Journal of Controlled Release 91 (2003) 477488

www.elsevier.com / locate / jconrel

An investigation into the structure and bioavailability of

a-tocopherol dispersions in Gelucire 44 / 14
S.A. Barker a , S.P. Yap b , K.H. Yuen b , C.P. McCoy a , J.R. Murphy a , D.Q.M. Craig a , *
a

The School of Pharmacy, The Queen s University of Belfast, 97 Lisburn Road, Belfast BT9 7 BL, UK
b
The School of Pharmacy, University of Science Malaysia, 11800 Penang, Malaysia
Received 17 May 2002; accepted 30 July 2002

Abstract
In this investigation we describe the preparation, physical characterisation and in vivo behaviour of solid dispersions of a
liquid nutraceutical, a-tocopherol, in Gelucire 44 / 14 with a view to establishing whether dispersion in this matrix may
provide a means of formulating a liquid drug in a solid dosage form while also improving the oral bioavailability. Using
Vitamin E Preparation USP as the source of a-tocopherol, dispersions were prepared using a melt-fusion method with active
loadings up to 50% (w / w) and characterised using differential scanning calorimetry and optical microscopy. Capsules
containing 300 IU a-tocopherol were manufactured and the absorption profiles compared to a commercial soft gelatin
capsule preparation in healthy human volunteers. Confocal laser scanning microscopy (CLSM) studies were performed in
order to elucidate the mechanism by which drug release may be occurring. Differential scanning calorimetry studies
indicated that the presence of the active had a negligible effect on the melting profile of the carrier, indicating limited
miscibility between the two components, a conclusion supported by the microscopy studies. Similarly, the dispersions were
shown to exhibit a glass transition corresponding to the incorporated drug, indicating molecular cooperativity and hence
phase separation from the lipid base. Despite the phase separation, it was noted that capsules stored for 18 months under
ambient conditions showed no evidence of leakage. Bioavailability studies in six healthy male volunteers indicated that the
Gelucire 44 / 14 formulation showed an approximately two-fold increase in total a-tocopherol absorption compared to the
commercial preparation. Confocal laser scanning microscopy studies indicated that, on contact with water, the dispersions
formed two interfacial layers, from which the Gelucire 44 / 14 disperses in the liquid medium as small particles. Furthermore,
evidence was obtained for the dispersed material becoming incorporated into the hydrated lipid. In conclusion, the dispersion
of the liquid drug in Gelucire 44 / 14 appears to allow the dual advantages of the preparation of a solid formulation and
improved bioavailability of this material.
2003 Elsevier B.V. All rights reserved.
Keywords: Bioavailability; Confocal; Differential scanning calorimetry; Gelucire; Glyceride; Microscopy; Solid dispersion; Tocopherol;
Vitamin E

1. Introduction
*Corresponding author. Tel.: 144-28-9027-2129; fax: 144-289024-7794.
E-mail address: duncan.craig@qub.ac.uk (D.Q.M. Craig).

a-Tocopherol [(6)-(2RS,49RS,89RS)-2,5,7,8-tetramethyl-2-(49,89,129-trimethyltridecyl)-6-chromanol)]

0168-3659 / 03 / $ see front matter 2003 Elsevier B.V. All rights reserved.
doi:10.1016 / S0168-3659(03)00261-X

478

S. A. Barker et al. / Journal of Controlled Release 91 (2003) 477488

is the most common naturally occurring member of

the Vitamin E family. It is a practically odourless
viscous oil obtained from distillation of vegetable
oils. The naturally occurring form is the d (R,R,R)
isomer while the synthetic analogue exists as the
racemic d,l mixture. a-Tocopherol is listed in the
main pharmacopoeias with a range of definitions and
nomenclature. For example, the British Pharmacopoeia and European Pharmacopoeia both contain separate monographs on a-tocopherol using
the chemical definition given above, RRR-atocopherol (defining the nature-identical form), atocopheryl acetate and RRR-a-tocopheryl acetate.
There is no monograph for Vitamin E. Conversely,
the United States Pharmacopoeia (USP) has a
generic monograph on Vitamin E which allows dand dl-a-tocopherol, d- and dl-a-tocopheryl acetate
and d- and dl-a-tocopheryl acid succinate. In addition, the USP has a monograph on Vitamin E
Preparation which is defined as a single form of
vitamin E in a stated concentration.
The main use of a-tocopherol is as a source of
Vitamin E for use in nutraceutical products, although
it also has a role as an antioxidant in pharmaceutical
formulations, such as Total Parenteral Nutrition
emulsions [1]. More recently, the antioxidant nature
of the Vitamin E family has prompted researchers to
speculate that large daily doses of these materials
may help to prevent or retard the development of
diseases such as cancer and cardiovascular disease,
by inhibiting the damaging effects of free radicals
[2,3]. However, the delivery of this molecule presents several significant formulation challenges. Oral
absorption is thought to proceed via the formation of
mixed bile salt emulsions in the small intestine with
consequent absorption into the lymphatic system
[4,5], hence there is potential for variable absorption
depending on the fed / fasted status of the individual
[6]. Secondly, a-tocopherol is a liquid at room
temperature, thus rendering formulation in a conventional tablet or capsule formulation impractical.
Indeed, the effective formulation of potent oral
liquids (and also low-melting point solids) remains a
serious difficulty within the pharmaceutical industry,
usually necessitating the use of liquid-filled soft
gelatin capsules.
Here we report the novel application of solid
dispersion technology as a means of incorporating

