1. Impact of formulation and process variables on solid-state stability of
theophylline in controlled release formulations$
Maxwell Korang-Yeboaha
, Ziyaur Rahmana
, Dhaval Shaha
, Adil Mohammada
,
Suyang Wua
, Akhtar Siddiquia
, Mansoor A. Khana,b,
*
a
Division of Product Quality Research, Center for Drug Evaluation and Research, Food and Drug Administration, MD, USA
b
Rangel College of Pharmacy, Texas A&M Health Science Center Reynolds Medical Building, Suite 159, College Station, TX 77843-1114, USA
A R T I C L E I N F O
Article history:
Received 21 August 2015
Received in revised form 10 November 2015
Accepted 26 November 2015
Available online 11 December 2015
Keywords:
Solid-state stability
Dissolution
Pseudopolymorphic transition
Theophylline
A B S T R A C T
Understanding the impact of pharmaceutical processing, formulation excipients and their interactions on
the solid-state transitions of pharmaceutical solids during use and in storage is critical in ensuring
consistent product performance. This study reports the effect of polymer viscosity, diluent type,
granulation and granulating fluid (water and isopropanol) on the pseudopolymorphic transition of
theophylline anhydrous (THA) in controlled release formulations as well as the implications of this
transition on critical quality attributes of the tablets. Accordingly, 12 formulations were prepared using a
full factorial screening design and monitored over a 3 month period at 40
C and 75%. Physicochemical
characterization revealed a drastic drop in tablet hardness accompanied by a very significant increase in
moisture content and swelling of all formulations. Spectroscopic analysis (ssNMR, Raman, NIR and PXRD)
indicated conversion of THA to theophylline monohydrate (TMO) in all formulations prepared by aqueous
wet granulation in as early as two weeks. Although all freshly prepared formulations contained THA, the
hydration–dehydration process induced during aqueous wet granulation hastened the pseudopoly-
morphic conversion of theophylline during storage through a cascade of events. On the other hand, no
solid state transformation was observed in directly compressed formulations and formulations in which
isopropanol was employed as a granulating fluid even after the twelve weeks study period. The transition
of THA to TMO resulted in a decrease in dissolution while an increase in dissolution was observed in
directly compressed and IPA granulated formulation. Consequently, the impact of pseudopolymorphic
transition of theophylline on dissolution in controlled release formulations may be the net result of two
opposing factors: swelling and softening of the tablets which tend to favor an increase in drug dissolution
and hydration of theophylline which decreases the drug dissolution.
Published by Elsevier B.V.
1. Introduction
Most pharmaceutical solids exist in more than one crystalline
form or in a disordered amorphous state. Some crystalline drugs
have the propensity of incorporating in their crystal lattice
solvents; either in a stoichiometric or non-stoichiometric way
(Hilfiker et al., 2006). The different solid forms exhibit dissimilar
physical, mechanical and chemical properties which may affect
their processability during product manufacturing, and alter
product performance, such as stability, dissolution, bioavailability
and clinical efficacy (Huang and Tong, 2004; Kobayashi et al., 2000;
Raw et al., 2004; Yu et al., 2003). Also, pharmaceutical processes
such as wet granulation, drying, milling; compression and
lyophilization can induce polymorphic transition during
manufacturing. Conversely, judicious selection of formulation
excipients can be employed to retard or inhibit solid-state
transition during processing and storage (Airaksinen et al.,
2004; Zhang et al., 2004).
Theophylline is a bronchodilator used in the management of
reversible airway obstruction associated with asthma and chronic
obstructive pulmonary disease. Currently, the therapeutic use of
theophylline in developed countries is restricted to patients whose
disease conditions are poorly controlled due to the drug's higher
incidence of side effects; nonetheless theophylline is the most
widely used bronchodilator due to its lower cost (Barnes, 2003;
ZuWallack et al., 2001). Theophylline exists either as a monoclinic
channel hydrate or an anhydrate. Additionally, anhydrous
$
Disclaimer: The views and opinions expressed in this paper are only those of the
authors, and do not necessarily reflect the views or policies of the FDA.
* Corresponding author at: Rangel College of Pharmacy, Texas AM Health
Science Center Reynolds Medical Building, Suite 159, College Station, TX 77843-
1114, USA. Fax: +1 979 436 0087.
E-mail address: mkhan@pharmacy.tamhsc.edu (M.A. Khan).
http://dx.doi.org/10.1016/j.ijpharm.2015.11.046
0378-5173/Published by Elsevier B.V.
International Journal of Pharmaceutics 499 (2016) 20–28
Contents lists available at ScienceDirect
International Journal of Pharmaceutics
journal homepage: www.elsevier.com/locate/ijpharm

2. theophylline has been identified to exist in three different forms;
Form I is the most stable state at high temperature, Form II, which
is used in pharmaceutical formulations is stable at room
temperature and Form III the metastable anhydrous intermediate
(Seton et al., 2010). However, the commonly encountered solid
state transition of theophylline during the pharmaceutical
manufacturing process and product storage is the transposition
between anhydrous Form II (THA) and theophylline monohydrate
(TMO) with or without Form III as an intermediate.
