International Journal of Pharmaceutics 469 (2014) 132–139

Capreomycin inhalable powders prepared with an innovative spray-drying technique

Aurélie Schoubben, Stefano Giovagnoli, Maria Cristina Tiralti, Paolo Blasi *,Maurizio RicciDipartimento di Scienze Farmaceutiche, Università degli Studi di Perugia, Via del Liceo 1, Perugia 06123, Italy

A R T I C L E I N F O

Article history:Received 8 March 2014Received in revised form 14 April 2014Accepted 16 April 2014Available online 18 April 2014

Keywords:InhalationDry powder inhalerCapreomycinTuberculosisLung administration

A B S T R A C T

The aim of the work was to produce inhalable capreomycin powders using a novel spray-dryingtechnology. A 23 factorial design was used to individuate the best working conditions. The maximumdesirability was identified at the smallest mean volume diameter (dv) and span, and the highest yield.Powders were characterized for size, morphology, flowability and aerodynamic properties. Mathematicalmodels showed a good predictivity with biases lower than 20%. The maximum conformity withdesirability criteria was obtained spraying a 10 mg/mL bacitracin solution at 111 �C with the 4 mm poresize membrane. By processing capreomycin sulfate with the parameters optimized for bacitracin, aninhalable powder was obtained (i.e., yield of 82%, dv of 3.83 mm, and span of 1.04). By furtheroptimization, capreomycin sulfate powder characteristics were improved (i.e., yield, �71%; dv, 3.25 mm;span, 0.95). After formulation with lactose, emitted dose and respirable fraction of 87% and �27% wereobtained, respectively. Two capreomycin sulfate powders with suitable properties for inhalation wereproduced using the nano spray-dryer B-90.

ã 2014 Elsevier B.V. All rights reserved.

Contents lists available at ScienceDirect

International Journal of Pharmaceutics

journa l home page : www.e l sev ier .com/ loca te / i jpharm

1. Introduction1

Drug pulmonary administration has become an appealingapproach to treat local and systemic diseases (Laube et al., 2011;Forbes et al., 2011; Patton and Byron, 2007). Among the differentdevices available to aerosolize a drug into the lungs, metered-doseinhalers are leaders on the market but dry powder inhalers (DPI)are easier to use, do not need a spacer and are more efficient indeep lung deposition (Smith et al., 2010; Geller, 2005). For thisreason, a number of studies nowadays are focused on theformulation of dry powders for inhalation (Islam and Cleary,2012; Islam and Gladki, 2008).

Particles, to show the best performances and to have areproducible bioavailability in inhalation therapy, must deposit in

* Corresponding author at: University of Perugia, Dipartimento di ScienzeFarmaceutiche, Via del Liceo 1, 06123 Perugia. Italy. Tel.: +39 755852057;fax: +39 755855123.

E-mail addresses: aurelie.schoubben@unipg.it (A. Schoubben),stefano.giovagnoli@unipg.it (S. Giovagnoli), maria.tiralti@unipg.it (M.C. Tiralti),paolo.blasi@unipg.it (P. Blasi), maurizio.ricci@unipg.it (M. Ricci).

1dv, mean volume diameter; CS, capreomycin sulfate; BAC, bacitracin; SEM,

scanning electron microscopy; DPI, dry powder inhaler; TSGI, twin-stage glass

http://dx.doi.org/10.1016/j.ijpharm.2014.04.0420378-5173/ã 2014 Elsevier B.V. All rights reserved.

the deep lung where large surface area and high vascularisation willgrant a high and fast systemic absorption. Similar consideration canbe done if macrophage targeting is wanted like in the case ofpulmonary tuberculosis (Blasi et al., 2009; Stegemann et al., 2013).

