Acta Urol. Jpn. 41; 289-296, 1995
EVALUATION OF EFFECTS OF CHITOSAN
IN PREVENTING HEMORRHAGIC
CYSTITIS IN RATS INDUCED
From the Department of Urology, Nagoya City University Medical School
From the Laboratory of Pathology, Aichi Cancer Center Research Institute
From the Department of Pathology and Microbiology, and the Eppley Institute for
Cancer Research, University of Nebraska Medical Center, Omaha, Nebraska, USA
Margaret K. St. John and Samuel M. Cohen
From Division of Urology, in the Department of Surgery, University
of Nebraska Medical Center, Omaha, Nebraska, USA
Rodney J. Taylor
Hemorrhagic cystitis is a common problem following cyclophosphamide or rad.i.ation therapy.
Chitosan has been shown to be an effective hemostatic agent and promoter of wound healing
in animal experiments. We evaluated the safety and efficacy of intravesical chitosan in an animal
model of cyclophosphamide cystitis. Hemorrhagic cystitis was induced in female F 344 rats by
intraperitoneal cyclophosphamide, IOOmg/kg. Chitosan solution (O.3m!) was instilled intravesi-
cally on day 1 (Group I), on days I, 3, and 5 (Group 2), or 1 hour after the administration of
cyclophosphamide (Group 3). The rats in group 4 were treated with chitosan diluent on day 1
after cyclophosphamide, and the rats in group 5 received intravesical chitosan without cyclophos-
phamide. Sequential examination revealed decreased mortality and lower incidences of severe
bladder bleeding, necrosis and inflammation in Group 3. Treatment delayed until after the
appearance of the cystitis, especially repeated treatments, appeared to make the cyclophosphamide-
induced changes worse. Used within 1 hour of cyclophosphamide administration, before the
cystitis develops, chitosan seemed to have the possibility to inhibit the appearance of hemorrha-
gic cystitis. In addition to the changes in the bladder, severe changes occurred in the kidneys
secondary to cyclophosphamide. (Acta Urol. Jpn. 41; 289-296, 1995)
Key Words: Chitosan, cyclophosphamide, hemorrhagic cystitis, bladder instillation.
Chitosan is a collective term applied to
chitin in various stages of deacetylation
and depolymerization. It is composed of
poly-N-acetyl glycosamine units.
material has been shown, when mixed
with whole blood, to form a tenacious
coagulum!). Taking note of this chemi-
eal, several investigators reported that
chitosan was a good hemostatic agent In
animal experimen tsl-3).
a result of
diffuse capillary fragility and bleeding
within the damaged urothelial lining of
the bladder. This clinical entity is a
frequent result of cyclophosphamide ther-
apy or radiation therapy4)
therapeutic modalities for treating this
entity include withholding the causative
agent, catheter drainage, bladder irriga-
tion with normal saline, formalin or silver
nitrate bladder instillations to cauterize
the bladder, transurethral electrocautery,
and occasionally, cystectomy and urinary
diversion in order to control severe hem-
Acta Urol. Jpn. Vol. 41, No.4, 1995
In this study we evaluated the safety
and efficacy of intravesical chitosan in
an animal model of cyclophosphamide-
MATERIALS AND METHODS
Chitosan was kindly supplied by Hoe-
chst-Roussel Pharmacauticals, Inc., (So-
merville, NJ). Sterile chitosan solution
contained 2 mg/ml chitosan and 2 mg/ml
acetic acid. For a placebo, sterile acetic
acid solution (2mg/mI) was also supplied
by Hoechst-Roussel Pharmaceuticals, Inc.,
Cyclophosphamide was purchased from
Elkins-Sinn, Inc. (Cherry Hill, NJ) and
administered dissolved sterile water.
Ninety-one weanling (4 weeks old),
female F344 rats were obtained from Ch-
arles River Breeding Laboratories, Inc ..
