Available via license: CC BY-NC-ND 4.0
Content may be subject to copyright.
ORIGINAL ARTICLE
Redefinition to bilayer osmotic pump tablets as
subterranean river system within mini-earth via
three-dimensional structure mechanism
Abi Maharjan
a,b,c,y
, Hongyu Sun
a,d,y
, Zeying Cao
a,c
,KeLi
c,e
,
Jinping Liu
f
, Jun Liu
a,g
, Tiqiao Xiao
e
, Guanyun Peng
e
, Junqiu Ji
g
,
Peter York
a
, Balmukunda Regmi
h
, Xianzhen Yin
a,g,
*,
Jiwen Zhang
a,c,d,
*,LiWu
a,b,
*
a
Center for Drug Delivery System, Shanghai Institute of Materia Medica, Chinese Academy of Sciences, Shanghai
201203, China
b
School of Pharmacy, Key Laboratory of Molecular Pharmacology and Drug Evaluation, Ministry of Education,
Yantai University, Yantai 264005, China
c
University of Chinese Academy of Sciences, Beijing 100049, China
d
NMPA Key Laboratory for Quality Research and Evaluation of Pharmaceutical Excipients, National Institutes for
Food and Drug Control, Beijing 100050, China
e
Shanghai Synchrotron Radiation Facility/Zhangjiang Lab, Shanghai Advanced Research Institute, Chinese
Academy of Sciences, Shanghai 201204, China
f
Hefei Lifeon Medication Group, Hefei 230088, China
g
Center for MOST and Image Fusion Analysis, Shanghai Institute of Materia Medica, Chinese Academy of
Sciences, Shanghai 201210, China
h
Maharajgung Medical Campus, Institute of Medicine, Tribhuvan University, Kathmandu 44606, Nepal
Received 10 August 2021; received in revised form 11 October 2021; accepted 24 October 2021
KEY WORDS
Bilayer osmotic pump
tablet;
Synchrotron radiation
micro-computed
Abstract Defining and visualizing the three-dimensional (3D) structures of pharmaceuticals provides a
new and important tool to elucidate the phenomenal behavior and underlying mechanisms of drug deliv-
ery systems. The mechanism of drug release from complex structured dosage forms, such as bilayer os-
motic pump tablets, has not been investigated widely for most solid 3D structures. In this study, bilayer
osmotic pump tablets undergoing dissolution, as well as after dissolution in a desiccated solid state were
*Corresponding authors. Tel./fax: þ86 21 20231980 (Li Wu); þ86 21 20231980 (Jiwen Zhang); þ86 2150805622 (Xianzhen Yin).
E-mail addresses: xzyin@simm.ac.cn (Xianzhen Yin), jwzhang@simm.ac.cn (Jiwen Zhang), wuli@simm.ac.cn (Li Wu).
y
These authors made equal contributions to this work.
Peer review under responsibility of Chinese Pharmaceutical Association and Institute of Materia Medica, Chinese Academy of Medical Sciences.
https://doi.org/10.1016/j.apsb.2021.11.008
2211-3835 ª2022 Chinese Pharmaceutical Association and Institute of Materia Medica, Chinese Academy of Medical Sciences. Production and hosting
by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
Chinese Pharmaceutical Association
Institute of Materia Medica, Chinese Academy of Medical Sciences
Acta Pharmaceutica Sinica B
www.elsevier.com/locate/apsb
www.sciencedirect.com
Acta Pharmaceutica Sinica B 2022;12(5):2568e2577
tomography;
Three-dimensional
microstructure;
Release kinetics;
Void formation;
Peripheral “roadways”;
Push-pull model;
Subterranean river model
examined, and visualized by synchrotron radiation micro-computed tomography (SR-mCT). In situ
formed 3D structures at different in vitro drug release states were characterized comprehensively. A
distinct movement pattern of NaCl crystals from the push layer to the drug layer was observed, beneath
the semi-permeable coating in the desiccated tablet samples. The 3D structures at different dissolution
time revealed that the pushing upsurge in the bilayer osmotic pump tablet was directed via peripheral
“roadways”. Typically, different regions of the osmotic front, infiltration region, and dormant region were
classified in the push layer during the dissolution of drug from tablet samples. According to the observed
3D microstructures, a “subterranean river model” for the drug release mechanism has been defined to
explain the drug release mechanism.
ª2022 Chinese Pharmaceutical Association and Institute of Materia Medica, Chinese Academy of Medical
Sciences. Production and hosting by Elsevier B.V. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/4.0/).
1. Introduction
A therapeutic effect is attained when a drug released and absorbed
from a dosage forms reaches the biologic target within the thera-
peutic window at a predetermined time. To produce effective ther-
apeutic responses without causing significant adverse effects in
patients, various dosage forms have been designed. Among them,
oral drug delivery is one of the most common and convenient routes
for drug administration in clinical practice. The releasing rate and
extent from conventional oral dosage forms could not be controlled
for attaining constant therapeutic level
1,2
. Moreover, bioavailability
from these formulations could be easily affected by the physio-
logical factors like food, gastric motility, and pH
3
. The above
shortcomings can be overcome by formulating controlled and sus-
tained drug release dosage forms such as osmotic pump tablets
4
,
multi particulate
5
, and erosion controlled drug delivery systems
6
.
