur-face free
energy materials has also been reported as apossible means of
increasing the aerosolisation effi-ciencies of dry powder inhaler
formulations19, 21, 22).The primary role of these materials is to
modify theinterfacial properties of the excipient particles
todecrease drug-excipient adhesion. These force con-trol agents
(FCAs) preferably exhibit anti-adherentand/or anti-friction
properties. Typical FCAs includeamino acids such as leucine,
phospholipids such aslecithin or fatty acid derivatives such as
magnesiumstearate (MgST)23).
To optimise the efficiency of a carrier based for-mulation, the
force control agent must be specifi-cally introduced into the dry
powder formulation toselectively target the particle interactions
to be modi-fied. In this study, the FCA was mechanically fusedvia a
highly intensive co-processing system termedMechanofusion to ensure
a nanometre thick coatingof the specific components of the
formulation24). Thisapproach was recently developed by Staniforth
andMorton for inhalation powders25). In contrast to otherlow
energetic mixing or even intensive mixing, thisdry coating process
is designed to provide a relativelycomplete ultra thin coating onto
the host particles viathe application of high shear forces (Fig.
1). Suc-cinctly, a Mechanofusion mixer is composed of alarge rotor
with rounded blades revolving in a steelvessel at very high speed
(typically of the order of5000rpm). The gap size between the rotor
bladesand the vessel wall can be adjusted in order to varythe
mixing energy delivered to the powder blend. Asa result, the
particles experience very high shear
110 KONA No.23 (2005)
-
forces as they are compressed between the innerdrum wall and the
rotor.
The aim of this study was to investigate the specificinf luence
of force control agents (leucine, lecithin andmagnesium stearate)
on the interfacial properties of asalbutamol sulphate-lactose dry
powder inhaler for-mulation. The inf luence of the FCAs on the
cohesiveand adhesive force balance was directly assessed viaan AFM
colloid probe technique, with a recentlydeveloped CAB-graph
analysis procedure. This novelprocedure allows quantification of
cohesive-adhesivebalances (CAB) in a dry powder formulation26),
and,thus, can be directly utilised to highlight their
specificaffect on formulation behaviour and delivery
charac-teristics27). The in vitro deposition profile of the
modelsalbutamol sulphate formulations was investigated toelucidate
any correlation between the cohesive andadhesive nature of the
modified formulations withtheir aerosol delivery performance.
MATERIALS AND METHODS
MaterialsMicronised salbutamol sulphate, donated by Vectura
Ltd., and Sorbalac 400 lactose (Meggle, Wasserburg,Germany) were
used as supplied. The use of Sorbalac400 lactose particles (10m)
with respect to moreconventional carrier sizes (63-90m) was
dictatedby the need to minimise the potential inf luence oflarger
carrier particles over f luidisation and de-aggre-gation processes
of particle agglomerates. L-Leucinewas supplied from Ajinomoto Co.
(batch number601FK72, Tokyo, Japan), lecithin from Lipoid
GmbH(batch number 25661113-1/14, Ludwigshafen, Ger-many) and
magnesium stearate from Avocado(batch number H1028A, Heysham, UK).
All materialswere used as supplied. Ultra pure water was producedby
reverse osmosis (MilliQ, Millipore, Molsheim,France).
Preparation of powder formulationsPowder mixing was achieved in
two successive
steps involving different energetic processes. Pre-blends of
salbutamol sulphate and FCA (5% w/w) orlactose and FCA (5% w/w)
were prepared using aMechanofusion system. (Hosokawa-Alpine,
Augsburg,Germany). Powders to be processed were sealed intothe
Mechanofusion system core. Cold-water circula-tion was applied
using an incorporated water jacketto dissipate localised heating.
Samples were mixedat 5000rpm for 10 minutes to achieve the
requiredprocess intensity and mechanically fuse the FCA tothe host
particles.
The formulations were subsequently prepared bygeometrically
mixing 1g of pre-blend and 1g of eitherlactose or drug depending on
the nature of the pre-blend in 100mg increments via a Whirlimixer
(FisonsScientific Apparatus, Loughborough, UK). The result-ing
mixture was further mixed in a Turbula (GlenCreston Ltd.,
Middlesex, UK) at 46rpm for 30 min-utes. This blend design was not
intended to ref lectany commercial available or relevant DPI powder
for-mulation. This formulation was selected solely to suitthe
objectives of the study of the cohesive-adhesivebalance between
drug and lactose components.
Preparation of compressed powder substratesModel surfaces of the
powder formulations were
prepared by high-pressure compression (TA HDITexture analyser,
Stable Micro Systems, Surrey, UK).Approximately 250mg of material
was weighed into a10mm stainless steel die and compacted over
3min,with a load of 500kg.
Scanning electron microscopyThe morphology of the various powder
formula-
tions of was investigated using a scanning electronmicroscope
(SEM) (Jeol 6310, Jeol, Tokyo, Japan).Samples were gold-coated
(Edwards Sputter Coater,Crawley, UK) prior to imaging.
