International Journal of Pharmaceutics 477 (2014) 57–63

Geometry of modified release formulations during dissolution—Influence on performance of dosage forms with diclofenac sodium

Przemysław Doro _zy�nski a,*, Piotr Kulinowski b,c, Witold Jamróz a, Ewelina Juszczyk a

aDepartment of Pharmaceutical Technology and Biopharmaceutics, Pharmaceutical Faculty, Jagiellonian University, ul. Medyczna 9, Kraków 30-688, PolandbDepartment of Magnetic Resonance Imaging, Institute of Nuclear Physics PAN, ul. Radzikowskiego 152, Kraków 31-342, Polandc Institute of Technology, The Pedagogical University of Cracow, ul. Podchora _zych 2, Kraków 30-084, Poland

A R T I C L E I N F O

Article history:Received 14 July 2014Received in revised form 2 October 2014Accepted 7 October 2014Available online 12 October 2014

Keywords:Diclofenac sodiumModified releaseMagnetic resonance imagingMRIFractal dimension

A B S T R A C T

The objectives of the work included: presentation of magnetic resonance imaging (MRI) and fractalanalysis based approach to comparison of dosage forms of different composition, structure, andassessment of the influence of the compositional factors i.e., matrix type, excipients etc., on propertiesand performance of the dosage form during drug dissolution. The work presents the first attempt tocompare MRI data obtained for tablet formulations of different composition and characterized by distinctdifferences in hydration and drug dissolution mechanisms. The main difficulty, in such a case stems fromdifferences in hydration behavior and tablet’s geometry i.e., swelling, cracking, capping etc. A novelapproach to characterization of matrix systems i.e., quantification of changes of geometrical complexityof the matrix shape during drug dissolution has been developed. Using three chosen commercialmodified release tablet formulations with diclofenac sodium we present the method of parameterizationof their geometrical complexity on the base of fractal analysis. The main result of the study is thecorrelation between the hydrating tablet behavior and drug dissolution – the increase of geometricalcomplexity expressed as fractal dimension relates to the increased variability of drug dissolution results.

ã 2014 Elsevier B.V. All rights reserved.

Contents lists available at ScienceDirect

International Journal of Pharmaceutics

journal homepage: www.elsev ier .com/locate / i jpharm

1. Introduction

Modified-release dosage forms are complex formulationsintended to release drug with controlled rate to achieve desiredefficacy and safety by reducing fluctuations in plasma concen-trations of drug. The variety of strategies of modified releasedevelopment as well as structural differences between formula-tions may affect their properties. Understanding of their techno-logical attributes of modified release formulations as well as thestructural and functional consequences of the composition andmanufacturing process is pertinent for better prediction of the invivo performance of modified release products. Nowadays thesolution of this problem is implemented at two levels: (i)understanding and controlling of manufacturing variables withquality by design (QbD) tools (Dickinson et al., 2008; Singh et al.,2011) and (ii) understanding of drug product performance with

* Corresponding author at: Department of Pharmaceutical Technology andBiopharmaceutics, Pharmaceutical Faculty, Jagiellonian University, ul. Medyczna 9,30-688 Kraków, Poland. Tel.: +48 126205603/603535371; fax: +48 126205619.

E-mail address: mfdorozy@cyf-kr.edu.pl (P. Doro _zy�nski).

http://dx.doi.org/10.1016/j.ijpharm.2014.10.0160378-5173/ã 2014 Elsevier B.V. All rights reserved.

advanced analytical techniques, such as magnetic resonanceimaging (MRI), FTIR microscopy, high-resolution computed micro-tomography (micro-CT) (Zeitler and Gladden, 2009; Huanabuttaet al., 2012). The results obtained by these techniques are referred(directly or indirectly) to drug dissolution studies (Doro _zy�nskiet al., 2012).