Particle size is one of the main factors affecting the site ofdeposition in the lungs. Particles that have an aerodynamicdiameter (dae) between 5 and 1 mm are able to reach therespiratory region of the lungs (Washington et al., 2001). Particleswith a mean diameter between 1 mm and 100 nm barely deposit inthe airways because of high exhalation (Washington et al., 2001).Particles lower than 100 nm are able to deposit in the respiratoryregion but production and aerosolization become an issue (highcohesiveness due to large surface-to-volume ratio). Powders withparticles having a dae in the range of 5–1 mm are preferred forinhalation. Particle size distribution, shape (Yu et al., 2011; Sandlerand Wilson, 2010), surface roughness (Tong and Chow, 2006), andrelative humidity (Yadav and Lohani, 2013) are other factorsaffecting aerosolization and lung deposition.

Micronization and spray-drying are the techniques mainlyemployed for the production of dry powders for inhalation(Schoubben et al., 2010; Saleem and Smyth, 2010). A new spray-drying technology has been developed for the production ofparticles in the micron and submicron range (Li et al., 2010;

A. Schoubben et al. / International Journal of Pharmaceutics 469 (2014) 132–139 133

Schafroth et al., 2012). One of the innovations is represented by thespraying technology that uses a piezoelectric actuator incorporat-ing a thin perforated stainless steel membrane (4.0, 5, or 7.0 mmpore size are available). An ultrasonic frequency (60 kHz) causesmembrane vibration and droplet generation in the micron/submicron range (Li et al., 2010; Heng et al., 2011). Optimizingthe operating settings it may be possible to produce particles withsize suitable for inhalation therapy (Lee et al., 2013; Bürki et al.,2011). A long drying chamber and an electrostatic system forparticle collection guarantee high yields with small samples.

The aim of the work was to individuate the best workingconditions to produce inhalable capreomycin sulfate (CS), a secondline antitubercular drug under clinical investigation for thetreatment of pulmonary tuberculosis by inhalation. Bacitracin(BAC), a peptide characterized by a similar cyclic structure and highwater solubility, was used as model compound to establish thesmallest working frame for CS. The influence of peptide concentra-tion, drying temperature, and membrane pore size on the powdercharacteristics was investigated using a 23 factorial design. Theoptimized CS powders were characterized for their aerodynamicbehavior in presence of different lactoses for inhalation.

2. Materials and methods

2.1. Materials

CS from Streptomyces capreolus and BAC from Bacillus lichen-iformis were purchased from Sigma–Aldrich Chemical (Milan,Italy). Acetonitrile was provided by Panreac Quimica S.A.U.(Castellar del Vallès, Spain). Lactohale1 LH200 (LH200) andLH100 (LH100) were kindly provided by DOMO1-Pharma (Goch,Germany), while FlowLac1 100 (FLL100), InhaLac1 70 (INL70), andInhaLac1 230 (INL230) were gifts of MEGGLE1-Pharma (Wasser-burg, Germany). Hydroxypropyl-methylcellulose capsules(Vcaps1) were a gift of Capsugel, Peapack (NJ, USA). Ultrapurewater was produced by a Human Power 1 ultrafiltration systemfrom Human Corporation (Caserta, Italy). Other chemicals andsolvents were of analytical grade and were used without furtherpurification.

2.2. Design of experiment

A 23 factorial design was used to evaluate the effects of thespray-drying process on the final particle characteristics. BAC andCS aqueous solutions were spray-dried with a nano spray-dryerB-90 (Büchi Italia Srl.). The instrument parameters are stainlesssteel membrane pore size, drying temperature, drying chamberpressure and spray rate. The process is also influenced by soluteconcentration of the nebulized solution. In this study, the dryingchamber pressure and the spray rate were maintained constant at50 mbar and 100% (corresponding to the maximum throughput)(Büchi, 2014), respectively. The parameters investigated were:membrane pore size (A), drying temperature (B) and solutionconcentration (C) (Table 1). To assess process and systemreproducibility, 2 replicates were performed. The dependentvariables considered were yield, mean volume diameter (dv),

Table 1Factors and factor level limits studied to build the model.

Factors Minimum value Maximum value

A Membrane pore size (mm) 4 7B Drying temperature (�C) 100 120C Solution concentration (mg/mL) 1 10

and distribution width, expressed as span. The process yield wascalculated according to Eq. (1).