Force measurements by atomic force microscopy(AFM)
Prior to force measurements, salbutamol sulphate,lactose and the
corresponding conditioned particles(n3 for each material) were
fixed onto standardV-shaped tipless cantilevers (DNP-020, Digital
Instru-ments, CA, USA) using an epoxy resin glue (Araldite,UK). The
spring constant (k) of the cantilevers wasdetermined by the thermal
noise method (k0.2820.039N/m)28, 29).
The AFM was housed in an environmental chamber
KONA No.23 (2005) 111
Fig. 1 Schematic representation of Mechanofusion particle
mix-ing mechanisms.
Simple mixing Intensive mixing Mechanofusion
-
to maintain constant temperature of 25C (0.2C)and relative
humidity of 35% RH (3%). The partialwater vapour pressure was
controlled via a custom-built perfusion unit coupled to a highly
sensitivehumidity sensor (Rotronic AG, CH). The interactionforces
were measured by recording the def lectionof the AFM cantilever as
a function of the substratedisplacement (z) by applying Hookes law
(Fkz).Individual force curves (n4096) were conductedover a 10m10m
at a scan rate of 4Hz and a com-pressive loading of 10nN. These
parameters werekept constant throughout the study.
Cohesive-adhesive balance (CAB) graphs The wealth of information
from AFM measure-
ments of the interparticulate forces were analysedusing a
recently developed cohesive-adhesive balanceprocedure. Detailed
information regarding the CABgraphical analysis is described
elsewhere26). Brief ly,the construction of a CAB-graph requires a
set ofprobes (n3) and well-defined substrates of eachrespective
material to investigate all possible interac-tions (drug-drug,
drug-excipient and excipient-excipi-ent).
The adhesive force measurements between drugand excipient are
plotted on the X-axis; the relatedcohesive forces of the respective
materials are plottedon the Y-axis. The relative position of the
alignedplots with respect to the bisector indicates an affinityfor
the probe material to develop adhesive interac-tions (below the
bisector.) or a dominancy of cohesiveproperties (above the
bisector).
To express the affinity of the drug (material 1) tointeract with
the carrier (material 2), the reducedintermixing coefficient (12)
was introduced. The 12corresponds to the ratio of the adhesive
interactions(F12) and cohesive interactions (F11) of two
interactingmaterials and can be directly calculated from
theslope:
12 (1)
The position of 12 with respect to unity is a directindication
of the predisposition (121) or the reluc-tance (121) for the drug
particles to blend withanother material.
For direct visualisation of the inf luence of the addi-tion of
the FCA on the interfacial behaviour of drugand excipient
interactions, the CAB graphs for the vir-gin and treated surfaces
have been superimposed.The graph in the foreground corresponds to
the inter-actions observed between the micronised drug ()
1k12
F12ad
F11co
and lactose () probes and the virgin substratesurfaces of the
drug and excipient (Fig. 1). Thebackground graph corresponds to the
interactionmeasurements when a ternary agent was processedeither
with the drug () or the lactose ().
Content uniformity measurements The content uniformity of the
salbutamol sulphate-
lactose blends was measured by analysing the quan-tity of active
in 10mg0.5mg samples (10). Relativestandard deviation between
samples was calculated toassess the homogeneity of the different
blends. Drugcontent was analysed by UV-spectrometry
(CECILinstruments, CE7200, Cambridge, UK). Salbutamolsulphate was
analysed using a 0.06M NaOH solventand a UV detection wavelength
set at 295nm.
In vitro aerosol deposition studiesApproximately 10mg of the
carrier formulations
was accurately weighed into a gelatine capsule to beloaded into
a Monohaler device (Miat SpA, Milan,Italy). In vitro deposition
investigations were per-formed using a next generation impactor
(NGI)(Copley Scientific, Nottingham, UK). The loaded de-vice was
connected to the throat of the NGI via amoulded mouthpiece. In
vitro analysis was performedupon each actuation of the device (n3).
Testing wasperformed at 60L.min1 f low rate with a 5
secondexposure. Each NGI plate was rinsed with solventand the
subsequent solution was collected in a 50mlvolumetric f lask.
Statistical analysis of the data wasperformed using one-way ANOVA.
The levels of sig-nificance are indicated in the legend of the
respectivegraphs.
RESULTS & DISCUSSIONS
Previous studies have emphasized the relativestrength of the
adhesive salbutamol sulphate-lactoseadhesive forces with respect to
salbutamol sulphatecohesive forces26). The initial part of this
study wasto investigate possible variations in formulation
behav-iour upon modifying the cohesive and adhesive
bonds,quantified by AFM measurements, via the introduc-tion of a
force control agent. This was achieved byfirst conditioning the
lactose with various force con-trol agents using a Mechanofusion
system, prior tomixing with the drug.