Osmotic pump tablets have been shown to be an attractive tool for
oral controlled drug delivery system
7
with different osmotic devices
invented over the past 60 years
8
. Osmotic pump tablets are efficient
formulations with a simple arrangement to achieve zero-order drug
release by generating hydraulic pressure via a swelling osmotic
agent inside the tablets
9
. Different types of osmotic pump tablets
have been reported such as elementary osmotic pump tablets
10
,
controlled porosity osmotic pump tablets
11
, sandwiched osmotic
pump tablets
12
, capsule-based osmotic systems
13,14
, and pushepull
osmotic pump (PPOP) tablets
15
. Among them, PPOP tablets have
received much attention for its success in delivering drugs with low
aqueous solubility
16,17
.
Various imaging techniques such as micro-computed tomog-
raphy (mCT)
18
, magnetic resonance imaging (MRI)
19
, terahertz
imaging
20
, confocal imaging technique and Raman imaging
21
facilitate in situ visualization of solid dosage forms, offering un-
derstanding of ongoing microstructural transformations in
different conditions. The mCT imaging offers special features for
non-destructive reconstruction of specimens at micrometer reso-
lution; MRI applies magnetic field gradients and radio waves to
generate the 3D images; terahertz imaging finds its application in
studying semiconductor material properties, biomedical cell im-
aging and structures of pharmaceutical dosage forms. The
confocal imaging technique is useful for tracking fluorescence
material within the specimens. Raman imaging allows observation
of molecular distribution and dynamic behaviors in the specimens.
Direct comprehensive visualization in submicron resolution in any
section and detailed microstructure analysis are the challenging
aspects in all of the above methods.
Synchrotron radiation micro-computed tomography (SR-mCT)
is an advanced technology that uses synchrotron radiation X-rays
having high photon flux and polarization as a light source to
achieve high-speed imaging, high energy, and high spatial reso-
lution. It is a valuable investigational tool for non-destructive in
situ three-dimensional (3D) visualizations of samples. Moreover,
this technology allows sectioning the sample in any direction for
the direct visualizations of the internal structure at micrometer
resolution. It has been used in pharmaceutics to reveal key in-
formation about the structure of active pharmaceutical ingredients
(API) and dosage forms
22e24
, drug release kinetics
25
, drug char-
acterization
26
, and the quality of pharmaceutical products
27
.
The underlying principle of drug delivery through an osmotic
pump system involves the controlled diffusion of water through a
semi-permeable membrane (surrounding the tablet unit) and drug
release through a drilled orifice
28,29
. For monolith osmotic pump
tablets, the 3D structures
24
and the controlled release kinetics
25,30
have been investigated based on SR-mCT. Due to the complex
geometry of PPOP tablets, the various microstructural transforms
occur within the tablets which may impede or promote drug
release kinetics. This aspect has not been considered to date in
defining drug release models from PPOP tablets.
A clear understanding regarding the release kinetics of PPOP
tablets with associated microstructural alteration has not been
established. In this study, dissolution conditioned bilayer osmotic
pump tablets before and after desiccation were investigated via
SR-mCT to discover detailed microstructural transformations at
different drug dissolution times. Static and dynamic structures of
bilayer osmotic pump tablets were characterized comprehensively.
Based on observed in situ structures, a new release model for the
bilayer osmotic pump tablets is presented which defines the drug
release mechanism.
2. Materials and methods
2.1. Materials
Bilayer osmotic pump tablets (CarduraXL) were purchased
from Pfizer Inc. (Batch number R61208, Wuxi, China), containing
doxazosin mesylate as an active ingredient, with polyethylene
oxide (PEO), sodium chloride (NaCl), hydroxyl propyl methyl-
cellulose (HPMC), red ferric oxide, titanium dioxide, magnesium
stearate, cellulose acetate, Macrogol, pharmaceutical glaze, and
black iron oxide as inactive ingredients. Synchrotron radiation was
applied for obtaining the micro-computed tomography scans from
Subterranean river model for elucidating release mechanism from bilayer tablets 2569
Shanghai synchrotron radiation facility (SSRF) in Shanghai
Institute of Applied Physics, Chinese Academy of Sciences
(Shanghai, China). Different software programs including Phase-
sensitive X-ray Image Processing and Tomography Reconstruc-
tion (PITRE, Version 3.1), Image-Pro Plus (Media Cybernetics,
Version 6), Image-Pro Premier 3D (Media Cybernetics, Version
9.1), and Amira (Thermo Fischer Scientific, Version 5.6) were
used for obtaining different qualitative and quantitative informa-
tion of the tablets
27
. Dissolution tests were carried out according
to United State Pharmacopoeia (USP, 2017), using dissolution
apparatus II (Distek Model 2500, Distek Inc., North Brunswick,
USA). The chemicals used for the dissolution test were of
analytical grade and purchased from the Sinopharm Chemical
Reagent Co., Ltd. (Shanghai, China). Water was purified by a
reverse osmosis using Milli-Qs system (Millipore, Bedford, MA,
USA).