Yield ð%Þ ¼ MassFMassI

� 100 (1)

where MassF is the amount of powder recovered after spray-dryingand MassI the starting amount of powder solubilized. The dv wasmeasured with a laser granulometer (Section 2.5.1) and the spanwas calculated using the following Eq. (2).

span ¼ dð0:9Þ � dð0:1Þdð0:5Þ (2)

where d(0.9), d(0.1), and d(0.5) represent the diameters lower thanor equal to those measured at the 90%, 10%, and 50% of the particlesize distribution.

The complete general third order polynomial regression modelchosen is described by the following Eq. (3):

Y ¼ b0þb1XA þ b2XB þ b3Xc þ b4XAXB þ b5XAXC þ b6XBXCþ b7XAXBXC þ e (3)

where Y is the dependent variable, A, B, and C are the three chosenfactors (Table 1), b1 is the regression coefficient related to the maineffects (b0–b3), the two-factor (b4–b6), and three-factor (b7)interactions, while e is the residual error of the model.

The regression model was applied to the different powders andevaluated in terms of statistical significance and lack of fit byANOVA. Statistical and factorial analyses were performed by usingDesign-Expert1 v. 8.0.1 (Stat-Ease Inc., Minneapolis, MN, USA).

2.3. Check-point analysis

Check-point analyses were carried out to establish modelreliability and predictivity. Four points were chosen arbitrarily andall experiments were performed in triplicate. Bias was estimated toevaluate the agreement between predicted and actual values.

2.4. Response surface methodology

A response surface methodology was used to plot responsebehaviour versus factor levels. The contour plots obtained for yield,dv, and span were investigated to frame the region of the workingspace satisfying the user-defined criteria. The overall desirabilityapproach (Derringer and Suich, 1980) is summarized in Eq. (4).

f ðdgÞ ¼ SMi ¼ 1wi di

SMi ¼ 1wi

(4)

where f(dg) is the overall desirability, di is the desirability of the i-thresponse, M is the number of response variables and w is the user-specified weight. This procedure was useful to draw theboundaries of the region corresponding to the highest yield andthe lowest dv and span.

2.5. Powder characterization

2.5.1. Particle size determinationAn Accusizer C770 (PSS Inc., Santa Barbara, CA, USA) was used to

determine powder mean particle size and distribution (Nicoli et al.,1992). Since both BAC and CS are soluble in water, measurementswere performed in acetonitrile using the gravity modality. Thesuspension was sonicated for 1 min before analysis. Particle sizewas expressed as dv and span (Eq. (3)).

2.5.2. Particle morphologyBAC and CS particle morphology was investigated by scanning

electron microscopy (SEM) using a Field Emission SEM (LEO 1525

134 A. Schoubben et al. / International Journal of Pharmaceutics 469 (2014) 132–139

equipped with a GEMINI column, ZEISS, Germany). Samples wereprepared depositing powders onto an aluminium specimen stubcovered with a double sided adhesive carbon disc. Samples weresputter coated with chromium prior to imaging (Quorum Q150TES East Grinstead, West Sussex, UK). Coating was done at 120 mAfor 30 s.

2.5.3. Flowability measurementTapped density (rtapped) was determined with a tap density

tester ERWEKA SMV 102 (Heusenstamm, Germany). Powders werestored in a desiccator cabinet at 20% relative humidity beforeanalysis. The powder rtapped was measured upon a 5000 tap cycleand each measurement was performed in duplicate. The rtapped

and the freely settled bulk density (rbulk) were calculated asfollow:

rtapped ¼ powder massVtapped

(5)

rbulk ¼ powder massVbulk

(6)

where Vtapped and Vbulk represent the tapped and bulk volume,respectively. The measured rtapped and rbulk values were used tocalculate the theoretical flowability of the powders quantified bythe Hausner’s ratio (H) (Hausner, 1967; Liu et al., 2008) (Eq. (7)).

