The Relation Between De Novo Atherosclerosis Remodeling and Angioplasty-Induced Remodeling in an Atherosclerotic Yucatan Micropig Model
Abstract—Geometric remodeling in de novo atherosclerosis and in restenosis after balloon angioplasty constitutes a change in total arterial circumference that, together with plaque growth or neointima formation, determines the lumen of the artery. The heterogeneous nature of arterial obstructions raises the question of whether early and late outcomes (restenosis) of angioplasty are affected by the degree and direction of de novo atherosclerotic remodeling. This study was designed to assess the relationship between atherosclerotic remodeling and the degree and mechanism of restenosis after balloon angioplasty. Atherosclerosis was induced in 27 peripheral arteries of 18 Yucatan micropigs by a combination of denudation and atherogenic diet. Balloon angioplasty was performed, with serial intravascular ultrasound and quantitative angiography before and after intervention and at 42 days’ follow-up. We used the relative media-bounded area (MBA), defined as the MBA of the treated site divided by the MBA of the reference, before angioplasty as a measure of remodeling in de novo atherosclerosis and late MBA loss as a measure of remodeling after balloon angioplasty. Relative MBA before angioplasty was not correlated with angiographic and echographic acute gain after balloon angioplasty (r=.22, P=.28 and r=.14, P=.48) or with late lumen loss (r=−.05, P=.81 and r=.19, P=.33). No correlation was found between relative MBA and late MBA loss (r=.14 and P=.48). In the atherosclerotic Yucatan micropig, remodeling during de novo atherosclerosis has no relevance for acute gain and late lumen loss after balloon angioplasty. Both the direction and the extent of remodeling after balloon angioplasty are not related to the direction and extent of remodeling during de novo atherosclerosis.
- Received May 12, 1997.
- Accepted November 7, 1997.
Geometric remodeling in de novo atherosclerosis1 2 3 4 and in restenosis after balloon angioplasty5 6 7 8 9 constitutes a change in total arterial circumference that, together with plaque growth or neointima formation, determines the lumen of the artery. The change in total arterial circumference ranges from enlargement, which may lead to an actual increase in lumen size, to shrinkage. In the latter case, remodeling contributes to luminal (re)narrowing. The different contributions of remodeling and plaque formation to the atherosclerotic process result in a heterogeneity of atherosclerotic arterial obstructions, with a large plaque mass accommodated by a locally enlarged artery at one extreme and a small plaque in a focally shrunken artery at the other. The heterogeneous nature of arterial obstructions with respect to remodeling raises the question of whether early and late outcomes (restenosis) of angioplasty are affected by the degree and direction of de novo atherosclerotic remodeling and whether, for instance, a lesion based predominantly on shrinkage will restenose by remodeling rather than by neointima formation.
In a recent serial IVUS study from our laboratory in human femoral arteries,10 we observed no difference in acute lumen gain between three categories of de novo atherosclerotic remodeling, ie, enlarged, unchanged, and shrunken arteries. However, the mechanism of acute gain by balloon angioplasty was different, showing more plaque reduction or axial redistribution of plaque in enlarged arteries and more stretch of the arterial wall (increase in MBA) in shrunken arteries.
Restenosis is the arithmetic sum of neointima formation,11 12 13 14 being a combination of smooth muscle cell hyperplasia, extensive matrix elaboration, and remodeling,5 6 7 8 9 and it is important for future therapeutic strategies to determine whether the relative contributions of neointima formation and remodeling are influenced by atherosclerotic remodeling. This study was designed to assess the relationship between atherosclerotic remodeling and the degree and mechanism of restenosis after balloon angioplasty. We performed serial IVUS, quantitative angiography, and histology in an atherosclerotic Yucatan micropig model and measured the geometric dimensions of lumen and area within the internal elastic membrane, in both reference and stenotic segments, before and after balloon angioplasty.
