Development of infection resistant polyurethane biomaterials using textile dyeing technology.

Infection is a major complication when using biomaterials such as polyurethane in the clinical setting. The purpose of this study was to develop a novel infection resistant polyurethane biomaterial using textile dyeing technology. This procedure results in incorporation of the antibiotic into the polymer, resulting in a slow, sustained release of antibiotic from the material over time, without the use of exogenous binder agents. Polycarbonate based urethanes were synthesized that contained either a non-ionic (bdPU) or anionic (cPU) chain extender within the polymer backbone and cast into films. The fluoroquinolone antibiotic ciprofloxacin (Cipro) was applied to bdPU and cPU using textile dyeing technology, with Cipro uptake determined by absorbance reduction of the "dyebath." These dyed bdPU/cPU samples were then evaluated for prolonged Cipro release and antimicrobial activity by means of spectrophotometric and zone of inhibition assays, respectively. Cipro release and antimicrobial activity by dyed cPU segments that were aggressively washed persisted over 9 days, compared with dyed bdPU and dipped cPU control segments that lasted < 24 hours. Dyed cPU segments, which remained in a static wash solution, maintained antimicrobial activity for 11 days (length of study), whereas controls again lost antimicrobial activity within 24 hours. Thus, application of Cipro to the cPU polymer by means of dyeing technology results in a slow sustained release of antibiotic with persistent bacteriocidal properties over extended periods of time.

Altered ubiquitin/proteasome expression in anastomotic intimal hyperplasia.

OBJECTIVE: Anastomotic intimal hyperplasia remains a leading mechanism of prosthetic arterial graft failure. Recent studies using messenger RNA differential display have demonstrated altered proteasome gene expression at the anastomoses in an expanded polytetrafluoroethylene canine carotid model. However, this technique is technically limited because of a paucity of available hyperplastic tissue at early time periods after arterial injury. Microarray gene chip technology offers a new and sensitive technique to assay early gene expression, requiring far less tissue for analysis. The purpose of this study was to screen for altered proteasome gene expression at 48 hours and 14 days after prosthetic arterial grafting. METHODS: Expanded polytetrafluoroethylene grafts (6-mm diameter, n = 9) were implanted into 25-kg mongrel dogs. The normal intervening carotid artery was used as control. At 48 hours and 14 days, RNA was extracted from the perianastomotic tissue and compared with RNA from the control carotid. Messenger RNA was then hybridized to microarray genomes screening for differential gene expression. RESULTS: Two 26S proteasome genes and five ubiquitin pathway genes were significantly underexpressed at 48 hours, among several hundred significantly expressed clones. The two 26S proteasome genes were 26S proteasomal subunit p55 (0.26), and 26S proteasomal subunit p40.5 (0.13). The underexpressed ubiquitin genes included ubiquitin (0.31), Nedd-4-like ubiquitin-protein ligase (0.30), ubiquitin conjugating enzyme UbcH2 (0.25), putative ubiquitin C-terminal hydrolase UHX1 (0.11), and ubiquitin-conjugating enzyme UbcH7 (0.12). At 14 days, six ubiquitin genes were underexpressed, and 17 26S proteasome genes were significantly downregulated. CONCLUSIONS: This study shows decreased expression of the ubiquitin/proteasome pathway 48 hours after graft implantation and similar diminished expression patterns after 14 days. This early and sustained underexpression after arterial bypass may lead to altered cell cycle control and matrix protein signaling, contributing to the unregulated proliferation of smooth muscle cells and extracellular matrix in anastomotic intimal hyperplasia after prosthetic arterial grafting.

Apolipoprotein J inhibits the migration, adhesion, and proliferation of vascular smooth muscle cells.

