European Journal of Pharmaceutics and Biopharmaceutics 81 (2012) 334–338

Contents lists available at SciVerse ScienceDirect

European Journal of Pharmaceutics and Biopharmaceutics

journal homepage: www.elsevier.com/locate/ejpb

Research paper

A novel liquefied gas based oral controlled release drug delivery system for
liquid drug formulations
Dorota Haznar-Garbacz a,b,⇑, Grzegorz Garbacz b, Friederike Eisenächer c, Sandra Klein b,
Werner Weitschies b
a
Department of Pharmaceutical Technology, Wroclaw Medical University, Wroclaw, Poland
b
Institute of Pharmacy, University of Greifswald, Greifswald, Germany
c
Institute of Pharmacy, University of Halle, Halle (Saale), Germany

a r t i c l e i n f o a b s t r a c t

Article history: A novel liquefied gas based drug delivery system for the oral delivery of liquid and semi-solid drug for-
Received 5 December 2011 mulations is presented. The capsule-shaped system is equipped with a capillary as an element controlling
Accepted in revised form 27 February 2012 the release rate. The delivery mechanism is based on a constant vapor pressure produced by isopentane
Available online 8 March 2012
as a low-boiling liquefied gas. The liquid drug valproic acid (VA) was used as a model compound. The vis-
cosity was increased by the addition of povidone (PVP). The VA–PVP gel exhibited pseudoplastic rheolog-
Keywords: ical properties, the shear rate was above 0.1 s 1, similar to a Newtonian liquid. The gels tested in the gas
Oral drug delivery system
based delivery system provided near-zero-order release kinetics. The longest delivery time was up to ca.
Liquefied gas based drug delivery system
Controlled delivery of liquids
8 h. The system is characterized by high flexibility of the delivery rate, which can be achieved by adjust-
Controlled delivery of semi-solids ing system parameters such as the diameter and length of the capillary, the vapor pressure of the propel-
Extended release lant and the viscosity of the drug formulation.
Ó 2012 Elsevier B.V. All rights reserved.

1. Introduction to mechanical stress events of physiological intensity [6]. Typically,

Oral extended release (ER) delivery systems are typically ap-
plied for two main reasons: (1) a reduction in the administration
frequency and/or (2) the avoidance of fluctuations in drug plasma
levels, which are often associated with unwanted side effects.
However, depending on the technology of the ER system, physio-
logical factors may influence the individual drug plasma profiles
(the distribution of intestinal liquids and their properties, the
mechanical forces during gastro-intestinal (GI) transit) [1–3]. The
impact of the GI transit conditions may result in substantial fluctu-
ations in the drug delivery profiles and in extreme cases in a loss of
the controlling mechanism, resulting in dose-dumping [4,5]. This
disadvantage of matrix-based or coated drug delivery systems
might be overcome by using oral osmotically driven systems
(OODSs). The release profiles of such formulations are typically
considerably independent of the composition and the volume of
liquids available in the gastrointestinal tract. Recently, it was also
shown that conventional OODS (OROS) are not prone (susceptible)
⇑ Corresponding author. Department of Pharmaceutical Technology, Wroclaw
Medical University, Szewska 38, 50-139 Wroclaw, Poland. Tel.: +49 3834515490/
+48 503447806; fax: +48 717840317.
E-mail addresses: dorota.haznar@wp.pl (D. Haznar-Garbacz), grzegorz.garbacz
@uni-greifswald.de (G. Garbacz), friederike.eisenaecher@pharmazie.uni-halle.de
(F. Eisenächer), sandra.klein@uni-greifswald.de (S. Klein), werner.weitschies@
uni-greifswald.de (W. Weitschies).
OODS are limited to solid APIs and excipients [7]. Until now, only
the so-called L-OROS was proposed for the controlled oral delivery
of non-aqueous liquids or semi-solid API formulations [8].
The novel delivery system described here has been designed for
the controlled oral delivery of hydro- and lipophilic liquids and
semi-solids, including drug suspensions, emulsions, microemul-
sions and self-emulsifying as well as self-microemulsifying formu-
lations. The system has been briefly introduced elsewhere [9].
The working principle of the system is based on the nearly con-
stant temperature along the GI tract [10]. The system utilizes a low
boiling liquid as a propellant for building up a constant pressure,
which is the driving force for the drug delivery process. The system
contains a capillary as an element for release rate control. The
delivery mechanism of the novel system can be described by the
Hagen–Poiseuille law. According to this law, under defined system
parameters (propellant type, capillary properties, formulation vis-
cosity) and temperature, a constant capillary flow and conse-
quently a constant delivery rate may be achieved.
Valproic acid (2-propylpentanoic acid, VA) is one of the natural
fatty acids of Valeriana officinalis. The mechanism of action is
mainly associated with an increased activity of c-amino butyric
acid (GABA) in the brain [11,12]. In its pure form, valproic acid is
a viscous, transparent, colorless or bright yellow liquid that is
slightly soluble in water, but well soluble in organic solvents.
Although the drug was introduced some 40 years ago, it is still
one of the most frequently-used drugs for all types of seizures and