The CAB-graph obtained for a salbutamol sulphate-lactose binary
system, without the presence of a FCA,is shown in Fig. 2. The
relative position of the databelow the bisecting line indicated a
stronger affinity
112 KONA No.23 (2005)
-
between salbutamol sulphate and lactose than theircohesive
forces. This suggested a predisposition foran ordered blend.
However, the quantitative measure-ment of the relative strength of
the cohesive-adhesiveratio indicated that the adhesive salbutamol
sulphate-lactose interaction was approximately six timesgreater
(126.25) than the cohesive salbutamol sul-phate bond. It should be
stressed that previousassessments performed with model crystal
substratesrevealed a salbutamol sulphate-lactose reduced
inter-mixing coefficient of 16.8827). This disparity may
beexplained by the inevitable increase in surface rough-ness by
using compressed powder substrates in con-trast to smooth
crystalline substrates. This wouldhave a considerable effect on
both van der Waalsforces and capillary forces30, 31). Nevertheless,
bothstudies revealed a consequent adhesively led system,suggesting
that a significant amount of energy wouldbe required to overcome
the adhesive interaction forefficient dispersion of the drug from a
lactose sur-face.
Thus, the introduction of a FCA was intended toadvantageously
lower the adhesive interactions be-tween drug and excipient to
facilitate the detachmentof the drug particles from the carrier
upon aerosolisa-tion provided that an adhesive-led system is
main-tained.
1. Carrier-based formulations with conditionedlactose
The CAB-graphs obtained for the interaction ofsalbutamol
sulphate probes and conditioned lactoseprobes with leucine,
lecithin and MgST are shownin Fig. 3A, 3B and 3C, respectively. As
expected,
KONA No.23 (2005) 113
Fadhesion (nN)
F coh
esio
n (nN
)
0
250
200
150
100
50
050 100 150 200 250
Salbutamol tipsLactose tipsFcohesionFadhesion
y0.4951xR20.9444
y0.16xR20.7928
Fig. 2 CAB-graph of a salbutamol sulphate lactose binary
system(with the use of compressed tablet substrates).
200 200
150
100
50
0150
10050
0
150
100
50
0
y0.16xR20.79
y0.86xR20.58
A
B
C
y1.04xR20.95
y0.50xR20.94
F coh
esio
n (nN
)
F coh
esio
n (nN
)
Fadhesion (nN)
Fadhesion
(nN)
0 50100 150
200
Salbutamol tipsLactose-leucine tips
FcohesionFadhesionSalbutamol tipsLactose tips
200 200
150
100
50
0150
10050
0
150
100
50
0
y0.16xR20.79
y0.50xR20.94
F coh
esio
n (nN
)
F coh
esio
n (nN
)
Fadhesion (nN)
Fadhesion
(nN)
0 50100 150
200
Salbutamol tipsLactose-lecithin tips
FcohesionFadhesionSalbutamol tipsLactose tips
y0.87xR20.61
y1.13xR20.92
200 200
150
100
50
0150
10050
0
150
100
50
0
y0.16xR20.79
y0.50xR20.94
F coh
esio
n (nN
)
F coh
esio
n (nN
)
Fadhesion (nN)
Fadhesion
(nN)
0 50100 150
200
Salbutamol tipsLactose-MgST tips
FcohesionFadhesionSalbutamol tipsLactose tips
y0.90xR20.99
y1.65xR20.88
Fig. 3 Inf luence of the coating of lactose with a force
controlagent on the cohesive-adhesive balances in a
salbutamolsulphate-lactose system.
-
the addition of FCAs significantly modified
thesalbutamol-lactose interactions. In all cases, the intro-duction
of the ternary agent significantly reduced theadhesive interactions
of the salbutamol probe withthe various modified lactose
substrates. However,particle adhesion decreased to such an extent
thatthe reduced intermixing coefficient (Fdrug-lactose/Fdrug-drug),
calculated from the gradient of the CAB plots,was below 1 (Table
1). This shift moved the CABsystem to one synonymous of a
cohesive-led system.The conditioning of lactose with MgST resulted
in thelowest intermixing coefficient value (120.61) whilethe
addition of leucine and lecithin reduced the inter-mixing
coefficient to 0.96 and 0.88, respectively.Thus, the
pre-conditioning of lactose particles withleucine, lecithin or MgST
transformed a systemwhich was dominated by the adhesive
drug-lactoseforces into a cohesive system. Such lowering of
theinteractions between drug and excipient via the intro-duction of
the FCAs may possibly lead to an unstableformulation, subjected to
undesirable segregation.
Consequently, to highlight the inf luence of themodifications of
the intermixing coefficient on theblending characteristics via the
introduction ofFCAs, scanning electron microscopy and drug con-tent
uniformity analyses of the blends were investi-gated.
Representative SEM images of formulations ofsalbutamol sulphate
mixed with lactose-leucine, lac-tose-lecithin and lactose-MgST
conditioned particlesare shown in Fig. 4A, 4B and 4C, respectively.
