In Vivo Angioplasty Prevents the Development of Vasospasm in Canine Carotid Arteries
Pharmacological and Morphological Analyses
Background and Purpose To study the effects of in vivo transluminal balloon angioplasty (TBA) on the structure and function of the arterial wall, a canine model of hemorrhagic cerebral vasospasm of the high cervical internal carotid artery (ICA) was used. This model was also used to determine whether TBA performed before clot placement could prevent the development of vasospasm.
Methods Twelve dogs underwent surgical exposure of both distal cervical ICAs, followed by baseline angiography. One randomly selected ICA in each dog was then subjected to in vivo TBA and repeated angiography. Both distal ICAs were then surrounded with blood clots held by silicone elastomer sheaths. Seven days later angiography was repeated, and all animals were killed. The ICAs in four animals were perfusion-fixed in situ for morphological analysis by electron microscopy, and the arteries in the remaining eight animals were removed and immediately immersed in oxygenated Krebs’ solution. Contractile responses of isolated arterial rings from each ICA were recorded after treatment with KCl, noradrenaline, serotonin, and prostaglandin F2α, while relaxations in response to the calcium ionophore A23187 and papaverine were recorded after tonic contraction to noradrenaline had been established. The morphology and pharmacological responses of ICAs that had been exposed to blood with or without prior TBA were compared with data obtained from control arterial segments of intact, more proximal regions of the ICAs from each animal.
Results TBA resulted in immediate angiographic enlargement of the ICA lumen that was still evident 7 days later despite the placement of clotted blood around the artery. Scanning and transmission electron microscopy demonstrated flattening of the intima and internal elastic lamina in these dilated arteries, associated with patchy losses of endothelial cells. In contrast, ICAs that had been exposed to clotted blood but had not undergone prior TBA developed consistent angiographic and morphological vasospasm. In comparison with control vessels and nondilated vasospastic vessels, vessels dilated with TBA and then exposed to clotted blood showed significantly diminished responses to all compounds tested, with the exception of prostaglandin F2α.
Conclusions These results indicate that in vivo TBA results in a degree of functional impairment of vascular smooth muscle that persists for at least 7 days. This result is consistent with previous observations of the acute effects of TBA in isolated arteries. Furthermore, these results support the hypothesis that normal smooth muscle function is required for the development of vasospasm. Finally, these results indicate that TBA performed before the onset of vasospasm prevents its development.
The development and maintenance of delayed-onset cerebral vasospasm after aneurysmal SAH is believed to represent a combination of sustained smooth muscle constriction and structural change, both of which cause narrowing of the arterial lumen diameter. The mechanisms involved have been subjected to intensive experimental investigation.1 Severe vasospasm causing symptomatic cerebral ischemia can now be treated with TBA, which is the mechanical dilatation of vasospastic arteries by means of inflatable microballoons attached to the tips of percutaneously inserted arterial catheters.2 3 4 5 6 7 8 While it has been recognized that TBA results in efficient arterial dilatation and that improvement in cerebral blood flow leads to clinical recovery in a high percentage of patients, the morphological and functional effects of TBA on the cerebral arterial wall have only recently been reported.9 10 11
Previous work in this laboratory has demonstrated that in vitro TBA of both normal and vasospastic canine basilar arteries causes an immediate and profound functional impairment of vascular smooth muscle, as well as providing structural evidence of arterial wall stretching.9 However, determination of longer-term effects of TBA requires an animal model of cerebrovascular spasm in which the affected artery is accessible to consistent and safe in vivo TBA. We have thus developed a model of vasospasm using placement of clotted blood around the distal cervical ICA.
While it is clear that TBA can dilate spastic vessels, it is not known whether prior TBA can prevent the development of vasospasm. This question is of more than academic interest; if it is possible to alter the function of normal arteries by stretching, and this then renders them insensitive to agents that would normally produce vasospasm, the observation would have important implications about the mechanism by which vasospasm develops. It would also have clinical implications, in that if prior TBA can prevent the development of cerebrovascular spasm, it would support early TBA to halt the progression of the condition. Finally, long-term effects of TBA remain uncertain, and the use of the model and the experimental protocol outlined below would enable additional information on this area to be obtained.
Materials and Methods
Twelve mongrel dogs of either sex weighing between 16 and 25 kg were used in this study. The protocol was evaluated and approved by the University of Alberta Animal and Ethics Review Committee, and experiments were conducted with strict adherence to the standards of the Canadian Council on Animal Care.