PURPOSE: Apolipoprotein J (ApoJ) is expressed during tissue injury and remodeling and has been implicated in vascular smooth muscle cell (VSMC) differentiation. Recently, the gene for ApoJ was identified as upregulated in distal anastomotic intimal hyperplasia after prosthetic arterial grafting. In this study we investigate the effect of ApoJ on VSMC migration, adhesion, proliferation, and gene expression. METHODS: To study the effect of ApoJ on cell migration, we used a microchemotaxis chamber with an intervening 8-microm pore semipermeable polycarbonate membrane. Assays were performed with ApoJ alone (1-50 microg/mL) or in combination with platelet-derived growth factor homodimer bb (PDGF-bb; 10 ng/mL) or 2% fetal bovine serum (FBS; n = 8) in the lower wells. The influence of extracellular matrix interactions on ApoJ and chemotaxis was studied in a similar way, except membranes were collagen coated. Furthermore, cells were exposed to ApoJ 15 minutes before or after seeding on the coated membrane (n = 10). The influence of ApoJ on cell adhesion to collagen was assessed with cell exposure to ApoJ 15 minutes before or after seeding on a collagen-coated membrane (n = 10). Migration and adhesion were quantified by counting the number of cells per three independent high-power fields with light microscopy. The effect of ApoJ in 0.4% FBS with or without PDGF-bb on VSMC proliferation (n = 12) was assessed by means of [Methyl-(3)H] thymidine incorporation. The transcriptional profile of VSMCs in 2% FBS exposed to ApoJ and a control for 24 hours was analyzed with an oligonucleotide microarray containing 12,560 genes. RESULTS: ApoJ alone was not chemotactic for VSMCs. Without collagen, ApoJ decreased the migration of VSMCs toward 2% FBS by 96% or more starting at 10 microg/mL (P < .05) and toward PDGF-bb by 60.9% or more starting at 25 microg/mL (P < .05) compared with the control. When collagen was introduced, ApoJ (25 microg/mL) decreased migration toward 2% FBS by 64% (P < .01) and toward PDGF-bb by 67.5% (P < .01) and decreased adhesion by 26.8% (P < .01) only when VSMCs in solution were exposed to ApoJ before placement on collagen. ApoJ did not induce VSMC proliferation. ApoJ alone decreased VSMC thymidine incorporation by 41.1% at 25 microg/mL (P < .05). ApoJ decreased thymidine incorporation of PDGF-bb stimulated VSMCs by 42.8% at 50 microg/mL (P < .05). Interleukin-8 and endothelin-1 were demonstrated by means of the microarray to be differentially expressed more than twofold in VSMCs that were exposed to ApoJ. CONCLUSION: ApoJ is a potent inhibitor of VSMC migration, adhesion, and proliferation. Its genetic targets are linked to cell senescence and differentiation. Therefore, ApoJ may play a role, in part, in modulating the VSMC response to injury.

In vivo magnetic resonance imaging of experimental thrombosis in a rabbit model.

The process of atherosclerotic plaque disruption has been difficult to monitor because of the lack of an animal model and the limited ability to directly visualize the plaque and overlying thrombus in vivo. Our aim was to validate in vivo magnetic resonance imaging (MRI) of the thrombus formation after pharmacological triggering of plaque disruption in the modified Constantinides animal model of plaque disruption. Atherosclerosis was induced in 9 New Zealand White male rabbits (3 kg) with aortic balloon endothelial injury followed by a high cholesterol (1%) diet for 8 weeks. After baseline (pretrigger) MRI, the rabbits underwent pharmacological triggering with Russell's viper venom and histamine, followed by another MRI 48 hours later. Contiguous cross-sectional T2-weighted fast spin echo images of the abdominal aorta were compared by histopathology. In all animals, aortic wall thickening was present on the pretrigger MRI. On MRIs performed 48 hours after triggering, a histologically confirmed intraluminal thrombus was visualized in 6 (67%) of the 9 animals. MRI data correlated with the histopathology regarding aortic wall thickness (R=0.77, P<0.0005), thrombus size (R=0.82, P<0.0001), thrombus length (R=0.86, P<0.005), and anatomic location (R=0.98, P<0.0001). In vivo, MRI reliably determines the presence, location, and size of the thrombus in this animal model of atherosclerosis and plaque disruption. The combination of in vivo MRI and the modified Constantinides animal model could be an important research tool for our understanding of the pathogenesis of acute coronary syndromes.

Apolipoprotein J inhibits the migration and adhesion of endothelial cells.

