Use of a highly standardized mixture of Vitis vinifera, Ginkgo biloba and Melilotus officinalis extracts in the treatment of cellulite: a biopharmaceutical approach

Abstract

Cellulite currently remains a controversial topic to be defined and fully explained. The genesis of cellulite has a highly complex pathophysiology, which includes expansion of the subcutaneous adipose tissue, fibrotic dermal septa formation, progressive skin laxity, atrophy and structural modification of some dermal components. These alterations could also be caused by insufficiency of the precapillary sphincters affected by the development of cellulite. The study here encompassed a retrospective clinical trial that analyzed the results of 90 days of treatment in 21 adult women aged between 24 and 53 who underwent a motor programme for the treatment of cellulite along with using two different food supplements. The first was based on extracts of Vitis vinifera, Ginkgo biloba and Melilotus officinalis, with two subgroups at a dosage of two and three tablets per day, respectively; the second was based on carnosine and beta-alanine, constituting a control group. The weight, height, waist, hip, proximal thigh, mid-thigh, distal thigh and calf circumferences were measured for all subjects. There were no appreciable variations with respect to weight, BMI, waist circumference or the waist-to-hip ratio. The consumption of the mixture of Vitis vinifera, Ginkgo biloba and Melilotus officinalis extracts was associated with a greater reduction in hip circumference and thigh circumferences with a dosage of two tablets/day compared to three tablets/day, and compared to the control. The sample size yielded interesting but still not completely unequivocal results. Further studies with a greater number of subjects will be required to clarify the extent of the results obtained, while investigating the posology to determine the most effective treatment within the timeframe considered.

Full text

Nutrafoods (2021) 1:220-226 DOI 10.17470/NF-021-0029
220
Use of a highly standardized mixture
of Vitis vinifera, Ginkgo biloba
and Melilotus ocinalis extracts
in the treatment of cellulite:
a biopharmaceutical approach
Abstract
Cellulite currently remains a controversial topic to be dened and fully
explained. The genesis of cellulite has a highly complex pathophysiology,
which includes expansion of the subcutaneous adipose tissue, brotic
dermal septa formation, progressive skin laxity, atrophy and structural
modication of some dermal components. These alterations could also
be caused by insuciency of the precapillary sphincters aected by the
development of cellulite. The study here encompassed a retrospective
clinical trial that analyzed the results of 90 days of treatment in 21 adult
women aged between 24 and 53 who underwent a motor programme for
the treatment of cellulite along with using two dierent food supplements.
The rst was based on extracts of Vitis vinifera, Ginkgo biloba and Melilotus
ocinalis, with two subgroups at a dosage of two and three tablets per
day, respectively; the second was based on carnosine and beta-alanine,
constituting a control group. The weight, height, waist, hip, proximal
thigh, mid-thigh, distal thigh and calf circumferences were measured for
all subjects. There were no appreciable variations with respect to weight,
BMI, waist circumference or the waist-to-hip ratio. The consumption of
the mixture of Vitis vinifera, Ginkgo biloba and Melilotus ocinalis extracts
was associated with a greater reduction in hip circumference and thigh
circumferences with a dosage of two tablets/day compared to three
tablets/day, and compared to the control. The sample size yielded
interesting but still not completely unequivocal results. Further studies
with a greater number of subjects will be required to clarify the extent of
the results obtained, while investigating the posology to determine the
most eective treatment within the timeframe considered.
Keywords: Cellulite, Leucoselect, Lymphaselect, Ginkgoselect,
Carvelin
Original Research
Alexander Bertuccioli 1*
Alfredo Bressan 2
Andrea Biagi 3
Marco Neri 3
Giordano Zonzini 3
1 Department of Biomolecular
Sciences, University of Urbino
Carlo Bo, 61029 Urbino, Italy
2 ASUR Area 1, Pesaro, Italy
3 AIFeM Ravenna, Italy
*Corresponding author:
Alexander Bertuccioli
alexander.bertuccioli@uniurb.it

221
Introduction
Despite numerous studies, cellulite re-
mains a controversial topic to be dened and
fully explained. Currently, the term ‘cellulitis’
is used in the scientic literature to indicate a
spreading gangrenous infection of the subcu-
taneous cellular tissue [1]; in contrast, the term
‘cellulite’ was used for the rst time according to
several authors between 1850 and 1920 [2, 3] to
describe an aesthetic alteration of the cutaneous
surface. In an attempt to provide a more cor-
rect denition over time, many more descrip-
tive terms have been used, and are considered
by some authors to be synonyms of the term
cellulite, including: panniculitis, lipodystrophy,
oedematobrosclerotic panniculitis, liposclero-
sis, lipoedema [1], adiposis oedematosa, dermo-
panniculosis deformans, status protrusus cutis,
gynoid lipodystrophy (GLD) [2] , nodular liposcle-
rosis, oedematous brosclerotic panniculopa-
thy and panniculosis [3]; a multitude of names
for a pathology not yet fully understood.
