(Stroke. 1997;28:1631-1638.)
© 1997 American Heart Association, Inc.
Articles |
From the Departments of Neurological Surgery (S.O., I.D., M.N., K. Onoda, K. Ogihara, T.S., S.A., T.O.) and Molecular Biology and Biochemistry (S.O., M.N., K. Onoda, T.S., Y.N.), Okayama (Japan) University Medical School, and Department of Neurological Surgery, Okayama Saiseikai General Hospital (N.Y.).
Correspondence to Shigeki Ono, MD, Department of Neurological Surgery, Okayama University Medical School, 2-5-1 Shikata-cho, Okayama 700, Japan.
| Abstract |
|---|
|
|
|---|
Methods After withdrawal of 0.1 mL cerebrospinal fluid, 0.2 mL hemolysate (n=9) or saline (n=10) was injected into the cisterna magna of male Sprague-Dawley rats weighing between 300 and 350 g. Ten minutes later, perfusion of a semipolymerized casting medium was performed at an injection pressure of 100 to 120 mm Hg. The brains were immersed and corroded in 10% NaOH solution. After these procedures, the basilar artery as well as peripheral vessels was analyzed morphologically with scanning electron microscopy. Conventional histological analysis with the use of paraffin-embedded section with hematoxylin-eosin staining was also performed, and the results were compared with those for the corrosion cast methods.
Results In the saline-injected group, SEM showed that the inner surface of the basilar artery was smooth and the form of the endothelial cell was printed on the surface of the cast. In the hemolysate-injected group, the basilar artery showed an apparent vasospasm over its entire length, and corrugation was observed on the inner surface of the basilar artery in a three-dimensional fashion. Higher magnification revealed that the nuclei of the endothelial cells were distorted. Local narrowing of the basilar artery and vasospasm in the arteries of the anterior circulation and in peripheral arteries were also observed. Measurement of the inner diameter of the basilar artery showed 37.8% contraction in the hemolysate-injected group compared with the saline-injected group by the corrosion cast method. This degree of vasospasm was similar to that observed by the conventional histological method.
Conclusions In this report we show that detailed three-dimensional observation in the rat can be performed qualitatively and quantitatively with the corrosion cast technique. We conclude that this method derives an accurate measurement of the diameter of rat major cerebral arteries and is more reliable for analyzing vasospasm in rats than angiography and other conventional procedures.
Key Words: corrosion casting subarachnoid hemorrhage vasospasm rats
| Introduction |
|---|
|
|
|---|
Although a few methods, such as angiography,6 7 8 conventional histological examination,9 16 monitoring by operating microscope,4 13 and injection of gelatin,17 have been used to evaluate the angioarchitecture or to measure the diameter of vasospastic vessels in small animals, there are some problems with these methods, namely, low resolving power in angiography, difficulty in cutting the appropriate section in conventional histological examination, and impossibility of observing these vessels from the inner surface by operating microscope. Therefore, accurate and detailed analysis of the spastic vessels in the brain of small animals has been limited.
The corrosion cast technique is very useful for three-dimensional observation of small vessels. We have reported the usefulness of this method for analyzing newly formed cerebral vessels in neural grafts,18 19 20 21 as well as for obtaining the three-dimensional angioarchitecture between host and donor tissue. However, the method has not been applied for examining vasospastic vessels in rats. The aim of the present study was to obtain views of the three-dimensional angioarchitecture from the inner surface, to evaluate vasospastic major cerebral arteries qualitatively and quantitatively, and to indicate the effectiveness of the corrosion cast technique for evaluating vasospastic vessels with the use of the acute vasocontraction model.