BACKGROUND: Apolipoprotein J (ApoJ) is expressed after vascular injury and remodeling and may inhibit endothelial cell activation in the vascular wall. Recently, ApoJ was identified as upregulated in hyperplastic lesions after prosthetic arterial grafting. This study analyzed the effect of ApoJ on human umbilical vein endothelial cell (HUVEC) migration, adhesion, and proliferation. METHODS: Cell migration towards ApoJ + fetal bovine serum (FBS) or vascular endothelial growth factor (VEGF) was evaluated with the use of a microchemotaxis chamber with or without a fibronectin-coated membrane. For migration that involved fibronectin, cells were exposed to ApoJ before or after placement on the membrane. Cell adhesion to fibronectin was studied similarly but without stimulant. The vital dye alamar blue assessed proliferation of ApoJ + FBS- or VEGF-stimulated HUVECs. RESULTS: ApoJ alone did not cause migration or proliferation of HUVECs. Without fibronectin, ApoJ decreased the migration of HUVECs towards FBS or VEGF. When fibronectin was introduced, ApoJ decreased cell migration toward FBS or VEGF and decreased adhesion only when HUVECs in solution were exposed to ApoJ before the placement on fibronectin. ApoJ had no effect on FBS- or VEGF-induced proliferation. CONCLUSIONS: ApoJ inhibits HUVEC migration and adhesion. By altering endothelial function during vascular injury, ApoJ appears to regulate, in part, the early development of intimal hyperplasia after prosthetic arterial grafting.

Human vascular smooth muscle cells of diabetic origin exhibit increased proliferation, adhesion, and migration.

PURPOSE: Patients with diabetes mellitus (DM) experience progressive macrovascular atherosclerosis and intimal hyperplastic restenosis with increased frequency as compared with nondiabetic patients. These observations suggest that vascular smooth muscle cells (VSMCs) behave in a phenotypically different and more aggressive manner in diabetic patients. In this study, we compared the in vitro rates of proliferation, adhesion, and migration of human VSMCs obtained from diabetic and nondiabetic patients. METHODS: Human VSMC cultures were isolated from 23 diabetic patients (9 artery, 14 vein) and 15 nondiabetic patients (9 artery, 6 vein) with extensive lower extremity atherosclerosis. All patients were between 61 and 78 years of age (average: 68.4 years [diabetic]; 67.3 years [nondiabetic]). All diabetic patients had type 2 DM. Vascular specimens were obtained at the time of amputation from infragenicular arteries and during arterial revascularization from saphenous veins. Cells from passages 2 and 3 were assayed for their proliferative capacity with total DNA fluorescence photometry and for adhesion and migration with a modified Boyden chamber. RESULTS: The average duration of diabetes was 11.6 +/- 4.1 years. The average number of diabetic complications (retinopathy, neuropathy, nephropathy, coronary artery disease) was 2.8 +/- 0.7 per patient. Diabetic VSMCs exhibited abnormal morphology in cell culture with loss of the normal hill and valley configuration. Proliferation was significantly increased in VSMCs of diabetic origin (156 +/- 57 absorption units) as compared with those of nondiabetic origin (116 +/- 42 absorption units) (P <.001). Diabetic VSMCs demonstrated significantly greater adhesion (63.6 +/- 24 per high-power field vs 37.9 +/- 13 per high-power field; P =.002) and migration (397 +/- 151 per low-power field vs 121 +/- 99 per low-power field; P =.001) rates. CONCLUSIONS: Diabetic VSMCs exhibit significantly increased rates of proliferation, adhesion, and migration as well as abnormal cell culture morphology suggestive of abnormal contact inhibition. These observations of human VSMCs in culture are consistent with the increased rate of infragenicular atherosclerosis and the increased rates of restenosis observed clinically in diabetic patients. The atherosclerosis- and intimal hyperplasia-promoting behavior exhibited appears to be intrinsic to the DM-VSMC phenotype and must be considered when designing methods to limit atherosclerosis and intimal hyperplasia in diabetic patients.

Coating of Dacron vascular grafts with an ionic polyurethane: a novel sealant with protein binding properties.