Cellulite is estimated to aect between 85 and
98% of the female population [2]: such a high
prevalence has led some authors to consider it
almost as a secondary sexual characteristic [4].
The genesis of cellulite has a very complex
pathophysiology, which includes expansion of
the subcutaneous adipose tissue, brotic der-
mal septa formation, progressive skin laxity,
atrophy and structural modication of some
dermal components [4].
It is believed that fat cells take up an arrange-
ment in chambers formed by bands of con-
nective tissue called septa, which connect the
muscular plane to the lower part of the dermis.
Various factors such as inammation and/or al-
teration of the microcirculation can favour ex-
pansion of the septa, stressing the connective
components that will end up acting on the skin,
which will be progressively tensed beyond its
elastic capacity.
This process, which can continue with further
weight gain and/or further gain of water,
manifests as the appearance of dimples
and skin with an ‘orange peel’ appearance,
mainly in the gluteal, trochanteric, pelvic and
abdominal areas [3, 4].
It is interesting that considerably older [3, 5]
analyses of the interstitial uid show an in-
creased protein concentration (0.8–1.2 mg/ml;
normal 0.2–5.1 mg/ml) and an increased inter-
stitial pressure (150–200 mmH20; normal phys-
iological pressure 75–91 mmH20).
These alterations could be caused by insu-
ciency of the precapillary sphincters aected
by the development of cellulite [5].
Laser Doppler uxmetry carried out on patients
with cellulite showed an average ow reduction
of 35% in the aected regions compared to the
unaected regions [6].
Genetic and epigenetic factors are involved in
the predisposition to development of cellulite:
Genetic factors [3, 7 ]:
• Sex (women are exclusively aected in the
classical pattern);
• ethnicity (Caucasian women have a greater
tendency than Asian, African or Afro-American
women);
• biotype (Latina women develop cellulite
on the hips, while Anglo-Saxon and Nordic
women develop cellulite on the abdomen);
• fat tissue distribution;
• number, isoform and sensitivity of hormone
receptors on the aected cells;
• predisposition to development of peripheral
angiopathy (or susceptibility to circulatory
insufciency).
Exposomic factors [3]:
• An unbalanced diet (excessive intake of fats
and carbohydrates);
• excessive salt intake (uid retention);
• a bre-poor diet (constipation and increased
venous resistance);

222
Nutrafoods (2021) 1:220-226 DOI 10.17470/NF-021-0029
vinifera, Ginkgo biloba and Melilotus ocinalis.
The subjects receiving carnosine and beta-
alanine constituted a control group.
The weight, height, waist, hip,
proximal thigh, mid-thigh, distal thigh and
calf circumferences were obtained for all
subjects. The means of the data obtained
within the respective groups were evaluated
using Student's t test for hypothesis testing.
Retrospective data studies and analysis were
conducted in accordance with good clinical
practice rules xed by the Declaration of
Helsinki and in accordance with the European
Union Directive 2001/20/EC [8].
Each patient signed a consent form and privacy
policy documents and approved data analysis
and publishing.
Evaluated products and evaluation
scheme
Of the 21 subjects whose data were
examined, 7 took 2 tablets/day and 7 took
3 tablets/day of a mixture of Vitis vinifera,
Ginkgo biloba and Melilotus ocinalis extracts
formulated in 3-layered tablets, produced by
SIIT (Trezzano sul Naviglio, Milan, Italy), and
notied to the Italian Ministry of Health as
a food supplement by Pharmextracta SpA
(Pontenure, PC, Italy) complying with law n°196-
2004 (notication number: 22435), marketed
under the name ‘Carvelin’.
