Pentoxifylline, a methyl xanthine derivative, reduces peritoneal adhesions and increases peritoneal fibrinolysis in rats.

Abstract

Peritoneum has an intrinsic fibrinolytic activity that breaks the peritoneal adhesions. Peritoneal injuries with ischemia interfere this fibrinolytic activity and cause adhesions. Pentoxifylline, a methyl xanthine derivative, improves blood flow by decreasing its viscosity and also increases fibrinolytic activity in plasma. We hypothesized that pentoxifylline would increase peritoneal fibrinolysis and ameliorate adhesions. A rat model of peritoneal adhesion (cecal abrasion with gauze, n = 15 for each group) was used to test this hypothesis and cardinal parameters of peritoneal fibrinolysis were measured in peritoneal samples. No medication was given in control animals, while pentoxifylline was administered intraperitonealy (IP) (25 mg/kg, before abdominal closure to whole abdomen) or intravenously (IV) (25 mg/kg, for 9 days after operation) in the experimental groups. At postoperative day 10, peritoneal biopsies were obtained and adhesions were graded qualitatively. Activities and concentrations of tissue plasminogen activator (tPA), plasminogen activator inhibitor type 1 (PAI-1), tPA/PAI-1 complex and hydroxyproline contents were determined. Total adhesion scores were decreased in both treated groups. Mean levels of tPA concentration and tPA activity were increased in the treated groups compared to controls (p < 0.001 and p = 0.001, respectively). The tPA/PAI-1 complex levels were similar among the three groups. PAI-1 levels were lower in animals receiving IP pentoxifylline compared to control animals and those treated with IV pentoxifylline (p = 0.048, p = 0.015, respectively). Peritoneal hydroxyproline levels were similar among the three groups. Our results suggest that pentoxifylline administration either through IV or IP may reduce peritoneal adhesion formation probably by altering peritoneal fibrinolytic activity.

Full text

Page 1

Pentoxifylline Reduces Peritoneal Adhesions249
249
Tohoku J. Exp. Med., 2006, 209, 249-255
Received October 17, 2005; revision accepted for publication May 2, 2006.
Correspondence: Omer Ridvan Tarhan, 6 Mart Ataturk Cad., Yasar Aksel Apt. 10/9, 32100, Isparta, Turkey.
e-mail: drtarhan@yahoo.com
Pentoxifylline, a Methyl Xanthine Derivative, Reduces
Peritoneal Adhesions and Increases Peritoneal
Fibrinolysis in Rats
OMER RIDVAN TARHAN, IBRAHIM BARUT, RECEP SUTCU, 1 YUSUF AKDENIZ and
ONUR AKTURK1
Department of General Surgery, 1Department of Biochemistry, Suleyman
Demirel University Medical School, Isparta, Turkey
TARHAN, O.R., BARUT, I., SUTCU, R., AKDENIZ, Y. and AKTURK, O. Pentoxifylline, a
Methyl Xanthine Derivative, Reduces Peritoneal Adhesions and Increases Peritoneal
Fibrinolysis in Rats. Tohoku J. Exp. Med., 2006, 209 (3), 249-255 ─ ─ Peritoneum has an
intrinsic fibrinolytic activity that breaks the peritoneal adhesions. Peritoneal injuries with
ischemia interfere this fibrinolytic activity and cause adhesions. Pentoxifylline, a methyl
xanthine derivative, improves blood flow by decreasing its viscosity and also increases
fibrinolytic activity in plasma. We hypothesized that pentoxifylline would increase perito-
neal fibrinolysis and ameliorate adhesions. A rat model of peritoneal adhesion (cecal
abrasion with gauze, n = 15 for each group) was used to test this hypothesis and cardinal
parameters of peritoneal fibrinolysis were measured in peritoneal samples. No medication
was given in control animals, while pentoxifylline was administered intraperitonealy (IP)
(25 mg/kg, before abdominal closure to whole abdomen) or intravenously (IV) (25 mg/kg,
for 9 days after operation) in the experimental groups. At postoperative day 10, peritoneal
biopsies were obtained and adhesions were graded qualitatively. Activities and concentra-
tions of tissue plasminogen activator (tPA), plasminogen activator inhibitor type 1 (PAI-1),
tPA/PAI-1 complex and hydroxyproline contents were determined. Total adhesion scores
were decreased in both treated groups. Mean levels of tPA concentration and tPA activity
were increased in the treated groups compared to controls (p < 0.001 and p = 0.001,
respectively). The tPA/PAI-1 complex levels were similar among the three groups. PAI-1
levels were lower in animals receiving IP pentoxifylline compared to control animals and
those treated with IV pentoxifylline (p = 0.048, p = 0.015, respectively). Peritoneal
hydroxyproline levels were similar among the three groups. Our results suggest that pent-
oxifylline administration either through IV or IP may reduce peritoneal adhesion formation
probably by altering peritoneal fibrinolytic activity. ──── pentoxifylline; peritoneal
adhesion; tPA; PAI-1; peritoneal fibrinolysis
© 2006 Tohoku University Medical Press
Peritoneal adhesions are defined as fibrous
bands of tissue that join together organs that are
normally separated. Today, most common cause
of peritoneal adhesions is previous abdominal
surgery. Abdominal surgery causes peritoneal
adhesions in 93% of patients (Menzies and Ellis