(Kingston, NY). Upon arrival, the rats
were weighed and randomly assigned to
an experimental group by a weight strat-
quarantine on their respective control
diets for 5 days prior to study. The rats
were housed 5 or 6 per cage on dry corn-
cob bedding in polycarbonate cages (16
x18x20 inches) with stainless steel wire
bar covers (Lab Products, Inc., Maywood,
NJ). Animal rooms were maintained at
a temperature of 22+3°C and 50+20%
They were kept in
# of rats
T : cyclophoshamide 100mg/kg J.P.
* : chitosan 0.3 ml (2 mg/ml) bladder instillation (nembutal anesthesia).
* : acetic acid 0.3 ml (2 mg/mI) bladder instillation (nembutal anesthesia) .
• : sacrifice 5/group/time.
humidity on a 12-h light/12-h dark cycle.
Food and water were available ad libitum.
Prolab 3000 diet was purchased from Ag-
way, Inc. (St. Mary's, OH).
A preliminary study demonstrated that
a single dose of cyclophosphamide, 100
mg/kg, i.p., produced hemorrhagic cysti-
tis in 100% of the rats.
Figure I shows the experimental design
Group 1 consisted of 20 rats treated with
a single intraperitoneal Injection of
cyclophosphamide at a dose of 100mg/kg
body weight. After 24 hours, transurethral
intravesical instillation of 0.3 ml chitosan
was performed through a 23 G catheter
(Argyl, Medicut "R", Sherwood Medical
Industries. St. Louis, MO) under Nembutal
(Abbott Laboratories, North Chicago,IL.)
anesthesia. Immediately prior to and one
hour after instillation, the bladder con-
tents were emptied by light abdominal mas-
sage so that the duration of exposure to
instilled compounds was consistent. Group
2 consisted of 21 rats injected with cyclo-
phosphamide as in group I. Intravesical
instillation of 0.3ml chitosan was perform-
ed after 24 hours, and after 3 and 5 days.
Group 3 contained 20 rats injected with
cyclophosphamide as in group 1. Intravesi-
cal instillation of 0.3 ml chitosan was per-
formed one hour after injection of cyclo-
phosphamide. Group 4 contained 20 rats
injected with cyclophosphamide as in
Fig. 1. Experimental design
Okamura et aI.: Effects of chitosan
group 1. After 24 hours, 0.3ml (mg/ml) of
acetic acid was instilled into the bladder
instead of chitosan. A fifth group of 10
rats was used as a negative control group
they were treated with intravesical instilla-
tion of 0.3 ml chitosan, but without prior
The rats were sequentially killed as indi-
cated in Fig. 1.
Two to five rats each
from groups 1, 3 and 4 were killed 3,6 and
9 days post-cyclophosphamide injection. In
group 2, 2 to 5 animals were killed 4, 6,
and 9 days post-cyclophosphamide treat-
ment. Five animals in group 5 were killed
on day 3 post-cyclophosphamide treatment.
As occasion demanded, moribund rats
were killed. The last day the rats were
killed was 21 days post-cyclophosphamide
At death, the liver and each kidney were
observed grossly, then carefully removed
and weighed. These organs were placed
directly into formalin fixative.
ders were inflated in situ with fixative,
and a day later were bisected sagittally and
each half cut into 3 to 4 longitudinal strips.
All tissues were embedded in paraffin, sec-
tioned, stained with hematoxylin and eosin
and examined histopathologically.
Experimental data were evaluated statis-
tically on an IBM 3090 mainframe opera-
ting system VM using the Wilcoxon rank
test (for mortality), and Fisher's exact test
1 tail (for macroscopic and microscopic
findings) from the Statistical Analysis Sys-
tem software package (SAS Institute, Inc.,
About 25 % of the rats died before they
were scheduled to because of cyclophos-
phamide toxicity excessive anesthesia. The
rats were killed or found dead as shown in
Table 1. The mortality in each group was
(including moribund killed) 7/20 (35%) in
group 1,8/21 (38%) in group 2,3/20 (15
%) in group 3, 5/20 (25 %) in group 4,
0/10 (0 %) in group 5. Groups 1 and 2
were significantly different from group 5
at p<0.05 (Wilcoxon rank test). In the
groups administered cyclophosphamide, the
mortality was the lowest in group 3 (how-
ever, no significant difference from the
Table 2 summarizes the macroscopic
findings. In groups 1,2 and 4, white nod-
ules, I mm to 5 mm in diameter, were ob-
Table 1. Numbers of rats killed or found dead at different
times after cyclophosphamide injection.