0939-6411/$ - see front matter Ó 2012 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.ejpb.2012.02.015
 D. Haznar-Garbacz et al. / European Journal of Pharmaceutics and Biopharmaceutics 81 (2012) 334–338 335

epilepsy syndromes. Valproic acid is also used in the treatment of

depression, migraine and schizophrenia and, as recently reported,
in the therapy of certain types of cancer [12,13].
VA has a high oral bioavailability of about 80–100%. It is typically
administered in form of ER formulations as they are associated with
decrease in tremor during daily activities [14]. Furthermore, it has
been described that many of the side effects and adverse effects
of valproic acid depend on the drug plasma concentration [15,16].
Daily doses of VA are in the range of 30 mg/kg body weight (BW)
for adults and 20–25 mg/kg BW for children. In some diseases even
higher doses are applied [11,17]. Currently, there are extended re-
lease formulations of solid valproic acid derivatives available, for
example, different salts and complexes [18,19]. However, an ER sys-
tem for zero-order release of free valproic acid or preparations
thereof has not been described before.
The objective of the work was to develop and characterize a no-
vel liquefied gas based drug delivery system. Valproic acid was se-
lected as a model drug substance for liquid drugs and formulations
thereof due to its physical and chemical properties, the wide dos-
ing range as well as delivery kinetics.