Model of Blood Clot–Induced Vasospasm in the Canine High Cervical Carotid Artery
This model is an adaptation of models used by others in the rat femoral artery12 and in the rabbit cervical carotid artery.10 11 13 Prior studies, performed in 20 animals in our laboratory, have shown that placement of a blood clot held in place by a silicone elastomer sheath around the canine ICA results in typical angiographic and structural vasospasm 7 days later, while placement of a silicone elastomer sheath alone does not significantly affect the vessel wall. Furthermore, the time course of vasospasm development and resolution in this model is consistent with that observed in human cerebral arteries. For this study the following surgical protocol was used:
Animals were anesthetized with sodium pentobarbital (0.5 mL/kg) and intubated on day 0 for angiography, TBA, and blood clot placement. Adequate anesthesia was maintained by administration of intravenous boluses of pentobarbital (0.05 mL/kg). The animals were allowed to breathe room air spontaneously, and arterial blood gases were determined on the first four animals to confirm that the protocol maintained normocarbia. A midline cervical incision was made, and both cervical ICAs were exposed and punctured with a 24-gauge angiocatheter for angiography (performed with 5 mL of iothalamate meglumine, injected at a rate of 0.75 mL/s). With the operating microscope used for magnification, 60-mm sections of both ICAs were then dissected free of adjacent tissues.
One of the ICAs was chosen by block randomization to undergo TBA, and a silicone elastomer balloon angioplasty catheter was introduced into this ICA lumen through a small arteriotomy proximal to the area of study. The balloon, which was 7 mm in diameter and 12 mm in length when inflated (303.9 to 506.5 kPa pressure), was used to dilate the ICA twice for 10 seconds, each time to approximately 150% of its original diameter along a length of 50 mm. Angiography was then repeated on this artery. Both ICAs were then covered along their dissected segments with 10 mL of autologous arterial clotted blood. The blood clot was placed within a 50-mm segment of silicone elastomer tubing sheath with a 10-mm inner diameter. The silicone elastomer tubing was secured with three silk ties along its length. Small cotton pledgets positioned at either end of the tube contained the clot.
The animals were cared for in the usual fashion for 7 days with daily neurological monitoring. There was no mortality or severe morbidity related to the procedures outlined above, and no animal developed a neurological deficit. On day 7 the animals were again anesthetized, the cervical incision was reopened, and angiography was repeated on both ICAs. The animals were then killed with sodium pentobarbital (30 mg/kg).
Four animals, designated as group A, underwent immediate in situ perfusion fixation of both ICAs with the use of 2.5% glutaraldehyde in 0.12 mol/L Millonig’s buffer solution (pH 7.2). In the remaining group of 8 animals (group B), both ICAs were immediately removed and placed in Krebs’ solution of the following composition: 120 mmol/L NaCl, 5 mmol/L KCl, 1.5 mmol/L CaCl2, 1 mmol/L KH2PO4, 1 mmol/L MgSO4, 25 mmol/L NaHCO3, 55 mmol/L dextrose, aerated with 95% O2/5% CO2 and maintained at 37°C. These arterial segments were used for pharmacological studies. In addition, a segment of ICA proximal to the original surgical field and the previously dissected ICA segment was removed on one side from each animal to provide a control preparation for both the morphological and pharmacological studies. Two rings were cut from each isolated ICA obtained from animals in group B, thus enabling us to examine the pharmacological properties of dilated and nondilated blood-coated ICAs and normal control ICAs. The inner diameters of the segments were measured with a micrometer under magnification before pharmacological analysis. The remainder of the arterial segments were prepared for morphological analysis.
Angiographic Measurement of Vasospasm and the Effects of TBA
For all angiograms the diameter of the carotid artery in millimeters was determined by direct measurement of the angiogram, at the point corresponding to the midpoint of the blood-filled silicone elastomer sheath. For the nondilated group, the degree of angiographic vasospasm was determined by comparing vessel diameters on day 0 and on day 7 and calculating the percent change in vessel diameter. For the dilated group, the initial magnitude of the TBA-induced dilatation was estimated by determining the percent change in vessel diameter on day 0, before and after angioplasty. After 7 days, the persistence of the dilatation induced on day 0 was estimated by calculating the percent change in vessel diameter between that seen on the day 7 angiogram and that seen on the day 0 preangioplasty angiogram.
