вход по аккаунту


Effects of pulse methylprednisolone on bone and marrow tissues. Corticosteroid-induced osteonecrosis in rabbits

код для вставкиСкачать
Vol 40, No 11, November 1007, pp 2055-2064
0 1997. American College ot Rhcumatology
Corticosteroid-Induced Osteonecrosis in Rabbits
Objective.To investigate the effects of pulse methylprednisolone acetate on bone and bone marrow tissues
and to clarify the causal factors of corticosteroidinduced osteonecrosis (ON) by using an experimental
animal model.
Methods. Male adult Japanese white rabbits were
injected once with 20 mg/kg of methylprednisolone into
the right gluteus medius muscle. Seven rabbits were
killed a t 4 weeks, 4 at 6 weeks, 4 at 8 weeks, and 6 a t 10
weeks. Both histopathologic and hematologic studies
were performed every week.
Results. By 4 weeks after the steroid injection,
43% of the rabbits studied had developed multifocal ON
lesions in the femur and/or humerus. In 1 rabbit, a
thrombus was detected in an arteriole adjacent to the
necrotic area at 4 weeks. After 6 weeks, there was also
progressive histologic evidence of revascularization,
with granulation tissue, and osteoblastic repair, with
appositional bone formation. Hyperlipemia, fatty liver,
and intraosseous fat embolism were observed in conjunction with thrombocytopenia and hypofibrinogenemia.
Conclusion. A single injection of high-dose corticosteroids was found to be capable of inducing thrombocytopenia, hypofibrinogenemia, and hyperlipemia with
multifocal ON in several bones.
Supported in part by a Research Grant for Intractable
Diseases from thc Ministry of Health and Welfare of Japan, by the Hip
Joint Foundation of Japan, Inc., and by a Grant-in-Aid in Scientific
Rcsearch (no. 07407041) from the Ministry of Education and Culture
of Japan.
Takuaki Yamamoto, MD, PhD, Takahiko Irisa, MD, Yoichi
Sugioka, MD, PhD, Katsuo Sueishi, MD, PhD: Kyushu University,
Fukuoka, Japan.
Address reprint requests to Takuaki Yamamoto, MD, PhD,
Department of Pathology I, Faculty of Medicine, Kyushu University,
3-1-1 Maidashi, Higashi-ku, Fukuoka 812-82, Japan.
Subrnittcd for publication February 19, 1997; accepted in
revised form June 23, 1997.
Osteonecrosis (ON) has been recognized as an
important complication in 3-40% of patients who have
received high- or relatively low-dose corticosteroids for
the treatment of such underlying diseases as systemic
lupus erythematosus, rheumatoid arthritis, leukemia,
and asthma, and after renal or cardiac transplants (1-4).
This corticosteroid-induced ON is known to involve
multifocal regions and to affect multiple bones (5-7).
Small regions of ON in the femoral head and long bones
are usually silent and do not collapse, while large
osteonecrotic regions occurring in the femoral head
often result in hip joint pain due to subchondral collapse
and joint destruction (8-1 0). Despite numerous efforts,
we have yet to develop any useful preventive measures
against this disease, mainly because of the uncertainty of
the pathogenesis of ON and the lack of a useful experimental model of ON.
Several risk factors for ON have been suggested,
including chronic alcoholism, sickle cell disease, decompression sickness, and corticosteroid therapy, and the
high-dose administration of corticosteroids has been
recently considered to be one of the major risk factors
for ON (3,5,11-14). Many experimental studies using
corticosteroids have suggested the possible pathogenesis
of corticosteroid-induced ON, such as an enlargement of
the size of fat cells (15), an accumulation of lipid within
the osteocytes (16), and fat emboli (12,13,15,17). However, to our knowledge, the studies have never succeeded in developing histologically definitive ON (1820) in animals by the injection of corticosteroids alone
(15,16,21-25). Perhaps the studies were focusing exclusively on the epiphyseal regions of the femoral and
humeral heads that result in subchondral collapse, and
were not thoroughly evaluating the metaphyseal and
diaphyseal regions. It has been clinically shown that
several ON regions never undergo collapse, even if they
are located in the femoral head (8-10). Every ON region
Table 1. The prevalence, location, and extent of osteonecrosis (ON) at 4 weeks*
No. (5%) of ON lesions
%I area of
mean -t SD
Distal femur or
humerus (condyle)
(n = 14)
(n = 14)
6 (43)
8 (57)
7 (50)
6 (43)
2 (14)
0.69 2 0.46
* Adult male Japanese white rabbits were given a single injection of methylprednisolone (20 mgikg).
