Angioplasty Reduces Pharmacologically Mediated Vasoconstriction in Rabbit Carotid Arteries With and Without Vasospasm
Background and Purpose We tested the hypothesis that vasospastic arteries do not reconstrict after angioplasty because angioplasty decreases smooth muscle contractility.
Methods Twenty-four rabbits had carotid angiography and placement of silicone elastomer sheaths around both carotid arteries in the neck. Sheaths were empty (control groups) or filled with clotted blood (vasospasm groups). Angiography was repeated 2 days later, and one carotid artery was dilated with a balloon catheter. Animals were killed 1, 5, or 28 days after angioplasty, and the carotid arteries were studied pharmacologically under isometric tension.
Results Before angioplasty, there was significant vasospasm in the vasospasm groups but not in the control groups (P<.05, ANOVA). Angioplasty produced significant, long-lasting dilation of arteries in the vasospasm groups. One and 5 days after angioplasty, arteries from control and vasospasm groups that had angioplasty had significantly reduced contractions to serotonin, KCl, and caffeine compared with arteries not subjected to angioplasty. Twenty-eight days after angioplasty, contractions were reduced in arteries subjected to vasospasm compared with controls, but there were no differences between arteries with or without angioplasty. At all times after angioplasty, vasospasm significantly decreased acetylcholine-induced relaxations of arteries contracted with serotonin. Relaxations were further decreased by angioplasty in the vasospasm group 1 day after angioplasty. Arterial wall compliance was significantly decreased in the vasospasm and control groups at all times after angioplasty, although there were no significant differences between arteries with and without angioplasty.
Conclusions These results suggest that arteries do not reconstrict after angioplasty because angioplasty decreases smooth muscle contractility. There was no evidence that angioplasty disrupted the arterial wall matrix, as judged by the lack of increase in arterial wall compliance after angioplasty.
The vasospasm that complicates aneurysmal subarachnoid hemorrhage (SAH) consists of at least two processes. Under some circumstances, intra-arterial papaverine can reverse experimental and clinical vasospasm, suggesting that smooth muscle contraction is important in vasospasm.1 2 A papaverine-insensitive component of vasospasm that is associated with decreased arterial contractility and compliance also develops after SAH and has been suggested to be due to structural changes in the arterial wall.2 Several treatments for vasospasm, such as calcium channel antagonists and intra-arterial papaverine infusions, are directed at preventing or reversing smooth muscle contraction. Transluminal balloon angioplasty is also efficacious in dilating vasospastic arteries, and recurrence of vasospasm after dilation by angioplasty has not been reported.3 4 Vasospastic arteries might not renarrow after angioplasty if angioplasty decreased smooth muscle con-tractility, damaged the extracellular matrix, or both. The effects of angioplasty on vasospastic arteries have not been examined in detail.5 6 This study tested the hypoth-
esis that vasospastic arteries do not reconstrict because angioplasty decreases smooth muscle contractility. A rabbit model of induced vasospasm in the cervical carotid arteries was used to provide arteries accessible to an endovascular approach and of sufficient size to allow angioplasty.7
Materials and Methods
The 48 carotid arteries of 24 male New Zealand White rabbits (weight, 2.5 to 4.1 kg) were randomly divided into 12 groups to undergo baseline carotid angiography and placement of silicone elastomer sheaths around both carotid arteries in the neck (day 0). Sheaths were either empty to serve as controls or filled with blood to induce vasospasm. Transluminal balloon angioplasty was performed on one carotid artery in each animal 2 days (day 2) after placement of silicone elastomer sheaths. Animals were killed by injection of sodium pentobarbital (100 mg/kg IV) 1 (day 3), 5 (day 7), or 28 (day 30) days after angioplasty. The cervical carotid arteries were removed and studied under isometric tension.
All animal procedures and care were according to standards set by the US Department of Health and Human Services and the National Institutes of Health. Experiments were approved by the Animal Care and Use Committee of the University of Chicago.
