Reduced Intimal Hyperplasia in Rabbits via Medical Therapy after Carotid Venous Bypass

Abstract

Intimal hyperplasia is a major cause of restenosis after the interventional or surgical treatment of occlusive arterial disease. We investigated the effects of clopidogrel, calcium dobesilate, nebivolol, and atorvastatin on the development of intimal hyperplasia in rabbits after carotid venous bypass surgery.

We divided 40 male New Zealand rabbits into 4 study groups and 1 control group. After occluding the carotid arteries of the rabbits, we constructed jugular venous grafts between the proximal and the distal segments of the occluded artery. Thereafter, group 1 (control) received no medication. We administered daily oral doses of clopidogrel to group 2, calcium dobesilate to group 3, nebivolol to group 4, and atorvastatin to group 5. The rabbits were killed 28 days postoperatively. The arterialized jugular venous grafts were extracted for histopathologic examination.

Intimal thicknesses were 42.87 ± 6.95 μm (group 2), 46.5 ± 9.02 μm (group 3), 34.12 ± 5.64 μm (group 4), and 48.37 ± 6.16 μm (group 5), all significantly less than the 95.12 ± 9.93 μm in group 1 (all P < 0.001). Medial thicknesses were 94 ± 6 μm (group 2), 101.5 ± 13.52 μm (group 3), 90.5 ± 9.69 μm (group 4), and 101.37 ± 7.99 μm (group 5), all significantly thinner than the 126.62 ± 13.53 μm in group 1 (all P < 0.001).

In our experimental model of carotid venous bypass grafting in rabbits, clopidogrel, calcium dobesilate, nebivolol, and atorvastatin each effectively reduced the development of intimal hyperplasia. Herein, we discuss our findings and review the medical literature.

Intimal hyperplasia is a major cause of restenosis after the interventional or surgical treatment of occlusive arterial diseases.^1–3 Intimal hyperplasia has been described as the adaptive process that takes place in response to hemodynamic stresses or injuries to the vascular bed.^2,4 Some degree of endothelial injury always occurs after any percutaneous or surgical treatment of vascular disorders. The vascular endothelial layer routinely forms an anticoagulant layer by secreting various mediators that enable blood flow.⁵ This anticoagulant and its homeostatic properties are lost in cases of vascular injury, and a procoagulant surface for the elements of blood is formed. The chief pathologic process involves platelet aggregation and degranulation with proliferation and migration of vascular smooth-muscle cells, along with an increase in matrix proteins at the site of endothelial injury. This adaptive process results in intimal hyperplasia, a phenomenon that is seen often upon clinical follow-up of invasively treated occlusive vascular disease. In numerous studies, restenosis due to intimal hyperplasia has been as high as 20% to 50%.^6–8

We investigated the effects of clopidogrel, calcium dobesilate, nebivolol, and atorvastatin on the development of intimal hyperplasia in rabbits that had undergone venous carotid artery bypass. This method was similar to clinical coronary artery or peripheral arterial bypass surgery, and it was practical, replicable, and inexpensive. Our findings and a literature review are presented here.

Materials and Methods

Study Protocol

This study was conducted at the Medical & Surgical Research Center of Ondokuz Mayis University, and the protocol was approved by the university's Ethical Committee of Animal Studies. We experimented upon 40 male New Zealand rabbits, each 2,500 to 3,000 g in weight and of similar age. The animals were furnished by the university's Surgical Research Center, whose staff provided humane care throughout. The rabbits were kept in a daylit environment, in specially designed cages, at a mean temperature of 20 °C. They were randomly divided into 4 study groups and 1 control group of 8 animals each.