The integrated analysis of modified release formulationproperties requires to develop the methodology that (i) allowsto quantify geometrical and/or morphological changes of dosageforms, (ii) enables to compare formulations of different structureand mechanism of action on the base of the results of imageanalysis and drug dissolution study, and (iii) gives the possibility torelate quantified structural changes to drug dissolution profiles.Although the need to develop methodology for multifacetedevaluation of such formulations has been widely discussed in thelast decade (Chen et al., 2010), these challenges remain stillunanswered.

The influence of geometry of modified release formulations onthe drug release behavior has been reviewed by Moodley et al.(2012). It must be pointed out that, geometrical aspects of oralmodified release dosage forms were considered only for drymatrices. So far there was no attempt to analyze geometrical

58 P. Doro_zy�nski et al. / International Journal of Pharmaceutics 477 (2014) 57–63

parameters of matrices during dissolution studies to comparecommercial formulations. Magnetic Resonance Imaging is the onlymethod, which allows for relatively unrestricted evolution ofpolymeric matrix and imaging of its internal structure. Previousworks concerning integration of MRI technique with drugdissolution studies were focused on formulations with essentiallysimilar structure, geometry, and behavior (Fyfe and Blazek-Welsh,2000; Baumgartner et al., 2005). In the current work, we proposethe first attempt to the quantification and comparison of MRimages of matrix systems based on various excipients. Thedeveloped methodology is proposed for characterization ofchanges, which occur during matrix hydration and drug dissolu-tion. Commercial formulations with diclofenac sodium (DS) wereused to illustrate the proposed methodology.

Fractal analysis was applied to describe the external boundaryor the surface structure of objects. It was previously used forcharacterization of pharmaceutical powders and particulateformulations. The description of the geometry of the matrixtablets during hydration using fractal analysis has not beenpreviously described. In 1998, Riippi et al. (1998) proposedcharacteristics of dry tablets surface with fractal dimensionparameter. They correlated surface geometry of the tablets withtheir mechanical parameters i.e., crushing strength and friability.The fractal analysis was applied to study the morphology of theeffective surface influencing the sweetener dissolution (Tromelinet al., 2001). It was also used for characterization of tabletformation composed of different types of lactose (Jelcic et al.,2007). Recently, Yin et al. (2013) proposed 3D fractal analysis toquantify simultaneously the shapes of internal and surfacestructures of felodipine osmotic pump tablets during the drugrelease process as studied by X-ray micro tomography.

The aims of the work were (i) to present MRI and fractal analysisbased approach to compare dosage forms of various compositionand structure, (ii) to characterize the properties of commercialmodified release formulations with diclofenac sodium, and (iii) toassess the influence of the compositional factors on properties andperformance of the dosage form during drug dissolution studies.

2. Materials and methods

2.1. Materials

Three commercial modified release matrix tablets containing100 mg of diclofenac sodium were chosen for the study. Thequalitative compositions of each formulation based on theinformation from summary of product characteristics (SPC)are presented in Table 1. All other materials used in the studywere of analytical grade.

2.2. Magnetic resonance imaging (MRI) and dissolution studies

The MRI and dissolution studies were carried out in the systemcomposed of 4.7T super-conducting magnet (Bruker, Germany)with TMX (NRC, IBD, Canada) digital MR console and a specialnon-magnetic flow-through cell designed specifically for use in

Table 1Qualitative composition of commercial formulations used in the study.

Formulation Core

I Ethylcellulose, magnesium stearate, povidone

II Eudragit RS 30D, Eudragit L 100–55, eudragit RS PO, cellulose powdecolloidal anhydrous, talc, povidone K-25, magnesium stearate

III Sucrose, cetyl alcohol, povidone, magnesium stearate, silica colloidal

high magnetic fields. A detailed description of the system waspresented elsewhere (Kulinowski et al., 2008).

The dissolution study was carried out using flow throughmethod (apparatus 4) with media exchange after 2 h. The testcovered two consecutive phases:

1. 2 h in 1000 mL of simulated gastric fluid (SGF) – 0.1 mol/L HClsolution pH 1.2.