Angiography, Induction of Vasospasm, and Angioplasty
Methods for angiography, induction of vasospasm, and angioplasty have been described.8 Briefly, rabbits were anesthetized on day 0 with ketamine (20 mg/kg IM) and xylazine (5 mg/kg IM). They were intubated and ventilated on a small-animal respirator (Harvard Apparatus) with oxygen and 0.5% to 2% isoflurane. Body temperature was maintained with a water-jacketed heating pad, and arterial blood gases were monitored on an arterial blood gas machine (STAT Profile 3 Analyzer, Nova Biomedical). Blood pressure and heart rate were continuously monitored with a noninvasive cuff (Criticon Dinamap Research Monitor, Criticon). A single anteroposterior arterial phase angiogram of both common carotid arteries was obtained by manual injection of 5 mL of 65% iohexol through a 4F catheter advanced from the femoral artery into the common brachiocephalic trunk. Both common carotid arteries were then exposed in the neck and dissected free of loose connective tissue. Nonrestricting silicone elastomer sheaths were placed around each common carotid artery. Both sheaths were empty in control groups; both were filled with fresh, nonheparinized, human blood in vasospasm groups.
On day 2, carotid angiography was repeated. A Stealth Dilatation Catheter System (Target Therapeutics) with a 2-mm-diameter by 1-cm-long balloon was advanced under fluoroscopic control from one femoral artery into one common carotid artery that was selected at random. The contralateral carotid artery was not subjected to angioplasty. Angioplasty was performed by inflating the balloon, which had a diameter equal to the diameter of the normal rabbit carotid artery, along the narrowed area of the carotid artery within the sheath. Angiography was repeated on days 3, 7, or 30, and animals were killed.
Arteries were removed and placed in Krebs-Henseleit buffer bubbled with 95% O2/5% CO2 at 4°C. Arteries were cut into rings 4 mm long and suspended between stainless steel hooks in 25-mL water-jacketed tissue baths (Radnoti Glass) filled with Krebs-Henseleit buffer (composition [mmol/L]: Na+ 132, K+ 5.9, Ca2+ 2.5, Mg2+ 1.2, Cl− 122.7, HCO3− 25, SO42− 1.2, H2PO4− 1.2, and glucose 11) bubbled with 95% O2/5% CO2 at pH 7.4 and 37.5°C. Tension was adjusted to 1.5 g as measured by strain gauges (model FT 03, Grass Instrument Corp), and rings were allowed to equilibrate for 1 hour. The bath solution was changed every 15 to 20 minutes. Output from the strain gauges was directed into a preamplifier and a Dell 486 personal computer (Dell Computer Corp) equipped with software to continuously record tension (Lakeshore Technologies). In addition to arteries encased in silicone elastomer sheaths, the carotid arteries were removed from four normal rabbits and studied under isometric tension.
Contractions were induced with KCl (5 to 75 mmol/L), caffeine (10 mmol/L), or serotonin (10−7 to 10−4 mol/L). Relaxation to contraction with serotonin (10−5 mol/L) was tested with acetylcholine (10−7 to 10−5 mol/L). Arterial wall compliance was measured according to Kim et al9 under inhibition of myogenic tone with papaverine (10−4 mol/L) and nicardipine (10−5 mol/L). Preliminary experiments showed that these drugs prevented contractions to KCl (60 mmol/L) and serotonin (10−5 mol/L). The distance between the hooks was decreased to a point where an increase in length produced a measurable increase in tension. Rings were stretched in increments of 0.5 mm, and the tension was recorded after equilibration at the new length. Resting tension also was measured at each length without addition of nicardipine or papaverine by increasing the distance between the hooks and recording the tension.
The following drugs were obtained from Sigma Chemical Co: serotonin, papaverine hydrochloride, caffeine, acetylcholine, and nicardipine. The concentration of drug reported is the final molar concentration in the tissue bath.
Data and Statistical Analysis
The diameter of the carotid arteries was measured four times at three points with the use of an optical micrometer. Compliance was expressed by dividing the difference in length between two measurements by the difference in tension between the two measurements.9 All data are expressed as mean±SEM. Comparisons between multiple groups were made by one-way ANOVA followed by a Bonferroni multiple comparison procedure if significant variation was found. P<.05 was considered significant.