Anesthetic and Surgical Technique

General anesthesia was achieved in each rabbit upon the administration of 10 mg/kg of ketamine hydrochloride intravenously and 3 mg/kg of xylazine intramuscularly. During each 2.5-hour operation, half of the above dosages were repeated every 45 minutes. Each animal was placed on the surgical table in supine position with cervical extension. Cefazolin sodium (10 mg/kg) was injected intramuscularly as prophylaxis against infection. A vertical 5-cm skin incision was performed, 1.5 cm to the right of and lateral to the anterior midline, in order to reach the carotid artery and jugular vein. After complete occlusion of the carotid artery was instituted by use of heavy silk ligature, each animal underwent a venous carotid artery bypass that was constructed between the proximal and distal portions of the carotid artery, with the accompanying jugular vein as the conduit. A single intravenous dose of 150 IU/kg of heparin was administered. Side branches of the jugular vein segment that was to be used as the bypass conduit were ligated with 4-0 silk sutures. Each bypass graft was at least 2.5 cm long. Proximal and distal anastomoses (length, 3 mm) were performed end-to-side with 9-0 polypropylene sutures. Pulsatile blood flow to the venous graft was confirmed. The carotid artery was ligated externally with 4-0 silk suture (Fig. 1). After hemostasis was achieved, the surgical site was closed. Intraoperatively, 1 rabbit died and was replaced. The subjects were protected from the effects of hypothermia for approximately 3 hours until they regained a normal activity level.

Fig. 1 Operative photograph of completed venous bypass graft (arrow) to the internal carotid artery

Medical Therapy

Postoperatively, the rabbits in group 1 (control group) were not medicated. Starting from the 5th postoperative hour, the 32 rabbits in the 4 study groups received their respectively assigned medication dosages once daily for 4 weeks, through nasoenteral tubes. The group 2 rabbits initially received 10 mg/kg of clopidogrel. The rabbits in group 3 were administered 100 mg/kg of calcium dobesilate. In group 4, the rabbits received 0.5 mg/kg of nebivolol. The group 5 rabbits were initially given 10 mg/kg of atorvastatin.

The effective dosages and safety limits of the pharmaceutical agents had been determined after a literature review. The animals in groups 3 and 4 tolerated the dosages; however, 2 rabbits each in groups 2 and 5 experienced diarrhea and lethargy and died on postoperative day 3. Necropsy studies revealed extensive liver necrosis. The dead rabbits were replaced, and the initial 10-mg/kg daily dosages of clopidogrel and atorvastatin, respectively, were reduced to 2.5 mg/kg in those 2 groups thereafter. The reduced dosages were tolerated well.

Graft patency was monitored daily with use of a hand-held Doppler ultrasonographic apparatus. Each graft's nonstenotic arterial waveform was recorded as evidence of patency. In addition, 1 rabbit in group 1 underwent carotid angiography (Fig. 2), and 1 each from groups 2 and 5 underwent duplex ultrasonographic examination. This last was deemed impractical for study purposes and was limited to those 2 animals.

Fig. 2 Postoperative carotid angiogram (arrow indicates venous bypass graft).

Results

After 4 weeks, the animals were placed under general anesthesia and were surgically explored. Arterialized pulsatile jugular vein grafts were seen in all. The carotid arteries were explanted 1 cm below and above the proximal and distal anastomoses, along with the graft and neighboring nerves, and the animals were humanely killed. One graft each in groups 1, 2, 3, and 5 was found to be occluded (overall patency, 90%).

All of the explanted grafts were examined histopathologically (Fig. 3). After the tissue samples were irrigated with formaldehyde solution and fixed in buffered neutral formalin, microslices were stained with hematoxylin and eosin and with Verhoeff's stain, in order to show the elastic fibers in the walls of the jugular vein. The distance between the vascular lumen and internal elastic lamina was recorded as the intimal thickness. The distance between the internal and external elastic laminae was the medial thickness. Intimal–medial thickness ratios were then calculated.

Fig. 3 Histologic views of the explanted grafts at the end of the study. Lesser intimal hyperplasia can be seen in all of the study groups in the A) control group: B) clopidogrel group, C) calcium dobesilate group, D) nebivolol group, and E) atorvastatin group (H & E, orig. ×10).

a = adventitia; i = intima; m = media; arrows indicate borders of intimal and medial areas

As Table I shows, intimal thicknesses were 95.12 ± 9.93 μm in group 1, 42.87 ± 6.95 μm in group 2, 46.5 ± 9.02 μm in group 3, 34.12 ± 5.64 μm in group 4, and 48.37 ± 6.16 μm in group 5. These were significantly lower in the study groups than in the control group (all P < 0.001). Intimal thickness was significantly lower in group 4 than in group 3 (P = 0.029) and in group 5 (P = 0.008). Similar results were noted when medial thicknesses were compared: 126.62 ± 13.53 mm in group 1, 94 ± 6 μm in group 2, 101.5 ± 13.52 mm in group 3, 90.5 ± 9.69 mm in group 4, and 101.37 ± 7.99 mm in group 5 (all study groups, P < 0.001 vs control group). Differences between the study groups were not statistically significant.

