Drug Design, Development and Therapy Dovepress

open access to scientific and medical research

Open Access Full Text Article Original Research

Optimization of a combined wet milling process

in order to produce poly(vinyl alcohol) stabilized
nanosuspension
This article was published in the following Dove Press journal:
Drug Design, Development and Therapy

Csaba Bartos 1 Purpose: The article reports a wet milling process, where the planetary ball mill was combined
Orsolya Jójárt-Laczkovich 1 with pearl milling technology to reach nanosize range of meloxicam (Mel; 100–500 nm). The main
Gábor Katona 1 purpose was to increase the dissolution rate and extent of a poorly water-soluble Mel as nonsteroidal
Mária Budai-Szűcs 1 anti-inflammatory drug as well as to study its permeability across cultured intestinal epithelial cell
Rita Ambrus 1 layers.
Methods: Viscosity of milled dispersion and particle size distribution and zeta potential of
Alexandra Bocsik 2
Mel were investigated and differential scanning calorimeter and X-ray powder diffractometer
Ilona Gróf 2
were used to analyse the structure of the suspended Mel. Finally in vitro dissolution test and in
Mária Anna Deli 2
vitro cell culture studies were made.
Piroska Szabó-Révész 1 Results: It was found that the ratio of predispersion and pearls 1:1 (w/w) resulted in the most effec-
1
Faculty of Pharmacy, Institute of tive grinding system (200-fold particle size reduction in one step) with optimized process parameters,
Pharmaceutical Technology and
437 rpm and 43 min. Nanosuspension (1% Mel and 0.5% poly[vinyl alcohol]) as an intermediate
Regulatory Affairs, University of
Szeged, Szeged, Hungary; 2Institute product showed a stable system with 2 weeks of holding time. This optimized nanosuspension
of Biophysics, Biological Research enhanced the penetration of Mel across cultured intestinal epithelial cell layers without toxic effects.
Centre, Hungarian Academy of
Sciences, Szeged, Hungary Conclusion: The dissolution rate of Mel from the poly(vinyl alcohol) stabilized nanosuspension
justified its applicability in the design of innovative per oral dosage form (capsule) in order to
ensure/give a rapid analgesia.
Keywords: nanonization, meloxicam, milled dispersion, milling efficiency, zeta potential,
intermediate product

Introduction
The planetary ball milling belongs to the group of high-energy milling methods. The
process is mainly used in laboratory-scale research work. It is a common technique
for dry milling,1,2 nevertheless it is also suitable for wet grind.3–5 Dry milling with this
technique is usually used for micronization with a particle size range of 1–2,000 µm.6
In general, additives are not required for micronization, but for dry nanonization,
application of them and a long milling time (2–4 h) can be necessary. Additives decrease
the cohesion between the nanosize particles and the collision energy during the mill-
ing process; thereby, the risk of the decomposition of the active agent can be reduced.7
Correspondence: Piroska Szabó-Révész Wet milling is a top–down process, where the raw material is broken down via
Faculty of Pharmacy, Institute of mechanical forces. In this method, a sufficiently concentrated dispersion of drug
Pharmaceutical Technology and
Regulatory Affairs, University of Szeged,
particles in an aqueous or nonaqueous liquid medium is treated. Increased mill
Eötvös u 6, Szeged H-6720, Hungary capacity, lower energy consumption, and easier handling of materials can be perceived
Tel +36 62 545 572
Fax +36 62 545 571
as advantages of the process. However, it must be said, in the course of the milling
Email revesz@pharm.u-szeged.hu process, increased wear of the milling medium and corrosion can occur, and the

submit your manuscript | www.dovepress.com Drug Design, Development and Therapy 2018:12 1567–1580 1567
Dovepress © 2018 Bartos et al. This work is published and licensed by Dove Medical Press Limited. The full terms of this license are available at https://www.dovepress.com/terms.php
http://dx.doi.org/10.2147/DDDT.S159965
and incorporate the Creative Commons Attribution – Non Commercial (unported, v3.0) License (http://creativecommons.org/licenses/by-nc/3.0/). By accessing the work you
hereby accept the Terms. Non-commercial uses of the work are permitted without any further permission from Dove Medical Press Limited, provided the work is properly attributed. For permission
for commercial use of this work, please see paragraphs 4.2 and 5 of our Terms (https://www.dovepress.com/terms.php).
Bartos et al Dovepress

instability of the active agent must be taken into account. were required. The influence of milling time was also studied
Wet milling is applicable for micronization in the particle on the particle size distribution, crystallinity, and dissolution
size range of 1–50 µm, in this case, colloid mill, toothed rate of Mel.
high-shear inline mixers, and cone mill can be used,8 but During the use of the combined wet milling technique,
for nanonization, high-pressure homogenization (1–20 µm) the type and amount of PVA are very important because it
and pearl milling technique (20–200 nm)9–11 can be applied. has a dual role. On the one hand, it promotes the grinding
It should be noted that in case of high-pressure homog- efficiency in the concentrated predispersion, and on the other
enization and pearl milling techniques, the preparation of hand, it stabilizes the milled dispersion and later the final
pretreated dispersions (particle size reduction to 1–10 µm) nanosuspension in a diluted medium.
is required to reach the nanosize range. PVA is a nonionic polymer with very different molecular
The pearl milling process has been proven to be a robust weights. It is frequently used as a stabilizer agent.20 Polymer
technique for the production of nanoparticle suspension of adsorption on the solid–liquid interface can be influenced
poorly water-soluble drugs.12 With this method, nanosuspen- by the various conformations of the polymer chains and the
sions are produced through the use of high-shear media or interaction of the polymer segments with the solvent and the
pearl mills. Pearl milling is a continuous process wherein the surface of the solid as well. PVA with low molecular weight
drug suspension is pumped through the milling chamber in (about 20,000 g/mol) is adsorbed on the colloid particles and
order to reduce the particle size of the suspended material. thus stabilizes the colloid suspension (coating the particles
The milling medium consists of glass, zirconia, or highly and providing the repulsion among them), whereas PVA with
cross-linked polystyrene resin.13 The physical characteris- high molecular weight (about 1,000,000 g/mol) flocculates
tics of the resulting nanocrystals depend on the number and the dispersed systems.24
size of the milling pearls, the amount of the drug, and the The prediction of permeability features of drug candi-
stabilizer(s).14,15 dates across biological barriers is of great importance in the
Thanks to the high efficiency of a smaller pearl size16 early phase of drug development.25 Oral drug formulations
and the high mechanical forces of the planetary mill, Retsch are the most widespread in human therapy, and therefore
GmbH (Haan, Germany) recommends the combination of the intestinal drug absorption is the most studied in pharma-
planetary ball and pearl milling as a novel milling technique ceutical research. The human Caco-2 cell line, presenting
in order to prepare drug nanodispersions.17 In the literature, many of the structural and functional aspects of the epithe-
there are few articles about the combinative method, where lium of small intestine, is a routinely used culture model of
various active agents were co-milled. In the presence of intestinal drug penetration showing good correlation with
d-tocopherol polyethylene glycol 1000 succinate as a stabi- in vivo data.26
lizer agent, nanoparticle range was achieved; however, the The aim of this work was to optimize the process param-
production efficiency was very low.18,19 eters (pearl amount, milling time, and rotation speed) of the
Our team uses different milling techniques (dry and wet) combined wet milling technique (planetary ball and pearl
in order to nanonize different water-insoluble drugs, eg, milling), using Mel as an active agent and PVA as a stabiliz-
meloxicam (Mel), to provide a faster dissolution, a higher ing agent in the predispersion. We basically investigated the
saturated concentration, a faster absorption, and, in this influence of the amount of low molecular weight PVA on
context, a better bioavailability. Mel as a nonsteroidal anti- the grinding efficiency in concentrated predispersion. Our
inflammatory agent is often used in malignant and nonma- aim was also to get to know the effect of mechanical forces
lignant pain therapy, but its bioavailability is unsatisfactory on polymer viscosity and drug–polymer interaction as well.
thanks to its poor solubility in the gastrointestinal tract. One To describe the stability of nanosuspensions, their particle
strategy to address these problems is the particle size reduc- size was monitored for 2 weeks. Furthermore, the optimized
tion (eg, nanonization), which increases the dissolution rate Mel formulations were tested on the cell culture model of
of the poorly soluble drug resulting in faster absorption and intestinal epithelium.
faster action in pain therapy. The novelty of this study is the application of PVA and
In our earlier studies, we investigated the applicability of the combined wet milling process and optimization of the
the combined wet milling technique,20–23 and these prelimi- amount of the additive and the process parameters in order
nary studies showed that for the nanonization of Mel, PVA to produce Mel nanosuspension (particle size range of
as a stabilizer agent and an increased milling time (.50 min) 100–500 nm) without any pretreating procedure.