Angiographic Vasospasm and Effect of Angioplasty
Comparison Between Groups at Different Times
There were no significant differences between groups in angiographic arterial diameters at baseline or before angioplasty (Fig 1⇓). The degree of narrowing that developed between day 0 and day 2 angiograms in the vasospasm groups (35±5% to 43±9%) was more severe than in the control groups (10±6% to 33±8%). On the day the animals were killed, there was significant variation in the arterial diameter of arteries with and without angioplasty, although the only pairwise difference was between the arteries of control animals without angioplasty killed at day 30 and animals in the vasospasm group killed at day 7.
In the groups with perivascular application of blood, there was a statistically significant reduction in arterial diameter of 35% to 43% on day 2 (P<.05, Fig 1⇑). After angioplasty, there was a statistically significant increase in the diameter of arteries that underwent angioplasty in the groups killed on days 3 and 7 (38±6% in each group, P<.05). There were no significant changes in the arterial diameters of arteries contralateral to the arteries that underwent angioplasty. Comparison of angiograms taken immediately after angioplasty and on day 3 showed that no significant changes in arterial diameter had occurred. There were significant differences between arterial diameters at baseline and on days 3 and 7 in arteries with and without angioplasty (Fig 1⇑). By day 30, there were no significant differences in the diameter of arteries with or without angioplasty compared with baseline (−10±6%, P=NS).
In the control groups, the reduction in diameter between day 0 and day 2 ranged from 10% to 33% and was not statistically significant in any group (Fig 1⇑). Angioplasty did not significantly alter angiographic diameters in control animals. The only significant intragroup difference in arterial diameter was between arterial diameter after angioplasty in arteries in the no-angioplasty group compared with the diameter on day 30 (P<.05).
Contractions to Serotonin, KCl, and Caffeine
Contractions of normal rabbit carotid arteries (not encased in silicone elastomer sheaths) to KCl and serotonin and relaxations to acetylcholine were not significantly different from control arteries not subjected to angioplasty. There was an insignificant difference in contraction to caffeine between normal (330±180 mg) and control arteries without angioplasty (560±160 mg). Because there were no significant differences, the results for normal arteries are not included in the figures, and the analysis is restricted to a comparison of control and vasospastic arteries that were placed in the silicone elastomer sheaths.
1 Day After Angioplasty (Day 3)
Contractions to serotonin were significantly decreased in both control and vasospastic arteries subjected to angioplasty compared with arteries not subjected to angioplasty (Fig 2⇓). Angioplasty of control arteries also significantly decreased contractions to KCl compared with control arteries that did not undergo angioplasty (Fig 3⇓). In arteries exposed to perivascular blood, angioplasty caused a similar decrease in contraction to KCl, although it was not significantly different from the contraction in the arteries without angioplasty. There were no significant differences in contractions to KCl between control and vasospastic arteries not subjected to angioplasty.
In control arteries, contraction to caffeine was significantly reduced by angioplasty. The contraction of vasospastic arteries to caffeine was also decreased by angioplasty, although this reduction was not significant (Table 1⇓).
5 Days After Angioplasty (Day 7)
Contractions to serotonin were significantly reduced in the vasospastic arteries with and without angioplasty compared with control arteries without angioplasty (Fig 2⇑). Contractions to KCl and caffeine were significantly reduced in vasospastic arteries with angioplasty compared with control arteries without angioplasty (Fig 3⇑ and Table 1⇑). Contractions to KCl, serotonin, and caffeine were always lower in the arteries of each group that had angioplasty, and contractions were generally lower in vasospastic arteries compared with control arteries. In contrast to day 3, when the largest differences in contractility were between arteries with and without angioplasty, on day 7 the largest differences in contractility were between arteries with and without vasospasm.
28 Days After Angioplasty (Day 30)
Contractions to serotonin were significantly lower in vasospastic arteries compared with control arteries (Fig 2⇑). Contractions to KCl and caffeine were also decreased in vasospasm arteries with and without angioplasty compared with control arteries with angioplasty (Fig 3⇑ and Table 1⇑).