Ando et al. (1986) first reported the conversion of THA to TMO
upon storage at 90% relative humidity (Ando et al., 1986). This
transition was also reported to occur upon storage at a relative
humidity of 75% (Alvarez-Lorenzo et al., 2000; Phadnis and
Suryanarayanan,1997; Zhu et al.,1996). Additionally, THA has been
shown to convert to the monohydrate form irrespective of the
choice of formulation excipient used, during wet granulation
(Airaksinen et al., 2004; Jørgensen et al., 2004; Wikström et al.,
2008). The formation of anhydrous Form III has also been identified
upon drying of the wet theophylline granules (monohydrate)
under low pressure (Nunes et al., 2006; Phadnis and Suryanar-
ayanan, 1997; Tantry et al., 2007). These transitions are associated
with decrease in the dissolution and drug bioavailability owing to
the lower solubility of the monohydrate crystals as well as an
increase in binding interaction between theophylline and formu-
lation excipients (Alvarez-Lorenzo et al., 2000; Herman et al., 1988,
1989; Rodriguez-Hornedo et al., 1992). Theophylline has a narrow
therapeutic window and is associated with a high incidence of
adverse events and sudden deaths (eHealthMe, 2015). Conse-
quently, minor alterations in theophylline product quality may
have significant impact on the clinical efficacy and toxicity
observed in patients. For this reason, a thorough understanding
of the impact of the manufacturing process, excipient choice as
well as their interactions on the solid-state stability of theophylline
during storage is vital.
Although several authors have reported on the hydration and
dehydration of theophylline, most of these studies were either
conducted in binary mixtures of theophylline and excipients or in
immediate release formulations. However, most marketed the-
ophylline products are controlled release formulations. Secondly,
the impact of these solid-state transitions during manufacturing
on the storage stability of the product remains unexplained.
The present study was an attempt to investigate the impact of
excipient selection, manufacturing process and their interactions
on solid-state transitions of theophylline in controlled release
formulation during storage and use. Physicochemical and spectro-
scopic characterizations were carried to monitor for any solid-state
transitions as well changes in product quality attributes.
2. Materials and method
2.1. Materials
Hydroxypropyl methylcellulose K4M and K100M (Colorcon,
Harleysville, PA, USA), theophylline anhydrous Form II (THA),
magnesium stearate (MgS) (Sigma, St. Louis, MO, USA), lactose
monohydrate (LM) and anhydrous lactose (LA) (Foremost farms,
Baraboo, WI, USA), Colloidal silicon dioxide (Aerosil 200)(Evonik,
Parsippany, NJ, USA), polyvinylpyrrolidone (PVP K15, K30 and K90)
(Sigma Aldrich, St. Louis, MO, USA) were used.
2.2. Methods
2.2.1. Design of experiment
The effect of formulation and process variables on solid-state
stability and product quality were assessed using a full factorial
design. The most commonly used formulation excipients and
process variables were chosen for this study. The formulations
variables studied were; the polymer viscosity/molecular weight
(HPMC K4M and HPMC K100M) and the diluent (LA and LM). The
formulation composition was as follows: THA 53.33%, K4M/K100M
33.33%, LA/LM 10.7%, Aerosil 0.1% and MgS 2.5%. In addition, the
impact of the manufacturing processes (wet granulation and direct
compression) and the granulating fluid employed (water and
isopropanol) during wet granulation were considered. The
experimental design and data analysis was conducted with JMP
software version 11.1.1 from SAS (Cary, NC, USA). In all twelve
formulations were prepared according to the experimental design
shown in Table 1.
2.2.2. Granulation and tableting
Mixing and granulation were performed with the KG-5 high
shear granulator/mixer (Key International, Cranbury, New Jersey,
USA). Solid-state transitions during granulation were monitored by
in-line Luminar 5030 AOTF-NIR probe (Sparks Glencoe, Maryland,
USA) and offline X-ray powder diffractometry. The wet granules
were dried at 50
C until the moisture content was below 2%. The
moisture content of the dried granules was determined by loss on
drying (MB 45 moisture analyzer, Ohaus Corporation, Parsippany,
NJ, USA). The dried granules were sieved, blended with magnesium
stearate and aerosil 200. The granules were compressed into
tablets with a rotary tablet press using a 10 mm punch size (Globe
Pharma, New Brunswick, New Jersey, USA). The initial tablet
hardness was 6–8 kp for all the formulations.
2.2.3. Physicochemical characterization
Moisture content analysis was performed with Karl Fisher
V30 compact titrator from Mettler Toledo- (Columbus OH, USA)
using Aquastar1
Comp-2 Karl fisher reagent (EMD Millipore,
Billerica, MA, USA). About 100 mg of powdered sample was used
for moisture analysis. tablet hardness was measured with the PTB
11EP hardness tester (Pharma Test, Hainburg, Germany). Scanning
electron microscopy (JSM-6390 LV- JEOL, Tokyo, Japan) images of
the tablets were taken before and after stability studies at a
magnification of 100X. The dissolution profiles were obtained with
USP dissolution apparatus I (basket) at 100 rpm. The dissolution
media used was 900 ml of 0.05 M phosphate buffer pH 6.6 main-
tained at 37
C. Sample collection was done over 24 hrs and
analyzed for their theophylline content. HPLC analysis was
conducted with an Agilent 1260HPLC system equipped with an
auto sampler, a quaternary pump, diode array detector set at
271 nm wavelengths, and column temperature maintained at
25
C. A Zorbax1
eclipse plus C-18 column (4.6 Â100 mm, 3.5 mm
packing) was used with a mobile phase composition of 7%
acetonitrile and 93% acetate buffer (10 mM pH 3.5) run isocratically
at 1 ml/min.
Table 1
DOE of formulations used (abbreviations used: IPA-isopropanol; DC-direct
compression).