Percentage area of involvement was calculated as the necrotic area divided by the total area in the
proximal one-third of eithcr the femur or humerus.
without collapse should therefore be carefully examined
throughout the whole area of the bone.
In this study, we investigated the hematologic and
histopathologic changes, not only in the epiphysis, but
also the metaphysis and diaphysis of corticostcroidtreated rabbits. We also discuss the possible causal
factors of ON development in this model.
Animals. Adult (growth plate already closed) male
Japanese white rabbits (Kyudo, Tosu, Japan), weighing from
3.0 kg to 4.5 kg, were used in this study at the Animal Center
of Kyushu University and were maintained on a standard
laboratory diet and water. All experiments were reviewed by
the Common Ethics Committee for Animal Experiments at
Kyushu University, and were conducted in accordance with the
Guidelines for Animal Experiments of Kyushu University, the
Law (no. 105), and the notification (no. 6) of the government
and the Committee on Ethics in Japan.
Treatments. Thirty rabbits were injected once with 20
mgikg of body weight of methylprednisolone acetate (Upjohn,
Tokyo, Japan) into the right gluteus medius muscle. Seven
rabbits were killed at 4 weeks, 4 at 6 weeks, 4 at 8 weeks, and
6 at 10 weeks after the injection of methylprednisolone. In
addition, 5 rabbits were injected once with physiologic saline (1
mlikg of body weight; Otsuka Pharmaceutical, Naruto, Japan)
into the right gluteus medius muscle and killed at 4 weeks as
the control. The animals were anesthetized with an intravenous injection of pentobarbital sodium (25 mgikg of body
weight; Abbott, Chicago, IL), and then were killed by exsanguination via aortectomy.
Closure of growth plates in both the femur and humerus was histologically confirmed in all rabbits. Three animals died at 2 weeks and 1 at 3 weeks after the injection of
methylprednisolone. These 4 rabbits were excluded from the
Tissue preparation. For light microscopic examination,
tissue samples were obtained from the femur, humerus, liver,
kidney, heart, lung, and spleen at the time of death, and then
were fixed for 1 week with 10% formalin-0.1M phosphate
buffer, p H 7.4. The bone samples were decalcified with 25%
formic acid for 3 days and then neutralized with 0.35M sodium
sulfate for 3 days. The specimens were embedded in paraffin,
cut into 4 p m sections, and stained with hematoxylin and eosin,
elastica van Gieson’s, and Masson’s trichrome.
Evaluation of ON. Bone samples from the femur and
humerus were cut along the coronal plane in the proximal part
and axial plane in the distal part, and were histopathologically
examined for the presence of hematopoietic cell necrosis
(cytolysis, karyorrhexis, or karyolysis), fat cell necrosis (the loss
of either nuclei or distinct cell borders), and ON. ON was
blindly assessed by three pathologists. based on the diffuse
presence of empty lacunae or pyknotic nuclei of osteocytes in
the bone trabeculae, accompanied by the surrounding bone
marrow cell necrosis. Only bone marrow cell necrosis showing
tissue debris consisting of both the hematopoietic cell necrosis
and fat cell necrosis in which no bone trabecula was included
was assessed as ON. However, lesions consisting of only a few
empty lacunae in the normal bone trabecula and/or fat cell
necrosis was excluded from the assessment of ON width and
incidence in this study. The presence of either granulation
tissue, fibrosis: or appositional bone formation against necrotic
tissue was also carefully determined in regard to the repair
Measurement of the extent of ON. For all rabbits with
ON, the percentage area of involvement (necrotic area/total
area of the proximal one-third of either the femur or the
humerus) was measured morphometrically by a previously
reported method (26). The necrotic area was defined as the
necrotic bone and bone marrow tissue; thus, the repair process
against ON, including granulation tissue and appositional bone
formation, was excluded from this area. Briefly, an image of
the sections was picked up by camera and then was electronically directed to the image processor, where it was then
digitized into numerical values and interpreted by a digital
computer. The projected image of the sections displayed on
the video monitor was measured using an interactive mousepad tracing instrument, while the corresponding morphometric
data were processed automatically by the computer system.