These contain: in the upper (30/45 minute-re-
leasing) layer 100 mg of Leucoselect Phyto-
some® (from Vitis vinifera), in the middle (8 hour
sustained-releasing) layer 50 mg of Lymphase-
lect® (from Melilotus ocinalis) and in the lower
(30/45 minute-releasing) layer 80 mg of Gink-
goselect Phytosome® (from Ginkgo biloba).
Leucoselect Phytosome® corresponds
to a highly standardized mixture of oligomeric
procyanidins (OPC – proanthocyanidins: 25.0–
30.0%), extracted from the outer part of Vitis
vinifera seeds, complexed with distearoylphos-
phatidylcholine from soy (Glycine max).
• a sedentary lifestyle (causing a decrease in
muscle mass, an increase in fatty mass, in-
creased accidity of tendons and muscles, a
decreased muscular pumping mechanism,
thereby increasing stasis).
There are dierent methods of classifying
cellulite. That proposed by Rossi and Vergnanini [3]
yields four grades, with their respective
characteristics as follows:
Grade I: The patient is asymptomatic, and there
are no clinical alterations.
Grade II: There are no relief alterations at rest
but after skin compression or after muscular
contraction there is pallor, decreased tempera-
ture and decreased elasticity.
Grade III: Evident at rest there is padded skin
and/or an orange peel appearance; palpable
thin granulations in the deep levels of the der-
mis; pain to palpation; decreased elasticity and
temperature; pallor.
Grade IV: As for grade III with more palpable,
visible and painful nodules, adherence to the
deep levels and an evident wavy appearance of
the skin surface.
The purpose of the study here was to evaluate
the eects of a formulation based on highly
standardized and bioavailable extracts of Vitis
vinifera, Ginkgo biloba and Melilotus ocinalis on
anthropometric parameters related to cellulite,
and consequently, on the course of the pathology.
Materials and methods
Study overview
This was a retrospective clinical trial that
analyzed the results of 90 days of treatment
with 2 dierent food supplements in subjects
who voluntarily underwent a structured lifestyle
change programme on their own initiative. The
experimental group was further divided into 2
dierent subgroups, based on the daily dosage
(2 vs 3 tablets per day) of extracts of Vitis

223
Carvelin
(3 tablets; n=7)
Carvelin
(2 tablets; n=7) Control (n=7)
Mean SD Mean SD Mean SD
Age
(years) 38 ± 9 35 ± 7 38 ± 4
BMI
(kg/m2)22.58 ± 1.27 25.18 ± 3.07 22.29 ± 2.42
WHR 0.69 ± 0.03 0.71 ± 0.04 0.74 ± 0.04
Table 1 Patient demographics and anthropometric
characteristics; data are represented as the mean ±
standard deviation (SD). BMI = body mass index; WHR =
waist-to-hip ratio
Inclusion and exclusion criteria
All women had to be aected by grade II or
III cellulite, according to the scale proposed by
Rossi and Vergnanini [3 ], and still in the presence
of menstrual activity. The motor programme
path had to be started no more than 3 days
since the beginning of data recording.
Ongoing pathologies, pregnancy, treatment with
oestrogen-progestogen drugs, anticoagulants,
anti-inammatories, phlebotropics, the execu-
tion of other specic treatments for cellulite,
changes with respect to usual food consumption
and behaviour (smoking, alcohol consumption)
and the carrying out of further physical activity
in addition to that scheduled (2 weekly aerobic
exercise sessions, lasting approximately 45 min)
constituted exclusion criteria.
Results
At the end of the evaluation period,
considering the extent of the variations and the
standard deviations, there were no appreciable
variations in weight, BMI, waist circumference
or the waist-to-hip ratio (WHR).
The circumference of the hips was
reduced by 2 cm ± 3.1 (p>0.05), 4.1 cm ± 2.1
(p>0.05) and 0.8 cm ± 1.6 (p>0.05), respectively,
in the Carvelin 3, Carvelin 2 and control groups.
Assessment of the circumferences of the
thigh showed a respective reduction for the
Lymphaselect® corresponds to a highly
standardized extract from aerial parts of
Melilotus ocinalis titrated to yield ≤20%
coumarins.
Ginkgoselect Phytosome® corresponds
to a highly standardized extract from Ginkgo
biloba leaves (ginkgoavonglycosides ≥7.0%,
ginkgoterpenes ≥2.0%, bilobalide ≥0.8%,
ginkgolides ≥0.8%, total ginkgolic acids ≤5 ppm),
complexed with distearoylphosphatidylcholine
from soy (Glycine max). Carvelin was used
under the rationale of supporting circulatory
function through the synergistic action of its
components.