Eighteen Yucatan micropigs with an average weight of 25 kg at termination were used for this study. All animals were started on an atherogenic diet at the age of 3 to 5 months. Two weeks thereafter, they underwent Fogarty denudation of the internal iliac, external iliac, and femoral arteries; were continued on the atherogenic diet; and then, 4 weeks later, underwent a second denudation of the same arterial segments. The atherogenic diet was continued for an additional 5 to 8 months, after which time selected arteries were balloon dilated and the atherogenic diet discontinued. Six weeks later, the animals were killed and the arteries harvested. The protocol complied with the regulations of the Dutch Animals Act of 1977 and was approved by the ethics committee on Animal Experiments of the Faculty of Medicine, Utrecht University.
Atherogenic Diet and Anesthesia
In addition to essential nutrients, vitamins, and salts, 1.5% cholesterol, 17.5% casein, 19.5% lard, and 0.5% bile salts formed the basic atherogenic components of the diet that had a daily nutritional value equivalent to 2400 kcal. In pilot experiments, it had been shown that this regimen resulted in a sustained 15-fold increase in total cholesterol and a 2.5-fold increase in HDL level. Water intake was not restricted. The diet during follow-up was a regular, nonatherogenic chow diet, with a nutritional value equal to that of the atherogenic diet.
For denudation, intervention, and termination, the animals were anesthetized with intravenous metomidate (4 mg/kg body weight) and ventilated (Servo, EM 902) with a mixture of O2/N2O=1:2 (vol/vol) and 1% to 2% halothane. During each procedure, the animals were heparinized (100 IU/kg thromboliquine, Organon Technika) to an activated partial thromboplastin time of >80 seconds. Every 15 minutes, 0.25 mg atropine was given intravenously.
One day before intervention, dipyridamole 2 dd 250 mg (twice daily) and acetylsalicylic acid (125 mg) PO was given, and this regimen was continued for 2 weeks after intervention. During intervention and termination, a continuous infusion of nitroglycerin (20 μg/min) was given, and 10 mg nifedipine was administered through the nasogastric tube to prevent arterial spasm.
For denudation, re-denudation, intervention, and termination, the arterial tree was accessed through a carotid cutdown and insertion of an arterial 8F sheath into the aorta descendens under fluoroscopic guidance. For denudation and redenudation, the left carotid artery was used. For intervention and termination, the right carotid artery was used. An 8F guiding catheter was advanced to the aortic bifurcation. Through this, contrast (Telebrix, Laboratoire Guerbet) angiography was performed and the Fogarty balloon catheter, the IVUS catheter, and the balloon angioplasty catheter were advanced. After the denudations and intervention, the carotid arteries were carefully sutured for future interventions.
For both denudation procedures, 3- to 4-cm segments (measured by a radio-opaque ruler) in the iliac and femoral arteries were denuded by triple withdrawal of a 4F Fogarty catheter that was manually inflated with a 50:50 (vol/vol) water/contrast material mixture.
Five to 8 months after the second denudation, selected sites of angiographic arterial narrowing were balloon dilated. Before and after each intervention, angiography and IVUS were performed under fluoroscopic guidance. For balloon dilation, a standard peripheral (length, 2 to 4 cm; outer diameter, 4 to 7 mm) or coronary (length, 2 cm; diameter, 2 to 4 mm) balloon catheter was advanced over a 0.03-in. or a 0.014-in. guide wire. Because the diameter of the target vessels ranged from 1.5 to 6 mm and because the dilation ratio needed to be standardized, we used different balloon sizes. The mean dilation ratio, defined as the diameter of the inflated balloon on fluoroscopy divided by the angiographic diameter of the reference segment, was 1.21±0.16. The balloon was inflated three times for 1 minute at a pressure of 10 atm.
