Reprinted article "Pathophysiology of vein graft failure: a review".

Vein bypass grafting is an integral component of cardiovascular surgical practice for both arterial and venous diseases. However, many of these grafts will eventually fail due to either intrinsic or extrinsic causes. This review examines the current understanding and knowledge of venous histology, vein graft pathology and the associated endothelial and smooth muscle cell physiology and pharmacology. In addition, the status of research on the therapeutic control of vein graft intimal hyperplasia and accelerated atherosclerosis is assessed.

Phospholipids of bacteria with extensive intracytoplasmic membranes.

Examination of the lipids of three species of nonphotosynthetic bacteria with extensive internal membranes revealed phosphatidyl choline (lecithin) in two species. In one of these there was an unusual accumulation of phosphatidyl N-dimethylethanolamine. (The relation between lecithin and membrane elaboration in microorganisms is discussed.)

The expression and function of G-proteins in experimental intimal hyperplasia.

G-proteins are membrane-bound signal transduction proteins which couple extracellular receptor signals to various effectors. This study examines the expression and the function of G-proteins (alpha i, alpha s, alpha q, and alpha o) in experimental intimal hyperplasia. Vein bypass grafts were placed in 30 New Zealand White rabbits and were harvested after 28 d. The contralateral jugular veins served as controls. Isometric tension studies were performed on rings from veins and vein grafts (n = 10), and Western blot and mRNA analyses were performed in another 20 vessels. There was a fivefold increase in alpha q, a 2.7-fold increase in the alpha i2, and a 3.3-fold increase in alpha s expressions in vein grafts compared with veins. Detectable expression of alpha i3 was observed in vein grafts but not in jugular veins. In addition, there was a 3.8-fold increase in beta subunits in the vein grafts compared with the veins. mRNA for alpha s, alpha i3, and alpha i2 were all elevated in the vein grafts. No detectable levels of the alpha i1 protein or its mRNA were present in either veins or vein grafts. Contractile responses in the veins were not inhibited by pertussis toxin. The contractile responses to norepinephrine were enhanced by twofold, and the responses to serotonin developed de novo in vein grafts compared with veins. The contractile responses to both norepinephrine and serotonin were only partially inhibited by pertussis toxin in the vein grafts even though there was 100% ADP ribosylation with pertussis toxin in both veins and vein grafts. These data suggest that intimal hyperplasia is associated with increased or novel expression of G-proteins in vivo which occur simultaneously with the development of pertussis toxin-sensitive contractile responses. Changes in G-proteins at a transcriptional level or at the level of RNA stability may be involved in the response of smooth muscle cells to injury and to intimal hyperplasia formation.

Remodeling of an acellular collagen graft into a physiologically responsive neovessel.

Surgical treatment of vascular disease has become common, creating the need for a readily available, small-diameter vascular graft. However, the use of synthetic materials is limited to grafts larger than 5-6 mm because of the frequency of occlusion observed with smaller-diameter prosthetics. An alternative to synthetic materials would be a biomaterial that could be used in the design of a tissue-engineered graft. We demonstrate that a small-diameter (4 mm) graft constructed from a collagen biomaterial derived from the submucosa of the small intestine and type I bovine collagen has the potential to integrate into the host tissue and provide a scaffold for remodeling into a functional blood vessel. The results obtained using a rabbit arterial bypass model have shown excellent hemostasis and patency. Furthermore, within three months after implantation, the collagen grafts were remodeled into cellularized vessels that exhibited physiological activity in response to vasoactive agents.

Efficient adenoviral gene transfer to early venous bypass grafts: comparison with native vessels.