Pattern Polymer Diluent Granulating fluid
F1 À+À HPMC K4M LM IPA
F2 +ÀÀ HPMC K100M LA IPA
F3 ÀÀ+ HPMC K4M LA Water
F4 +++ HPMC K100M LM Water
F5 À++ HPMC K4M LM Water
F6 +À+ HPMC K100M LA Water
F7 ++À HPMC K100M LM IPA
F8 ÀÀÀ HPMC K4M LA IPA
F9 22 HPMC K100M LM DC
F10 21 HPMC K100M LA DC
F11 12 HPMC K4M LM DC
F12 11 HPMC K4M LA DC
M. Korang-Yeboah et al. / International Journal of Pharmaceutics 499 (2016) 20–28 21

3. 2.2.4. Spectroscopic characterization
Raman spectra of THA, TMO and the stability samples were
obtained with non-contact Raman probe (RamanRXN2TM Multi-
Channel Raman Analyzer, Kaiser Optical System Inc., Ann Arbor, MI,
USA). NIR spectra were also collected over the wavelength range of
1100–2500 nm (FOSS NIR System Inc., Laurel, MD, USA) as
previously described (Korang-Yeboah et al., 2015; Rahman et al.,
2015a). X-ray diffraction and C-13 ssNMR spectra patterns were
measured with Bruker D8 Advance (Bruker AXS, Madison, WI, USA)
and Varian VNMR 400 spectrometer (Varian Inc. Palo Alto,
California) respectively as described in earlier studies. (Rahman
et al., 2015b) The extent of transformation was quantified from the
XRPD data by the univariate peak area approach using characteristic
THA and TMO peaks at 7.2, and 11.5 2u (Otsuka and Kinoshita, 2010).
3. Results
3.1. Spectroscopic characterization
The differences between THA and TMO were studied using NIR,
Raman, XRPD and ssNMR. These differences were exploited in
monitoring for solid state transitions of theophylline during
manufacturing and in storage.
3.1.1. NIR
NIR spectra of THA, TMO, freshly prepared formulations and
stability samples are shown in Fig. 1. NIR absorption bands are
mainly due to overtones and combination bands of C-H, O-H, N-H
and S-H bonds and hence are highly susceptible to the presence of
moisture. THA and freshly prepared tablets showed combination
peaks of water and ÀÀOH vibration (first overtone) at 1930 nm and
1450 nm respectively. However, these peaks are also common to
LM and THA and not very useful in monitoring of pseudopoly-
morphic changes. Anhydrate to hydrate transition of theophylline
anhydrous Form II led to the generation of newer peaks and shifts
in already present peaks. Notable was the presence of an
absorption maxima at 1970 nm (water of crystallization) which
was absent in the initial samples. Variations in these peak maxima
were used to monitor the solid state transition in the formulations.
3.1.2. Raman spectra
Raman spectra of THAand freshly prepared tablets showed peaks
in the spectral regions of 3200–2800 cmÀ1
and 100–1750 cmÀ1
.
The spectra consisted of stretching modes of CÀÀH (2970 cmÀ1
, and
3120 cmÀ
1), C¼O (1662 cmÀ1
and 1704 cmÀ1
), C¼C (1610 cmÀ1
),
C¼N (1571 cmÀ1
), O¼CÀÀN (554 cmÀ1
), ÀÀCH3 deformation
(1424 cmÀ1
) and rocking bands (928 cmÀ1
) (Ahlneck and Zografi,
1990; Edwards et al., 2005; Jørgensen et al., 2002). Although Raman
spectroscopy is insensitive to the presence of water, hydration of
theophylline leads to changes in hydrogen bond interactions which
directly alter molecular vibrations. Transition of THA to TMO led to
band modifications along the entire spectral range. The most
prominent modificationwas the replacementof the double carbonyl
peaks at 1662 cmÀ1
and 1704 cmÀ1
with a single peak at 1686 cmÀ1
(Fig. 2). Also majority of the Raman bands shifted either to a higher
wavelength or a lower wavelength. These variations were observed
with time in some formulations during storage at accelerated
stability conditions.
3.1.3. XRPD analysis
The XRPD pattern of THA, TMO, freshly prepared tablets and
stability samples are shown in Fig. 3. The diffraction curves of THA
and all freshly prepared tablets closely resembled previously
reported patterns for orthorhombic THA with characteristic peaks
at 7.2
and 12.5
(Edwards et al., 2005). However, the XRPD pattern
of TMO showed distinctive peaks at 8.8
, 11.5
and 27
2u and the
absence of characteristic THA peaks at 7.2
, and 12.5
. The presence
of excipients did not interfere with the XRPD peaks of THA or TMO.
Consequently, variations in these peaks were used as the measure
of the solid state stability of the formulations during manufactur-
ing and storage.
3.1.4. ssNMR
ssNMR spectra for THA, freshly prepared formulation F3, TMO
and stability samples are shown in Fig. 4. The 13C ssNMR spectra
assignments of the freshly prepared tablets were similar to
previously reported(Edwards et al., 2005; Nolasco et al., 2006).
THA showed peaks due to carbonyl carbons C-6 and C-2 at
155.12 ppm, 150.93 ppm, respectively and methine carbons, C-
4 and C-8 at 146.14 ppm, 140.96 ppm correspondingly. In addition,
peaks corresponding to methyl group at C-10 and the methine
carbon at position 5 were obtained at 29.46 and 107 ppm. The most
notable difference in the spectra of the TMO crystals was observed
as a shift of the carbonyl peaks from 150.93 ppm to 148.4 ppm.
