- Email: [email protected]
Efficacy of corticosteroids in suppression of intimal hyperplasia
Efficacy of corticosteroids in suppression of intimal hyperplasia A r u n Chervu, M D , Wesley S. M o o r e , M D , William J. Quifiones-Baldrich, M D , and T h e o d o r e H e n d e r s o n , BS, Los Angeles, Calif. The effect of steroids and immunosuppression on the development of intimal hyperplasia was studied in 24 New Zealand white rabbits. Staged bilateral arteriotomy and stripping ofintima of the common carotid artery was performed by means o f a 2F balloon catheter under 700 mm Hg of pressure. At 3 months, after the control side artery was harvested (N = 24), the rabbits were assigned to three groups of eight animals each: (1) dexamethasone 0.1 mg/kg, (2) cyclophosphamide (Cytoxan) 1 mg/kg, or (3) azathioprine 1 mg/kg intramus~xtlarly. The animals were treated on the day before the contralateral intimal stripping and then were treated six times a week for 8 weeks; the vessels were harvested at 3 months. Twelve-week patency rates were 62.5% in the control group, 83.3% in the dexamethasone group, 100% in the Cytoxan group, and only 33.3% in the azathioprine group. The ratio of the luminal area to the area enclosed by the internal elastic lamina was used as an index of intimal hyperplasia, with a higher ratio indicating less intimal thickening. This ratio of the patient vessels was 0.74 +-- 0.14 (AT = 15) for the controls, 0.79 - 0.11 (N = 6) for the Cytoxan group, and 0.91 - 0.06 (AT = 5, p < 0.05) for the dexamethasone group, which is statistically significant by means of analysis of variance. Occluded vessels had evidence of organized thrombus without any intimal hyperplasia. The decrease in intimal hyperplasia seen in the steroid group suggests a potential role for steroids in small vessel revascularization, but further studies are required. (J VAsc SURG 1989;10:129-34.)
Intimal hyperplasia is one o f the most important causes o f stenosis and late thrombosis o f arterial reconstructions. The pathophysiologic mechanisms are still unclear. Ross ~ has proposed the response-toinjury hypothesis o f atherosclerosis with intimal hyperplasia as an early lesion. After injury there is a complex interaction o f smooth muscle cells (SMCs), adjoining endothelial" cells, and blood-borne~ele ments such as platelets and white blood cells (WBCs). Extensive work on antiplatelet agents (aspirin, di, pyridamole, nonsteroidal antiinflammatory drugs) has yielded mixed results. 2s White blood cells have also been shown to be deposited on the injured vessel wall. 6,7 For this reason it was hypothesized that the modification o f white blood cell function by corticosteroids and immunosuppressive agents could have a salutary effect in decreasing the hyperplastic response. The purpose of this study is to investigate From the Department of Surgery/Vascular, UCLA Medical Center. Presented at the SVS/ISCVS Forum on Fundamental Problems in Vascular Surgery II, Chicago, Ill., June 15, 1988. Reprint requests: Wesley S. Moore, MD, Department of Surgery/Vascular, UCLA Medical Center, 72-160 CHS, 10833 LeConte Ave., Los Angeles, CA 90024-1749.
this hypothesis i n a balloon catheter model o f intimal hyperplasia. MATERIAL AND METHODS Twenty-four adult New Zealand white rabbits weighing 3 to 4 kg each were anesthetized with 10 to 15 m g / k g chlorpromazine (acepromazine, Chorazine) and 40 to 50 m g / k g ketamine (Ketalar); both were injccted intramuscularly (IM). The animals were given 50 m g / k g cephapirin sodium (Cefadyl) for wound prophylaxis. The common carotid artery was isolated through a right or left longitudinal neck incision without undue dissection from its bed. Microvascular clamps were placed 3 to 4 cm apart, just proximal to the carotid bifurcation. A transverse arteriotomy was made with a No. 11 blade, and a 2F balloon catheter was introduced distally. Intraballoon pressure was monitored with a pressure gauge to obtain a reproducible arterial injury. The balloon was inflated to an opening pressure o f 1200 m m H g and was allowed to stabilize; then three passes were made with the pressure at 700 m m Hg. This pressure was selected from pilot studies in which complete denudation o f the endothelium with intact media and adventitia was noted, Care was taken to rotate the 129
Journal of VASCULAR SURGERY
130 Chervu et al.
Fig. 1. A, Rabbit 15 with intimal hyperplasia at 12 weeks after control stripping of the carotid intima. (Elastica van Gieson stain; original magnification x 10.) B, Higher power view shows cellular composition of the hyperplastic response (arrow points to IEL). Hematoxylin and eosin stain, original magnification x 100.)