5 6 8
Mb• moribund death
Acta Urol. Jpn. Vol. 41, No.4, 1995
Table 2. Macroscopic findings of kidney and bladders.
Nodule Hydronephrosis Bleeding
Day I, 3, 5
Acid Day I
20 6 (30%)d
21 6 (29%)d 3 (14%)
8 (38%) "d,.
3 20 0 ( 0%)
10 0 ( 0%) (10%)
0 ( 0%)
0 ( 0%)
• Statistical analyses were performed using Fisher's exact test (I-tail),
b CP: cyclophosphamide.
, Significantly different from Group I, p<0.05.
d Significantly different from Group 3, p<0.05.
< Significantly different from Group 3, p<0.OO5.
f Significantly different from Group 3, p<O.OOI.
• Significantly different from Group 4, p<0.05.
h Significantly different from Group 4, p<O.OI.
Table g. Histopathological findings of kidney and bladders.
Acid Day I
10 (50%)" 5 (25%)
II (52%)" 8 (38%)
3 (15%) 3 (15%)
0 ( 0%)
0 ( 0%)
20 12 (60%)h 5 (25%)
0 ( 0%)
0 ( 0%)
10 0 ( 0%)
0 ( 0%)
0 ( 0%) (10%)
0 ( 0%)
• Statistical analyses were performed using Fisher's exact test (I-tail).
h CP: cyclophosphamide.
, 18 rats for bladder.
d Significantly different from Group I, p<0.05.
< Significantly different from Group 3, p<0.05.
f Significantly different from Group 3, p<O.OI.
" Significantly different from Group 3, p< 0.005.
h Significantly different from Group 3, p<O.OOI.
• Significantly different from Group 4, p< 0.05.
j Significantly different from Group 4, p<O.OI.
• Significantly different from Group 4, p< 0.005.·
I Significantly different from Group 4, p<O.OOI.
served in the kidneys in 29 to 35% of the
rats. The nodules were homogeneous and
not encapsulated. A few cases of hydrone-
phrosis were observed in each group. Al-
most all of these cases involved adhesions
of the bladder. Various grades of bladder
hemorrhage were observed in the cyclo-
phosphamide administered groups.
incidence was highest in group 2 (81%) (p
<0.05, p<O.OOI, and p<O.OI, compared to
groups I, 3, 4, respectively) and lowest in
group 3 (20%). The incidence was similar
in groups I and 4 (45% and 40%, re-
spectively). Some rats had adhesions
caused by severe inflammation, and this
was the most common in group 2 (38 %)
(p < 0.05, compared to the other groups).
Macroscopically, no significant findings
were found in the liver.
Table 3 summarizes the histopathologi-
Okamera et at.:' Effects of chitosan
Fig. 2. Severe hemorrhage of bladder in rat
from group 2, involving the entire
thickness of the wall.
face is at the top. H&E, x 40.
Fig. 3. Erosion and ulceration of bladder in
rat from group 2, H&E, x 20
cal findings. In group 2, the incidence of
severe bleeding (52%) (p < 0.05, p < 0.05,
and p < 0.001, compared to groups I, 3, 4,
respectively) (Fig. 2) , erosion or ulceration
(81 %) (p <0.0 I, compared to group 3)
(Fig. 3), and severe necrosis or inflam-
mation (52%) (p<0.05, p< O.OOI, and p<
0.005, compared to groups I, 3, 4, respec-
tively) of the bladder was higher than that
in the other three groups injected with
cyclophosphamide. None of the group 3 ani-
mals had severe necrosis or inflammation.
Simple hyperplasia of the bladder epitheli-
um was seen in all groups. Papillary or
nodular hyperplasia was seen in groups I,
2, and 5 (11%, 11%, and 10%, respectively)
on day 9 or later.