Asanticipated from the intermixing coefficient measure-ments,
scanning electron micrographs highlighted ahigh degree of drug
segregation resulting from intro-duction of the FCAs. A very
limited adhesive interac-tion was apparent between agglomerated
salbutamolsulphate particles and conditioned lactose-leucine
(Fig. 4A). However, large quantities of drug particleswere
present as loose agglomerates. This segregationwas even more
pronounced for lecithin (Fig. 4B) andMgST (Fig. 4C) conditioned
lactose particles. Virtu-
114 KONA No.23 (2005)
Conditioning MixingReduced intermixing coefficient (12)
Content uniformity
Mechanofusion, 10min @ 5200rpm Turbula, 30min @ 46rpm (%
RSD)
Salbutamol sulphate (non conditioned) Sorbalac 400 6.25
04.20
Sorbalac 400 Leucine Salbutamol sulphate 0.96 08.92
Sorbalac 400 Lecithin Salbutamol sulphate 0.88 09.31
Sorbalac 400 MgST Salbutamol sulphate 0.61 15.51
Salbutamol sulphate Leucine Sorbalac 400 1.89 02.92
Salbutamol sulphate Lecithin Sorbalac 400 2.13 03.00
Salbutamol sulphate MgST Sorbalac 400 1.52 03.62
Table 1 Mixing sequences of salbutamol, lactose, force control
agent formulations and resulting reduced intermixing coefficients
and contentuniformities.
Fig. 4 Representative scanning electron micrographs of
ternarymixtures of salbutamol sulphate and lactose pre-con-ditioned
with leucine (A), lecithin (B) or magnesiumstearate (C).
-
ally no interaction was observed between drug andthe conditioned
lactose surfaces, which resulted inthe formation of large drug
agglomerates. Interest-ingly, the mechanofused Sorbalac 400 lactose
parti-cles appeared to be smoother after pre-conditioningwith
leucine, lecithin or magnesium stearate, com-pared to the as
supplied Sorbalac27). This suggested asmooth continuous coating of
the FCAs over the sur-face of the lactose particles.
As anticipated, dose content uniformity measure-ments revealed
an increase in the relative standarddeviation of salbutamol
sulphate of each blend. TheRSD of 4.2% for micronised salbutamol
sulphatemixed with lactose increased to 8.92% with the
condi-tioning of lactose with leucine, 9.31% for
lactose-lecithinand 15.51% for lactose-MgST. These observations
sug-gested a good correlation between the reduction ofthe
intermixing coefficients, which were all signifi-cantly below 1,
and the content uniformity measure-ments of salbutamol sulphate in
the correspondingcarrier-based formulations.
The de-agglomeration and dispersion behaviour ofthe salbutamol
sulphate particles from the model car-rier based formulations are
shown in Fig. 5. Theemitted dose of the salbutamol sulphate-lactose
for-mulation via the low resistance Monohaler devicewas quite high
(76.57%). However, a significant per-centage of the drug was
recovered in the throat andthe first stage of the NGI apparatus.
These results
were in accordance with a previous in vitro studyconducted with
a Rotahaler and Turbuhaler DPIdevices2). This study suggested that
the observeddeposition pattern was due to the limited detachmentof
the drug from the carrier upon actuation caused bythe highly
adhesive salbutamol sulphate-lactose inter-actions.
The mechanical fusion of leucine with lactoseresulted in a
similar drug emission efficiency to theconventional blend. However,
the amount of salbutamolsulphate recovered in the first stage of
the NGI signif-icantly increased with respect to the non coated
lac-tose blend. The coating of lactose with either lecithinor MgST
slightly reduced the device retention of theactive ingredient from
23.43% to 18.73% and 18.99%,respectively. Although the interaction
between drugand coated lactose surfaces were significantly
de-creased, a large amount of drug was still recoveredon the upper
stage of the in vitro apparatus.
These results were consistent with the assessmentfrom the
CAB-graphs. The CAB analysis indicatedthat the energy of
interaction between the drug andcoated lactose would reduce to such
an extent thatthe adhesively led salbutamol sulphate-lactose
formu-lation would shift to an unfavourable cohesive system.The
aerosolisation performances of the modifiedlactose carrier-based
formulations may have not im-proved since the dramatic shift in the
force balancewas shown to lead to poor blend homogeneity and
the
KONA No.23 (2005) 115
Device Throat Stage 1 Stage 2 Stage 3 Stage 4 Stage 5 Stage 6
Stage 7 Stage 8
40
30
20
10
0
Salbutamol
sulphatelactoseSalbutamol(lactose-leucine)Salbutamol(lactose-lecithin)Salbutamol(lactose-MgST)
Perc
enta
ge o
f nom
inal
dos
e
**
**
**
*
* *
Fig. 5 In vitro deposition of salbutamol sulphate carrier-based
formulations with conditioned lactose (meanS.D., n3). * p0.05, **
p0.01, *** p0.001: significant difference compared to without force
control agent by ANOVA one-way.