Table I. Results of lumen/internal elastic lamina ratio Control Dexamethasone Cyclophosphamide Azathiprine
Mean
SEM
No.
0.737 0.915" 0.787 0.951
0.037 0.025 0.047 --
15 5 6 1
*p < 0.05, analysisof variance, Bonferronipairwisecomparison. catheter after each pass to account for balloon asymmetry. Under the operating microscope (× 10 to x 20). The arteriotomy was closed with three to four interrupted sutures of 10-0 Dermalon (Ethicon, Somerville, N.J,)The microvascular clamps were removed, any bleeding was controlled with slight pressure, and the wound was closed. Twelve weeks after the stripping of the intima of the common carotid artery on the control side, the segments with the denuded intima were excised (2 to 3 cm). After the harvesting of the control specimens, the animals were prepared for the stripping of the intima on the experimental side. One day before the experimental intervention, the animals were
randomized into three treatment groups as follows: group I received dexamethasone 0.1 mg/kg IM; group II received cyclophosphamide (Cytoxan) 1 mg/kg IM; and group III received azathioprine (Imuran) 1 mg/kg IM. The rationale of the use of these drugs will be explained. On the second day of administration of the drugs, the intima was stripped from the common carotid artery on the experimental side in all three groups of animals. The medication was continued for 8 weeks; the doses were given six times per week. Twelve weeks after the completion of the intimal stripping, the animals were killed humanely for removal of the experimental arterial segments. All animal care complied with the "Principles of Laboratory Animal Care" and the "Guide for the Care and Use of Laboratory Animals" (NIH Publication No. 80-23, revised 1978). Both control and experimental harvested vessels were fixed ex vivo by perfusion with 1% glutaraldehyde at 100 mm H g for 10 minutes to maintain anatomic dimensions, and then they were placed in 10% buffered formaldehyde. Serial cross sections were stained with hematoxylin-eosin and Verhoeff van Geison's stain. Photomicrographs of the stained
Volume 10 Number 2 August 1989
Suppression, of initial hyperplasia by steroids 131
Fig. 2. Rabbit 15 treated with dexamethasone shows absence of intimal hyperplasia in (A) low-power view (Elastica van Gieson stain; original magnification × 10) and (B) high-power view (Hematoxylin and eosin stain; original magnification × 100). Arrow points to IEL.
sections were used for planimetric measurement of the cross-sectional area of the lumen and the area enclosed by the internal elastic lamina (IEL). Resuits are expressed as the ratio of the cross-sectional area of the lumen to the area enclosed by the IEL. The higher the ratio, the less the intimal thickening and the morc favorable the result. Statistical analysis was performed by means of Levene's test for equality of variances, one-way analysis of variance (ANOVA), and the Bonferroni multiple comparison test. RESULTS There were no deaths after control carotid intimectomy. There were nine deaths in the experimental groups: Five deaths occurred in the azathioprine group, and two deaths each occurred in the dexamethasone and cyclophosphamide groups. Six of these deaths occurred in the first 48 hours after operation and were probably a result of cerebral ischemia from thrombosis of the one remaining carotid artery. One late death occurred in the dexamethasone group, and two deaths occurred in the azathioprine group from unknown causes. No wound complications were noted.