Interestingly, necrotizing papillitis of the
kidneys was seen in only 5% of the rats in
group 3, in contrast to 50 to 60% inciden-
ces in groups I, 2, and 4 (p< 0.005, p<
0.005, and p < 0.001, respectively). The
white nodules in the kidneys were com-
posed of inflammatory granulation tissue
by microscopic examination. Severe hyper-
plasia of the renal pelvis was seen in 15 to
38 % of the rats in the cyclophosphamide-
administered groups, and it was highest in
group 2 (38 %) and lowest in group 3 (15
Because of advanced autolytic changes,
two bladders in group I were excluded
from histopathological examination.
group 5, one rat had multiple stones com-
posed of calcium phosphate.
DISCUSSION AND CONCLUSIONS
Chitosan is a unique hemostatic agent.
Chitosan-induced coagulum does not de-
pend on the normal clotting cascade mech-
anism, and the resulting "clot" does not
undergo retraction like a normal clot. Fur-
thermore, reaction does not occur with al-
bumin, globulin, or platelets. However,
the same coagulum occurs with heparin-
ized blood, washed red cells, and defibrin-
ated blood2). Mallette et aI.2) have shown in
studies involving dacron grafts placed into
dogs that the chitosan coagulum remains
in place around the graft until replaced by
ingrowth of normal smooth muscle, vaso-
vasorum, and nerve fibrils. This is in di-
rect contrast to the extensive fibrotic reac-
tion and fibrosis that occurs around un-
treated dacron grafts! ,2). In another study
involving the effects of chitosan on capil-
lary bleeding, Branden berg et a].3) showed it
to be an effective topical hemostatic agent
when applied to surgically created central
nervous system wounds in cats. In a re-
cent study, Bartone and Adickes6) demon-
strated the effect of chitosan on wounds of
the genitourinary system in dogs. Wounds
were made in the kidney, ureter and penile
foreskin, with decreased fibrosis observed
with chitosan in all tissues studied.
Hemorrhagic cystitis is a fairly common
clinical problem. It is frequently caused
by cyclaphosphamide therapy or radiation
therapy, but may be associated with vesical
malignancy, bladder amyloidosis, or viral
infection. Current therapeutic modalities
are directed at controlling the bleeding
Acta Urol. Jpn. Vol. 41, No, 4, 1995
until healing occurs. Use of caustic agents
,such as I % formalin or silver nitrate. to
control the acute bleeding often results in
further damage to the urotheli urn 7).
tients in whom this entity occurs are fre-
quently immunosuppressed, chronically ill
cancer patients in whom normal healing is
retarded at best and for whom the risk of
infection and persistent bleeding is much
greater. In addition, patients are known
to be at greater risk to develop bladder
cancer following successful treatment with
cyclophosphamide4,7-9). A further problem
involved with hemorrhagic cystitis regard-
less of the cause is that agents currently
used to treat the hemorrhagic cystitis (i.e.
formalin or silver nitrate) are caustic in
nature and do not infrequently result in
scarring of the bladder with a significant
decrease in the pliability and volume of
the bladder. Such patients may develop
severe urinary problems after treatment
related to a poorly functioning bladder.
Furthermore, the risk of repeated bleeding
episodes is significant.
In this study, we evaluated whether chi-
tosan, an agent of proven value in control-
ling capillary bleeding, can, when used
either prophylactically or therapeutically,
decrease the incidence and sequelae of
hemorrhagic cystitis in an animal model.
Chitosan was not therapeutic as it did
not improve the hemorrhagic cystitis once
it had already developed. Surprisingly, in
the group treated with three instillations,
the bleeding and inflammation of the blad-
der were increased in severity. The chito-
san solution has very high viscosity, which
may cause it to be adherent and irritating.
Once inflammation has occurred, chitosan
might have the potential to make it worse.
The chitosan diluent, acetic acid, may also
have irritating properties because of its
acidity, which might make the inflam-
mation worse. Furthermore, intravesical in-
stillation itself may produce inflammation.