1568 submit your manuscript | www.dovepress.com Drug Design, Development and Therapy 2018:12
Dovepress
Dovepress Combined wet milling process to produce nanosuspension

Materials and methods &RQFHQWUDWHGSUHGLVSHUVLRQ

Materials RI0HO
±RI39$
Mel was obtained from EGIS Ltd. (Budapest, Hungary). XSWRRIZDWHU
PVA-Mowiol 4-98 (Mw ~27,000) (Sigma-Aldrich Co. LLC,
&RPELQHGZHWPLOOLQJZLWKRSWLPL]HG
St Louis, MO, USA) was used as a stabilizing agent. Reagents SURFHVVSDUDPHWHUV
were purchased also from Sigma-Aldrich. Zirconium oxide
&RQFHQWUDWHGPLOOHGGLVSHUVLRQ
(ZrO2) beds with a diameter of 0.3 mm were obtained from
Netsch (Netsch GmbH, Selb, Germany). All reagents were IROGGLOXWLRQ HOLPLQDWLRQRIWKH
PLOOLQJPHGLD
purchased from Sigma-Aldrich, Ltd. (Budapest, Hungary)
for the in vitro cell culture experiments, unless otherwise 'LOXWHGGLVSHUVLRQ
RI0HO
indicated.
RI39$
XSWRRIZDWHU
Methods 6HOHFWLRQRIVDPSOHZLWKSUHGHWHUPLQHG
Combined wet media milling VL]HUDQJH ±QP
Optimization of process parameters
1DQRVXVSHQVLRQDVVWDEOHLQWHUPHGLDWH
The samples were milled with the steel jar with 50 mL SURGXFW KROGLQJWLPHZHHNV
volume of the Retsch PM 100 planetary ball mill (Retsch
PM 100 MA, Retsch GmbH) combined with 0.3 mm ZrO2 Figure 1 Protocol of sample preparation for the optimization of PVA content.
Abbreviations: Mel, meloxicam; PVA, poly(vinyl alcohol).
beads as the grinding media. The concentrated (10% w/w)
predispersions (2 g of Mel suspended in 18 g of dispersant
medium containing PVA) were added to the ZrO2 beads. products with 1% of Mel (w/w) content. The samples were
In the first step, the effect of different amount of ZnO2 pearls selected on the basis of the particle size range (100–500 nm)
on the particle size reduction was investigated. The ratio of and the holding time (2 weeks).
the amount of predispersion and pearls (w/w) was 1:0.5, 1:1,
Investigation of the samples
1:2, and 1:4; and the milling times were 10, 30, and 50 min.
Particle size measurement
In these cases, the concentration of PVA solution was 2.5%
The volume-based particle size distribution was measured by
(w/w), and the rotation speed was 400 rpm. In the second
laser diffraction (Mastersizer S 2000, Malvern Instruments
step, design and analysis of experiments with 3 levels were
Ltd, Worcestershire, UK) with the following parameters:
used to optimize the milling time (10, 30, and 50 min) and
300RF lens; small volume dispersion unit (1,000 rpm);
the rotation speed (200, 350, and 500 rpm) as independent
refractive index for dispersed particles 1.596; and refractive
variables. The amount of the pearls was chosen on the basis
index for dispersion medium 1.330. Water was used as a
of the optimization study. The temperature of the samples
dispersant, and the obscuration was in the range of 11%–16%
was measured immediately after milling. This value did not
for all measurements. In all cases, the particle size distribu-
exceed 39°C.
tions were characterized by the d(0.1), d(0.5), and d(0.9)
Optimization of PVA concentration (where, eg, d(0.5) is the maximum particle diameter below
Various amounts of PVA (2.5%–7.5%) were applied to pre- which 50% of the sample volume exists), and the Span values
pare the concentrated predispersions. The concentration of were calculated according to equation. A high Span value
the PVA solutions was increased in the half percent range. (.1) denotes a broad particle size distribution.27
Mel content was 10% (w/w), and the optimized process
parameters were used during the milling. The degree of d (0.9) − d (0.1)
Span =
particle size reduction and particle size distribution were the d (0.5) 
main factors for the optimization of the PVA concentration
(Figure 1). Rheological measurement
The concentrated milled dispersions were filtered by a To investigate the viscosity changes during the milling
sieve with 150 μm mesh size in order to remove the pearls. process, the initial and milled PVA solutions and the con-
For the washing of the pearls, 180 g of distilled water was centrated milled dispersions were used. Rheological measure-
used. In all cases, the milled dispersions were 10-fold diluted ments were carried out with Physica MCR101 rheometer

Drug Design, Development and Therapy 2018:12 submit your manuscript | www.dovepress.com
1569
Dovepress
Bartos et al Dovepress