Twelve-week patency rates were 62.5% in the control group (15 / 24), 83.3% in the dexamethasone group (5/6), 100% in the cyclophosphamide group (6/6), and only 33.3% in the azathioprine group (1/3). The time of occlusion was unknown, but the occluded vessels had evidence of organized thrombus without intimal hyperplasia, which suggested early thrombosis. Occluded vessels were excluded from planimetric analysis. In all groups the segment of the vessel proximal to the arteriotomy showed no intimal hyperplasia, with the maximal response present in the middle of the denuded area. The results of the lumen (L) to IEL ratio (L/IEL) are outlined in Table I (L/IEL of 1.0 indicates no intimal hyperplasia). The L/IEL ratio was 0.737 +_ 0.037 in the control group compared to 0.915 ___ 0.025 in the rabbits treated with dexamethasone, and 0.787 + 0.047 in the cyclophosphamide group. The only patent vessel in the azathioprine group had a ratio of 0.951. Inadequate data excluded this group from further analysis. The L/IEL is significant for the dexamethasone-treated animals as compared with control animals (p < 0.05). No significant difference was noted between the control and cyclophosphamide groups.
Journal of VASCULAR SURGERY
132 Chcrvu et aL
Fig. 31 A, Rabbit 18 with moderate intimal hyperplasia. Elastica van Gieson stain; original magnification × 25.) B, Intima shows a similar cellular response to previous control stripping of the intima (Hematoxylin-eosin stain; original magnification × 250.)Arrow points to IEL. Fig. 1 shows a control rabbit with a moderate amount of intimal hyperplasia after balloon catheter injury. The same rabbit showed no intimal hyperplasia (Fig. 2) when treated with dexamethasone. Figs. 3 and 4 illustrate the results in an animal after a control injury and with cyclophosphamide treatment. DISCUSSION
Balloon catheter injury has been used as a model of intimal hyperplasia, s,9 with reendothclialization postulated to occur from adjacent areas. In the absence of endothelial coverage, migration and proliferation of SMCs has been thought to cause intimal hyperplasia. However, the total number of SMCs stays constant after 2 weeks from initial injury, with continued intimal proliferation caused by accumulation of connective tissue.I° The initial balloon injury causes deposition and aggregation of platelets 3 as well as WBCs (polymorphonuclear leukocytes 6 and monocytes 7) on the denuded surface. This may be due in part to chemotactic factors secreted by the SMCs. 11 It is postulated that the interaction of these cells with elaboration of chemotactic factors and mi-
togens, for example, platelet-derived growth factor 12 and monocyte-derived growth factor, is: may influence SMC hyperplasia. Corticostcroids have been shown to prevent both granulocyte adherence and aggregation in response to chemotactic factors, prirnar@ the complement system. 14'ls Since adherence and aggregation of WBCs are part of the vessel response to injury, there may be a mediating role for the WBC in intimal hyperplasia. If this is the case, then modification of WBC function with corticosteroids could potentially suppress intimal hyperplasia. Our results suggest a significant decrease in the hyperplastic response when the animal is treated with high-dose steroids (equivalent to 200 mg hydrocortisone per day in a 70 kg man) for an 8-week interval. Dexamethasone was chosen because of its high antiinflammatory potency, long half-life (36 to 72 hours), and absence ofmineralocorticoid effects.16 For these regimens to bc clinicaUy useful, further animal studies with decreased doses of steroids and shorter intervals are necessary to avoid the toxicity of adrenocorticosteroids. One previous study 17has investigated the efficacy ofcorticosteroids in a canine femoropopliteal Dacron
Volume 10 Number 2 August 1989
Suppression of initial hyperplasia by steroids 133