2. Materials and methods

2.1. Construction of the drug delivery system

The novel drug delivery device was manufactured from poly-

methacrylate polymer (PlexiglasÒ), a robust, resistant and inert
polymer. Its shape and size correspond to a capsule of standard
size 000. Schematic representation is given in Fig. 1. The device
consists of two cylindrical parts equipped with leak-proof thread
enabling a tight mounting of the components into a capsule-like
drug delivery device. The main part is a cylindrical reservoir con-
taining the liquid or semi-solid drug formulation {4}. The second
and smaller part contains the propellant, which is preferably not
miscible with the drug-containing vehicle {9}. The walls of the
drug reservoir are built of two layers {1,2} grooved in such a way
that a spiral capillary of constant pitch (pitch 3 mm, ø 0.5 mm,
l = 32 mm) is formed {3}. During release, the liquid is emptied by
the propellant pressure via the capillary as the main flow control-
ling element. Twenty-five microliters of isopentane with a boiling Fig. 1. A schematic representation of the liquid gas based drug delivery system (all
point of 27.5 °C was used as a propellant, leading to a pressure of dimensions are given in mm). 1. external wall, 2. internal wall, 3. capillary, 4. drug
about 1000 mm Hg at 37 °C [20]. formulation, 5. liquefied gas, 6. saturated vapor, 7. output gap, 8. gasket, and 9. cap.
The filling of the device with the drug containing vehicle proce-
eded manually by using a 10 mL plastic luer-lock syringe equipped
with a custom made needle (c 1.2 mm, l = 50 mm). The filling level Table 1
was controlled gravimetrically and amounted to 300 ± 10 mg. After Composition of VA/PVP gels.
the filling procedure, the drug reservoirs were allowed to stand VA/PVP 90 gel Weighted PVP 90 Weighted VA Concentration of
at room temperature in upright position in order to deaerate. The labeling (%) sample (g) sample (g) PVP 90 in gel (%)
cold propellant (5 °C) was added immediately prior the experiment. 5 0.2472 4.5153 5.1906
7.5 0.3757 4.2297 8.1578
2.2. Gel preparation 10 0.4989 4.4974 9.9854
12.5 0.625 4.4534 12.3070
15 0.7505 4.349 14.7171
The investigated gels were prepared in 10 mL plastic luer-lock
syringes (Braun) by mixing VA (Sigma Aldrich) and povidone 90
(PVP 90, Sigma Aldrich) as a gelling agent. First, PVP 90 was added
directly to the VA and vortexed for 3 min in order to obtain a supported by Rheoplus application, cone CP25-1, gap d = 0.052).
homogenous dispersion. Subsequently, the mixture was heated The kinematic viscosity of the gel was determined at shear
to 50 °C in a water bath, then vortexed for further 3 min, cooled rates ranging from 0.1 up to 10 s 1 for samples with a volume of
down slowly and finally stored at 5 °C for at least 24 h prior to test- approximately 1 mL. The measurement was performed at 37 °C
ing. Prepared gels contained 5%, 7.5%, 10%, 12.5% and 15% of PVP in triplicate for each formulation (Fig. 2).
90. The compositions of the gels are given in Table 1.
2.4. HPLC determination of valproic acid
2.3. Viscosity analysis
The drug amount in the samples was determined by means of
The rheological properties of the VA/PVP gels were determined RP-HPLC by using a method validated according to ICH guidelines
by using a rotational cone-plate viscometer (Anton Paar MRC, [21]. A KONTRON HPLC system with single wave length UV
336 D. Haznar-Garbacz et al. / European Journal of Pharmaceutics and Biopharmaceutics 81 (2012) 334–338
Fig. 2. The dependence of the dynamic viscosity of v VA/PVP 90 gels on the shear
rate. Presented are means of n = 3 and the standard deviation is indicated by the
error bars.
detection set at k = 200 nm was used for the drug determination.
The system was equipped with a C-18 column (Lichrospher
LiChroCART 125-4). The mobile phase was composed of a mixture
of acetonitrile (ACN) and 50 mM sodium phosphate buffer of pH
6.8. The system was operated under isocratic conditions at a flow
rate of 1 mL/min and an oven temperature of 30 °C. The injection
volume was 40 lL. The retention time of VA was 2.9 min. The VA
peak was clearly separated from the injection peak and impurities
(base line separation). The method was characterized by a
good linearity (line equation y = 104.74 x, regression coefficient
R2 = 0.9997), sensibility and moreover by a sufficient robustness
to temperature changes (<5 °C), pH variations (<0.5 pH) and the
ACN content (<2%) of the mobile phase. This enabled utilization
of the method for routine measurements. The limits of detection
and quantification were 1.12  10 3 and 9.01  10 3 mg/mL,
respectively. The method reproducibility calculated for internal
standards containing 14.265, 28.530 and 42.795  10 3 mg/mL
and 10 subsequent assays was 98.16% (RSD 4.34%), 98.24% (RSD
3.25%) and 97.90 (RSD 4.75%).
2.5. In vitro dissolution
The drug delivery characteristics of the novel system were
investigated by using an USP paddle apparatus with a rotational
speed of 75 rpm and a temperature of 37 °C. The dissolution med-
ium (500 mL) used to assure sink conditions and solubilization of
the gels consisted of 50 mM sodium phosphate buffer, pH 6.8, with