(Anton Paar, Graz, Austria). A concentric cylinder measuring range of 25°C–270°C. The heating rate was 10°C/min in the
device with a diameter of 17 mm was used. The flow curves presence of argon as a carrier gas with a flow rate of 10 L/h.
of the samples were determined at 25°C, where the shear rate
was increased from 0.1 to 100 L/s. The shearing time was X-ray powder diffraction analysis (XRPD)
300 sec. In this paper, viscosity values were given at 100 L/s The crystallinity state of Mel in the dried samples was evalu-
shear rate. Two parallel measurements were run. ated by XRPD. XRPD patterns were produced by a Bruker
D8 Advance diffractometer (Bruker AXS GmbH, Karlsruhe,
Zeta potential Germany) system with Cu K λI radiation (λ=1.5406 Å).
The zeta potential of the dispersions was measured using The samples were scanned at 40 kV and 40 mA from 3 to
a Malvern Zeta Nano ZS (Malvern Instruments Ltd). For 40 2θ, at a step time of 0.1 sec, and a step size of 0.010°.
the zeta potential determination, Malvern DTS 1070 folded The instrument was calibrated by using SI standard.
capillary cell was used. The diluted milled dispersions were The semiquantitative determination of Mel crystallinity
further diluted with water (25-fold) for the measurements. (Cryst. %) was performed using the total area under the
curve of 3 characteristic peaks (13.06, 14.94, and 18.61
Raman spectroscopy 2θ) of Mel. The area under the curve value of the peak of
For the investigation of Mel degradation as a function of raw material without milling (rawMel) and the dried milled
the pearl amount and milling time in the dispersion, Raman dispersions (MelD) was calculated and compared according
spectra were acquired with a Thermo Fisher DXR Dispersive the following formula:
Raman (Thermo Fisher Scientific Inc., Waltham, MA, USA)
equipped with a CCD camera and a diode laser operating
AUCMelD
at a wavelength of 532 nm. Raman measurements were Cryst. % = × 100
AUCrawMel 
carried out with a laser power of 4 and 8 mW at 25-µm slit
aperture size on a 2 µm spot size. The spectra of the indi-
vidual substances as Mel and PVA were collected using a Drug content determination
2-sec exposure time, a total of 48 scanning in the spectral The loss of weight of Mel was controlled in the milled
range of 3,300–200 cm–1 with cosmic ray and fluorescence suspension. Seventy-five milligram of the liquid products
corrections. with 0.75 mg of theoretical Mel was dissolved in 100 mL of
phosphate buffer pH 7.4±0.1. The sample was stirred with a
Morphology of the particles (scanning electron magnetic stirrer at 25°C for 24 h and then filtered (0.1 μm, Fil-
microscopy) terBio PES Syringe Filter) (Labex Ltd., Budapest, Hungary),
For the investigation of the morphology of the particles, and the concentration of the dissolved Mel was analyzed
the diluted milled dispersions were dried in a vacuum dryer spectrophotometrically (Unicam UV/VIS) (Thermo Fisher
(Binder GmbH, Tuttlingen, Germany) at 40°C in order to Scientific Inc.) at 364 nm wavelength. The investigations
obtain solid products for physicochemical investigations. were repeated 3 times.
After drying, the shape and surface characteristics of the
samples were visualized using a scanning electron micro- In vitro dissolution test
scope (Hitachi S4700, Hitachi Scientific Ltd., Tokyo, Japan). To determine the dissolution extent of Mel from disper-
The samples were sputter-coated with gold–palladium under sions, the paddle method (USP dissolution apparatus, type II
an argon atmosphere, using a gold sputter module in a high- Pharma Test, Heinburg, Germany) was used. About 750 mg
vacuum evaporator, and the samples were examined at 10 kV of the dispersion with 7.5 mg of Mel (therapeutic dose) was
and 10 mA. The air pressure was 1.3–13 MPa. filled into hard gelatin capsules within 5 sec and put promptly
into the medium. The medium contained 900 mL of artificial
Differential scanning calorimetry (DSC) gastric fluid at pH 1.2±0.1 and intestinal fluid (pH 6.8±0.1).
DSC measurements were carried out with a Mettler Toledo The paddle was rotated at 100 rpm, and sampling was
DSC 821e thermal analysis system with the STARe thermal performed up to 60 min. The Mel contents of the samples
analysis software V9.0 (Mettler Inc., Schwerzenbach, were determined using a spectrophotometer (ATI-UNICAM
Switzerland). Approximately 2–5 mg of pure Mel and PVA UV/VIS Spectrophotometer) at 362 nm (gastric juice) and
as well as dried samples were examined in the temperature 364 nm (enteric fluid). The number of parallel runs was 3.

1570 submit your manuscript | www.dovepress.com Drug Design, Development and Therapy 2018:12
Dovepress
Dovepress Combined wet milling process to produce nanosuspension

In vitro cell culture studies check the barrier integrity by an epithelial volt-ohmmeter
Human Caco-2 intestinal epithelial cell line (World Precision Instruments, Sarasota, FL, USA) combined
Caco-2 intestinal epithelial cell line was purchased from with STX-2 electrodes and was expressed relative to the
ATCC (Manassas, VA, USA) (cat. no HTB-37) and used surface area of the monolayers as Ω×cm2.
until passage 60 for the experiments. The cells were grown Caco-2 cells were seeded onto Transwell inserts (polycar-
in DMEM (Gibco, Life Technologies, Carlsbad, CA, USA) bonate membrane, 0.4 µm pore size, 1.12 cm2 surface area;
and supplemented with 10% fetal bovine serum (Pan-Biotech Corning Life Sciences, Tewksbury, MA, USA) and cultured
GmbH, Aidenbach, Germany) and 50 μg/mL gentamycin for 3 weeks.31,32 The culture medium was changed and TEER
in a humidified incubator with 5% CO2 at 37°C. All plastic was checked every second day.
surfaces were coated with 0.05% rat tail collagen in sterile For the permeability experiment, inserts were transferred
distilled water before cell seeding in culture dishes and the to 12-well plates containing 1.5 mL Ringer–Hepes buffer
medium was changed every 2 days. The stock solutions for in the lower (basal) compartments. In the upper (apical)
cell culture experiments were the following: Mel, 100 mg/ compartments, the culture medium was replaced by 0.5 mL
mL; PVA, 5% (w/w), and 3 products containing 100 mg/mL buffer containing treatment solutions of Mel, PVA, and Mel
Mel with various amounts of PVA (2.5%, 5%, or 7.5%). The formulation groups for 1 h. Permeability marker molecules
working solutions were diluted in the cell culture medium fluorescein (10 μg/mL; Mw: 376 Da) and albumin (10 mg/mL;
or Ringer–Hepes buffer and contained 1 mg/mL of Mel for Mw: 65 kDa) labeled with Evans blue (167.5 μg/mL) were
the Mel and formulation groups. The final concentrations of used for verifying the cell layer integrity. Treatment solu-
PVA were 0.025%, 0.05%, and 0.075% (w/w). tions from both compartments were collected and the Mel
level was detected using a Thermo Spectronic Helios Alpha
Cell viability measurement by impedance UV-Vis spectrophotometer (Thermo Fisher Scientific Inc.).
Impedance was measured at 10 kHz using the RTCA-SP The concentrations of the permeability marker molecules
instrument (RTCA-SP instrument, ACEA Biosciences, San of collected samples were determined by a fluorescence
Diego, CA, USA). This method is label-free, noninvasive multi-well plate reader (Fluostar Optima, BMG Labtech,
and follows cell adherence, growth, number, and viability Offenburg, Germany; for fluorescein: excitation wavelength,
real time. We have successfully tested the cellular effects 485 nm; emission wavelength, 535 nm and for Evans blue-
of peptides and pharmaceutical excipients by impedance labeled albumin: excitation wavelength, 584 nm; emission
kinetics.28–30 For background measurements, a 50 μL cell wavelength, 680 nm).
culture medium was added to the wells; then, cells were The apparent permeability coefficients (Papp) were calcu-
seeded at a density of 6×103 cells/well to 96-well plate with lated as described previously.28 Briefly, the cleared volume was
gold electrodes (E-plate 96, ACEA Biosciences) coated with calculated from the concentration difference of the tracer in
collagen. Cells were cultured for 5 days in a CO2 incubator the lower/basal compartment (Δ[C]B) after 30 min and upper/
at 37°C and monitored every 10 min until the end of experi- apical compartments at 0 h ([C]A), the volume of the lower/
ments. When cells reached the plateau phase of growth, they basal compartment (VB, 1.5 mL) and the surface area available
were treated with Mel, PVA, and Mel+PVA samples diluted for permeability (A, 1.1 cm2) using the following equation:
in a cell culture medium, and the effects were followed for
8 h. Triton X-100 detergent (1 mg/mL) was used as a refer- ∆[C ]A × VA
ence compound to induce cell toxicity. Cell index was defined Papp (cm/s) =
A × [C ]L × ∆t
as Rn-Rb at each time point of measurement, where Rn is the 
cell–electrode impedance of the well when it contains cells
and Rb is the background impedance of the well with the Determination of holding time
medium alone. Since the diluted dispersions are intermediate products, it
was necessary to specify the “holding time” of the sample
Permeability study on cell culture model through the particle size distribution. The products were
Transepithelial electrical resistance (TEER) reflects the tight- stored in sealed glass bottles at room temperature (25°C±1°C)
ness of the intercellular junctions closing the paracellular for 2 weeks. The particle size distribution of the Mel in the
cleft, and therefore reflects the overall tightness of cell layers samples was analyzed on the day of production (day 0) and
of biological barriers. TEER was measured every 2 days to after 14 days of storage.