Fig. 4. No decrease in the hyperplaStic response is noted in rabbit 18 treated with cyclOphosphamide (A, Elastica Van Gieson stain; original magnification × 25; B, Hematoxylin and eosin stain; original magnification × 250). Arrow points to IEL. arterial bypass graft model, and has found no improvement in intimal hyperplasia over the control animals. However, only one fifth of the dose of the steroid methylprednisolone sodium succinate (1 mg/kg Solu Medrol IM) with a shorter half-life was used every other day. In addition, the drug regimen was started on the first postoperative day rather than preoperatively. This may account for the apparent contradictory results. However, they did report a significant decrease in intimal hyperplasia, with a higher patency rate in the dogs treated with azathioprine 1 mg/kg orally daily. We were not able to substantiate this finding in our study since many of the animals died during the drug regimen. Whereas the two late deaths may have been from drug toxicity, the three early deaths were most likely from thrombosis of the intimectomized vessel. This could be avoided in a model where only one carotid artery is used at the cost of losing the advantages of an internal control. Cyclophosphamide is an alkylating agent in the nitrogen mustard group. Its main cytotoxic effect is against mitotic activity, differentiation, and function of actively proliferating cells. 16 With active prolifer-
ation of SMCs after an intimal injury, low-dose cyclophosphamide was used as an immunosuppressive agent to decrease activity of these cells. However, no decrease in intimal hyperplasia could be documented in our experimental group. The use of corticosteroids has potential benefit in both vein and prosthetic bypass grafts. Pearce et al.18 have shown that vein grafts implanted into the arterial circuit develop a massive, acute inflammatory response with subendothelial and transmural accumulation of neutrophils. This causes endothelial sloughing; the process is worse in low-flow states. They showed that high doses of 0~-methylprednisolone sodium succinate (10 mg) almost completely prevented the inflammatory reaction. It is possible that the use of corticosteroids in our model limited the area of endothelial sloughing in addition to modifying WBC function. Applicability to the clinical arena is yet to be established. In conclusion, in our balloon catheter model of intimal hyperplasia, high-dose steroids significantly reduce the intimal lesion. Further studies are needed to elucidate the mechanisms involved in this suppression of intimal hyperplasia.
134
Journal of VASCULAR SURGERY
Chervu el; al.
We thank Burroughs-Wellcome for their generous contribution o f azathioprine. 10. REFERENCES
1. Ross ~ Glomset JA. The pathogenesis of atherosclerosis: an update. N Engl J Med 1986;314:488-500. 2. McCready KA, Price MA, Kryscio KJ, Hyde GL, Mattingly SS, Griffen WO. Failure of antiplatelet therapy with ibuprofen (Motrin) to prevent neointimal hyperplasia. J VAse St~R~ 1985;2:205-13. 3. Plate G, Stanson AW, Hoilier LH, Dewanjee MK. Drug effects on platelet deposition after endothelial injury of the rabbit aorta. J Surg Res 1985;39:258-66. 4. McCann RL, Hagen PO, Fuchs JCA. Aspirin and dipyridamole decrease intimal hyperplasia in experimental vein grafts. Ann Surg 1980;191:238-43. 5. Quinones-Baldrich WQ, Ziomek S, Henderson T, Moore WS. Patency and intimal hyperplasia: the effect of aspirin on small arterial anastomosis. Ann Vase Surg 1988;2!50-6. 6. Cole CW, Lucas IF, Mikat EM, Hagen PO, McCann ILL. Adherence ofpolymorphonuclear leukocytes to injured rabbit aorta. Surg Forum 1984;35:440-2. 7. Lucas JF, Makhoul RG, Cole CW, Mikat EM, McCann RL, Hagen PO. Mononuclear cells adhere to sites of vascular balloon catheter injury. Curr Surg 1986;43:112-5. 8. Stemerman MB, Spaet TH, Pitlick F, Cintron J, Lejnieks I, Tiell ML. Intimal healing: the pattern of reendothelialization and intimal thickening. Am J Pathol 1977;87:125-37. 9. Christensen BC, Chenmitz J, Tkocz I, Kim CM. Repair in
11.
12.
13.
14.
15.
16.
17.
18.