addition of 0.05% SLS. The dissolution tests were continued until
the complete drug dose was released by the system. The sampling
scheme was therefore dependent on the gel type. Samples of 2 mL
volume were withdrawn through a 0.1 lm pore size PES filter and
transferred directly into HPLC vials. The withdrawn volume was
immediately replaced with blank medium and considered in the
Table 2
Slopes of mean dissolution profiles with corresponding regression factors and half
delivery times calculated as geometrical means (⁄calculated without lag time in the
range 0–60% of the amount of the drug dissolved).
VA/PVP 90 gel Slope of the Regression Calculated
labeling (%) dissolution profile coefficient R2 t0.5 (min)
5 3.644 0.9906 13.37
7.5 1.9256 0.9902 23.70
10 0.7732 0.9909 59.17
⁄
12.5 0.3879 0.9438 114.44
⁄ shear rates were correlated with the slope of the mean dissolution
15 0.2521 0.9626 353.33
Fig. 3. Dissolution profiles of valproic acid from the drug delivery system filled with
gels of different PVP 90 concentration. Presented are means of n = 6, standard
deviation is indicated by the error bars.
calculations. All dissolution tests were performed for n = 6 in par-
allels (Fig. 3).
3. Results
The gels obtained from VA and PVP 90 were clear, transparent
and colorless or bright yellow. During storage of the gels, no visu-
ally detectable changes were observed. The dynamic viscosities are
shown in Fig. 2. The gels containing 5–10% PVP 90 were Newtonian
liquids at shear rates higher than 0.1 s 1. For the gels containing
12.5% and 15% PVP, pseudoplastic behavior was observed for the
entire applied shear rate range. The release profiles of the five sys-
tems are presented in Fig. 3. A clear correlation between VA release
rate and the viscosity of the VA/PVP 90 gel could be observed. For a
gel containing 5% PVP, the time for complete release was about
30 min, whilst in the case of the 15% PVP 90 gel more than
480 min was required to release the same volume of the liquid
drug formulation. In our study, especially for more viscous gels,
we observed a characteristic lag time in the delivery process, which
was 5 min for a 10% gel and about 25 min for a 15% gel, respec-
tively. This lag time is due to the situation that the gel was always
only placed in the main chamber of the delivery system with the
capillary remaining empty in every case.
The results indicate that if an immediate start of drug release is
required, both elements, that is the main chamber and the capil-
lary, have to be filled with the gel formulation. However, indepen-
dent of viscosity the gels are emptied from the system as a thin
stream coming out form the capillary. The gel is subsequently dis-
persed in the dissolution medium and does not agglomerate on the
capsule surface. The preliminary tests yielded linear relationship
between system mass loss, emptied volume and the increase in
the amount of the drug dissolved (data not shown).
The gels containing 5%, 7.5% and 10% of PVP 90 were delivered
by the system with almost zero-order release kinetics. The gel con-
taining 12.5% and 15% of PVP 90 were delivered in a linear manner
in the range of 0–60% of the amount of drug released. Subse-
quently, an increase in the delivery rate was observed. Therefore,
the linearity coefficients of the drug delivery process were calcu-
lated for the range of 0–100% drug released for formulations con-
taining 5%, 7.5% and 10% PVP 90 and for 0–60% drug released
from gels containing 12.5% and 15% of PVP 90. The effective release
time of each of the obtained profiles, the calculated linear regres-
sion factors and coefficients of determination (R2) are shown in Ta-
ble 2. The viscosities of the different gels determined at different
profiles (k). The resulting exponential relationships between these
 D. Haznar-Garbacz et al. / European Journal of Pharmaceutics and Biopharmaceutics 81 (2012) 334–338
Fig. 4. Correlation of the viscosity (mean of n = 3) and the slope of the mean VA
dissolution profiles obtained from novel drug delivery system for gels of different
viscosities.
parameters are shown in Fig. 4, and the equations describing the
curves and their fitting coefficients are given in Table 3. Since the
mathematical relationship derived from the experimental data
show a very good fit, the coefficients can be used in the design
phase to achieve the desired release rate from the novel drug deliv-
ery system.
4. Discussion
The use of liquid or semi-solid drug formulations may reduce
the problems associated with drug solubilization in the gastroin-
testinal lumen, where both the fluid composition and volume are
highly variable, which, particularly for poorly soluble drugs, might
have a significant impact on dissolution and subsequent drug
absorption [22,23]. In addition to eliminating the step of drug dis-
solution, the use of semi-solid or liquid vehicles also gives the
opportunity to formulate poorly soluble APIs with surfactants or
as self-emulsifying or micro-emulsifying vehicles, which may en-
hance not only the solubility but also the bioavailability of the
administered drugs [24,25]. Due to the construction of the novel