Anhydrate to hydrate transition of THA was monitored in the
Fig. 1. (A) Raw and (B) 2nd derivative NIR spectra of formulation F3 before and during accelerated stability studies.
22 M. Korang-Yeboah et al. / International Journal of Pharmaceutics 499 (2016) 20–28

4. region of 160–130 ppm due to the absence of interferences from
the excipients.
3.1.5. Process induced transitions
NIR spectra obtained in-line during wet granulation indicated
conversion of THA to TMO for formulations F3–F6. This was
detected by the appearance of crystalline OH peaks at 1970 nm and
1468 nm. XRPD analysis conducted off-line also confirmed the
transformation of THA to TMO when water was used for
granulation. XRPD diffractogram of the wet granules showed the
presence of characteristic TMO peaks at 8.8
and 11.5
and the loss
of the distinct anhydrous peak at 7.1
. However, no transitions were
observed when IPA was used as the granulating fluid or in directly
compressed formulations (data not shown). The observed tran-
sitions were in agreement with already published studies on
pseudo polymorphic transitions of theophylline (Airaksinen et al.,
2004; Herman et al., 1988). In contrast to earlier studies, oven
drying resulted in conversion of TMO toTHA without the formation
of Form III (Airaksinen et al., 2004; Phadnis and Suryanarayanan,
1997; Tantry et al., 2007). This difference could be attributed to the
differences in pressure conditions and duration of the drying
process. Also, tablet compression did not alter the solid state form
of the drug as no differences were observed in the XRPD patterns
before and after tableting.
3.1.6. Physicochemical characterization
There was a significant (p 0.05) increase in tablet weight for
all 12 formulations after 3 months of accelerated stability testing
(40
C 75% RH). Additionally, all the formulations swelled
significantly in both axial and radial directions, as indicated by
increase in tablet diameter thickness (about 15% and 4%
respectively), and almost 7% increase in average tablet weight.
This was also accompanied by a statistically significant (p 0.05)
increase in tablet moisture content. The average moisture content
in F3, F4, F5 and F6 was about 6.32%, representing about a 4-fold
increase from the initial average value of 1.42%. Moreover, the
average moisture content for F1, F2 and F7–F12 doubled from the
initial content of 1.40% to about 3.02% (Table 2).
Furthermore, the hardness of all the tablets dropped signifi-
cantly after two weeks of storage with moderate changes observed
afterwards. SEM images showed an increase in surface roughness
and tablet porosity due to surface erosion, gelation of the polymer
and swelling of the tablets (Fig. 5).
3.1.7. Solid state stability
All spectra techniques employed showed the presence of TMO
in F3–F6 after two weeks of the accelerated stability studies.
However there were no changes in the spectral patterns of F1,
F2 and F7–F12. Anhydrate to hydrate conversion of THA to TMO
during storage was observed in all tablets prepared by wet
granulation by the appearance of characteristic peaks at 8.8,
11.5 and 27
in the XRPD pattern. The ssNMR spectra also showed
an extra carbonyl peak at 148.66 in addition to the two anhydrous
carbonyl peaks at 155.12 ppm and 150.93 ppm. The peak intensity
of the carbonyl carbon due to theophylline hydration was found to
increase with time.
However, there were no transformations in directly compressed
formulations and in formulations in which isopropanol was used
as the granulating fluid even after 12 weeks of accelerated stability
conditions. There was about 18% pseudopolymorphic conversion in
formulations F3 and F4 and about 10% in F4 and F5 after two weeks.
The transformation of THA to TMO in the tablets was made up of
three phases. An initial phase of rapid transition was followed by
the next phase in which very little to no change was observed and
Fig. 2. Changes in Raman spectra of formulations A. F1 and B. F3 during 12 weeks accelerated stability studies.
M. Korang-Yeboah et al. / International Journal of Pharmaceutics 499 (2016) 20–28 23

5. the final phase from week 8 to 12 in which significant changes were
observed in the extent of transformation. At the end of the study
period, about 90% of THA had converted to TMO in F3, F4 and F5;
and 77% conversion in F6.
3.2. Dissolution studies
The mechanism of the drug release was determined from the
dissolution profiles using the semi-empirical model developed by
Korsmeyer et al. (1983). The release exponent (n) which describes
the drug transport mechanism was calculated from regression
analysis using the equation
Mt
M1
¼ Kn
t
where Mt/M1 is the fractional drug release and K, a constant
incorporating structural and geometric characteristic of the tablet.
An n value of 0.45 is indicative of diffusion controlled drug release
while swelling controlled drug release are characterized by an n
value of 0.89. A value of n between 0.45 and 0.89 suggests the
mechanism of drug release is by anomalous transport. The
dissolution profiles of the formulations were well described by
the Korsmeyer–Peppas model with an R2
value greater than 0.99 in
all samples. The average diffusional exponent n was 0.67 (0.64–
0.73) indicating the transport mechanism of THA from the
formulation was by a combination of diffusion-controlled and
swelling-controlled release.
Additionally, the impact of the observed pseudopolymorphic
transition of theophylline on the drug dissolution profile was
studied by comparing the dissolution profiles of F3–F6 after
12 weeks to the initial samples and unhydrated THA formulations
F1 and F2. In contrast to the dissolution profile of the initial
samples there was a moderate drop in dissolution in formulations
F–F5 after 12 weeks. However, the decrease in dissolution of F6 at
the 24 h time point was not as pronounced as in F3–F5 although
over 75% of THA had converted to TMO. The extent of dissolution
was lowered by 12.33% in F3, 13.70% in F4, 12.21% in F5 and about
5.60% in F6. Conversely, there was an increase in the extent of
dissolution in F1 and F2 after the 12 week period (Fig. 6).