liquid pharmaceuticals and nutraceuticals into solid

dosage forms while also improving the bioavailability profile of the active. Gelucire 44 / 14 is one of
a family of lipid-based excipients, the Gelucires,
comprising a mixture of pegylated fatty acid esters
and glycerides, the two numbers of their names
corresponding to the approximate melting point and
HLB value, respectively [7]. These materials are
normally associated with controlled (slow) release
applications, with incorporated drugs being released
via diffusion and / or erosion processes over a period
of hours [8]. However, for reasons that have not yet
been elucidated, Gelucire 44 / 14 appears to promote
rapid drug release and bioavailability. This is exemplified by the studies of Dordunoo et al. [9] who
reported increases in the dissolution rate of
temazepam and triamterene when formulated in
Gelucire 44 / 14 compared to polyethylene glycol
dispersions. Similarly, Serajuddin et al. [10] showed
improved dissolution of a poorly soluble drug, REV
5901
[a-pentyl-3-(2-quinolinylmethoxy)
benzenemethanol], when dispersed in Gelucire 44 / 14
compared to polyethylene glycols. Studies have also
indicated improved oral absorption of drugs dispersed in Gelucire 44 / 14 [1113]. For example,
Aungst et al. [13] reported marked improvements in
the absorption in dogs of DMP 323 from a dispersion
in Gelucire 44 / 14 compared to equivalent polyethylene glycol or polyethylene glycol / polyvinylpyrrolidine dispersions. To date, however, there
have been no bioavailability studies on the absorption of liquid drugs from Gelucire 44 / 14 preparations. It is by no means obvious that improvements
seen for solids would be mirrored in liquids when
dispersed in these matrices, as the nature of the drug
distribution within the matrix and the dispersal
behaviour of the drug on contact with an aqueous
phase may be expected to be profoundly different
compared to a solidsolid dispersion. Previous
studies involving the incorporation of liquids into
Gelucire 44 / 14 have focused on manufacturing and
in vitro behaviour [14], hence there remains a need
to assess the potential of this approach for improving
bioavailability.
In this investigation, we describe the possibility of
using Gelucire 44 / 14 as a matrix material for a
model liquid drug, Vitamin E Preparation USP, from
three perspectives. In the first instance we have

S. A. Barker et al. / Journal of Controlled Release 91 (2003) 477488

studied the nature of the physical interaction between

the active and the carrier using differential scanning
calorimetry (DSC), T zero DSC and optical microscopy. Such investigations are of importance for the
elucidation of the mechanism of drug release, particularly in terms of ascertaining whether the active
is dispersed on a molecular basis through the matrix
or whether it is present as distinct domains. In
addition, the distribution of the active may be
expected to influence the mechanical properties and
leakage potential of the dispersions [14]. Secondly,
we describe a comparative oral bioavailability study
in healthy human volunteers between a Gelucire
44 / 14 formulation containing 300 IU a-tocopherol
and an equivalent commercially available capsule
formulation, based on an oily solution. Finally, in the
light of findings from the above we have performed
confocal laser scanning microscopy studies on equivalent dispersions in order to obtain information on
the behaviour of the system on hydration.

2. Materials and methods

Gelucire 44 / 14 was obtained from Gattefosse s.a.,
France. Vitamin E Preparation USP (d-a-tocopherol
1000 IU / g) was obtained from Archer Daniels
Midland, USA.

2.1. Manufacture of the solid dispersions

Solid dispersions containing 10, 25 and 50% (w /
w) Vitamin E Preparation USP were manufactured in
10 g batches as follows. The Gelucire 44 / 14 was
dispensed into a glass vial and placed in a waterbath
held at 60 8C for 15 min in order to melt the
material. It was then allowed to cool for 5 min (to
minimise degradation of the Vitamin E without
allowing solidification), at which point the Vitamin E
Preparation (d-a-tocopherol 1000 IU / g) USP was
added. To minimise solidification prior to adequate
mixing, the sample was then placed back in the
waterbath (60 8C) for 1 min, followed by 30 s gentle
agitation using a magnetic flea. The sample was then
allowed to stand to cool and solidify. Three separate
samples were manufactured at each concentration.
Homogeneity was indicated by the excellent reproducibility of the DSC data.

479

2.2. Manufacture of the capsule formulations

Liquid-filled hard gelatin capsules were prepared
for the 50% (w / w) solid dispersions. The above
process was followed up to the point of agitation and
then 600 mg of the molten mixture was filled into the
bodies of size 0 capsules. These were then allowed
to cool for 2.5 h before being capped and stored at
ambient temperature until use. Each capsule therefore contained 300 IU of a-tocopherol. It was noted
that no leakage or visible change in appearance was
apparent after 18 months storage under ambient
conditions.