The purpose of this study was to develop a novel sealant that would seal prosthetic vascular graft interstices and be accessible for protein binding. Crimped knitted Dacron vascular grafts were cleaned (CNTRL) and hydrolyzed in boiling sodium hydroxide (HYD). These HYD grafts were sealed using an 11% solids solution of a polyether-based urethane with carboxylic acid groups (PEU-D) via a novel technique that employs both trans-wall and luminal perfusion. Carboxylic acid content, determined via methylene blue dye uptake, was 2.3- and 4.2-fold greater in PEU-D segments (1.0+/-0.27 nmol/mg) as compared to HYD and CNTRL segments, respectively. Water permeation through PEU-D graft (1.1+/-2 ml/cm2 min(-1)) was comparable to collagen-impregnated Dacron (9.8+/-10 ml/cm2 min(-1)). Non-specific 125I-albumin (125I-Alb) binding to PEU-D segments (18+/-3 ng/mg) was significantly lower than HYD and CNTRL segments. 125I-Alb linkage to PEU-D using the crosslinker EDC resulted in 5.7-fold greater binding (103+/-2 ng/mg) than non-specific PEU-D controls. However, covalent linkage of 125I-Alb to PEU-D was 4.9- and 5.9-fold less than CNTRL and HYD segments with EDC, respectively. Thus, ionic polyurethane can be applied to a pre-formed vascular graft, seal the interstices and create "anchor" sites for protein attachment.

Transatrial access to the normal pericardial space for local cardiac therapy: preclinical safety testing with aspirin and pulmonary artery hypertension.

The reliability, rapidity, and safety of nonsurgical, transatrial pericardial access for local cardiac therapy have been demonstrated in healthy animals. Since many patients take aspirin or have increased right-sided pressures, we evaluated the procedure's safety under these conditions. Transatrial pericardial access was performed in anesthetized pigs following aspirin administration (162 mg p.o., n = 6) or during experimental pulmonary artery hypertension (n = 4 different animals) and required only 3 minutes following guide catheter positioning. Platelet aggregability testing with arachidonic acid confirmed aspirin effectiveness. Mean pericardial fluid hematocrit was 0.1 +/- 0.1% after 2 days of aspirin therapy and 1.9 +/- 1.1% at sacrifice 24 hours later (NS). Mean pericardial fluid hematocrit was 1.0 +/- 0.5% after 45 minutes of pulmonary artery hypertension and 4.3 +/- 0.8% at sacrifice 30 minutes later (NS). Histologic analysis in both groups revealed a small thrombus and localized inflammation at the site of puncture. Neither aspirin use nor pulmonary artery hypertension causes significant bleeding into the pericardial space following transatrial access and thus does not preclude this route for local cardiac drug delivery.

Relationship of the 20S proteasome and the proteasome activator PA28 to atherosclerosis and intimal hyperplasia in the human vascular system.

Down-regulation of the proteasome activator PA28 results in abnormal proteasome activation and has been implicated in the development of intimal hyperplasia (IH) in animal models. Demonstration of proteasome and PA28 expression has not yet been documented in the human vascular system. This study sought to define the distribution of the 20S proteasome and its activator PA28 in human vessels and determine the relationship between the expression of the proteasome and PA28 and the development of atherosclerosis and IH. Vascular biopsies were obtained from 70 patients at the time of surgery, were snap frozen and sectioned in 5-micron sections, and prepared using standard histological techniques. The immunoperoxidase technique was used to identify 20S proteasome and PA28 expression in diseased and normal human arteries and veins as well as in patent bypass grafts with and without IH. Expression was graded by a blinded pathologist (scale: 1-4). Repeat quantification of the immunopositive cells was also performed. Expression of 20S proteasome and PA28 was identified in all vascular tissues examined. The proteins were identified predominately within the cytoplasm of vascular smooth muscle cells and endothelial cells. PA28 was more intensely expressed in quiescent regions of the vessel wall as compared to areas undergoing active proliferation and remodeling. PA28-mediated activation of the proteasome may be necessary to maintain normal cellular homeostasis and prevent excessive cellular proliferation in the human vascular system. Abnormalities of proteasome activation may have a significant role in the development of atherosclerosis and IH.

Effect of porosity on small-diameter vascular graft healing.

Experimental studies have reported that complete healing of small-diameter expanded polytetrafluoroethylene (ePTFE) grafts occurs only if the porosity of the graft is increased, thereby allowing ingrowth of perigraft capillaries yielding endothelial cells. This study investigates the effects of varied graft porosity on the healing characteristics of 2-mm internal diameter (ID) ePTFE grafts interposed in the rabbit common carotid artery. Four groups were evaluated: Group A (n = 8) standard (30-microm pores) ePTFE grafts; Group B (n = 8) increased porosity (60-microm pores) ePTFE grafts; Group C (n = 8) standard ePTFE; and Group D (n = 8) 60-microm ePTFE external graft surface was externally coated with an impermeable layer of polyurethane. Patency was 100% for all groups at 8 weeks. At explantation, the neointima was composed of primarily modified smooth muscle cells. Endothelial cells were only identified at the perianastomotic region using the endothelial cell-specific antibody CD31. The impermeable external polyurethane coating of ePTFE grafts had no effect on neointima formation, regardless of porosity.