Hematologic examination. All blood samples were
obtained with the animals in a fasting state and were collected
into 4.5-ml vacuum tubes containing 0.5 ml of 3.8% sodium
citrate anticoagulant from auricular arteries. Hematologic and
chemical evaluations were made for plasma levels of platelets,
fibrinogen, free fatty acid (FFA), triglycerides, cholesterol,
glutamic oxaloacctic transaminasc (GOT), and glutamic pyruvic transaminase (GPT) in all animals before and after the
injection of methylprednisolone.
Statistical analysis. The data on the extent of the
necrotic area and the hematologic examinations are given as
the mean +- SD. All data on the hematologic examinations
were statistically assessed by a repeated measurement analysis
and Student’s t-test. Statistical differences were considered
significant when the P value was less than 0.05.
Prevalence and location of ON. Multifocal ON
lesions were identified in both the femur and humerus.
ON was mainly localized in the metaphysis and diaphysis, but not in the epiphysis. In the femoral bones, the
prevalence of ON was 43% at 4 weeks, 13% at 6 weeks,
25% at 8 weeks, and 25% at 10 weeks. The incidence of
ON in the femoral condyle was SO% at 4 weeks (Table
1). ON gradually accompanied the repair process after 6
weeks, and thereafter, the necrotic marrow was almost
wholly replaced by the repair tissue in 4 of 12 femoral
bones examined at 10 weeks. These completely repaired
necrotic regions were excluded from the count of ON,
and thus, the prevalence of ON decreased gradually.
In the humeral bones, the prevalence of ON at 4
weeks was also 43%, 25% at 6, 13% at 8, and 10% at 10
weeks. ON occurred in neither the epiphysis nor the
distal part of the humerus (Table 1). No ON lesions
could be seen in any of the rabbits in the control group.
Extent of ON. The extent of O N was shown as the
percentage area of involvement (necrotic area/total area
of the proximal one-third of either the femur or humerus). It was about 0.75 f 0.58% in the femur and 0.69 -f
0.46% in the humerus at 4 weeks, when ON was not
associated with the formation of the repair tissue. Thus,
these rates of necrosis reflect only the necrotic area
(Table 1). The repair process gradually appeared after 6
weeks, and thus the precise necrotic area could only be
measured at 4 weeks after the methylprednisolone
Histopathologic features. Macroscopically, ON
in both the femur and humerus was only observed in the
metaphysis and diaphysis as yellowish-colored areas
(Figures 1A and B). No ON could be seen in the
epiphysis. Histologically, typical ON showed an accumulation of bone marrow cell debris and bone trabeculae
demonstrating empty lacunae and occasionally containing some pyknotic nuclei of osteocytes (Figures 2A and
B). The repair process represented by the granulation
tissue and appositional bone formation around ON sites
varied based on the number of weeks after the corticosteroid injection. At 4 weeks, the granulation tissue was
scarcely formed around ON sites, and only the accumulation of serofibrinous exudate was seen around the
Figure 1. Gross sections of the femur (A) and humerus (B) of adult
male Japanese white rabbits 4 weeks after corticosteroid injection. The
yellowish-colored areas (arrows) located mainly in the metaphysis
underwent osteonecrosis.