2.3. Differential scanning calorimetry

DSC was performed using a DSC 2920 (TA
Instruments, Leatherhead, UK), using helium as the
carrier gas for all experiments due to its greater
thermal conductivity at low temperatures than nitrogen. Weight-matched hermetically sealed aluminium
pans (Perkin-Elmer) were used for all runs. A
scanning rate of 5 8C / min was used over the
temperature range from 225 to 65 8C. Full calibration was performed prior to the acquisition of
experimental data. Baseline calibrations were performed using weight-matched empty pans. Temperature calibration was performed using cyclohexane
(theoretical melting point 6.54 8C) and n-octadecane
(theoretical melting point 28.24 8C); these were
chosen to cover the low temperature transitions
undergone by the Gelucire and to facilitate analysis
of any melting point lowering induced by the vitamin
E preparation. A temperature calibration check was
then performed using n-octadecane.
Sample weights of 6.5 to 14 mg were used. The
sample and reference pans were loaded into the
sample chamber at ambient temperature and the
chamber sealed. The sample was then cooled to 225
8C, a process taking approximately 5 min, and then
held at this temperature for 10 min. The sample
temperature was then increased at 5 8C / min to 65
8C. One measurement was made for each combination preparation to give a total of three measurements
at each concentration. Gelucire 44 / 14 alone was
studied as is and with a preheating step to replicate
the production of the combination formulations. In
this latter case, the sample and reference pans were

480

S. A. Barker et al. / Journal of Controlled Release 91 (2003) 477488

placed into the sample chamber at ambient temperature, then heated to 60 8C and held there for 15 min
prior to cooling at 5 8C / min to 225 8C. After
holding at 225 8C for 10 min, the sample temperature was increased at 5 8C / min to 65 8C. Samples of
Gelucire 44 / 14 alone and the three bulk solid
dispersions were tested immediately after manufacture (the time of the test varied from 100 to 130 min
after manufacture) and at weekly intervals up to 6
weeks post-preparation.
T zero DSC was performed using a DSC Q1000
(TA Instruments). Temperature calibration was performed using indium, tin and n-octadecane, with the
enthalpic response being calibrated with indium.
Weight-matched hermetically sealed aluminium pans
(Perkin-Elmer) were used for all runs. The sample
and reference pans were loaded into the sample
chamber at ambient temperature and the chamber
sealed. The sample was then cooled to 290 8C, a
process taking approximately 5 min, and then equilibrated at 290 8C for 15 min with an applied
modulation of 60.796 8C every 60 s (i.e. the
modulation conditions used for the heating stage of
the experiment). Following equilibration, the sample
was heated to 0 8C with an underlying heating rate of
5 8C / min and an applied modulation of 60.796 8C
every 60 s.

2.4. Microscopy
Optical microscopy studies were performed using
an Olympus BX50 optical microscope fitted to a
Linkam temperature controller, with optical, polarising light and differential interference contrast (DIC)
lenses and filters used as indicated. Confocal laser
scanning microscopy (CLSM) was performed using
a Leica TCS SP2 CLSM, with a 103 objective.
Rhodamine B base stain was incorporated at 0.1%
(w / w) into Gelucire 44 / 14 using the melting method
as described above. The sample was smeared into
half the indentation of a cavity slide prior to solidification and a drop of water introduced to the other
half of the cavity. Excitation of the sample was from
bright field illumination from a xenon lamp, with
detection of emitted light between 566 and 651 nm.
The interface of the doped Gelucire 44 / 14 and the
water was monitored over a period of 10 min, with
images being taken every 45 s.

2.5. Human oral bioequivalence study

A human volunteer study was carried out comparing the bioavailability of a-tocopherol from the
Gelucire 44 / 14 formulation to that from a proprietary product containing equivalent amounts of
active. The study was conducted with the approval of
the Joint School of Pharmaceutical Sciences, University of Science Malaysia and General Hospital
Penang Committee on Bioavailability Studies; all
volunteers gave written informed consent. Six healthy adult male volunteers aged between 22 and 38
years old and weighing from 50 to 78 kg participated
in a standard two period, two sequence crossover
study. The volunteers were not receiving any medication or vitamin E supplements from 2 weeks
before and during the study. The volunteers were
randomly divided into two groups of three, with each
group taking two capsules of either the proprietary or
the Gelucire 44 / 14 preparation in the first period.
After a washout period of 1 week each volunteer
then received the alternative product. The total dose
of a-tocopherol per period was therefore 400 mg
(approximately 600 IU). All capsules were given
with 240 ml of water at approximately 10 a.m. after
an overnight fast. Food and drink (other than water,
given ad libitum) were withheld for at least 4 h after
dosing. Lunch and dinner comprising chicken and
rice were served 4 and 10 h after dosing. Five
millilitre blood samples were collected at 0 (before
dosing), 1, 2, 3, 4, 5, 6, 7, 8, 10, 12, 14, 18, 24, 30
and 36 h after dosing via an in-dwelling cannula
placed in the forearm. Samples were centrifuged at
2000 g for 15 min and the plasma frozen until
analysis. Plasma levels of d-a-tocopherol were analysed using the HPLC method described by Julianto
et al. [15].
Maximum plasma concentration (Cmax ) and time
to reach maximum plasma concentration (T max ) were
obtained directly from the plasma values [16] while
the total area under the plasma concentrationtime
curve (AUC 0 ` ) was calculated by adding the area
from time zero to time t (AUC 0 t ) and the area from
time t to infinity (AUC t ` ). The former was calculated using a trapezoidal function and the latter by
dividing the last measurable plasma drug concentration by the elimination rate constant (k e ), calculated from the terminal slope of the individual

S. A. Barker et al. / Journal of Controlled Release 91 (2003) 477488

profiles after logarithmic transformation of the plasma concentration and application of linear regression
[17]. An analysis of variance (ANOVA) procedure
that distinguished effects due to subjects, periods and
treatments [18] was used to analyse the values of
Cmax and AUC 0 ` from the two preparations. T max
values were analysed using the Wilcoxon Signed
Rank Test for paired samples.