arterial tissue. Acta Pathol Microbiol Immunol Scand [A] 1979;87:275-83. Clowes AW, Reidy MA, Clowes MM. Kinetics of cellular proliferation after arterial inujury: I. Smooth muscle growth in the absence of endothelitmq. Lab Invest 1983;49:327-33. Mazzone T, Jensen M, Chait A. Human arterial wall cells secrete factors that are chemotactic for monocytes. Proc Nail Acad Sci USA 1983;80:5094-7. Ross R, Glomset JA, Kariya B, Harker L. A plateletdependent serum factor that stimulates the proliferation of arterial smooth muscle ceils in vitro. Proc Natl Acad Sci USA 1974;71:1207-10. Glenn KC, Ross R. Human monocyte-derived growth factor(s) for mesenchymal cells: activation of secretion by endotoxin and concanavalin A. Ceil 1981;25:603-15. Oseas RS, Allen J, Yang HH, Baehner RL, Boxer LA. Mechanism of dexamethasone inhibition of chemotactic factor induced granulocyte aggregation. Blood 1982;59:265-9. Hammerschmidt DE, White IG, Craddock PR, Jacob HS. Corticosteroids inhibit complement-induced granulocyte aggregation. J Clin Invest 1979;63:798-803. Gilman AG, Goodman LS, Gilman A, eds. The pharmacologic basis of therapeutics. New York: MacMillan Publishing Co, 1980:1256-68, 1470-96. Hoepp LM, Elbadawi A, Cohn M, Dachelet R, Peterson C, DeWeese JA. Steroids and immunosuppression: effect of anastomotic intimal hyperplasia in femoral arterial Dacron bypass grafts. Arch Surg 1979;114:273-6. Pearce JE, Dujovny M, Ho KL, et al. Acute inflammation and endothelial injury in vein grafts. Neurosurgery 1985; 17:626-34.
Intimal hyperplasia is one o f the most important causes o f stenosis and late thrombosis o f arterial reconstructions. The pathophysiologic mechanisms are still unclear. Ross ~ has proposed the response-toinjury hypothesis o f atherosclerosis with intimal hyperplasia as an early lesion. After injury there is a complex interaction o f smooth muscle cells (SMCs), adjoining endothelial" cells, and blood-borne~ele ments such as platelets and white blood cells (WBCs). Extensive work on antiplatelet agents (aspirin, di, pyridamole, nonsteroidal antiinflammatory drugs) has yielded mixed results. 2s White blood cells have also been shown to be deposited on the injured vessel wall. 6,7 For this reason it was hypothesized that the modification o f white blood cell function by corticosteroids and immunosuppressive agents could have a salutary effect in decreasing the hyperplastic response. The purpose of this study is to investigate From the Department of Surgery/Vascular, UCLA Medical Center. Presented at the SVS/ISCVS Forum on Fundamental Problems in Vascular Surgery II, Chicago, Ill., June 15, 1988. Reprint requests: Wesley S. Moore, MD, Department of Surgery/Vascular, UCLA Medical Center, 72-160 CHS, 10833 LeConte Ave., Los Angeles, CA 90024-1749.
this hypothesis i n a balloon catheter model o f intimal hyperplasia. MATERIAL AND METHODS Twenty-four adult New Zealand white rabbits weighing 3 to 4 kg each were anesthetized with 10 to 15 m g / k g chlorpromazine (acepromazine, Chorazine) and 40 to 50 m g / k g ketamine (Ketalar); both were injccted intramuscularly (IM). The animals were given 50 m g / k g cephapirin sodium (Cefadyl) for wound prophylaxis. The common carotid artery was isolated through a right or left longitudinal neck incision without undue dissection from its bed. Microvascular clamps were placed 3 to 4 cm apart, just proximal to the carotid bifurcation. A transverse arteriotomy was made with a No. 11 blade, and a 2F balloon catheter was introduced distally. Intraballoon pressure was monitored with a pressure gauge to obtain a reproducible arterial injury. The balloon was inflated to an opening pressure o f 1200 m m H g and was allowed to stabilize; then three passes were made with the pressure at 700 m m Hg. This pressure was selected from pilot studies in which complete denudation o f the endothelium with intact media and adventitia was noted, Care was taken to rotate the 129
Journal of VASCULAR SURGERY
130 Chervu et al.
Fig. 1. A, Rabbit 15 with intimal hyperplasia at 12 weeks after control stripping of the carotid intima. (Elastica van Gieson stain; original magnification x 10.) B, Higher power view shows cellular composition of the hyperplastic response (arrow points to IEL). Hematoxylin and eosin stain, original magnification x 100.)
Table I. Results of lumen/internal elastic lamina ratio Control Dexamethasone Cyclophosphamide Azathiprine
Mean
SEM
No.