Drug Design, Development and Therapy 2018:12 submit your manuscript | www.dovepress.com
1571
Dovepress
Bartos et al Dovepress

Statistical analyses
Data were expressed as mean±SD, and groups were com-
pared by using Student’s t-test. For the evaluation of cell
culture results, GraphPad Prism 5.0 software (GraphPad
Software Inc., San Diego, CA, USA) was used. All cul- 
ture data presented are mean±SD; values were compared 
using analysis of variance followed by Bonferroni post- 

3DUWLFOHVL]H —P
test. Differences were considered statistically significant 
when p,0.05. 

Results and discussion 

Optimization of process parameters (pearl ±


±
amount, milling time, and rotation speed) 


P

The effect of pearl amount and milling time on particle size 

US

 

G
of d(0.5) was investigated. This study was performed using 0LO 

HH

OLQ  
JW

VS
2.5% PVA solution.23 The ratio of the concentrated predisper- LP  

Q
H 

WLR
sion (2.0 g of Mel+18.0 g of PVA aqueous solution) and pearl PL  
Q 

WD
 

5R
amount was different: 1:0.5, 1:1, 1:2, and 1:4 (w/w). It can be
established that the pearl amount and the milling time have Figure 2 Three-dimensional illustration of the particle size changes during the
second factorial experimental design.
a great effect of on the d(0.5) value (Table 1). The milling
efficiency was not satisfactory in case of ratio 1:0.5, but it
increased linearly on increasing the amount of the milling efficiency of the particle size reduction was improved by
media and the milling time except in case of ratio 1:4. It was increasing the milling time and the rotation speed. Based on
found that the increase of the amount of the milling media the results, 437 rpm and 43 min are the optimal parameters
(up to 1:2 w/w or more) causes gray coloring because of the of the milling process.
high friction between the pearls and the wall of the steel jar.
Therefore, the pearl milling amount was optimized at the Optimization of PVA concentration
ratio of 1:1 (20 g of concentrated predispersion and 20 g of Influence of PVA amount on the milling effectiveness
pearls), and the milling time was investigated as an indepen- Concentrated predispersions with different PVA amounts
dent variable in the factorial experimental design. Another (2.5%–7.5%) were milled with optimized parameters
advantage of the small amount of the grinding media may (2 g of Mel+18 g of PVA aqueous solution, 20 g pearls,
be the reduction of the product loss. 437 rpm, 43 min). The results show (Figure 3) that the lower
During the factorial experimental design, the influence
of the milling time and the rotation speed on the particle size
 
reduction was investigated. The ratio of the predispersion G 
 
and the pearls was also 1:1 (w/w). Figure 2 shows that the G 
3DUWLFOHVL]H —P


6SDQ 


6SDQ

Table 1 Particle size of Mel (d[0.5]) in milled dispersion as a func­ 

tion of different pearl amounts and milling time (d[0.5] of raw Mel 

was 34.260±4.860 µm) 

Ratio of predispersion and pearl amount (w/w)  
1:0.5 1:1 1:2 1:4  
          
Particle size (µm)
39$FRQFHQWUDWLRQ 
10 min 4.015±0.06 2.426±0.029 2.383±0.016 0.149±0.03
30 min 0.293±0.008 0.145±0.007 0.190±0.003 0.137±0.006 Figure 3 Particle size reduction effectiveness according to PVA concentrations and
the Span values for the demonstration of particle size distribution.
50 min 0.202±0.003 0.140±0.004 0.140±0.002 0.130±0.004
Notes: d(0.5) and d(0.9) is the maximum particle diameter below which 50% and
Note: d(0.5) is the maximum particle diameter below which 50% of the sample 90%, respectively, of the sample volume exists. Span is the calculated value for a
volume exists. broad particle size distribution.
Abbreviations: Mel, meloxicam; w/w, weight/weight (concentration). Abbreviation: PVA, poly(vinyl alcohol).

1572 submit your manuscript | www.dovepress.com Drug Design, Development and Therapy 2018:12
Dovepress
Dovepress Combined wet milling process to produce nanosuspension