3. Results

3.1. DSC and optical microscopy studies

Fig. 1 shows the DSC trace for Gelucire 44 / 14
alone, previously heat-treated in the manner described above, and the three formulations with
Vitamin E Preparation USP. While the reproducibility for these samples was satisfactory, it was noted
that samples studied as-received showed the same
basic peak profiles but exhibited poorer reproducibility, demonstrating the necessity for controlling the
thermal history of this material in order to obtain
reliable results, as has been noted for other Gelucires
[7,18,19]. The trace shows a main peak at circa 42
8C with a broad shoulder at approximately 35 8C and

481

Table 1
Main peak maximum temperatures and enthalpies of transition for
mixtures of Gelucire 44 / 14 and Vitamin E Preparation USP
% of Vitamin E
Preparation USP

Peak maximum
(mean6S.D.) (8C)

Enthalpy of transition
(mean6S.D.) (J / g)

0
10
25
50

42.860.4
43.460.1
42.760.2
41.161.4

95.964.8
84.467.2
57.262.8
32.861.9

a smaller secondary peak with a mean peak maximum at 20 8C. The mean enthalpies of the main and
secondary transitions were 95.9 J / g (S.D. 4.8 J / g)
and 9.6 J / g (S.D. 0.5 J / g), respectively, using the
minimum seen between the two sets of thermal
events as a baseline reference. As Gelucire 44 / 14 is
a mixture of materials, it is logical to speculate that
the secondary peak and the shoulder on the main
peak are due either to melting of specific minor
components of the mixture or to mixed crystals
formed between different components, rather than
different polymorphs of specific materials [19,20].
Comparison of the results for Gelucire 44 / 14 and
the mixes shows that the main Gelucire 44 / 14 peak
is retained even as the content of Vitamin E Preparation USP is increased to 50% (w / w). Table 1
illustrates the main peak maximum temperature and

Fig. 1. DSC profiles of representative samples of Gelucire 44 / 14 alone and with increasing quantities of Vitamin E Preparation.

482

S. A. Barker et al. / Journal of Controlled Release 91 (2003) 477488

the enthalpy of transition for the various samples.

There is no marked change in the peak maximum
temperature as the percentage of the drug system is
increased, while the enthalpy of the main transition
decreases in an approximately linear fashion as the
percentage of drug system is increased, although the
magnitude of the decrease is not directly proportional
to the decrease in Gelucire 44 / 14 content. However,
it is not possible to obtain completely reliable
enthalpy data for individual peaks when overlap
occurs, as is the case here, hence these data should
be regarded as indicative only. It is interesting to
note that the secondary lower temperature peak
becomes progressively broadened in the presence of
the drug.
The results indicate that even at a level of 50%
(w / w), the Vitamin E Preparation USP is not having
a profound effect on the melting behaviour of the
Gelucire 44 / 14, particularly when one compares
these data to previous studies on the effect of
paracetamol dispersions in solid polyethylene glycol
(PEG) whereby temperature and enthalpy of the
higher melting peak were dramatically reduced by
the presence of the PEG [21]. Overall, therefore, the
lack of effect of the Vitamin E Preparation USP on
the melting behaviour of Gelucire 44 / 14 when it is
incorporated at a level of up to 50% (w / w) is
striking and unexpected. This suggests that there
may exist discrete regions of Gelucire 44 / 14 and the
drug system which then behave thermally as separate
entities.
This hypothesis was explored further by examining the thermal behaviour of the Vitamin E Preparation USP itself. This has been the subject of a
separate study [22] and, in brief, the material shows
a glass transition at circa 260 8C and a recrystallisation endotherm at circa 227 8C, the presence and
relative magnitude of the two events depending on
the heating rate used. Therefore, if phase separation
is indeed occurring between the Vitamin E Preparation USP and the Gelucire 44 / 14 in the solid
dispersions, it should be possible to see the glass
transition of Vitamin E Preparation USP in the total
thermal response of the dispersions. However, as the
transitions for the Vitamin E Preparation USP alone
are subtle (an enthalpy of 1.21 J / g was measured for
the endotherm at circa 227 8C), it is likely to be
difficult to observe these in a mixed system using