The remaining 7 participants took a galenic
preparation containing 250 mg of carnosine
and 250 mg of beta-alanine, taken under the
rationale of supporting the buer function of
skeletal muscle by promoting the execution of
the physical activity programme; this group was
used as a control. Anthropometric data such as
height, weight and circumferences were ob-
tained according to the standard methods.
Patients
Data from 21 adult women aged between
24 and 53 who underwent a motor programme
for the treatment of cellulite were examined.
Normal weight and overweight subjects were
considered (BMI between 19 and 29.7 kg/m2),
with gynoid fat accumulation according to
the method proposed by Björntorp [9], as the
literature seems to indicate that these parame-
ters are not decisive for the development and
progression of cellulite [1, 10, 11]
. All 21 subjects
evaluated according to the method proposed
by Rossi and Vergnanini [3] had a cellulite
framework, with 17 classiable as grade II and
4 classiable as grade III. No subjects classied
with grade I or IV were included in the study, as
they were characterized by excessively dier-
ent conditions compared to the other subjects
evaluated. Patient characteristics are reported
in Table 1.

224
Nutrafoods (2021) 1:220-226 DOI 10.17470/NF-021-0029
grade II. In the control group, at T0 there were
7 subjects classied as grade II; by T90 1 subject
originally classied as grade II was reclassied
as grade I. Therefore, even in the evaluation ac-
cording to the Rossi and Vergnanini scale, the
Carvelin 2 group demonstrated a superior out-
come. All data are summarized in Table 2.
Discussion
At the end of the 90 days, the two
subgroups that took Carvelin showed a greater
reduction in the circumferences of the hips
and thigh compared to the group that took
the control product, although this was not
statistically signicant. Compared with each
other, the group that took two tablets/day of
Carvelin showed superior results to the group
that took three tablets. In the initial speculative
hypothesis, it is possible to consider the role
of the components of Carvelin as inducers of
the activity of dierent cytochrome enzymes;
Carvelin 3, Carvelin 2 and control groups as
follows: proximal thigh 1.1 cm ± 1.1 (p>0.05),
2.3 cm ± 1.3 (p>0.05), 0.1 cm ± 1.5 (p>0.05);
mid-thigh 1.3 cm ± 2.1 (p>0.05), 1.4 cm ± 1.3
(p>0.05), 0.6 cm ± 1.6 (p>0.05); distal thigh 1.1 cm
± 0.9 (p>0.05), 1.3 cm ± 0.5 (p>0.05), 0.3 cm ±
0.5 (p>0.05). At the calf level, no signicant
alterations were detectable.
The scale of Rossi and Vergnanini [ 3] is based on
clinical and symptomatic evaluations aimed at
assigning the patient to 1 of 4 classes.
As the possibility of intermediate scores is not
envisaged, means and standard deviations are
not applicable. At T0 in the Carvelin 3 group
there were 5 subjects classied as grade II and
2 subjects as grade III; by T90 1 subject originally
classied as grade II was reclassied as grade I
and 1 subject was reclassied from grade III to
grade II. Similarly, in the Carvelin 2 group, at T0
there were 5 subjects classied as grade II and 2
subjects as grade III; by T90 2 subjects originally
classied as grade II were reclassied as grade
I and 1 subject was reclassied from grade III to
Carvelin (3 tablets; n=7) Carvelin (2 tablets; n=7) Control (n=7)
T0 ± T90 ± ∆ ± pT0 ± T90 ± ∆ ± pT0 ± T90 ± ∆ ± p