liquefied gas-based drug delivery system, it is possible to minimize
the exposure time of the drug-containing vehicle to the GI liquids
before drug absorption. In addition, the system might protect the
formulation ingredients from physicochemical impact and enzy-
matic degradation, which may occur in the GI tract [26,27]. Fur-
thermore, the release mechanism of the system is independent of
physiological factors like the water content of the intestines at
the site of residence of the delivery system or the pH of the sur-
rounding gastrointestinal medium. Therefore, it is likely that the
novel system should at least be as robust as osmotically driven
drug delivery systems with regard to the generation of fluctuations
in plasma profiles.
The amount of gas (propellant) that was required to assure a
constant drug release from the novel delivery device in the present
study was approximately 25 lL of liquefied isopentane (this corre-
sponds to 5.425 mL gaseous isopentane and the same amount of
Table 3
Correlation between the slopes (k) of the dissolution profiles and the gel viscosity obtained at various shear rates.
Shear rate (1/s) Mean dynamic viscosity (Pa s) of the tested gels
5% 7.5% 10% 12.5%
0.1 0.603 1.080 3.740 12.712
1 0.307 1.047 3.681 12.567
10 0.284 1.03 3.525 10.933
337
formulated drug according to Avogadro’s law). Twenty-five micro-
liters of isopentane represented an excess of liquefied gas to elim-
inate imperfections of the model. When optimizing the system,
amounts as small as 10 lL of liquefied isopentane might be suffi-
cient for the controlled release of 2.17 mL of formulated drug. This
still exceeds the filling volume of a capsule size 000, which corre-
sponds to 1.37 mL. In comparison with the total volume of hydro-
carbon gases (115–1000 mL; mostly methane) that are produced
by bacteria under natural conditions in the gastrointestinal tract,
the 5.425 mL of gaseous isopentane produced by the drug delivery
device (from 25 lL liquefied isopentane) is an extremely small vol-
ume [28,29] and should have no impact on the patient.
In the present work, due to the manual filling procedure of the
drug containing vehicle, the capillary was not prefilled with the gel
prior to the dissolution experiments. Thus, a lag time was observed
in the delivery profiles. However, if the capillary were prefilled,
it would be also possible to achieve systems without lag times.
In the present setup, the system is activated by the temperature
increase above the boiling point of isopentane at intake. This
requires storage at a temperature below the boiling point.
However, it is likely that by sealing the capillary with materials
providing defined dissolution kinetics or pH dependent solubility,
it should also be possible to provide variable triggers for the
start of drug release.
The presented data indicate that the new system is highly flex-
ible in terms of delivery rate adjustment and assures the achieve-
ment of predictable release profiles with close to zero-order
delivery kinetics. The increase the delivery rate observed for the
gels containing 12.5% and 15 % PVP may be a consequence of the
gel inhomogeneity, water uptake during residence in the capillary
and a partial mixing with the propellant, which seems to change in
the flow properties of the gel. It is likely that the separation of the
gel from the propellant by the impermeable plunger manufactured
from an inert material or the saturation of the gel with the propel-
lant during the preparation might preclude such problems. More-
over, it is likely that the decrease in the capillary diameter and
length would also allow the reduction of the undesired changes
of the dissolution profiles. In this work, changes in the release rates
were achieved only by modifying the viscosity of the VA formula-
tions. The relationship between the viscosity of the drug formula-
tion and the delivery rate can be described and to some extent
even predicted by simple mathematical equations. Since some of
the investigated gels exhibited pseudoplastic flow properties at
low shear rates and high PVP contents, their dynamic viscosity
seems to vary depending on the applied shear rate. Thus, it is not
possible to calculate the theoretical dissolution profiles for all the
tested formulations because the shear rate in the novel drug deliv-
ery system depends on the flow velocity of the gel along the capil-
lary. However, based on the test results, theoretical viscosities of
the gels during the dissolution process were calculated (Table 4).
The results obtained for gels containing 5%, 7.5% and 10% PVP are
in good accordance with the viscosities obtained experimentally
with a cone plate viscometer (Fig. 2). The observed differences
are most likely related to the gel interaction with the capillary
material and some imperfections in the design or manufacturing
of the device. The theoretical viscosities calculated for the gels
Curve regression equation Regression coefficient R2
15%
1,38
21.921 y = 3.088x 0.991
1,57
20.532 y = 2.572x 0.996
1,51
16.611 y = 2.349x 0.992
338 D. Haznar-Garbacz et al. / European Journal of Pharmaceutics and Biopharmaceutics 81 (2012) 334–338