2. 6 h in 1000 mL of simulated intestinal fluid (SIF) – phosphatebuffer (pH 6.8).

The dissolution medium was circulated in the closed loop withthe circulation rate 40 mL/min, the temperature in the dissolutioncell was maintained at 37 �C � 0.5 �C using the thermostatic waterbath (LW 502M, AJL Electronic, Poland). The samples of solutionwere withdrawn every 30 min. The volume of the sample was 5 mL.Equal amounts of temperature-equilibrated dissolution mediawere replaced immediately after withdrawal of the sample. Theamounts of released diclofenac sodium were determined bymeasuring the UV absorption at 280 nm and calculated using acalibration curve of the drug. For each of the formulations,dissolution was repeated six times.

The MRI studies were carried out with flow compensated spin-echo sequence to obtain images of a dosage form, in theflow-through cell at continuous flow condition in the sameconditions as dissolution study. Imaging sequence parameterswere as follows: echo time (TE) of 19 ms, repetition time (TR) of625 ms, number of excitations (NEX) – 4. Matrix size of256 � 256 for the 3.5 cm field of view (FOV) was used and theslice thickness was 1 mm, resulting in spatial resolution of0.14 � 0.14 �1 mm and temporal resolution of 30 min (total scantime of 10 min).

2.3. Image analysis

The images were analyzed using public domain software Fijidistribution of ImageJ v.1.44 (Schindelin et al., 2012). For fractalanalysis, FracLac plug-in for ImageJ was used (Karperien, 1999).

Images obtained in SGF suffered from lower signal to noiseratio, due to high probehead electrical loading. For these images,Gaussian blur (sigma of 1.7) was applied to reduce noise. Theschematic diagram of image processing and analysis is presentedin the Fig. 1. All images were interpolated to 512 � 512 usingbilinear interpolation (Fig. 1A). Linear Kuwahara edge preservingfilter was applied (number of angles: 30, line length 11, criterionvariance/mean square) for noise reduction and improving themanual region selection for segmentation (Fig 1B). Imagesegmentation was performed using trainable segmentation basedon the Weka data mining software implemented as Fiji plug-in(Schindelin et al., 2012). The two regions were assumed, i.e., matrixas a whole together with a holder and solvent around the matrix.Segmentation resulted in binary images of matrix (Fig. 1C). Next,the region of interest (ROI) containing the shape of the tablet onlywas selected. The region outside the ROI was cleared (Fig. 1D) andthe edge detection filter was applied to obtain an outline of thehydrating matrix (Fig. 1E). Finally, outlines of the matrices were

Coating

Hypromellose, diethyl phtalate, magnesium stearate, titaniumdioxide

r, silica Hypromellose, macrogol 6000, titanium dioxide, orange–yellowlack, propylene glycol

anhydrous Hypromellose, purified talc, polysorbate 80, titanium dioxide, rediron oxide, sucrose, polyethylene glycol 8000

Fig.1. Diagram of image analysis procedure; (A) raw image, (B) image after filtration (Kuwahara filter), (C) image segmentation, (D) ROI, (E) edge detection, (F) example of boxcounting.

P. Doro _zy�nski et al. / International Journal of Pharmaceutics 477 (2014) 57–63 59

analyzed with FracLac software. The box counting dimensionranged from 2 pixels (a minimum box size) to 230 pixels (amaximum box size). As a result, fractal dimension (DB) of thetablet’s border region was obtained.

3. Results and discussion

As it was mentioned above, for the purpose of the study threecommercial matrix formulations of diclofenac sodium (DS) withdifferent excipients were selected (Table 1). Formulation I was basedon ethylcellulose as water insoluble drug delivery matrix platform(Streubel et al., 2000). Formulation II was based on cellulose andEudragit RS as pH independent customized release matrix formerand Eudragit L100-55 as pH dependent polymer dissolving at pH 5.5(Streubel et al., 2000; Grund et al., 2014). Formulation III wascomposed of water insoluble cetyl alcohol and sucrose as dissolutionmodulator (Dvorackova et al., 2009). All formulations were coatedwith non-functional hypromelose based coatings.