Serum lipids and arterial plaque load are altered independently with high-dose progesterone in hypercholesterolemic male rabbits.

Antiatherogenic effects of sex steroids in premenopausal women are not well defined. Therefore, we employed an established rabbit model for atherosclerosis to study the effects of exogenous estrogen and a progesterone analogue (P) on serum lipids and aortic plaque load. Serum cholesterol (C) and triglyceride (T) levels and atherosclerotic plaque loads were compared in 5 groups of male New Zealand White rabbits fed a 12-week, C-rich diet: 1 control group (CG) and 4 groups treated with estriol (E), haloperidol (H), low-dose 17-hydroxyprogesterone (LDP), or high-dose 17-hydroxyprogesterone (HDP). Serum P was measured in the LDP and HDP groups. Serial histologic sections (15 each of 27 ascending aortas) were studied by light microscopy and computerized morphometric analysis. Plaque load is defined as the ratio of intimal area to medial area (I/M). Exogenous E (p<0.001), H (P = 0. 02), LDP and HDP (P<0.001, each) were found to be significantly associated with less aortic plaque load than controls. In a multivariate analysis, after controlling for the differences in serum C and T levels, HDP (p = 0.014) was found to be associated with less aortic plaque load than controls, and this association approached statistical significance in the E (p = 0.052) and H (p = 0.069) groups. These data suggest that the mechanism(s) involved with the antiatherogenic effect of HDP in this animal model is, or are, independent of an alteration in serum lipids.

Preclinical safety testing of percutaneous transatrial access to the normal pericardial space for local cardiac drug delivery and diagnostic sampling.

The safety of a percutaneous method and streamlined catheter system to access the normal pericardial space via the right atrial appendage for drug delivery and diagnostic sampling was demonstrated in 20 anesthetized pigs. Access was successfully accomplished in all animals within 3 min of guide catheter positioning and was documented by fluoroscopic imaging and pericardial fluid sampling. The animals were sacrificed at 24 hr (n = 10) and 2 weeks (n = 10) for histopathologic analysis. Mean pericardial hematocrit was 1.1% +/- 0.3% at initial sampling, 4.3% +/- 1.4% at 24 hr (P = 0.005 vs. baseline), and 0.4% +/- 0.2% at 2 weeks (P = 0.13 vs. baseline). At 24 hr, there was local inflammatory reaction in the atrial wall and a small thrombus at the site of puncture. At 2 weeks, no significant inflammatory changes or pericarditis were evident. The technique is well tolerated with no apparent adverse complications. Advances in intrapericardial therapeutics and diagnostics will direct the clinical application of this novel approach in human subjects.

Interferon gamma up-regulates a novel protein in vascular smooth muscle cells.

PURPOSE: By means of the technique of messenger RNA (mRNA) differential display, we previously isolated a partial DNA clone found to be down-regulated at the polytetrafluoroethylene (PTFE) hyperplastic arterial anastomosis compared with the normal artery. The partial DNA gene sequence was found to be homologous with interferon gamma up-regulated protein (IGUP) first found in human psoriatic keratinocytes. We cloned the entire IGUP gene from human vascular smooth muscle cells (VSMCs) to determine its regulation by gamma interferon (gamma-IFN) and other cytokines in cultured human VSMCs. METHODS: By means of polymerase chain reaction, the IGUP gene was amplified from a QUICK-Clone complementary DNA human aorta kit using 5' and 3' oligonucleotide primers to the known IGUP sequence. Immunohistocytochemistry studies compared normal artery and distal anastomotic IH. Human VSMCs were stimulated with 1000 U/mL of gamma-IFN, 5 ng/mL of platelet-derived growth factor BB (PDGF-BB), 3. 2 ng/mL basic fibroblast growth factor, 3.3 ng/mL transforming growth factor beta(TGF-beta), 10 ng/mL of vascular endothelial growth factor, and 10% fetal bovine serum (FBS) for zero, 24, 48 and 72 hours. Western blot analysis of lysates of the stimulated VSMCs was performed to determine up-regulation of IGUP. RESULTS: DNA sequencing confirmed the cloning of the entire coding region of the IGUP gene with 100% homology to the known IGUP DNA sequence. There was strong expression of IGUP in quiescent VSMCs and marked reduction of expression of IGUP in proliferating smooth muscle cells. gamma-IFN was the only cytokine, of the cytokines evaluated, to up-regulate production of IGUP in VSMCs. CONCLUSION: IGUP is a novel protein in VSMCs found to be down-regulated in areas of anastomotic IH, as compared with a normal artery. We have now shown IGUP to be up-regulated only by gamma-IFN in human VSMCs. IGUP may, therefore, be the intermediary for the known gamma-IFN inhibition of human VSMC proliferation.