Responses of arterial rings were recorded isometrically with force-displacement transducers connected to a polygraph. Rings of cerebral arteries were suspended between two stainless steel hooks, under a resting tension of 1 g in organ baths of 10 mL working volume containing Krebs’ bicarbonate solution maintained at 37°C and bubbled with 95% O2/5% CO2. After an equilibration period of 1 hour, during which the Krebs’ solution was changed every 15 minutes, the response to KCl (60 mmol/L) was recorded, and preparations were washed until resting tension was again obtained. Cumulative dose-response curves for noradrenaline (10−9 to 10−5 mol/L), 5-HT (10−9 to 10−5 mol/L), and PGF2α (10−9 to 10−6 mol/L) were then recorded for each arterial ring. Tonic contraction of preparations was established with the use of noradrenaline (10−5 mol/L) before vasorelaxation studies in which cumulative dose-response curves to the calcium ionophore A23187 (10−8 to 10−5 mol/L) were obtained. The response to papaverine (5×10−4 mol/L) was also recorded after tonic contraction with the use of noradrenaline (10−5 mol/L) had been established. Preparations were washed until the resting tension had been restored, before another agent was tested. The ring preparations were studied in the organ baths for approximately 8 to 10 hours. At the end of each experiment the ring preparations were tested with KCl (60 mmol/L) to confirm that the responses were not different from the initial responses obtained to KCl at the beginning of the experiment.
Segments from the three groups of vessels were examined with transmission electron microscopy and scanning electron microscopy. Cross sections of the vessel wall were examined with transmission electron microscopy. The intact vessel wall (luminal and cross-sectional aspects) was examined with scanning electron microscopy.
All specimens were prefixed in 2.5% glutaraldehyde in 0.12 mol/L Millonig’s buffer solution (pH 7.2) overnight at room temperature. After samples had been washed three times for 15 minutes each in Millonig’s buffer, they were postfixed with 1% osmium tetroxide in the same buffer for 2 hours. These samples were briefly washed in distilled water and dehydrated in a graded series of ethanol solutions (50%, 60%, 70%, 80%, and 90%; 10 to 15 minutes each grade) before the final two 10-minute rinses with absolute ethanol. From this point onward, preparation for scanning electron microscopy and transmission electron microscopy samples differed.
For the scanning electron microscopy study, samples in absolute ethanol were dried in a CO2 critical-point dryer at 31°C for 5 to 10 minutes and then mounted on aluminum stubs. All samples were sputter-coated with gold before examination under the scanning electron microscope.
For the transmission electron microscopy study, absolute ethanol bathing the samples was replaced with propylene oxide, which was changed three times at 10-minute intervals. Samples were then embedded in Araldite CY212 mixture/propylene oxide for 1 hour and subsequently in complete Araldite CY212 mixture overnight in a vacuum desiccator. The mixtures embedding the samples were allowed to polymerize at 60°C for 2 days before sectioning. Sections were stained with 4% uranyl acetate for 30 minutes and lead citrate for 5 minutes. Photomicrographs of samples were taken at 75 kV with a transmission electron microscope.
We assessed the morphological appearance of nondilated and dilated groups separately; that is, nondilated arteries and dilated arteries were compared with normal arteries. Based on electron micrographs, a pairwise semiquantitative comparison of morphological differences between nondilated and normal arteries and dilated and normal arteries for each dog was performed by three independent researchers blinded to specimen identification. Specific features in scanning electron micrographs of the intact vessel wall were identified as follows: degree of luminal narrowing, wall thickness, and corrugation of intima and IEL. For transmission electron microscopy of vessel cross sections, observations included the degree of thinning of the entire vessel wall and its component layers; stretching and breakage of the IEL; straightening, stretching, and surface rippling of smooth muscle cells; and the amount of endothelial changes such as cell loss and flattening of the luminal surface.
For the pharmacological study, comparisons between groups at each concentration for each vasoconstrictor or vasodilator were assessed with one-way ANOVA, followed by a Scheffé’s test of multiple comparisons if a significant probability was reached. Data were expressed as mean±SE of the mean. A level of P<.05 was considered significant.