H ¼ rtapped

rbulk(7)

2.6. Aerodynamic assessment

Aerodynamic characteristics were evaluated using a twin-stageglass impinger (TSGI) (Disa, Milano, Italy) and lactose as carrier.The two stages of the TSGI were loaded with 7 mL (stage 1) and30 mL (stage 2) of 0.1 M phosphate buffer at pH 7.4. Powderblending and capsule filling were done manually. After theintroduction and the perforation of the capsule into the Handi-Haler1, the pump was set at an aspiration rate of 60 � 5 L/min andturned on for 5 s following the European Pharmacopoeia (appara-tus A) guidelines (Council of Europe, 2008). Aerodynamicperformances were assessed using 3 capsules Vcaps1 number 3,different type lactose, and a 1:50 blend weight ratio between CSand lactose. Each determination was performed in triplicate.

Samples were collected and diluted with phosphate buffer andanalyzed at 268 nm by UV spectrophotometry using a UV/visAgilent 8453 Spectrophotometer (Agilent, Germany) as elsewherereported (Rossi et al., 2004). The emitted fraction (EF) and the

Fig. 1. BAC response surface plots showing the desirability for the different factor

respirable fractions expressed as percentage of the nominal dose(RFN) and of the emitted dose (RFE) were calculated using thefollowing equations.

EF% ¼ Emitted dose of CS ðstage 1 and 2Þ � 100Nominal dose

(8)

RFN% ¼ Dose of CS stage 2 � 100Nominal dose

(9)

RFE% ¼ Dose of CS stage 2 � 100Emitted dose

(10)

3. Results and discussion

3.1. BAC powder nano spray-drying

A 23 factorial design was used to investigate the effect onprocess performance of the main parameters affecting the spray-drying process (Table 1). The results obtained spray-drying BACshow a high variability for process yield (18–82%), dv(3.12–6.60 mm) and span (0.91–2.70) (tabulated results areincluded in the Supporting Information; Table 1). Three non linearregression models were elaborated for each investigated depen-dent variable and validated by ANOVA (tabulated results areincluded in the Supporting Information Table 2). All themathematical models were characterized by a high significance(p < 0.0001) and by a non significant lack-of-fit (p = 0.09, p = 0.4533,p = 0.2248) at 95% significance level (tabulated results are includedin the Supporting Information Table 2).

Fig. 1 shows that changing membrane pore size, the relativeinfluence of the other two factors changes as well. Concentrationhad a linear effect and mainly influenced yield, while dryingtemperature had a larger effect on span and dv. Switching from 4 to7 mm pore size the overall influence of the factors on the responsesdecreased. The process yield was significantly influenced by dryingtemperature (B), its interaction with membrane pore size (AB) andBAC concentration (BC), and the term A2. For dv and span, the mostinfluential factors were BAC concentration (C coeff. 1.37 and �0.13),pore size quadratic term (A2 coeff. 0.44 and �0.34), and itsinteraction with concentration (A2C coeff. �0.84 and 0.20). The 3dependent variables were significantly affected by the term A2

(tabulated results are included in the Supporting InformationTable 2) indicating a non linear influence of this factor, as it isoutlined by the curvature on the effect of A (plotted results areincluded in the Supporting Information Fig. 1). The dryingtemperature had a significant influence on the span while itsinteraction with BAC concentration (BC) affected dv.

combinations. (A) 4 mm membrane pore size, (B) 7 mm membrane pore size.

Fig. 3. Scanning electron microscopy photomicrographs and particle size distribution profiles of BAC before (A) and after (B) spray-drying.

Fig. 2. Phase contour plot over the response space showing the distribution of the 25 experiments reported in Table 1 of the Supporting Information. The color map accordingto span values is reported aside.

A. Schoubben et al. / International Journal of Pharmaceutics 469 (2014) 132–139 135

136 A. Schoubben et al. / International Journal of Pharmaceutics 469 (2014) 132–139

To confirm the regression model predictivity, four check pointswere arbitrarily chosen and the actual results were compared withthose predicted by the models (tabulated results are included in theSupporting Information Table 1). Since the models possess a goodpredictivity (bias always below 18% and in some case as low as 1%)(tabulated results are included in the Supporting InformationTable 3), they were used to navigate the working space by buildingthe respective response surfaces. Bias values generally express thediscrepancy between predicted and actual values. Commonly, biasesbelow 20% are considered acceptable. The desirability function(Eq. (4)) was employed to obtain the desirability plots combining theindividual response surfaces. Maximum desirability was intended asthe smallest dv and span, and the highest yield.

