Elastic and viscoelastic properties of porcine subdermal fat using MRI and inverse FEA.

There is a scarcity of investigation into the mechanical properties of subdermal fat. Recently, progress has been made in the determination of subdermal stress and strain distributions. This requires accurate constitutive modelling and consideration of the subdermal tissues. This paper reports the results of a study to estimate non-linear elastic and viscoelastic properties of porcine subdermal fat using a simple constitutive model. High-resolution magnetic resonance imaging (MRI) was used to acquire a time series of coincident images during a confined indentation experiment. Inverse finite element analysis was used to estimate the material parameters. The Neo Hookean model was used to represent the elastic behaviour (µ = 0.53 ± 0.31 kPa), while a single-element Prony series was used to model the viscoelastic response (α = 0.39 ± 0.03, τ = 700 ± 255 s).

The control of Staphylococcus epidermidis biofilm formation and in vivo infection rates by covalently bound furanones.

In order to overcome the continuing infection rate associated with biomaterials, the use of covalently bound furanones as an antibiofilm coating for biomaterials has been investigated. Furanones have previously been shown to inhibit growth of Gram-positive and Gram-negative bacteria. The aim of these studies were to covalently bind furanones to polymers and to test their efficacy for inhibiting biofilm formation of Staphylococcus epidermidis and in vivo infection rate. Two methods of covalent attachment of furanones were used. The first, a co-polymerisation with a styrene polymer, and second, a plasma-1-ethyl-3-(dimethylaminopropyl) carbodiimide (EDC) reaction to produce furanone-coated catheters. Biofilm formation by S. epidermidis in vitro was inhibited by 89% for polystryene-furanone disks and by 78% by furanone-coated catheters (p<0.01). In an in vivo sheep model we found furanones were effective at controlling infection for up to 65 days. Furanones have potential to be used as a coating for biomaterials to control infection caused by S. epidermidis.

New methods for the assessment of in vitro and in vivo stress cracking in biomedical polyurethanes.

This article describes a new test method for the assessment of the severity of environmental stress cracking of biomedical polyurethanes in a manner that minimizes the degree of subjectivity involved. The effect of applied strain and acetone pre-treatment on degradation of Pellethane 2363 80A and Pellethane 2363 55D polyurethanes under in vitro and in vivo conditions is studied. The results are presented using a magnification-weighted image rating system that allows the semi-quantitative rating of degradation based on distribution and severity of surface damage. Devices for applying controlled strain to both flat sheet and tubing samples are described. The new rating system consistently discriminated between the effects of acetone pre-treatments, strain and exposure times in both in vitro and in vivo experiments. As expected, P80A underwent considerable stress cracking compared with P55D. P80A produced similar stress crack ratings in both in vivo and in vitro experiments, however P55D performed worse under in vitro conditions compared with in vivo. This result indicated that care must be taken when interpreting in vitro results in the absence of in vivo data.

Fluid dynamics of a textured blood-contacting surface.

This study examined the fluid dynamics of a textured blood-contacting surface using a computational fluid-dynamic modeling technique. The texture consisted of a regular array of microfibers of length 50 or 100 microm, spaced 100 microm apart, projecting perpendicularly to the surface. The results showed that the surface texture served as a flow-retarding solid boundary for a laminar viscous flow, resulting in a lowered wall shear stress on the hase-plane surface. However, the maximum wall shear stress on the fibers was much higher than the shear stress on the nontextured phase plane. At all fractions of fiber height down past 10 microm, the permeability of the textured region greatly exceeded the analytically predictable permeability of an equivalent array of infinite-height fihers. The lowered suiface shear stress appears to explain in part the enhanced deposition of formed blood elements on the textured surface.

Acute cellular interaction with textured surfaces in blood contact.

Textured blood-contacting surfaces can promote the formation of a blood-compatible pseudo-neointima. We hypothesized that by controlling the surface texturing, the pseudo-neointima thickness could be controlled. The hypothesis was tested experimentally by fabricating the polyurethane textured surfaces with three different fiber lengths, and exposing them simultaneously to the flowing blood in an ovine ex vivo carotid-jugular series shunt for periods up to 4 h. The textured surface consisted of regularly spaced tapered micro-fibers of defined length on a smooth base-plane surface. Because of the simple surface topography, detailed computational fluid-dynamic modeling of the surface could be obtained as a parallel study. Experimental results showed that white cell was the predominant cell type deposited on the textured surfaces, whereas macroscopic thrombus formation occurred only in one of nine blood-contacting experiments. White cell density on the textured base-plane surface was subsequently quantified by image-analyzing the electron micrographs of blood-contacted textured surfaces. The statistical analysis of cell densities on individual textured surfaces showed effects of wall shear stress on the textured base plane (which was obtained from the fluid-dynamic modeling), the longitudinal position of the test section in the series shunt, and blood-contact time.