TABLE I. Intimal and Medial Thicknesses by Group, and Intimal–Medial Thickness Ratios

Intimal–medial thickness ratios were 0.75 ± 0.097 (group 1), 0.46 ± 0.099 (group 2), 0.46 ± 0.096 (group 3), 0.37 ± 0.03 (group 4), and 0.47 ± 0.064 (group 5). The ratios were significantly less in the study groups than in the control group (all P < 0.001). There were no significant differences between the study groups.

Statistical Analysis

Data were evaluated by use of the Statistical Package for the Social Sciences version 10.0 (SPSS Inc.; Chicago, Ill) and were stated as mean ± SD. Sample-size determination was made before the study was conducted. Levene's test of equality of error variances showed homogenous distribution of the data. Differences of means were calculated by using univariate analysis of variance. Homogenous subset analysis (with Bonferroni corrections) and post hoc Tukey HSD (honestly significantly different), Bonferroni, and Tamhane tests were also applied.

Discussion

Two major hypotheses have emerged regarding the pathologic processes that underlie the development of intimal hyperplasia after the treatment of occlusive vascular disease: response to injury, and adaptive remodeling. Currently, pharmaceutical agents (such as statins, nitric oxide [NO] donors or precursors, and clopidogrel) and surgical techniques (external mechanical barriers or internal degradable stents for grafts) are being applied to inhibit or retard intimal hyperplasia. Although many recent experimental studies in the area of molecular biology have produced seemingly effective results in animals (including the techniques of heparin bonding to polytetrafluoroethylene, free-radical scavenging, and photochemical therapies), no agent or technique has proved to be safe, applicable, and efficient in human beings.^2–4,9

In different reports of intermediate and long-term follow-up, vein-graft restenosis rates range from 20% to 50%.2,6,10,11 In a rabbit model of balloon injury, it was shown that re-endothelization started as early as the 3rd day and that intimal hyperplasic activity peaked by the end of the 1st month.¹² In a more recent study of the development of intimal hyperplasia, examination was recommended at the end of the 4th week.¹³ The pathogenesis may differ for arterial and venous allografts, but the same time period was used for venous allografts in different studies.^9,14 In accordance with these data, we decided to examine the grafts after 4 weeks. We chose animals of similar age and weight. Because studies have indicated an impact of estrogens on the development of intimal hyperplasia,^15,16 we worked with male rabbits.

Basic fibroblast growth factor (bFGF) and platelet-derived growth factor (PDGF) are probably the most important mediators in the proliferation of smooth-muscle cells and in migration initiation,¹⁷ with bFGF showing its effects on smooth-muscle-cell proliferation solely after endothelial injury.¹⁸ In vitro studies on human and animal smooth-muscle cells showed that nebivolol inhibits the smooth-muscle-cell proliferation and migration triggered by bFGF, and the smooth-muscle-cell proliferation initiated by PDGF.^19,20 Oral treatment with nebivolol inhibited the formation of intimal hyperplasia in a venous bypass graft model, as in our study.

Calcium dobesilate was reported to protect endothelial functions and increase vascular relaxation in a model of aortic endothelial injury in rabbits.²¹ Dosage-related inhibition of rat aortic smooth-muscle-cell proliferation and DNA synthesis by calcium dobesilate (daily oral dosages, 30–200 mg/kg for 7 days) was shown in a study by Parés-Herbuté and colleagues.²² It was stated that this inhibition was reversible upon removal of the drug, in ex vivo and in vitro models. In the same study, it was shown that calcium dobesilate increases NO production by increasing NO synthase activity in addition to protecting against low-density-lipoprotein cholesterol oxidation.²² We inhibited intimal hyperplasia through the daily oral dosage of 100 mg/kg of calcium dobesilate in our venous graft model, seemingly in agreement with the earlier study.

