The Porcine Restenosis Model Using Thermal Balloon Injury:
Comparison with the Model by Coronary Stenting

Animal models have assumed a central position in understanding the mechanisms of angioplasty, stenting, restenosis and development of medical devices.^1–5 The porcine coronary vascular injury models, by either stenting or overstretching injury alone, are now accepted standards by which potential restenosis therapies are studied. Radiofrequency thermal balloon angioplasty was introduced as a new technique for percutaneous arterial dilatation in the 1990s,^6–8 however, increased restenosis rates were observed in these thermal balloon angioplasty patients.^9,10
The purpose of this study was to systemically evaluate the porcine restenosis model by using thermal balloon injury and to compare it with the model of coronary overstretch stenting.

Methods
A total of 22 swine were used for this study. All experiments were performed according to the procedures described under the Stanford University Administrative Panel on Laboratory Animal Care approved protocol.
Anesthesia and procedure. Juvenile Yorkshire swine (25–40 kg) were pretreated with aspirin (650 mg), clopidogrel (75 mg) and nifedipine (30 mg) orally at least 12 hours prior to the procedure. The animals were sedated with telazol (6 mg/kg intramuscularly), followed by inhalant mask induction (isoflurane 4–5%). Following intubation, anesthesia was maintained via ventilation using 2 L/minute of oxygen mixed with 1–3% isoflurane. Animals were placed in the supine position, and continuous electrocardiography (anterior and inferior leads) and hemodynamic monitoring were performed throughout all procedures.
Vascular access was obtained using a standard 8 Fr vascular sheath placed in the right or left carotid and/or femoral arteries, as necessary. All animals received preprocedural heparin intra-arterially (300 IU/kg) and an activated clotting time was maintained at > 300 seconds.
After advancing an 8 Fr guiding catheter through the aorta into the coronary ostia, an intracoronary injection of 0.2 mg nitroglycerin was administered and baseline coronary angiography (CAG) was performed to identify the desirable location for the development of restenotic lesions according to the coronary artery size. A 0.014 inch percutaneous transluminal coronary angioplasty (PTCA) guidewire was advanced through the guiding catheter into the coronary artery. A Radiofrequency (RF) Thermal Balloon Catheter (Innovation in Medicine, LLC, Inver Grove heights, Minnesota), with a balloon-to-artery ratio equal to a range of 1.2–1.3, was placed at the desired location in the coronary artery and inflated to a pressure of 2 atm. The RF generator was then turned on and heat was transmitted to the balloon until 80^°C was reached and maintained for 80 seconds (this includes the ramp-up time as the balloon is heated to the desired temperature). As RF energy was being delivered, the inflation pressure was maintained at 2 atm, and once the timer completed its countdown, the RF generator was automatically shut off.
To compare this thermal balloon injury method, coronary stents were implanted to achieve a 1.3–1.4:1 stent-to-artery ratio. In all swine, 1 stent was implanted in 1 coronary artery and 1 thermal balloon injury was induced in the remaining 2 coronary arteries.
Following thermal balloon and stent deployment, coronary artery patency was confirmed by CAG. Subsequently, the animals were weaned from mechanical ventilation and allowed to recover in their normal housing at the animalcare facility. The animals were monitored daily until the time of planned restudy at 4 weeks and at euthanasia.
Planned restudy and histology. At the end of the experimental protocol (4 weeks post lesion creation), the animals were returned to the cardiac catheterization laboratory for further evaluation. Following induction of anesthesia as previously described, CAG was documented to evaluate the degree of coronary artery stenosis. A 0.014 inch PTCA guidewire was advanced into the coronary arteries and intravascular ultrasound (IVUS) was performed. Using a 2.9 Fr 40 MHz IVUS catheter (Boston Scientific) and automated pullback (0.5 mm/second), IVUS images were obtained and recorded on S-VHS videotape for offline quantitative analysis.
At the procedure’s end, animals were euthanized with an overdose bolus of potassium chloride and the hearts were excised. The vessels were perfusion-fixed with lactated ringers solution followed by 10% buffered formalin and were placed in the same solution for further fixation. All segments of injured coronary arteries were step-sectioned and stained with hematoxylin- eosin, elastic-van Gieson stain and Masson Trichrome stain. All histological slices treated with coronary stents were stained with hematoxylin-eosin and a Movat pentachrome.
Quantitative coronary angiography (QCA) and quantitative coronary ultrasound (QCU) analysis. All angiograms were analyzed using the computer-assisted, automated, edge-detection algorithm software, the PlusPlus (Sanders Data, Palo Alto, California). Lesion length, reference diameter, minimal lumen diameter and diameter stenosis were calculated.
For QCU analysis, the EchoPlaque imaging system (Indec Systems, Inc., Mountain View, California) was utilized for all measurements. After digitization of IVUS recordings, lumen, external elastic membrane (EEM) areas and stent areas were manually traced. Plaque area (PA) was calculated using the following formula: PA = EEM area (stent area) - lumen area (LA). Remodeling of coronary arteries after thermal injury was determined using the remodeling index (RI). The RI was defined as the ratio of the EEM area of minimum LA (MLA) site to the EEM area of the proximal reference site, as previously described,^11,12 and calculated as the EEM area of MLA site divided by that of the proximal reference site. Positive remodeling was defined as RI > 1.05, intermediate remodeling as 0.95 £ RI £ 1.05, and negative remodeling as RI < 0.95.^11,13–15