OBJECTIVES: Gene therapy may provide new approaches to reduce vein graft failure following coronary or peripheral bypass surgery. The aim of this study was to investigate the relative efficacy of intraoperative adenoviral gene transfer to vein grafts, comparing transgene expression in vein grafts with that in matched native vessels in the same animal. In addition, we assessed the impact of bypass grafting on the cellular targets of gene transfer. METHODS: New Zealand White rabbits underwent interposition bypass grafting of the carotid artery, using the ipsilateral external jugular vein, which was infected with an adenovirus expressing beta-galactosidase immediately prior to bypass grafting (n = 16). The contralateral native jugular vein (n = 16) and carotid artery (n = 8) were infected concurrently with the same adenoviral preparation. After 3, 7 or 14 days, beta-galactosidase protein expression was quantified by ELISA, and specific cell types expressing beta-galactosidase were identified by X-Gal staining and by immunohistochemistry. RESULTS: After 3 days, endothelial cells were efficiently transduced in all vessels; medial smooth muscle cells were transduced infrequently. In contrast to jugular veins after gene transfer, endothelium in vein grafts showed expression of VCAM-1 and ICAM-1, and intense inflammation with CD18+ leukocytes. Transgene expression in vein grafts at day 3 was maintained at levels approximately 50% of that in ungrafted jugular veins, but continued to decrease through day 7. CONCLUSIONS: Although vascular injury in early venous bypass grafts reduces gene transfer efficacy, significant transgene expression is maintained for at least 7 days. These findings have important implications for intraoperative gene transfer strategies in vein grafts.

Effects of rabbit platelets on vascular smooth muscle cell DNA synthesis.

During platelet aggregation a factor is released which stimulates DNA synthesis in vascular smooth muscle cells. We have examined the capacity of plasma free suspensions of intact rabbit platelets to stimulate DNA synthesis in vascular smooth muscle cells in tissue culture. Platelet-rich plasma was obtained by low-speed centrifugation of citrated blood and passed through sterile columns of sepharose 2B beads. The resultant plasma-free platelet suspension was adjusted to 400,000 platelets/mm3. An aliquot was completely aggregated with thrombin, centrifuged and the supernatant examined for ability to stimulate incorporation of [3H]thyrmidine into DNA by cultured vascular smooth muscle cells. Sonication of an aliquot of platelet suspension released an equivalent amount of this stimulating factor. Another aliquot of platelet suspension of identical volume was added directly to the test smooth muscle cells. Stimulation by intact platelets co-cultured with smooth muscle cells was more than twice as great as that of the thrombin-aggregated supernatant. The latter was not increased by increasing the thrombin concentration nor by increasing the time of exposure. The potency of intact platelets was decreased only slightly by segregating them in Millipore chambers indicating the direct contact was not responsible for its enhancement.

Association of polymorphonuclear leukocytes with sites of aortic catheter-induced injury in rabbits.

The kinetics of the association of polymorphonuclear leukocytes (PMNs) with arterial balloon catheter-induced injury have been examined. An average of 6 X 10(7) PMNs were isolated from 20 ml of blood and labelled with 111In-oxine for reinfusion into the donor rabbit. The cells remained viable as demonstrated by both in vitro and in vivo tests of cell function. The abdominal aorta of rabbits was denuded of endothelium and immediately, 24 h, or 5 weeks later, exposed to autologous radiolabelled PMNs for 1 h. The presence of PMNs at sites of denudation was demonstrated by detection of the radioactive label and was confirmed by light and electron microscopy after 24 h, but not at 5 weeks. Immediately following denudation radioactivity was 2.44 +/- 0.33 times control (P = 0.006); 2.52 +/- 0.18 at 24 h (P = 0.005); and 1.88 +/- 0.32 times control at 5 weeks (P = 0.045). The presence of PMNs, or their products, 5 weeks after denudation suggests a more complex role of PMNs and possibly a direct involvement in the long term changes resulting from arterial balloon catheter injury.

Postoperative reduction of high serum cholesterol concentrations and experimental vein bypass grafts. Effect on the development of intimal hyperplasia and abnormal vasomotor function.