0.737 0.915" 0.787 0.951
0.037 0.025 0.047 --
15 5 6 1
*p < 0.05, analysisof variance, Bonferronipairwisecomparison. catheter after each pass to account for balloon asymmetry. Under the operating microscope (× 10 to x 20). The arteriotomy was closed with three to four interrupted sutures of 10-0 Dermalon (Ethicon, Somerville, N.J,)The microvascular clamps were removed, any bleeding was controlled with slight pressure, and the wound was closed. Twelve weeks after the stripping of the intima of the common carotid artery on the control side, the segments with the denuded intima were excised (2 to 3 cm). After the harvesting of the control specimens, the animals were prepared for the stripping of the intima on the experimental side. One day before the experimental intervention, the animals were
randomized into three treatment groups as follows: group I received dexamethasone 0.1 mg/kg IM; group II received cyclophosphamide (Cytoxan) 1 mg/kg IM; and group III received azathioprine (Imuran) 1 mg/kg IM. The rationale of the use of these drugs will be explained. On the second day of administration of the drugs, the intima was stripped from the common carotid artery on the experimental side in all three groups of animals. The medication was continued for 8 weeks; the doses were given six times per week. Twelve weeks after the completion of the intimal stripping, the animals were killed humanely for removal of the experimental arterial segments. All animal care complied with the "Principles of Laboratory Animal Care" and the "Guide for the Care and Use of Laboratory Animals" (NIH Publication No. 80-23, revised 1978). Both control and experimental harvested vessels were fixed ex vivo by perfusion with 1% glutaraldehyde at 100 mm H g for 10 minutes to maintain anatomic dimensions, and then they were placed in 10% buffered formaldehyde. Serial cross sections were stained with hematoxylin-eosin and Verhoeff van Geison's stain. Photomicrographs of the stained
Volume 10 Number 2 August 1989
Suppression, of initial hyperplasia by steroids 131
Fig. 2. Rabbit 15 treated with dexamethasone shows absence of intimal hyperplasia in (A) low-power view (Elastica van Gieson stain; original magnification × 10) and (B) high-power view (Hematoxylin and eosin stain; original magnification × 100). Arrow points to IEL.
sections were used for planimetric measurement of the cross-sectional area of the lumen and the area enclosed by the internal elastic lamina (IEL). Resuits are expressed as the ratio of the cross-sectional area of the lumen to the area enclosed by the IEL. The higher the ratio, the less the intimal thickening and the morc favorable the result. Statistical analysis was performed by means of Levene's test for equality of variances, one-way analysis of variance (ANOVA), and the Bonferroni multiple comparison test. RESULTS There were no deaths after control carotid intimectomy. There were nine deaths in the experimental groups: Five deaths occurred in the azathioprine group, and two deaths each occurred in the dexamethasone and cyclophosphamide groups. Six of these deaths occurred in the first 48 hours after operation and were probably a result of cerebral ischemia from thrombosis of the one remaining carotid artery. One late death occurred in the dexamethasone group, and two deaths occurred in the azathioprine group from unknown causes. No wound complications were noted.
Twelve-week patency rates were 62.5% in the control group (15 / 24), 83.3% in the dexamethasone group (5/6), 100% in the cyclophosphamide group (6/6), and only 33.3% in the azathioprine group (1/3). The time of occlusion was unknown, but the occluded vessels had evidence of organized thrombus without intimal hyperplasia, which suggested early thrombosis. Occluded vessels were excluded from planimetric analysis. In all groups the segment of the vessel proximal to the arteriotomy showed no intimal hyperplasia, with the maximal response present in the middle of the denuded area. The results of the lumen (L) to IEL ratio (L/IEL) are outlined in Table I (L/IEL of 1.0 indicates no intimal hyperplasia). The L/IEL ratio was 0.737 +_ 0.037 in the control group compared to 0.915 ___ 0.025 in the rabbits treated with dexamethasone, and 0.787 + 0.047 in the cyclophosphamide group. The only patent vessel in the azathioprine group had a ratio of 0.951. Inadequate data excluded this group from further analysis. The L/IEL is significant for the dexamethasone-treated animals as compared with control animals (p < 0.05). No significant difference was noted between the control and cyclophosphamide groups.
