(Stroke. 1999;30:140-147.)
© 1999 American Heart Association, Inc.
Original Contributions |
Inhibition of Experimental Vasospasm in Rats With the Periadventitial Administration of Ibuprofen Using Controlled-Release Polymers
From the Department of Neurosurgery, Division of Vascular Neurosurgery, The Johns Hopkins University School of Medicine, Baltimore, Md.
Correspondence to Rafael J. Tamargo, MD, Department of Neurosurgery, Division of Vascular Neurosurgery, The Johns Hopkins Hospital, 600 N Wolfe St, Meyer 7-113, Baltimore, MD 21287-7713.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferencesIntroduction |
|---|
Background and Purpose—The chronic phase of vasospasm after an aneurysmal subarachnoid hemorrhage may be mediated in part by early leukocyte-endothelial cell interactions. Ibuprofen is an anti-inflammatory agent that inhibits expression of certain cell adhesion molecules and therefore disrupts leukocyte-endothelial cell interactions. Its systemic administration, however, has dose-limiting side effects. We evaluated the effect of the periadventitial delivery of ibuprofen using controlled-release polymers in the rat femoral artery model of chronic posthemorrhagic vasospasm.
Methods—Before the animal studies, the release pharmacokinetics of the ibuprofen-loaded ethylene–vinyl acetate polymers were determined in vitro. Subsequently, the femoral arteries (n=266) of Fischer 344 rats (n=133) were enclosed in latex pouches bilaterally. In the toxicity study (n=15 rats), the animals were randomized into 5 dose groups in which 0%-, 10%-, 20%-, 30%-, or 50%-loaded ibuprofen polymers were evaluated. In the efficacy study, the animals were randomized into 5 time groups in which 50%-loaded ibuprofan polymers were inserted at 0 (n=58 rats), 6 (n=16), 12 (n=13), 24 (n=11), or 48 hours (n=12) after blood injection into the pouch. The rats were killed 12 days after blood exposure, at the time of maximal vasospasm in this model. Vasospasm was expressed as percent lumen patency. To evaluate the effect of ibuprofen on leukocyte migration, 8 rats were randomized into 2 groups. Macrophages and granulocytes were stained by immunohistochemistry with the use of a mouse OX-41 monoclonal antibody and counted in the periadventitial space 24 hours after blood exposure.
Results—In vitro pharmacokinetics showed that the 50%-loaded ibuprofen polymer released its total drug load over a 12-day period. In the toxicity study, a nonsignificant arterial vasodilatation with ibuprofen treatment was seen at higher doses, and no deleterious effects were noted on the vessel wall histologically. In the efficacy study, ibuprofen treatment resulted in significant vasospasm inhibition when treatment was initiated at 0 hour (73.7±4.9% versus 94.5±3.3% [mean±SEM percent lumen patency]; P<0.001) and 6 hours (69.2±5.7% versus 98.0±3.9%; P=0.002) after blood exposure, but not at 12, 24, or 48 hours. Leukocyte immunohistochemistry showed that ibuprofen treatment resulted in significantly lower periadventitial macrophage and granulocyte counts of 25.0±3.9 cells per high-powered field compared with counts of 140.5±18.2 cells per high-powered field in the untreated vessels (P<0.001).
Conclusions—The periadventitial, controlled release of ibuprofen from surgically implanted polymers significantly inhibits chronic posthemorrhagic vasospasm in this model when treatment is initiated within 6 hours of blood exposure. Vasospasm inhibition with ibuprofen correlates with a significant decrease in the number of macrophages and granulocytes in the periadventitial space. This study supports the hypothesis that inflammation mediates in part the chronic phase of posthemorrhagic vasospasm and suggests a potential alternative treatment for this condition.
Key Words: cell adhesion molecules • ibuprofen • inflammation • polymers • vasospasm
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferencesIntroduction |
|---|
Leukocyte–endothelial cell interactions may play a role in the etiology of chronic vasospasm after an intracranial aneurysmal subarachnoid hemorrhage (SAH). Chronic vasospasm is the delayed narrowing of the cerebral vasculature after SAH that often results in ischemic injury and death after aneurysmal rupture.1 Although the etiology of chronic vasospasm remains unclear, recent work in our laboratory and by others has implicated cell adhesion molecules (CAMs) in experimental chronic vasospasm and suggested that leuko- cyte–endothelial cell interactions may be important in this process.2 3 4 Specifically, we have previously demonstrated that the increased expression of intercellular adhesion molecule-1 (ICAM-1, CD54) on endothelial cells of blood-exposed vessels is correlated with the subsequent development of chronic vasospasm3 and that chronic vasospasm can be prevented by the systemic administration of an anti–ICAM-1 monoclonal antibody (MAb).4 In addition, we have documented a correlation between the early migration of macrophages and granulocytes across the vessel wall after the hemorrhage and the subsequent development of chronic vasospasm.3 4
Ibuprofen is an anti-inflammatory agent that has been recently shown to be an inhibitor of both ICAM-1 expression and also of vascular cell adhesion molecule-1 (VCAM-1, CD106) expression.5 Ibuprofen's inhibition of leukocyte–endothelial cell binding5 6 and of ICAM-1 and VCAM-1 expression5 appears to be unrelated to its well-described cyclooxygenase inhibition, in which prostaglandin synthesis from arachidonic acid is disrupted.7 This prostaglandin-independent effect of ibuprofen on leukocyte-endothelial cell interactions is becoming increasingly recognized as a shared property of several NSAIDs.8 This observation may also explain why higher NSAID doses are needed to inhibit inflammation than to disrupt prostaglandin synthesis.8 9 At higher doses, however, the systemic side effects of NSAIDs, such as gastric erosion, thrombasthenia or thrombocytopenia, and fluid retention, may become severe.