In vivo assessment of a novel dacron surface with covalently bound recombinant hirudin.

Prosthetic arterial graft surfaces are relatively thrombogenic and fail to heal with a cellular neointima. The goal of this study was to characterize the in vivo antithrombin properties of a novel Dacron surface with covalently linked recombinant hirudin (rHir) implanted in a canine thoracic aorta with high flow and shear rates. rHir was bound to a knitted Dacron patch using crosslinker-modified bovine serum albumin (BSA) as a basecoat protein. BSA was first reacted with the heterobifunctional crosslinker, sulfo-SMCC. This BSA-SMCC complex was then bound to the carboxylic acid groups of hydrolyzed Dacron patches using the carbodiimide crosslinker, 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide hydrochloride. Iodinated, Traut's-modified rHir (125I-rHir-SH) was then reacted with the Dacron-BSA-SMCC surface, thereby covalently binding 125I-rHir. Graft segments were washed and sonicated to remove any nonspecifically bound 125I-rHir. Dacron-BSA-SMCC-S-125I-rHir patches (n = 5) and control Dacron-BSA patches (n = 5) were implanted in series in the thoracic aortas of canines. These patches were exposed to nonheparinized, arterial blood flow for 2 hours. Patches were explanted and assessed for 125I-rHir loss. Antithrombin activity of explanted 1-cm2 patch segments was evaluated using a chromogenic assay with 1, 5, 10, 15 units of added thrombin. Light microscopy was performed to qualitatively examine the pseudointima. Two animals were excluded from the study owing to excessive bleeding through the knitted 125I-rHir patch. Comparison of preoperative and postoperative 125I-rHir gamma counts revealed an overall decrease of 20+/-5.4% over the period studied. Explanted 125I-rHir patch segments were able to inhibit 1, 5, and 7 NIHU of thrombin, demonstrating retained antithrombin activity. Gross and microscopic examination of the control and test Dacron surfaces showed marked differences. Dacron surfaces with covalently bound 125I-rHir had no gross thrombus and a thin pseudointima of platelets and plasma proteins. In contrast, the control patches had a thick pseudointima composed of fibrin-rich thrombus. rHir, covalently bound to Dacron patches, maintains its biologic activity as well as prevents thrombus formation on the graft surface. This novel antithrombin coating, by modifying the blood/ graft interface, may improve both short- and long-term patency in small-diameter prosthetic arterial grafts and has applications with respect to other implantable or indwelling biomaterials.

Noninvasive characterization of plaque morphology using helical computed tomography.

BACKGROUND: Evidence suggests that atherosclerotic plaque composition and morphology are more important predictors of plaque stability and clinical behavior than the degree of vessel stenosis. Unfortunately, noninvasive methods of plaque characterization in situ using duplex ultrasound have been largely unsuccessful. We investigated the accuracy of helical computed tomography (HCT) in defining plaque morphology in patients undergoing carotid endarterectomy. METHODS: Candidates for endarterectomy were enrolled prospectively and underwent HCT with intravenous contrast injection. The attenuation values (in Hounsfield Units) of plaque features such as calcium, fibrous tissue, and lipid were measured and correlated with the tissue types seen on microscopic examination. Percent luminal stenosis derived from transverse HCT images was also compared to values obtained from duplex ultrasound. SETTING: Ambulatory patients referred to a tertiary, metropolitan medical center. RESULTS: A total of 21 out of 22 cases were suitable for comparison. Qualitatively, HCT accurately defined plaque features containing calcium, fibrous stroma, and lipid. Using tissue attenuation values (Hounsfield Units - H), HCT distinguished between lipid and fibrous stroma (means 39+/-12 H and 90+/-24 H, respectively, p<0.001). HCT-derived calculations of percent stenosis also correlated with the ultrasound data (p<0.001). CONCLUSIONS: HCT is an accurate, noninvasive means for studying detailed plaque morphology and composition. This technique provides the basis for further studies correlating plaque anatomy with clinical behavior.