Journal of VASCULAR SURGERY
132 Chcrvu et aL
Fig. 31 A, Rabbit 18 with moderate intimal hyperplasia. Elastica van Gieson stain; original magnification × 25.) B, Intima shows a similar cellular response to previous control stripping of the intima (Hematoxylin-eosin stain; original magnification × 250.)Arrow points to IEL. Fig. 1 shows a control rabbit with a moderate amount of intimal hyperplasia after balloon catheter injury. The same rabbit showed no intimal hyperplasia (Fig. 2) when treated with dexamethasone. Figs. 3 and 4 illustrate the results in an animal after a control injury and with cyclophosphamide treatment. DISCUSSION
Balloon catheter injury has been used as a model of intimal hyperplasia, s,9 with reendothclialization postulated to occur from adjacent areas. In the absence of endothelial coverage, migration and proliferation of SMCs has been thought to cause intimal hyperplasia. However, the total number of SMCs stays constant after 2 weeks from initial injury, with continued intimal proliferation caused by accumulation of connective tissue.I° The initial balloon injury causes deposition and aggregation of platelets 3 as well as WBCs (polymorphonuclear leukocytes 6 and monocytes 7) on the denuded surface. This may be due in part to chemotactic factors secreted by the SMCs. 11 It is postulated that the interaction of these cells with elaboration of chemotactic factors and mi-
togens, for example, platelet-derived growth factor 12 and monocyte-derived growth factor, is: may influence SMC hyperplasia. Corticostcroids have been shown to prevent both granulocyte adherence and aggregation in response to chemotactic factors, prirnar@ the complement system. 14'ls Since adherence and aggregation of WBCs are part of the vessel response to injury, there may be a mediating role for the WBC in intimal hyperplasia. If this is the case, then modification of WBC function with corticosteroids could potentially suppress intimal hyperplasia. Our results suggest a significant decrease in the hyperplastic response when the animal is treated with high-dose steroids (equivalent to 200 mg hydrocortisone per day in a 70 kg man) for an 8-week interval. Dexamethasone was chosen because of its high antiinflammatory potency, long half-life (36 to 72 hours), and absence ofmineralocorticoid effects.16 For these regimens to bc clinicaUy useful, further animal studies with decreased doses of steroids and shorter intervals are necessary to avoid the toxicity of adrenocorticosteroids. One previous study 17has investigated the efficacy ofcorticosteroids in a canine femoropopliteal Dacron
Volume 10 Number 2 August 1989
Suppression of initial hyperplasia by steroids 133
Fig. 4. No decrease in the hyperplaStic response is noted in rabbit 18 treated with cyclOphosphamide (A, Elastica Van Gieson stain; original magnification × 25; B, Hematoxylin and eosin stain; original magnification × 250). Arrow points to IEL. arterial bypass graft model, and has found no improvement in intimal hyperplasia over the control animals. However, only one fifth of the dose of the steroid methylprednisolone sodium succinate (1 mg/kg Solu Medrol IM) with a shorter half-life was used every other day. In addition, the drug regimen was started on the first postoperative day rather than preoperatively. This may account for the apparent contradictory results. However, they did report a significant decrease in intimal hyperplasia, with a higher patency rate in the dogs treated with azathioprine 1 mg/kg orally daily. We were not able to substantiate this finding in our study since many of the animals died during the drug regimen. Whereas the two late deaths may have been from drug toxicity, the three early deaths were most likely from thrombosis of the intimectomized vessel. This could be avoided in a model where only one carotid artery is used at the cost of losing the advantages of an internal control. Cyclophosphamide is an alkylating agent in the nitrogen mustard group. Its main cytotoxic effect is against mitotic activity, differentiation, and function of actively proliferating cells. 16 With active prolifer-
ation of SMCs after an intimal injury, low-dose cyclophosphamide was used as an immunosuppressive agent to decrease activity of these cells. However, no decrease in intimal hyperplasia could be documented in our experimental group. The use of corticosteroids has potential benefit in both vein and prosthetic bypass grafts. Pearce et al.18 have shown that vein grafts implanted into the arterial circuit develop a massive, acute inflammatory response with subendothelial and transmural accumulation of neutrophils. This causes endothelial sloughing; the process is worse in low-flow states. They showed that high doses of 0~-methylprednisolone sodium succinate (10 mg) almost completely prevented the inflammatory reaction. It is possible that the use of corticosteroids in our model limited the area of endothelial sloughing in addition to modifying WBC function. Applicability to the clinical arena is yet to be established. In conclusion, in our balloon catheter model of intimal hyperplasia, high-dose steroids significantly reduce the intimal lesion. Further studies are needed to elucidate the mechanisms involved in this suppression of intimal hyperplasia.