Therefore, since posthemorrhagic chronic vasospasm is correlated with ICAM-1 expression and leukocyte migration,3 4 and since ibuprofen is a specific inhibitor of ICAM-1 expression and leukocyte–endothelial cell interaction,5 we hypothesized that ibuprofen could be used to treat chronic vasospasm. Indeed, in the canine double hemorrhage model of chronic vasospasm, the systemic administration of high doses of ibuprofen (37.5 mg/kg per day) and methylprednisolone (90 mg/kg per day) starting 1 hour after the initial hemorrhage was reported to inhibit chronic vasospasm.10 11 The systemic administration of these agents, however, resulted in serious complications such as hemorrhagic pneumonitis and duodenal ulcer perforation in some animals.10
The toxicity associated with the systemic administration of high doses of ibuprofen could be avoided by delivering the drug in high concentration only at the site of the hemorrhage with the use of controlled-release polymers. Controlled-release polymers are biologically inert matrices capable of releasing extremely high drug concentrations locally (thus minimizing systemic drug exposure and side effects) with highly reproducible release kinetic profiles.12 13 14 15 16 We have previously demonstrated that controlled-release polymers have several specific advantages for drug delivery in the brain.14 15
In this study we examine the effect of the local, periadventitial delivery of ibuprofen, with the use of controlled-release polymers surgically implanted at the site of blood deposition, on chronic posthemorrhagic vasospasm in the rat femoral artery model.3 4 17 We correlate the inhibition of chronic vasospasm with the decreased migration of macrophages and granulocytes across the vessel wall in the presence of ibuprofen. In addition, we describe an ibuprofen controlled-release polymer that releases its complete drug load over a 12-day period and document the lack of vascular toxicity associated with this form of ibuprofen administration.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferencesIntroduction |
|---|
Experimental Design
Three experiments are described in this report. In the first experiment, the pharmacokinetics of the drug-polymer formulation were evaluated by measuring the controlled release of ibuprofen from the polymer in vitro. In the second experiment, the potential toxicity of the controlled release of high doses of ibuprofen adjacent to the femoral artery was evaluated in a dose-escalation study in which the histology and lumen diameter of the drug-exposed vessels were compared. In the third experiment, the efficacy of the controlled release of ibuprofen in preventing chronic vasospasm of the femoral vessels exposed to blood was evaluated. The polymers were implanted next to the vessels at 0, 6, 12, 24, or 48 hours after the initial blood deposition. At each time point, 2 cohorts were studied: a control cohort that received empty polymers and a treatment cohort that received polymers loaded with ibuprofen. In this experiment, granulocytes and macrophages in the periadventitial space were identified immunohistochemically 24 hours after blood deposition, and their counts around ibuprofen-treated vessels were compared with those in the control group.
Polymer Preparation
Ibuprofen was incorporated into controlled-release polymers by a
previously described technique.15 16 Briefly, ibuprofen
(Sigma Chemical Co) and ethylene–vinyl acetate copolymer (EVAc;
40% vinyl acetate by weight; Dupont Co) in 5 different weight ratios
were dissolved in methylene chloride (1:9 [wt/vol] ratio of EVAc to
methylene chloride), resulting in ibuprofen-EVAc–methylene chloride
suspensions that yielded dry polymers 0%, 10%, 20%, 30%, and 50%
loaded (wt/wt) with ibuprofen. The suspension was poured into
cylindrical glass molds at -70°C. The frozen polymer cylinders were
transferred to glass plates at -30°C, and the solvent methylene
chloride was allowed to evaporate over 6 days. Residual methylene
chloride was extracted in a vacuum desiccator over another 2 days. The
polymers were cut into 5.0-mg cylindrical fragments (diameter=1
mm, height=6 mm) and exposed to ultraviolet irradiation for 1 to 2
hours before implantation to reduce the risk of infection.
Femoral Artery Isolation and Polymer Implantation
Adult male Fischer 344 rats (n=133; Harlan Sprague Dawley, Inc,
Indianapolis, Ind) weighing 200 to 250 g were used in 3 different
experiments: (1) toxicity study (n=15), (2) efficacy study (n=110), and
(3) immunohistochemistry study (n=8). The animals were
anesthetized (3 mL/kg IP) with a stock solution containing
ketamine hydrochloride (25 mg/mL), xylazine (2.5 mg/mL), and
14.25% ethyl alcohol in normal saline. Each animal received a dose of
ampicillin (25 mg/kg IP) at the time of surgery. The inguinal regions
were shaved and prepared in sterile fashion with alcohol and
povidone-iodine. With the use of microsurgical technique, the femoral
artery was dissected out and wrapped in an inert latex pouch measuring
8x8 mm, as previously described.4 EVAc polymers were
inserted into the latex pouch either at the time of blood deposition or
at later time points during reoperation. The latex pouches were sealed
with cyanoacrylate (Borden, Inc). Autologous venous blood was drawn
from the adjacent femoral vein and allowed to clot spontaneously.
Approximately 0.1 mL of clotted blood was injected into the latex
pouch. The contralateral vessel served as an internal matched control
or reference vessel in each rat. Normal saline (0.1 mL) instead of
autologous blood was injected into the reference pouch. The incisions
were closed with surgical staples. For reoperations at 6, 12, 24, and
48 hours, the animals were reanesthetized as described above,
the incisions were reopened, and the polymers were inserted into the
latex pouches through a small incision in the pouch. The wounds were
closed as described above, and the animals were allowed to recover.
Given that the implanted polymers weighed on average 5.0 mg, that they
were 50% loaded with ibuprofen, and that the smallest animals used in
the experiment weighed 200 g, then the maximum ibuprofen dose used
in this experiment was 12.5 mg/kg over the 12-day period. All
procedures were performed in accordance with guidelines established by
the Animal Care and Use Committee of The Johns Hopkins University
School of Medicine.