Table 4 release formulation marketed in the European Union, Int. J. Clin. Pharmacol.
Gel viscosities obtained at various shear rates and the calculated theoretical viscosity Ther. 44 (2006) 38–48.
of the gels during dissolution. [5] G. Garbacz, S. Klein, W. Weitschies, A biorelevant dissolution stress test device
– background and experiences, Expert Opin. Drug Deliv. 11 (2010) 1251–1261.
% Of PVP Dynamic viscosity of the tested gels [6] V. Malaterre, J. Ogorka, N. Loggia, R. Gurny, Oral osmotically driven system:
content 30 years of development and clinical use, Eur. J. Pharm. Biopharm. 73 (2009)
g at shear g at shear g at shear g 311–323.
rate 0.01/s rate 1/s rate 10/s Calculated [7] G. Garbacz, B. Golke, R.S. Wedemeyer, M. Axell, E. Soederlind, B. Abrahamsson,
5 0.603 0.307 0.284 0.532 W. Weitschies, Comparison of dissolution profiles obtained from nifedipine
7.5 1.082 1.047 1.031 1.001 extended release once a day products using different dissolution test
10 3.741 3.682 3.525 3.451 apparatuses, Eur. J. Pharm. Sci. 38 (2009) 147–155.
[8] R. Conley, S.K. Gupta, G. Sathyan, Clinical spectrum of the osmotic-controlled
12.5 12.741 12.567 10.933 5.288
release oral delivery system (OROS), an advanced oral delivery form, Curr.
15 21.912 20.500 16.677 8.373
Med. Res. Opin. 10 (2006) 1879–1892.
[9] D. Haznar-Garbacz, G. Garbacz, F. Eisenächer, S. Klein, W. Weitschies, An oral
controlled release drug delivery system for liquid and semisolid drug
containing 12.5% and 15% PVP differed markedly from the test re- formulations, AAPS PharmSciTech 12 (2011) 1183–1185.
sults. This is most likely due to the changes of the flow properties [10] C. O’Brien, R.W. Hoyt, M.J. Buller, J.W. Castellani, A.J. Young, Telemetry pill
measurement of core temperature in humans during active heating and
of the gel during release caused by partial mixing with the propel- cooling, Med. Sci. Sports Exerc. 30 (1998) 468–472.
lant and water uptake during the relative long prevailing time in [11] T. Gerstner, N. Bell, S. König, Oral valproic acid for epilepsy-long-term
the capillary. experience in therapy and side effects, Expert Opin. Pharmather. 9 (2008)
285–292.
By adjustment of other formulation parameters, that is, the
[12] B. Monti, E. Polazzi, A. Contestabile, Biochemical, molecular and epigenetic
length and diameter of the capillary (under investigation), use of mechanism of valproic acid neuroprotection, Curr. Mol. Pharmacol. 2 (2009)
a capillary sealing with desired solubility behavior or the proper- 95–109.
[13] S. Chateauvieux, F. Morceau, M. Dicato, M. Diederich, Molecular and
ties of the propellant, it might be possible to further improve the
therapeutic potential and toxicity of valproic acid, J. Biomed. Biotech. (2010),
release characteristics of the presented gas based drug delivery de- http://dx.doi.org/10.1155/2010/479364.
vice in order to meet the clinical needs for a given drug. [14] S.J. Pierre-Louis, R.T. Brannegan, A.T. Evans, Seizure control and side-effect
profile after switching adult epileptic patients from standard to extended-
release divalproex sodium, Clin. Neurol. Neurosurg. 111 (5) (2009 Jun) 437–
5. Conclusion 441.
[15] M.D. Sztajnkrycer, Valproic acid toxicity: overview and management, Clin. Tox.
40 (2002) 789–801.
The novel gas-based drug delivery system presented here is [16] J. Haymond, M.H.H. Ensom, Does valproic acid warrant therapeutic drug
dedicated to control oral delivery of liquid and semi-solid drug for- monitoring in bipolar affective disorder?, Ther Drug. Monit. 32 (2010) 19–29.
[17] S.R. Winkler, M.S. Luer, Antiepileptic drug review: Part 1, Surg. Neurol. 49
mulations. The system is characterized by a high flexibility of the
(1998) 449–452.
delivery rate, which can be achieved by adjustment of system [18] G. Marcelín-Jiménez, A.P. Angeles-Moreno, L. Contreras-Zavala, M. Morales-