134
Journal of VASCULAR SURGERY
Chervu el; al.
We thank Burroughs-Wellcome for their generous contribution o f azathioprine. 10. REFERENCES
1. Ross ~ Glomset JA. The pathogenesis of atherosclerosis: an update. N Engl J Med 1986;314:488-500. 2. McCready KA, Price MA, Kryscio KJ, Hyde GL, Mattingly SS, Griffen WO. Failure of antiplatelet therapy with ibuprofen (Motrin) to prevent neointimal hyperplasia. J VAse St~R~ 1985;2:205-13. 3. Plate G, Stanson AW, Hoilier LH, Dewanjee MK. Drug effects on platelet deposition after endothelial injury of the rabbit aorta. J Surg Res 1985;39:258-66. 4. McCann RL, Hagen PO, Fuchs JCA. Aspirin and dipyridamole decrease intimal hyperplasia in experimental vein grafts. Ann Surg 1980;191:238-43. 5. Quinones-Baldrich WQ, Ziomek S, Henderson T, Moore WS. Patency and intimal hyperplasia: the effect of aspirin on small arterial anastomosis. Ann Vase Surg 1988;2!50-6. 6. Cole CW, Lucas IF, Mikat EM, Hagen PO, McCann ILL. Adherence ofpolymorphonuclear leukocytes to injured rabbit aorta. Surg Forum 1984;35:440-2. 7. Lucas JF, Makhoul RG, Cole CW, Mikat EM, McCann RL, Hagen PO. Mononuclear cells adhere to sites of vascular balloon catheter injury. Curr Surg 1986;43:112-5. 8. Stemerman MB, Spaet TH, Pitlick F, Cintron J, Lejnieks I, Tiell ML. Intimal healing: the pattern of reendothelialization and intimal thickening. Am J Pathol 1977;87:125-37. 9. Christensen BC, Chenmitz J, Tkocz I, Kim CM. Repair in
11.
12.
13.
14.
15.
16.
17.
18.
arterial tissue. Acta Pathol Microbiol Immunol Scand [A] 1979;87:275-83. Clowes AW, Reidy MA, Clowes MM. Kinetics of cellular proliferation after arterial inujury: I. Smooth muscle growth in the absence of endothelitmq. Lab Invest 1983;49:327-33. Mazzone T, Jensen M, Chait A. Human arterial wall cells secrete factors that are chemotactic for monocytes. Proc Nail Acad Sci USA 1983;80:5094-7. Ross R, Glomset JA, Kariya B, Harker L. A plateletdependent serum factor that stimulates the proliferation of arterial smooth muscle ceils in vitro. Proc Natl Acad Sci USA 1974;71:1207-10. Glenn KC, Ross R. Human monocyte-derived growth factor(s) for mesenchymal cells: activation of secretion by endotoxin and concanavalin A. Ceil 1981;25:603-15. Oseas RS, Allen J, Yang HH, Baehner RL, Boxer LA. Mechanism of dexamethasone inhibition of chemotactic factor induced granulocyte aggregation. Blood 1982;59:265-9. Hammerschmidt DE, White IG, Craddock PR, Jacob HS. Corticosteroids inhibit complement-induced granulocyte aggregation. J Clin Invest 1979;63:798-803. Gilman AG, Goodman LS, Gilman A, eds. The pharmacologic basis of therapeutics. New York: MacMillan Publishing Co, 1980:1256-68, 1470-96. Hoepp LM, Elbadawi A, Cohn M, Dachelet R, Peterson C, DeWeese JA. Steroids and immunosuppression: effect of anastomotic intimal hyperplasia in femoral arterial Dacron bypass grafts. Arch Surg 1979;114:273-6. Pearce JE, Dujovny M, Ho KL, et al. Acute inflammation and endothelial injury in vein grafts. Neurosurgery 1985; 17:626-34.