necrotic area (Figures 2A and B). At 6 weeks, fibrosis
and vascular- or cellular-rich granulation tissue surrounded the necrotic area, but the presence of appositional bone formation was still unclear (Figure 2C). At
10 weeks, necrotic bone tissue was surrounded by prom-
Figure 2. Histologic features of osteonecrosis (ON) 4,6, and 10 weeks after corticosteroid injection. A, At 4 weeks, the bone trabeculae show empty
lacunae, while the surrounding hone marrow tissue also underwent necrosis. Neither granulation tissue nor appositional bone formation can be seen
between the ON (Nec) and living (Liv) bone marrow tissue. Only serofibrinous exudate (arrows) surrounds the area of ON. B, At 4 weeks, bone trabeculae
show empty lacunae, and surrounding marrow tissue consists of necrotic marrow cell debris. C, At 6 weeks, vascular-rich granulation tissue (arrows)
surrounds the ON area. D, At 10 weeks, the repair process becomes more prominent. Necrotic hone trabeculae are surrounded by the appositional bone
formation (arrows) and granulation tissue. E, The exudate of amorphous material is seen in thc space between the adipocytes, and some adipoqtcs also
appear to be small and atrophic. (Hematoxylin and eosin stained; original magnification X 100 in A, C, and D, X 400 in B, and X 150 in E.)
Figure 3. Morphologic changes in the vessels 4 weeks after corticosteroid injection. A, An organizing thrombus is secn in the intraosseous arteriole,
which is accompanied by fibrinous exudate in its wall, just adjacent to the osteonecrosis in the metaphysis of thc fcmur. B, The intraosseous vein
occluded by fat emboli and red thrombi is noted in the diaphysis of the humerus, (Hematoxylin and eosin stained; original magnification X 200 in
A and X 100 in B.)
(x I 0
* : p<0.05 (both ON + and -)
No. of animals examined
O N + : 8
O N - 1
3 3
Figure 4. Sequential changes in blood platelet levels. Although there was no significant difference
between the rabbits with osteonccrosis (ON+) and those without ON (ON-), the blood platelet
levels significantly dccreased 1 wcek aftcr the injcction of methylprednisolone ( P < O.OS), and then
gradually recovered, reaching almost normal levels or even a littlc higher after S weeks. Bars show
the mean 5 SD.
o . l . , . , . , , , , , . , . , . , . I . , , , . I , ,
No. of animals examined
O N + :
O N - : 6
8 7
Figure 5. Free fatty acid levels in the plasma. A significant increase occurred at 2 weeks ( P <
0.01), and then gradual recovery began at 8 weeks. ON = osteonecrosis. Bars show the mean f SD.
inent appositional bone formation and dense fibrotic
granulation tissue (Figure 2D), and some ON regions
were completely replaced by the reparative tissue. Bone
marrow edema was occasionally seen in both the femur
and humerus at 4 weeks (Figure 2E).
A thrombus was noted in the intraosseous arteriole of the metaphysis just adjacent to the necrotic area
in 1 rabbit at 4 weeks. Fat emboli accompanied by
thrombus formation was noted in the intraosseous vein
in the diaphysis in 1 rabbit with ON at 4 weeks (Figures
3A and B). No evidence of angiitis or rupture was seen.
Histopathology in the other organs. Four weeks
after the injection of methylprednisolone, almost all
rabbits had a fatty liver, and in 3 of 8 rabbits, focal
necrosis of the liver parenchyma, with calcification, was
occasionally noted. However, this prominent fatty metamorphosis in the liver gradually recovered by 10 weeks.
Neither ischemic lesions nor thrombosis was noted in
any other organs examined weekly. In 1 rabbit, a small
calcified region was noted in the superficial parts of the
myocardium at 4 weeks.
Hematologic examination. There was no significant difference in any of the factors we examined
between the rabbits with ON (ON+) and those without
ON (ON-). The blood platelet levels significantly decreased 1week after the injection of methylprednisolone
( P < 0.05), and then gradually recovered and reached
almost normal levels or even a little higher at 5 weeks
(Figure 4). The sequential changes in fibrinogen were
almost the same as those of the platelets.
Levels of free fatty acids (FFAs) in the plasma
significantly increased at 2 weeks (P < O.Ol), and then
gradually recovered at 8 weeks (Figure 5). However, the
individual differences varied, and thus, the SD showed a
wide range. Triglycerides, cholesterol, GOT, and GPT
levels showed the same sequential changes as those of
the FFAs (Figures 6 and 7). At 10 weeks, no apparent
abnormal findings were detected in any of the factors
examined. In the control group, no significant changes
were found in any of the factors examined.