Page 2

O.R. Tarhan et al.250
1990). Peritoneal adhesions are an unsolved sur-
gical problem. Abdominal operations cause peri-
toneal injury and ischemia, contamination of for-
eign body (talk, cotton and starch powder),
desiccation and diathermy in peritoneum.
Peritoneal healing differs from that of skin.
The two main differences are epithelization and
consequences of fibrin deposition. First, the peri-
toneal epithelization takes place by the entire sur-
face of the defect simultaneously. Skin epitheli-
zation results from migration of epithelial cells
from dermal edges. Second, fibrin matrix that
was formed after peritoneal injury is broken down
via peritoneal fibrinolysis and absorbed under
normal circumstances. Fibrin matrix is organized
to scar tissue in skin healing.
All injuries to the peritoneum lead to the for-
mation of fibrinous exudate, which is essential for
normal tissue repair. Such fibrinous material
might either reabsorb completely or become orga-
nized into fibrous adhesions. The question is
what factor determines the reabsorbsion or orga-
nization. Ischemia is the most important determi-
nant (Holtz 1984). If ischemia is absent, epitheli-
zation proceeds and fibrin matrix is broken down
by fibrinolysis. Normal peritoneum has an inher-
ent fibrinolytic activity. Fibrinolysis is mainly
activated by tissue plasminogen activator (tPA)
and principal tPA inhibitor is plasminogen activa-
tor inhibitor (PAI) type 1. Ischemia reduces fibri-
nolytic capacity and fibrin matrix changes perma-
nent adhesions (Raftery 1981; Holtz 1984;
Menzies 2004).
Pentoxifylline, a methyl xanthine derivative,
enhances plasma fibrinolytic activity, reduces
plasma fibrinogen levels, inhibits platelet aggre-
gation and increases erythrocyte and leukocyte
flexibility (Jarret et al. 1977; Alkharfy et al. 2000;
Schroer 1985). Eventually, pentoxifylline
improves the flow properties of blood by decreas-
ing its viscosity (Jarret et al. 1977; Schroer 1985).
If pentoxifylline increases local (peritoneal)
fibrinolytic activity, it possibly reduces peritoneal
adhesions. To test this hypothesis, adhesions,
fibrinolytic parameters, and hydroxyproline con-
tents of the peritoneal samples were investigated
after the pentoxifylline administration in a rat
peritoneal adhesion model.
MATERIALS AND METHODS
Animals
The Ethics Committee of Suleyman Demirel
University Medical School approved the experimental
procedures in this study (Suleyman Demirel University
Research Fund, Project Number: 864 M 04). Animals
were obtained from the breeding unit of Suleyman
Demirel University, School of Medicine and “All of the
guiding principles in the care and use of laboratory ani-
mals” were strictly adhered throughout the entire study.
Forty-five female adult Wistar rats weighing 136 -
218 g were used. Throughout the experiment, animals
were given food and water without any limitation.
Animals were assigned in three groups randomly (n = 15
each); control group, intraperitoneal (IP) pentoxifylline
group and intravenous (IV) pentoxifylline group.
Anesthesia and experimental design
The rats were anesthetized with 100 mg/kg ket-
amine hydrochloride (Ketalar®, Parke - Davis, Morris
Plains, NJ, USA), and 25 mg/kg xylazine hydrochloride
(Rompun®, Bayer, Mefar Ilac Sanayi, Istanbul, Turkey)
intramuscularly. The abdomen was shaved and disin-
fected with 70% povidone iodine. All surgical proce-
dures were carried out under sterile conditions. After a
3-cm midline incision had been made, antimesenteric
border of cecum was abraded with a dry sterile gauze
until punctuate bleeding occurred. In control group no
medication was given. Abdominal wall was repaired
with 2 - 0 polypropylene (Prolene®, Ethicon, Woluwe,
Belgium) continuously. In the IP pentoxifylline group
pentoxifylline (25 mg/kg; Trental ampul®, 100 mg/5 ml,
Aventis Pharma, Istanbul, Turkey) was administered to
the whole abdominal cavity, with 1:4 dilution using ster-
ile water (5 ml/kg drug solution). In the IV pentoxifyl-
line group pentoxifylline (25 mg/kg) was administered
on a daily basis via the dorsal tail vein, without dilution,
from the day of operation, until 9 days postoperatively
with the first dose given preoperatively. Abdominal wall
was repaired with 2 - 0 polypropylene continuously. The
skin was closed with 2 - 0 polyglactine suture (Vicryl®,
Ethicon, Woluwe, Belgium) continuously.
At postoperative day 10, a U-shaped incision was
made under the anesthesia that was described previously.
Adhesions were classified into four groups according to
their severity (Evans et al. 1993) (Table 1).
Peritoneal biopsy specimens were taken from the