Synthesis of a novel small diameter polyurethane vascular graft with reactive binding sites.

Development of a small diameter (4 mm inner diameter [ID]) prosthetic vascular graft with functional groups accessible for covalent binding of recombinant hirudin (a potent anticoagulant) should create a more hemocompatible surface. The purpose of this study was to develop a technique for generating carboxylic acid groups on the surface of precast 4 mm ID poly-(carbonate urea)-urethane vascular grafts and to evaluate the accessibility of these groups. A polycarbonate based urethane with the chain extender 2,2-bis(hydroxymethyl)propionic acid was synthesized. A precast 4 mm ID poly(carbonate urea)-urethane vascular graft (Chronoflex [CF]; CardioTech International, Woburn, MA) was then placed into a 4% carboxylated polyurethane (cPU) solution (in 1% dimethyl acetamide) and incubated for 30 minutes (cPU graft). To determine the accessibility of the carboxylic acid groups, a standard textile technique using methylene blue dye was used. Macroscopic cross-sections, which were cut and evaluated for dye penetration, showed greatest concentration of carboxylic acid groups at the luminal and capsule surfaces, with minimal penetration into the mid-portion of the graft. Analysis of dye baths for absorbance reduction resulted in the cPU grafts having 3.7-fold and 5.4-fold more accessible carboxylic acid groups compared with untreated and dimethyl acetamide dipped CF grafts. Thus, a novel small diameter vascular graft has been developed that contains reactive carboxylic acid groups accessible for protein binding.

Bioengineering of a novel small diameter polyurethane vascular graft with covalently bound recombinant hirudin.

Development of a small diameter prosthetic vascular graft with surface based antithrombin properties should aid in maintaining early graft patency in small vessel reconstruction. The purpose of this study was to bind covalently a basecoat protein (canine serum albumin [CSAJ) and a potent antithrombin agent (recombinant hirudin [rHir]) to 4 mm inner diameter poly(carbonate urea) urethane grafts with reactive carboxylic acid groups (cPU). 125I-CSA was covalently bound to 1 cm length segments of cPU grafts using the carbodimide cross-linker, 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide hydrochloride (EDC). To bind 125I-rHir covalently, CSA was modified with the heterobifunctional cross-linker sulfosuccinimidyl 4-(N-maleimidomethyl) cyclohexane-1-carboxylate (sulfo-SMCC) before linkage to the cPU surface with EDC (cPU-CSA-SMCC). 125I-rHir was modified with Traut's reagent and reacted with the cPU-CSA-SMCC surface, covalently linking 125I-rHir to surface bound CSA. 125I-CSA binding to the cPU graft surface (34,235 ng/segment) was ninefold, sevenfold, and 10-fold greater than controls with nonspecifically bound 125I-CSA. Covalent linkage of 125I-rHir to the cPU-CSA-SMCC surface (9,974 ng/segment) was 172, 192, and 142-fold greater than controls with nonspecifically bound 125I-rHir. Surface antithrombin properties were characterized using a chromogenic assay to measure residual thrombin activity. Evaluation of surface antithrombin activity showed significantly greater 131I-thrombin inhibition and binding by the cPU surface with covalently bound 125I-rHir, as compared with controls. Release of 125I-rHir from the cPU surface was minimal as compared with controls. Therefore, rHir can be covalently linked to a novel small diameter polyurethane vascular graft surface while maintaining its potent antithrombin properties.

Covalent linkage of recombinant hirudin to a novel ionic poly(carbonate) urethane polymer with protein binding sites: determination of surface antithrombin activity.