2.2. In vitro dissolution test
In vitro drug release of CarduraXL was measured using the
paddle method at a rotation speed of 50 rpm according to USP
2017 (nZ6). The dissolution media was simulated gastric fluid,
without enzyme at pH 1.2 at 37 C. Following the addition of the
tablets in the dissolution vessel, 2 mL of the aliquots were with-
drawn from the dissolution vessel at the predetermined time points
as 1, 2, 4, 6, 8, 12, and 16 h. The withdrawn samples were
immediately filtered using a microporous membrane with the pore
size of 0.22 mm and also, an equal volume of dissolution medium
was replaced in the dissolution vessel to maintain sink condition.
The concentration of the doxazosin mesylate at each time point
was analyzed by using high-performance liquid chromatography
(HPLC, Agilent, 1290, USA) on column Platisil (C18,
250 mm 4.6 mm, 5 mm, Agilent Technologies, USA). Buffer
solution for mobile phase was prepared by dissolving monobasic
potassium phosphate in 800 mL water, 4 mL of triethylamine was
added and its pH was adjusted to 4.5 by phosphoric acid, and
finally diluted to 1000 mL with water. The mobile phase was
methanol/buffer solution at a ratio of 55:45, and the flow rate was
1 mL/min. An injection volume of 20 mL of the sample was
injected through the column. The detection wavelength was
245 nm. Calibration curve was prepared over the concentration
range of 0.2, 0.5, 1.0, 2.0, 5.0, 10.0, and 20.0 mg/mL against the
area under the curve.
2.3. Samples pretreatment for SR-mCT
Three tablets were sampled at predetermined time points of 1, 2,
4, 6, 8, 12, and 16 h from the dissolution medium to study the
temporal changes in the microstructure of the bilayer osmotic
pump tablets. Owing to the interference of the dissolution medium
during the mCT scans for acquiring images, the samples were
desiccated. Different methods of drying have been reported in the
literature
25
. In this experiment, the silica drying method was
applied due to the minimal changes in the internal microstructure
of the osmotic pump tablet. The tablets were placed in an airtight
vessel containing dried silica for assisting rapid drying for 72 h to
ensure they completely dried before the acquisition of the tomo-
graphic image.
To investigate microstructural detail directly of the tablet at
different dissolution stages, wet dissolution conditioned tablet
samples were also prepared. The samples at predetermined time
were taken out from the dissolution media and mCT was per-
formed immediately.
2.4. Image acquisition by SR-mCT
SR-mCT tomographic images were acquired from the beam line
BL13W1 at SSRF of the osmotic pump tablets. The tablets were
scanned by the synchrotron radiation at the photon energy level of
18 keV, and X-ray window dimensions were maintained at
45 mm 5 mm for the exposure of rays to the tablets. The
double-crystal monochromator with Si (111) and Si (311) crystals
was used to monochromatize the X-rays. Synchrotron radiation
source was placed at the distance of 34 cm from the sample and
after penetrating through the sample the X-rays were converted to
the visible light by using a cleaved Lu
2
SiO
5
: Ce single crystal
scintillator (10 mm thickness). The obtained projections were
magnified by diffraction-limited microscope optics (1.25 ) and
digitized by high-resolution with an effective pixel size of 5.2 mm
(ORCA Flash 4.0 Scientific CMOS, Hamamatsu K.K., Shizuoka
Pref., Japan, physical pixel size: 6.5 mm). The exposure period
was maintained for 0.8 s and the distance between the detector and
the sample was set at 30 cm to obtain good phase contrast. For
each sample, 1080 projection images were captured with an
angular step size of 0.17for 180rotation. Flat-field (tomogram
obtained without sample) and dark-field (tomogram obtained after
shutting off the X-rays) images were collected during each
acquisition procedure to correct the electronic noise and variations
in the X-ray light source brightness
31
.
PITRE was used for the phase contrast extractions and quality
enhancement of reconstructed slices. Filtered back projection al-
gorithm was used for the reconstruction of projections by using
PITRE software. In situ visualization of the tablet and different
qualitative and quantitative parameters were analyzed by using
commercial software like Image-Pro Plus, Image-Pro Premier 3D,
and Amira.
2.5. Video actuation by the camera
Tablets were divided into 7 groups containing 3 tablets in each
group. The dissolution test was carried out to each group using
paddle with 50 rpm speed following USP dissolution test
method. At the predetermined time (1, 2, 4, 6, 8, 12, and 16 h),
tablets were taken out and placed in the petri dish contained
dissolution medium. Then videos were captured by using a
digital video camera (Microsoft LifeCam Studio™)for5min
using a 140 mm 20 mm petri dish containing about 200 mL of
dissolution medium at 37 C. Then videos were forwarded by
20 to view the release pattern of the drugs from the orifice
effectively.
