(Stroke. 1995;26:2347-2352.)
© 1995 American Heart Association, Inc.
Articles |
From the Department of Neurosurgery, Klinikum Mannheim, Ruprecht-Karls-Universität Heidelberg, Mannheim, Germany.
Correspondence to Dr Axel Piepgras, Neurochirurgische Klinik, Klinikum Mannheim/Ruprecht-Karls-Universität Heidelberg, Theodor-Kutzer-Ufer 1-3, 68135 Mannheim, Germany. E-mail fa8@ix.urz.uni-heidelberg.de.
| Abstract |
|---|
|
|
|---|
Methods Subarachnoid hemorrhage was induced by transorbital injection of 0.3 mL of nonheparinized autologous arterial blood into the chiasmatic cistern. Changes in regional cerebral blood flow were continuously recorded with the use of laser-Doppler flowmetry over the parietal cortex. Angiographic verification of middle cerebral artery diameter was performed by carotid catheterization at baseline and 2 days after injection of blood or artificial cerebrospinal fluid. We monitored intracranial and systemic blood pressure during and after injections.
Results Injection of artificial cerebrospinal fluid in the control group did not change the diameter of the middle cerebral artery. Injection of blood caused a significant arterial narrowing of 17.5%, from 0.37±0.04 mm to 0.31±0.04 mm after 2 days (P=.0001). In the control group regional cerebral blood flow decreased to 75.9±16.8% of preinjection control but quickly recovered to 99.7±19.4%. Intracranial pressure increased for 5 minutes after the injection to a maximum of 27.3±8.9 mm Hg, accompanied by a 10% decrease in mean arterial pressure. A fall in cerebral blood flow to 53.1±26.3% in blood-injected animals that recovered to only 80.7±16.9% of baseline values during the observation period of 30 minutes was noted. A peak intracranial pressure of 45.7±11.5 mm Hg occurred 2 minutes after injection with a decrease in mean arterial pressure of 13%, resulting in a markedly lower cerebral perfusion pressure than in the control group.
Conclusions An angiographically controlled model of subarachnoid hemorrhage primarily involving the anterior circulation is feasible in the rat. The resulting narrowing of the middle cerebral artery reflects moderate vasospasm and will allow further microcirculatory studies with cranial windows.
Key Words: angiography cerebral blood flow intracranial pressure subarachnoid hemorrhage rats vasospasm
| Introduction |
|---|
|
|
|---|
The subsequent changes in the macrocirculation after SAH have been known since the original description by Ecker and Riemenschneider in 1951.1 The varying degree of constriction of major cerebral arteries is often used as an explanation for the ischemic syndrome of cerebral vasospasm. The role of the cerebral microcirculation in either promoting or counteracting the constriction of major cerebral arteries after SAH, however, remains unclear. Several attempts, limited to histopathological examinations,2 3 have been made thus far. The detection of cerebral ischemia after SAH requires continuous measurements of rCBF. In vivo, continuous rCBF measurements may be accomplished by LDFM, primarily reflecting changes in the microcirculation.
The rat is a frequently used animal to investigate the physiology and pathophysiology of the cerebral circulation, including SAH. However, angiographic studies of the intracranial circulation are rare. The majority of studies have been performed for determination of diameter changes in the posterior circulation, ie, the basilar artery.4 Although a technique for carotid angiography in the rat has been described,5 thus far it has not been used for a controlled study.
The aim of this study was to prove the feasibility of an angiographically controlled SAH model in the rat that allows for microcirculatory studies. In a first step, changes in the cortical microcirculation were measured by LDFM in the acute stage after SAH. These measurements were performed through a modified cranial window technique in the distribution area of the MCA. The effect of SAH on MCA caliber was studied by repetitive carotid angiography.
| Materials and Methods |
|---|
|
|
|---|
Carotid Angiography
The rats were placed in a supine position, the ventral portion
of the neck was shaved, and under local anesthesia
(lidocaine 1% SC) a 1.5-cm paramedian skin incision below the hyoid
bone was made. With the use of an operating microscope, the cervical
musculature and thyroid gland were retracted. Blunt dissection of the
carotid bifurcation was followed by exposure of the ECA up to its third
branch, the ascending pharyngeal artery (Fig 1
7 ). The ECA was ligated distal to the
origin of the ascending pharyngeal artery and cannulated with a blunted
radiopaque polyethylene tube (PE 50). The catheter tip was placed at
the carotid bifurcation. The skin incision was sutured and the catheter
rinsed continuously with saline solution (NaCl 0.9%) at a flow rate of
1 mL/h. The animals were then placed supine on a Plexiglas plate for
baseline angiography. Iopromide (Ultravist 370, Byk Gulden) was used as
a nonionic contrast agent.8 We manually injected 0.5 mL
over 1 second for a retrograde carotid artery angiogram, using a
standard mammographic unit (Siemens) with a 0.1x0.1-mm focus spot.