3.1. Qualitative description

Fig. 2 shows the comparison among cross-sections of thetablets obtained with MRI technique, at two representative timepoints of the dissolution study – the first, at early stage ofhydration, after 2 h in hydrochloric acid solution and the second at5th hour of hydration (2 h in pH 1.2 and 3 h in pH 6.8).

The hydration patterns of the matrix systems varied. In the caseof tablets based on ethylcellulose (formulation I), only limitedwater penetration into the lateral sides of the matrix was observed(Fig 2A). The presence of hydrophobic polymer in the modifiedrelease matrix tablets slowed water penetrability (Badshah et al.,2011). As a result of water ingress into the matrix, partial capping ofthe tablets was observed (Fig. 2B).

In the case of the matrix composed of cellulose and Eudragits(formulation II), in the acidic phase of dissolution, the restricted

hydration of the matrix was observed (Fig. 2C). After mediaexchange, the dissolving at pH 6.8 Eudragit L100-55 promotedhydration of the matrix. Medium penetration toward the core ofthe tablet together with swelling of the polymeric matrix led tocracking of hydrated external layer of the tablets. Subsequently, thegeometry of the matrix started to change. The geometrical changesof the matrix were clearly visible starting from the fifth hour of theexperiment (Fig. 2D).

The presence of sucrose in the formulation III resulted in thefast formation of the highly hydrated external layer (Fig. 2E). Thesucrose acted within the insoluble cetyl alcohol matrix as pHindependent pore forming agent. After 1 h in SGF, a thick externallayer (characterized by high MR signal intensity) on the surface ofthe tablets was identified. The formation of capillary structurewithin the tablets together with osmotic properties of sucrosefacilitated fast water ingress into the matrix. The increase of thewater concentration in the external zone of the tablet and lack ofswelling properties of cetyl alcohol weakened the structure of thematrix and resulted in the increase of its fragility. It led tofragmentation of the external layer (Fig. 2F). A similar observationwas made by Abd-Elbary et al. (2013) using the scanning electronmicroscope (SEM). The micrographs of cross-sections of CRetodolac lipid matrix tablets with sucrose revealed the increaseof porosity of the tablets after dissolution.

3.2. Dissolution profiles

The dissolution profiles of DS are presented in Fig. 3. Insimulated gastric fluid the release of diclofenac sodium wasrestricted, due to the acidic nature (pKa� 4) of the active substance.At the end of acidic phase (2 h), the mean amounts of released DSdid not exceed 1.05% (up to 2.2% in individual measurements).After media exchange to SIF (pH 6.8), gradual increase ofcumulative concentrations of released DS was observed. The drugdissolution after eight hours of the study was incomplete for all

Fig. 2. Magnetic resonance images of DS tablets at specific time points presenting evolution of geometry and changes of internal structure of analyzed formulations: (A)formulation I, 2 h, (B) formulation I, 5 h, (C) formulation II, 2 h, (D) formulation II, 5 h, (E) formulation III, 2 h, (F) formulation III, 5 h.

60 P. Doro_zy�nski et al. / International Journal of Pharmaceutics 477 (2014) 57–63

formulations. For the formulation III, about 39% of the drug werereleased. The amounts of drug released from the formulations I andII were smaller – about 16 and 22%, respectively.

Internal variability of the dissolution results depended on thecomposition of modified release tablets. In the case of formulationI, the standard deviation (SD) values of cumulative concentrationwere below 1% throughout the experiment. Whilst in the cases offormulations II and III the gradual increase in SDs was observed –

up to about 3 and 6%, respectively (Fig. 4), at the eighth hour of thestudy.