4. Discussion
Theophylline for the past seven decades has been the most
widely used bronchodilator due to its lower cost. Although the four
different forms of anhydrous theophylline have been identified so
far, anhydrous theophylline form II is the only form used in
pharmaceutical preparations of theophylline due to its superior
stability at room temperature. The crystalline packing of theoph-
ylline Form II is characterized by hydrogen bonding between
NÀÀHÁ Á ÁN and two bifurcated hydrogen bonds between CÀÀHÁ Á ÁO,
forming a bilayer structure. However, at higher storage humidity or
during aqueous granulation, the theophylline molecule undergoes
dimerization in the presence of water to form a monoclinic channel
hydrate. This leads to changes in the physicochemical
Fig. 3. PXRD pattern of formulations (A) F1 and (B) F3.
24 M. Korang-Yeboah et al. / International Journal of Pharmaceutics 499 (2016) 20–28

6. characteristics of the drug and ultimately drug quality attributes
such as stability, bioavailability and clinical efficacy. In the current
study, anhydrate to hydrate transition of THA was observed during
wet granulation using in-line NIR probes and off-line XRPD. Oven
heating led to reconversion of TMO to THA. However, unlike in
previous reports, the formation of anhydrous theophylline Form III
as an intermediate was not observed. This could be attributed to
the longer duration of drying used and pressure differences.
Although this hydration–dehydration transition seems innocuous,
it had direct impact on the storage stability and product quality.
Theophylline formulations F3–F6 manufactured using aqueous
wet granulation technique had significantly higher moisture
content after 12 weeks of accelerated stability studies than in
directly compressed formulations and when isopropanol was used
for granulation (Table 2). The hydration–dehydration process is
known to reduce both surface and bulk crystallinity of drug
products(Hüttenrauch et al., 1985; Murphy et al., 2002). Since THA
is moderately soluble in water, it dissolves during the wet
granulation process and precipitates as the monohydrate crystals
but reconverts to a more amorphous form of THA upon drying.
Upon drying, there is a loss of crystalline structure and formation
of amorphous THA hence a drop in crystallinity. The absence of
moisture and poor solubility of THA in isopropanol precludes this
transition in directly compressed formulations and when IPA was
Fig. 4. ssNMR spectra of THA, TMO and formulations (A) F1 and (B) F3.
Table 2
Moisture content, extent of pseudopolymorphic conversion of theophylline and hardness of theophylline formulations over the study period.
Moisture content (% w/w) Theophylline Monohydrate(%w/w) Hardness(kP)
Initial WK 2 WK 4 WK 8 WK 12 WK 2 WK 4 WK 8 WK 12 Initial WK 12
F1 1.41(Æ0.09) 2.22(Æ0.22) 2.75(Æ0.03) 3.08(Æ0.01) 3.79(Æ0.09) – – – – 8.13(7.9–8.6) 4.05(3.1–4.6)
F2 0.88(Æ0.01) 2.20(Æ0.08) 2.77(Æ0.07) 2.65(Æ0.34) 3.38(Æ0.01) – – – – 7.05(5.5–8.3) 4.50(3.8–5.2)
F3 1.42(Æ0.05) 4.28(Æ0.29) 5.94(Æ0.03) 6.24(Æ0.00) 7.30(Æ0.09) 18 72 73 88 6.75(6.0–7.7) 5.91(5.0–6.5)
F4 1.51(Æ0.06) 5.65(Æ0.17) 5.87(Æ0.03) 6.51(Æ0.17) 7.38(Æ0.00) 17 72 74 93 7.15(6.8–7.3) 6.22(5.8–6.2)
F5 1.48(Æ0.01) 4.59(Æ0.20) 5.72(Æ0.02) 6.27(Æ0.02) 7.57(Æ0.80) 12 67 65 88 7.88(7.0–8.5) 5.48(4.4–6.6)
F6 1.28(Æ0.01) 3.90(Æ0.11) 5.34(Æ0.15) 6.26(Æ0.05 7.41(Æ0.05) 10 48 51 77 7.05(6.1–7.6) 4.50(3.4–5.7)
F7 1.75(Æ0.28) 2.41(Æ0.04) 2.77(Æ0.09) 3.15(Æ0.01) 3.98(Æ0.20) – – – – 6.93(5.9–6.4) 3.54(3.1–3.9)
F8 1.22(Æ0.15) 2.39(Æ0.08) 2.64(Æ0.09) 2.99(Æ0.05) 3.66(Æ0.30) – – – – 6.83(6.3–8.7) 4.58(3.8–5.2)
F9 2.06(Æ0.10) 2.35(Æ0.34) 3.03(Æ0.20) 2.94(Æ0.01) 3.88(Æ0.04) – – – – 6.77(5.9–7.3) 3.25(3.0–3.4)
F10 1.03(Æ0.05) 1.95(Æ0.34) 2.92(Æ0.10) 3.00(Æ0.02) 4.01(Æ0.01) – – – – 7.58(6.2–9.0) 3.55(3.2–3.9)
F11 1.65(Æ0.10) 2.26(Æ0.08) 3.24(Æ0.27) 3.17(Æ0.07) 4.2(Æ0.28) – – – – 7.55(7.3–7.9) 3.50(3.1–3.9)
F12 1.15(Æ0.10) 1.94(Æ0.06) 3.12(Æ0.18) 3.24(Æ0.03) 4.28(Æ0.34) – – – – 6.77(6.4–7.5) 3.31(2.9–3.5)
M. Korang-Yeboah et al. / International Journal of Pharmaceutics 499 (2016) 20–28 25

7. used as a granulating fluid. Earlier studies by Debnath and
Suryanarayanan (2004) reported a reduction of drug crystallinity
by 25% due to the hydration–dehydration process (Debnath and
Suryanarayanan, 2004). This increase in amorphous content might
account for the significantly higher moisture uptake kinetics in
formulations F3–F6. Also, all formulations increased in tablet size
and weight after 12 weeks of accelerated stability studies due to
the hygroscopic nature of the polymer. tablet hardness dropped
significantly in all formulations. The surface erosion and cracks
observed in SEM images may be due to tablet swelling and
dissolution of the polymer and drug in the sorbed water (Fig. 5).