3. Results and discussion
3.1. Structure characterization by the SR-mCT
The detailed in situ structures of the bilayer osmotic tablets are
revealed as shown in Fig. 1, with the drug layer and push layer
identified within the tablet. The crystals of NaCl were seen in the
push layer for creating the osmotic pressure and facilitating water
absorption. Definite arrangement of PEO as a swelling agent could
be seen in drug layer and push layer. The diameter of the tapering
2570 Abi Maharjan et al.
orifice in the surface coating was approximately 500 mm. Two types
of coating layers were identified, with the inner layer as the semi-
permeable membrane and the outer layer as a dense protective layer.
The 3D microstructures within the tablet were revealed via tomo-
graphic images and utilized for comprehensive structural analysis.
3.2. In vitro drug release kinetics from bilayer osmotic pump
tablet
The in vitro dissolution profile of the bilayer osmotic pump
tablets at different periods is illustrated in Fig. 2A, indicating 3
kinetic phases of during the drug release process. At the initial
stage, i.e., lag stage (0e2 h), only small amount of the drug was
released, then drug release progressively increased up to 12 h.
After this period, the drug release rate dramatically decreased,
indicating the end phase of drug release. More than 90% of the
drug was released within 16 h. Typically, desiccated tablets at
6h(Fig. 2A and B) show the different structural transformation
from a sectioned axial view. The hollow space between two
layers revealed that a swollen layer formed and subsequently
dried out during the desiccation process. A “U” shaped
morphological structure in the push layer after 4 h dissolution
time indicated that the pressure applied to the drug layer is
exerted from the peripheral extension of push layer. At the
initial stage, during the lag period of release, water absorption
inside the tablet took place. This resulted in the formation of
gelatinized PEO. Low molecular weight PEO in the drug layer
will dissolve quickly resulting in a drug suspension. The high
molecular weight PEO present in the push layer swells when it
absorbs water
32,33
and the swollen gelatinized PEO in the push
layer functions as a driving force for expelling the gelatinized
PEO in the drug layer. The movement of push layer from the
core region to the peripheral region of the tablet was visualized
(from the axial slice of the tablet) representing the pushing route
for drug release through the orifice. The water content and
fluidity observed at the peripheral region of the tablet were high
such that the push layer and drug layer in this region expanded
rapidly and shifted towards the orifice for drug release. Pressure
was generated from the push layer to the drug layer for drug
releaseafter2h.Thereleaseratereducedafter12hduetothe
maximum expansion of the push layer and reduced drug levels
in the drug layer.
From Fig. 2B, salt crystals were also identified in the push
layer of the tablet which would create an osmotic gradient for
water penetration. From the visualization of push layer of the 6 h
tablet, distinct regions can be described fined based on the salt
crystal distribution. Spatial distribution of the crystals in the tablet
was due to the shift of salt after dissolving. The dissolved salt
preferentially moved to the outer peripheral region with the high
the water content as compared to the core region, and also due to a
high osmotic gradient. The outermost peripheral region consisted
of the relatively larger dimension of recrystallized salt indicating
the presence of concentrated solution, while that in the core re-
gions were the same as that of the original tablet.
From this analysis of the images and release pattern, the
observed release mechanism can be interpreted. The pushing force
in the bilayer osmotic pump tablet initiated from the peripheral
region of the tablet. With recrystallized salts concentrated in the
peripheral region of the tablets, the consequence would be that the
osmotic pressure within the push layer be greater in the peripheral
region rather than from the core region.
3.3. Characterization of empty voids at different dissolution
time points
To provide detailed 3D visualization of voids and vacant spaces
within the tablets, photos were randomly segmented with different
colors depending on the threshold difference of grey value as
shown in Fig. 3. The vacant space was observed within the bilayer
osmotic pump tablet indirectly, indicating the released drug layer
portion. The vacant space created by the released portion of drug
layer and drug release rate was highly correlated (Supporting
Information Fig. S1), which is also in agreement with the study
carried out in felodipine monolith osmotic pump tablets
30
. Defi-
nite channels (typically at 4, 6, and 8 h) in the drug layer were
observed indicating preferred routes for the drug release. From
these observations, the drug release pathways and the release
process can be understood and defined.
3.4. In situ visualization of channels during drug release
From the analysis of the voids within the tablet channel-like
structures were identified in the drug layer. Also, at different
dissolution times, well defined channels were observed through
the semipermeable membrane. Through that channel, drug
movement through these channels can be visualized (Fig. 4 and
Supporting Information Movie). The channels were not identified
after 1 and 2 h dissolution time, due to the overall swelling of the
tablets or lag time for the formation of the channels. Distinct
channels were visualized at 4 h. As the push layer of the tablets
contained particles of red iron oxide, dark red color, minute traces
are seen, indicating the release channels. The formation of distinct
channels from the peripheral to the orifice of the tablets had
revealed that the pressure in the osmotic pump tablets was
generated through the peripheral region of the tablets together
with channels formed from the peripheral due to excess pressure
generated by the push layer at peripheral region. This finding
supports the view that the pressure in the bilayer osmotic is
initiated solution flow from the peripheral region of the tablets to
the orifice of tablets. After 12 and 16 h, the drug layer was
exhausted and only swollen push layer is present (Fig. 4). The
drug release channels originating from the peripheral region of the
tablets to the orifice indicated that a definite pathway was created
for releasing the drug through the orifice.