Formulations with characteristics closer to the maximumdesirability (dark blue area) were n. 6, 7, 8, 17, 19, and 24(Fig. 2). Among these, n. 17 and 19 gave the lowest dv (3.12 and3.14 mm), while n. 6 gave the highest yield (82.89%) (Fig. 2). Theprocess conditions corresponding to the maximum desirabilitywere 4 mm membrane pore size, 111 �C of drying temperature, and10 mg/mL of BAC.

SEM analysis revealed that commercial BAC powder wascharacterized by irregular particles with rough edges, a highpolydispersity, and dimensions around 5 mm (Fig. 3). Spray-driedBAC powder showed collapsed spherical particles with meandiameter around 3 mm.

3.2. CS powder nano spray-drying

CS was spray-dried using the process conditions optimized forBAC obtaining a yield of 82.4%, a dv of 3.83 mm, and span of 1.04(Powder 1).

To confirm the superimposability of the best working con-ditions for BAC and CS, a reduced design of experiment was set up.This study evaluated the area around the maximum desirabilityvalues of BAC and all the values investigated gave dv values lowerthan 3.83 mm. Fig. 4 shows that CS concentration affected more thepowder characteristics than drying temperature. Optimized

Fig. 4. CS response surface plots showing the desirability for the d

conditions were: membrane pore size, 4 mm; drying temperature,102 �C; and CS concentration, 6.3 mg/mL. In these conditions, ayield of 71.4%, a dv of 3.25 mm, and span of 0.95 were obtained(Powder 2). Fig. 5 shows the powders before and after spray-dryingcomposed by spherical particles with a smooth surface but withlower particle size after processing.

Dimensionless numbers (e.g., Peclet number) have been used togain better understanding of particle formation by spray drying. ForPeclet number <1, the solute diffusive motion is faster than the radialvelocity of the retreating droplet surface and the solutes remainhomogeneously distributed in the droplet during the evaporation.Particles do not collapse and, generally, maintain shape and sizesimilar to that of the original droplets (Vehring, 2008). Judging fromCS particle morphology (data not sown), in our experiments thePeclet numberwas always lower than 1 and size and sizedistributionwere the particle characteristics really variable.

CS dae, calculated considering perfectly spherical particles and atap density of 0.369 g/mL, was of 1.95 mm. Spray-dried CS shouldbe more cohesive than the commercial powder since H ratios(Hausner, 1967) increased from 1.0997 to 1.2957.

Considering the differences between the nano spray-dryer andconventional spray-dryer, one of the main advantage of the formeris the possibility to achieve high yields processing low amount ofmaterial and/or low volumes of feed solutions (Heng et al., 2011;Schoubben et al., 2013). Here, amounts of peptide processedranged from 20 to 200 mg while the feed solution was 20 mL. Thesetiny amounts of material are generally not suitable for processingusing conventional spray-dryer with acceptable yields(Heng et al., 2011). This peculiarity allows optimizing theparameters also with expensive materials, such as monoclonalantibody and recombinant proteins.

3.3. CS aerodynamic characterization

The EF and RFE of bulk spray-dried CS powder (Powder 2) werearound 70% and 14%, respectively. Since low EF and RFN areprobably due to high powder cohesiveness, different lactoses were

ifferent factor combinations using 4 mm membrane pore size.

Fig. 5. Scanning electron microscopy photomicrographs and particle size distribution profiles of CS before (A) and after (B) spray-drying.

Table 2EF, RFE and RFN of spray-dried CS with different lactoses.

CS/lactose EF (%) � s.d. RFN (%) � s.d. RFE (%) � s.d.