-
potential for significant segregation between drugand carrier
particles.
2. Carrier-based formulations with conditioneddrug
The conditioning of excipient particles with a FCAmodified the
adhesive interaction between drug-car-rier and excipient-excipient
interactions. In contrast,the processing of the micronised drug
particles withthe FCA would alter both the cohesive (drug-drug)and
adhesive (drug-lactose) interactions. To furtherinvestigate the
possibility of selectively modifyingboth these interactions within
a carrier based formu-lation, the drug was mechanofused with the
FCAs.
The CAB-graphs obtained for the interaction be-tween conditioned
salbutamol sulphate with leucine,lecithin or MgST and lactose are
shown in Fig. 6A,6B and 6C, respectively. The balance remained
adhe-sive for all three systems, although the drug-lactoseforces
decreased by more than a half of its originalvalue. The
conditioning of salbutamol sulphate withleucine, lecithin and MgST
led to an intermixingcoefficient of 1.89, 2.13 and 1.52,
respectively. Thesemixtures would therefore be expected to form
homoge-nous interactive blends with advantageously weakdrug-lactose
interactions. As expected, the lactoseforce balance transformed
from an adhesive to a cohe-sive system. Nevertheless, it can be
speculated thatthis shift should not greatly affect the
formulationproperties as this change of behaviour is predomi-nately
due to a decrease of the adhesive (drug-lactose)forces and not in
an increase of the lactose cohesivebonds.
Representative scanning electron micrographs ofthe lactose
particles blended with conditionedsalbutamol sulphate-leucine,
salbutamol sulphate-lecithin, and salbutamol sulphate-MgST are
shown inFig. 7A, 7B and 7C, respectively. In contrast to theternary
mixture of drug and conditioned lactoseshown in Fig. 4, the
conditioned drug particlesstrongly interacted with the lactose
particles for allthree FCAs. This suggested an effective adhesive
dis-position, in agreement with the CAB data analyses.The
corresponding content uniformity measurementsof the ternary
mixtures ref lected an adhesive led sys-tem with low relative
standard deviations for lactosemixed with salbutamol
sulphate-leucine (2.92%),salbutamol sulphate-lecithin (3.00%) and
salbutamolsulphate-MgST (3.62%) conditioned particles.
These observations suggested good correlation be-tween the
reduced intermixing coefficients and thecharacteristics of the
respective carrier-based formu-
116 KONA No.23 (2005)
200 200
150
100
50
0150
10050
0
150
100
50
0
y0.16xR20.79
y0.53xR20.76
y2.07xR20.91
y0.50xR20.94
F coh
esio
n (nN
)
F coh
esio
n (nN
)
Fadhesion (nN)
Fadhesion
(nN)
0 50100 150
200
Salbutamol tipsLactose tips
FcohesionFadhesionSalbutamol-leucine tipsLactose tips
200 200
150
100
50
0150
10050
0
150
100
50
0
y0.16xR20.79
y0.47xR20.92
y1.30xR20.95
y0.50xR20.94
F coh
esio
n (nN
)
F coh
esio
n (nN
)
Fadhesion (nN)
Fadhesion
(nN)
0 50100 150
200
Salbutamol tipsLactose tips
FcohesionFadhesionSalbutamol-lecithin tipsLactose tips
200 200
150
100
50
0150
10050
0
150
100
50
0
y0.16xR20.79
y0.66xR20.98
y2.58xR20.54
y0.50xR20.94
F coh
esio
n (nN
)
F coh
esio
n (nN
)Fadhesion
(nN)
Fadhesion
(nN)
0 50100 150
200
Salbutamol tipsLactose tips
FcohesionFadhesionSalbutamol-MgST tipsLactose tips
A
B
C
Fig. 6 Inf luence of the coating of salbutamol sulphate with
aforce control agent on the cohesive-adhesive balances in
asalbutamol sulphate-lactose system.
-
lations. Such formulations would be expected to bestable during
handling and storage, and may lead to agreater de-agglomeration and
dispersion efficiency ofthe respirable particles.
The in vitro deposition profile of conditionedsalbutamol
sulphate carrier-based formulations areshown in Fig. 8. The
Mechanofusion of the forcecontrol agents to the salbutamol sulphate
particlesresulted in a significant decrease in device retentionfrom
23.42% for the FCA free formulation to 12.51%with leucine, 14.52%
with lecithin and 8.23% withMgST. These results suggested a
lubrication effect ofthe FCA and subsequent reduction in
interactionbetween the powder bed and the capsule, while
theformulation conserved its metastability as suggestedby the CAB
analyses. More dramatic was the signifi-cant decrease of the
percentage of drug deposited onthe first stage (cut off diameter
8.06m) of the NGIapparatus. Stage 1 deposition decreased from
22.89%to 6.16% for the conditioning of salbutamol sulphatewith
leucine, 9.20% for salbutamol sulphate-lecithinand 7.99% for
salbutamol sulphate-MgST. This indi-cated a greater
de-agglomeration efficiency of thecoated salbutamol sulphate
particles in the carrierbased formulations. This was further
highlighted bythe increase deposition of the active ingredient
inthe lower stages of the in vitro apparatus. Thesedata clearly
indicated that the characteristic proper-ties of carrier based
formulations can be controllably
KONA No.23 (2005) 117
Fig. 7 Representative scanning electron micrographs of
ternarymixtures of lactose and salbutamol sulphate pre-con-ditioned
with leucine (A), lecithin (B) or magnesiumstearate (C).