In this study, we succeeded for the first time in
developing an experimental corticosteroid-induced ON
in rabbits, showing the essential histopathologic features
of ON in humans, namely, osteocytic necrosis surrounded by necrotic bone marrow cells and/or the repair
No. of animals examined
ON-: 6
Figure 6. Sequential changes in plasma triglyceride levels. which significantly increased at 2 weeks.
P < 0.01 in rabbits with osteonecrosis (ON); P < 0.05 in rabbits without ON. After 5 weeks, rabbits
with ON tended to show higher triglyceride levels than rabbits without ON. Bars show the mean
i- SD.
(u) 400
* : p<O.O1
(both GOT and GPT)
~ ~
l ~
~ l
No. of animals examined
1 3 6
1 4 1 4 1 4 1 3
Figure 7. Levels of glutamic oxaloacetic transaminase (GOT) and glutamic pyruvic transaminase
(GPT) in plasma. A significant increase (P< 0.01) occurred at 2 weeks, and then gradual recovery
began at 6 weeks. Bars show the mean t SD.
process (5,18-20,27). Since the relationship between the
occurrence of ON and the systemic administration of
corticosteroids was first reported in 1957 (4), there have
been several experiments that demonstrated many
steroid-related histopathologic and pathophysiologic alterations in the bone tissue, such as fat embolism
(21,22), an increased pressure of bone marrow in the
femoral head (24), an enlargement of the fat cell size
(15), and an accumulation of lipid within the osteocytes
(1 6). However, no histopathologically definitive ON has
been described in those steroid-treated animals. We do
not know any definite reasons for the discrepancies
between the previous experiments and our study, but the
following possibilities can be considered.
First, ON-prone sites may differ in mammalian
species, and thus the distribution of ON should be
characterized by examining all parts of the bonc. Most
previous studies have mainly examined the femoral
head, expecting a collapsed lesion (15,16,22,24,25).
However, Matsui et a1 (28) reported that ON was
recognized in only the metaphysis and diaphysis in rabbit
models of ON produced by serum sickness with horse
serum and corticosteroid treatment. We (26) have also
reported that ON induced by the Shwartzman reaction
and corticosteroids was mainly recognized in the rabbits’
metaphysis, although some ON regions were also observed in the epiphysis of the femur and humerus. For
the above-stated reasons, the differences observed
among species may result in the difference in the
location of ON. Moreover, in humans, ON is known to
occur in multifocal regions in various part of bone tissue
(5-7), and even the ON occurring in the femoral head
sometimes remains silent and does not collapse if the
region is small or is located in the central portion (8-10).
We therefore should not exclude ON occurring in the
metaphysis or diaphysis, just because these regions are
not located in the area of the epiphysis or do not
undergo collapse. Whether the ON region undergoes
collapse or not is mainly due to the size and location of
the ON against the weight-bearing area, and we should
also keep in mind that ON undergoing collapse is only
one part of all the ON that occurs in humans.
Second, it is still controversial as to whether or
not only osteocytic necrosis without surrounding bone
marrow cell necrosis can be histologically assessed as
ON. Gold et a1 (22) reported that daily or alternate-day
injections of corticosteroid (0.8 mg) produce osteocyte
death, necrotic debris, osteoporosis, and intravascular
fat emboli in the femur, while ON was individually
evaluated by the percentage of empty lacunae or the
presence of necrotic bone debris, respectively. AlthoLgh
Kawai et a1 (16) also reported that corticosteroid admin-
istration (4.2 mg/kg of body weight) could induce osteocytic necrosis in the femoral heads of rabbits, they did
not report whether marrow necrosis surrounding the
bone that had osteocytic necrosis occurred in their
experiments. The histologic features of ON in humans
are characterized by the subchondral necrotic area in the
femoral head, namely, osteocytic death resulting in
empty lacunae and surrounding bonc marrow necrosis.
which commonly accompany the reparative responseappositional bone formation, vascular-rich or cellular
granulation tissue, and the infiltration of chronic inflammatory cells-around the necrotic region (3,5,18-20,29).