| Materials and Methods |
|---|
|
|
|---|
Surgical Procedures
The present study used the rat model for vasospasm developed
by Ram et al,13 with a few modifications as follows. Male
Sprague-Dawley rats weighing between 300 and 350 g were
anesthetized with pentobarbital sodium (30 mg/kg IP) and
allowed to breathe spontaneously. Under sterile macrosurgical
conditions each rat was placed in a prone position with its head fixed
on a stereotaxic frame. A midline scalp incision was made
over the parieto-occipital bone and descending to the neck. Nuchal
musculatures were partly excised and retracted laterally to expose the
surface of the occipital bone, atlanto-occipital junction, and lamina
of atlas. When the atlanto-occipital membrane was partially removed,
the cisterna magna could be observed through the dura. A 26-gauge
needle (Terumo Syringe, Terumo Corp) bent at the tip for a length of
2 mm was carefully inserted into the cisterna magna from the
lateral side of the exposed dura under direct vision. After withdrawal
of 0.1 mL cerebrospinal fluid (CSF), 0.2 mL hemolysate
(hemolysate-injected group), which was prepared as mentioned above, or
0.2 mL saline (control group) was injected into the cisterna magna.
Hemolysate or saline injection was performed slowly over 5 minutes.
After these procedures, the syringe was withdrawn, and the hole was
sealed with glue. The rat was kept in a 20° head-down position, and
its body temperature was maintained between 37°C and 39°C until the
injection was completed. During the procedure, the blood pressure and
heart rate were monitored with an automatic detector (Omniace RT3108J,
NEC Corp) via the catheter in the femoral artery, and the blood gases
were analyzed by an automatic analyzer. These
parameters were maintained within
physiological conditions. All the experimental
procedures performed in this report were in accordance with the
institutional guidelines of Okayama University Medical School.
Vascular Casting and Electron Microscopic Studies
We used the corrosion cast technique as previously reported with
some modifications.24 25 26 Ten minutes after the injection
of hemolysate or saline into the cisterna magna, the rats were deeply
anesthetized with injections of 100 mg/kg IP sodium
pentobarbital. After perfusion with 60 mL Ringer's solution via the
cannulated ascending aorta after cross-clamping of the descending
aorta, perfusion fixation was performed with the same amount of 4%
paraformaldehyde as used before casting. Five minutes
after fixation, 20 mL polyester resin (Mercox CL2B, Dainippon-ink &
Chemical, Inc) mixed with a polymerizer in the volume ratio of 50:1 was
injected with the use of a nonpulsatile technique into the ascending
aorta via a 12-mL plastic syringe. The casting medium was slowly
injected (Fig 1
) at an injection pressure
of 100 to 120 mm Hg to avoid technical artifacts, such as
change in arterial diameter,27 distortion of
vessels, or extravasation of the casting medium.28 Both
the ascending aorta and superior vena cava were clamped with forceps.
The pressure was measured by a manometer connecting with the catheter
inserted through the aorta. After a waiting period of 30 minutes at
room temperature, the decapitations were performed, and the heads were
heated in a warm water bath 60°C for 6 hours to allow complete
polymerization of the casting medium. Subsequently, they were immersed
and corroded for 36 hours in hot 10% NaOH solution and washed for 24
hours in water. The remaining bones or tissue were gently removed. The
vascular casts were freeze-cut by razor blade into pieces suitably
sized for SEM observation, then air-dried. After gold-coating the casts
with an ion sputter coater (E-1030, Hitachi), we observed them with SEM
(S-2300, Hitachi) using an accelerating voltage of 5 to 8 kV.
|
We also observed the structure of the spastic major arteries from the transverse section using a "conventional" SEM technique, and these findings were compared with those by corrosion cast. In conventional SEM technique, the spastic BA receives perfusion fixation and dehydration in graded alcohols, then is cut transversely to the longitudinal axis with a razor blade. After gold-coating the samples with an ion sputter coater (E-1030, Hitachi), we then observed them by SEM (S-2300, Hitachi) using an accelerating voltage of 5 to 15 kV.
The diameter of BA was measured at three different points: at 0.2
mm above the union of the bilateral VAs, at just below the anterior
inferior cerebellar arteries, and at 0.2 mm below the
top of the BA (Fig 2
). The mean of the
three points was taken as the diameter of the BA. All measurements were
made in a blinded manner by two examiners. The results were denoted as
mean±SD and assessed by Student's t test.
|
Measurement of the Diameter of Major Arteries by Conventional
Histology
To compare the BA diameter obtained by corrosion cast with that
obtained by conventional histology, we performed H&E staining and
observed the slices of BA under the light microscope. Ten minutes after
the injection of hemolysate or saline (0.2 mL each volume) into the
cisterna magna, perfusion fixation was performed as described above.