concentration of PVA (from 2.5% to 3.5%) was not suitable Mel amount in the dispersions increased the viscosity of
to reach the nanosize range (100–500 nm). A higher concen- the systems and did not change the flow behavior – they
tration of PVA (6.0%–6.5%) also resulted in unsatisfactory remained Newtonian. Adding Mel to the polymer solutions,
milling effectiveness. Over 6.5% a robust protecting layer the smallest shift in the viscosity could be detected in case
was probably formed on the solid particles, and nanoniza- of the highest polymer concentration. This can be explained
tion was not possible. The curve of the Span follows the by the more considerable viscosity changing effect of the
different particle size distribution of the samples. The best polymer concentration than that of the Mel particles. It can
particle size distribution was measured in the range of 4.0% be concluded that the mechanical influence did not change
and 5.5% of PVA. Based on the results, the concentrated the viscosity of the polymer solutions, and therefore, the
milled dispersion containing 5.0% of PVA was selected as structure of the PVA chains did not change.
the optimized PVA amount.
Zeta potential changes
Influence of PVA amount on the physicochemical To determine the electrokinetic changes of the diluted dis-
properties of milled dispersions persions, the zeta potential of the samples was measured.
After the optimization of the wet milling process and the PVA The results show that the increase of the PVA amount
amount, in order to understand the influence of the amount of decreases the zeta potential in comparison to the sample
PVA on the physicochemical properties of the samples, the without PVA (Table 3). The main reason for the zeta potential
milled dispersions with 3 different concentrations of PVA reduction can be linked to the nonionic polymer adsorption
were investigated. These were the following: 2.5%, 5.0% on the surface of the solid particles, which causes a decrease
(as optimized), and 7.5% of PVA. of the diffuse layer charge. A greater zeta potential-lowering
effect can be observed between 0% and 0.25% of PVA than
Viscosity changes during the milling between 0.25% and 0.50. At 0.50% of PVA, the surface of
In order to exclude the viscosity changes because of the high particles is saturated by the PVA chain; therefore, the change
mechanical forces during the milling, the viscosity of 3 raw of zeta potential is smaller. In case of a higher concentration
PVA solutions (2.5%, 5%, and 7.5%) was investigated before of PVA (.0.50%), the steric hindrance stabilizes the system
the milling process, after the process (437 rpm, 43 min), but hampers the disintegration/abrasion of the particles.
and after the milling with the addition of Mel. The polymer
solutions showed Newtonian flow behavior as their shear Investigation of the diluted dispersions to select
viscosity was independent of the applied shear rate. The nanosuspension as an intermediate product
viscosity of the polymer solutions increased with increas- Raman investigation
ing the polymer concentration (Table 2). There were no Raman spectrograms and chemical maps of raw materials
remarkable differences between the viscosity of the polymer and products are presented in Figures 4 and 5. The individual
solution before and after the milling procedure, which may spectrum of Mel (a) shows that the absorption peaks are con-
indicate there are no changes in the polymer structure. 10% centrated in the region from 1,600 to 1,000 cm-1 (fingerprint
region), whereas the individual spectra of PVA (b and c)
Table 2 Viscosity values (η) of the raw PVA solutions before
milling (PVA %) and after milling (PVA % milled), and the PVA
solutions after milling with the addition of Mel (Mel PVA % Table 3 Zeta potential values of the diluted dispersions as a
milled) function of the PVA concentration and particle size distribution
of Mel (±SD)
η (mPa⋅s) SD (mPa⋅s)
Samples Particle size distribution Zeta
PVA 2.5% 3.14 0.02
d(0.1) d(0.5) d(0.9) potential
PVA 5.0% 7.38 0.02
(mV)
PVA 7.5% 20.55 0.07 Particle size (μm)
PVA 2.5% milled 3.12 0.27 Mel PVA 0% 2.508±1.100 5.762±2.700 135.640±12.900 -30.7
PVA 5.0% milled 7.09 0.02 Mel PVA 0.25% 0.070±0.001 0.150±0.009 1.478±0.0400 -20.9
PVA 7.5% milled 20.05 0.07 Mel PVA 0.50% 0.067±0.001 0.130±0.005 0.371±0.010 -16.1
Mel PVA 2.5% milled 4.42 0.07 Mel PVA 0.75% 1.235±0.006 2.611±0.018 5.560±0.070 -15.7
Mel PVA 5.0% milled 8.46 0.01
Note: d(0.1), d(0.5), d(0.9) is the maximum particle diameter below which 10%,
Mel PVA 7.5% milled 21.65 0.07
50%, and 90%, respectively, of the sample volume exists.
Abbreviations: Mel, meloxicam; PVA, poly(vinyl alcohol); η, viscosity. Abbreviations: Mel, meloxicam; PVA, poly(vinyl alcohol).

Drug Design, Development and Therapy 2018:12 submit your manuscript | www.dovepress.com
1573
Dovepress
Bartos et al Dovepress

$ %

I
5HODWLYH5DPDQLQWHQVLW\

5HODWLYH5DPDQLQWHQVLW\
H
F

G
E
E

D D
            
5DPDQVKLIW FP± 5DPDQVKLIW FP±

Figure 4 Investigation with Raman spectroscopy: (A) (a) spectrum of raw Mel, (b) spectrum of raw PVA, (c) spectrum of raw PVA (0.50%) containing solution.
(B) Comparison study of raw materials (Mel and PVA) and the dispersions, (a) spectrum of raw Mel, (b) spectrum of raw PVA, (d) spectrum of dispersion containing 1% Mel
and 0.25% PVA, (e) spectrum of dispersion containing 1% Mel and 0.50% PVA, (f) spectrum of dispersion containing 1% Mel and 0.75% PVA.
Abbreviations: Mel, meloxicam; PVA, poly(vinyl alcohol).

show just 1 characteristic and extended peak from 3,000 of Enol which is presented when Mel is in an aqueous
to 2,800 cm-1. This is the CH stretching vibration region medium, but it is disappeared in dried form, so the change
that results a medium-to-strong intensity in Raman spectra. is reversible. The chemical map of dispersion profiled to
The spectrograms of diluted dispersions (d, e, and f) show Mel spectrogram shows homogenous distribution of Mel.
the same characteristic region with the spectra of Mel – there It can be summarized that there is no chemical degradation
are no detectable differences among them. Two peaks (one or interaction in dispersion which could be detectable with
more characteristic in 2,437 cm-1 and another smaller one Raman technique.
in 482 cm-1) appear in the spectra of PVA-containing dis-
persions and in spectra of aqueous Mel dispersion as well. Morphology of particles (scanning electron microscopy)
The chemical mapping of dried dispersions profiled to this In order to investigate the effect of milling and PVA amount
peak in 2,437 cm-1 shows that this peak cannot detected on the morphology of milled Mel, the water was evaporated
in this map. This peak can show a dissociated -OH group from dispersion and the dried samples were characterized.

$ %
5HODWLYH5DPDQLQWHQVLW\

&KHPLFDOPDSRI
 PHOR[LFDP



   

H
G &KHPLFDOPDSRI
F SHDNLQFP±

E
D
              
    
5DPDQVKLIW FP±
Figure 5 Investigation with Raman spectroscopy: (A) Comparing raw Mel and aqueous dispersion of Mel- and PVA-containing dispersions, (a) spectrum of raw Mel, (b)
spectrum of aqueous 1% Mel-containing dispersion without PVA, (c) spectrum of dispersion containing 1% Mel and 0.25% PVA, (d) spectrum of dispersion containing 1% Mel
and 0.50% PVA, (e) spectrum of dispersion containing 1% Mel and 0.75% PVA. (B) Chemical mapping of Mel-containing dispersion (1% Mel and 0.50% PVA) and chemical
mapping of its dried form profiled to peak in 2,437 cm-1.
Abbreviations: Mel, meloxicam; PVA, poly(vinyl alcohol).