This reparative process depends on the clinical course,
including the time after ON onset, steroid treatment,
and other factors (18-20). In our study, the repair
process was extremely weak at 4 weeks, but the apparent
repair tissue around the necrotic foci could be recognized after 6 weeks. Therefore, we believe that both
bone marrow cell necrosis and the repair tissue around
necrotic bone trabeculae are fundamental features in the
histologic cvaluation of ON; thus, in this study, only the
empty lacunae without bone marrow necrosis werc not
assessed as ON.
Third, the timing of the examinations seems to be
important. The interval between corticosteroid administration and the onset of ON (not symptoms) is still
unknown in both humans and rabbits, but some clinical
studies have reported that early cases of ON were
detected 2-3 months after the steroid treatment (2,30).
In our study, ON regions can be clearly detected at 4
weeks, and thereafter are gradually rcplaced by reparative tissue. Most previous studies have conducted histopathologic examinations at 6-20 weeks after steroid
treatment (153422). The duration after the steroid
treatment may thus affect the histopathologic features
of ON.
In our study, a single dose was used, and the
dosage (20 mg/kg of body weight) was much higher than
that used in previous animal studies (15,16,22,24,25).
Abeles et a1 (1) suggested that the high initial corticosteroid dosage in patients with systemic lupus erythematosus may be apt to induce osteonecrosis of the femoral
head. The hasty introduction of high-dose corticosteroid
therapy has also been suggested to participate in the
occurrence of steroid-induced ON (14). Moreover, the
prevalence of ON after renal transplantation has also
decreased since the use of cyclosporin A, which can
allow for the reduction of the corticosteroid dosage
during the first postoperative week (2). Based on these
findings, high initial doses of corticosteroids are considered to be one of the major risk factors for ON, and we
therefore used the high-dose administration of methyl-
prednisolone in the same manner as that of steroid pulse
therapy in humans. However, in the rabbits treated with
high-dose corticosteroids, only 40% developed ON,
while the others did not. Moreover, the fact that lower
doses of corticosteroids, even by intraarticular injection,
can also result in ON in humans, and that only 3-40% of
persons treated with corticosteroids have developed this
condition, suggests the necessity of further investigations
of other ON factors, such as the variability of the
response and the metabolism of corticosteroids.
Regarding the pathogenesis of ON in this model,
corticosteroids are known to modulate not only the
immune and inflammation systems (3 l), but also lipid
metabolism (25). Several investigators have thus concluded that intravascular fat embolism is the critical
event in the development of corticosteroid-induced ON.
There have been numerous reports showing the presence of fat emboli in association with bone lesions in
experimental animal models (12,13,15,16,22,25). Moreover, high plasma or serum levels of lipids after highdose steroid therapy have also been reported to significantly correlate with the occurrence of ON in humans
(32). On the other hand, corticosteroids have been
considered to cause a hypercoagulable and hypofibrinolytic state of plasma (33-35). It thus seems clear that no
one factor adequately accounts for the development of
ON. However, in this model, necrotic lesions were often
recognized in the organs containing prominent fat tissue,
such as bone marrow and liver with fatty metamorphosis. The serial hematologic studies revealed that significant changes in the coagulation system, the platelet
counts and fibrinogen levels in plasma, and the lipid
levels were detected 1-2 weeks after the corticosteroid
injection in rabbits with and without ON. These histologic and hematologic studies suggest that both a hypercoagulable state and high lipid levels in the early period
after the injection of methylprednisolonc seem to be
important for the development of ON in this model.
However, in this study no significant differences were
observed in either the plasma levels ol' platelets or
fibrinogen or the lipid levels between the rabbits with
and those without ON. Abnormalities of plasma lipids
have been reported to enhance the coagulation process
and thrombogenesis (36), and bone marrow contains
prominent adipose tissue. These findings may indicate
that the local changes that occur in the bone marrow
tissue may be more significant than those revealed by
systemic hematologic examination. Further evaluations,
including local expression of such coagulationfibrinolysis-related factors as tissue factor, plasminogen
aclivator inhibitor 1, and plasminogen activators, and
histopathologic examinations in the early weeks may
therefore be necessary to clarify the direct causal mechanisms of ON in this model.
The ON observed in this rabbit model has several
similarities to ON in humans. First, the histopathologic
features of ON in this model are characterized by empty
lacunae accompanied by surrounding bone marrow cell
necrosis and such resultant reparative changes as granulation tissue and appositional bone formation. These
features are analogous to those in humans (5,18-20,27).