Weight (kg) 60.4 2.8 60.7 3.4 0.3 1.1 >0.05 68.5 3.0 66.7 2.5 -1.8 2.1 >0.05 62.8 4.4 62.1 4.8 -0.7 1.6 >0.05
BMI (kg/m2)22.6 1.3 22.7 1.4 0.1 0.4 >0.05 25.2 3.1 24.5 2.9 -0.7 0.8 >0.05 22.3 2.4 22.1 2.7 -0.2 0.6 >0.05
Waist (cm) 68.3 2.1 69.3 2.6 1.0 1.5 >0.05 75.6 5.4 74.2 4.7 -1.4 1.4 >0.05 74.2 5.8 73.9 6.2 -0.3 1.6 >0.05
Hips (cm) 99.3 2.7 97.3 5.2 -2.0 3.1 >0.05 106.9 4.5 102.8 3.9 -4.1 2.1 >0.05 99.9 3.8 99.1 4.2 -0.8 1.6 >0.05
WHR 0.7 0.0 0.7 0.0 0.0 0.0 >0.05 0.7 0.0 0.7 0.0 0.0 0.0 >0.05 0.7 0.0 0.7 0.0 0.0 0.0 >0.05
Thigh P (cm) 57.3 3.1 56.3 3.9 -1.1 1.1 >0.05 61.4 3.3 59.1 3.2 -2.3 1.3 >0.05 56.2 3.5 56.2 3.4 -0.1 1.5 >0.05
Thigh M (cm) 52.2 2.4 50.8 3.2 -1.3 2.1 >0.05 53.6 4.0 52.1 4.2 -1.4 1.3 >0.05 50.6 3.3 50.0 3.8 -0.6 1.6 >0.05
Thigh D (cm) 39.1 1.7 38.0 2.4 -1.1 0.9 >0.05 41.6 2.1 40.4 2.4 -1.3 0.5 >0.05 38.1 2.4 37.8 2.4 -0.3 0.5 >0.05
Calf (cm) 35.2 1.6 35.1 1.7 -0.1 0.2 >0.05 36.8 1.7 36.6 2.0 -0.2 0.6 >0.05 34.9 2.6 35.1 2.8 0.3 0.4 >0.05
Rossi and
Vergnanini
grade [3]
No. Subjects
T0
No. Subjects
T90 ∆ Subjects No. Subjects
T0
No. Subjects
T90 ∆ Subjects No. Subjects
T0
No. Subjects
T90 ∆ Subjects
I0 1 +1 0 2 +2 0 1 +1
II 5 5 -1/+1 5 4 -2 /+1 7 6 -1
III 2 1 -1 2 1 -1 0 0 0
IV 0 0 0 0 0 0 0 0 0
Table 2 Comparison of the parameters evaluated at time 0 (T0) and 90 days after the start of the programme (T90); data are
represented as the mean ± standard deviation (SD) or the number of subjects (cellulite grading). BMI = body mass index;
WHR = waist-to-hip ratio; Thigh P = proximal thigh; Thigh M = mid-thigh; Thigh D = distal thigh. The Rossi and Vergnanini
scale [3] is based on clinical and symptomatic evaluations aimed at assigning the patient to 1 of 4 classes; as the possibility
of intermediate scores is not envisaged, the mean and SD are not applicable

225
Ginkgo biloba has a known in vitro role as an
inducer of CYP2B1/2, CYP3A1, CYP3A2 and
CYP3A4 [12–15]. Vitis vinifera is known to bring
about the in vitro induction of CYP3A4 [16 ], while
Melilotus ocinalis is known to be metabolized
in vitro by CYP2C9 [ 17]. These actions could
only potentially play a role in the metabolism
of substances with reduced bioavailability,
such as those associated with plant extracts,
compared to the metabolism of drugs for which
there are no clinically conrmed interactions.
For example, in the study of the Ginkgo biloba
special extract EGb 761® no relevant eect on
the in vivo activity of the major CYP enzymes in
humans has been reported, and therefore, this
has no associated potential to cause metabolic
drug– drug interactions [15] .
This possibly explains how a lower dosage
in this case could potentially correspond to a
greater eect. Although size reductions were
found, a p value <0.05 was not reached in any of
the statistical evaluations. Among other factors,
this most likely could be related to the small
sample size of the study. Another signicant
limitation to take into consideration is the fact
that the study was retrospective, based on data
collected during the execution of a structured
and homogeneous lifestyle programme.
The future execution of a double-blind,
randomized, placebo-controlled study where,
in addition to anthropometric evaluation, a bio-
impedance and haematochemical evaluation
is also performed, could contribute signicant-
ly to the solidication of our results while also
clarifying, if conrmed, the reason for greater
ecacy with a lower dose.