The brain stem, including the BA and distal segment of the VAs, was cut
out and postfixed for 24 hours in 4% paraformaldehyde.
After dehydration in ethanol, it was embedded in paraffin, cut into
4-µm sections on a microtome, and observed under a light microscope
after H&E staining.
The inner diameter of the BA was measured by a microscale at the same three points as in the corrosion cast procedure, and the mean values were used for comparison.
| Results |
|---|
|
|
|---|
|
In contrast, cerebral vasospasm was observed in all of the animals in
the hemolysate group (Fig 3B
). The image of vasospasm along the
longitudinal axis could be obtained with the use of the corrosion cast
technique, and corrugation caused a severe narrowing of the BA (Figs 3B
and 5A
). Vasospasm tended to be stronger in the region between the
posterior inferior cerebellar artery and VA union than in
the corresponding area of the BA (data not shown). Since hemolysate was
injected through the cisterna magna, we observed vasospasm not only in
the BA but also in other major cerebral arteries bilaterally, that is,
in the internal carotid artery, middle cerebral artery, anterior
cerebral artery, and posterior cerebral artery (Fig 4
). Higher magnification also showed
local vasospasm in the limited area of the BA, and the narrowest
diameter was approximately 100 to 150 µm (Fig 5A
), compared with 350 to 400 µm
in the normal BA (Fig 9
). This finding was very similar to angiographic
findings in patients who suffered from SAH.
|
|
|
In regard to the change of vessel wall in spastic arteries, the
characteristic folds were observed at regular intervals of 10 to
20 µm along the longitudinal axis (Fig 6
), and the nuclear indentations of
endothelial cells were also observed along the edge of
corrugation at the same intervals (Fig 6
, arrows), while these
characteristics were not observed in the normal BA. By this
three-dimensional analysis, we observed the folds of
endothelium caused by indentations of nuclei along the
transverse axis as well as along the longitudinal axis.
|
The vasospastic features were also observed in small arteries and
arterioles (diameter, 30 to 100 µm)29 (Fig 5B
), and
we frequently observed the shallow folds without narrowing of the inner
diameter in small arteries.
Morphological changes were not observed in the capillary network or venous circulation.
Findings by Conventional SEM Observation
The normal vessel wall of the BA was thinner than that of the
spastic BA (Fig 7A
). The ovoid humps,
representing endothelial nuclei, were lined
up at regular intervals of 10 to 20 µm (Fig 7B
, arrows). The
junctions between each endothelial cell were not
clearly detected in the normal BA.
|
Corrugation was also observed in the hemolysate-induced BA along the
longitudinal axis in conventional SEM (Fig 7C
), as was seen in
corrosion cast technique. The smooth muscle layer was relatively thick.
Humps that indicate endothelial nuclei lay in the
valley of the curved endothelium, sandwiched and
flattened between the endothelial hills along the
longitudinal axis (Fig 7D
, arrows). These findings were
consistent with those by the corrosion cast method, which in
some ways is a replication of conventional SEM.
Measurement of the Diameter of the BA
Conventional histological analysis using a
specimen stained with H&E showed that the hemolysate caused the
narrowing of BA diameter, resulting in the corrugation, and that the
vessel wall was thickened as a result of the smooth muscle contraction
(Fig 8A
) seen in conventional SEM (Fig 7C
). Such change was not observed in the normal vessels (Fig 8B
).
|
The diameter of the BA in the saline- and hemolysate-injected groups
was 369.08±26.24 µm (mean±SD) (n=9) and 214.10±19.80
µm (n=7), respectively, as observed in the H&E-stained specimen. By
SEM observation of the corrosion cast method, the diameter of the BA in
the saline- and hemolysate-injected groups was 381.70±21.19 µm
(n=12) and 237.21±37.11 µm (n=6). The difference between the
diameter of the BA in the saline-injected group and in the
hemolysate-injected group was statistically significant by both methods
(Fig 9
). There was no significant
difference between methods in the reduction rate of the diameter of the
BA (42.3% reduction by conventional histology [H&E staining]; 37.5%
by the corrosion cast method [SEM]).