Hypercholesterolemia is an important contributor to the development of intimal hyperplasia and superimposed accelerated atherosclerosis in vein bypass grafts. This study examines the effect of dietary modification of serum cholesterol on the development of intimal hyperplasia and vasomotor function of vein grafts. Thirty male New Zealand White rabbits had a right carotid vein bypass graft and were put to death 28 days after the operation. Twenty animals received a 1% cholesterol diet for 4 weeks before the operation. In 10 animals this diet was continued until harvest (hypercholesterolemia group). In another 10 animals the diet was changed to standard rabbit chow on the day of the surgical procedure and continued until harvest (cholesterol reduction group). The last 10 animals were control subjects. Vein grafts were harvested either for histologic study or for in vitro isometric tension studies. Cumulative dose response curves to norepinephrine, serotonin, bradykinin, and endothelin-1 were determined. After in situ pressure fixation, intimal thicknesses of the vein grafts were measured by videomorphometry. The change in diet produced a 74% reduction in serum cholesterol concentration within 28 days. There was a 26% reduction in the intimal thickness of vein graft intimal hyperplasia and the macroscopic disappearance of atheromatous lesions from the graft wall, which are always observed in vein grafts from the hypercholesterolemia group. Cholesterol reduction did not change hypercholesterolemia-induced agonist supersensitivity. Therefore, cholesterol reduction slows the formation of intimal hyperplasia in vein grafts but does not prevent the persistence of the hypercholesterolemia-associated smooth muscle phenotype.

Time course of the regression of intimal hyperplasia in experimental vein grafts.

A previous study in which vein grafts were removed from the arterial circulation and reimplanted into the venous circulation of the same animal demonstrated regression of vein graft intimal hyperplasia and medial thickening within 14 days. The present study was designed to characterize the kinetics of the morphological and ultrastructural changes over this 14-day period. Twenty-one male New Zealand White rabbits received a reversed vein interposition bypass graft of the right common carotid artery. Fourteen days after the procedure, 21 vein grafts were isolated, removed, and reimplanted into the contralateral external jugular venous system as veno-venous interposition bypass grafts (reversal grafts). The grafts were harvested at 60 minutes, 1 day, 3 days, 5 days, 7 days, and 14 days after reversal. Before insertion into the venous circulation, the vein graft had a confluent endothelial cell surface with multiple layers of smooth muscle cells representing intimal hyperplasia. After 1 hour, the reversal graft retained an intact endothelial cell layer with no evidence of tissue edema or cellular disruption. By 24 hours, there were a few blood cells on the endothelial cell surface. There was no inflammatory infiltrate seen in the subendothelium, and the smooth muscle cells were unaltered. At 3 days, the endothelial cell lining remained intact with no polymorphonucleocytes in the subendothelium or within the graft wall. Underlying smooth muscle cells at this time were noted to contain cytoplasmic vacuoles. At 5 days, there were no inflammatory cells seen on the surface or within the vein graft wall, but many of the underlying smooth muscle cells within the intimal hyperplasia were noted to be fragmented and to have clumping of chromatin. After 7 days, the endothelial cells remained intact and there was widespread evidence of apoptosis beneath the subendothelium with highly fragmented smooth muscle cells, some of which were histologically in the process of breaking up. At 14 days, the grafts retained uniform endothelial cell surfaces. Most of the smooth muscle cells that composed the intimal hyperplasia seen before implantation as a reversal graft were gone. Areas of newly laid down collagen could be observed. There were no acute inflammatory cells but for some mast cells seen in the graft wall. This study demonstrates that in this model, regression of intimal hyperplasia was associated with apoptosis of the smooth muscle cells and the deposition of collagen. There was no evidence that this process is mediated by an acute inflammatory response. Regression therefore appears to be due to induction of smooth muscle cell apoptosis by either a reduction in pressure or flow or a combination of both factors. The findings will enable a systematic cellular and molecular analysis of the biology of regression, which may afford clues to better understand the biology of the developing intimal hyperplasia.

Regression of intimal hyperplasia with restoration of endothelium-dependent relaxing factor-mediated relaxation in experimental vein grafts.

BACKGROUND: The reversibility of the morphologic and functional alterations that occur in veins transplanted into the arterial circulation was examined in this study. METHODS: Common carotid vein bypass grafts (VG) were performed in 20 male New Zealand White rabbits. Ten VG and jugular veins (CV) were harvested after 14 days, and ten VG were reimplanted as venovenous bypass grafts (REV) and harvested after an additional 14 days. Vessels were taken for structural or isometric tension studies to norepinephrine, serotonin, and bradykinin and to acetylcholine and sodium nitroprusside after precontraction. RESULTS: There was a decrease in the thickness of the intima (p = 0.02) and the media (p = 0.002) in REV compared with VG. In REV, sensitivity to norepinephrine decreased (p = 0.0007) with a reduced maximal tension to norepinephrine (p = 0.02) and to serotonin (p = 0.0001). Bradykinin sensitivity increased in REV (p = 0.003 vs VG) and was greater than in CV. Only the precontracted CV and REV relaxed to acetylcholine. All tissues relaxed to sodium nitroprusside. CONCLUSIONS: This study suggests that intimal hyperplasia can be reversed with restoration of endothelium-dependent relaxing factor-mediated relaxation but that only a partial regression of the contractile abnormalities can be achieved.