CS/LH200 87.03 � 4.72 26.50 � 1.13 30.50 � 2.33CS/LH100 82.80 � 1.27 21.10 � 0.99 25.50 � 0.85CS/INL230 83.40 � 0.85 9.40 � 1.98 11.20 � 2.26CS/INL70 77.20 � 3.96 19.00 � 0.28 24.65 � 0.92CS/FLL100 84.90 � 1.13 15.65 � 1.63 18.42 � 1.67

A. Schoubben et al. / International Journal of Pharmaceutics 469 (2014) 132–139 137

formulated with CS to enhance powder flowability and respira-bility (Thalberg et al., 2004). With the exception of INL230, lactosesincreased the EF and RFN which were in the ranges 77–87% and15–26%, respectively (Table 2). CS/LH200 mixture provided thebest results with a RFN around 26%. Lactose is generally used to geta higher powder flowability obtaining a better capsule emptyingand aerosolization that are of paramount importance in drypowders for inhalation. Among the lactoses investigated, LH200 isthe only one comprising fine particles in addition to the coarseones. Fine particles are adsorbed on the high energy sites of thecoarse particles making available only the low energy adsorptionsites (Chow et al., 2008). In this way, CS is easily detached from thecarrier during the test (Pilcer and Amighi, 2010; Maas et al., 2010).Fig. 6A shows CS/LH200 formulation consisting of large andirregular lactose particles on which smaller irregular particles areadsorbed (Fig. 6B). Spherical CS particles are observable on lactosesurface at higher magnification (Fig. 6C).

Since Powder 1 and 2 had similar dimensions, Powder 1 wasalso characterized for its aerodynamic propertied using LH200,which led to the best result with Powder 2. Powder 1 had similar EFand RFN than Powder 2. This result evidenced the goodness of themodel employed. In fact, the maximum desirability individuated

for the model drug, BAC, were also appropriate to obtain CS powderwith optimal characteristics for inhalation. The data obtained fromthe aerodynamic tests demonstrated the need of lactose toimprove powder respirability and the importance of lactose typeto achieve the best performances. Considering the data reported inliterature in which RFE were between 7.3 and 46% (Le et al., 2012)and 11.2 and 30.66% (Flament et al., 2004) for different lactose/drugmixtures, the results of the present study can be consideredinteresting to develop a new inhalable formulation, different fromthe one in clinical evaluation (Dharmadhikari et al., 2013).

Fig. 6. Scanning electron microscopy photomicrographs of CS/LH200 mixture.

138 A. Schoubben et al. / International Journal of Pharmaceutics 469 (2014) 132–139

4. Conclusions

By using a new spray-drying technology, it was possible toproduce 2 optimal inhalable CS powders. A 23 design of experimentallowed individuating the best working conditions with a reducednumber of experiments. CS particles formulated with lactoseLH200 showed suitable features for an administration by a DPI.These results are particularly interesting in light of the recentenrolment of spray-dried CS in clinical trial for the treatment ofmulti-drug resistant pulmonary tuberculosis (Dharmadhikariet al., 2013).

Acknowledgement

The Authors would like to thank LUNA (Laboratorio Universi-tario di NAnomateriali) laboratory of the University of Perugia forthe support in performing SEM analysis.

Appendix A. Supplementary data

Supplementary data associated with this article can be found, inthe online version, at http://dx.doi.org/10.1016/j.ijpharm.2014.04.042.

References

Büchi, 2014. Operation Manual Nano Spray Dryer B -90. Büchi, Switzerland.Bürki, K., Jeon, I., Arpagaus, C., Betz, G., 2011. New insights into respirable protein

powder preparation using a nano spray dryer. International Journal ofPharmaceutics 408, 248–256.

Blasi, P., Schoubben, A., Giovagnoli, S., Rossi, C., Ricci, M., 2009. Fightingtuberculosis: old drugs, new formulations. Expert Opinion on Drug Delivery6, 977–993.

Chow, K.T., Zhu, K., Tan, R.B., Heng, P.W., 2008. Investigation of electrostatic behaviorof a lactose carrier for dry powder inhalers. Pharmaceutical Research 12, 2822–2834.