Vessel Histology and Cross-Sectional Area Measurements
We have previously shown that the time of peak chronic vasospasm
in the rat femoral artery model is 12 days after blood
exposure.3 Therefore, at 12 days the animals were
anesthetized as described above, and the abdominal aorta was
exposed in preparation for in situ perfusion-fixation. The aorta was
isolated and cannulated with polyethylene tubing (Intramedic PE-90,
Clay Adams). The inferior vena cava was pierced for
drainage. The lower extremities were then perfused with 40 mL of
heparinized normal saline followed by 40 mL of ice-cold freshly
depolymerized 4% paraformaldehyde in 0.1 mol/L
phosphate buffer (pH 7.4). The femoral arteries were dissected out of
the pouch and placed in 4% paraformaldehyde for 1
hour. For cryoprotection, the vessels were then placed in a solution of
20% sucrose in PBS for 24 hours at 4°C. The vessels were then frozen
in dry ice–equilibrated isopentane at -50°C and stored at -40°C
until sectioning. Vessel cross-sectional slices (14 µm) in a
transverse orientation were sectioned in a cryostat (Microm) resulting
in 10-µm slices taken at 200-µm intervals, thaw-mounted onto
Superfrost Plus–pretreated microscope slides, and allowed to air dry.
Standard staining with hematoxylin and eosin was then performed.
Lumen cross-sectional areas were calculated by computerized image
analysis on an Apple Macintosh 8100/80 AV computer using the
public domain NIH Image Program (developed at the US National
Institutes of Health and available from the Internet by anonymous file
transfer protocol from zippy.nimh.nih.gov). The areas were calculated
by measuring the actual circumference of the vessel lumen and then
calculating the area of a generalized circle
(
r2, where r=radius) based on
the calculated r value from the circumference measurement (r=measured
circumference/2), thus correcting for vessel deformation and
off-transverse sections. For each vessel, 3 separate sections at least
200 µm apart were measured and averaged. Results of the
cross-sectional area measurements were expressed as percent lumen
patency, defined as the ratio of the area of the blood-exposed or
ibuprofen-exposed vessel to the area of the contralateral
saline-exposed vessel and expressed as a percentage. According to this
protocol, each animal served as its own control, and vessel size
variability was thus taken into account. All ratios are
presented as mean±SEM. Statistical analysis was
performed with the Student's t test; P<0.05 was
considered significant.
Controlled-Release Pharmacokinetics of Ibuprofen
The release pharmacokinetics of the 50%-loaded ibuprofen-EVAc
polymers were evaluated in vitro as previously
described.16 Ibuprofen-EVAc polymers (n=3) weighing 35 mg
were suspended in individual glass vials containing 3 mL of PBS at
pH=7.4. The polymers were sequentially transplanted to new vials each
with 3 mL of fresh PBS at 1, 3, 6, 12, and 24 hours and then daily for
a total of 12 days. Ibuprofen release in the PBS aliquots was
quantified spectrophotometrically by comparison against standardized
ibuprofen concentration curves.
Controlled-Release Toxicity of Ibuprofen on the Femoral
Artery
The potential toxicity on the femoral vessels of the
periadventitial, controlled-release of ibuprofen was evaluated. With
the use of a Fibonacci dose escalation, ibuprofen-EVAc polymers (n=15),
0%, 10%, 20%, 30%, and 50% loaded with ibuprofen, were inserted in
a latex pouch encasing the femoral artery (n=15 rats, 30 vessels).
Normal saline (0.1 mL) was injected into the pouch. The contralateral
vessel was encased in a latex pouch with normal saline. Empty polymers
were inserted in the control pouch. The animals were killed, and the
vessels were harvested 12 days after surgery. The vessels were
processed as described above and examined
histologically. Vessel histology was evaluated, and
cross-sectional area ratios were calculated as described above.
Effect of Ibuprofen on Lumen Patency of Blood-Exposed
Vessels
To determine the effect of ibuprofen on lumen patency, a total
of 110 animals (220 femoral vessels) were randomized into 5 groups.
Each of these 5 groups in turn contained 2 cohorts, one in which
vessels were exposed to empty polymer (n=54 rats) and the other to
ibuprofen polymer (n=56). Ibuprofen polymers 50% loaded with the drug
were used in the treatment cohorts. After implantation of femoral latex
pouches bilaterally and subsequent injection of normal saline into the
control pouch and of autologous venous blood into the other,
controlled-release polymers were inserted in the pouches during
reoperation at the following time points after initial blood exposure:
0 (ie, at the time of blood injection, n=58 rats), 6 (n=16), 12 (n=13),
24 (n=11), and 48 hours (n=12). In the empty polymer (control) cohort,
polymers without drug were inserted into both the blood-containing
femoral pouch and the contralateral saline-containing pouch. In the
ibuprofen polymer (treatment) cohort, empty polymers were inserted in
the saline-containing femoral pouch, and 50%-loaded ibuprofen polymers
were inserted in the blood-containing femoral pouch. The animals were
killed on day 12 after initial blood deposition, previously determined
to be the time of maximum vasospasm in this model,3 4 and
the vessels were processed as described above.
Immunohistochemical Staining for Granulocytes and
Macrophages
Immunohistochemistry was used to quantitatively assess the
migration of granulocytes and macrophages into the
periadventitial space. For this purpose, rats were randomized into 2
groups, one exposed to empty polymer (n=4 rats, 8 vessels) and the
other exposed to 50%-loaded ibuprofen polymers (n=4 rats, 8 vessels).
We have previously shown that 24 hours after blood deposition is the
peak time for leukocyte migration into the periadventitial
space.3 Therefore, the animals were killed 24 hours after
blood deposition, and the vessels were harvested, sectioned, and
mounted as described above. Slides were hydrated in PBS for 5 minutes
then permeabilized in Triton X-100 (0.2% in PBS) for
15 minutes at room temperature. After they were washed in PBS for 5
minutes, slides were incubated in 3% normal horse serum for 20
minutes. After the slides were washed again in PBS for 5 minutes, the
mouse OX-41 MAb (MAS 369, Harlan Sera Laboratory, Ltd, Sussex, England)
was applied to each section and allowed to incubate for 1 hour in a
humidified chamber at 37°C. The mouse OX-41 MAb was diluted to 1:200
in PBS with 1.0% BSA and 3% normal horse serum. OX-41 detects a
membrane antigen found on rat granulocytes and the majority of
macrophages.18 After they were washed in PBS for 5
minutes, the slides were incubated with biotinylated, horse anti-mouse
IgG (rat adsorbed, BA-2001, Vector Laboratories, Burlingame, Calif)
diluted 1:100 in PBS with 1.0% BSA and 3% normal horse serum for 1
hour. After they were washed again in PBS for 5 minutes, slides were
incubated in 0.3% hydrogen peroxide in methanol for 30 minutes. Slides
were subsequently washed in PBS for 5 minutes, then incubated in avidin
and biotinylated horseradish peroxidase macromolecular complex (ABC
Elite solution; Vectastain Elite, Vector Laboratories) for 30 minutes.