Effects of distention and short-term grafting on vasoreactivity in rabbit external jugular veins.

The relative effects of distention, intraluminal pressure, and wall tension on venous smooth muscle and endothelial cell function were examined in 40 external jugular veins from New Zealand white rabbits. Vein grafts (n = 5) were interposed in the common carotid artery and explanted after 10 minutes. Distended veins were inflated in vitro with modified Krebs' solution at 37 degrees C for 10 minutes at pressures of either 20 mm Hg (D-20; n = 5) or 80 mm Hg (D-80; n = 5). Externally supported veins (ES-80; n = 5) were inflated at 80 mm Hg pressure, but distention was prevented by covering with a 3 mm internal diameter polytetrafluoroethylene sleeve. Bradykinin-induced in vitro maximal tension was attenuated significantly in vein grafts (0.13 +/- 0.04 g) and D-80 rings (0.27 +/- 0.07 g) compared with D-20 rings (1.20 +/- 0.14 g), ES-80 rings (0.99 +/- 0.13 g), or nondistended control rings (n = 40; 1.19 +/- 0.10 g; p less than 0.001). The attenuation in contraction in the vein graft and D-80 groups was nonspecific (i.e., similar results were obtained with respect to other smooth muscle agonists). Contractile function was inversely associated with wall tension, the product of pressure and radius (r2 = 0.7438; p = 0.06). In contrast, there were no differences in endothelium-dependent or endothelium-independent relaxation among the five groups. It is concluded that, in this experimental system, (1) venous smooth muscle function is significantly attenuated after short-term in vitro distention or grafting although endothelial function is largely preserved, and (2) the decrement in contraction is due to elevated wall tension.

Reduction of experimental vein graft intimal hyperplasia and preservation of nitric oxide-mediated relaxation by the nitric oxide precursor L-arginine.

BACKGROUND: Previous studies in animals and human beings have shown that vein bypass grafts exhibit diminished endothelium-dependent relaxation and the development of intimal hyperplasia. This study examines the effect of L-arginine on experimental vein graft endothelial cell function and the development of intimal hyperplasia. METHODS: Common carotid vein bypass grafts were performed in 24 New Zealand White rabbits: 12 were controls and 12 received L-arginine (2.25%) orally 7 days before operation and thereafter until harvest 28 days after operation. Intimal and medial dimensions were determined by planimetry on pressure-fixed vessels. Relaxation to acetylcholine, serotonin, calcium ionophore (A23187), and sodium nitroprusside was performed on precontracted vessel rings. RESULTS: Arginine-treated vein grafts showed a 47% reduction in mean intimal thickness (p < 0.001) compared with controls. By scanning and transmission electron microscopy, all vein grafts showed a confluent endothelium. In contrast to control grafts, which do not relax to acetylcholine and serotonin, arginine-treated vein grafts relaxed in response to both agonists. There was a significant increase (p < 0.05) in the maximal relaxation to calcium ionophore (A23187) in arginine-treated vein grafts compared with control grafts. Non-endothelium-dependent responses to sodium nitroprusside were equivalent in all vein grafts. CONCLUSIONS: This study shows that oral L-arginine supplementation significantly reduces intimal hyperplasia and preserves nitric oxide-mediated relaxation in experimental vein grafts, suggesting a role for nitric oxide in the regulation of the cellular events that lead to intimal hyperplasia.

Adenoviral-mediated inhibition of G beta gamma signaling limits the hyperplastic response in experimental vein grafts.