Page 3

Pentoxifylline Reduces Peritoneal Adhesions251
right lower quadrant while the animals were alive.
Biopsy specimens were rinsed with normal saline solu-
tion and dried with a piece of blotting paper to remove
blood. The biopsy specimens were stored at –80°C in
airtight tubes until the homogenization.
Homogenization
For the homogenization of the biopsy specimens, 1
ml homogenizing solution (0.1% Tritron X 100, 2.5 mM
sodium dihydrogen phosphate, 0.075 M sodium chloride)
was added for every 40 mg tissue. After the biopsy spec-
imens were homogenized in an Ultra Thurrax IKA T - 25
(Janke & Kunken, Staufen, Germany) homogenizer for 3
min, the homogenates were sonicated (Bandelin
Electronic, Berlin, Germany) for 30 sec. The homoge-
nates were centrifuged at 4,000 g for 20 min and super-
natants were stored at –80°C until assay.
Assays for tPA, tPA activity, PAI-1, hydroxyproline and
protein
tPA (IMUBIND® Plasma tPA ELISA, American
Diagnostica Inc., Stamford, CT, USA), tPA activity
(Chromolize® tPA Assay Kit, Biopool, Trinity Biotech
plc., Bray, Co., Wicklow, Ireland), PAI-1 (IMUBİND®
Plasma PAI-1 ELISA, American Diagnostica Inc.) and
tPA/PAI-1 complex (Haemochrom Diagnostica ELISA
tPA-PAI-1 complex GmbH Kleverkämpchen, Essen,
Germany) were determined by enzyme-linked immuno-
sorbent assay (ELISA) in accordance with the manufac-
turer’s instructions. The absorbances of the wells were
measured in a microplate reader. The standard curves
were drawn for each parameter separately through absor-
bance values of standard solutions. The concentration of
the tPA, tPA activity, PAI-1 and tPA/PAI-1 complex were
calculated in view of this standard curves.
The hydroxyproline values were calculated in dry
tissue samples with the partially modified Woessner
method (Woessner 1961).
The protein contents of the supernatants were
assayed with Lowry method. tPA, tPA activity, PAI-1 and
tPA/PAI-1 complex concentrations were normalized with
the protein contents in the supernatants. The results of
tPA, PAI-1 and tPA/PAI-1 complex parameters were
described as ng/mg protein and tPA activity as IU/mg
protein.
Statistical analysis
Results are given as mean ± S.D. Mann Whitney’s
U-test was used to compare the adhesion grades in
groups. Mann Whitney’s U-test was also used to deter-
mine the significance of differences in the hydroxypro-
line content and normalized values of tPA, tPA activity,
PAI-1 and tPA/PAI-1 complex assays. P < 0.05 was con-
sidered significant. The statistical analysis was per-
formed by using the Statistical Package for Social
Sciences (SPSS for Windows Release 10.0, Chicago, IL,
USA).
RESULTS
Adhesion formation
The adhesion grades were significantly lower
in IP (0.80 ± 0.77) and IV pentoxifylline groups
(1.07 ± 0.80) than the control group (2.0 ± 0.76)
(p = 0.001 and p = 0.006 respectively, Fig. 1).
Adhesion scores tended to be lower in IP pentoxi-
fylline than the IV pentoxifylline group, but the
difference was not significant (p = 0.353) (Fig. 1).
tPA
In both IP (21.72 ± 11.09 ng/mg protein) and
IV (21.60 ± 12.18 ng/mg protein) pentoxifylline
groups, tPA levels were twofold higher in perito-
neal samples compared with the control group
(9.61 ± 4.22 ng/mg protein) (p < 0.001 and p =
0.001, respectively). No significant differences
were encountered in tPA levels regarding the
route of pentoxifylline administration (p = 0.723)
(Table 2, Fig. 2).
TABLE 1. Adhesion grading scale.
Adhesion grade Definition
Grade 0
Grade 1
Grade 2
Grade 3
No adhesions.
Filmy and avascular adhesions separated easily by blunt dissection.
Firm and limited vascular adhesions separated by aggressive blunt dissection.
Dense and well-vascularized adhesions separated by sharp dissection.