1574 submit your manuscript | www.dovepress.com Drug Design, Development and Therapy 2018:12
Dovepress
Dovepress Combined wet milling process to produce nanosuspension

Figure 6 shows the raw Mel in physical mixture (Mel peak is the melting temperature.33 The DSC curves of the
PVA 0.50% PM) which has an irregular shape with dried products exhibited lower melting points of Mel than
34.260±4.860 µm as average particle size. In this PM, the that of raw Mel. It is connected to the premelting of PVA,
PVA particles with size ,6 µm are located on the surface which induces the earlier melting of Mel in proportion to
of Mel crystals. After drying of the aqueous dispersions, the amount of PVA and decreases the crystallinity degree of
the polymer formed non-coherent and coherent think film Mel. Therefore, the XRPD investigation was used to check
with Mel particles. In case of 0.25% containing PVA (Mel the amorphization of the active agent.
PVA 0.25%), the particle size of Mel has decreased, but The XRPD investigations justified the change of crystal-
aggregation of the fragmented particles can be observed. linity degree of Mel in the dispersions with different PVA
Mel particles are in homogeneous disperse distribution in amounts. Figure 8 shows the fingerprints of raw materials
sample (Mel PVA 0.50%), resulting in ,500 nm (aver- (Mel and PVA) and the dried dispersions. The samples
age particle size: 0.130±0.005 µm). High concentration show the characteristic peaks of Mel at 2θ values: 13.06°,
of PVA (Mel PVA 0.75%) helped the recrystallization of 14.94°, and 18.61°. It was found that the PVA content of
Mel (nanocrystals with smooth surface) thanks to increased the samples fundamentally influenced the decrease of the
solubility of Mel in aqueous PVA solution. Otherwise, the crystallinity degree of Mel. As it was established earlier
sample shows heterogeneous disperse system with nano- and in this study, low (2.5%) and high (7.5%) concentrations
microparticles. of PVA in the milled dispersion did not result in suitable
milling efficiency. In this case, the crystallinity degree
Crystallinity characterization of Mel in the dried of Mel was 75.82% at low PVA content (2.5%), and it
dispersions (DSC and XRPD) decreased to 51.44% at high concentration of PVA (7.5%).
DSC was used to investigate the melting of raw Mel and raw These results are connected to the milling effectiveness.
PVA and the dried samples (Figure 7). The DSC curve of In this study, the 5.0% PVA-containing milled dispersions
the raw Mel revealed a sharp endothermic peak at 268.66°C, showed smaller crystallinity (13.43%) and the highest
reflecting its melting point and an instantly following exo- milling efficiency.
thermic peak at 278.09°C can be observed. The DSC curve
of raw PVA as a semi-crystalline polymer has 2 endothermic Drug content determination
peaks at 169.51°C and at 222.74°C. The first peak of PVA The spectrophotometrically measured drug content of the sam-
signifies a particular decrystallization of PVA and the second ples was less than the theoretical value (7.5 mg Mel). The drug

N9PPî6( 8 —P

0HO39$30

N9PPNî6( 8 —P N9PPNî6( 8 —P N9PPNî6( 8 —P

0HO39$ 0HO39$ 0HO39$

Figure 6 SEM pictures of physical mixture (Mel PVA 0.25% PM) and different PVA concentrations containing dried dispersions.
Abbreviations: SEM, scanning electron microscopy; Mel, meloxicam; PVA, poly(vinyl alcohol); PM, physical mixture.

Drug Design, Development and Therapy 2018:12 submit your manuscript | www.dovepress.com
1575
Dovepress
Bartos et al Dovepress

P:

             °&

              PLQ

0HO 39$ 0HO39$

0HO39$ 0HO39$

Figure 7 DSC curves of raw Mel and PVA and dried dispersions with different PVA concentrations.
Abbreviations: DSC, differential scanning calorimetry; Mel, meloxicam; PVA, poly(vinyl alcohol).

contents, converted to 750 mg dispersion quantity, are

as follows: 7.36 mg for aqueous sample (without PVA),
7.19 mg for 0.25% PVA, 7.12 mg for 0.50% PVA, and
7.23 mg for 0.75%. It can be stated that the washing method
of the pearls resulted in greater loss of Mel in PVA-containing
dispersion (3.60%–5.06%) than in case of the aqueous sample
without PVA (1.80%). The amount of PVA did not signifi-
cantly affect the loss of weight of Mel.
/LQ FRXQWV

In vitro dissolution study

In case of each sample, the liberation from the hard gela-
tin capsules occurred within 2 sec. Mel has a weak acidic
character (pKa 4.8), and therefore, its solubility in gastric
juice (pH=1.2) is very poor (1.6±0.2 mg/L, at 37°C). In this
medium, the greatest dissolved amount of Mel with 0.25%
and 0.50% PVA content was maximum, 37%, within 20 min
(Figure 9A). This result is due to the wetting effect of PVA
(0.25% and 0.50%), which could increase the solubility
of Mel, and the reduction of the particle size of Mel in the
    
dispersions. In contrast, a higher amount of PVA (0.75%)
θVFDOH hinders the dissolution because a thicker polymer layer is
formed on the Mel particles.
0HO 39$ 0HO39$
0HO 0HO39$ In intestinal fluid (pH=6.8), the dissolved amount was
higher in all cases because of the better solubility of Mel
Figure 8 XRPD diffractograms of the Mel, PVA, and the dried dispersion.
Abbreviations: XRPD, X-ray powder diffraction; Mel, meloxicam; PVA, poly(vinyl
(0.272±0.001 mg/mL, at 37°C). Figure 9B shows that the
alcohol). concentration of PVA influences the amount of dissolved

1576 submit your manuscript | www.dovepress.com Drug Design, Development and Therapy 2018:12
Dovepress
Dovepress Combined wet milling process to produce nanosuspension

$  % 





 

'LVVROYHG0HO 

'LVVROYHG0HO 
 


 

 


 

 





 
             
7LPH PLQ 7LPH PLQ

39$ 39$ 39$ 39$

Figure 9 In vitro dissolution curves of Mel in artificial gastric juice (A) and intestinal juice (B).
Abbreviations: Mel, meloxicam; PVA, poly(vinyl alcohol).

Mel as in the case of gastric juice, but its extent is bigger. comparison, cells treated with the detergent Triton X-100
In this study, the dispersion containing 0.50% of PVA had were lysed and a 100% toxicity was measured. The curves
the most satisfying dissolution behavior from among the show the kinetics of the cellular effects of treatment solu-
4 samples. It is followed by the dispersions containing 0.25% tions (Figure 10A), whereas the columns show the effect
and 0.75% of PVA, and finally the dispersion without PVA. of Mel, PVA and Mel formulations at the 8-h time point
The results justify the need of the polymer (PVA) and the (Figure 10B).
correct choice of its quantity.
Permeability study on intestinal barrier model
Cell viability assay Caco-2 monolayers showed high TEER values (2,660±181
Impedance measurement, as a sensitive method to detect Ω×cm2, n=20) before permeability experiments, indicating
cellular effects, did not show significant cell damage after tight barrier properties. The average apparent permeability
treatments with Mel, PVA, and Mel formulation groups, as coefficients of marker molecules were also low (fluores-
reflected by unchanged cell index values (Figure 10). As a cein: 0.81±0.13×10-6 cm/s; albumin: 0.08±0.03×10-6 cm/s),

$  % 


1RUPDOL]HGFHOOLQGH[

1RUPDOL]HGFHOOLQGH[




 


 
&RQWURO 39$
 39$0HO 0HO
39$0HO 7; 
 39$0HO
 
     &RQWURO    39$ 0HO 7;
7LPH K RI39$LQ
0HOIRUPXODWLRQV
Figure 10 Cell viability kinetics (A) and results 8 h after treatment (B) in Caco-2 intestinal epithelial cells with Mel, PVA, and formulations measured by impedance.
Notes: Values are presented as mean±SD, n=6–12. Statistical analysis: ANOVA followed by Dunett’s test. Statistically significant differences are: ***p,0.001, compared to
control group.
Abbreviations: Mel, meloxicam; PVA, poly(vinyl alcohol); TX-100, Triton X-100; ANOVA, analysis of variance.