| Discussion |
|---|
|
|
|---|
Several methods have been reported for the measurement of BA diameter in small animals such as rats. Delgado et al7 demonstrated the conventional angiography method by injecting contrast medium from bilateral axillary arteries. This method is advantageous in chronological observation of the cerebral vessels of living rats, but it requires high-pressure injection and risks interobserver error due to its poor resolution. Barry et al30 reported the method of measurement of the BA using a histological procedure. While the advantage of this method is that it analyzes the vessel directly and with minimum error, disadvantages include the inequality of sections, the compaction of the tissue due to long fixation, and the necessity for long fixation itself. Although in vivo video monitoring under an operating microscope has been reported,4 13 there are limitations to applying this method for analysis of vasospastic cerebral vessels in rats: this method does not measure the inner diameter of the vessels, and CSF obscures the view. For the reasons mentioned above, there have been limitations in investigating vasospasm in small animals, especially in relation to measurement of the diameter of major cerebral vessels.
Murakami25 first reported detailed three-dimensional analysis of vascular structure in rats using the corrosion cast technique. With this method, detailed analyses of rat spinal vessels and arterioles in rat cerebral cortex have also been reported.24 31 32 33 34 We have previously used the technique to investigate the vascular structure between the host and grafted tissue in rats.20 21 Based on this experience, we here applied this technique to the three-dimensional analysis of major and peripheral rat vasospastic vessels. Our goal was to establish a new method for measuring vascular diameter in small animals.
The merits of the corrosion cast technique in analyzing vasospastic major as well as peripheral vessels are as follows. (1) This method is the most accurate for measuring the diameter of the vessels. Since this method makes it possible to observe rat vessels on the order of up to 10-6 m, interobserver error will be negligible. (2) It is possible to observe different portions of the cast or to observe from different angles with the same material. Also, the observation of arteries, veins, and peripheral vessels can be conducted simultaneously. (3) Three-dimensional observation from the interior of the vessels can be performed along the long axis. Other advantages include the simplicity of the technique and the fact that it does not use radiation. These features suggest that this technique can be applied not only for analyzing vasospastic vessels but also for analyzing all the cerebral vessels and even the change of cerebral blood flow in small animals.
Despite these advantages of the corrosion cast technique, several issues should be kept in mind to ensure the homogeneous quality of the cast: (1) injection pressure: several reports27 28 mentioned that extremely high or extremely low injection pressure causes an artificial change in the diameter of the vessels. In our preliminary study, nonphysiological expansion of the vessels or extravasation of the medium was observed at an injection pressure of 200 mm Hg, and arterial collapse was observed at extremely low injection pressure; (2) cast compaction on hardening of resin: although Lametschwandtner et al27 noted that the natural compaction rate due to hardening is less than 1%, the possibility of very local compaction cannot be ruled out depending on the injection speed or temperature; (3) postmortem artificial spastic change of the major vessels: if perfusion of the casting medium is not performed immediately after heart arrest, the diameter of the vessels obtained from the corrosion cast technique does not reflect the real diameter of the vessel, and spasm-like change may even be acquired; and (4) the necessity of complete removal of tissue around the vessel after the corrosion procedure: if tissues surrounding the vessels remain, these artifacts may obscure the true depiction of the vessels. The present study took these issues into consideration, and the obtained images were satisfactory, homogeneous, and without artifact.
At higher magnification, we showed corrugation of the inner surface of
the vessel along the longitudinal axis, the characteristic folds of
endothelial cells at regular intervals, and the
indentations of endothelial nuclei at each peak of
those folds (Fig 6
). These indentations corresponded to the ovoid humps
observed in conventional SEM analysis (Fig 7D
). The width of
the folds coincided with that of endothelial cells.
These findings suggest that the mechanical force of corrugation
compressed the endothelial cells, flattened their
nuclei, and eventually disturbed their function. Ovoid humps of
endothelial nuclei not only cause the visible narrowing
of the vessels but also may disturb the local blood flow at the level
of micrometers. These mechanical disturbances of
endothelial cells and the local disorder of blood flow
may evoke the disturbance of blood coagulation and the adhesion
of white blood cells and platelets to the
endothelium, ultimately resulting in thrombus formation
and inflammatory response of the vessels. Thus, arteries with minimal
corrugation can show the characteristic change of their walls and
eventually may cause angiographic vasospasm.