conventional DSC. We have used the novel technique

of T zero DSC to study this issue, as this approach is
believed to exhibit superior baseline stability to
conventional DSC and therefore allow the detection
of smaller transitions. In brief, this method operates
on the same principle as modulated temperature DSC
in that the total heat flow signal is resolved into
reversing and non-reversing components, with glass
transitions seen in the reversing heat flow signal, but
the inclusion of a further thermocouple allows
correction for the main errors associated with
baseline curvature and poor resolution.
The T zero DSC results are shown in Fig. 2,
whereby the glass transition for the Vitamin E
Preparation USP is seen for all the disperse samples.
It should be noted that baseline drift is an issue at
such temperatures and hence care is required when
attempting to interpret what appear to be minor
transitions, hence in the present case only the clear
T g is considered. The observation that the T g is not
only detected but is also seen over the same temperature range as was the case for the pure drug further
supports the suggestion that the Vitamin E Preparation USP is largely or completely phase separated
from the Gelucire 44 / 14. Additionally, this study
indicates that T zero DSC can successfully be used to
distinguish small thermal transitions in mixed systems. It was not possible to use the heat capacity
change through T g in a quantitative manner in this
case due to the baseline slope obscuring the onset
and offset temperatures, although it can be clearly
seen that the expected increase in DCp with increasing drug content was indeed observed.
Microscopy studies were performed on the Gelucire 44 / 14 alone and on the dispersions containing
10, 25 and 50% (w / w) Vitamin E Preparation USP,
using both polarising light and DIC microscopy, the
latter being conducted such that the vitamin E was
coloured blue to allow visualisation of the drug in
contrast to the lipid matrix. Typical results for the
Gelucire 44 / 14 alone under polarisers at 25 8C are
shown in Fig. 3a. In essence, a mass of microcrystals
is seen which subsequently melted over a narrow
temperature range at approximately 40 to 45 8C; no
distinct events were seen corresponding to the shoulder and the main peak seen on the DSC trace. Fig. 3b
and c shows the corresponding image for the system
containing 10% (w / w) and 50% (w / w) Vitamin E

S. A. Barker et al. / Journal of Controlled Release 91 (2003) 477488

483

Fig. 2. Heat capacity as a function of temperature, obtained from the reversing heat flow (reversing Cp ) obtained from T zero modulated
temperature DSC studies for Vitamin E Preparation USP alone and its 10, 25 and 50% dispersions in Gelucire 44 / 14.

Preparation USP. In this case, there are discontinuities in the mass of crystals that may correspond
to regions of Vitamin E Preparation USP. This is
supported by DIC microscopy whereby the dark
regions seen under the cross-polarisers were found to
correspond to blue regions seen under DIC, again
indicating that these areas correspond to the separate
liquid phase. It is interesting to note that the crystals
seen for the Gelucire 44 / 14 alone under crosspolarisers appeared to be considerably finer than
those seen for systems containing Vitamin E Preparation USP, whereby comparatively large needleshaped crystals were observed. Taken together with
the similarity in the DSC endothermic responses for
the systems with and without Vitamin E Preparation
USP, it is reasonable to suggest that the presence of
the liquid is altering the kinetics of nucleation and
crystal growth, leading to the formation of a smaller
number of larger (but essentially structurally identical) crystals.

3.2. Bioavailability studies

The mean plasma a-tocopherol concentration versus time profiles of the commercial and Gelucire
44 / 14 formulations are shown in Fig. 4, with
individual and mean plasma concentrations being
detailed in Tables 2 (commercial product) and 3
(Gelucire 44 / 14 product). Individual and mean
values of the associated pharmacokinetic parameters
are given in Table 4.
It is clear from these profiles that the oral absorption of a-tocopherol from the Gelucire 44 / 14 formulation was considerably higher than from the commercial product, with a more rapid onset. Both the
AUC 0 ` and Cmax values of the Gelucire 44 / 14
formulation were significantly higher (P,0.01) than
those of the commercial product. In addition, the
90% confidence interval for the ratio of the logarithmic transformed AUC 0 ` values of the Gelucire
44 / 14 formulation over those of the commercial

484

S. A. Barker et al. / Journal of Controlled Release 91 (2003) 477488

Fig. 4. The mean (6S.E.M.) plasma a-tocopherol concentration

versus time profiles of the commercial and Gelucire 44 / 14
formulations.

preparation was estimated to be between 1.91 and

2.22 (median 2.07) while that of Cmax was between
1.59 and 2.06 (median 1.83). There was no statistically significant difference in T max or k e values
between the two preparations (P.0.05) when analysed using the Wilcoxon Signed Rank Test. Overall,
therefore, the Gelucire 44 / 14 preparation showed an
approximately two-fold increase in the extent of
absorption, with a more rapid onset than was seen
for the commercial preparation. Additionally, the lag
phase prior to absorption of a-tocopherol from the
Gelucire 44 / 14 product was markedly shorter than
for the commercial product. For four of the volunteers, there was a 2 h difference in this lag phase
(commercial product 5 h, and Gelucire 44 / 14 product 3 h) and for the remaining two volunteers the
difference was 1 h. It is interesting to note that the
absorption of a-tocopherol appears to be biphasic,
particularly for the Gelucire 44 / 14 product, although
this may simply be a reflection of the fact that more
a-tocopherol is being absorbed, hence allowing for a
finer distinction of absorption patterns than is possible with the conventional formulation. The second
phase of the absorption of a-tocopherol occurs just
after the volunteers were fed (at 10 h post-dose) and
may be due in part to the mobilisation of the lipidabsorbing pathway in response to food. These observations are in agreement with previous studies on
a-tocopherol bioavailability [23].
Fig. 3. Polarised light images of (a) Gelucire 44 / 14 alone, (b)
10% (w / w) dispersions of Vitamin E Preparation, and (c) 50%
(w / w) dispersions of Vitamin E Preparation at 25 8C. Horizontal
axis represents 250 mm.

3.3. Confocal laser scanning microscopy

The study thus far had indicated that the Vitamin E

S. A. Barker et al. / Journal of Controlled Release 91 (2003) 477488

485

Table 2
Individual and mean plasma concentrations of a-tocopherol after dosing with the commercial product for volunteers (Vol.) as indicated
Time
(h)
1
2
3
4
5
6
7
8
10
12
14
18
24
30
36

Plasma concentration of a-tocopherol (mg / ml)

Vol. 1

Vol. 2

Vol. 3

Vol. 4

Vol. 5

Vol. 6

Mean

S.E.M.