Statistical analysis. All data are expressed as mean ± standard deviation. The correlation was analyzed by simple linear regression with 95% confidence intervals. A p-value < 0.05 was considered statistically significant.

Results
A total of 18 swine survived until the 4-week follow-up procedure. Mortality was 18.2%; 1 animal died immediately following the procedure and 3 animals died within the 4-week period. Although temporary spasm (resolution occurred without the use of medication over 10 minutes) of the coronary arteries was observed in a few cases, flow-limiting dissections due to balloon inflation were not observed. In some cases, ventricular fibrillation occurred immediately after deflation of the thermal balloon, however, cardioversion was successful in all animals.
A total of 54 coronary arteries (thermal balloon injury: n = 43, coronary stenting: n = 11) at a 4-week timepoint were analyzed for this study.
QCA analysis. Representative CAG is shown in Figure 1. Coronary artery stenoses had consistently developed at 4 weeks after thermal balloon injury. However, as shown in Figures 1B and C, a different intimal response was observed in each animal.

Table 1 demonstrates the results of QCA analysis. The mean minimum lumen diameter (MLD) and percent diameter stenosis (%DS) of the coronary stenting group (Stent) postprocedure were significantly greater than those of the thermal balloon injury group (Thermo) (MLD 3.24 ± 0.40 mm vs. 2.85 ± 0.50 mm; p = 0.02; %DS -22.5 ± 13.3% vs. -3.9 ± 14.2%; p = 0.0003, respectively). At 4 weeks postprocedure, significantly greater coronary stenoses were observed in the Thermo group. Thirty-five of 43 coronary arteries (81.4%) in the Thermo group were angiographically overstretched at postprocedure evaluation, since the mean of %DS was a negative value. Interestingly, the mean of %DS at 2 weeks following thermal balloon injury still remained a negative value, as shown in Figure 2. To investigate the relationship between thermal balloon injury and development of coronary artery stenosis, the correlation between QCA analysis at the acute phase and %DS at 4 weeks after thermal balloon injury was calculated as shown in Figures 3A and B. There were significant linear correlations between the balloon-to-artery ratio, post %DS and DS at 4 weeks (balloon-to-artery ratio: r = 0.538, p = 0.0012; post %DS: r = –0.744, p < 0.0001, respectively). However, no correlation between the heating time and %DS at 4 weeks was observed.

QCU analysis. Representative IVUS images of the Thermo group are shown in Figure 4. Table 2 represents the results of QCU analysis of the Thermo group. The mean remodeling index was 0.67 ± 0.25, and negative remodeling was detected in 88.9% of analyzed coronary arteries. There was no coronary artery stenosis associated with positive remodeling.
Histology. Figure 5 represents photomicrography of histological specimens. In the histological analysis of the Thermo group, neointimal hyperplasia was observed in 82.9% of specimens and mainly composed of smooth muscle cells, proteoglycan matrix, and a number of lesions were found to be fibrotic with traces of calcium. Adventitial inflammation and fibrosis were frequently observed (62.9% and 80.0%, respectively). In 88.6% of cases in the Thermo group, the internal elastic lamina was at least focally disrupted and was diffusely disrupted in the segment with severe stenosis, as shown in Figure 5. Both proximal and distal native vessel sections were intact in all specimens. All stent sections demonstrated surface ablation of the intima (Figure 5). The ablated surface was irregular with focal platelet thrombus deposition on the lumen. Deep-vessel injury was not observed to extend beyond the proximal or distal margins of the stent.