With regard to the location of vasospasm, the lower BA, internal carotid bifurcation, horizontal segment of the anterior cerebral artery, and horizontal segment of the middle cerebral artery showed more severe vasospasm than other areas in the present study. This means that vasospasm is more severe in the vessels located in the major cisterns, where arteries are exposed to a greater amount of hemolysate than in the area of the cortical surface or brain parenchyma. This also means that hemolysate diffused the area of the supratentorial cisterns. Although we did not observe changes in the venous system in this study, it should be considered that there is a change in venous perfusion in addition to that in arterial blood flow in patients with SAH when increased intracranial pressure exists.
In the present study we demonstrated the reliability of the corrosion cast technique as follows: (1) conventional SEM observation proved that the corrosion cast was the mold of the inner surface of the vessels; (2) conventional histology with the use of H&E staining confirmed the diameter of vasospastic vessels; and (3) there was no statistical difference between the values of the BA diameter obtained by conventional histology and by the corrosion cast technique. Although the corrosion cast technique has been used mainly for morphological evaluation of cerebral arterioles and veins,20 21 25 26 27 28 29 31 32 33 34 35 36 the present study showed that this technique can be applied to analyze quantitative change of cerebral major arteries in SAH.
We conclude that the corrosion cast technique is very useful for analyzing vasospasm of major arteries in small animals, particularly in allowing both qualitative and quantitative three-dimensional analysis. We consider that this method is expected to be applied for the SAH model of mice or rats and to be helpful for a molecular biological approach in relation to the pathophysiology of cerebral vasospasm after SAH.
| Selected Abbreviations and Acronyms |
|---|
|
| Acknowledgments |
|---|
Received March 5, 1997; revision received April 9, 1997; accepted April 30, 1997.
| References |
|---|
|
|
|---|
| Editorial Comment |
|---|
|
|
|---|
Department of Neurology, University of Miami School of Medicine, Miami, Fla
This study used a corrosion cast technique to evaluate SAH-induced cerebral vasospasm in rats. Ten minutes after the infusion of hemolysate into the cisterna magna, the basilar artery showed apparent vasospasm and disortion of the endothelial cells. The ability to perform three-dimensional observations on the cerebral vasculature after SAH has many advantages in small-animal studies. Although the study provides no new information regarding the pathophysiology of vasospasm, the methodology allows for the quantitative assessment of the vascular response to brain injury. Thus, the corrosion cast technique would be useful in the assessment of treatment strategies directed at vascular abnormalities.
Vasospasm is delayed onset of focal or diffuse narrowing of major cerebral vessels days after SAH, and there is little evidence that acute vasospasm occurs in humans.1 In future studies, the authors should investigate longer survival periods to evaluate the more chronic vascular effects of SAH using this powerful morphological approach.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
A. Ivanov, V. Gerzanich, S. Ivanova, R. DenHaese, O. Tsymbalyuk, and J. M. Simard Adenylate cyclase 5 and KCa1.1 channel are required for EGFR up-regulation of PCNA in native contractile rat basilar artery smooth muscle J. Physiol., January 1, 2006; 570(1): 73 - 84. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Nakajima, I. Date, K. Takahashi, Y. Ninomiya, S. Asari, and T. Ohmoto Effects of Aging on Cerebral Vasospasm After Subarachnoid Hemorrhage in Rabbits Stroke, March 1, 2001; 32(3): 620 - 628. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. C. Carpenter, L. Miao, Y. Miyagi, E. Bengten, J. H. Zhang, and J. P. Muizelaar Altered Expression of P2 Receptor mRNAs in the Basilar Artery in a Rat Double Hemorrhage Model Editorial Comment Stroke, February 1, 2001; 32(2): 516 - 522. [Abstract] [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Stroke Home | Subscriptions | Archives | Feedback | Authors | Help | AHA Journals Home | Search Copyright © 1997 American Heart Association, Inc. All rights reserved. Unauthorized use prohibited. |