Furthermore, tablets manufactured by aqueous wet granulation
technique underwent anhydrate to hydrate transition as early as in
two weeks at accelerated stability conditions. About 90% of THA in
F3–F5 had undergone anhydrate to hydrate transition at the end of
the study and more than 75% of THA transformed to TMO in F6.
However the rest of the formulations (directly compressed and
granulation with isopropanol) exhibited no change in the solid
state nature of THA after 12 weeks of accelerated stability studies.
The significant difference in the stability of the formulations could
be attributed to a combination of process induced factors. The
transformation of THA to TMO is known to be preceded by the
dissolution of THA form II in sorbed water to form a supersaturated
solution (Alvarez-Lorenzo et al., 2000; Ando et al., 1992; Otsuka
et al., 1990; Zhu et al., 1996). The next stage of the transformation
involves nucleation of the monohydrate crystals followed by
crystal growth until the drug concentration of the solution has
decreased to the solubility of the stable form (theophylline
monohydrate). Interestingly, the hydration–dehydration tends to
promote each of these phases. First, the reduction in both surface
and bulk crystallinity of THA results in a faster super saturation and
subsequently a higher rate of crystallization. Secondly, the
presence of seeds of the monohydrate crystals hastens the process
by lowering the nucleation barrier required to advance from a
supersaturated solution to crystal growth (Cacciuto et al., 2004;
Giron, 2001; Giron et al., 1990; Kelton, 1991). A similar observation
was reported in phenylbutazone in which no solid state
transformation was observed after 3 years of storage, however
in the presence of trace amounts of Form B, significant levels of
transformation was observed in only 6 months (Giron et al., 1990).
Also, the anhydrate to hydrate transition of theophylline in
controlled release matrices of HPMC may involve three phases
(Table 2). An initial phase in which a dramatic increase in the rate
of transformation is observed followed by a tapering off period in
Fig. 5. SEM images of tablet surfaces before and after 12 weeks of accelerated stability studies.
26 M. Korang-Yeboah et al. / International Journal of Pharmaceutics 499 (2016) 20–28

8. which no significant transformation and the final phase where
there was a further increase in the extent of anhydrate to hydrate
conversion. The initial higher rate of hydration of theophylline
could be attributed to the faster reactivity of THA with the sorbed
bulk water than the HPMC or water migration from HPMC. The
existence of a moisture gradient between the tablet surface in
contact with sorbed moisture and the dry inner core might account
for the second phase in which no significant transition of THA to
TMO is observed. This gradient limits the availability of moisture to
the unaltered THA. As the sorbed water seeps further into the
matrix, an increase in the amount of TMO is observed once more as
the remaining THA reacts with the bulk sorbed water.
The DOE study showed the use of aqueous wet granulation
technique as the single most important factor that affects
anhydrate to hydrate transition of theophylline in the controlled
release formulations during storage. The polymer molecular
weight, the diluent used and the interaction between the
formulation factors did not have significant effect (p 0.05) on
the transition of THA to TMO.
Furthermore, there was a drop in the dissolution of F3–F5,
however, no significant change was observed in the dissolution of
F6. On the other hand there was an increase in dissolution of the
unhydrated tablets. The extent at which the pseudopolymorphic
transition of theophylline impacts the dissolution profile was not
as remarkable as previously reported by Ando et al. (1995) in
immediate (IR) release formulations of theophylline in which
hydration of theophylline led to over 50% drop in dissolution (Ando
et al.,1995). This may be due to the presence of factors that oppose
the effects of theophylline hydration on drug dissolution. The
swelling and softening of the tablets due to moisture uptake favors
an increase in drug dissolution, in contrast, hydration of
theophylline leads to a decrease in dissolution due to a decrease
in solubility and dissolution rate. In addition, hydration of
theophylline during storage leads to increase in binding between
theophylline and cellulose polymers which causes a decrease in
dissolution (Herman et al., 1989). The observed effect on the
dissolution is possibly the net results of the above factors. In
theophylline formulations F3–F5 the observed decrease was
possibly due to the dominating effects of theophylline hydration
and increase in theophylline polymer binding. In F6, these factors
could not override the effect due tablet swelling and the decrease
in hardness hence lowering the impact of theophylline hydration
in F6.