Relaxations to Acetylcholine
One day after angioplasty, acetylcholine-induced relaxations of arteries precontracted with serotonin were progressively and significantly diminished by exposure to perivascular blood clot and exposure to angioplasty (Fig 4⇓). Vasospastic arteries exposed to angioplasty showed significantly less relaxation to acetylcholine compared with control arteries with and without angioplasty.
On days 7 and 30, there were no differences within control and vasospasm groups in relaxations to acetylcholine between arteries with and without angioplasty. At both times, however, vasospastic arteries had significant reductions in relaxation to acetylcholine compared with control arteries.
Comparisons at Each Time
On days 3, 7, and 30, there were significant decreases in arterial wall compliance in both control and vasospastic arteries with and without angioplasty compared with normal rabbit carotid arteries (Table 2⇓). There were no significant differences in compliance between control and vasospastic arteries, although there was a trend for compliance to be lower in vasospastic compared with control arteries.
Comparisons Within Groups Across Time
In control arteries not subjected to angioplasty, compliance was significantly lower at each time compared with normal rabbit carotid arteries (P<.0005, Table 2⇑). Compliance was significantly lower on day 30 compared with day 3 (P<.05). In control arteries that underwent angioplasty, compliance was lower compared with normal rabbit carotid arteries (P<.0005). There was a trend for compliance to decrease with increasing time.
For vasospastic arteries with or without angioplasty, compliance was significantly lower than in normal rabbit carotid arteries (P<.0001), although there were no significant differences over time between vasospastic arteries. There was a trend for compliance to decrease with increasing time.
Periarterial blood clot placement around the carotid arteries of rabbits produced vasospasm that lasted for at least 7 days. Placement of empty sheaths resulted in 50% as much narrowing as placement of sheaths filled with blood. The vasospasm produced in this model, therefore, is partly due to periarterial blood and partly due to dissection and placement of the sheaths. Transluminal angioplasty increased the diameter of the vasospastic arteries. The arteries remained dilated for at least 1 day, and vasospasm recurred 5 days after angioplasty. Although systemic arteries were examined and their responses to agonists are known to differ from those of intracranial arteries, it seems unlikely that the contractile mechanisms in the smooth muscle cells and the extracellular matrix of the arterial wall differ enough to make the action of angioplasty different between the two types of arteries.
Mechanism of Action of Angioplasty
The results of this experiment and of previous pathological observations from the rabbit model8 are consistent with the theory that vasospasm does not recur after angioplasty because angioplasty decreases arterial contractility. The difference in contractility between arteries with and without angioplasty correlated with arterial narrowing in that it was most marked 1 day after angioplasty and was less prominent or absent (for contractions to serotonin) on day 7, when vasospasm recurred. The reason for the decrease in contractility is unclear, but the increase in endothelial proliferation that was observed after angioplasty suggests that there was increased damage to the smooth muscle cells beyond that induced by vasospasm itself.8 Endothelial proliferation is a response of the arterial wall smooth muscle cells to injury, although it may also result from injury to the endothelial cells. The transient inhibitory effect of angioplasty on endothelium-dependent relaxation suggests that endothelial cell function is changed by angioplasty. Decreased contractility has been observed after angioplasty of normal or pharmacologically contracted arteries10 11 and at an unspecified time after angioplasty of vasospastic dog basilar arteries.12
Contractility measurements do not rule out the possibility that arteries do not recontract after angioplasty because there is disruption of the extracellular matrix of the arterial wall or of the connections between the smooth muscle cells and the extracellular matrix.13 14 Measurements of arterial wall compliance and of resting tension at different arterial diameters were obtained because damage to the extracellular matrix severe enough to prevent vasospasm would be expected to increase compliance and decrease resting tension. Evidence for these changes was not found pharmacologically, nor was it found in a previous study of the pathological effects of angioplasty.8 It is uncertain whether the tears in the collagen fibrils of the arterial wall that Yamamoto et al13 observed in arteries that underwent angioplasty are severe enough to prevent arteries from renarrowing after angioplasty but not severe enough to be detected pharmacologically.