3.3. Fractal analysis

The majority of MRI studies of modified release formulationswas previously focused on analysis of the structurally similarformulations (Fyfe and Blazek-Welsh, 2000; Tajarobi et al., 2009).The main difficulty is that image analysis and data interpretationshould allow comparing dosage forms of different geometry andhydration behavior. In our study, internal structure of hydratingmatrix systems based on various polymers was totally differente.g., swelling front and hydration zones usually identified in

0

5

10

15

20

25

30

35

40

45

50

0 0,5 1 1,5 2 2,5 3 3,5 4 4,5 5 5,5 6 6,5 7 7,5 8

Rel

ease

d D

S [%

]

Time [h]

Formulation I Formulation II Formulation III

Fig. 3. Dissolution profiles of DS formulations. (Error bars indicate standard deviation).

P. Doro _zy�nski et al. / International Journal of Pharmaceutics 477 (2014) 57–63 61

hydroxypropyl methylcellulose matrices (Kulinowski et al., 2011,2012, 2014Doro _zy�nski et al., 2012) not occurred in polymericmatrices used in the study. Therefore, the changes in the matriceswere difficult to compare. In this study, we propose parameteriza-tion of the changes of hydrating matrix systems using matrixoutline complexity with the fractal dimension (DB) parameter. Thechanges of DB values over the time are presented in Fig. 5. Thevalues of fractal dimension of the formulation I were only slightlychanged during 8 h of hydration, they were concentrated around1.145. For the formulation II, DB time course up to 3.5 h reflectsrestricted matrix hydration – DB is concentrated at 1.139. Afterswitching to SIF, matrix started to “open” resulting in relatively fastchange in DB. Between 4.5 and 7 h of hydration, the values of DB areconcentrated at 1.189. For the formulation III, gradual increase offractal dimension was observed from 1.137 to 1.194.

3.4. Fractal dimension DB versus variability of dissolution results

The increase of DB during dissolution corresponded to theincreasing variability of dissolution results. The juxtaposition ofstandard deviations for dissolution results, in particular timepoints, and fractal dimensions (Fig. 6) showed that the distributionof the standard deviation values correlated with changes of valuesof DB.

The formulation I was characterized by low standard deviationof dissolution results and restricted changes in matrix geometry. It

Fig. 4. Changes of standard deviation value for p

led to the narrow distribution of the data on the graph (cluster A inFig. 6). For the formulation II and III, two separate data clusterswere identified; (i) initial, characterized by low SD values and lowmatrix geometric complexity expressed as low DB values (clustersB1, C1) and (ii) subsequent, characterized by higher SD values andmore complex geometry of the matrix (clusters B2, C2). Therelationship between internal variability of dissolution results andthe geometry of the matrix was common for all formulations. Thelinear correlation (R2 = 0.7064) between the SD and DB values wasobserved.

The importance of the relation between morphology of dosageforms, quantified by fractal dimension, and dissolution results hasbeen indicated by Pippa et al. (2014). They analyzed the relationbetween the fractal dimension of self-assembling liposomalstructures and different release patterns of indomethacine. Thetrend of decreasing drug release rate with increasing fractaldimension was demonstrated.

The in vitro studies carried out by Bertocchi et al. (2005) oncommercial multisource modified release tablets with diclofenacsodium shown that the release profiles varied considerablybetween formulations. The authors suggest serious implicationsfor the bioequivalence (and in consequence for interchangeability)of the CR dosage forms. Interestingly, besides the differences indissolution profiles between formulations, in some cases, theintra-batch variability of dissolution results was observed.Although the paper did not discuss broader this phenomenon,

articular time points in dissolution profiles.

Fig. 5. Changes of MR images fractal dimension (DB) value for particular time points in dissolution study.

Fig. 6. Fractal dimension (DB) versus standard deviation of the dissolution results. The trend line shows/illustrates the global relationship between variability of dissolutionresults and changes in external matrix complexity. The ovals cover separate data clusters.