After they were washed again in PBS for 5 minutes, slides were
incubated for 8 minutes with peroxidase substrate (Vector VIP, Vector
Laboratories). Slides were then washed in tap water for 5 minutes, then
counterstained with nuclear fast red for 30 seconds. Slides were
finally dry mounted.
Positively stained macrophages and granulocytes were counted by
light microscopy (Olympus BH-2, Optical Elements Corp) with the x40
objective, and the results were expressed as number of cells per
high-powered field. A representative area in the
adventitia of the vessel sections was chosen, and 3 sections
(
100 µm apart) of each vessel were counted and averaged.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferencesIntroduction |
|---|
Controlled-Release Pharmacokinetics of Ibuprofen
The release kinetics of 50%-loaded ibuprofen-EVAc polymers (n=3) were evaluated to determine the amount of drug released per unit time and the span of the release period. We found that this polymer formulation released its entire ibuprofen load in 12 days. Approximately half of the ibuprofen load was released within the first 24 hours, and the remainder was released over the next 11 days (Figure 1
).
|
Controlled-Release Toxicity of Ibuprofen on the Femoral
Artery
The controlled release of ibuprofen next to the adventitial
surface of the femoral artery resulted in no
histologically identifiable deleterious changes of the
vessel wall. Vessel cross-sectional area measurements showed
vasodilatation that peaked at the 30% loading (Figure 2). The effect was not statistically
significant.
|
Effect of Ibuprofen on Lumen Patency of Blood-Exposed
Vessels
The controlled release of ibuprofen resulted in a significant
inhibition of chronic vasospasm when treatment was initiated at 0 and 6
hours after blood exposure. When treatment was initiated 12 hours after
blood exposure, a similar trend was observed, but the difference did
not reach statistical significance (P=0.131). Microscopic
examination of the untreated blood-exposed vessels revealed the
characteristic histological changes of chronic
vasospasm in the form of corrugation of the internal elastic lamina,
thickening of the vessel wall in general and the tunica media in
particular, and intimal proliferation. At the 0-hour time point in
which the polymer was introduced at the time of blood deposition,
whereas the control empty polymer group (n=27) had a lumen patency of
73.7±4.9% (mean±SEM), the treated ibuprofen polymer group (n=31) had
a lumen patency of 94.5±3.3%. This difference was significant
(Student's t test, P<0.001) and
represents a 79% inhibition of vasospasm. At the 6-hour time
point, whereas the control empty polymer group (n=9) had a lumen
patency of 69.2±5.7%, the treated ibuprofen polymer group (n=7) had a
lumen patency of 98.0±3.9%. This difference was also significant
(P=0.002). At the 12-hour time point, however, the
difference between the 2 groups showed a similar trend but was not
statistically significant (P=0.131). At the 24- and 48-hour
time points, no difference was observed between lumen patencies in the
2 groups (Figure 3). Therefore, vasospasm
inhibition by ibuprofen was not effective if the treatment was
initiated beyond 6 hours after blood exposure.
|
Immunohistochemical Staining for Granulocytes and
Macrophages
The controlled release of ibuprofen resulted in significant
reduction of periadventitial macrophage and granulocyte
infiltration around blood-exposed vessels at 24 hours (Figure 4). Whereas the control vessels exposed
to empty polymers (n=4) had macrophage and granulocyte counts
of 140.5±18.2 cells per high-powered field (mean±SEM), the treated
vessels exposed to ibuprofen polymers (n=4) had significantly reduced
macrophage and granulocyte counts of 25.0±3.9 cells per
high-powered field (P<0.001).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferencesIntroduction |
|---|
Inflammation has long been suspected to play a role in chronic posthemorrhagic vasospasm, mostly on the basis of indirect evidence. In clinical studies, this evidence has consisted of the correlation of significantly higher systemic temperatures in patients with severe vasospasm after aneurysmal SAH,19 20 the accumulation of macrophages and granulocytes in the wall of arteries exposed to blood after SAH,21 22 23 24 25 and increased levels of immunoglobulins and complement fractions in the serum and vessel walls of patients in vasospasm.25 26 27 28 29 30 In experimental studies, this evidence has consisted of the inhibition of chronic vasospasm with the administration of methylprednisolone,10 11 31 ibuprofen,10 11 cyclosporin A,32 or complement depletion33 and the reproduction of chronic vasospasm with the introduction in the subarachnoid space of inflammatory agents such as latex beads24 or talc powder.34
Recent experimental work provides direct evidence that leukocyte–endothelial cell interactions may play a role in the etiology of chronic posthemorrhagic vasospasm. It has been shown that ICAM-1 expression is increased in blood-exposed vessels that subsequently develop chronic vasospasm,3 that endothelial ICAM-1 receptor blockade with an anti–ICAM-1 MAb prevents chronic vasospasm,4 that periadventitial migration of macrophages and granulocytes correlates with the subsequent development of chronic vasospasm,4 and that ICAM-1 expression is increased in the cerebral vasculature in ischemia35 and after SAH.2 On the basis of these studies, we have concluded that ICAM-1 may play an important role in the development of chronic posthemorrhagic vasospasm.
Although the anti-inflammatory properties of ibuprofen have been known
for a long time, only recently has it been recognized that ibuprofen
inhibits interleukin-1
– and tumor necrosis factor-–induced
expression of endothelial VCAM-1 and
ICAM-15 and thus specifically inhibits
leukocyte–endothelial cell interactions.6
This effect of ibuprofen and of other NSAIDs appears to be unrelated to
their cyclooxygenase inhibition and requires higher
drug concentrations.8 9 The decreased expression of these
endothelial CAMs disrupts the interactions between
endothelial cells and leukocyte integrins such as
lymphocyte function–associated antigen-1 (LFA-1, CD11a) and
macrophage antigen-1 (Mac-1, complement receptor 3,
CD11b).8 This prevents the irreversible adhesion of
leukocytes to the endothelial membrane, which is a
required step before diapedesis and extravasation, and thus inhibits
the inflammatory cascade.