Second, the multifocal nature of the ON, which includes
the femoral condyle and humerus, is also similar to that
in humans (5-7). However, ON in this model was not
found in the epiphysis, and the extent was small and not
sufficient to induce a collapse of the ON region. These
features are different from those in humans, in whom
joint destruction due to collapse occurs. The reason why
ON does not occur in the subchondral area remains an
open question, and further examinations should be
performed, with consideration for species differences.
Recent animal models of O N were produced by
serum sickness (28) and the Shwartzman reaction (26) in
addition to corticosteroid treatment, but the O N model
reported herein, which is simple and reproducible, may
be useful not only in researching the effects of corticosteroids on the bone and bone marrow tissue, but also in
clarifying the etiology and early pathogenesis of steroidinduced ON. Moreover, in order to design appropriate
drug treatment for the prevention of ON after corticosteroid treatment in humans, further research into the
pathogenesis of this model may thus result in many
useful findings, while taking into account the several
differences between ON in this model and ON in
humans, such as the distribution and extent of ON.
We would like to thank Hiroshi Fujii and Shizuko
Y ugawa for their very helpful technical assistance.
1. Abeles M, Urman JD. Rothfield NF: Aseptic necrosis of bone in
systemic lupus erythematosus: relationship to corticosteroid therapy. Arch Intern Med 138:750-754, 1978
2. Landmann J, Rcnner N, Gather A, Thild G, Harder F: Qclosporin A and osteonecrosis of the fcmoral head. J Bone Joint Surg
An1 69A:1226-1228, 10x7
3. Mont MA, Hungerford DS: Non-traumatic avascular necrosis of
thc femoral head. J Bone Joint Surg Am 77A:459-473, 1905
4. Pietrogrande V, Mastromarino R: Osteopatia da prolungato trattamento cortisonico. Orthop Travmatol Protez 28:791-810, 19.57
5. Maiikin HJ: Nontraumatic necrosis of bone. N Engl J Med
326: 1473-1479, 1992
6. Milgram JW: Steroid induced avascular necrosis of bonc in
eighteen sites (abstract). Bull Hosp Jt Dis 37:11, 1976
7. Taylor L1: Multifocal avascular nccro
ftcr short-term high-dose
steroid therapy: a report of three cases. J Bone Joint Surg Br
66B:431-433, 1984
Ohzono K, Saito M, Takaoka K, Ono K, Saito S, Nishina T,
Kadowaki T: Natural history of nontraumatic avascular necrosis of
the femoral head. J Bone Joint Surg Br 73B:68-72, 1991
Shimizu K, Moriya H, Akita T, Sakamoto M, Suguro T: Prediction
of collapse with magnetic resonance imaging of avascular necrosis
of the femoral head. J Bone Joint Surg Am 76A:215-223, 1994
Koo K, Kim R: Quantifying the extent of osteonecrosis of the
femoral head. J Bone Joint Surg Br 77B:875-880, 1995
Glueck CJ, Freiberg RF, Glueck HI, Henderson C, Welch M,
Tracy T, Stroop D, Hamer T, Sosa F, Levy M: Hypofibrinolysis: a
common, major cause of osteonecrosis. Am J Hematol 45:156166, 1994
Jones JP Jr: Fat embolism, intravascular coagulation and ostconccrosis. Clin Orthop 292:294-308, 1993
Jones JP Jr: Concepts of etiology and early pathogenesis of
osteonecrosis. Am Acad Orthop Surg Instructional Course Lect
43:499-512, 1994
Ono K, Tohjima T, Komazawa T: Risk factors of avascular
necrosis of the femoral head in patients with SLE under high-dose
of corticosteroid therapy. Clin Orthop 277:89-97, 1992
Wang GJ, Sweet DE, Reger SI, Thompson RC: Fat-cell changes as
a mechanism of avascular necrosis of the femoral head in