Discussion
In this study, we compared two strategies for inducing coronary artery neointimal thickening in swine. The major finding of this study was that thermal balloon injury could induce a significantly higher degree of coronary stenoses than the accepted standard model of intracoronary overstretch stenting.
Thermal balloon injury results in cell death and adventitial fibrosis. The mechanism of this model might be that adventitial fibrosis prevents positive remodeling by producing a thick, densely collagenous adventitial collar around the vessel that physically limits vessel expansion. Staab et al reported that remodeling in fibrotic compared to inflammatory lesions differed markedly and that remodeling indices for the thermally injured lesions were negative.¹⁶ Similarly, negative remodeling was detected in 88.9% of analyzed coronary arteries in the Thermo group.
In this study, severe arterial injury due to a higher balloonto- artery ratio induced considerable coronary artery stenosis, and the balloon-to-artery ratio correlated well with coronary artery stenosis. Histological analysis also confirmed that the internal elastic lamina was diffusely disrupted in the segment with severe stenosis. Thus, the balloon-to-artery ratio applied in the thermal balloon injury method may be effective in controlling the serverity of coronary artery stenosis. This coronary restenosis model induced by thermal balloon injury is advantageous for coronary imaging studies such as magnetic resonance imaging (MRI), computed tomography (CT), IVUS and optical coherence tomography (OCT), since artifacts by foreign material such as stents are avoided. In addition, this model can be useful for technical training in PCI.

Conclusion
This methodology involving thermal balloon injury may provide reproducible and consistent coronary stenoses. Furthermore, this model can be useful for the evaluation of medical devices, technical training in PCI and development of coronary imaging technologies such as CT, MRI, IVUS and OCT.
Acknowledgements. In the preparation of this manuscript, we thank Heidi Bonneau, RN, MS, for her editorial review, and Alfred Green and Hossein Shenasa, for their technical assistance.

 

References
1. Schwartz RS, Murphy JG, Edwards WD, et al. Restenosis after balloon angioplasty. A practical proliferative model in porcine coronary arteries. Circulation 1990;82:2190–2200.
2. Schwartz RS, Huber KC, Murphy JG, et al. Restenosis and the proportional neointimal response to coronary artery injury: Results in a porcine model. J Am Coll Cardiol 1992;19:267–274.
3. Schwartz RS, Edwards WD, Huber KC, et al. Coronary restenosis: Prospects for solution and new perspectives from a porcine model. Mayo Clin Proc 1993;68:54–62.
4. Suzuki T, Kopia G, Hayashi S, et al. Stent-based delivery of sirolimus reduces neointimal formation in a porcine coronary model. Circulation 2001;104:1188–1193.
5. Schwartz RS, Edelman ER, Carter A, et al. Drug-eluting stents in preclinical studies: Recommended evaluation from a consensus group. Circulation 2002;106:1867–1873.
6. Becker GJ, Lee BI, Waller BF, et al. Radiofrequency balloon angioplasty. Rationale and proof of principle. Invest Radiol 1988;23:810–817.
7. Lee BI, Becker GJ, Waller BF, et al. Thermal compression and molding of atherosclerotic vascular tissue with use of radiofrequency energy: Implications for radiofrequency balloon angioplasty. J Am Coll Cardiol 1989;13:1167–1175.
8. Barry KJ, Kaplan J, Connolly RJ, et al. The effect of radiofrequency-generated thermal energy on the mechanical and histologic characteristics of the arterial wall in vivo: Implications for radiofrequency angioplasty. Am Heart J 1989;117:332–341.
9. Yamashita K, Satake S, Ohira H, Ohtomo K. Radiofrequency thermal balloon coronary angioplasty: A new device for successful percutaneous transluminal coronary angioplasty. J Am Coll Cardiol 1994;23:336–340.
10. Saito S, Arai H, Kim K, Aoki N. Initial clinical experiences with rescue unipolar radiofrequency thermal balloon angioplasty after abrupt or threatened vessel closure complicating elective conventional balloon coronary angioplasty. J Am Coll Cardiol 1994;24:1220–1228.
11. Schoenhagen P, Ziada KM, Kapadia SR, et al. Extent and direction of arterial remodeling in stable versus unstable coronary syndromes: An intravascular ultrasound study. Circulation 2000;101:598–603.
12. Nakamura M, Nishikawa H, Mukai S, et al. Impact of coronary artery remodeling on clinical presentation of coronary artery disease: An intravascular ultrasound study. J Am Coll Cardiol 2001;37:63–69.
13. Pasterkamp G, Borst C, Gussenhoven EJ, et al. Remodeling of de novo atherosclerotic
lesions in femoral arteries: Impact on mechanism of balloon angioplasty. J Am Coll Cardiol 1995;26:422–428.
14. Schoenhagen P, Ziada KM, Vince DG, et al. Arterial remodeling and coronary artery disease: The concept of “dilated” versus “obstructive” coronary atherosclerosis. J Am Coll Cardiol 2001;38:297–306.
15. Ehara S, Kobayashi Y, Yoshiyama M, et al. Spotty calcification typifies the culprit plaque in patients with acute myocardial infarction: An intravascular ultrasound study. Circulation 2004;110:3424–3429.
16. Staab ME, Srivatsa SS, Lerman A, et al. Arterial remodeling after experimental percutaneous injury is highly dependent on adventitial injury and histopathology. Int J Cardiol 1997;58:31–40.