0.00
0.00
0.00
0.00
1.32
1.58
1.89
2.37
3.22
2.64
2.80
1.54
1.14
0.80
0.55

0.00
0.00
0.00
0.00
1.51
1.68
2.44
2.85
4.21
3.39
2.69
2.36
1.70
1.28
0.94

0.00
0.00
0.00
0.77
1.60
2.58
2.83
3.36
3.06
2.89
3.31
2.78
2.12
1.64
1.28

0.00
0.64
0.83
1.69
1.69
1.27
1.56
1.55
2.78
2.79
2.55
2.73
2.72
1.58
1.27

0.00
0.00
0.00
0.00
0.70
1.03
1.48
1.86
2.85
2.70
1.91
1.26
2.63
1.69
1.25

0.00
0.00
0.00
0.00
1.49
1.84
2.03
2.28
3.80
3.81
2.55
2.98
3.51
2.93
1.45

0.00
0.11
0.14
0.41
1.38
1.66
2.03
2.38
3.32
3.04
2.64
2.28
2.30
1.65
1.12

0.00
0.11
0.14
0.29
0.15
0.22
0.21
0.27
0.23
0.19
0.18
0.29
0.34
0.29
0.13

S.E.M., standard error of the mean.

USP and Gelcuire 44 / 14 exist largely or completely

as a phase separated system. However, the oral
bioavailability of the drug was clearly enhanced
when incorporated into the lipid matrix in comparison with the commercial preparation. This therefore raised the question of how the drug was being
released from the matrix and to this effect CLSM
studies were conducted to examine the behaviour of
the Gelcuire 44 / 14 on contact with water, using the

fluorescent dye rhodamine (base) as a model drug.

Examination of the dispersions indicated that almost
full miscibility was obtained, with a small number of
rhodamine particles remaining distinct; this miscibility was anticipated due to the lipophilic nature of the
dye and the low concentration in the matrix. Observation of the corresponding data indicated that the
system appeared to be forming particles with an
approximate diameter of 10 mm that were then

Table 3
Individual and mean plasma concentrations of a-tocopherol after dosing with the Gelucire 44 / 14 product for volunteers (Vol.) as indicated
Time
(h)
1
2
3
4
5
6
7
8
10
12
14
18
24
30
36

Plasma concentration of a-tocopherol (mg / ml)

Vol. 1

Vol. 2

Vol. 3

Vol. 4

Vol. 5

Vol. 6

Mean

S.E.M.

0.00
0.00
0.51
1.46
2.34
2.80
3.27
2.95
4.75
5.02
5.12
4.56
3.70
2.47
1.50

0.00
0.00
0.47
1.43
3.44
3.73
5.68
6.67
7.89
7.44
7.11
6.52
4.84
3.70
2.90

0.00
0.00
0.89
1.80
4.17
5.43
8.08
7.22
6.83
7.03
5.37
6.58
5.97
4.97
3.10

0.75
0.59
0.82
2.03
1.09
2.63
4.37
3.48
4.42
5.13
4.27
4.25
4.84
3.42
2.34

0.00
0.00
0.59
0.92
1.94
2.44
3.33
3.49
4.19
4.90
4.71
3.50
3.03
2.50
1.50

0.00
0.00
0.55
1.72
3.15
4.16
5.72
4.08
5.88
5.00
4.76
6.12
3.90
2.41
2.03

0.13
0.10
0.64
1.56
2.69
3.53
5.07
4.65
5.66
5.76
5.22
5.25
4.38
3.25
2.23

0.13
0.10
0.07
0.16
0.46
0.47
0.74
0.74
0.60
0.47
0.41
0.54
0.43
0.41
0.28

S.E.M., standard error of the mean.

S. A. Barker et al. / Journal of Controlled Release 91 (2003) 477488

486

Table 4
Individual and mean pharmacokinetic parameters of the two preparations for volunteers (Vol.) as indicated
Vol. 1

Vol. 2

Vol. 3

Vol. 4

Vol. 5

Vol. 6

Mean

S.E.M.

49.5
9.6
59.0
3.2
10

65.7
19.3
85.0
4.2
10

75.0
29.7
104.6
3.4
8

72.1
20.0
92.1
2.8
12

58.8
20.2
78.9
2.9
10

89.5
19.7
109.2
3.8
12

68.4
19.8
88.1
3.4
10.3

5.65
2.60
7.47
0.2
0.6

Gelucire 44 /14 product

AUC 0 t (mg h / ml)
114.0
AUC t ` (mg h / ml)
24.1
AUC 0 ` (mg h / ml)
138.1
Cmax (mg / ml)
5.1
T max (h)
14

169.3
64.5
233.8
7.9
10

182.8
84.5
267.3
8.1
7

127.2
38.7
165.9
5.1
122

101.9
32.9
134.9
4.9
12

132.0
32.1
164.2
6.1
18

137.9
46.1
184.0
6.2
12.2

12.9
9.53
22.1
0.6
1.5

Ratio of Gelucire 44 /14: commercial products

AUC 0 `
2.3
2.8
Cmax
1.6
1.9

2.6
2.4

1.8
1.8

1.7
1.7

1.5
1.6

2.1
1.8

Commercial product
AUC 0 t (mg h / ml)
AUC t ` (mg h / ml)
AUC 0 ` (mg h / ml)
Cmax (mg / ml)
T max (h)

released into the aqueous medium (Fig. 5a and b).