Polydimethylsiloxane/polyether-mixed macrodiol-based polyurethane elastomers: biostability.

A series of four thermoplastic polyurethane elastomers were synthesized with varying proportions of poly(hexamethylene oxide) (PHMO) and poly(dimethylsiloxane) (PDMS) macrodiols. The macrodiol ratios (by weight) employed were (% PDMS:% PHMO) 100:0, 80:20, 50:50 and 20:80. The weight fraction of macrodiol in each polymer was fixed at 60%. The mixed macrodiols were reacted with 4,4'-methylenediphenyl diisocyanate (MDI) and 1,4-butanediol (BDO) chain extender. The biostability of these polymers was assessed by strained subcutaneous implantation in sheep for three months followed by microscopic examination. Pellethane 2363-80A and 2363-55D were employed as control materials. The mechanical properties of the polymers were tested and discussed along with the biostability results. The results showed that soft, flexible PDMS-based polyurethanes with very promising biostability can be successfully produced using the mixed macrodiol approach. The formulation with 80% PDMS macrodiol produced the best result in terms of a combination of flexibility, strength and biostability.

A novel textured surface for blood-contact.

Blood-contacting surface modifications aimed at reduction of thromboembolic complications have included the texturing of surfaces so as to promote the formation of a stable pseudo-neointima. A technique has been developed whereby a textured surface consisting of regularly spaced micro-fibres was produced on a smooth base plane. Polyurethane vascular patches with and without the textured luminal surface were fabricated and implanted bilaterally in ovine carotid arteries for 1- and 3-week implantation periods (n = 6 per period). One of 6 arteries with textured patches in the 1-week group was occluded. All other arteries were patent. At 1 week, all patent textured patches had adherent thrombus covering the entire patch surface. By 3 weeks, the thrombus had organised to form a stable pseudo-neointima. Non-textured patches at 1 week had only partial surface coverage of thrombus. At 3 weeks, 4 of 6 non-textured patches had significant red thrombus in the lumen. At 3 weeks, there was also evidence of cellular migration from artery onto both textured and non-textured patches. These findings suggest that the major role of the textured surface was as a promoter of a stabilised thrombus base onto which subsequent cellular migration and tissue healing occurred more rapidly than onto a smooth polyurethane surface.

Nitrous acid pretreatment of tendon xenografts cross-linked with glutaraldehyde and sterilized with gamma irradiation.

Collagenous xenografts made from kangaroo tail tendon cross-linked with glutaraldehyde have a potential application in the reconstruction of massive digital tendon deficits. However, a limitation to the clinical use of these xenografts has been the optimization of collagen cross-linking, and subsequent bio-incorporation and retention of mechanical properties following implantation. The purpose of this study was to evaluate the effect of nitrous acid on modulating the biologic and mechanical properties of tendon xenografts cross-linked with glutaraldehyde. Tendon xenografts were pretreated with 0.1 or 0.01 M nitrous acid solution, prior to cross-linking in 2% glutaraldehyde and sterilization by gamma irradiation. Xenografts were implanted intramuscularly in rabbits to examine biocompatability, and also used to repair ovine digital extensor tendon deficits to evaluate functional incorporation. Histologically, intramuscularly implanted nitrous acid pretreated xenografts in rabbits had a greater degree of diffuse cellular infiltration into interstitial splits in the graft than controls after 12 weeks. Xenografts implanted in an ovine extensor tendon deficit were evaluated after 26 and 52 weeks. Rate of failure of tenorrhaphies between host tendon and xenografts overall (15/21) was significantly greater (P < 0.05) than for autografts (1/21), suggesting that the holding power of sutures in xenografts was inferior to that obtained in autografts. Tensile failure stress of midsections of both nitrous acid pretreated and control xenografts was about 100 MPa prior to implantation (time zero). After 26 and 52 weeks, failure stress of both types of xenografts was significantly less than at time zero (P < 0.05). At 52 weeks, failure stress of nitrous acid pretreated xenografts (47.4 +/- 3.1 MPa) was significantly less than control xenografts (63.7 +/- 5.4 MPa); (P < 0.05). However, nitrous acid pretreated xenografts were similar to control xenografts in failure load (357 +/- 29 and 354 +/- 26 N, respectively), but they tended to have larger cross-sectional areas (7.6 +/- 0.5 versus 5.7 +/- 0.6 mm2, respectively) which were responsible for the lower calculated value for failure stress. Histologically, autografts maintained their normal tissue architecture and evoked a more limited cellular response in surrounding tissues than xenografts (P < 0.05). Both types of xenograft were surrounded by a thicker cuff of cellular response than autografts. However, compared to control xenografts, nitrous acid pretreated xenografts had more extensive fragmentation and splitting of collagen bundles, and more diffuse cellular and vascular infiltration into these interstitial splits, and these alterations were apparently contributing to the greater 'swelling' of these xenografts. It was concluded that pretreatment of tendon xenografts with nitrous acid modulated their biologic and material properties. Further studies are needed to elucidate the mechanism of these effects, and to determine if the protocol for tendon xenograft preparation could be optimized for improved clinical performance.