Council of Europe, 2008. Preparations for inhalation: aerodynamic assessment offine particles, European Pharmacopoeia. sixth ed. Council of Europe, Strasbourg(France), pp. 287–288.

Derringer, G., Suich, R., 1980. Simultaneous optimization of several responsevariables. Journal of Quality Technology 12, 214–219.

Dharmadhikari, A.S., Kabadi, M., Gerety, B., Hickey, A.J., Fourie, P.B., Nardell, E.,Phase, I., 2013. single-dose, dose-escalating study of inhaled dry powdercapreomycin: a new approach to therapy of drug-resistant tuberculosis.Antimicrobial Agents and Chemotherapy 57, 2613–2619.

Flament, M.P., Leterme, P., Gayot, A., 2004. The influence of carrier roughness onadhesion, content uniformity and the in vitro deposition of terbutaline sulphatefrom dry powder inhalers. International Journal of Pharmaceutics 275, 201–209.

Forbes, B., Asgharian, B., Dailey, L.A., Ferguson, D., Gerde, P., Gumbleton, M.,Gustavsson, L., Hardy, C., Hassall, D., Jones, R., Lock, R., Maas, J., McGovern, T.,Pitcairn, G.R., Somers, G., Wolff, R.K., 2011. Challenges in inhaled productdevelopment and opportunities for open innovation. Advanced Drug DeliveryReviews 63, 69–87.

Geller, D.E., 2005. Comparing clinical features of the nebulizer, metered-doseinhaler, and dry powder inhaler. Respiratory Care 50, 1313–1322.

Hausner, H.H., 1967. Friction conditions in a mass of metal powders. InternationalJournal of Powder Metallurgy 3, 7–13.

Heng, D., Lee, S.H., Ng, W.K., Tan, R.B.H., 2011. The nano spray dryer B-90. ExpertOpinion on Drug Delivery 8, 965–972.

Islam, N., Cleary, M.J., 2012. Developing an efficient and reliable dry powder inhalerfor pulmonary drug delivery a review for multidisciplinary researchers. MedicalEngineering & Physics 34, 409–427.

Islam, N., Gladki, E., 2008. Dry powder inhalers (DPIs) areview of device reliabilityand innovation. International Journal of Pharmaceutics 360, 1–11.

Laube, B.L., Janssens, H.M., de Jongh, F.H., Devadason, S.G., Dhand, R., Diot, P.,Everard, M.L., Horvath, I., Navalesi, P., Voshaar, T., Chrystyn, H., 2011. What thepulmonary specialist should know about the new inhalation therapies.European Respiratory Journal 37, 108–131.

Le, V.N.P., Bierend, H., Robins, E., Steckel, H., Flament, M.P., 2012. Influence of thelactose grade within dry powder formulations of fluticasone propionate andterbutaline sulphate. International Journal of Pharmaceutics 422, 75–82.

Lee, S.H., Teo, J., Heng, D., Ng, W.K., Chan, H.K., Tan, R.B., 2013. Synergisticcombination dry powders for inhaled antimicrobial therapy: Formulation,characterization and in vitro evaluation. European Journal of Pharmaceutics andBiopharmaceutics 83, 275–284.

Li, X., Anton, N., Arpagaus, C., Belleteix, F., Vandamme, T.F., 2010. Nanoparticles byspray drying using innovative new technology: the Büchi nano spray dryer B-90.Journal of Controlled Release 147, 304–310.

Liu, L.X., Marziano, I., Bentham, A.C., Litster, J.D., White, E.T., Howes, T., 2008. Effectof particle properties on the flowability of ibuprofen powders. InternationalJournal of Pharmaceutics 362, 109–117.

Maas, S.G., Schaldach, G., Walzel, P.E., Urbanetz, N.A., 2010. Tailoring dry powderinhaler performance by modifying carrier surface topography by spray drying.Atomization Spray 20, 763–774.