After a follow-up of 42 days, the pigs were anesthetized and angiograms and IVUS measurements were made. The animals were then killed (by bleeding) and the arteries were harvested for histology and histomorphometry. To avoid collapse and contraction of the arteries, they were pressure (60 mm Hg) infused with a 50°C 3%/48.5%/48.5% (wk/vol/vol) agar-agar/contrast/water gel that congealed at room temperature within the arteries. After solidification of this mixture, the arteries were submerged in situ under 4% formalin for 4 to 6 hours. With the aid of anatomic markers (side branches) on intervention angiograms and agar/contrast postmortem fluoroscopy, the balloon-dilated segments were identified and marked with adventitial sutures.15 The arterial tree was then taken out en bloc and postfixed for at least 24 hours.
Angiography and IVUS
Angiograms were performed before and after each intervention and at follow-up. Contrast (Telebrix) was injected selectively into the artery under study through an 8F guiding catheter. The fluoroscopy was recorded at a cine rate of 12 images/s by using a digital C-arm (Philips). The image with the highest contrast was selected and stored on DAT tape for later analysis.
The angiographic diameters of the arteries were measured using a semi-automated program. The quantitative edge detection algorithm is applied to the digitized gray value of a proposed line perpendicular to the center axis of the lumen. The gray value distribution along the perpendicular line has its maximum outside the lumen and its minimum in the middle of the lumen. The edge of the lumen was defined by the position of the pixel with a gray value equal to the average of the maximum and minimum. The diameter of the artery was calculated by this full-width/half-maximum distance. In each artery, lumen diameters were measured at intervals of 0.5 cm, including the treated segment and a proximal and distal reference segment. Reference segments were chosen at a distance of at least 1 cm proximal to and distal from the balloon-dilated site. To use equal positions at different time points (preprocedure, postprocedure, and at follow-up), these positions were documented relative to an anatomic landmark. Angiography was calibrated by using a radio-opaque ruler.
The mean and minimum lumen diameters (MLD) of the lesion were determined. Angiographic acute gain was defined as the difference between postprocedure and preprocedure lumen diameters, and angiographic late lumen loss was defined as the difference between postprocedure and follow-up lumen diameters. Angiographic percent stenosis was calculated as (1−MLD/RLD)×100, where RLD is the distal reference segment. The site of stenosis was area of the treated site with the minimum lumen diameter.
IVUS recordings were made before and after intervention and at follow-up by using a 30-MHz ultrasound transducer (Du-MED), which rotated up to 16 times per second within a 4.1 French catheter. The axial resolution of the system was 0.1 mm. The images were displayed on a monitor and recorded on VHS videotape (Fig 1⇓). IVUS images were analyzed with a digital video analyzer as described previously.16 Fluoroscopy was performed during IVUS so that the IVUS images were documented relative to an anatomic landmark to match the IVUS images at different time points (preprocedure, postprocedure, and follow-up). In the IVUS images, the area circumscribed by the interface between the echodense intimal layer and the echolucent media was manually traced and was designated as the MBA. In addition, the LA was traced. Echographic acute gain was defined as the difference between postintervention and preintervention LA, and echographic late lumen loss as the difference between postintervention and follow-up LA. In addition to late lumen loss, late MBA loss was introduced to measure the remodeling after angioplasty. Late MBA loss was defined as the difference between the postintervention MBA and follow-up MBA at either the site of the initial stenosis or the reference sites. To measure remodeling from de novo atherosclerosis before balloon angioplasty, we used an MBA index (relative MBA): MBA at the treated site divided by the MBA at the reference site. A relative MBA >1.0 indicates enlargement, and relative MBA <1.0 indicates shrinkage compared with the reference site. Intimal area was defined as the difference between MBA and LA. At follow-up, the intima is the sum of plaque and neointima formation. Neointima formation is therefore calculated as the difference between the intima at follow-up and that at preintervention. Echographic percent stenosis is calculated as 1−minimum lumen area (MLA) divided by reference lumen area (RLA) ×100. The site of stenosis was the area of the treated sited with the smallest LA.