Conclusions
The consumption of a mixture of Vitis
vinifera, Ginkgo biloba and Melilotus ocinalis
extracts with high bioavailability, in a
biopharmaceutical formulation, is related to
a greater reduction in hip circumference and
thigh circumferences when taking a dosage
of two tablets/day compared to three tablets/
day, and compared to a control consisting of
a galenic preparation containing 250 mg of
carnosine and 250 mg of beta-alanine. Taking
into account the sample size here, the study
has yielded promising results. This approach
constitutes a potentially interesting solution
within the perspective of a global programme
for the treatment of cellulite, which can
provide a further element to support medical
treatments, pharmacological and nutritional
therapies and structured physical activity.
Further studies with a greater number of
subjects will be needed in the future to clarify
the results obtained, while investigating which
dose represents a more eective treatment in
the timeframe considered.
Author contributions
All authors contributed equally to
writing of the manuscript; all authors read and
approved the nal version of the manuscript.
Funding
This research received no external funding.
Conict of Interest
Alexander Bertuccioli works as a scientic
consultant for the company responsible for
developing Carvelin.

226
15. Zadoyan G, Rokitta D, Klement S, Dienel A, Hoerr R,
Gramatté T, Fuhr U (2012) Eect of Ginkgo biloba special
extract EGb 761® on human cytochrome P450 activity: a
cocktail interaction study in healthy volunteers. Eur J Clin
Pharmacol 68(5):553–560
16. Nishikawa M, Ariyoshi N, Kotani A, Ishii I, Nakamura H,
Nakasa H, Ida M, Nakamura H, Kimura N, Kimura M,
Hasegawa A, Kusu F, Ohmori S, Nakazawa K, Kitada M
(2004) Eects of continuous ingestion of green tea or
grape seed extracts on the pharmacokinetics of midaz-
olam. Drug Metab Pharmacokinet 19(4):280–289
17. Oga EF, Sekine S, Shitara Y, Horie T (2016) Pharmacoki-
netic herb-drug interactions: insight into mechanisms
and consequences. Eur J Drug Metab Pharmacokinet
41(2):93–108
References
1. Terranova F, Berardesca E, Maibach H (2006) Cellulite: na-
ture and aetiopathogenesis. Int J Cosmet Sci 28(3):157–167
2. Avram MM (2004) Cellulite: a review of its physiology and
treatment. J Cosmet Laser Ther 6(4):181–185
3. Rossi AB, Vergnanini AL (2000) Cellulite: a review. J Eur
Acad Dermatol Venereol 14(4):251–262
4. Rossi AM, Katz BE (2014) A modern approach to the treat-
ment of cellulite. Dermatol Clin 32(1):51–59
5. Merlen JF, Curri SB (1983) Rapporti vasculo-tessutali. In:
Ribuo A, Bartoletti CA (eds) La cellulite. Salus, Rome, pp
37–46
6. Smith WP (1995) Cellulite treatments: snake oils or skin
science. Cosmet Toil 110:61–70
7. Isidori A (1983) Fattori predisponenti. In: Ribuo A, Barto-
letti CA (eds) La cellulite. Salus, Rome, pp 49–54
8. World Medical Association (2002) World Medical Associ-
ation Declaration of Helsinki: ethical principles for med-
ical research involving human subjects. J Postgrad Med
48(3):206–208
9. Björntorp P (1990) "Portal" adipose tissue as a generator
of risk factors for cardiovascular disease and diabetes.
Arteriosclerosis 10(4):493–496
10. Quatresooz P, Xhauaire-Uhoda E, Piérard-Franchimont
C, Piérard GE (2006) Cellulite histopathology and related
mechanobiology. Int J Cosmet Sci 28(3):207–210
11. Querleux B, Cornillon C, Jolivet O, Bittoun J (2002) Anat-
omy and physiology of subcutaneous adipose tissue by
in vivo magnetic resonance imaging and spectroscopy:
relationships with sex and presence of cellulite. Skin Res
Technol 8(2):118–124
12. Cho HJ, Yoon IS (2015) Pharmacokinetic interactions of
herbs with cytochrome p450 and p-glycoprotein. Evid
Based Complement Alternat Med 2015:736431
13. Wanwimolruk S, Prachayasittikul V (2014) Cytochrome
P450 enzyme mediated herbal drug interactions (Part 1).
EXCLI J 13:347–391
14. Tang J, Sun J, Zhang Y, Li L, Cui F, He Z (2007) Herb-drug
interactions: eect of Ginkgo biloba extract on the phar-
macokinetics of theophylline in rats. Food Chem Toxicol
45(12):2441–2445