A novel method to quantify the edge contour of RGP contact lenses.

A new method of accurate profile measurement of rigid gas permeable lenses and edges is described. A purpose built optical profilometer non-destructively scans complete cross-sections in several meridians of any rigid lens. The accuracy of the instrument was verified to be better than +/-2.5 microm. The profile data are obtained and stored in digital format for further analysis. The seven most commonly used edge features were defined by means of mathematical algorithms to extract unique numerical values for these parameters from the measured profiles. Applications of the new edge quantification method in clinical contact lens research and manufacturing quality control are shown.

The bionic eye (electronic visual prosthesis): a review.

The concept of a visual prosthesis for the blind or partially sighted is not a new one. Indeed, for more than three decades this technology based treatment for blindness has appeared imminent. Despite the concerted efforts of numerous physicians, scientists and engineers, the successful application of a useful visual prosthesis remains elusive. The present review will endeavour to describe past efforts, investigate the present state of the art and indicate the obstacles that must be overcome in order to bring an electronic visual prosthesis to fruition.

Nutritional requirements of the corneal epithelium and anterior stroma: clinical findings.

PURPOSE: To monitor the health of the epithelium and the anterior stroma when porous membranes are implanted into the feline cornea and to determine membrane diffusivity characteristics needed to maintain corneal integrity. METHODS: Filtration membranes in a range of effective pore sizes of less than 15 nm (groups 1 and 2, n = 11), 25 nm (group 3, n = 8), 50 nm (group 4, n = 16), and 100 nm (group 5, n = 15) were implanted into an interlamellar corneal pocket of the stroma. The implanted membranes ranged in thickness from 6 nm to 15 nm and were between 8 mm and 12 mm in diameter. Animals were monitored for clinical signs of intolerance to the implants. RESULTS: At 1 month, thinning and ulceration had occurred in the epithelium and the anterior stroma of all animals in groups 1 and 2; epithelial changes, anterior stromal thinning, and ulceration had developed in 75% of animals of group 3; 50% of animals showed vascularization and only 7% showed epithelial degeneration in group 4; and local anterior stromal thinning was observed in 7% of animals in group 5, indicating clinical acceptance of the implanted membrane. In the long term (greater than 50 days), 30% and 73% of the group 4 and 5 corneas, respectively, were clinically quiet. Analysis of the failure times indicated an inverse relation between failure rate and pore size: less than 15 nm > 25 nm > 50 nm > 100 nm. The difference between the 100-nm and 50-nm membranes was significant (P = 0.03). CONCLUSIONS: A corneal implant must have a porosity greater than that provided by 50-nm membranes. The 100-nm membranes used in this study establish the porosity needed to satisfy the nutritional requirements of the cornea.

Dialysis neutropenia: the role of the cytoskeleton.