In addition to inhibiting leukocyte–endothelial cell
interactions, ibuprofen may have a direct vasodilatory effect in the
femoral artery. We observed a trend toward vasodilation at higher
ibuprofen doses in the toxicity experiment (Figure 2
). Although
the effect of ibuprofen on vascular smooth muscle has been shown to be
variable and to depend on the species and organ system under study,
a vasodilatory effect of ibuprofen has been reported in some models and
has been attributed its cyclooxygenase
inhibition.36 37 We do not think, however, that the direct
vasodilatory effect of ibuprofen observed in this study was primarily
responsible for the inhibition of chronic vasospasm, since we saw no
effect in the groups in which treatment was initiated 24 and 48 hours
after blood deposition (Figure 3).
Controlled-release polymers such as EVAc can be implanted surgically at the site of the hemorrhage, thus delivering high doses of ibuprofen at this site but avoiding the systemic toxicity of high doses of ibuprofen. Selective surgical implantation of ibuprofen polymers in the subarachnoid space would expose the arteries in this compartment to high doses of ibuprofen but spare tissues elsewhere, thus avoiding unwanted systemic side effects. Previous studies in the canine double hemorrhage model of chronic vasospasm reported that the systemic administration of high doses of ibuprofen (37.5 mg/kg per day) starting 1 hour after the initial hemorrhage inhibited chronic vasospasm.10 11 The systemic administration of this agent, however, resulted in serious complications such as hemorrhagic pneumonitis and duodenal ulcer perforation in the animals.10 The ibuprofen-EVAc preparation (50% loading) used in this study resulted in a polymer that releases approximately half its drug load within the first 24 hours and the remainder over the next 11 days. This release profile is ideal for the treatment of chronic posthemorrhagic vasospasm in patients, since the process typically spans a 4- to 14-day period1 and since theoretically the highest ibuprofen doses are required within the first 24 hours of the hemorrhage at the time of peak ICAM-1 upregulation and leukocyte extravasation.3
Although the rat femoral artery is different from a cerebral vessel in its endothelial permeability, response to vasoactive agents, and structure of its adventitial matrix, the inflammatory mechanisms considered in this study appear to be conserved in both the peripheral and central nervous system vasculature. This model has been validated by Okada and colleagues,17 who documented that the rat femoral artery shares many morphological and physiological features of cerebral arteries. In our laboratory, we have confirmed that this vessel, in response to periadventitial blood deposition, displays acute vasospasm within the first 24 hours and then goes on to gradually develop corrugation of the internal elastic lamina, medial thickening, and endothelial proliferation that starts 4 days and peaks 10 to 12 days after blood deposition and results in significant luminal narrowing, similar to that seen angiographically in patients with chronic vasospasm of the cerebral arteries.3 We recognize, however, that the results of this study must be replicated in an intracranial model of chronic posthemorrhagic vasospasm before generalizing them to chronic vasospasm after aneurysmal SAH.
In this study, ibuprofen therapy was effective when initiated within 6 hours after blood deposition. Vasospasm inhibition was significant only when treatment was initiated at 0 and 6 hours after blood exposure (P<0.001 and P=0.002, respectively). A nonsignificant but suggestive trend was seen when treatment was initiated at 12 hours after blood exposure (P=0.131). It is likely that initiating treatment between 6 and 12 hours after blood exposure may be similarly effective. We hypothesize that since the extravasation of macrophages and granulocytes occurs primarily 3 to 24 hours after blood deposition during the period of peak ICAM-1 upregulation,3 a treatment aimed at disrupting leukocyte–endothelial cell binding but initiated after a significant portion of these leukocytes have extravasated may be ineffective. In this study, no inhibition was seen when treatment was started 24 hours after blood exposure. These results suggest that in chronic posthemorrhagic vasospasm there may be a therapeutic window of at least 6 hours for the initiation of anti-inflammatory therapy. Although this limited time window is a potential clinical obstacle to the treatment of vasospasm using ibuprofen, it is less restrictive than the 3-hour therapeutic window for thrombolytic therapy in ischemic stroke.38
A therapeutic window for the treatment of chronic vasospasm has been similarly observed in the monkey intracranial model of SAH and chronic vasospasm. In studies of clot evacuation in this model, Handa et al39 and Nosko et al40 demonstrated that removal of the clot from the subarachnoid space followed by copious irrigation with normal saline prevented chronic vasospasm only when clot evacuation was performed within 48 hours of blood deposition. In a similar study of clot fibrinolysis in this model, Findlay and colleagues41 demonstrated that intrathecal administration of recombinant tissue plasminogen activator prevented chronic vasospasm only when it was administered within 48 hours of blood deposition. The therapeutic window for initiating ibuprofen treatment in our study, however, was only 6 hours. The difference between our 6-hour therapeutic window and the 48-hour therapeutic window in these other studies could be explained as follows. We hypothesize that chronic posthemorrhagic vasospasm is primarily a result of the increased expression of endothelial CAMs of the cerebral arteries exposed to the hemorrhage. The increased expression of endothelial CAMs results in the extravasation of macrophages and granulocytes that participate in erythrocyte phagocytosis but cannot be cleared from the subarachnoid space because of the presence of the clot and impaired cerebrospinal fluid flow. These extravasated leukocytes are then primarily responsible for the subsequent events leading to chronic vasospasm because they release endothelins42 43 and lysosomal toxins44 45 during enhanced phagocytosis and after they die. Ibuprofen therapy prevents leukocyte extravasation but does not clear the leukocytes that are already present in the subarachnoid space. By contrast, clot evacuation followed by irrigation removes not only the clot but also the extravasated leukocytes. Similarly, intrathecal fibrinolysis dissolves the clot and allows for more effective clearance from the subarachnoid space by bulk flow of both erythrocytes and leukocytes. Therefore, in order to be effective, ibuprofen therapy has to be started earlier to prevent the accumulation in the subarachnoid space of a critical number of leukocytes that will then cause chronic vasospasm. By contrast, clot removal or fibrinolytic therapy can be initiated later since both will clear the subarachnoid space of extravasated leukocytes. When considered together, the results of these studies suggest that in SAH a critical number of leukocytes reaches the subarachnoid space 6 to 12 hours after the hemorrhage and that these leukocytes start to release the toxic products that eventually injure the vessel wall and lead to chronic vasospasm by 48 hours after the hemorrhage.