Endothelial Changes in Arteries Undergoing Angioplasty
Endothelium-dependent relaxations to acetylcholine and other substances are inhibited after SAH and after angioplasty of normal arteries.15 16 In this model they were progressively inhibited with increasing time after periarterial blood clot placement and were further inhibited 1 day after angioplasty, an effect that would be expected to promote vasoconstriction after angioplasty. The effects of angioplasty on the release of other endothelium-derived substances that might have opposite effects, such as endothelins, were not examined. The effect of angioplasty could be due to inhibition of the release of endothelium-derived relaxing factor from endothelial cells or to a decrease in the response of the smooth muscle cells to endothelium-derived relaxing factor. Further studies using bioassay for endothelium-derived relaxing factor and tests of the relaxant effects of agents that cause relaxation independent of the endothelium, such as sodium nitroprusside, would help to differentiate between these two possibilities.
Relation of Pharmacological Changes to Vasospasm
Although many studies have examined pharmacological changes in arteries at single times after SAH,17 18 19 few have systematically studied these changes over time,20 and few have investigated whether the processes that are supposed to cause vasospasm persist after vasospasm resolves. This study found that periarterial blood clot caused progressive decreases in contractility, arterial wall compliance, and endothelium-dependent relaxation that were most severe more than 30 days after vasospasm had reversed. The lack of correlation between these features and vasospasm has been observed before, at least for endothelium-dependent relaxation,20 and suggests that while these changes may contribute to vasospasm, they may not be of primary importance in its development.
Vorkapic et al2 described an early papaverine-sensitive vasospasm that was followed by a progressive papaverine-insensitive vasospasm after SAH in rabbits. The latter phase was associated with decreased arterial contractility and compliance. In the present study angioplasty was performed early, possibly during the predominantly papaverine-sensitive phase of vasospasm. It remains speculative whether the pharmacological effects of angioplasty on contractility would be the same if it were performed, for example, 7 days after placement of periarterial blood clot. Our histopathological studies showed that angioplasty was effective when performed after 7 days in this model.8
Studies of the pathology and pharmacological effects of angioplasty suggest that vasospastic arteries do not reconstrict after angioplasty because angioplasty inhibits smooth muscle contraction. Further studies will be required to elucidate the mechanism or mechanisms that underlie these effects.
This study was supported by a grant from the Physicians’ Services Incorporated Foundation. We thank Target Therapeutics for generously supplying angioplasty catheters.
Reprint requests to R. Loch Macdonald, MD, Section of Neurosurgery, MC3026, University of Chicago Medical Center, 5841 S Maryland Ave, Chicago, IL 60637. E-mail firstname.lastname@example.org.
- Received October 10, 1994.
- Revision received December 28, 1994.
- Accepted March 9, 1995.
- Copyright © 1995 by American Heart Association
Macdonald RL, Wallace MC, Montanera WJ, Glen J. Pathological effects of angioplasty on vasospastic rabbit carotid arteries. J Neurosurg. In press.
Pile-Spellman J, Berenstein A, Bun T, Oot R, Baker K, Peterson J. Angioplasty of canine cerebral vessels. AJNR Am J Neuroradiol. 1987;8:938. Abstract.
Schweitzer DH, Koek GH, van der Laarse A. Spontaneous and pharmacologically-induced vasoconstrictive responses of rat aortic rings are attenuated by balloon angioplasty. Res Commun Chem Pathol Pharmacol. 1991;72:285-294.
Smith RR, Connors JJ III, Yamamoto Y, Bernanke DH. Balloon angioplasty for vasospasm: theoretical and practical considerations. In: Sano K, Takakura K, Kassell NF, Sasaki T, eds. Cerebral Vasospasm. Tokyo, Japan: University of Tokyo Press; 1990:415-420.
Bevan JA, Bevan RD, Frazee JG. Functional arterial changes in chronic cerebrovasospasm in monkeys: an in vitro assessment of the contribution to arterial narrowing. Stroke. 1987;18:472-481.
Debdi M, Seylaz J, Sercombe R. Early changes in rabbit cerebral artery reactivity after subarachnoid hemorrhage. Stroke. 1992;23:1154-1162.
Vorkapic P, Bevan RD, Bevan JA. Pharmacologic irreversible narrowing in chronic cerebrovasospasm in rabbits is associated with functional damage. Stroke. 1990;21:1478-1484.