Dialysis neutropenia is the result of pulmonary sequestration of neutrophils after complement activation by the dialyzer membrane. Increased expression of neutrophil adhesion receptors, such as CD11b/CD18, suggests that neutrophil adhesion to the capillary endothelium is a possible mechanism. An alternative hypothesis is that the complement fragment C5a modulates neutrophil mechanical properties via the cytoskeleton-largely filamentous actin (F-actin)-stiffening them and thereby slowing their passage through the pulmonary capillaries. To investigate this hypothesis, we developed an assay to measure the F-actin content of neutrophils in whole blood using flow cytometry and the stain NBD-phallacidin. We measured neutrophil F-actin content during hemodialysis of patients with polysulfone (N = 6), Hemophan (N = 6), and Cuprophan membranes sterilized with either ethylene oxide (N = 5) or steam (N = 6). Cell counts, neutrophil and monocyte CD11b expression and plasma C5a concentrations were also measured. The results confirm the strong relationship between the degree of neutropenia, increases in CD11b expression and plasma C5a levels reported by previous researchers. Modulation of the F-actin content of neutrophils was also strongly related to C5a levels, indicating that the neutrophil cytoskeleton is active during dialysis. Modeling of cell counts suggests that with Cuprophan a substantial fraction of neutrophils and monocytes are sequestered before they even pass through the dialyzer, suggesting some form of systemic activation of these cells. Evidence for systemic activation was also seen in measurements of F-actin content, but not CD11b expression, a finding that strengthens the case for the involvement of the cytoskeleton in dialysis neutropenia.

Oxygen permeability of a new type of high Dk soft contact lens material.

Lotrafilcon A is a biphasic block-copolymer, comprising a highly permeable siloxane-based polymeric phase, coupled with a water phase (hydrogel phase). The high oxygen permeability of this material, and the fact that it is a hydrogel, places it outside the applicability of both the polarographic ISO standard and coulometric ISO draft standards for contact lens Dk determination. The oxygen permeability (Dk) and transmissibility (Dk/t) of lotrafilcon A lenses were determined by an adaptation of the standard coulometric method. Lenses with a thickness (t) range from 30 microns to over 300 microns were measured in a liquid-to-gas and a gas-to-gas configuration in an effort to combine features of the ISO (draft) standards to yield a valid measurement of the intrinsic material's oxygen transmission characteristics. The following results, for lotrafilcon A, are the mean values and SE for the oxygen permeability coefficient (Dk) determined at 34 degrees C: 140 +/- 2 barrer with water overlay (liquid-to-gas) at 2100 rpm stirring speed, 150 +/- 5 barrer with water overlay (liquid-to-gas) at "infinite" stirring speed, and 170 +/- 2 barrer without water overlay (gas-to-gas), where barrer = 10(-11) (mlO2.cm)/(sec.cm2.mm Hg). Clinical lenses produced from this material are expected to be in the 60 to 90 microns thickness range, with an average center thickness of 80 microns. Given that the Dkmaterial is greater than or equal to 140 barrers, for this material, a parallel sided lens of 80 microns central thickness would exhibit a lens transmissibility (Dk/t) of at least 170 x 10(-9) mlO2/(sec.cm2.mm Hg) at 34 degrees C. This transmissibility is well in excess of the 87 x 10(-9) mlO2/(sec.cm2.mm Hg) value postulated to be sufficient to prevent overnight lens-induced corneal swelling and places it in the hyper-permeable material category.

Anterior cruciate ligament graft fixation. Initial comparison of patellar tendon and semitendinosus autografts in young fresh cadavers.

The initial biomechanical properties of semitendinosus and patellar tendon autografts and their fixation strengths were investigated. Twenty fresh cadaveric knees from donors under 42 years of age were used in the study. After removing all soft tissues other than the anterior cruciate ligament, we determined the ultimate tensile strength (2195 +/- 427 N) and stiffness (306 +/- 80 N/mm) of the anterior cruciate ligament in nine knees. In six knees, anterior cruciate ligaments were reconstructed using an autologous patellar tendon graft with proximal and distal interference fit screws; this resulted in an ultimate tensile strength of 416 +/- 66 N. Five knees were reconstructed with quadruple-stranded (double-looped) semitendinosus tendons fixed proximally by a titanium button and braided tape and distally by tibial post screw. This resulted in an ultimate tensile strength of 612 +/- 73 N, which was significantly higher than the strength in the patellar tendon group. Graft stiffness did not differ between the groups and was 47 +/- 19 N/mm (N = 11). This study demonstrates that the reconstructed knees had only 20% to 30% of the ultimate tensile strength of the normal anterior cruciate ligament. In summary, the semitendinosus reconstruction using a button for proximal fixation is, at the time of surgery, approximately 50% stronger than patellar tendon reconstructions with similar stiffness.