5. Conclusion
The solid state stability of many pharmaceuticals during storage
and in-use is highly affected by the processing factors employed
during manufacture, the choice of excipients and interactions
between these factors. For drugs with narrow therapeutic index
such as theophylline, minor alterations in the product quality may
either result in loss of clinical efficacy or toxicity hence the need for
a thorough understanding of how formulation excipients and
process affects the storage stability. The use of aqueous wet
granulation led to hydrate formation of theophylline. Although this
process was reversible during drying, the hydration–dehydration
of theophylline significantly affected the solid state stability in
controlled release formulations through a sequence of events that
ultimately increases the rate of hydrate conversion. Conversely, the
molecular weight of HPMC, the diluent type and the interaction
between these factors and process did not affect the rate of
anhydrous to hydrate transition. There was more than 10%
Fig. 6. Dissolution curves for hydrated and unhydrated formulations before and after 12 weeks of accelerated stability studies.
M. Korang-Yeboah et al. / International Journal of Pharmaceutics 499 (2016) 20–28 27

9. decrease in the dissolution rate of most of the hydrated tablets.
However the clinical significance of this reduction needs to be
further evaluated as theophylline is a narrow therapeutic index
drug. In addition, most spectroscopic techniques were effective in
monitoring and characterizing the anhydrate to hydrate transition
of theophylline in controlled release formulations. The authors are
currently exploring various analytical techniques for quantifying
these transitions in theophylline products.
References
Ahlneck, C., Zografi, G.,1990. The molecular basis of moisture effects on the physical
and chemical stability of drugs in the solid state. Int. J. Pharm. 62, 87–95.
Airaksinen, S., Karjalainen, M., Rasanen, E., Rantanen, J., Yliruusi, J., 2004.
Comparison of the effects of two drying methods on polymorphism of
theophylline. Int. J. Pharm. 276, 129–141.
Alvarez-Lorenzo, C., Gómez-Amoza, J.L., Martínez-Pacheco, R., Souto, C., Concheiro,
A., 2000. The stability of theophylline tablets with a hydroxypropylcellulose
matrix. Drug Dev. Ind. Pharm. 26, 13–20.
Ando, H., Ishii, M., Kayano, M., Ozawa, H., 1992. Effect of moisture on crystallization
of theophylline in tablets. Drug Dev. Ind. Pharm. 18, 453–467.
Ando, H., Ishii, M., Kayano, M., Watanabe, S., 1995. Effect of crystallization of
theophylline on physical properties of tablets. Drug Dev. Ind. Pharm. 21,
2227–2237.
Ando, H., Takayuki, O., Masaaki, I., Suniio, W., Yasuo, M., 1986. Crystallization of
theophylline in tablets. Int. J. Pharm. 34, 153–156.
Barnes, P.J., 2003. Theophylline. Am. J. Respir. Crit. Care Med. 167, 813–818.
Cacciuto, A., Auer, S., Frenkel, D., 2004. Onset of heterogeneous crystal nucleation in
colloidal suspensions. Nature 428, 404–406.
Debnath, S., Suryanarayanan, R., 2004. Influence of processing-induced phase
transformations on the dissolution of theophylline tablets. AAPS PharmSciTech
5, 39–49.
Edwards, H.G.M., Munshi, T., Anstis, M., 2005. Raman spectroscopic
characterisations and analytical discrimination between caffeine and
demethylated analogues of pharmaceutical relevance. Spectrochim. Acta A 61,
1453–1459.
eHealthMe, 2015. Review: theophylline side effects eHealthMe.
Giron, D., 2001. Investigations of polymorphism and pseudo-polymorphism in
pharmaceuticals by combined thermoanalytical techniques. J. Therm. Anal.
Calorim. 64, 37–60.
Giron, D., Edel, B., Piechon, P., 1990. X-ray quantitative determination of
polymorphism in pharmaceuticals. Mol. Cryst. Liq. Cryst. Incorp. Nonlin. Opt.
187, 297–307.
Herman, J., Remon, J.P., Visavarungroj, N., Schwartz, J.B., Klinger, G.H., 1988.
Formation of theophylline monohydrate during the pelletisation of
microcrystalline cellulose-anhydrous theophylline blends. Int. J. Pharm. 42, 15–
18.
Herman, J., Visavarungroj, N., Remon, J.P., 1989. Instability of drug release from
anhydrous theophylline-imcrocrystalline cellulose formulations. Int. J. Pharm.
55, 143–146.
Hilfiker, R., Blatter, F., Raumer, M.v., 2006. Relevance of Solid-state Properties for
Pharmaceutical Products, Polymorphism. Wiley-VCH Verlag GmbH Co. KGaA,
pp. 1–19.
Huang, L.-F., Tong, W.-Q., 2004. Impact of solid state properties on developability
assessment of drug candidates. Adv. Drug Deliv. Rev. 56, 321–334.
Hüttenrauch, R., Fricke, S., Zielke, P., 1985. Mechanical activation of pharmaceutical
systems. Pharm. Res. 2, 302–306.
Jørgensen, A., Rantanen, J., Karjalainen, M., Khriachtchev, L., Räsänen, E., Yliruusi, J.,
2002. Hydrate formation during wet granulation studied by spectroscopic
methods and multivariate analysis. Pharm. Res. 19, 1285–1291.
Jørgensen, A.C., Airaksinen, S., Karjalainen, M., Luukkonen, P., Rantanen, J., Yliruusi,
J., 2004. Role of excipients in hydrate formation kinetics of theophylline in wet
masses studied by near-infrared spectroscopy. Eur. J. Pharm. Sci. 23, 99–104.