BACKGROUND: Vein graft intimal hyperplasia is associated with changes in G protein expression. The carboxyl terminus of the beta-adrenergic receptor kinase-1 (beta ARKCT) is known to inhibit G beta gamma-mediated mitogen-activated signaling pathways. This study examines the effects of adenoviral-mediated beta ARKCT infection on the development of intimal hyperplasia in vein grafts. METHODS: New Zealand White rabbits underwent bypass grafting of the carotid artery with the jugular vein. Vein grafts were infected with adenoviral vectors encoding for beta ARKCT (n = 19), beta-galactosidase (n = 3), or empty viral constructs (n = 12). In control animals, vein grafting was performed without infection (n = 10). RESULTS: The efficacy of beta ARKCT infection in vein grafts was verified by reverse transcriptase-polymerase chain reaction. X-gal staining of beta-galactosidase-infected vein grafts demonstrated the transgene in cells throughout the vessel wall. Adenoviral infection of vein grafts without gene transfer did not alter wall thicknesses or sensitivities to contractile agonists, compared with control grafts. beta ARKCT infection, however, reduced intimal thickness by 36% (P < .001) and medial thickness by 24% (P < .001), compared with empty viral infection. beta ARKCT-infected vein grafts also demonstrated increased sensitivity in response to contractile agonists. CONCLUSIONS: These results show that inhibition of G beta gamma signaling with adenoviral-mediated beta ARKCT in vivo infection effectively modifies the structural and functional hyperplastic abnormalities in vein grafts.

Localized versus systemic angiotensin II receptor inhibition of intimal hyperplasia in experimental vein grafts by the specific angiotensin II receptor inhibitor L158,809.

BACKGROUND: This study examines the effect of the angiotensin II receptor (type 1) antagonist (L158,809) on the formation of vein graft intimal hyperplasia in vivo, by both localized and systemic administration. METHODS: Forty New Zealand White rabbits underwent carotid interposition bypass grafting with the external jugular vein and were killed on postoperative day 28. To determine the effect of L158,809 on the development of intimal hyperplasia, 10 animals received long-term oral therapy with L158,809 (10 mg/kg/day, begun 5 days before operation and continued until harvest), 10 animals underwent coating of the grafts with a pluronic gel containing L158,809 (10(-5) mol/L), and 20 animals were controls (10 with and 10 without pluronic gel). These grafts were harvested for either histologic analysis (n = 6 per group) or in vitro isometric tension studies to angiotensin II (n = 4 per group). RESULTS: Long-term oral treatment with L158,809 produced a 43% decrease in intimal thickness from 76 +/- 6 microns (mean +/- SEM) in the control animals to 43 +/- 7 microns in the treated vein grafts (p = 0.002). There was also a significant decrease (44%) in the medial thicknesses between the control (75 +/- 4 microns) and L158,809-treated (42 +/- 6 microns) vein grafts (p = 0.007). The contractile responses to angiotensin II were abolished in the vein grafts by long-term L158,809 therapy. Local application by gel of L158,809 produced a significant decrease (33%) in the intimal thickness (48 +/- 3 microns) but no change in medical thicknesses (76 +/- 6 microns) compared with control grafts. The contractile responses to angiotensin II were unchanged in the vein grafts by local L158,809 therapy. CONCLUSIONS: This study shows that a local single application of L158,809 will reduce the intimal response but not the medial response in vein grafts, whereas long-term treatment will reduce intimal hyperplasia and the medial response in experimental vein grafts. Therefore angiotensin II acting through AT1 receptors mediates a significant part of the intimal hyperplastic response in vein grafts that appears to involve two phases: an acute intimal response requiring short-term therapy and a long-term medial response that requires prolonged therapy.

Effects of experimental hypertension and hypercholesterolemia on the lipid composition of the aorta.

This study evaluates alterations in canine aortic lipid composition under conditions of hypertension, hypercholesterolemia, or a combination of these factors. Hypertension was produced in the proximal thoracic aorta by creation of an aortic coarctation, whereas hypercholesterolemia was brought about by a lard-cholesterol diet in combination with thyroidectomy. Hypertension alone produced only minor changes in the lipid content of the arterial wall, but hypercholesterolemia yielded modest increases. The combination of hypertension and hypercholesterolemia, however, produced a striking increase in the total lipid content in the arterial wall. This change was most marked in the cholesteryl ester fraction, and a shift in cholesteryl ester fatty acids from linoleate to oleate was found. These data indicate that the interaction of hypertension and hypercholesterolemia produces alterations in lipid composition in a relationship which appears to be more geometric than arithmetic in nature.