Design of hollow fiber modules for uniform shear elution affinity cell separation.

Large-scale monoclonal antibody based systems for the selection of cell subsets will play a prominent role in the development of hematotherapy and graft engineering. Hollow fiber systems for affinity cell separation rely on the generation of uniform fluid shear stress at the lumenal attachment interface. Potential mechanisms for nonuniformity of lumenal wall shear stress are fiber wall permeation fluxes driven by the pressure gradient along individual fibers and the influence of inlet header dynamic pressure on the radial distribution of axial flow within the fiber module. Dimensional analysis and numerical solution of the flow field within the lumen of a hollow fiber module illustrate the main physical criteria for design of hollow fiber modules. There will be a nearly uniform distribution of flow within the fiber bundle provided that the dynamic inlet pressure is small in comparison with the pressure drop along fibers. Fiber wall permeation fluxes will have a negligible effect on axial flow rate for nonporous membranes such as Cuprophan.

Resorbable and non-resorbable augmentation devices for tenorrhaphy of xenografts in extensor tendon deficits: 12 week study.

Resorbable (poly-L-lactide) and non-resorbable (polyethylene terephathalate) tendon augmentation devices (TAD) in conjunction with a pericardial adhesion barrier, were designed to strengthen tenorrhaphies and were evaluated in an ovine extensor tendon deficit model in a short term study. Fifteen centimetres of tendon were resected and replaced with kangaroo tail tendon xenografts that had been cross-linked with 0.075% glutaraldehyde (GA) at 4 degrees C for one or seven days. Compared with tenorrhaphies performed with Kessler sutures alone, both types of TAD were more effective at preventing tenorrhaphy dehiscence, and thus maintaining tendon function. Furthermore, tensile strength of TAD tenorrhaphies increased significantly between zero and twelve weeks. For xenografts cross-linked in GA for one day, the tensile strength of tenorrhaphies with the resorbable TAD rose from 38 +/- 9 N at time zero, to 116 +/- 46 N at twelve weeks, while non-resorbable TAD tenorrhaphy strength at time zero was 42 +/- 16 N and 99 +/- 27 N at twelve weeks. For xenografts cross-linked with GA for seven days, similar increases in tensile strength of tenorrhaphies, with the two types of TAD were found. As there was no significant difference in mechanical performance or tissue response between the two TAD types in the first 12 weeks, use of the resorbable poly-L-lactide device may be advantageous clinically. Tensile strengths of midsections of the tendon xenograft cross-linked for 7 days was not significantly diminished 12 weeks after implantation and these xenografts were partially remodelled around the periphery. However, the tensile strength of xenografts cross-linked for one day declined significantly between time zero (319 +/- 80 N) and twelve weeks (239 +/- 92 N), suggesting that this degree of cross-linking was inadequate for maintenance of mechanical strength. Evaluation of the performance of tenorrhaphy augmentation devices with xenografts, over a longer implantation period, is required to further understand their usefulness for reconstruction of traumatic tendon injuries.

In-vivo degradation of polyurethanes: transmission-FTIR microscopic characterization of polyurethanes sectioned by cryomicrotomy.

A combination of cryomicrotomy and transmission Fourier transform infrared (FTIR) microscopy was used to investigate chemical changes in unstrained sheets of Pellethane 2363-80A, Tecoflex EG80A and Biomer caused by biodegradation (18 month subcutaneous ovine implant). Cryomicrotomy was used to obtain thin sections (ca. 2.5 microm) from the surface into the bulk, parallel to the plane of the surface. FTIR microscopy was then used to obtain infrared absorbance spectra in the range 4000-600 cm(-1). Comparisons between the infrared spectra (by spectral subtraction) from implant surface, implant interior and non-implanted controls were used to detect chemical changes. Scanning electron microscopy was used to assess microstructural changes owing to biodegradation. Biodegradation in Biomer was observed as uniform pitting and superficial fissuring (<2.0 microm depth) over the implant surface. Biodegradation in Pellethane 2363-80A and Tecoflex EG 80A was observed as severe localized embrittlement of the surface with fissures infiltrating up to 40 microm into the bulk. The chemical changes associated with biodegradation were observed as localized oxidation of the soft segment and hydrolysis of the urethane bonds joining hard and soft segments. Tecoflex EG80A was also found to be susceptible to localized hydrolysis of the urethane bond within the aliphatic hard segment. Biomer showed evidence of a significant non-specific degradation in the non-implanted wet control (37 degrees C phosphate buffered saline at pH 7.3) samples and in the implant bulk.