Figure 1 In situ structures of bilayer osmotic pump tablet as
revealed by SR-mCT. (A) 3D view of the tablet. (B) Orthoslice view of
the tablet.
Subterranean river model for elucidating release mechanism from bilayer tablets 2571
3.5. In situ visualization of crystals inside the tablets
The pattern of NaCl crystal movement in the peripheral region of
the tablet from the push layer to the drug layer is shown in
Supporting Information Fig. S2. The peripheral upward movement
of the salt from the push layer to the drug layer indicated that the
osmotic pressure was developed peripherally. The vacant spaces at
the center region of the desiccated tablet support the finding that
Figure 3 3D structures of extracted voids from desiccated tablet samples. The voids were randomly segmented with different colors, which
represent the individual voids (The color is randomly selected by the software, and the voids are indicated by the colors (nZ3 for the tablet at
each dissolution time point).
Figure 2 In vitro dissolution profile of bilayer osmotic pump tablet at different periods (nZ6 for the tablet of dissolution experiment). (A)
Pictures represent the axial orthoslice at the respective dissolution time point. (B) Internal structure visualization of desiccated bilayer osmotic
pump tablets at different dissolution time. Upper, middle, and lower represent the radial slice at the respective position.
2572 Abi Maharjan et al.
the push layer moved to the peripheral regions and subsequently to
the drug layer.
The crystals were characterized by dividing the tablet into
three regions, viz., upper, middle, and lower region as shown in
Fig. 5A. Each section was analyzed from the respective region
with a height of 1.5 mm (289 pixels of slices), crystals were
extracted by using Image-Pro Premier 3D software to reveal
special distribution and visualization (Fig. 5B and C). A pattern of
crystal movement from the push layer to the drug layer could be
visualized with movement from the peripheral region indicating
presence of more concentrated solution of NaCl for creating more
push force to the drug layer as shown in Fig. 5C.
Relatively high amount of solution in the peripheral region
resulted in rapid swelling and prompt movement of the push
layer towards the drug layer. When salts at the push layer in the
peripheral region dissolved, they created a more osmotic pres-
sure gradient, attracting more dissolution fluid inside the tablets
that resulted in swelling of PEO. As compared to the outer pe-
ripheral region, the inner region had less water content that
resulted into the formation of a higher concentration salt solu-
tion, which progressively diffused towards the outer peripheral
region. Subsequently, this region moved towards the drug layer
due to high fluidity thus creating osmotic upsurge for drug layer
release. When the tablet was dried, the salts recrystallized ac-
cording to their distribution in the dissolution conditioned state.
This spatial distribution of the crystals in the peripheral region in
the osmotic push layer and their movement pattern indicated that
osmotic push force was created from the peripheral region of the
tablets.
3.6. Study of dissolution conditioned PPOP tablets
The dissolution conditioned samples were visualized via SR-mCT
in an attempt to define the exact microstructural transformations at
different dissolution times. Due to the presence of dissolution
media within the tablet and its interference with the synchrotron
X-ray absorption, the quality of the image obtained was not in an
optimum state. However, different transformation states were
visualized to understand the drug release mechanism.
The typical release state from the bilayer osmotic pump tablet
at 6 h is shown in Fig. 6A. The gap between two layers of the
tablet identified in the desiccated tablet was not seen. On the
contrary, a swollen state region inside the tablet is identified.
Three regions in the push layer were distinguished as an osmotic
front, infiltration state, and dormant state which had functionally
different attributes. However, in the case of drug layer, only two
regions namely, the osmotic region and dormant region could be
visualized.
The osmotic front was characterized by high fluidity and
mobility which consisted of dissolved salt and PEO for creating
the osmotic push force to the drug layer. In the drug layer, an
osmotic front was formed beneath the semipermeable membrane
and also between two layers of the tablet. The infiltration region
was characterized by the less compacted mass of PEO beneath the
osmotic front of the push layer. A “bubble like” spherical structure
was visualized in this region (Supporting Information Fig. S3).
Bubbles formed during the swelling of high molecular weight
PEO in the push layer. The infiltration region was also charac-
terized by low salt concentration (comparatively low brightness in
images) as salt from this region moved to the osmotic front. In the
dormant region, water had not penetrated and it remained in its
original state.
The mass movement and the osmotic pressure gradient in the
osmotic front were high due to the presence of more fluid, dis-
solved NaCl, and PEO. There was a continuous tendency to draw
water into this region that resulted in swelling. High molecular
weight PEO in the push layer resulted in a high swelling index
which created a higher osmotic push force from the osmotic front.