62 P. Doro_zy�nski et al. / International Journal of Pharmaceutics 477 (2014) 57–63

the conclusion is that the internal variability of dissolution resultscould be an important factor influencing DS in vivo performance.Our study confirmed intra-batch variability of dissolution resultsbetween formulations. By extension of DS formulations studieswith a set of new methods (MRI, fractal analysis), we have shownthat the repeatability of drug dissolution results was dependent onthe structural and compositional properties of the dosage form.

The studies after oral administration of DS tablets to healthyvolunteers by Lötsch et al. (2000) showed the high variability of drugplasma levels. Moreover, the lack of bioequivalence of Voltaren LP1

and generic Xenid LP1 was presented by Hooper et al. (1996). Theauthors have shown that the reference and generic batches were notbioequivalent but also generic batches were not bioequivalent whencompared with each other. The observations presented in our papermay help to explain one of potential reasons of such observations.Although the pharmacokinetics of the drug is undoubtedly the resultof physiological factors, the differences in dissolution in vitro may bealso an important source of variability of DS plasma levels.

In the work of Chen et al. (2010), it is pointed out that “drugrelease time course and dissolution profile are the fundamental

properties for modified release formulation, which needs to bewell understood in order to predict the performance of theproduct.” While dissolution profiles give only indirect informationabout matrix performance, there is a need to supplementdissolution profiles with data, which allow additional differentia-tion of formulations (Doro _zy�nski et al., 2011) and can giveadditional knowledge of formulation physicochemical propertiesand microstructure (Kulinowski et al., 2012, 2014). In the presentedexample, the standard deviation of drug dissolution stronglydepends upon the matrix shape complexity and integrity of thehydrated matrix. Presented study is another confirmation that theuse of MRI may be useful as prerequisite to establish the predictivedissolution testing methodology for modified release generic drugdevelopment.

4. Conclusions

In order to compare the behavior of different types of modifiedrelease DS matrix systems, the magnetic resonance imaging basedmethod with dedicated image processing and analysis was

P. Doro _zy�nski et al. / International Journal of Pharmaceutics 477 (2014) 57–63 63

proposed. Fractal analysis was applied to quantify geometricalchanges of modified release matrix systems during drug dissolu-tion. The method allowed to characterize the behavior of chosencommercial formulations with DS during drug dissolution studiesand to explain it on the base of analysis of the qualitativecomposition of each drug product. In the presented study, evenmore important than cumulative dissolution profiles were theirstandard deviations. The changes of matrix systems geometryexpressed by changes in fractal dimension were related to thevariability of drug dissolution results. Matrices composed ofswelling, soluble polymers, which changed their shape and sizeduring hydration, were characterized by higher fractal dimensionand with higher standard deviations of dissolution resultscomparing to non-swelling matrix based on insoluble polymer.

Acknowledgement

The project was co-financed by the European RegionalDevelopment Fund under the Infrastructure and EnvironmentProgram UDA-POIS.13.01-023/09-00.

References

Abd-Elbary, A.L., Tadros, M.I., Alaa-Eldin, A.A., 2013. Sucrose stearate-enriched lipidmatrix tablets of etodolac: modulation of drug release, diffusional modeling andstructure elucidation studies. AAPS PharmSciTech 14, 656–668.

Badshah, B., Subhan, F., Rauf, K., Bukhari, N.I., Shah, K., Khan, S., Ahmed, Z., Khan, I.,2011. Development of controlled-release matrix tablet of risperidone: influenceof methocel1- and ethocel1-based novel polymeric blend on in vitro drugrelease and bioavailability. AAPS PharmSciTech 525–533.

Baumgartner, S., Lahajnar, G., Sepeb, A., Kristl, J., 2005. Quantitative evaluation ofpolymer concentration profile during swelling of hydrophilic matrix tabletsusing 1H NMR and MRI methods. Eur. J. Pharm. Biopharm. 59, 299–306.