However, two unexpected observations were made.
In the first instance, two distinct interfacial regions
were observed, with the outer region remaining
approximately constant in thickness (|200 mm)
throughout the experiment. The inner region extended into the solid phase at approximately 20 mm / min
under the experimental conditions used here. Examination of the colour intensity showed a clear decreasing gradation from the solid to the aqueous phase
between adjacent layers, indicating that the two
interfacial regions corresponded to hydrated layers
with a greater proportion of water in the outer layer.
The second observation related to the small crystals
of rhodamine remaining in the solid dispersion. It
was noted that when the inner interfacial layer came
into contact with this region the lipophilic base
dispersed into the hydrated layer, as indicated in Fig.
5b.

4. Discussion
The study has highlighted several issues of interest
to the formulation of liquid drugs and nutraceuticals
in Gelucire 44 / 14. In the first instance, it is clearly
possible to formulate this material as a liquid-filled
hard gelatin capsule without discernible leakage; this

0.2
0.1

is of clear significance for the manufacture of any

commercial product using this technology. Secondly,
there are clear improvements in a-tocopherol bioavailability when formulated in Gelucire 44 / 14
compared to a standard commercial softgel preparation. Taken together these studies suggest that the
formulation of liquid drugs and neutraceuticals in
Gelucire 44 / 14 solid dispersions represents a viable
production approach. However, the study has also
yielded insights into the mechanism by which this
matrix material may exert its effect. The only study
discussing the mechanism by which Gelucire 44 / 14
may improve bioavailability is that of Mehta [24], in
which it is suggested that this material emulsifies on
contact with water and therefore acts as a form of
self-emulsifying drug delivery system (SEDDS), the
important difference being that SEDDSs have to date
been considered as liquid dosage forms; indeed, the
question arises as to whether any such particles
would be liquid or solid, with the latter appearing
more likely. The emulsification explanation is based
on anecdotal evidence mentioned by the manufacturers of Gelucire in sales literature and has not been
studied as an issue in its own right to date. Nevertheless, previous studies have indicated that the presence of lipids improves a-tocopherol absorption [6],
while other studies have shown that liquid SEDDS
formulations may also improve the bioavailability of
this material [23]. Mehta [24] also suggests that

S. A. Barker et al. / Journal of Controlled Release 91 (2003) 477488

487

(with reference to solid drugs) the proportion of

Gelucire 44 / 14 should be high compared to the drug
in order for this improvement to be apparent, while
also suggesting that the drug must be molecularly
dispersed in the matrix in the solid state in order for
incorporation into the oil droplets to occur on contact
with water. Interestingly, our studies have indicated
that, in this case, loadings as high as 50% (w / w)
(liquid) drug may show bioavailability improvements compared to a standard lipid formulation.
On examining the data from the structural studies,
our initial prediction was that bioavailability improvement was unlikely as, if the emulsification or
disintegration hypothesis is correct, the drug would
probably not be incorporated into any such particles
due to the observed phase separation. This hypothesis was clearly incorrect, as indicated by the human
volunteer study, hence the drug must be interacting
with the lipid at a subsequent stage in the absorption
process. However, the CLSM study has provided a
putative mechanism whereby the (clearly complex)
hydration behaviour of the Gelucire 44 / 14 is such
that drug incorporation may be possible during the
hydration process itself. Clearly, it is not yet known
whether this is applicable to the drug under study but
does nevertheless provide a plausible explanation for
the observed behaviour.

5. Conclusion

Fig. 5. CLSM images of rhodamine B base dispersions in

Gelucire 44 / 14 on (a) initial hydration and (b) further hydration
after 15 min (horizontal scale represents 1.5 mm). (1) Aqueous
phase containing droplets or particles, (2) outer hydration layer,
(3) inner hydration layer, (4) solid Gelucire 44 / 14, (5) crystalline
rhodamine B base.

The study has demonstrated that dispersions of

Vitamin E Preparation USP in Gelucire 44 / 14 may
be prepared as a liquid-filled hard gelatin capsule
formulation, with loadings as high as 50% (w / w)
showing no leakage after 18 months storage under
ambient conditions. Thermal analysis and microscopy studies indicated that the drug system and
matrix are at least partially phase separated, while
the human bioavailability study indicated an approximately two-fold increase in absorption compared to a
commercial preparation. Subsequent CLSM studies
have suggested that this increase may be at least
partially associated with emulsification or disintegration of the Gelucire base on contact with water,
with evidence being obtained for the drug being
incorporated into the hydration layer of the Gelucire.
Overall, therefore, the study has demonstrated that

488

S. A. Barker et al. / Journal of Controlled Release 91 (2003) 477488

the incorporation of liquid drugs into Gelucire 44 / 14

presents a viable means of formulating solid dosage
forms from such materials, with the possibility of
concomitant improvement of bioavailability in
humans.