Drug Design, Development and Therapy 2018:12 submit your manuscript | www.dovepress.com
1577
Dovepress
Bartos et al Dovepress

  Conclusion

 Based on our results, it can be stated that the combination
3DSSRIPHOR[LFDP

of planetary ball mill and the pearl milling technology is

î±FPV

 a new possibility to nanonize the Mel as a nonsteroidal

anti-inflammatory agent to reach the particle size range of
 100–500 nm. This combinative wet technology resulted
in significant particle size reduction without premilling
 (pretreatment of raw agent). In addition to the process
parameters (the pearl amount, the milling time, and the

0HO   
rotation speed of jar), the amount of PVA was also a critical
parameter because it affected the milling effectiveness, the
RI39$LQ
0HOIRUPXODWLRQV particle size distribution, and the crystallinity of Mel. The
different concentrations of PVA in the aqueous dispersion
Figure 11 Evaluation of permeability of Mel across Caco-2 epithelial cell layers
treated with Mel and optimized Mel PVA formulations for 1 h. also influenced the viscosity and the electrokinetic property
Notes: Values are presented as mean±SD, n=4. Statistical analysis: ANOVA of the particle, according the DLVO theory,34 and thus, the
followed by Bonferroni posttest. Statistically significant differences are: ***p,0.001,
compared to control group; ###p,0.001 compared to the indicated columns. stability of the dispersions.
Abbreviations: Mel, meloxicam; PVA, poly(vinyl alcohol); Papp, apparent permeability
coefficients.
It was found that the milling effectiveness of low con-
centration of PVA (,4%) was not satisfactory, because the
crushing/breaking effect of the pearls was less prevalent. High
in agreement with the TEER values and the formation of concentration of PVA (.5%) also resulted in unsatisfactory
tight cell layers. milling effectiveness because of the formation of polymer
The permeability of Mel suspension and Mel formulations layer on surface of particles, which protects the particles from
was significantly higher than that of marker molecules. The the fragmentation. In connection with this, the crystallinity
penetration of Mel from the 3 investigated products across of Mel decreased with the increase of milling effectiveness,
cell layers was significantly increased as compared to Mel which plays an important role in the fast drug release.
suspension. From all the tested samples, the Papp value of In this work, the combined wet milling process was also
Mel was the highest in the formulation containing 0.05% used successfully to prepare Mel-containing nanosuspen-
PVA (Figure 11). sion as an intermediate product to design the final dosage
form(s) for per oral administration. The optimized process
Holding time determination parameters (1:1 ratio of predispersion and pearls, 437 rpm,
Since the investigated dispersions are intermediate products, and 43 min) resulted in 200-fold particle size reduction
the change of particle size distribution and the crystallinity of Mel. Considering the effectiveness of milling, 5% PVA
index are very important during storage. In general, the time was proved to be an optimal quantity to meet the expected
period before the dispersion used for the preparation of differ- value (100–500 nm). The optimized nanosuspension (1%
ent dosage forms is 1 or 2 h, or possibly longer. The measure- Mel and 0.50% PVA) as an intermediate product showed a
ments have proven that 0.50% of PVA-containing dispersion stable system with 2 weeks of holding time.
had no significant changes in the particle size, particle The human Caco-2 cell culture studies justified that the
size distribution, and crystallinity up to 2 weeks (Table 4). penetration of Mel from different PVA-containing products
was significantly increased as compared to Mel suspension
Table 4 Particle size distribution changes (µm) during the stability
without toxic effects. From all the tested samples, the Papp
testing value of Mel was the highest in the investigated sample
Day 0 2 weeks
containing 0.05% PVA, which belongs to the optimized
d(0.1) d(0.5) d(0.9) d(0.1) d(0.5) d(0.9) nanosuspension.
Mel PVA 0.25% 0.070 0.150 1.478 0.080 0.152 2.073 Based on the above results, the milling process and the
Mel PVA 0.50% 0.067 0.136 0.371 0.068 0.140 0.427 composition of the nanosuspension can be recommended
Mel PVA 0.75% 1.207 2.232 5.224 0.244 2.611 5.560 to produce innovative dosage forms (eg, capsule) with fast
Note: d(0.1), d(0.5), d(0.9) is the maximum particle diameter below which 10%,
dissolution rate of Mel. During the development of dosage
50%, and 90%, respectively, of the sample volume exists.
Abbreviations: Mel, meloxicam; PVA, poly(vinyl alcohol). form(s), the stabilization of the amorphous Mel particle may

1578 submit your manuscript | www.dovepress.com Drug Design, Development and Therapy 2018:12
Dovepress
Dovepress Combined wet milling process to produce nanosuspension