Device Throat Stage 1 Stage 2 Stage 3 Stage 4 Stage 5 Stage 6
Stage 7 Stage 8
40
30
20
10
0
Salbutamol
sulphatelactose(Salbutamol-leucine)lactose(Salbutamol-lecithin)lactose(Salbutamol-MgST)lactose
Perc
enta
ge o
f nom
inal
dos
e
**
**
***
***
*** *
**
*** *
**
***
***
***
***
***
****
**
**
**
**
*
*
*
*
*
*
Fig. 8 In vitro deposition of salbutamol sulphate carrier-based
formulations with conditioned drug (meanS.D., n3).* p0.05, **
p0.01, *** p0.001: significant difference compared to without force
control agent by ANOVA one-way.
-
enhanced by judicious selection of the interparticu-late
interactions to be modified by the introduction ofthe FCAs.
A summary of the device retention, fine particlefraction (%
respirable particle of the emitted dose)and total fine particle
fraction (% respirable particlefrom total recovered dose) of the
salbutamol sulphatecarrier-based formulations is shown in Fig. 9. A
clearpattern of formulation performance was observeddepending on
whether the force control agent wasfused either with the drug or
the lactose. Formula-tions with pre-conditioned drug, in contrast
to con-ditioned lactose, offered the best drug
deliveryperformances. It is suggested that the conservation ofan
adhesive system for the pre-conditioned drug par-ticles directly
led to the increased de-aggregation per-formance. Meanwhile, the
selective decrease of thedrug-lactose interfacial interaction for
conditionedlactose particles led to a dominant cohesive (drug-drug)
system, which resulted in poor blend homo-geneity and poor f
luidisation. The highest %FPF ofthe emitted dose was obtained for
formulations withleucine and lecithin coated salbutamol sulphate
parti-cles (73.72% and 71.87%, respectively). The low drugretention
of salbutamol sulphate-MgST conditionedparticles (8.23% device
retention) contributed to deliveran equivalent total fine particle
dose as for leucineand lecithin (64.43% for leucine, 61.41% for
lecithinand 63.79% for MgST).
CONCLUSIONS
The inf luence of force control agents on the proper-ties and
performances of model salbutamol sulphatecarrier based formulations
was investigated. Thecohesive and adhesive dependencies were
controlledby conditioning either the drug or the carrier
beforemixing in order to create selective modifications ofthe
inter-particulate interactions within a dry powderformulation. The
conditioning of these fine inhalationpowders was conducted via a
Mechanofusion system.This new technique was intended to enable
effectiveparticle covering with a nano-scale coating, a
processwhich is difficult to achieve via conventional
ap-proaches.
The colloid probe AFM technique together with thenovel
cohesive-adhesive balance (CAB) analysis pro-cedure was utilised to
measure the variations in inter-particulate forces of binary and
ternary blends. TheCAB-graph method successfully predicted
substantialmodifications in the behaviour of the
formulations,dependant on whether the FCAs were conditionedwith the
drug or the lactose. This novel approach ofapplying FCA to the drug
is in contrast to previouswork in this area where force control
agents have tra-ditionally been applied to carrier particles.
This work emphasized that the CAB analysismethod can be utilised
for pre-formulation studiesand in the design of new formulation
systems for drypowder inhalers. The work also confirmed the
poten-
118 KONA No.23 (2005)
100
80
60
40
20
0
Perc
enta
ge o
f nom
inal
dos
e
***
***
***
***
***
***
***
*
**
***
***
*
*
Salbutamol
sulphate-lactoseSalbutamol-(lactose-leucine)Salbutamol-(lactose-lecithin)Salbutamol-(lactose-MgST)(Salbutamol-leucine)-lactose(Salbutamol-lecithin)-lactose(Salbutamol-MgST)-lactose
Device retention (%) Fine particle Fraction (%) Total Fine
particle (%)
Fig. 9 Fine particle fraction and emission efficiency of
salbutamol sulphate carrier-based formulations (meanS.D., n3). *
p0.05, ** p0.01, *** p0.001: significant difference compared to
without force control agent by ANOVA one-way.
-
tial value of the use of force control agents togetherwith the
Mechanofusion process in DPI formulations.
REFERENCES
1) D. Ganderton and N. M. Kassem. Advances in Pharma-ceutical
sciences., Academic Press, London, 1992.