Equipment and Supplies
Force-displacement transducers (model FT.03) and a polygraph (model 7D) were obtained from Grass Instrument Co. The critical-point dryer was manufactured by Seevac, Inc. The sputter-coater, model S150B, was manufactured by Edwards Vacuum. The scanning electron microscope (model S-2500) and the transmission electron microscope (H-7000) were obtained from Hitachi Ltd.
Angiographic Measurement of Vessel Diameter
In the absence of arterial dilation with TBA, the placement of a clot of autologous blood around the artery resulted in a highly significant constriction of the ICA. On day 7 after clot placement, the angiographic vessel diameters in the nondilated group were 48% to 89% of their original size, with a mean reduction to 63±3% (P<.05) (Fig 1B⇓). Dilation by TBA had an immediate effect on arterial diameter. On day 0, immediately after TBA and before clot placement, the angiographic vessel diameters in the dilated group were 148% to 200% of their original size, with a mean of 166±6% (P<.05) (Fig 2B⇓). On day 7 after angioplasty and clot placement, the angiographic vessel diameters in the dilated group showed that they remained dilated, with a diameter of 125% to 203% of their original size, giving a mean of 164±9% (P<.05) (Fig 2C⇓). Similar results were obtained when the actual vessels were measured with a micrometer. On day 7 after clot placement, the measured vessel diameters in the nondilated group were 51% to 97% the size of normal vessels, with a mean of 69±2% (P<.05), while vessels that had been dilated with TBA on day 0 had a diameter of 120% to 163% of control, with a mean of 144±2% (P<.05).
Thus, these results show that periarterial clot placement around segments of canine cervical ICAs results in significant reduction of vessel diameter after 7 days and that angioplasty of arterial segments immediately before the clot placement results in a vasodilatation that is sustained for 7 days even in the presence of surrounding clot.
Pharmacological Effects of Vasoconstrictor Agents
Responses to a single dose of KCl (60 mmol/L) and cumulative dose-response curves for noradrenaline, 5-HT, and PGF2α were recorded for the three groups of vessels: normal, nondilated, and dilated.
Dilated vessels showed significantly diminished responses in comparison to nondilated and normal vessels when exposed to KCl at 60 mmol/L. The response of nondilated vessels in comparison to normal vessels was diminished, but the results did not reach statistical significance. These results are shown in Fig 3A⇓.
Dilated vessels showed significantly diminished responses in comparison to nondilated and normal vessels when exposed to noradrenaline at concentrations of 10−8 to 10−5 mol/L. The responses of nondilated vessels in comparison to normal vessels were slightly diminished at all concentrations, but the diminution did not reach statistical significance at any point. These results are shown in Fig 3B⇑.
Dilated vessels showed significantly diminished responses in comparison to nondilated and normal vessels when exposed to 5-HT at concentrations of 10−7 to 10−5 mol/L, while at the lower concentrations of 10−8 and 10−9 mol/L the diminution was present but only significant when the comparison was between dilated and nondilated vessels. As in the case of noradrenaline, the responses of nondilated vessels in comparison to normal vessels were slightly diminished at concentrations of 10−7 to 10−5 mol/L, but not at a statistically significant level. These results are shown in Fig 3C⇑.
The responses of vessels in all study groups was very small. There were no responses of any arterial rings at the lower concentrations of 10−9 and 10−8 mol/L, nor was there any significant difference between normal vessels and those exposed to blood with prior TBA at any concentration. Vessels in vasospasm appeared to be more responsive than control or dilated vessels, and this reached statistical significance at a concentration of 10−6 mol/L. These results are shown in Fig 3D⇑.
Although the responses to PGF2α differ from those to other vasoconstrictor agents in that they are smaller, and vasospasm seems to result in enhanced responsiveness to PGF2α in this model, the observation that TBA pretreatment reduces the reactivity of arteries surrounded by perivascular clot is common to all vasoconstrictors tested. The observation that the reduction in response of nondilated (vasospastic) arteries compared with normal arteries (when tested with KCl, noradrenaline, and 5-HT) did not reach statistical significance is consistent with other models of vasospasm.9
Pharmacological Effects of Vasodilators
After tonic contraction with noradrenaline (10−5 mol/L) had been established, cumulative dose-response curves for the calcium ionophore A23187 and the response to a single dose of papaverine (5×10−4 mol/L) were recorded for the three groups of vessels. In most cases, pretreatment with noradrenaline produced a large enough tonic contraction in dilated vessels to allow comparative percent relaxation data to be obtained.