Histology and Histomorphometry
After harvesting, the arteries were cut into 0.5-cm blocks and embedded in paraffin. Serial sections of 5-μm thickness were cut at intervals of 1 mm and stained with hematoxylin/eosin and van Gieson’s elastin stains. For morphometry, the elastin van Gieson’s–stained sections were used. Images of the sections were filmed on a black-and-white film (camera model MX5/5010) mounted on a Leitz microscope. The frames were digitized using a videograbber and computer (Silicon Graphics) and analyzed with Analyze (Biomedical Imaging Resource, Mayo Foundation). The lumen boundary and the internal and external elastic laminas were manually traced, and their perimeters and areas were measured. To correct for form artifacts, the LA, internal elastic lamina area, and external elastic lamina area were calculated from their perimeters after assuming circular geometry. Analogous to the IVUS measurements, the intima was calculated as the difference between the internal elastic lamina area and LA.
For statistical calculations, we used an average of the entire treated site, the length of which varied with balloon length. For IVUS and angiography, the number of cross sections analyzed varied from 4 to 8 and for histology, from 5 to 10, both depending on balloon length. The data were analyzed separately for the MLD. All data in text and the Table⇓ are presented as mean±SD. SPSS 6.1 was used for all statistical calculations. Pearson’s correlation coefficients were calculated when indicated.
Balloon angioplasty was performed in 46 arteries in 18 Yucatan micropigs. Complete IVUS and angiographic imaging was performed in 41 of 46 arteries. In 5 arteries, IVUS images were either unavailable or of low quality. The analyses in this study were based on 27 of 41 balloon-dilated arteries in which a positive echographic acute gain was achieved. The negative gain in the remaining arteries resulted from spasm after balloon angioplasty, which potentially influences the relation between acute gain and late loss and introduces a group of postangioplasty arteries that are enlarged but not necessarily remodeled. The 14 excluded arteries were comparable to the arteries that were successfully dilated before balloon angioplasty with respect to angiographic and echographic reference diameter, angiographic and echographic percent stenosis before angioplasty, relative MBA before balloon angioplasty, and dilation ratio.
The average echographic stenosis was 21.7±24.6% (Table⇑). The echographic acute gain at the treated site was 3.9 mm2 and ranged from 0.4 to 8.6 mm2 The echographic late lumen loss ranged from to −2.0 to 15.4 mm2 and had an average of 4.5 mm2 The mean plaque area was 1.7±0.9 mm2
Histology Versus IVUS and IVUS Versus Angiography
Intimal area, which comprised preexisting plaque and neointima formation as measured by IVUS and histology at follow-up, respectively, was significantly correlated (r=.62, P<.001), as illustrated by the example of two corresponding cross sections (Fig 1⇑). The mean intimal area measured by IVUS was 2.6±1.2 mm2. With histology, the mean intimal area was 2.4±1.4 mm2. Angiographic lumen diameter at the treated site was significantly correlated with echographic LA at the treated site (r=.73, P<.001). Also, angiographic MLD was significantly correlated with echographic minimum LA (r=.65, P<.001).
Acute Gain–Late Lumen Loss Relationships for the Treated Site
Acute gain and dilation ratio are considered to reflect the severity of injury imparted to the arterial wall. Echographic acute gain was correlated significantly with echographic late lumen loss (r=.48, P=.011). Also for angiographic acute gain and late lumen loss, a significant correlation was found (r=.67, P<.001). The dilation ratio was also correlated significantly with angiographic acute gain (r=.43, P=.027) but not with echographic acute gain (r=−.09, P=.66). The average neointima formation at follow-up was 0.8±1.4 mm2 and was not correlated with angiographic or echographic late lumen loss. Instead, late MBA loss was correlated with angiographic late lumen loss and very strongly with echographic late lumen loss (r=.39, P=.046 and r=.95, P<.001, respectively).