Bertocchi, P., Antoniella, E., Valvo, L., Alimonti, S., Memoli, A., 2005. Diclofenacsodium multisource prolonged release tablets – a comparative study on thedissolution profiles. J. Pharm. Biomed. Anal. 37, 679–685.

Chen, M.L., Shah, V.P., Ganes, D., Midha, K.K., Caro, J., Nambiar, P., Rocci Jr., M.L.,Thombre, Abrahamsson, A.G., Conner, B., Davit, D., Fackler, B., Farrell, P., Gupta,C., Katz, S., Mehta, R., Preskorn, M., Sanderink, S.H., Stavchansky, G., Temple, S.,Wang, R., Winkle, Y., Yu, H., 2010. Challenges and opportunities in establishingscientific and regulatory standards for assuring therapeutic equivalence ofmodified-release products: workshop summary report. Eur. J. Pharm. Sci. 40,148–153.

Dickinson, P.A., Lee, W.W., Stott, P.W., Townsend, A.I., Smart, J.P., Ghahramani, P.,Hammett, T., Billett, L., Behn, S., Gibb, R.C., Abrahamsson, B., 2008. Clinicalrelevance of dissolution testing in quality by design. AAPS J. 10, 280–290.

Doro _zy�nski, P., Kulinowski, P., Mendyk, A., Jachowicz, R., 2011. Gastroretentive drugdelivery systems with Ldopa based on carrageenans and hydroxypropylmethylcellulose. Int. J. Pharm. 404, 169–175.

Doro _zy�nski, P.P., Kulinowski, P., Młynarczyk, A., Stanisz, G.J., 2012. Foundationreview: MRI as a tool for evaluation of oral controlled release dosage forms.Drug Disc Today 17, 110–123.

Dvorackova, K., Rabiskova, M., Masteikova, R., Muselik, J., Krejcova, K., 2009. Solublefiller as a dissolution profile modulator for slightly soluble drugs in matrixtablets. Drug Dev. Ind. Pharm. 35, 930–940.

Fyfe, C.A., Blazek-Welsh, A.I., 2000. Quantitative NMR imaging study of themechanism of drug release from swelling hydroxypropyl methylcellulosetablets. J. Control Release 68, 313–333.

Grund, J., Koerber, M., Walther, M., Bodmeier, R., 2014. The effect of polymerproperties on direct compression and drug release from water-insolublecontrolled release matrix tablets. Int. J. Pharm. 469, 94–101.

Hooper, I.T., Allen, E., McLaughlin, K., Ward, C., Sioufi, A., 1996. Bioavailability of ageneric sustained-release formulation of diclofenac compared with thestandard sustained-release formulation. Clin. Drug Invest. 12, 259–270.

Huanabutta, K., Teradac, K., Sriamornsak, P., Nunthanidb, J., 2012. Advancedtechnologies for assessment of polymer swelling and erosion behaviors inpharmaceutical aspect. Eur. J. Pharm. Biopharm. 83, 315–321.

Jelcic, Z., Hauschild, K., Ogiermann, M., Picker-Freyer, K.M., 2007. Evaluation of tabletformation of different lactoses by 3d modeling and fractal analysis. Drug Dev.Ind. Pharm. 33, 353–372.

Karperien, A. V 251e. Charles Sturt University, 1999. FracLac. http://rsb.info.nih.gov/ij/plugins/frac-lac.html.

Kulinowski, P., Doro _zy�nski, P., Jachowicz, R., Weglarz, W., 2008. Integrated systemfor dissolution studies and magnetic resonance imaging of the hydrodynami-cally balanced systems – a tool for quantitative assessment of hydrogelformation process. J. Pharm. Biomed. Anal. 48, 685–693.

Kulinowski, P., Doro _zy�nski, P., Młynarczyk, A., Weglarz, W.P., 2011. Magneticresonance imaging and image analysis for assessment of hpmc matrix tabletsstructural evolution in usp apparatus 4. Pharm. Res. 28, 1065–1073.