2) A. R. Clark. Medical Aerosol Inhalers Past, Present,and
Future. Aerosol Science and Technology 22: 374-391(1995).
3) D. Ganderton and T. Jones. Drug Delivery to the Respira-tory
Tract., Camelot Press, Southampton, 1987.
4) W. C. Hinds. Aerosol technology: Properties, Behaviourand
measurements of airborne particles, Wiley, NewYork, 1999.
5) J. C. Feeley, P. York, B. S. Sumby, and H. Dicks.
Deter-mination of surface properties and f low characteristicsof
salbutamol sulphate, before and after micronisation.International
Journal of Pharmaceutics 172: 89-96(1998).
6) X. M. Zeng, G. P. Martin, and C. Marriott.
ParticulateInteractions in Dry Powder Formulations for
Inhalation,Taylor & Francis, London, 2001.
7) X. M. Zeng, G. P. Martin, C. Marriott, and J. Pritchard.The
inf luence of crystallization conditions on the mor-phology of
lactose intended for use as a carrier for drypowder aerosols.
Journal of Pharmacy and Pharmacol-ogy 52: 633-643 (2000).
8) X. M. Zeng, G. P. Martin, S. K. Tee, A. Abu Ghoush, andC.
Marriott. Effects of particle size and adding se-quence of fine
lactose on the deposition of salbutamolsulphate from a dry powder
formulation. InternationalJournal of Pharmaceutics 182: 133-144
(1999).
9) R. Price, P. M. Young, S. Edge, and J. N. Staniforth. Theinf
luence of relative humidity on particulate interac-tions in
carrier-based dry powder inhaler formulations.International Journal
of Pharmaceutics 246: 47-59(2002).
10) M. D. Louey and P. J. Stewart. Particle interactions
in-volved in aerosol dispersion of ternary interactive mix-tures.
Pharmaceutical Research 19: 1524-1531 (2002).
11) M. D. Louey, P. Mulvaney, and P. J. Stewart.
Characteri-sation of adhesional properties of lactose carriers
usingatomic force microscopy. Journal of Pharmaceutical
andBiomedical Analysis 25: 559-567 (2001).
12) Y. Kawashima, T. Serigano, T. Hino, H. Yamamoto, andH.
Takeuchi. Effect of surface morphology of carrierlactose on dry
powder inhalation property of pranlukasthydrate. International
Journal of Pharmaceutics 172:179-188 (1998).
13) L. Yu. Amorphous pharmaceutical solids:
preparation,characterization and stabilization. Advanced Drug
Deliv-ery Reviews 48: 27-42 (2001).
14) P. York. Strategies for particle design using supercriticalf
luid technology. Pharmaceutical Science and Technol-ogy Today 2:
430-440 (1999).
15) X. M. Zeng, K. H. Pandhal, and G. P. Martin. The inf lu-
ence of lactose carrier on the content homogeneity
anddispersibility of beclomethasone dipropionate from drypowder
aerosols. International Journal of Pharmaceutics197: 41-52
(2000).
16) N. Islam, P. J. Stewart, I. Larson, and P. G. Hartley.
Lac-tose surface modification by decantation: Are drug-finelactose
ratios the key to better dispersion of salmeterolxinofate from
lactose interactive mixtures? Pharmaceu-tical Research 21: 492-499
(2004).
17) S. K. Tee, C. Marriott, X. M. Zeng, and G. P. Martin.The use
of different sugars as fine and coarse carriersfor aerosolised
salbutamol sulphate. International Jour-nal of Pharmaceutics 208:
111-123 (2000).
18) S. K. Tee, G. P. Martin, A. R. Leeds, C. Walker, A. Kicman,
D. A. Cowan, and C. Marriott. The inf luenceof a tertiary component
on the in vivo disposition ofsalbutamol isomers aerosolised from a
dry powderinhaler formulation. Thorax 56: P51 (2001).
19) X. M. Zeng, G. P. Martin, S. K. Tee, and C. Marriott.The
role of fine particle lactose on the dispersion anddeaggregation of
salbutamol sulphate in an air stream invitro. International Journal
of Pharmaceutics 176: 99-110 (1998).
20) P. Lucas, K. Anderson, and J. N. Staniforth.
Proteindeposition from dry powder inhalers: Fine particlemultiplets
as performance modifiers. PharmaceuticalResearch 182: 562-569
(1998).
21) M. J. Clarke, M. J. Tobyn, and J. N. Staniforth. The
for-mulation of powder inhalation systems containing ahigh mass of
nedocromil sodium trihydrate. Journal ofPharmaceutical Sciences 90:
213-223 (2001).
22) J. N. Staniforth. Improvement in dry powder
inhalerperformance: surface passivation effects. Proceedings ofDrug
Delivery to the Lungs VII 86-89 (1996).
23) P. Begat, M. Green, D. A. V. Morton, A. Whittock, and J.N.