A series of transformation states from the dormant region to
infiltration region and finally to the osmotic front is visualized at
6 h dissolution time as shown in Fig. 6A. The dormant region
progressively decreased its size and finally disappeared after 8 h
dissolution time as shown in Fig. 6B. Similarly, the osmotic front
increased its dimension showing the swollen state of PEO.
Functionally different intermediate states within osmotic
bilayer tablets provide insight into the drug release mechanism.
Movement of the osmotic front towards the drug layer through the
peripheral region indicated that the pushing upsurge force was
created through the peripheral push region. Less compacted mass
Figure 4 In situ visualization of tablets via a camera at different dissolution time. The arrows in the figure represent channels through which the
drug was released. Drug layer was released till 8 h through definite channels. Reddish brown color after 12 h indicated that drug layer was
exhausted and push layer was released. The scales in the figure represented 1 mm.
Subterranean river model for elucidating release mechanism from bilayer tablets 2573
in the infiltration region indicated that PEO from the infiltration
region moved to the osmotic front region, where swelling took
place and the push force was created. The dormant region slowly
converted to the infiltration region and osmotic front to create
pressure on the drug layer. The drug layer also absorbed the
dissolution fluid that resulted in an increase in the fluidity of this
region. As the push layer triggered the pushing action through the
peripheral region, the drug layer moved through the same route
towards the orifice to create controlled drug delivery.
3.7. Subterranean river model over the traditional pushepull
model
The pushepull theory
3
was postulated based on the osmotic
phenomenon without taking into account in situ formed 3D
structures within the tablets. The theory is briefly explained as
follow; when the bilayer tablet was placed in the dissolution
media, the media penetrated in both layers of the bilayer osmotic
tablet leading to the formation of the suspension in the drug layer
which is referred to the pull layer. The formed suspension from
this layer will be released through the orifice when the push layer,
having a high expandable function, expands in volume after the
dissolution media absorbed in this layer
34,35
.
From the 3D observation of the bilayer osmotic pump tablet,
complex phenomena within the bilayer osmotic pump tablet
during the drug release process were identified with a number of
different structures present within the tablet during the drug
release process. The 3D view of a desiccated osmotic pump tablet
revealed that a larger portion of the push layer structure was
located in the upper region of the tablet having moved via the
peripheral pathway as illustrated in Fig. 2B. The core region of the
push layer translocated towards the peripheral region of the tablet
due to high fluidity at the periphery during the release process.
From the dissolution conditioned tablet visualization, movement
of the osmotic front from the push layer to the drug layer was
visualized. The loose swollen infiltration region and the dormant
region beneath the osmotic front were shown in push layer at
different states of expansion as shown in Fig. 6. According to
these 3D observations, it is clear that the osmotic front pushed the
suspension formed in the drug layer towards the orifice at a
relatively higher rate compared to the infiltration region and
dormant region through the peripheral routes. The release mech-
anism can be described as a ‘Subterranean river model’ to
represent an understanding of the drug release mechanism from
the bilayer osmotic pump tablets as illustrated in Fig. 7.
During drug dissolution and with the underground river flow-
ing, no morphological changes are seen externally. However,
under the surface and layers, the current of water would be
flowing in different layers, as revealed by this study. The water
taken up by the tablet primarily moved from the upper layer, even
Figure 5 Spatial distribution of NaCl crystals in different regions within desiccated bilayer osmotic pump tablets during the drug release
process. (A) Division of the tablet into 3 regions. (B) Upper, middle, and lower regions of segmented extracted crystals. (C) Extracted crystals at
different dissolution times. Different colors represent a specific single crystal (nZ3 for the tablet at each dissolution time point).
2574 Abi Maharjan et al.
Figure 6 In situ view of dissolution conditioned wet bilayer tablets at different dissolution time. (A) Representation of different regions
(osmotic front, infiltration state, and dormant state) in the tablet at 6 h by axial orthoslice. White color represents high-density material (NaCl). (B)
Representation of internal structure of the bilayer osmotic pump tablet at different dissolution time points.
Figure 7 A schematic model for the release kinetics from bilayer osmotic pump tablets over the traditional push‒pull model.
Subterranean river model for elucidating release mechanism from bilayer tablets 2575
though it appeared to move in an integrated manner from all the
regions of the layers. The flowrate of water in subterranean river
below the upper layer was low and, at the inner side, did not
appear to move at all due to resistance force created via viscosity.
In the same manner, the push layer seemed to move to the drug
layer in an integrated manner but in reality, it moved by creating
different push layers in the bilayer osmotic pump tablet.
4. Conclusions
Bilayer osmotic pump tablets were investigated for comprehensive
analysis of 3D microstructural transformations at different dissolu-
tion time using SR-mCT. By the application of various imaging
algorithms, microstructural transformations including crystal move-
ment pathways, voids, and channel pathways were characterized.