cortisone-treated rabbits. J Bone Joint Surg Am 59A729-735,
Kawai K, Tamaki A, Hirohata K: Steroid-induced accumulation of
lipid in the osteocytes of the rabbit femoral head: a histological
and electron microscopic study. J Bone Joint Surg Am 67A:755763, 198.5
Fisher DE: The role of fat embolism in the etiology of corticosteroid-induced avascular necrosis. Clin Orthop 130:68-80, 1978
Ohzono K, Takaoka K, Saito S, Saito M, Matsui M, Ono K:
Intraosseous artcrial architecture in nontraumatic avascular necrosis of the femoral head. Clin Orthop 277:79-88, 1992
Saito S, Ohzono K, Ono K Early arteriopathy and postulated
pathogenesis of osteonecrosis of the femoral head. Clin Orthop
277:98-110, 1992
Bullough PG: Atlas of Orthopaedic Pathology. New York, Gower
Medical Publishing, 1992
Fisher DE, Bickel WH, Holley KE, Ellefson RD: Corticosteroidinduced aseptic necrosis. 11. Experimental study. Clin Orthop
84:200-206, 1972
Gold EW, Fox OD, Weissfeld S, Curtiss PH: Corticosteroidinduced avascular necrosis: an experimental study in rabbits. Clin
Orthop 135:272-280, 1978
23. Surat A: Isolation of prostaglandin E,-like material from osteonecrosis induced by steroids and its prevention by kallikrein inhibitor,
aprotinin. Prostaglandins Leukot Med 13:159-167, 1984
24. Wang GJ, Lennox DW, Regcr SI, Stamp WG, Hubbard SL:
Cortisone-induced intrafemoral head pressure change and its
response to a drilling decompression method. Clin Orthop 159:
274-278. 1981
25. Wang GJ, Moga DB, Richemer WG, Sweet DE, Regcr ST,
Thompson RC: Cortisone induced bone changes and its response
to lipid clearing agents. Clin Orthop 130:81-8.5, 1978
26. Yamamoto T, Hirano K, Tsutsui H, Sugioka Y, Sueishi K:
Corticosteroid enhances the expcrimcntal induction of ostconccrosis in rabbits with Shwartzman reaction. Clin Orthop 316:235243, 1995
27. Catto M: A histological study of avascular necrosis of the femoral
head after transcervical fracture. J Bone Joint Surg Br 47B:749773, 1965
28. Matsui M, Saito S, Ohzono K, Sugano N, Saito M, Takaoka K,
Ono K: Experimental steroid-induced osteonecrosis in adult
rabbits with hypersensitivity vasculitis. Clin Orthop 277:61-72,
29. Jcrgesen HE, Lang P, Mosclcy M, Gcnant HK: Histologic corrclation in magnetic rcsonance imaging of fernoral head osteonecroClin Orthop 253:lSO-163, 1990
30. Bradbury G, Benjamin J, Thompson J, Klccs E, Copeland J:
Avascular necrosis of bone after cardiac transplantation. J Bone
Joint Surg Am 76A:1385-1388, 1994
31. Bailey JM: New mechanisms for effects of anti-inflammatory
glucocorticoids. Biofactors 3:97-102, 1991
32. Sakamoto M, Akita T, Shimizu K, Tanaka Y, Gotoh K, Moriya H:
Hematologic examination of the patients with collagen disease
suffered osteonecrosis (abstract). J Jpn Orthop Assoc 66 (suppl
2):S31, 1992
33. Kisker CT, Robillard JE, Bohlken DP: Glucocorticoid stimulation
of blood coagulation factor activities in the fetal lamb. J Lab Clin
Med 101569-575, 1983
34. Latour JG, Prejean JB, Margaretten W: Corticosteroids and the
generalized Shwartzman reaction. Am J Pathol 65:189-200, 197I
35. Smith RW, Margulis RR, Brcrnan MJ, Monte RW: The influence
of ACTH and cortisone on certain factors of blood coagulation.
Science 112:295-297, 1950
36. Nakashima H: Thrombosis and lipids. In, Rccent Advances in
Thrombosis and Fibrinolysis. Edited by K Tanaka. San Diego,
Academic Prcss, 1991
Без категории
Размер файла
969 Кб
pulse, effect, marrow, induced, rabbits, methylprednisolone, corticosterone, osteonecrosis, tissue, bones
Пожаловаться на содержимое документа