Kelton, K.F., 1991. Crystal nucleation in liquids and glasses. In: Henry, E., David, T.
(Eds.), Solid State Physics. Academic Press, pp. 75–177.
Kobayashi, Y., Ito, S., Itai, S., Yamamoto, K., 2000. Physicochemical properties and
bioavailability of carbamazepine polymorphs and dihydrate. Int. J. Pharm. 193,
137–146.
Korang-Yeboah, M., Akhtar, S., Siddiqui, A., Rahman, Z., Khan, M.A., 2015. Application
of NIR chemometric methods for quantification of the crystalline fraction of
warfarin sodium in drug product. Drug Dev. Ind. Pharm. 1–11.
Korsmeyer, R.W., Gurny, R., Doelker, E., Buri, P., Peppas, N.A., 1983. Mechanisms of
solute release from porous hydrophilic polymers. Int. J. Pharm. 15, 25–35.
Murphy, D., Rodrı’guez-Cintrón, F., Langevin, B., Kelly, R.C., Rodrı’guez-Hornedo, N.,
2002. Solution-mediated phase transformation of anhydrous to dihydrate
carbamazepine and the effect of lattice disorder. Int. J. Pharm. 246,
121–134.
Nolasco, M.M., Amado, A.M., Ribeiro-Claro, P.J.A., 2006. Computationally-assisted
approach to the vibrational spectra of molecular crystals: study of hydrogen-
bonding and pseudo-polymorphism. ChemPhysChem 7, 2150–2161.
Nunes, C., Mahendrasingam, A., Suryanarayanan, R., 2006. Investigation of the
multi-step dehydration reaction of theophylline monohydrate using 2-
dimensional powder X-ray diffractometry. Pharm. Res. 23, 2393–2404.
Otsuka, M., Kaneniwa, N., Kawakami, K., Umezawa, O., 1990. Effect of surface
characteristics of theophylline anhydrate powder on hygroscopic stability. J.
Pharm. Pharmacol. 42, 606–610.
Otsuka, M., Kinoshita, H., 2010. Quantitative determination of hydrate content of
theophylline powder by chemometric X-ray powder diffraction analysis. AAPS
PharmSciTech 11, 204–211.
Phadnis, N.V., Suryanarayanan, R.,1997. Polymorphism in anhydrous theophylline—
implications on the dissolution rate of theophylline tablets. J. Pharm. Sci. 86,
1256–1263.
Rahman, Z., Mohammad, A., Akhtar, S., Siddiqui, A., Korang-Yeboah, M., Khan, M.A.,
2015a. Chemometric model development and comparison of Raman and 13
C
solid-state nuclear magnetic resonance—chemometric methods for
quantification of crystalline/amorphous warfarin sodium fraction in the
formulations. J. Pharm. Sci. 104, 2550–2558.
Rahman, Z., Mohammad, A., Siddiqui, A., Khan, M., 2015b. Comparison of univariate
and multivariate models of 13
C SSNMR and XRPD techniques for quantification
of nimodipine polymorphs. AAPS PharmSciTech 1–9.
Raw, A.S., Furness, M.S., Gill, D.S., Adams, R.C., Holcombe Jr, F.O., Yu, L.X., 2004.
Regulatory considerations of pharmaceutical solid polymorphism in
abbreviated new drug applications (ANDAs). Adv. Drug Deliv. Rev. 56,
397–414.
Rodriguez-Hornedo, N., Lechuga-Ballesteros, D., Wu, H.J., 1992. Phase-transition
and heterogeneous epitaxial nucleation of hydrated and anhydrous
theophylline crystals. Int. J. Pharm. 85, 149–162.
Seton, L., Khamar, D., Bradshaw, I.J., Hutcheon, G.A., 2010. Solid state forms of
theophylline: presenting a new anhydrous polymorph. Cryst. Growth Des. 10,
3879–3886.
Tantry, J.S., Tank, J., Suryanarayanan, R., 2007. Processing-induced phase transitions
of theophylline—implications on the dissolution of theophylline tablets. J.
Pharm. Sci. 96, 1434–1444.
Wikström, H., Carroll, W., Taylor, L., 2008. Manipulating theophylline monohydrate
formation during high-shear wet granulation through improved understanding
of the role of pharmaceutical excipients. Pharm. Res. 25, 923–935.
Yu, L., Furness, M.S., Raw, A., Outlaw, K.W., Nashed, N., Ramos, E., Miller, S.F., Adams,
R., Fang, F., Patel, R., Holcombe Jr., F., Chiu, Y.-y., Hussain, A., 2003. Scientific
considerations of pharmaceutical solid polymorphism in abbreviated new drug
applications. Pharm. Res. 20, 531–536.
Zhang, G.G.Z., Law, D., Schmitt, E.A., Qiu, Y., 2004. Phase transformation
considerations during process development and manufacture of solid oral
dosage forms. Adv. Drug Deliv. Rev. 56, 371–390.
Zhu, H., Yuen, C., Grant, D.J.W., 1996. Influence of water activity in organic
solvent + water mixtures on the nature of the crystallizing drug phase. 1.
Theophylline. Int. J. Pharm. 135, 151–160.
ZuWallack, R.L., Mahler, D.A., Reilly, D., Church, N., Emmett, A., Rickard, K., Knobil, K.,
2001. SAlmeterol plus theophylline combination therapy in the treatment of
copd*. Chest 119, 1661–1670.
28 M. Korang-Yeboah et al. / International Journal of Pharmaceutics 499 (2016) 20–28