Calcium Ionophore A23187
Dilated vessels showed significantly diminished relaxations in comparison to nondilated and normal vessels when exposed to the calcium ionophore A23187 at concentrations of 10−6 and 10−5 mol/L. The responses of normal vessels in comparison to nondilated vessels were slightly diminished at all concentrations, but not at a statistically significant level. This is consistent with other models of vasospasm.9 These results are shown in Fig 4⇓.
All vessel groups showed 100% relaxation after exposure to papaverine (Fig 5⇓).
These studies suggest that for the calcium ionophore A23187 (10−7 to 10−5 mol/L), an endothelium-dependent vasorelaxant, vessels treated with periarterial blood but first dilated with TBA have an impaired response compared with vessels treated similarly but not dilated on day 0. However, endothelium-independent relaxation to papaverine is preserved even in dilated vessels.
Changes Observed With Scanning Electron Microscopy
Photomicrographs of normal, dilated, and nondilated vessels are shown in Fig 6⇓. Scanning electron microscopy of nondilated vasospastic vessels showed moderate diminution of the vessel lumen, decreased luminal diameter/wall thickness ratio (Fig 6B⇓, bottom), corrugation of the IEL, and folding of the endothelial surface (Fig 6B⇓, top). Similar observations of dilated vessels showed moderate enlargement of the vessel lumen (Fig 6C⇓, bottom), patchy endothelial denudation, and straightening and thinning of the IEL (Fig 6C⇓, top). Scanning electron micrographs of normal vessels are shown in Fig 6A⇓ (top and bottom).
Changes Observed With Transmission Electron Microscopy
Photomicrographs of normal, nondilated, and dilated vessels are shown in Fig 7⇓. Results were consistent with those obtained by scanning electron microscopy; nondilated vasospastic vessels showed corrugation of the IEL, folding of the endothelial surface, and thickening of the vessel wall, especially the tunica media (Fig 7B⇓). There was some swelling and vacuolation of endothelial cells, with cellular rounding and some cell separation. Rounding of smooth muscle cells with surface rippling and occasional breaks in the IEL were also noted. Transmission electron microscopy of dilated vessels showed flattening of the endothelial cells; straightening, thinning, and occasional rupturing of the IEL; and straightening and crowding of smooth muscle cells in the tunica media (Fig 7C⇓). A transmission electron micrograph of a normal vessel is shown in Fig 7A⇓.
Use of TBA for Cerebral Vasospasm
Conventional treatment for symptomatic cerebral vasospasm has usually involved the use of hypervolemic and hypertensive therapy. However, recent advances in interventional neuroradiology have allowed for the development of newer therapeutic modalities. One technique that has gained favor since its first description by Zubkov et al2 in 1984 is TBA. It is successful in reversing the clinical effects of vasospasm in approximately 75% of selected patients4 5 and generally produces results that are immediate and long-lasting. Aneurysmal repair is a prerequisite for the use of TBA. The technique is not without complications, including vessel rupture and occlusion.6 At present TBA is generally reserved for cases of moderate to severe symptomatic vasospasm that have been refractory to conventional treatment measures. However, it is not clear that these are the ideal conditions under which TBA should be employed, and earlier implementation may optimize patient outcomes. In a retrospective study, Coyne et al7 concluded that the best results of angioplasty occurred when it was performed within a short time of the onset of symptoms. Little benefit was seen in patients of poor clinical grade or when a new neurological deficit had become established. They suggested that angioplasty be considered if hypervolemic, hypertensive therapy did not reverse the symptoms of vasospasm within 6 to 12 hours.
Models of Vasospasm
To perform prospective, time-course studies of vasospasm, in vivo animal models have been developed. These have involved the use of cerebral arteries in animals including the cat,14 dog,9 15 and monkey.16 The advantages of these models include the presence of a cisternal space into which autologous blood can be introduced and retained and the fact that they use intracranial cerebral arteries, thus facilitating comparison with SAH in humans. However, because of their small size and intracranial location, it is sometimes difficult to access these arteries in the course of endovascular techniques such as TBA. This has led to the development of models of vasospasm in which extracranial arteries such as the femoral artery of the rat12 and the cervical carotid artery of the rabbit10 11 13 are used. These models can contribute valuable information, but it is important to remember that there are morphological and pharmacological differences between extracranial and intracranial arteries17 18 and that some caution must be used when results obtained from experiments involving these arteries are extrapolated to human cerebral arteries. Despite our concerns about these limitations, we have developed a model of vasospasm in the canine high cervical carotid artery. As demonstrated by the angiograms and vessel caliber measurements outlined earlier, the approach produces angiographic and morphological vessel constriction after 7 days in a consistent and reproducible fashion. The vessels under study are of approximately the same caliber as the larger cerebral arteries at the base of the human brain and, most important from the perspective of this study, the vessels are suitable for endovascular manipulation.