Yamanouchi and colleagues²³ reported that 10 mg/kg daily of a hydrophilic statin, pravastatin, significantly decreased intimal hyperplasia after 4 weeks in comparison with a control group, probably due to pleiotropic effects, in rabbits whose jugular veins were interpositionally grafted to the carotid artery. We preferred to use atorvastatin, which has well-known pleiotropic effects. In previous reports, 10-mg/kg daily dosages of atorvastatin were thought to be safe and effective,^24–26 but we were obliged to reduce this to 2.5 mg/kg. We observed that this lower dosage of atorvastatin effectively inhibited the development of intimal hyperplasia.

Bledsoe and associates²⁷ reported that a combination of clopidogrel and pravastatin in a rat carotid-endarterectomy model significantly decreased intimal hyperplasia and C-reactive protein levels. In another study,²⁸ clopidogrel alone was not found to inhibit intimal hyperplasia effectively. On the other hand, Herbert and coworkers²⁹ showed a dose-related inhibition of platelet aggregation and smooth-muscle-cell proliferation in a rabbit model of carotid artery injury, after 3 weeks of the daily administration of 25 mg/kg of clopidogrel. Zurbrügg and co-authors³⁰ also reported that the combination of aspirin and clopidogrel inhibited smooth-muscle-cell proliferation and intimal-hyperplasia development more effectively than did aspirin treatment alone, in an experimental venous bypass graft model. Even though dosages of clopidogrel between 1 and 25 mg/kg daily have been reported to be safe and effective,^29,31 our initial administration of 10 mg/kg daily caused death in our subject animals.

External absorbable synthetic materials, when used as barriers for tissue growth, are shown to inhibit neointimal formation and intimal hyperplasia.^9,32 Immunosuppressive antineoplastic agents, such as leflunomide analogue FK778 or paclitaxel, were reported to be effective in the inhibition of intimal hyperplasia.^33,34 Delayed re-endothelization was described as the reason for in-stent thrombosis. Probucol was effective in enhancing re-endothelization and inhibiting intimal hyperplasia.³⁵ Irradiation is known to inhibit atherosclerotic-lesion formation, especially after coronary artery stenting.^36–38

In many clinical and experimental studies, it was emphasized that NO donors can prevent the development of intimal hyperplasia. Kown and colleagues¹¹ showed experimentally that polymers of a NO precursor, L-arginine, can inhibit neointimal hyperplasia formation and intimal–medial thickness ratios in a dose-related fashion.

All of the pharmaceutical agents that we used in our study groups inhibited the development of intimal hyperplasia. This effect might be due to specific or common metabolic pathway alterations. One of the most commonly used new-generation antiaggregant agents, clopidogrel, has also been found to inhibit smooth-muscle-cell proliferation, causing less intimal hyperplasia. Calcium dobesilate, which is chiefly used clinically to treat venous disorders (and which was reported to effect endothelial functional regulation, antiaggregation, and augmentation of NO synthase activity, and to reduce smooth-muscle-cell proliferation), also inhibited intimal hyperplasia in our experimental model. A new-generation β-blocking agent, nebivolol, is known to increase NO secretion and inhibit endothelin-1 release, in addition to its previously reported effects.¹⁹ In our study, we observed its powerful intimal hyperplasia-inhibiting effect. Administration of the cholesterol-lowering drug atorvastatin is almost routine in the treatment of atherosclerotic vascular diseases. Given its other pleiotropic effects (anti-inflammatory, endothelial protector, NO-secretion increase, smooth-muscle-cell proliferation and migration inhibition, and antiaggregation), atorvastatin was also effective in our study.

The treatment of occlusive vascular disease is not flawless, and the high restenosis rate due to intimal hyperplasia during intermediate and long-term follow-up constitutes an ongoing challenge. Clopidogrel, calcium dobesilate, nebivolol, atorvastatin, and other agents should be studied experimentally and clinically in order to gain a better understanding of the pathologic processes that are involved, and, eventually, to achieve a clear solution.

Acknowledgment

We thank Prof. Yuksel Bek of our faculty's Public Health Department for his invaluable help in the statistical analysis of our data.