Norepinephrine-induced vasoconstriction is increased in intimal hyperplastic arteries of the dog.

Norepinephrine-induced arterial contraction in vitro is known to be increased after endothelial denudation and during the subsequent development of intimal hyperplasia. We now report the response to norepinephrine in intimal-thickened iliac vessels in conscious dogs. Seven dogs underwent balloon catheter deendothelialization of the right iliac artery. Three weeks later, ultrasonic transducers were implanted on both iliac arteries to record dynamic vessel dimension. A catheter was inserted into the terminal abdominal aorta for drug infusion and blood pressure monitoring. The dogs were studied unsedated 2 days after instrumentation. Norepinephrine was infused at doses that did not affect blood pressure (0.01 to 0.05 micrograms/kg/min). Control vessel diameter decreased from 5.2 mm +/- 0.3 to 5.1 mm +/- 0.3 (2.4% +/- 1.0% when standardized to baseline diameter) and intimal hyperplastic vessels from 6.2 mm +/- 0.4 to 5.7 mm +/- 0.4 (7.6% +/- 1.7%). The difference between control and intimal hyperplastic vessel vasoconstriction was significant at p less than 0.0025. The calculated reduction in total vessel cross-section area for control vessels was 4.5% +/- 1.9%. In intimal hyperplastic vessels total cross-sectional area was reduced by 14.5% +/- 3.3% by vasoconstriction and the luminal cross-sectional area was reduced by 17.9% +/- 0.7% by the intimal hyperplasia. These data suggest that luminal compromise due to intimal hyperplasia is compounded by increased sensitivity to norepinephrine. This effect, demonstrated in a large elastic artery shortly after endothelial denudation, may be of even greater significance in a smaller vessel with advanced intimal hyperplasia.

External support modulates G protein expression and receptor coupling in experimental vein grafts.

BACKGROUND: Intimal hyperplasia remains the leading cause of vein graft failure. Various external stenting devices have been shown to reduce the development of intimal hyperplasia in vein grafts. Mitogenic and mechanotransduction signals are known to be mediated by G protein-coupled receptors. Therefore in this study we examined the alterations in G protein expression and receptor coupling in vein grafts stented with external tube support. METHODS: Thirty New Zealand White male rabbits had a right carotid interposition bypass graft with use of the ipsilateral jugular vein. Fifteen animals received external support and 15 were controls. In a subset the animals either had removal of the external support or a sham-control neck exploration at 14 days after the initial implantation (n = 5 per group). RESULTS: External support reduced G alpha i3 proteins by 30% in vein grafts without changes in G alpha s by Western blot. Vein grafts with external support were significantly less sensitive to pertussis toxin inactivation than controls were in response to both norepinephrine and serotonin. A 24% decrease in intimal thickness was maintained after withdrawal of the initial external support. CONCLUSIONS: The placement of an external support is associated with alternations in G protein expression and receptor coupling function in vein grafts. The results of this study suggest that the development of vein graft intimal hyperplasia may involve G protein-mediated events.

Increased vascular contraction and sensitivity to norepinephrine after endothelial denudation is inhibited by prazosin.

The vasomotor function of rabbit aorta was examined after deendothelialization with a Fogarty balloon catheter and during the subsequent development of intimal hyperplasia. Helical strips of injured lower abdominal aortic tissue showed an increase in norepinephrine-induced contraction when compared with control strips from normal upper abdominal aorta. This increase was 235% +/- 40% of control immediately after injury and 341% +/- 51% at 28 days after injury. Standardized dose-response curves demonstrated that the injured tissue was increasingly sensitive over time to norepinephrine and that this was more marked at physiologic levels of norepinephrine. Contraction was blocked by prazosin hydrochloride but not by yohimbine or propranolol. Furthermore, a single intramuscular dose of prazosin hydrochloride 2 hours before injury significantly (p = 0.001) reduced the maximal contraction and sensitivity. These results imply an increase in vasomotor reactivity after deendothelialization mediated through the alpha 1-adrenergic receptor. These functional changes are related to the pertinent morphologic observations.