There are many toxic products of macrophages and granulocytes that can cause chronic vasospasm. It is apparent that nitric oxide and endothelins play a major role in chronic posthemorrhagic vasospasm.46 47 48 The balance in vascular tone exerted by endothelial cells may be altered by the exogenous endothelin production from macrophages42 and neutrophils.43 In addition, activated neutrophils may directly inhibit endothelium-dependent relaxation.49 50 Furthermore, activated macrophages during enhanced phagocytosis can secrete other toxic products that have been implicated in chronic vasospasm, such as hydrogen peroxide, superoxide anion, hydroxyl radical, perhydroxyl radical, and singlet oxygen.44 45 48
In summary, we describe in this report the inhibition of experimental chronic posthemorrhagic vasospasm with the periadventitial administration of ibuprofen using controlled-release polymers when therapy is initiated within 6 hours of blood deposition. We correlate this inhibition with a decreased concentration of extravasated macrophages and granulocytes in the periadventitial space of ibuprofen-treated vessels. In addition, we describe an ibuprofen-EVAc polymer preparation that delivers ibuprofen in a controlled fashion over a 12-day period with an initial 24-hour burst, which may be ideally suited for the treatment of chronic vasospasm in aneurysmal SAH.
|
Acknowledgments |
|---|
This work was supported in part by the Cerebrovascular Scholarship from the American Heart Association Stroke Council as well as by the Dr Harold Lamport Award from the Johns Hopkins School of Medicine (Q-A.T.) and by the Richard S. Ross Clinician Scientist Award from the Johns Hopkins University School of Medicine (R.J.T.). The authors thank Betty Tyler and Michael Pinn for technical assistance.
Received September 1, 1998; accepted October 7, 1998.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferencesIntroduction |
|---|
1. Tamargo RJ, Walter KA, Oshiro EM. Aneurysmal subarachnoid hemorrhage: prognostic features and outcomes. New Horiz. 1997;5:364–375.[Medline] [Order article via Infotrieve]
2. Handa Y, Kubota T, Kaneko M, Tsuchida A, Kobayashi H, Kawano H, Kubota T. Expression of intercellular adhesion molecule 1 (ICAM-1) on the cerebral artery following subarachnoid hemorrhage in rats. Acta Neurochir (Wien). 1995;132:92–97.[Medline] [Order article via Infotrieve]
3. Sills AK Jr, Clatterbuck RE, Thompson RC, Cohen PL, Tamargo RJ. Endothelial cell expression of intercellular adhesion molecule 1 in experimental posthemorrhagic vasospasm. Neurosurgery. 1997;41:453–461.[Medline] [Order article via Infotrieve]
4.
Oshiro EM, Hoffman PA, Dietsch GN, Watts MC, Pardoll
DM, Tamargo RJ. Inhibition of experimental vasospasm with
anti-intercellular adhesion molecule-1 monoclonal antibody in rats.
Stroke. 1997;28:2031–2038.
5. Kapiotis S, Sengoelge G, Sperr WR, Baghestanian M, Quehenberger P, Bevec D, Li SR, Menzel EJ, Muhl A, Zapolska D, Virgolini I, Valent P, Speiser W. Ibuprofen inhibits pyrogen dependent expression of VCAM-1 and ICAM-1 on human endothelial cells. Life Sci. 1996;58:2167–2181.[Medline] [Order article via Infotrieve]
6. Nielsen VG, Webster RO. Inhibition of human polymorphonuclear leukocyte functions by ibuprofen. Immunopharmacology. 1987;13:61–71.[Medline] [Order article via Infotrieve]
7. Austyn JM, Wood KM. Principles of Cellular and Molecular Immunology. Oxford, England: Oxford University Press; 1993.
8. Diaz-Gonzalez F, Sanchez-Madrid F. Inhibition of leukocyte adhesion: an alternative mechanism of action for anti-inflammatory drugs. Immunol Today. 1998;19:169–172.[Medline] [Order article via Infotrieve]
9.
Abramson S, Korchak H, Ludewig R, Edelson H, Haines K,
Levin RI, Herman R, Rider L, Kimmel S, Weissmann G. Modes of action of
aspirin-like drugs. Proc Natl Acad Sci U S A. 1985;82:7227–7231.
10. Chyatte D, Rusch N, Sundt TM. Prevention of chronic experimental cerebral vasospasm with ibuprofen and high-dose methylprednisolone. J Neurosurg. 1983;59:925–932.[Medline] [Order article via Infotrieve]
11.
Chyatte D. Prevention of chronic vasospasm in dogs with
ibuprofen and high-dose methylprednisolone. Stroke. 1989;20:1021–1026.
12. Langer R, Folkman J. Polymers for the sustained release of proteins and other macromolecules. Nature. 1976;263:797–799.[Medline] [Order article via Infotrieve]
13.
Langer R. New methods of drug delivery.
Science. 1990;249:1527–1533.
14. Tamargo RJ, Brem H. Drug delivery to the central nervous system: a review. Neurosurg Q. 1992;2:259–279.
15.
Tamargo RJ, Myseros JS, Epstein JI, Yang MB, Chasin M,
Brem H. Interstitial chemotherapy of the 9L gliosarcoma:
controlled-release polymers for drug delivery in the brain.
Cancer Res. 1993;53:329–333.