In the group treated only with chitosan
instillation without cyclophosphamide, one
rat had severe inflammation with stones,
providing some support for this sugges·
By contrast, treatment with chitosan be-
fore the cystitis developed seemed to pro-
tect the bladder. The causative agent of
hemorrhagic cystitis following cyclophos-
phamide injection has been identified as
acrolein that is generated in the urine10,Il)
If chitosan really pro stects the bladder,
protection by chitosan might be caused
either by adhesion to the bladder surface,
preventing contact with acrolein, or by a
direct interaction with acrolein, inactiva-
ting it. The incidence of severe bleeding,
severe necrosis or inflammation, simple
hyperplasia in the urinary bladder was
significantly less than the chitosan-treated
groups (Group I and 2) in Group 4 (ace-
tic acid treatment after cyclophosphamide
administration). No substantial difference
but necrotizing papillitis was obtained in
Group 4 when compared with Group 3.
Acetic acid itself might have the preven-
ting effect on cyclophosphamide-induced
In the group (Group 3) involving pro-
phylactic intravesical administration of chi-
tosan, necrotizing papillitis of the kidney
was almost completely inhibited.
mechanism of protection is unknown but
perhaps occurred by reflux of the chitosan,
protecting the kidney pelvis like the blad-
der, or by preventing reabsorption and re-
nal excretion of the active metabolites of
cyclophosphamide and preventing bladder
The finding of necrotizing papillitis was
an unexpected complication of cyclophos-
phamide. Necrotizing papillitis is rarely a
clinical problem, and it is usually observed
only in patients with diabetes mellitus,
pyelonephritis, or dehydration12). The ex-
act mechanism by which necrotizing papilli-
tis develops remains uncertain12,i3).
animal experiments, Molland14)
that aspirin or phenacetin administration
induced necrotizing papillitis, and ische-
mia and dehydration exacerbated it. Ani-
mals may have become dehydrated due to
the toxicity of cyclophosphamide, resulting
in ischemia of the kidney and necrotizing
papillitis. More likely, cyclophosphamide
in the female rat produced severe, ulcera-
Okamura et al.: Effects of chitosan
ting inflammation secondary to toxicity in
the kidney similar to that in the bladder.
In conclusion, chitosan does not have an
active therapeutic effect for hemorrhagic
cystitis once it develops, but it may act as
a prophylactic measure against the devel-
opment of hemorrhagic cystitis and necro-
tizing papillitis secondary to cyclophospha-
mide in a rat model. Additional studies
are required to further evaluate the mecha-
nism involved and the potential usefulness,
espesially for patients receiving very high
doses, such as those in bone marrow trans-
plantation protocols for the treatment of
We gratefully acknowledge the advice and
technical assistance of Dr. Emily Garland and the
assistance of Deboraha Coleman with the prepa-
ration of this manuscript. The studies were sup-
ported in part by a grant from Hoechst-Roussel
Pharmaceuticals, Inc., Somerville, N.J ..
I) Malette WG, Quigley HJ, Gaines RD, et al.:
Chitosan: A new hem astatic. Ann Thorac
Surg 36: 55-58, 1983
2) Malette WG, Quigley HJ and Addickes ED:
Chitosan effect in vascular surgery, tissue
culture and tissue regeneration.
zarelli R, Jeuniaux C, Gooday C (eds) Chi-
tin in nature and technology.
Plenum Publishing Corporation 435-442,
3) Brandenberg G, Leibrock LG, Shuman R,
et al.. Chitosan: A new topical hemostatic
agent for diffuse capillary bleeding in brain
tissue. Neurosurgery 15: 9-13, 1984
4) Stillwell T J and Renson RC Jr: Cyclophos-
phamide-induced hemorrhagic cystitis. Can-
cer 61: 451-457, 1988
5) Martin RA, Daly AM, Difonzo CJ, et al.:
Randomization of animals by computer pro-
gram for toxicity studies. J Am ColI Toxi-
col 3: I-II, 1984
6) Bartone FF and Adickes ED: Chitosan: Ef-
fects on wound healing in urogenital tissue:
Preliminary report. J Urol 140: 1134-1137,
7) Levine LA and Riche JP: Urological com-
plications of Cychophosphamide.