Focus-objective and focus-film distances were standardized to
obtain a linear twofold magnification.
|
After baseline angiography and induction of SAH or injection of artificial CSF in the control group, the ECA catheter was left in place with a neck incision and a flexible metal spring. The catheter was constantly perfused with saline solution (NaCl 0.9%, 0.5 mL/h) for the following days.
Follow-up angiography was performed on day 2 after injection of
subarachnoid material. A good and reproducible filling of the
CCA, ICA, MCA, and SA was required (Fig 2
) for morphometric angiographic
analysis.
|
ICP Measurements
Under local anesthesia (lidocaine 1% SC) the
posterior cervical muscles were separated in the midline to allow for
blunt exposure of the atlanto-occipital membrane. A 27-gauge,
saline-filled needle was passed stereotaxically into
the cisterna magna under microscopic vision and connected to a pressure
transducer (Statham P50). LDFM measurements and ICP were recorded
for 30 minutes after SAH, after which the needle was withdrawn. In some
cases a small leakage of CSF was noticed and controlled with an
absorbable gelatin sponge.
Laser-Doppler Flowmetry
Changes in rCBF were continuously recorded with the use of
LDFM.9 10 11 A midline scalp incision of approximately 2.5 cm
in length was made over the parieto-occipital bone descending to
the neck. A rectangular trepanation, 0.5 cm in length, was carried out
over the parietal region, above the vascular territory of the MCA. The
inner table of the parietal bone was left intact and continuously
rinsed with saline solution (NaCl 0.9%) at body temperature for a
clear optical medium. We mounted a laser-Doppler flow probe
(Perimed) on a micromanipulator (Kopf Instruments) and positioned it
0.5 mm above the surface, carefully avoiding large dural or pial
vessels. Once a suitable placement was obtained, the probe was kept in
place throughout the measurement, and mean rCBF values were
recorded every 10 seconds. The flow probe we used has a sample
volume of approximately 1 mm3.11 rCBF values
measured by LDFM were expressed as a percentage of each animal's
control values, measured before injection of subarachnoid
material. After the measurements, 30 minutes after injection, the bone
defect was sealed with wax and the wound closed.
Induction of SAH
After the first angiography and measurement of ICP, the animals
were fixed in a prone position with the use of a
stereotaxic head frame (Kopf Instruments).
Arterial blood pressure was monitored with a Statham P50
pressure transducer and was recorded on-line on a personal
computer (Apple Macintosh II cx, Apple Computers) with an
eight-channel data-acquisition board (National
Instruments).
In 28 animals (SAH group), the subarachnoid space in the perichiasmatic cistern was punctured with a 27-gauge needle via the orbit and the optic foramen. For induction of SAH, 0.3 mL autologous blood was drawn from the ECA catheter and injected manually over a period of 1 minute.
In 13 animals (control group), the same operation as in the SAH group was performed. Instead of blood, however, 0.3 mL artificial CSF was injected. The artificial CSF had the following composition (in mmol/L): Na+ 153, K+ 3, Ca2+ 1.5, Mg2+ 0.6, Cl- 140, glucose 3.7, urea 6, and HCO3- 25. Osmolarity was 315 mOsm/L. This fluid was equilibrated with 6.6% O2/5.9% CO2 and the balance N2, resulting in a pH of 7.39 at 37°C.
Image Analysis
The angiographic films were scanned (Agfa Arcus, Agfa) into a
personal computer (Apple Macintosh, s.a.) and morphometrically
analyzed with the National Institutes of Health
IMAGE software package (NIH IMAGE 1.50 by W.
Rasband).
Mean intraluminal diameters were measured by integration over the M1 segment and the catheter tip at the carotid bifurcation. To avoid artifacts related to different object-film distances at the two time points of angiography, the relative intraluminal diameter of the M1 segment to the intraluminal diameter of the catheter tip was calculated. Changes in diameter are expressed as percent changes from baseline to control angiography.
Statistical Analysis
All values are expressed as mean±SD. Statistical
analysis was performed with the use of Student's
two-sample t test for paired data for intragroup
differences and for unpaired data for intergroup differences. A value
of P
.05 was accepted as statistically significant.
| Results |
|---|
|
|
|---|
Macroscopically, 2 days after injection of blood through the optic foramen a consistent bilateral clot on the ventral surface of the brain was found at necropsy. No hemorrhage was seen in the CSF-injected rats, and no epidural hematoma occurred in either group. Gross macroscopic inspection revealed no signs of damage to the cortex or brain stem related to the various measurement techniques.