Mast cell infiltration: a possible mechanism for vein graft vasospasm.

Mast cell infiltration of the arterial wall has been demonstrated in atherosclerotic vessels and implicated in coronary artery spasm. Spasm of vein bypass grafts has also been reported. In this study we performed vein bypass grafting of the carotid arteries in rabbits and examined the grafts for the presence of mast cells. We also determined vein graft vasoreactivity to histamine, to assess whether mediators of mast cells may have a functional role in vivo. In the control veins no mast cells were identified in 80 high-power fields (400X). In the vein bypass grafts an average of 2.6 +/- 0.8 (p = 0.01) mast cells were identified in the same number of high-power fields. Isometric tension studies of control vein and vein bypass grafts treated with histamine resulted in sigmoid dose-response curves. The ED50 for control vein was 4.69 +/- 0.63 X 10(-5) mol/L. Compared with control vein, the vein bypass grafts showed a rightward shift in the dose-response curve to histamine (ED50 11.6 +/- 1.7 X 10(-5) mol/L, p = 0.01). The histaminergic response in both vessels was blocked by the H1 receptor antagonist pyrilamine (10(-7) mol/L) and was not altered by the H2 receptor antagonist cimetidine (10(-5) mol/L). The decreased sensitivity of vein bypass grafts to histamine suggests receptor down-regulation and is possibly the result of increased histamine in the vein bypass grafts. The presence of mast cells and histamine receptors, as well as altered histamine sensitivity, in vein bypass grafts suggests that infiltration by these cells may contribute to vein bypass graft vasospasm.

Intracavernosal injections of vascular endothelial growth factor protects endothelial dependent corpora cavernosal smooth muscle relaxation in the hypercholesterolemic rabbit: a preliminary study.

Atherosclerosis is a major risk factor for erectile dysfunction, and loss of endothelium-dependent vasodilation appears early in the development of this disorder. Nitric oxide (NO) appears to be the principle mediator of erectile function and is generated in part by the sinusoidal endothelium. Vascular endothelial growth factor (VEGF) is an angiogenic growth factor and an endothelial cell-specific mitogen and the actions of VEGF are coupled to NO. In this preliminary study, we investigated whether VEGF could be used to protect endothelial dependent cavernosal relaxation from the atherosclerotic injury induced by a hypercholesterolemic diet.Two groups of New Zealand white adult male rabbits received a 1% cholesterol diet for four weeks, and two groups consumed normal rabbit chow. Half of the rabbits consuming the 1% cholesterol diet received weekly penile injections of 0.3 mg VEGF (n=8), and half injections of normal saline (n=8). Rabbits fed normal chow followed a similar protocol, half received weekly penile injections of 0.3 mg VEGF (n=6) and half were given weekly penile injections of normal saline (n=6). Isometric tension studies (with norepinephrine, acetylcholine, sodium nitroprusside and histamine) were performed on isolated strips of corpora cavernosa. The degree of corporal smooth muscle relaxation in response to ACH and SNP administration was recorded and compared. Significant elevation in serum total cholesterol levels occurred in rabbits receiving 4 weeks of the 1% cholesterol diet (727+/-75.6 mg/dl vs 38.7+/-5.53 mg/dl) P<0.01. There were no significant differences in cavernosal contraction in any group, while cavernosal smooth muscle from rabbits on normal chow retained the ability to relax in response to ACH and SNP in tissue bath. The hypercholesterolemic rabbits receiving VEGF had a significantly higher maximal per-cent relaxation to ACH (111+/-28.9) compared to the hypercholesterolemic rabbits that received NS (77+/-23.1, P<0.001). This difference in percent maximal relaxation to SNP was also present for hypercholesterolemic/VEGF rabbits (129.4+/-24) versus the hypercholesterolemic/NS rabbits (115.0+/-18, P=0.033). In conclusion, intracavernosal injections of VEGF appear to protect corporal endothelium from hypercholesterolemia induced injury, thus preserving endothelial dependent corporal smooth muscle relaxation in hypercholesterolemic rabbit.