Page 4

O.R. Tarhan et al. 252
tPA activity
The levels of tPA activity were found higher
in IP (53.85 ± 10.33 IU/mg protein) and IV (41.58
± 10.85 IU/mg protein) pentoxifylline groups than
the control group (30.58 ± 7.14 IU/mg protein)
(p < 0.001, p = 0.003, respectively). On the other
hand, mean tPA activity was higher in animals
receiving IP pentoxifylline compared to animals
receiving IV pentoxifylline (p = 0.005) (Table 2,
Fig. 3).
PAI-1
While PAI-1 levels in IP pentoxifylline group
(5.74 ± 9.11 ng/mg protein) were lower than the
control group (17.8 ± 19.35 ng/mg protein)
(p = 0.048), there was no significant difference
Fig. 1. Adhesion scores. Values are median (hori-
zontal bars), with interquartile range (boxes)
and 10 - 90th centile (error bars). Mann Whit-
ney’s U-test was used.
TABLE 2. Parameters of peritoneal fibrinolysis and hydroxyproline levels in the biopsy specimens.
ParametersControlIP PTX and p valueIV PTX and p value
tPA
(ng/mg protein)
9.61 ± 4.22 a
21.72 ± 11.09 < 0.001b
21.60 ± 12.18 0.001b
tPA activity
(IU/mg protein)
30.58 ± 7.1453.85 ± 10.33 < 0.001 41.58 ± 10.85 0.003
PAI-1
(ng/mg protein)
17.8 ± 19.35 5.74 ± 9.11
0.04811.42 ± 8.34 0.739
tPA/PAI-1 complex
(ng/mg protein)
6.50 ± 2.28 6.32 ± 1.17 0.819 5.62 ± 0.59 0.109
Hydroxyproline
(μg/mg tissue)
52.70 ± 14.0453.57 ± 4.27 0.26950.57 ± 3.33 0.775
Mann Whitney’s U-test was used to determine the significance of differences in the normalized
values of tPA, tPA activity, PAI-1 and tPA/PAI-1 complex assays and HP contents.
a Results are mean ± S.D.
b P values are concerning about the comparisons between control and experimental groups.
Fig. 2. tPA levels in peritoneal samples. Values are
median (horizontal bars), with interquartile
range (boxes) and 10 - 90th centile (error bars).
Mann Whitney’s U-test was used.

Page 5

Pentoxifylline Reduces Peritoneal Adhesions 253
between IV pentoxifylline (11.42 ± 8.34 ng/mg
protein) and control group (p = 0.739). PAI-1
levels differed significantly among the IP and IV
pentoxifylline groups (p = 0.015) (Table 2, Fig. 4).
tPA/PAI-1 complex
There were no differences in the levels of
tPA/PAI-1 complex between the control group
(6.50 ± 2.28 ng/mg protein) and both IP (6.32 ±
1.17 ng/mg protein) and IV (5.62 ± 0.59 ng/mg
protein) pentoxifylline group (p = 0.819 and p =
0.109). However, tPA/PAI-1 complex levels were
slightly lower in IP pentoxifylline group than IV
pentoxifylline group (p = 0.043) (Table 2, Fig. 5).
Hydroxyproline
Peritoneal hydroxyproline measurements did
not differ significantly among the groups (p =
0.269, p = 0.755, p = 0.128, respectively) (Table
2).
DISCUSSION
Present investigation demonstrated that IP
and IV administrations of pentoxifylline reduced
peritoneal adhesion formation. Higher tPA and
tPA activity levels and lower PAI-1 levels in
experimental groups indicate that both application
manners improved fibrinolysis within the perito-
neum. The improvement was more prominent in
IP pentoxifylline group (Figs. 1 - 5). Although
adhesion scores were slightly lower in the IP
pentoxifylline group than the IV pentoxifylline
group, this was not statistically different.
Pentoxifylline did not change hydroxyproline
Fig. 3. tPA activity levels in peritoneal samples.
Values are median (horizontal bars), with inter-
quartile range (boxes) and 10 - 90th centile
(error bars). Mann Whitney’s U-test was used.
Fig. 4. PAI-1 levels in peritoneal samples. Values
are median (horizontal bars), with interquartile
range (boxes) and 10 - 90th centile (error bars).
Mann Whitney’s U-test was used.
Fig. 5. tPA/PAI-1 complex levels in peritoneal sam-
ples. Values are median (horizontal bars), with
interquartile range (boxes) and 10 - 90th cen-
tile (error bars). Mann Whitney’s U-test was
used.