141: 1063-1069, 1989
8) FairChild WV, Spence CR, Solomon HD, et
al.: The incidence of bladder cancer after
Cyclophosphamide therapy. J Urol 122: 163
9) Ansell ID and Castro JE: Carcinoma of the
bladder complicating Cycloshosphamide ther-
apy. Br J Urol 47: 413-418, 1975
10) Cox P J: Cyclophosphamide cystitis-identi-
fication of acrolein as the causative agent.
Biochem Pharmacol 28: 2045-2049, 1979
II) Brock N, SteKar J, Pohl J, et aL Acrolein,
the causative factor of urotoxic side-effects
of cyclophosphamide, ifosfamide, trofosfami-
de and sufosfamide. Arzneimittelforschung
Drug Research 29: 659-661, 1979
12) EKnoyan G, Qunibi WY, Grissom RT, et
al.: Renal papillary nicrosis: An update.
Medicine 61: 55-73, 1982
13) Nanra RS, Chirawong P and Kincaid-Smith
P: Medullary ischemia in experimental anal-
gestic nephropathy. The pathogenesis of re-
nal papillary necrosis. Aust N Z J Med 3:
14) Molland EA: Experimental renal papillary
necrosis. Kidney Int 13: 5-14, 1978
Received on September 21, 1994)
Accepted on January
296 Download full-text
Acta Urol. Jpn. Vol. 41, No. 4, 1995
ラ ッ トCycl ophospham i de誘発 出血 性 膀 胱 炎 に 対 す る
Chi tosan膀 胱 内注 入 効 果 の 検 討
名古屋市立大学医学部泌尿器科学教室(主 任: 郡 健二郎教授)
愛知県がんセンター研究所病理学第一部(部 長: 立 松生衛)
K. St. John and Samuel
and the Eppley
of Nebraska Nebraska USA
of Urology in the Department
USA Medical Omaha,
メ ス ラ ッ トcycl ophospham i de誘発 出血 性膀 胱 炎
に対 して, chi ti nの 一 種 で ユ=一 ク な 止 血 作用 を有
す るchi tosanの 膀 胱 内注 入 を行 い, そ の 効果 につ い
て 検 討 を 行 った. F344)ス ラ ッ トを用 い, 100m g/
Kgのcycl ophospham i deを腹 腔 内投 与 す るこ とに よ
り, 出血 性膀 胱 炎 を誘 発 し, ch三tosan(O . 3m l )を以
下 の よ うに膀 胱 内注 入 した. 第1日目(グ ル ー プ1),
第1・3・5日目(グ ル ー プ2), cycl ophospham i de
投 与1時 間後(グル ープ3). グ ル ー プ4は第1日目
にchi tosan溶解 液 の みを 膀 胱 内 注入 し, グ ル ー プ5
はcycl ophospham i deを投 与 せ ず, chi tosanの み を
膀 胱 内 注 入 した. 経時 的 に 屠 殺 し, 肉 眼 的, 病理 組 織
学 的 に 検 討 した結 果, グル ー プ3におい て 死 亡 率 の減
少 と, 肉 眼 的 膀 胱 出 血, 壊 死 お よ び炎 症 の発 生 率 の低
下 が認 め られ た. 膀胱 炎発 生 後 のchi tosan膀 胱 内注
入 は, 特に繰 り返 し施 行 した 場 合, cycl ophospham i -
deによる変 化 を 悪 化 させ る傾 向 が み られ た. cycl o-
phospham 五de投与1時間 後 の投 与 群 では, chi tosan
は 出 血 性膀 胱 炎 の発 生 を 抑 制 す る可 能 性 が 示 唆 され
た. さらに, こ れ ら 膀 胱 の 変 化 以外 に, cycl ophos-
pham i deに よる 腎 の変 化 も認 め られ た.
(泌 尿 紀 要41=289-296. 1995)