Acute Gain–Late Lumen Loss Relationships for the MLD
Subanalysis with data for MLD revealed similar results, with positive correlations between angiographic and echographic acute gain and late lumen loss (r=.53, P=.005 and r=.77, P<.001, respectively) and a positive correlation between dilation ratio and angiographic acute gain (r=.65, P=.002) but no correlation between dilation ratio and echographic acute gain (r=−.18, P=.44). Also at the MLD, late MBA loss was correlated with angiographic late lumen loss and echographic late lumen loss (r=.44, P=.036 and r=.94, P<.001).
Remodeling in De Novo Atherosclerosis
Relative MBA before angioplasty ranged from 0.56 (shrinkage) to 1.66 (enlargement), with a mean of 0.99±0.23. The cumulative distribution of the relative MBA is shown in Fig 2⇓. For the treated site, relative MBA before angioplasty was not correlated with angiographic or echographic acute gain after balloon dilation (r=.22, P=.28 and r=.14, P=.48) or with angiographic or echographic late lumen loss (r=−.05, P=.81 and r=.19, P=.33) and echographic neointima formation at follow-up (r=.14, P=.47). No correlation was found between relative MBA before angioplasty and plaque before angioplasty (r=.04, P=.85), as shown in Fig 3⇓.
Subanalysis of data at the MLD revealed similar results, with no correlations between relative MBA before angioplasty with angiographic and echographic acute gain (r=.27, P=.18 and r=−.11, P=.60) or with angiographic and echographic late lumen loss (r=−.11, P=.60 and r=.03, P=.90) and with echographic neointima formation at follow-up (r=.14, P=.52).
Remodeling in De Novo Atherosclerosis and Remodeling After Balloon Angioplasty
We used the relative MBA before angioplasty as a measure of remodeling in de novo atherosclerosis and late MBA loss as a measure of remodeling after balloon angioplasty. Late MBA loss ranged from −2.6 to 17.0 mm2, with an average of 3.7±4.3 mm2. The cumulative distribution of the late MBA loss is shown in Fig 4⇓. No correlation was found between relative MBA and late MBA loss (r=.14 and P=.48, as shown in Fig 5⇓). Also at the MLD, no correlation was found between relative MBA and late MBA loss (r=.027, P=.90).
Because vessel size can be a potentially confounding parameter, we also performed a multiple regression analysis entering vessel size and relative MBA as independent variables and MBA loss as the dependent variable. The multiple regression coefficient improved only slightly when relative MBA was entered (from r=.43 to r=.49), with an actual decrease in P value from .027 to .035 (due to one extra degree of freedom). Although vessel size is a predictor of MBA loss, it does not affect the weak and insignificant relation between relative MBA and MBA loss.
Remodeling in de novo atherosclerosis, ranging from enlargement to shrinkage, has previously been demonstrated.1 2 3 In contrast to enlargement, shrinkage of an atherosclerotic artery enhances arterial obstruction that is caused by atherosclerotic plaque.2 3 Remodeling after balloon angioplasty has also been reported and has been recognized as a major determinant of restenosis. After balloon angioplasty, shrinkage seems to be the predominant, although certainly not the only, mode of remodeling. Postdilation remodeling, therefore, usually leads to a reduction in lumen size5 6 7 8 9 that is additional to the lumen encroachment by neointima formation.
In this combined angiographic, serial ultrasound, and histological study of an atherosclerosis model in the Yucatan micropig, we studied the influence of remodeling in de novo atherosclerosis on the acute and long-term outcomes after balloon angioplasty. The de novo atherosclerosis was induced by a combination of mechanical denudation and an atherogenic diet, 5 to 8 months before balloon angioplasty. Our principal findings are the following: (1) Both compensatory enlargement and shrinkage were observed in this model of de novo atherosclerosis in peripheral pig arteries. (2) Remodeling in de novo atherosclerosis was not correlated with acute gain or late lumen loss after balloon angioplasty. (3) Remodeling in atherosclerosis and remodeling after balloon angioplasty were not related.