Comprehensive analysis of in situ formed 3D structures and their
retrospective analysis gave direct understanding about the drug
release mechanism from the tablet. Osmotic push force was more
pronounced at the periphery region of the tablets as high crystal
concentrations accumulated at the site. Definite channels, originating
in the peripheral region to the orifice indicated the pathways for drug
movement to the orifice. In situ formed functionally different inter-
mediate states during drug dissolution of wet tablet have been iden-
tified providing insight that the osmotic push force was primarily due
to the osmotic front at the peripheral region. A subterranean river
model for the drug release kinetics from bilayer osmotic pump is
proposed, taking into account the several microstructural trans-
formations identified from a 3D structural view over pushepull
model. This study focuses on in situ formed 3D structures and their
transformation to unveil the concealed release mechanism, opening a
new door in the pharmaceutics arena.
Acknowledgements
The authors are grateful for the National Nature Science Foun-
dation of China (Nos. 81803446, 81803441 and 81773645), Key
Program for International Science and Technology Cooperation
Projects of China (2020YFE0201700), and the Youth Innovation
Promotion Association of CAS (2018323). Thanks also go to the
staff from Shanghai Synchrotron Radiation Facility for the assis-
tance during the experiment and data collection.
Author contributions
Abi Maharjan and Hongyu Sun performed research and equally
contributed to this research. Li Wu, Jiwen Zhang and Xianzhen
Yin designed research. Tiqiao Xiao, Guanyun Peng and Ke Li
collected data. Zeying Cao and Jun Liu contributed in preparation
of manuscript. Junqiu Ji, Peter York and Balmukunda Regmi
contributed in preparation of manuscript and guided during
research. All authors have read and approved the final manuscript.
Conflicts of interest
The authors declare no conflict of interest.
Appendix A. Supporting information
Supporting data to this article can be found online at https://doi.
org/10.1016/j.apsb.2021.11.008.
References
1. Ahmed K, Shoaib MH, Yousuf RI, Siddiqui F, Qazi F, Iftikhar J, et al.
Comparative pharmacokinetics of osmotic-controlled and immediate-
release eperisone tablet formulation in healthy human subjects using a
sensitive plasma LC‒ESI-MS/MS method. Sci Rep 2020;10:1867.
2. Thakkar HP, Pancholi N, Patel CV. Development and evaluation of a
once-daily controlled porosity osmotic pump of tapentadol hydro-
chloride. AAPS J 2016;17:1248e60.
3. Verma RK, Krishna DM, Garg S. Formulation aspects in the devel-
opment of osmotically controlled oral drug delivery systems. J Control
Release 2002;79:7e27.
4. Eckenhoff B, Yum SI. The osmotic pump: novel research tool for
optimizing drug regimens. J Biomater 1981;2:89e97.
5. Sandberg A, Ragnarsson G, Jonsson UE, Sjogren J. Design of a new
multiple-unit controlled-release formulation of metoprolol-metoprolol
CR. Eur J Clin Pharmacol 1988;33:S3e7.
6. Daniel BS, Niels B, Ture KL. Egalet, a novel controlled release
system. Ann N Y Acad Sci 1991;618:578e80.
7. Mandal AS, Biswas N, Karim KM, Guha A, Chatterjee S, Behera M,
et al. Drug delivery system based on chronobiologyda review. J
Control Release 2010;147:314e25.
8. Santus G, Baker RW. Osmotic drug-deliveryda review of the patent
literature. J Control Release 1995;35:1e21.
9. Malaterre V, Ogorka J, Loggia N, Gurny R. Oral osmotically driven
systems: 30 years of development and clinical use. Eur J Pharm
Biopharm 2009;73:311e23.
10. Kumar P, Singh S, Mishra B. Development and evaluation of
elementary osmotic pump of highly water soluble drug: tramadol
hydrochloride. Curr Drug Deliv 2009;6:130e9.
11. Emara LH, Taha NF, El-Ashmawy AA, Raslan HM, Mursi NM.
Controlled porosity osmotic pump system for the delivery of diclo-
fenac sodium: in-vitro and in-vivo evaluation. Pharm Dev Technol
2014;19:681e91.
12. Liu L, Ku J, Khang G, Lee B, Rhee JM, Lee HB. Nifedipine controlled
delivery by sandwiched osmotic tablet system. J Control Release
2000;68:145e56.
13. Thombre AG, Cardinal JR, DeNoto AR, Gibbes DC. Asymmetric
membrane capsules for osmotic drug delivery II. In vitro and in vivo
drug release performance. J Control Release 1999;57:65e73.
14. Waterman KC, Goeken GS, Konagurthu S, Likar MD,
MacDonald BC, Mahajan N, et al. Osmotic capsules: a universal oral,
controlled-release drug delivery dosage form. J Control Release 2011;
152:264e9.
15. Wu C, Zhao ZZ, Zhao Y, Hao YN, Liu Y, Liu C. Preparation of a
push-pull osmotic pump of felodipine solubilized by mesoporous
silica nanoparticles with a core-shell structure. Int J Pharm 2014;475:
298e305.