References
[1] M.C. Allwood, Compatibility and stability of TPN mixtures
in big bags, J. Clin. Hosp. Pharm. 9 (1984) 181198.
[2] J. Bauersachs, I. Fleming, D. Fraccarollo, R. Busse, G. Ertl,
Prevention of endothelial dysfunction in heart failure by
vitamin E: attenuation of vascular superoxide anion formation and increase in soluble guanylyl cyclase expression,
Cardiovasc. Res. 51 (2001) 344350.
[3] B.J. Schmitz-Drager, M. Eichholzer, B. Beiche, T. Ebert,
Nutrition and prostate cancer, Urol. Int. 67 (2001) 111.
[4] L.J. Machlin, E. Gabriel, Kinetics of tissue a-tocopherol
uptake and depletion following administration of high levels
of vitamin E, Ann. N.Y. Acad. Sci. 393 (1983) 4860.
[5] K.J. Palin, Lipids and oral drug delivery, Pharm. Int.
November (1985) 272275.
[6] P. Tso, T. Lee, S.J. DeMichele, Randomized structured
triglycerides increase lymphatic absorption of tocopherol and
retinol compared with the equivalent physical mixture in a
rat model of fat malabsorption, J. Nutr. 131 (2001) 2157
2163.
[7] D.Q.M. Craig, The use of glycerides as controlled release
matrices, in: D.R. Karsa, R.A. Stephenson (Eds.), Excipients
and Delivery Systems For Pharmaceutical Formulations,
Royal Society of Chemistry, London, 1995, pp. 148173.
[8] W. Sutananta, D.Q.M. Craig, J.M. Newton, An evaluation of
the mechanisms of drug release from glyceride bases, J.
Pharm. Pharmacol. 47 (1995) 182187.
[9] S.K. Dordunoo, J.L. Ford, M.H. Rubinstein, Preformulation
studies on solid dispersions containing triamterene or
temazepam in polyethylene glycols or Gelucire 44 / 14 for
liquid filling of hard gelatin capsules, Drug Dev. Ind. Pharm.
17 (1991) 16851713.
[10] A.T.M. Serajuddin, P.C. Sheen, D. Mufson, D.F. Bernstein,
M.A. Augustine, Effect of vehicle amphiphilicity on the
dissolution and bioavailability of a poorly water-soluble drug
from solid dispersions, J. Pharm. Sci. 77 (1988) 414417.
[11] P.C. Sheen, S.I. Kim, J.J. Petillo, A.T.M. Serajuddin, Bioavailability of a poorly water-soluble drug from tablet and

[12]

[13]

[14]

[15]

[16]

[17]
[18]
[19]

[20]

[21]

[22]

[23]

[24]

solid dispersions in humans, J. Pharm. Sci. 80 (1991) 712

714.
F. Pozzi, A. Longo, C. Lazzarini, A. Carenzi, Formulations
of ubidecarenone with improved bioavailability, Eur. J.
Pharm. Biopharm. 37 (1991) 243246.
B.J. Aungst, N.H. Nguyen, N.J. Rogers, S.M. Rowe, M.A.
Hussain, S.J. White, L. Shum, Amphiphilic vehicles improve
the oral bioavailability of a poorly soluble HIV protease
inhibitor at high doses, Int. J. Pharm. 156 (1997) 7988.
C. Doelker, E. Doelker, P. Buri, L. Waginaire, The incorporation and in vitro release profiles of liquid, deliquescent or
unstable drugs with fusible excipients in hard gelatin capsules, Drug Dev. Ind. Pharm. 12 (1986) 15531565.
T. Julianto, K.H. Yuen, A.M. Noor, Simple high performance
liquid chromatographic method for determination of atocopherol in human plasma, J. Chromatogr. B 732 (1999)
227231.
D.L. Weiner, Design and analysis of bioavailability studies,
in: C.R. Buncher, J.Y. Tsay (Eds.), Statistics in the Pharmaceutical Industry, Marcel Dekker, New York, 1981, pp.
205209.
M. Gibaldi, D. Perrier (Eds.), Pharmacokinetics, 2nd Edition,
Marcel Dekker, New York, 1982, pp. 145149.
J.G. Wagner, Fundamentals of Clinical Pharmacokinetics, 1st
Edition, Drug Intelligence, Hamilton, 1975, pp. 285305.
W. Sutananta, D.Q.M. Craig, J.M. Newton, An investigation
into the effect of preparation conditions on the structure and
mechanical properties of pharmaceutical glycerides, Int. J.
Pharm. 110 (1994) 7591.
W. Sutananta, D.Q.M. Craig, J.M. Newton, The effects of
ageing on the thermal behaviour and mechanical properties
of pharmaceutical glycerides, Int. J. Pharm. 111 (1994)
5162.
G.R. Lloyd, D.Q.M. Craig, A. Smith, An investigation into
the melting behaviour of binary mixes and solid dispersions
of paracetamol and PEG 4000, J. Pharm. Sci. 86 (1997)
991996.
S.A. Barker, K.H. Yuen, D.Q.M. Craig, An investigation into
the low temperature thermal behaviour of Vitamin E Preparation USP using differential scanning calorimetry and low
frequency dielectric analysis, J. Pharm. Pharmacol. 52
(2000) 941947.
T. Julianto, K.H. Yuen, A.M. Noor, Improved bioavailability
of vitamin E with a self emulsifying formulation, Int. J.
Pharm. 200 (1999) 5357.
S.C. Mehta, Issues and approaches for improving the solubility and bioavailability of poorly water soluble compounds,
Bull. Technique Gattefosse 91 (1998) 6571.