Remodeling in De Novo Atherosclerosis
Compensatory enlargement of human coronary arteries1 17 and human femoral arteries2 in the presence of developing intimal plaque has been reported by many investigators. Incomplete enlargement and shrinkage have been observed in human femoral2 and human coronary3 arteries. In accordance with these human studies, we observed the whole range of atherosclerotic remodeling, from compensatory enlargement to shrinkage. Previously, we reported that remodeling after balloon angioplasty importantly determines restenosis in this model,7 to an extent that has been reported in human coronary arteries.5 Thus, this animal model seems to be highly suited for the study of remodeling before and after angioplasty.
The mechanism of remodeling in de novo atherosclerosis is still unknown. Zarins et al18 proposed two explanations of compensatory enlargement: (1) The local increase in wall shear stress caused by plaque development may stimulate endothelium-dependent arterial dilation. and/or (2) The development of plaque may lead to degradation of the media and adventitia, resulting in passive bulging of the plaque. We are currently studying proteolytic activity of plaque in this model to address this question.
Remodeling in Atherosclerosis: Angiographic and Echographic Outcome
In this study, remodeling in de novo atherosclerosis did not seem to be related to angiographic and echographic acute gain. This finding is in accordance with a recent study in human femoral arteries from our laboratory by Pasterkamp et al.10 In human femoral arteries, we observed less stretch of the arterial wall and more plaque regression or axial plaque redistribution in the group wherein either no de novo remodeling or compensatory enlargement was observed than in the group with shrinkage. However, no differences in acute gain were found.
Remodeling in De Novo Atherosclerosis and Remodeling After Balloon Angioplasty
Several investigators have suggested that remodeling in atherosclerosis may have relevance for acute and long-term outcomes after balloon angioplasty.3 10 In this study, however, we found no relation between remodeling in atherosclerosis and remodeling or late lumen loss after balloon angioplasty. Whether this result is specific for the animal model used in this investigation remains to be determined, although as mentioned previously remodeling in this model seems to parallel remodeling in human coronary arteries closely.
Comparison of IVUS and Histological Measurements
Although we observed a discrepancy between intimal area at follow-up found on IVUS and histomorphometry, these measurements were correlated significantly. Differences between these methodologies can be explained by two factors. First, histological tissue may shrink by 10% to 20% with dehydration, in spite of measures to preserve lumen dimensions.19 This explains possible underestimation of the intima by histology. Second, in ultrasound measurements, not only the intima but also the lumen appears larger than in histological or angiographic measurements, suggesting systematic overestimation by IVUS. Other investigators have observed a similar overestimation by IVUS.20 21
Limitations of the Study
This study was performed in peripheral arteries in an animal model of complex atherosclerosis, and it remains to be determined whether our findings are representative of the human situation. Recent serial IVUS studies by Mintz et al4 5 and Di Mario et al6 showing that remodeling after balloon angioplasty contributes significantly to restenosis confirmed their initial findings7 8 9 22 in earlier clinical and animal studies. Since our model involves an initial balloon injury, we cannot exclude the possibility that the mechanism of de novo remodeling in our model differs from true atherosclerotic remodeling. However, by creating this model, we succeeded in mimicking the human condition as closely as possible, at least with regard to the presence of a range of remodeling. It is unlikely that the initial injury still influences the response to the second injury, as these events took place 4 months apart. Although the arteries contained considerable plaques, with an average of 1.7±0.9 mm2, the angiographic stenoses were mild, and therefore it remains unknown whether progression of the atherosclerotic lesion to a more severe stenosis would have influenced the results.
The late lumen loss and late MBA loss observed in this study might be explained by acute and delayed elastic recoil, as we did not include early (eg, 24-hour) follow-up. However, in previous experiments, we did include a 48-hour follow-up group and found no early lumen loss or early MBA loss by elastic recoil.23 Furthermore, Kimura et al24 showed in the SURE study using serial IVUS that there was no difference between LA immediately after balloon dilation and the lumen 24 hours after balloon angioplasty. Taken together, it is highly unlikely that acute or delayed elastic recoil accounted for late MBA loss in this study.