Nicoli, D.F., Wu, J.S., Chang, Y.J., McKenzie, D.C., Hasapidis, K., 1992. Automatic, highresolution particle size analysis by single-particle optical sensing. AmericanLaboratory 24, 39–44.

Patton, J.S., Byron, P.R., 2007. Inhaling medicines: delivering drugs to the bodythrough the lungs. Nature Reviews Drug Discovery 6, 67–74.

Pilcer, G., Amighi, K., 2010. Formulation strategy and use of excipients in pulmonarydrug delivery. International Journal of Pharmaceutics 392, 1–19.

Rossi, C., Fardella, G., Chiappini, I., Perioli, L., Vescovi, C., Ricci, M., Giovagnoli, S.,Scuota, S., 2004. UV spectroscopy and reverse-phase HPLC as novel methods todetermine capreomycin of liposomal formulations. Journal of Pharmaceuticaland Biomedical Analysis 36, 249–255.

Saleem, I.Y., Smyth, H.D.C., 2010. Micronization of a soft material: air-jet and micro-ball milling. AAPS PharmSciTech 11, 1642–1649.

Sandler, N., Wilson, D., 2010. Prediction of granule packing and flow behavior based onparticle size and shape analysis. Journal of Pharmaceutical Sciences 99, 958–968.

Schafroth, N., Arpagaus, C., Jadhav, U.Y., Makne, S., Douroumis, D., 2012. Nano andmicroparticle engineering of water insoluble drugs using a novel spray-dryingprocess. Colloids and Surfaces B: Biointerfaces 90, 8–15.

A. Schoubben et al. / International Journal of Pharmaceutics 469 (2014) 132–139 139

Schoubben, A., Blasi, P., Giovagnoli, S., Rossi, C., Ricci, M., 2010. Development of ascalable procedure for fine calcium alginate particle preparation. ChemicalEngineering Journal 160, 363–369.

Schoubben, A., Blasi, P., Marenzoni, M.L., Barberini, L., Giovagnoli, S., Cirotto, C., Ricci,M., 2013. Capreomycin supergenerics for pulmonary tuberculosis treatment:preparation, in vitro, and in vivo characterization. European Journal ofPharmaceutics and Biopharmaceutics 83, 388–395.

Smith, I.J., Bell, J., Bowman, N., Everard, M., Stein, S., Weers, J.G., 2010. Inhalerdevices: what remains to be done? Journal of Aerosol Medicine and PulmonaryDrug Delivery 23, 25–37.

Stegemann, S., Kopp, S., Borchard, G., Shah, V.P., Senel, S., Dubey, R., Urbanetz, N.,Citterio, M., Schoubben, A., Hippchen, C., Cade, D., Fuglsang, A., Morais, J.,Borgström, L., Farshi, F., Seyfang, K.-H., Hermann, R., van de Putte, A., Klebovich, I.,Hincal, A., 2013. Developing and advancing dry powder inhalation towardsenhanced therapeutics. European Journalof Pharmaceutical Sciences 48,181–194.

Thalberg, K., Lindholm, D., Axelsson, A., 2004. Comparison of different flowabilitytests for powders for inhalation. Powder Technology 146, 206–221.

Tong, H.H.Y., Chow, A.H.L., 2006. Control of physical forms of drug particles forpulmonary delivery by spray drying and supercritical fluid processing. KONA 24,27–40.

Vehring, R., 2008. Pharmaceutical particle engineering via spray drying. Pharma-ceutical Research 25, 999–1022.

Washington, N., Washington, C., Wilson, C.G., 2001. Factors affecting particledeposition in the lung. In: Washington, N., Washington, C., Wilson, C.G. (Eds.),Physiological Pharmaceutics. Taylor & Francis, New York, pp. 239–242.

Yadav, N., Lohani, A., 2013. Dry powder inhalers: a review. Indo Global Journal ofPharmaceutical Sciences 3, 142–155.

Yu, W., Muteki, K., Zhang, L., Kim, G., 2011. Prediction of bulk powder flowperformance using comprehensive particle size and particle shape distribu-tions. Journal of Pharmaceutical Sciences 100, 284–293.