Surface thrombus formation on implantable biomaterials such as polyurethane is a major concern when utilizing these materials in the clinical setting. Thrombin, which is responsible for thrombus formation and smooth muscle cell activation, has been the target of numerous surface modification strategies in an effort to prevent this phenomenon from occurring. The purpose of this study was to covalently immobilize the potent, specific antithrombin agent recombinant hirudin (rHir) onto a novel polyurethane polymer synthesized with carboxylic acid groups which served as protein attachment sites. The in vitro efficacy of thrombin inhibition by this novel biomaterial surface was then evaluated. Bovine serum albumin (BSA), which was selected as the basecoat protein, was reacted with sulfo-SMCC in a 1:50 molar ratio. This BSA-SMCC complex was then covalently linked to the carboxylated polyurethane (cPU) surface via the crosslinker EDU (cPU-BSA-SMCC). This cPU-BSA-SMCC surface was then reacted with Traut's-modified 125I-rHir, a procedure which created free sulfhydryl groups on rHir (cPU-BSA-SMCC-S-125I-rHir). Using these crosslinking procedures, the cPU-BSA-SMCC-S-125I-rHir segments bound 188 +/- 40 ng/cm2 (n = 60) whereas the controls with non-specifically bound 125I-rHir (Mitrathane + EDC + BSA + 125I-rHir-SH and cPU-BSA + 125I-rHir-SH) bound 13 +/- 8 ng/cm2 and 4 +/- 8 ng/cm2, respectively. Evaluation of these cPU-BSA-SMCC-S-125I-rHir segments for 131I-thrombin inhibition using a chromogenic assay for thrombin showed that a maximum of 2.64 NIHU thrombin was inhibited in contrast to the controls which inhibited bound 0.76 and 0.70 NIHU. Controls with nonspecifically bound 125I-rHir also had 0.31 and 0.76 NIHU 131I-thrombin adherent to their respective surfaces whereas the maximum 131I-thrombin binding to the cPU-BSA-SMCC-S-rHir segments was 1.51 NIHU. Exposure to 131I-thrombin did not result in any release of covalently bound 125I-rHir from the cPU-BSA-SMCC-S-125I-rHir segments. Thus, these results demonstrate that rHir can be covalently bound to this novel polyurethane surface and still maintain potent antithrombin activity.

Identification of multiple genes with altered expression at the distal anastomosis of healing polytetrafluoroethylene grafts.

PURPOSE: Anastomotic intimal hyperplasia remains a significant cause of delayed prosthetic arterial graft failure. Prior studies have identified several genes with altered expression within the hyperplastic region at the downstream polytetrafluoroethylene arterial anastomosis as compared with normal arteries. The purpose of the current study was to determine the sequence of early gene-related events at the distal anastomosis of an in vivo prosthetic arterial graft model. Messenger RNA (mRNA) differential display was used to screen for alterations in gene expression between anastomotic sites and control arterial segments. METHODS: Six carotid interposition 6-mm expanded polytetrafluoroethylene grafts were placed in mongrel dogs, with the intervening carotid artery segment serving as the baseline control. Five days after graft implantation, the distal anastomotic artery segments were harvested and total RNA was isolated from both the intervening normal arteries and anastomotic segments. Differential mRNA display was used to identify candidate complementary DNA (cDNA) clones with expression that differed in anastomotic segments as compared with normal intervening arteries. Northern blot analysis confirmed alteration of gene expression. The cDNA clones were sequenced, and gene databases were searched. Novel sequences were used as probes for screening human cDNA libraries. RESULTS: Approximately 7000 mRNA species were screened, and 26 candidate clones were obtained. Northern blot analysis showed altered gene expression in 10 (38%) of the clones, undetectable signals in 13 (50%), and nonregulation in 3 (12%). Seven clones with 92% homology at the nucleotide level to human alpha1 (III) procollagen gene and novel sequence were expressed only at the distal anastomosis. A clone representing apolipoprotein J and a novel sequence had increased expression at the distal anastomosis of 364% +/- 236% and 156% +/- 47%, respectively (mean percentage, control +/- standard deviation). CONCLUSIONS: These studies identified genes with expressions that increased or were exclusive to the distal anastomosis of healing prosthetic arterial grafts in an in vivo prosthetic arterial graft model. Type III collagen may contribute significantly to the composition of the extracellular matrix associated with intimal hyperplasia by increasing lesion volume. Apolipoprotein J, through its association with proteases, may modulate some of the matrix changes seen early after grafting.