Hollow-fibre affinity cell separation system for CD34+ cell enrichment.

A hollow-fibre immunoadsorption system has been developed for the purification of CD34+ cells from mononuclear cells. This cell separation technique is based on the use of uniform surface fluid shear stress to fractionate cells that attach to the inside surface of hollow fibres. Monoclonal antibody to the CD34 antigen was covalently coupled to the lumenal surface of cuprophan minidialysers (surface area 220 cm2). After the selective adsorption of CD34+ cells (28 min), a depleted fraction was collected at 5 dynes/cm2 followed by washes at 10 and 25 dynes/cm2. Antigen-positive cells were recovered after incubation with chymopapain. The device was tested by using peripheral blood mononuclear cells from seven patients who had received granulocyte colony-stimulating factor and chemotherapy. The average number of cells processed was 1.3 +/- 0.2 x 10(8) (+/- S.E.M.), and the preselection incidence of CD34+ cells ws 1.6 +/- 0.6% (range 0.21-4.13%; n = 7). The enrichment purity was 94.4 +/- 3.1%, and 61 +/- 9% of input CD34+ cells were recovered in the enriched fraction (n = 4). Enrichment resulted in a 3.3 +/- 0.1% log10 depletion of CD34- cells (n = 4). Hollow-fibre affinity cell separation has potential as a medium to large-scale cell enrichment technology.

Ex vivo manipulation of cell subsets for cell therapies.

Large-scale cell separation and ex vivo expansion technologies will form the basis for development of new cellular products for the treatment of cancer and fatal viral diseases. The cell subsets that are likely to play a significant role in cellular therapy include hematopoietic stem cells, platelet and granulocyte precursors, cytotoxic lymphocytes, and genetically modified hematopoietic or lymphoid precursors. Cell enrichment techniques are required to eliminate tumor cells from autologous stem cell grafts and to reduce the size of culture systems required for expansion or gene transfection. The consumption of expensive culture components such as cytokines and serum may be reduced by the use of perfusion bioreactor devices. Methods that have been developed for the production of cell subsets for cellular therapy are reviewed.

Performance of small diameter synthetic vascular prostheses with confluent autologous endothelial cell linings.

Autologous grafts are superior to their synthetic counter-parts for grafting arteries smaller than 6-mm diameter both in terms of acute thrombogenicity and chronic intimal hyperplasia. Endothelial cell (EC) coating of the blood contacting surface may reduce thrombogenicity of synthetic small diameter vascular prostheses. In this study, the survival of EC monolayers on synthetic 4-mm diameter arterial prostheses over short-term implantations (< or = 6 weeks) was examined. Graft types examined were expanded polytetra-fluoroethylene (ePTFE) and microporous polyurethane (PU). Lumenal coverage with ECs was achieved by culturing ovine ECs on prostheses treated by either physical adsorption or covalent binding of ovine fibronectin (Fn). An ovine carotid interposition model was used to examine the performance of EC coated ePTFE and microporous PU over implantation periods of 1, 3, and 6 weeks. Outcomes assessed at the end of each experiment were graft patency, area covered by ECs, and thrombus free surface area (TFSA). Fn concentration, cell density at the time of coating and prostacyclin production in vitro were similar for both graft types. Occlusion occurred more frequently in unseeded grafts compared with EC coated grafts over 3 and 6 week implantation periods; however, the difference was not significant (p = 0.099). In prostheses precoated with ECs, approximately 40-60% of the surface area remained covered with endothelial-like cells following the first postoperative week. Recovery of EC layers occurred rapidly thereafter with 80-90% coverage at 3 weeks. TFSA remained low in comparison to EC cover in these prostheses until between 3 and 6 weeks postoperatively, suggesting a lag phase in recovery of EC function of seeded cells. In contrast, EC cover of unseeded prostheses only achieved 10-30% at 3 weeks, primarily by pannus EC ingrowth from the adjacent artery. TFSA of unseeded grafts increased in direct proportion to EC cover over time suggesting that there was no lag phase in function of these ingrowing cells.