16. Tamargo RJ, Langer R, Brem H. Interstitial drug delivery to the central nervous system using controlled-release polymers: chemotherapy for brain tumors. Methods Neurosci. 1994;21:135–149.
17. Okada T, Harada T, Bark DH, Mayberg MR. A rat femoral artery model of vasospasm. Neurosurgery. 1990;27:349–356.[Medline] [Order article via Infotrieve]
18. Robinson AP, White TM, Mason DW. Macrophage heterogeneity in the rat as delineated by two monoclonal antibodies MRC OX-41 and MRC OX-42, the latter recognizing complement receptor type 3. Immunology. 1986;57:239–247.[Medline] [Order article via Infotrieve]
19. Rousseaux P, Scherpereel B, Bernard MH, Graftieaux JP, Guyot JF. Fever and cerebral vasospasm in ruptured intracranial aneurysms. Surg Neurol. 1980;14:459–465.[Medline] [Order article via Infotrieve]
20. Weir B, Disney L, Grace M, Roberts P. Daily trends in white blood cell count and temperature after subarachnoid hemorrhage from aneurysm. Neurosurgery. 1989;25:161–165.[Medline] [Order article via Infotrieve]
21.
Crompton MR. The pathogenesis of cerebral infarction
following the rupture of cerebral berry aneurysms.
Brain. 1964;87:491–510.
22. Hughes JT, Schianchi PM. Cerebral artery spasm: a histological study at necropsy of the blood vessels in cases of subarachnoid hemorrhage. J Neurosurg. 1978;48:515–525.[Medline] [Order article via Infotrieve]
23. Tanabe Y, Sakata K, Yamada Y. Cerebral vasospasm and ultrastructural changes in cerebral arterial wall. J Neurosurg. 1978;49:229–238.[Medline] [Order article via Infotrieve]
24. Peterson JW, Kwun BD, Hackett JD, Zervas NT. The role of inflammation in experimental vasospasm. J Neurosurg. 1990;72:767–774.[Medline] [Order article via Infotrieve]
25. Handa Y, Kabuto M, Kobayashi H, Kawano H, Takeuchi H, Hayashi M. The correlation between immunological reaction in the arterial wall and the time course of the development of cerebral vasospasm in a primate model. Neurosurgery. 1991;28:542–549.[Medline] [Order article via Infotrieve]
26. Pellettieri L, Carlson CA, Lindholm L. Is the vasospasm following subarachnoid hemorrhage an immunoreactive disease? Experientia. 1981;37:1170–1171.[Medline] [Order article via Infotrieve]
27. Hoshi T, Shimizu T, Kito K, Yamasaki N, Takahashi K, Takahashi M, Okada T, Kasuya H, Kitamura K. Immunological study of late cerebral vasospasm in subarachnoid hemorrhage: detection of immunoglobulins, C3, and fibrinogen in cerebral arterial walls by immunofluorescence method [in Japanese]. Neurol Med Chir (Tokyo). 1984;24:647–654.[Medline] [Order article via Infotrieve]
28. Pellettieri L, Nilsson B, Carlsson CA, Nilsson U. Serum immuno-complexes in patients with subarachnoid hemorrhage. Neurosurgery. 1986;19:767–771.[Medline] [Order article via Infotrieve]
29. Ostergaard JR, Kristensen BO, Svehag SE, Teisner B, Miletic T. Immune complexes and complement activation following rupture of intracranial saccular aneurysms. J Neurosurg. 1987;66:891–897.[Medline] [Order article via Infotrieve]
30. Kasuya H, Shimizu T. Activated complement components C3a and C4a in cerebrospinal fluid and plasma following subarachnoid hemorrhage. J Neurosurg. 1989;71:741–746.[Medline] [Order article via Infotrieve]
31. Yamakawa K, Sasaki T, Tsubaki S, Nakagomi T, Saito I, Takakura K. Effect of high-dose methylprednisolone on vasospasm after subarachnoid hemorrhage. Neurol Med Chir (Tokyo). 1991;31:24–31.[Medline] [Order article via Infotrieve]
32.
Peterson JW, Nishizawa S, Hackett JD, Bun T, Teramura
A, Zervas NT. Cyclosporine A reduces cerebral vasospasm
after subarachnoid hemorrhage in dogs.
Stroke. 1990;21:133–137.
33. German JW, Gross CE, Giclas P, Watral W, Bednar MM. Systemic complement depletion inhibits experimental cerebral vasospasm. Neurosurgery. 1996;39:141–146.[Medline] [Order article via Infotrieve]
34. Nagata K, Sasaki T, Mori T, Nikaido H, Kobayashi E, Kim P, Kirino T. Cisternal talc injection in dog can produce delayed and prolonged arterial constriction resembling cerebral vasospasm morphologically and pharmacologically. Surg Neurol. 1996;45:442–447.[Medline] [Order article via Infotrieve]
35. Okada Y, Copeland BR, Mori E, Tung MM, Thomas WS, del Zoppo GJ. P-selectin and intercellular adhesion molecule-1 expression after focal brain ischemia and reperfusion. Stroke. 1994;25:202–211.[Abstract]
36.
Apstein CS, Vogel WM. Coronary
arterial vasodilator effect of ibuprofen. J
Pharmacol Exp Ther. 1982;220:167–171.
37. Askholt J, Nielsen-Kudsk R. Ibuprofen, pharmacokinetics and pharmacodynamics in the isolated rabbit heart. Acta Pharmacol Toxicol. 1985;56:99–107.[Medline] [Order article via Infotrieve]
38.
The National Institute of Neurological Disorders and
Stroke rt-PA Stroke Study Group. Tissue plasminogen
activator for acute ischemic stroke. N
Engl J Med. 1995;333:1581–1587.
39. Handa Y, Weir BKA, Nosko M, Mosevich R, Tsuji T, Grace M. The effect of timing of clot removal on chronic vasospasm in a primate model. J Neurosurg. 1987;67:558–564.[Medline] [Order article via Infotrieve]
40. Nosko M, Weir BKA, Lunt A, Grace M, Allen P, Mielke B. Effect of clot removal at 24 hours on chronic vasospasm after SAH in the primate model. J Neurosurg. 1987;66:416–422.[Medline] [Order article via Infotrieve]
41. Findlay JM, Weir BKA, Kanamaru K, Grace M, Baughman R. The effect of timing of intrathecal fibrinolytic therapy on cerebral vasospasm in a primate model of subarachnoid hemorrhage. Neurosurgery. 1990;26:201–206.[Medline] [Order article via Infotrieve]
42.