Follow-up angiograms of the anterior cerebral circulation were performed in all rats. Comparison of the MCA diameter in baseline and follow-up angiograms was feasible in 16 of 28 animals in the SAH group and 10 of 13 in the control group. In one case in which a thrombosis of the ICA was noted, follow-up angiography was nevertheless possible after thrombectomy.
The angiographically determined intraluminal diameter of the MCA at
baseline was 0.35±0.04 mm (range, 0.29 to 0.43 mm) in the control
group and 0.37±0.04 mm (range, 0.31 to 0.47 mm) in the SAH group
(P=NS). Two days after injection of artificial CSF, the MCA
diameter was 0.35±0.03 mm (range, 0.30 to 0.39 mm; P=NS
versus control diameter). The perichiasmatic injection of blood caused
a significant constriction in the MCA at day 2: intraluminal diameter
decreased to 0.31±0.04 mm (range, 0.23 to 0.39 mm; P=.0001;
Fig 3
).
|
Fig 4
represents the changes in
physiological parameters. In the
CSF-injected control rats, a small increase in ICP to 27.3±8.9 mm Hg
occurred, lasting for 5 minutes after the procedure. Blood pressure
decreased by 10% and cerebral perfusion pressure decreased to 46 mm Hg
for 2 minutes after the injection. Changes in rCBF as measured by LDFM
were mild, and there was no longer-lasting decrease in flow: after
an initial drop to 75.9±16.8% for 2 minutes, flow quickly recovered
to 99.7±19.4% 30 minutes after injection. In the blood-injected
rats, during the injection ICP rose to a peak of 45.7±11.5 mm Hg, then
sloped down exponentially to a level of 11.4±4.4 mm Hg 30 minutes
later. At the peak of ICP increase, rCBF decreased to 53.1±26.3% of
baseline flow. The short-lasting recovery period was followed by a
lasting mild decrease to 80.7±16.9% of baseline levels at the end of
the observation period of 30 minutes. Mean arterial blood
pressure decreased by 13%, and the minimum cerebral perfusion pressure
was 42 mm Hg 2 minutes after injection.
|
| Discussion |
|---|
|
|
|---|
The advantages of this model include low mortality, lack of a complicated intracranial surgical procedure, injection of a standardizable amount of fresh blood into the subarachnoid space, and a route for easy follow-up angiograms. The injection of artificial CSF instead of blood allows for control of normal vascular reactions. Furthermore, as shown by the application of LDFM, microcirculatory studies with a cranial window technique are possible.
The major disadvantage of this model is the blind induction of the SAH that includes the possibility of parenchymal damage. As in this series, this can easily be avoided by training.
The rat serves as a small-animal model for a variety of intracranial diseases, including SAH. The most striking difference from the human clinical setting appears to be the time course in changes of major vessel diameter. In humans, maximum angiographic vasospasm is known to occur between days 6 to 8 after the bleed,12 whereas in the rat the time course appears to be shorter, with maximal severity at day 2.4 However, development of proliferative angiopathy at day 713 suggests delayed pathophysiological changes in the rat vessel walls similar to those occurring in humans.
Several rat models for the study of SAH have been proposed for the anterior14 15 and posterior4 16 17 18 19 20 circulation. Models that use puncturing of a major cerebral artery17 21 probably most closely reflect the clinical setting, including a vascular injury resembling rupture of an aneurysm with subsequent release of potentially spasmogenic substances in the subarachnoid space. The major disadvantages are difficulties in standardizing the release of blood, the lack of a control group, and a high mortality rate.22 In our experience, puncturing the MCA at an accessible site did not cause formation of a major clot in the basal cisterns (A.P. et al, unpublished data, 1992).
Other models that are based on the injection of blood, leaving the vessel wall intact, provide several advantages, such as good reproducibility of blood placement and acceptable mortality rates. Lacy and Earle14 introduced a silicone elastomer tube through a frontal burr hole to the skull base for injection of heparinized blood. Problems arose from catheter placement that may easily be advanced in the brain parenchyma. A similar model mainly produces a cortical SAH.15 Because of these technical drawbacks, injection of whole blood or blood components after direct puncture is widely used for simulating SAH. Most investigators have injected blood into the cisterna magna. This is a straightforward, standardizable, and reproducible procedure. Various amounts of blood have been used, ranging from 0.07 mL4 to 0.7 mL,23 with 0.3 mL producing the best results with respect to formation of a clot, acceptable mortality, and avoidance of global ischemia. To date, no experiences with multiple injections have been reported for the rat.