In the atherosclerotic Yucatan micropig model, remodeling during de novo atherosclerosis has no relevance for the acute gain and late lumen loss after balloon angioplasty. Both the direction of remodeling, ie, whether the artery enlarges or shrinks, and the extent of remodeling after balloon angioplasty are unrelated to the direction and extent of remodeling during de novo atherosclerosis.
Selected Abbreviations and Acronyms
|MLD||=||minimum lumen diameter|
This study was supported by the Netherlands Heart Foundation (grant NHS 92.365, to M.J.P.) and in part by the Interuniversity Cardiology Institute of the Netherlands. The authors thank Jolanda van der Zande for biotechnical support.
Pasterkamp G, Wensing PJW, Post MJ, Hillen B, Mali WPTM, Borst C. Paradoxical arterial wall shrinkage contributes to luminal narrowing of human femoral arteries. Circulation. 1995;91:1444–1449.
Mintz GS, Kent KM, Pichard AD, Satler LF, Popma JJ, Leon MB. Contribution of inadequate arterial remodeling to the development of focal coronary artery stenoses: an intravascular ultrasound study. Circulation. 1997;95:1791–1798.
Mintz GS, Popma JJ, Pichard AD, Kent KM, Satler LF, Wong SC, Hong MK, Kovach JA, Leon MB. Arterial remodeling after coronary angioplasty: a serial intravascular ultrasound study. Circulation. 1996;94:35–43.
Di Mario C, Gil R, Camenzind E, Ozaki Y, Von Birgelen C, Umans VA, De Jaegere P, de Feyter PJ, Roelandt JRTC, Serruys PW. quantitative assessment with intracoronary ultrasound of the mechanisms of restenosis after percutaneous transluminal coronary angioplasty and directional atherectomy. Am J Cardiol. 1995;75:772–777.
Post MJ, Borst C, Kuntz RE. The relative importance of arterial remodeling compared with intimal hyperplasia in lumen renarrowing after balloon angioplasty: a study in the normal rabbit and the hypercholesterolemic Yucatan micropig. Circulation. 1994;89:2816–2821.
Kakuta T, Currier JW, Haudenschild CC, Ryan TJ, Faxon DP. Differences in compensatory vessel enlargement, not intimal formation, account for restenosis after angioplasty in the hypercholesterolemic rabbit model. Circulation. 1994;89:2809–2815.
Lafont A, Guzman LA, Whitlow PL, Goormastic M, Cornhill JF, Chisolm GM. Restenosis after experimental angioplasty: intimal, medial, and adventitial changes associated with constrictive remodeling. Circ Res. 1995;76:996–1002.
Liu MW, Roubin GS, King SB3. Restenosis after coronary angioplasty: potential biologic determinants and role of intimal hyperplasia. Circulation. 1989;79:1374–1387.
McPherson DD, Sirna SJ, Hiratzka LF, Thorpe L, Armstrong ML, Marcus ML, Kerber RE. Coronary arterial remodeling studied by high-frequency epicardial echocardiography: an early compensatory mechanism in patients with obstructive coronary atherosclerosis. J Am Coll Cardiol. 1991;17:79–86.
Post MJ, De Smet BJGL, van der Helm YJ, Borst C, Kuntz RE. Arterial remodeling after balloon angioplasty or stenting in an atherosclerotic experimental model. Circulation. 1997;96:996–1003.
Kimura T, Kaburagi S, Yokoi H, Nakagawa Y, Tamura T, Nosaka H, Nobuyoshi M, Mintz GS, Popma JJ, Leon MB. Time course of vessel response after coronary angioplasty: final result of serial ultrasound restenosis (SURE) study. Circulation. 1996;94:I-135.