In Vivo Mechanism of Action of TBA
There is accumulating evidence, both in extracranial10 11 19 20 and in intracranial9 21 arteries, that TBA results in sustained arterial dilatation through a mechanism of smooth muscle cell injury and paralysis. This is consistent with other evidence that vasospasm represents active smooth muscle cell contraction22 and that it can be acutely reversed in the presence of vasodilators such as papaverine.23 24 The results of our study suggest that in vivo balloon angioplasty performed immediately before the induction of vasospasm prevents vasoconstriction after SAH and produces a functional impairment in vascular reactivity that is sustained for at least 7 days. These results are consistent with our previous report on the effects of immediate in vitro TBA9 and those of others investigating in vivo TBA performed after SAH.10 11 21
High millimolar concentration of KCl produces smooth muscle cell contraction through electromechanical coupling mechanisms with depolarization of the sarcolemma,25 whereas noradrenaline, 5-HT, and PGF2α all act through pharmacomechanical coupling mechanisms with activation of second messengers such as 1,2-diacylglycerol and inositol 1,4,5-trisphosphate.25 26 27 The results of the present experiment indicate that both of these mechanisms are affected. After 7 days, normal arteries subjected to TBA showed responses to KCl, noradrenaline, and 5-HT that were markedly attenuated compared with vasospastic and normal arteries. This suggests a sustained functional impairment of smooth muscle cell function that does not reverse rapidly. The precise nature of this impairment cannot be determined from these experiments but may involve altered calcium homeostasis that prevents the intracellular signaling leading to contraction.
The minimal response of all vessel groups to the presence of PGF2α is in contrast to previous results found in preparations of other canine cerebral arteries.9 It is possible that the canine cervical carotid artery expresses fewer PGF2α receptors than the canine basilar artery. The persistence of vasospasm in arteries with a rather weak response to PGF2α would appear to provide additional evidence against a pivotal role for this compound in cerebrovascular spasm.
The calcium ionophore A23187 produces vasorelaxation through an endothelium-dependent mechanism. This mechanism requires the elevation of intracellular calcium within endothelial cells. This leads to the stimulation of NOS, which converts l-arginine to NO, a potent vasodilator. The NO then rapidly diffuses both within the endothelial cells and across membranes to nearby smooth muscle cells, where it interacts with soluble guanylyl cyclase, to cause relaxation.28 29 30 The calcium ionophore A23187 directly promotes entry of calcium across the endothelial cell membrane.29 In the present experiments the response of dilated vessels to the calcium ionophore was attenuated in comparison to the responses of vasospastic and normal arteries. This may be because the patchy endothelial damage seen with scanning and transmission electron microscopy after TBA reduces the amount of NO generated and/or because there is diminished NO synthesis and release in the remaining endothelium. Also, the responsiveness to NO of the guanylyl cyclase in the smooth muscle cells of dilated vessels may be reduced.
Endothelium-independent relaxation to papaverine is preserved after TBA. This agent has rapid and profound effects on most smooth muscle cells, and it is not surprising that the residual response to vasoconstrictors was reversed by papaverine.
It has been suggested that vasospastic arteries may remain dilated after TBA because angioplasty disrupts arterial wall components such as smooth muscle cells, myofibroblasts, or the extracellular matrix, thereby causing a mechanical impairment of contraction.31 The results of this experiment indicate that TBA causes alterations in the IEL, tunica intima, and tunica media, but that it does not cause frank disruptions in those layers. These results are generally consistent with previous in vitro and in vivo studies.9 10 11 21 32 33 However, although no gross mechanical disruptions of the vessel wall were seen, it is possible that the smooth muscle cell functional impairment observed represents a cellular or subcellular mechanical disruption. Indeed, it may be that after TBA the artery consists of different cell populations, some of which may be approximately normal in function, while others have gross impairment of normal function.