Kulinowski, P., Mlynarczyk, A., Dorozynski, P., Jasi�nski, K., Gruvel, M., Tomanek, B.,Weglarz, W.P., 2012. Magnetic resonance microscopy for assessment ofmorphological changes in hydrating hydroxypropylmethyl cellulose matrixtablets in situ. Pharm. Res. 29, 3420–3433.

Kulinowski, P., Młynarczyk, A., Jasi�nski, K., Talik, P., Gruwel, M.L., Tomanek, B.,Weglarz, W.P., Doro _zy�nski, P., 2014. Magnetic resonance microscopy forassessment of morphological changes in hydrating hydroxypropylmethylcellu-lose matrix tablets in situ-is it possible to detect phenomena related to drugdissolution within the hydrated matrices? Pharm. Res. 31, 2383–2392.

Lötsch, J., Kettenmann, B., Renner, B., Drover, D., Brune, K., Geisslinger, G., Kobal, G.,2000. Population pharmacokinetics of fast release oral diclofenac in healthyvolunteers: relation to pharmacodynamics in an experimental pain model.Pharm. Res. 17, 77–84.

Moodley, K., Pillay, V., Choonara, Y.E., du Toit, L.C., Ndesendo, V.M.K., Kumar, P.,Cooppan, S., Bawa, P., 2012. Oral drug delivery systems comprising alteredgeometric configurations for controlled drug delivery. Int. J. Mol. Sci. 13, 18–43.

Pippa, N., Dokoumetzidis, A., Pispas, S., Demetzos, C., 2014. The interplay betweenthe rate of release from polymer grafted liposomes and their fractalmorphology. Int. J. Pharm. 465, 63–69.

Riippi, M., Antikainena, O., Niskanenb, T., Yliruusia, T., 1998. The effect ofcompression force on surface structure crushing strength,friability anddisintegration time of erythromycin acistrate tablets. Eur. J. Pharm. Biopharm.46, 339–345.

Schindelin, J., Arganda-Carreras, I., Frise, E., Kaynig, V., Longair, M., Pietzsch, M.,Preibisch, S., Rueden, C., Saalfeld, S., Schmid, B., Tinevez, J.-Y., James White, D.,Hartenstein, V., Eliceiri, K., Tomancak, P., Cardona, A., 2012. Fiji: an open-sourceplatform for biological-image analysis. Nat. Methods 9, 676–682.

Singh, B., Kapil, R., Nandi, M., Ahuja, N., 2011. Developing oral drug delivery systemsusing formulation by design: vital precepts, retrospect and prospects. ExpertOpin. Drug Deliv. 8, 1341–1360.

Streubel, A., Siepmann, J., Dashevsky, A., Bodmeier, R., 2000. pH-Independentrelease of a weakly basic drug from water-insoluble and -soluble matrix tablets.J. Controlled Release 67, 101–110.

Tajarobi, F., Abrahmsén-Alami, S., Carlsson, A.S., Larsson, A., 2009. Simultaneousprobing of swelling, erosion, and dissolution by NMR-microimaging – effect ofsolubility of additives on HPMC matrix tablets. Eur. J. Pharm. Sci. 37, 89–97.

Tromelin, A., Hautbout, G., Pourcelot, Y., 2001. Application of fractal geometry todissolution kinetic study of a sweetener excipient. Int. J. Pharm. 224, 131–140.

Yin, X., Lia, H., Liu, R., Chen, J., Ji, J., Chen, J., Shao, O., Xiao, T., York, P., He, Y., Zhanga, J.,2013. Fractal structure determines controlled release kinetics of monolithicosmotic pump tablets. J. Pharm. Pharmacol. 65, 953–959.

Zeitler, J.A., Gladden, L.F., 2009. In-vitro tomography and non-destructive imaging atdepth of pharmaceutical solid dosage forms. Eur. J. Pharm. Biopharm. 71, 2–22.