parameters such as diameter and length of the capillary, vapor Martínez, R.L. Espinosa, A single-dose, three-period, six-sequence crossover
study comparing the bioavailability of solution, suspension, and enteric-
pressure of the propellant and the viscosity of the drug formula-
coated tablets of magnesium valproate in healthy Mexican volunteers under
tion. With this novel delivery system, liquid or semi-solid formula- fasting conditions, Clin. Ther. 9 (2009) 2002–2011.
tions can be administered in fixed doses over a long time range. [19] C.A. Zarate Jr., M. Tohen, R. Narendran, E.C. Tomassini, J. McDonald, M. Sederer,
The system advantages allow for the adjustment of the delivery A.R. Madrid, The adverse effect profile and efficacy of divalproex sodium
compared with valproic acid: a pharmacoepidemiology study, J. Clin.
rates according to the therapeutic requirements. Psychiatry. 60 (1999) 232–236.
[20] Shuzo Ohe, Distilation, vapor pressure, vapour–liquid equilibria. <www.s-
ohe.com>.
Acknowledgments [21] EMEA, Note for Guidance on Validation of Analytical Procedures: CPMP/ICH/
381/95.
The work was partly founded by the German Federal Ministry of [22] L. Kalantzi, K. Goumas, V. Kalioras, B. Abrahamsson, J.B. Dressman, C. Reppas,
Characterization of the human upper gastrointestinal contents under
Education (InnoProfile IP0612) and a research fellowship within conditions simulating bioavailability/bioequivalence studies, Pharm. Res. 1
the ‘‘Development program of Wroclaw Medical University’’ (2006) 165–176.
funded from ‘‘European Social Fund, Human Capital, National [23] C. Schiller, C.P. Froehlich, T. Giessmann, W. Siegmund, H. Moennikes, N.
Hosten, W. Weitschies, Intestinal fluid volumes and transit of dosage forms as
Cohesion Strategy (Contract No. UDA-POKL.04.01.01-00-010/08-
assessed by magnetic resonance imaging, Aliment. Pharmacol. Ther. 22 (2005)
00).’’ 971–979.
[24] T.D. Thi, M. Van Speybroeck, V. Barillaro, J. Martens, P. Annaert, P. Augustijns, J.
Van Humbeeck, J. Vermant, G. Van den Mooter, Formulate-ability of ten
References compounds with different physicochemical profiles in SMEDDS, Eur. J. Pharm.
Sci. 38 (2009) 479–488.
[1] W. Weitschies, O. Kosch, H. Moennikes, L. Trahms, Magnetic marker [25] D. Patel, K.K. Sawant, Self micro-emulsifying drug delivery system:
monitoring: an application of biomagnetic measurement instrumentation formulation development and biopharmaceutical evaluation of lipophilic
and principles for the determination of the gastrointestinal behavior of drugs, Curr. Drug. Deliv. 4 (2009) 419–424.
magnetically marked solid dosage forms, Adv. Drug. Deliv. Rev. 57 (2005) [26] A. Christiansen, T. Backensfeld, W. Weitschies, Effects of non-ionic surfactants
1210–1222. on in vitro triglyceride digestion and their susceptibility to digestion by
[2] M. Kamba, Y. Seta, A. Kusai, K. Nishimura, Comparison of the mechanical pancreatic enzymes, Eur. J. Pharm. Sci. 41 (2010) 376–382.
destructive force in the small intestine of dog and human, Int. J. Pharm. 37 [27] A. Christiansen, T. Backensfeld, S. Kuehn, W. Weitschies, Investigating the
(2002) 139–149. stability of the non-ionic surfactants tocopheryl polyethylene glycol succinate
[3] D. Cassilly, H.P. Parkman, Gastric emptying of a non-digestible solid: and sucrose laurate by HPLC-MS, DAD, and CAD, J. Pharm. Sci. 100 (2011)
assessment with simultaneous SmartPill pH and pressure capsule, 1773–1782.
antroduodenal manometry, Gastric Emptying Scintigraphy. [28] M. Montalto, M. Di Stefano, A. Gasbarrini, G.R. Corazza, Intestinal gas
Neurogastroenterol. and Motil. 20 (2008) 311–319. metabolism, Dig. Liver. Dis. 3 (Suppl 1) (2009) 27–29.
[4] M. Wonnemann, B. Schug, K. Schmücker, E. Brendel, P.A. van Zwieten, H. [29] A. Strocchi, M.D. Levitt, Factors affecting hydrogen production and
Blume, Significant food interactions observed with a nifedipine modified- consumption by human fecal flora, J. Clin. Invest. 89 (1992) 1304–1311.