In rats, angiographic verification of changes in basilar artery diameter was performed by Delgado et al in 1985.4 They report a biphasic pattern of constriction with an acute spasm of 40% 10 minutes after SAH and a maximum delayed spasm by day 2 of 20%, with return to baseline diameter by day 5 after SAH. These findings were confirmed by Verlooy et al in 1992.20 D'Avella et al19 used the angiographic technique initially reported by Boullin et al in 19815 to perform carotid angiography with blood injections into the cisterna magna 48 hours after SAH. These authors ligated the CCA, a procedure that alters hemispheric CBF24 and normal physiological responses of the ipsilateral cerebral vessels. Repetitive studies were not performed; however, a decrease to 79% of control diameter in the MCA was found 2 days after SAH. We found a comparable mean decrease in MCA diameter of 17.5% in the same animal using retrograde carotid filling via the ECA. The amount of contrast medium injected (0.5 mL compared with 0.7 mL used for vertebrobasilar angiography)4 20 had no adverse effects. However, matching of control and follow-up angiograms could not be performed in 15 of 41 rats because of incomplete filling of the CCA, ICA, MCA, and SA in one of the paired sets. This was due to the fact that the x-ray machine allowed only one exposure during injection of contrast medium.
Different degrees of arterial constriction are used to define angiographic vasospasm. In this series a mean reduction in MCA diameter of 17.5% was found, comparable to other reports.4 19 20 However, the clinically relevant syndrome of delayed cerebral ischemia due to vasospasm may extend beyond mere constriction of large cerebral arteries. Clinical25 26 27 28 and experimental29 data suggest a decreased cerebral perfusion concomitant with impaired autoregulation,30 31 possibly caused by an increase of vascular resistance, impaired capillary perfusion,2 32 and brain edema.15 Measurements of CBF in the rat after induction of SAH have yielded inconsistent results, but there is agreement for an acute decrease in CBF (up to 3 hours after SAH) to approximately 50% of control values.18 33 34 Some authors have described a chronic CBF depression for up to several days,21 35 but this was not confirmed by others.30 33 34
LDFM measures CBF changes in a small sample volume of approximately 1 mm3.11 The sample volume in this setup includes dural, pial, and mainly superficial cortical penetrating vessels. The early changes we noted with LDFM (a decrease on the order of 50%) are most likely part of a Cushing's response to a sudden rise in intracranial pressure.36 37 The differences between the SAH and the control groups are probably based on different fluid viscosity of the injected material and their respective effects on CSF outflow resistance. Butler et al38 and Jackowski et al34 have previously described similar phenomena. However, it is currently impossible to quantify CBF changes over a long period of time, and therefore we could not prove a delayed reduction of CBF in this model. Our results indicate a mild vasoconstriction in a large cerebral artery (MCA) and a moderate decrease in rCBF after experimental SAH. These effects may support the findings of an impaired capillary perfusion after SAH that Johshita et al2 described in rabbits.
We believe that the described model is useful to study early pathophysiological derangements after a subarachnoid bleed that might trigger secondary ischemia rather than chronic vasospasm. Our data support the advantage of this model for further microcirculatory studies.
| Selected Abbreviations and Acronyms |
|---|
|
| Acknowledgments |
|---|
Received May 25, 1995; revision received August 9, 1995; accepted August 30, 1995.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
G. W. Britz, J. R. Meno, I.-S. Park, T. J. Abel, A. Chowdhary, T.-S. K. Nguyen, H. R. Winn, and A. C. Ngai Time-Dependent Alterations in Functional and Pharmacological Arteriolar Reactivity After Subarachnoid Hemorrhage Stroke, April 1, 2007; 38(4): 1329 - 1335. [Abstract] [Full Text] [PDF] |
||||
![]() |
K. Kidoguchi, M. Tamaki, T. Mizobe, J. Koyama, T. Kondoh, E. Kohmura, T. Sakurai, K. Yokono, and K. Umetani In Vivo X-Ray Angiography in the Mouse Brain Using Synchrotron Radiation Stroke, July 1, 2006; 37(7): 1856 - 1861. [Abstract] [Full Text] [PDF] |
||||
![]() |
H. Kamii, I. Kato, H. Kinouchi, P. H. Chan, C. J. Epstein, A. Akabane, H. Okamoto, T. Yoshimoto, and C. Y. Hsu Amelioration of Vasospasm After Subarachnoid Hemorrhage in Transgenic Mice Overexpressing CuZn–Superoxide Dismutase • Editorial Comment Stroke, April 1, 1999; 30(4): 867 - 872. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. Ono, I. Date, M. Nakajima, K. Onoda, K. Ogihara, T. Shiota, S. Asari, Y. Ninomiya, N. Yabuno, T. Ohmoto, et al. Three-Dimensional Analysis of Vasospastic Major Cerebral Arteries in Rats With the Corrosion Cast Technique Stroke, August 1, 1997; 28(8): 1631 - 1638. [Abstract] [Full Text] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Stroke Home | Subscriptions | Archives | Feedback | Authors | Help | AHA Journals Home | Search Copyright © 1995 American Heart Association, Inc. All rights reserved. Unauthorized use prohibited. |