Ehrenreich H, Anderson RW, Fox CH, Rieckmann P, Hoffman
GS, Travis WD, Coligan JE, Kehrl JH, Fauci AS. Endothelins, peptides
with potent vasoactive properties, are produced by human
macrophages. J Exp Med. 1990;172:1741–1748.
43. Sessa WC, Kaw S, Hecker M, Vane JR. The biosynthesis of endothelin-1 by human polymorphonuclear leukocytes. Biochem Biophys Res Commun. 1991;174:613–618.[Medline] [Order article via Infotrieve]
44. Janeway CA, Travers P. Immunobiology: The Immune System in Health and Disease. New York, NY: Current Biology Ltd/Garland Publishing, Inc; 1996.
45. Nathan CF. Secretory products of macrophages. J Clin Invest. 1987;79:319–326.
46. Faraci FM. Endothelium-derived vasoactive factors and regulation of the cerebral circulation. Neurosurgery. 1993;33:648–659.[Medline] [Order article via Infotrieve]
47. Hino T, Tokuyama Y, Weir B, Takeda J, Yano H, Bell GI, Macdonald RL. Changes in endothelial nitric oxide synthase m RNA during vasospasm after subarachnoid hemorrhage in monkeys. Neurosurgery. 1996;39:562–568.[Medline] [Order article via Infotrieve]
48. Zabramski JM. Vasospasm After Subarachnoid Hemorrhage. Park Ridge, Ill: American Association of Neurological Surgeons; 1997.
49. Akopov SE, Grigorian MR, Gabrielian ES. The endothelium-dependent relaxation of human middle cerebral arteries: effects of activated neutrophils. Experientia. 1992;48:34–36.[Medline] [Order article via Infotrieve]
50. Csaki C, Szabo C, Benyo Z, Reivich M, Kovach AGB. Activated neutrophils inhibit cerebrovascular endothelium-dependent relaxations in vitro. Life Sci. 1991;49:1087–1094.[Medline] [Order article via Infotrieve]
Editorial Comment
Department of Neurosurgery, University of California at Davis, Sacramento, California
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferencesIntroduction |
|---|
Controlled-release polymers will undoubtedly become the "darlings" of physicians treating a wide variety of central nervous system afflictions, and the present work is the first suggesting they may one day have a use in the prevention of cerebral aneurysmal vasospasm. I cannot help but congratulate the authors on a beautiful piece of all-encompassing work: pharmacokinetic studies, dose-response toxicity studies, efficacy studies, and explanatory studies, all compressed into one, easy-to-read paper! I only wish that similar studies had been done with the carmustine-containing polymers for the treatment of brain tumors, so that we would not now have the low-dose (<4%) carmustine-containing polymers when higher doses might have been more effective. As the authors point out, intracranial experimental studies with this technique need to be done before it can be applied to humans, and one can only hope that before embarking on time- and resources-consuming human studies, the best drug and dose for maximal clinical safety and effect can be chosen with confidence. The present work demonstrates ways to reach such decisions.
Received September 1, 1998; accepted October 7, 1998.
This article has been cited by other articles:
 |
|
 |
|
  K. L. Chaichana, A. P. Levy, R. Miller-Lotan, S. Shakur, and R. J. Tamargo Haptoglobin 2-2 Genotype Determines Chronic Vasospasm After Experimental Subarachnoid Hemorrhage Stroke, December 1, 2007; 38(12): 3266 - 3271. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. O. Borel, A. McKee, A. Parra, M. M. Haglund, A. Solan, V. Prabhakar, H. Sheng, D. S. Warner, L. Niklason, and A. Bhardwaj Possible Role for Vascular Cell Proliferation in Cerebral Vasospasm After Subarachnoid Hemorrhage * Editorial Comment Stroke, February 1, 2003; 34(2): 427 - 433. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Nakai, Y. Numaguchi, T. H. Foster, K. Shima, and M. Kikuchi Endovascular Treatment Using Low-Power Ultraviolet Laser for Delayed Vasospasm in the Rabbit Carotid Artery Model AJNR Am. J. Neuroradiol., November 1, 2002; 23(10): 1725 - 1731. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Gabikian, R. E. Clatterbuck, C. G. Eberhart, B. M. Tyler, T. S. Tierney, and R. J. Tamargo Prevention of Experimental Cerebral Vasospasm by Intracranial Delivery of a Nitric Oxide Donor From a Controlled-Release Polymer: Toxicity and Efficacy Studies in Rabbits and Rats Stroke, November 1, 2002; 33(11): 2681 - 2686. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C.J.M. Frijns and L.J. Kappelle Inflammatory Cell Adhesion Molecules in Ischemic Cerebrovascular Disease Stroke, August 1, 2002; 33(8): 2115 - 2122. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. Noel, J. E. Conway, and R. J. Tamargo Vascular Complications of Cocaine Use * Response Stroke, July 1, 2002; 33(7): 1747 - 1748. [Full Text] [PDF]
|
|
|
|
|
|
M. Satoh, A. D. Parent, and J. H. Zhang Inhibitory Effect With Antisense Mitogen-Activated Protein Kinase Oligodeoxynucleotide Against Cerebral Vasospasm in Rats Stroke, March 1, 2002; 33(3): 775 - 781. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Satoh, I. Date, M. Nakajima, K. Takahashi, K. Iseda, T. Tamiya, T. Ohmoto, Y. Ninomiya, S. Asari, and R. L. Macdonald Inhibition of Poly(ADP-Ribose) Polymerase Attenuates Cerebral Vasospasm After Subarachnoid Hemorrhage in Rabbits Editorial Comment Stroke, January 1, 2001; 32(1): 225 - 231. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||