Staniforth. PowderHale: A Novel High PerformanceDry Powder Inhaler
Formulation Technology for Tar-geted and Systemic Drug Delivery.
Drug Delivery to theLung XII 119 (2001).
24) R. Pfeffer, R. N. Dave, D. G. Wei, and M.
Ramlakhan.Synthesis of engineered particulates with tailored
prop-erties using dry particle coating. Powder Technology117: 40-67
(2001).
25) J. N. Staniforth and D. A. V. Morton. Powder Technol-ogy
Research Leading to Improvements in InhalerProducts. Powder Science
and Engineering 34: 60-64(2002).
26) P. Begat, D. A. V. Morton, J. N. Staniforth, and R.
Price.The cohesive-adhesive balances in dry powder
inhalerformulations I: Direct quantification by atomic
forcemicroscopy. Pharmaceutical Research 21: 1591-1597(2004).
27) P. Begat, D. A. V. Morton, J. N. Staniforth, and R.
Price.The cohesive-adhesive balances in dry powder
inhalerformulations II: Inf luence on fine particle delivery
char-acteristics. Pharmaceutical Research 21: 1826-1833(2004).
28) J. L. Hutter and J. Bechhoefer. Calibration of Atomic-
KONA No.23 (2005) 119
-
Force Microscope Tips. Review of Scientific Instruments64:
1868-1873 (1993).
29) T. J. Senden and W. A. Ducker. Experimental-Determi-nation
of Spring Constants in Atomic-Force Microscopy.Langmuir 10:
1003-1004 (1994).
30) R. Price, M. Tobyn, and J. N. Staniforth. Variation
inParticle Adhesion Due to Capillary and Electrostatic
Forces. Respiratory Drug Delivery VII (2000).31) J. C. Hooton,
C. S. German, S. Allen, M. C. Davies, C. J.
Roberts, S. J. B. Tendler, and P. M. Williams. An atomicforce
microscopy study of the effect of nanoscale con-tact geometry and
surface chemistry on the adhesion ofpharmaceutical particles.
Pharmaceutical Research 21:953-961 (2004).
120 KONA No.23 (2005)
Authors short biography
Philippe Begat
Dr Philippe Begat is currently an Inhalation Project leader at
Pfizer Inc. Philippehas an MSc in Chemistry. He joined Vectura Ltd
in 2001, working on the optimiza-tion of PowderHaleTM technology.
In 2002, he joined the pharmaceutical surfacescience research group
at the University of Bath, as an Experimental Officer, todevelop
novel methods for characterizing and improving dry powder
formulations.He was awarded his PhD in 2005. Philippe has since
joined Pfizer Inc. to managethe development of new drug products
for their dry powder inhaler programmes.
Robert Price
Dr Robert Price is a senior lecturer at the Department of
Pharmacy and Pharma-cology at the University of Bath. He gained a
BSc in Physics and a PhD in PhysicalChemistry from Cardiff
University. He leads the pharmaceutical surface scienceresearch
group, investigating physico-chemical properties governing
inter-particu-late interactions, surface stability issues of
processes particles and the general areaof particle engineering and
crystal growth. He has published a series of originalresearch
articles in the areas of surface electrochemistry, crystal growth,
atomicforce microscopy and pharmaceutical technology.
Haggis Harris
Haggis is currently a postgraduate student at Bath University in
the Department ofPharmacy and Pharmacology investigating
inter-particle interactions in dry pow-der inhaler
formulations.Prior to studying at Bath University, Haggis was
employed at Vectura for four yearswhere his primary focus was
research and development of high intensity blendingtechniques to be
used in dry powder inhaler formulation.
David A.V. Morton
Dr David Morton is currently Head of Intellectual Property and
Technology atVectura Group plc. He gained a PhD from Bristol
University in Structural Chem-istry. In 1997, David joined the
Centre for Drug Formulation Studies at theUniversity of Bath to
manage their dry powder inhaler product development pro-grammes. In
1999, this group spun out into the drug delivery company
Vectura,and David was appointed Head of Pulmonary Research, where
he co-developed theemerging PowderHaleTM technology. He is also
known in the inhalation field forhis lead role in organising the
Aerosol Society Drug Delivery to the Lung confer-ence series.
-
KONA No.23 (2005) 121
Authors short biography
John N. Staniforth
Professor John Staniforth is Chief Scientific Officer and a
Founder of VecturaGroup plc. He has a BSc in Pharmacy and a PhD in
Pharmaceutical Technology.He became Professor of Pharmaceutical
Technology and Head of Pharmaceutics atthe University of Bath,
Department of Pharmacy. In 1999, he led the spin-out of twogroups
based at the University to establish Vectura. John is a registered
Pharma-cist and a Chartered Chemist. He has been elected a Fellow
of a number of interna-tional scientific societies, including the
American Association of PharmaceuticalScientists and is a Member of
the Royal Pharmaceutical Society of Great Britain.