A number of questions remain to be answered. The work reported here involved a considerable dilatation of the artery. This was done because initially it was thought that the higher balloon pressures were required to dilate the thicker walled carotid artery and because it was possible to actually observe the process of dilatation through the operating microscope. The results were entirely successful in preventing the development of cerebrovascular spasm, but it is certainly possible that a less forceful dilatation would have been equally effective and perhaps safer. Studies of varying the effects of TBA with various levels of distension or duration have important clinical implications. Early results of work from our laboratory, in which lower balloon dilatation pressures (101.3 to 202.6 kPa) were used, are similar to those reported here. The protocol described here did not contain a study group in which TBA was performed on arteries without subsequent blood clot placement. Current experiments in our laboratory include such a group, and preliminary results indicate a pharmacological impairment similar to that observed in dilated arteries surrounded with blood clot. Finally, we do not yet know long-term effects in normal arteries that have been distended by TBA, and the 7-day period studied here should be extended to examine the effects weeks or months later, to determine whether normal function returns.
In conclusion, this canine in vivo model of vasospasm using the ICA produces consistent and reproducible vasospasm, as demonstrated by both angiography and direct measurement of vessel caliber. This region of the cerebral circulation is readily accessed by catheter and is thus suitable for studies of TBA. In this model, TBA performed before the induction of vasospasm prevents its development, lending experimental support to the growing clinical evidence that balloon angioplasty may have optimal benefit when performed early in the course of symptomatic vasospasm. Examination of the pharmacological and gross morphological changes present 7 days after TBA indicates a greater impairment in vasoreactivity than structural disruption of the vessel wall. Balloon dilatation and stretching of the arterial wall, which can be observed morphologically on scanning electron microscopy, result in an alteration of smooth muscle contraction and vasodilation that is sustained for at least 7 days. Further studies are required to determine whether and when vessels subjected to angioplasty return to normal contractile behavior.
Selected Abbreviations and Acronyms
|ICA||=||internal carotid artery|
|IEL||=||internal elastic lamina|
|NOS||=||nitric oxide synthase|
|TBA||=||transluminal balloon angioplasty|
This study was supported by a grant from the Department of Surgery, Division of Neurosurgery, University of Alberta.
Reprint requests to J. Max Findlay, University of Alberta, 2D1.02 W.C. Mackenzie Health Sciences Centre, 8440-112 St, Edmonton, Alberta, Canada T6G 2B7.
Presented in part at the 21st International Joint Conference on Stroke and Cerebral Circulation, San Antonio, Tex, January 25-27, 1996, and the 1996 American Association of Neurological Surgeons Meeting, Minneapolis, Minn, April 27-May 2, 1996.
- Received December 6, 1996.
- Revision received February 17, 1997.
- Accepted March 12, 1997.
- Copyright © 1997 by American Heart Association
Findlay JM. Vth International Conference on Cerebral Vasospasm: an overview. In: Findlay JM, ed. Cerebral Vasospasm. Proceedings of the Vth International Conference on Cerebral Vasospasm. Amsterdam, Netherlands: Elsevier Science Publishers BV; 1993:1-3.
Higashida RT, Hieshima GB, Tsai FY, Halbach VV, Norman D, Newton TH. Transluminal angioplasty of the vertebral and basilar artery. AJNR Am J Neuroradiol.. 1987;8:745-749.
Macdonald RL, Zhang J, Han H. Angioplasty reduces pharmacologically mediated vasoconstriction in rabbit carotid arteries with and without vasospasm. Stroke. 1995:26:1053-1060.
Pile-Spellman J, Berenstein A, Bun T, Oot R, Baker K, Peterson J. Angioplasty of canine cerebral arteries. AJNR Am J Neuroradiol.. 1987;8:938. Abstract.
Macdonald RL, Weir BK. A review of hemoglobin and the pathogenesis of cerebral vasospasm. Stroke.. 1991;22:971-982.
Nishizuka Y. Intracellular signalling by hydrolysis of phopholipids and activation of protein kinase C. Science.. 1992;258:607-614.
Dinerman JL, Lowenstein CJ, Snyder SH. Molecular mechanisms of nitric oxide regulation: potential relevance to cardiovascular disease. Circ Res.. 1993;73:217-222.
Yamamoto Y, Smith RR, Bernanke DH. Mechanism of action of balloon angioplasty in cerebral vasospasm. Neurosurgery. 1992:30:1-6.