16. Zhang ZH, Dong HY, Peng B, Liu HF, Li CL, Liang M, et al. Design
of an expert system for the development and formulation of push-pull
osmotic pump tablets containing poorly water-soluble drugs. Int J
Pharm 2011;410:41e7.
17. Malaterre V, Ogorka J, Loggia N, Gurny R. Approach to design push-
pull osmotic pumps. Int J Pharm 2009;376:56e62.
18. Wang Y, Wertheim DF, Jones AS, Coombes AG. Micro-CT in drug
delivery. Eur J Pharm Biopharm 2010;74:41e9.
19. Richardson JC, Bowtell RW, Mader K, Melia CD. Pharmaceutical
applications of magnetic resonance imaging (MRI). Adv Drug Deliv
Rev 2005;57:1191e209.
20. Markl D, Wang P, Ridgway C, Karttunen AP, Chakraborty M,
Bawuah P, et al. Characterization of the pore structure of functional-
ized calcium carbonate tablets by terahertz time-domain spectroscopy
and X-ray computed microtomography. J Pharm Sci 2017;106:
1586e95.
21. Kandpal LM, Cho B-K, Tewari J, Gopinathan N. Raman spectral
imaging technique for API detection in pharmaceutical microtablets.
Sens Actuators B Chem 2018;260:213e22.
2576 Abi Maharjan et al.
22. Yin XZ, Wu L, Li Y, Guo T, Li HY, Xiao TQ, et al. Visualization and
quantification of deformation behavior of clopidogrel bisulfate poly-
morphs during tableting. Sci Rep 2016;6:21770.
23. Yin XZ, Li HY, Guo Z, Wu L, Chen FW, de Matas M, et al. Quan-
tification of swelling and erosion in the controlled release of a poorly
water-soluble drug using synchrotron X-ray computed micro-
tomography. AAPS J 2013;15:1025e34.
24. Wu L, Wang LB, Wang SX, Xiao TQ, Chen M, Shao Q, et al. Three
dimensional structural insight of laser drilled orifices in osmotic pump
tablets. Eur J Pharm Sci 2016;93:287e94.
25. Yin XZ, Li HY, Liu RH, Chen J, Ji JQ, Chen J, et al. Fractal structure
determines controlled release kinetics of monolithic osmotic pump
tablets. J Pharm Pharmacol 2013;65:953e9.
26. Zhang L, Wu L, Wang CF, Zhang GQ, Yu L, Li HY, et al. Synchrotron
radiation microcomputed tomography guided chromatographic anal-
ysis for displaying the material distribution in tablets. Anal Chem
2018;90:3238e44.
27. Fang LW, Yin XZ, Wu L, He YP, He YZ, Qin W, et al. Classification
of microcrystalline celluloses via structures of individual particles
measured by synchrotron radiation X-ray micro-computed tomogra-
phy. Int J Pharm 2017;531:658e67.
28. Xu HM, Li Z, Pan H, Zhang ZH, Liu DD, Tian BC, et al. A novel bi-
layer ascending release osmotic pump tablet: in vitro investigation and
in vivo investigation in pharmacokinetic study and IVIVC evaluation.
Int J Pharm 2013;458:181e7.
29. Malaterre V, Metz H, Ogorka J, Gurny R, Loggia N, Mader K.
Benchtop-magnetic resonance imaging (BT-MRI) characterization of
push‒pull osmotic controlled release systems. J Control Release 2009;
133:31e6.
30. Li HY, Yin XZ, Ji JQ, Sun LX, Shao Q, York P, et al. Microstructural
investigation to the controlled release kinetics of monolith osmotic
pump tablets via synchrotron radiation X-ray microtomography. Int J
Pharm 2012;427:270e5.
31. Sun X, Wu L, Maharjan A, Sun HY, Hu XX, York P, et al. Static and
dynamic structural features of single pellets determine the release
behaviors of metoprolol succinate sustained-release tablets. Eur J
Pharm Sci 2020;149:105324.
32. Nakajima T, Takeuchi I, Ohshima H, Terada H, Makino K. Push-
Pull controlled drug release systems: effect of molecular weight of
polyethylene oxide on drug release. J Pharm Sci 2018;107:
1896e902.
33. Apicella A, Cappello B, Del Nobile MA, La Rotonda MI,
Mensitieri G, Nicolais L. Poly(ethylene oxide) (PEO) and different
molecular weight PEO blends monolithic devices for drug release. J
Biomater 1993;14:83e90.
34. Malaterre V, Ogorka J, Loggia N, Gurny R. Evaluation of the
tablet core factors influencing the release kinetics and the loadability
of push‒pull osmotic systems. Drug Dev Ind Pharm 2009;35:433e9.
35. Swanson DR, Barclay BL, Wong PSL, Theeuwes F. Nifedipine
gastrointestinal therapeutics system. Am J Med 1987;83:1e9.
Subterranean river model for elucidating release mechanism from bilayer tablets 2577