also be an important viewpoint to ensure the dissolution 16. Johansson A. Correlation Between Process Parameters and Milling
Efficiency [dissertation]. Uppsala: Uppsala University; 2012.
profile of the active ingredient. 17. Web reference 6 Retsch®. Available from: http://pdf.directindustry.
com/pdf/retsch/the-sample-high-energy-ballmills/19308-518973.html.
Author contribution Accessed on March 20, 2016.
18. Van Eerdenbrugh B, Froyen L, Van Humbeeck J, Martensc JA,
All authors contributed toward data analysis, drafting and Augustijnsa P, Van Den Mootera G. Alternative matrix formers for
critically revising the paper and agree to be accountable for nanosuspension solidification: dissolution performance and X-ray
microanalysis as an evaluation tool for powder dispersion. Eur J Pharm
all aspects of the work. Sci. 2008;35(4):344–353.
19. Van Eerdenbrugh B, Froyen L, Van Humbeeck J, Martens JA, Augustijns P,
Van den Mooter G. Drying of crystalline drug nanosuspensions – the
Acknowledgment importance of surface hydrophobicity on dissolution behavior upon
This work was supported by Gedeon Richter Ltd – GINOP redispersion. Eur J Pharm Sci. 2008;35(1–2):127–135.
20. Mártha C, Kürti L, Farkas G, et al. Effects of polymers on the crys-
project (2.2.1-15-2016-00007).
tallinity of nanonized meloxicam during a co-grinding process. Eur
Polym J. 2013;49(9):2426–2432.
Disclosure 21. Bartos C, Ambrus R, Sipos P, et al. Study of sodium hyaluronate-based
intranasal formulations containing micro- or nanosized meloxicam
The authors report no conflicts of interest in this work. particles. Int J Pharm. 2015;491(1–2):198–207.
22. Gieszinger P, Csóka I, Pallagi E, et al. Preliminary study of nanonized
lamotrigine containing products for nasal powder formulation. Drug
References Des Devel Ther. 2017;11:2453–2466.
1. Andric L, Terzic A, Timovic-Pavlovic Z, Trumic M, Petrov M, Pavlovic L. 23. Bartos C, Szabó-Révész P, Bartos C, Katona G, Jójárt-Laczkovich O,
A kinetic study of micronization grinding of dry mica in a planetary Ambrus R. The effect of an optimized wet milling technology on the
ball mill. Adv Mater Sci Eng. 2013;2013:543857. crystallinity, morphology and dissolution properties of micro- and
2. Szafraniec J, Antosika A, Knapik-Kowalczuk J, et al. Planetary ball nanonized meloxicam. Molecules. 2016;21(4):507–518.
milling and supercritical fluid technology as a way to enhance dissolu- 24. Chibowski S, Paszkiewicz M, Krupa M. Investigation of the influence
tion of bicalutamide. Int J Pharm. 2017;533(2):470–479. of the polyvinyl alcohol adsorption on the electrical properties of Al2O3
3. Liu P, Rong X, Laru J, et al. Nanosuspensions of poorly soluble drugs: solution interface, thickness of the adsorption layers of PVA. Powder
preparation and development by wet milling. Int J Pharm. 2011; Technol. 2000;107(3):251–255.
411(1–2):215–222. 25. Vastag M, Keseru GM. Current in vitro and in silico models of blood-
4. Ghosh I, Schenck D, Bose S, Ruegger C. Optimization of formulation brain barrier penetration: a practical view. Curr Opin Drug Discov
and process parameters for the production of nanosuspension by wet Devel. 2009;12(1):115–124.
media milling technique: effect of vitamin E TPGS and nanocrystal par- 26. Hellinger E, Bakk ML, Pocza P, Tihanyi K, Vastag M. Drug penetration
ticle size on oral absorption. Eur J Pharm Sci. 2012;47(4):718–728. model of vinblastine-treated Caco-2 cultures. Eur J Pharm Sci. 2010;
5. George M, Ghosh I. Identifying the correlation between drug/stabilizer 41(1):96–106.
properties and critical quality attributes (CQAs) of nanosuspension for- 27. Li Q, Wang K, Zheng Y, Ruan M, Mei X, Lin B. Experimental research
mulation prepared by wet media milling technology. Eur J Pharm Sci. of particle size and size dispersity on the explosibility characteristics
2013;48(1–2):142–152. of coal dust. Powder Technol. 2016;292:290–297.
6. Sushant S, Archana K. Methods of size reduction and factors affecting 28. Bocsik A, Walter FR, Gyebrovszki A, et al. Reversible opening of
size reduction in pharmaceutics. Int Res J Pharm. 2013;4(8):57–64. intercellular junctions of intestinal epithelial and brain endothelial
7. Bhakay A, Merwade M, Bilgili E, Dave RN. Novel aspects of wet milling cells with tight junction modulator peptides. J Pharm Sci. 2016;105(2):
for the production of microsuspensions and nanosuspensions of poorly 754–765.
water-soluble drugs. Drug Dev Ind Pharm. 2011;37(8):963–976. 29. Kiss L, Walter FR, Bocsik A, et al. Kinetic analysis of the toxicity of
8. Rasenack N, Müller BW. Micron-size drug particles: common and novel pharmaceutical excipients Cremophor EL and RH40 on endothelial and
micronization techniques. Pharm Dev Technol. 2004;9(1):1–13. epithelial cells. J Pharm Sci. 2013;102(4):1173–1181.
9. Bilgili E, Hamey R, Scarlett B. Nano-milling of pigment agglomerates 30. Kürti L, Veszelka S, Bocsik A, et al. The effect of sucrose esters on
using a wet stirred media mill: elucidation of the kinetics and breakage a culture model of the nasal barrier. Toxicol In Vitro. 2012;26(3):
mechanisms. Chem Eng Sci. 2006;61(1):149–157. 445–454.
10. Loh ZH, Samanta AK, Heng PWS. Overview of milling techniques for 31. Hellinger E, Veszelka S, Tóth AE, et al. Comparison of brain capillary
improving the solubility of poorly water-soluble drugs. Asian J Pharm endothelial cell-based and epithelial (MDCK-MDR1, Caco-2, and VB-
Sci. 2015;10(4):255–274. Caco-2) cell-based surrogate blood-brain barrier penetration models.
11. Pawar VK, Singh Y, Meher JG, Gupta S, Chourasia MK. Engineered Eur J Pharm Biopharm. 2012;82(2):340–351.
nanocrystal technology: in-vivo fate, targeting and applications in drug 32. Kiss L, Hellinger É, Pilbat AM, et al. Sucrose esters increase drug pen-
delivery. J Control Release. 2014;183:51–66. etration, but do not inhibit p-glycoprotein in caco-2 intestinal epithelial
12. Azad AM, Afolabi A, Patel N, Dave R, Bilgili E. Preparation of stable cells. J Pharm Sci. 2014;103(10):3107–3119.
colloidal suspensions of superdisintegrants via wet stirred media mill- 33. Tretinnikov ON, Zagorskaya SA. Determination of the degree of crystal-
ing. Particuology. 2014;14:76–82. linity of poly(vinyl alcohol) by FTIR spectroscopy. J Appl Spectrosc.
13. Patravale VB, Date AA, Kulkarni RM. Nanosuspensions: a promising 2012;79(4):521–526.
drug delivery strategy. J Pharm Pharmacol. 2004;56(7):827–840. 34. Derjaguin B, Landau L. Theory of the stability of strongly charged
14. Peltonen L, Hirvonen J. Pharmaceutical nanocrystals by nanomilling: lyophobic sols and of the adhesion of strongly charged particles in
critical process parameters, particle fracturing and stabilization methods. solutions of electrolytes. Acta Physicochim URSS. 1941;14:633–662.
J Pharm Pharmacol. 2010;62(11):1569–1579.
15. Chen H, Khemtong C, Yang X, Chang X, Gao J. Nanonization strate-
gies for poorly water-soluble drugs. Drug Discov Today. 2011;16(7–8):
354–360.

Drug Design, Development and Therapy 2018:12 submit your manuscript | www.dovepress.com
1579
Dovepress
Bartos et al Dovepress

Drug Design, Development and Therapy Dovepress

Publish your work in this journal
Drug Design, Development and Therapy is an international, peer- has also been accepted for indexing on PubMed Central. The manu-
reviewed open-access journal that spans the spectrum of drug design script management system is completely online and includes a very
and development through to clinical applications. Clinical outcomes, quick and fair peer-review system, which is all easy to use. Visit
patient safety, and programs for the development and effective, safe, http://www.dovepress.com/testimonials.php to read real quotes from
and sustained use of medicines are the features of the journal, which published authors.
Submit your manuscript here: http://www.dovepress.com/drug-design-development-and-therapy-journal

1580 submit your manuscript | www.dovepress.com Drug Design, Development and Therapy 2018:12
Dovepress