Vein grafts: haemodynamic forces on the endothelium--a review

Osteopontin Activation and Microcalcification in Venous Grafts Can Be Modulated by Dexamethasone

BACKGROUND

Osteopontin has been implicated in vascular calcification formation and vein graft intimal hyperplasia, and its expression can be triggered by pro-inflammatory activation of cells. The role of osteopontin and the temporal formation of microcalcification in vein grafts is poorly understood with a lack of understanding of the interaction between haemodynamic changes and the activation of osteopontin.

METHODS

We used a porcine model of vein interposition grafts, and human long saphenous veins exposed to ex vivo perfusion, to study the activation of osteopontin using polymerase chain reaction, immunostaining, and 18F-sodium fluoride autoradiography.

RESULTS

The porcine model showed that osteopontin is active in grafts within 1 week following surgery and demonstrated the presence of microcalcification. A brief pretreatment of long saphenous veins with dexamethasone can suppress osteopontin activation. Prolonged culture of veins after exposure to acute arterial haemodynamics resulted in the formation of microcalcification but this was suppressed by pretreatment with dexamethasone. 18F-sodium fluoride uptake was significantly increased as early as 1 week in both models, and the pretreatment of long saphenous veins with dexamethasone was able to abolish its uptake.

CONCLUSIONS

Osteopontin is activated in vein grafts and is associated with microcalcification formation. A brief pretreatment of veins ex vivo with dexamethasone can suppress its activation and associated microcalcification.

Impact of the no-touch harvesting technique on the vessel diameter of saphenous vein grafts for coronary artery bypass grafting

Objectives

To explore the impact of the no-touch harvesting technique on the vessel diameter of saphenous vein grafts.

Methods

This retrospective, single-center study enrolled 166 patients who underwent isolated coronary artery bypass grafting using saphenous vein grafts. Saphenous vein grafts were harvested conventionally in 83 patients (conventional group) and using the no-touch technique in 83 patients (no-touch group). We analyzed graft patency and the vessel diameters of saphenous vein grafts in the pre- and postoperative states. The diameter mismatch between the saphenous vein grafts and the coronary artery at the anastomotic site was also measured; preoperative diameter was measured using ultrasound imaging, and the postoperative diameter was measured using electrocardiogram-gated enhanced computed tomography.

Results

A total of 135 saphenous vein grafts (66 and 69 grafts in the conventional and no-touch groups, respectively) were evaluated for postoperative patency. Graft patency was equivalent in the 2 groups (conventional, 96.9% vs no-touch, 100%; P = .24). A detailed evaluation was performed in 109 saphenous vein grafts (52 and 57 grafts in the conventional and no-touch groups, respectively). Saphenous vein graft diameter was significantly distended in the conventional group (preoperative, 2.6 ± 0.7 mm vs postoperative, 3.4 ± 0.5 mm; P < .0001). However, saphenous vein graft diameter did not change in the no-touch group (preoperative, 2.9 ± 0.4 mm vs postoperative 2.8 ± 0.4 mm, P = .33). The diameter mismatch was significantly smaller in the no-touch group (conventional 1.4 ± 0.6 mm vs no-touch 1.0 ± 0.4 mm, P < .0001).

Conclusions

The no-touch technique avoids the expansion of graft diameter and diameter mismatch between the saphenous vein grafts and coronary artery.

Surface modification of decellularized bovine carotid arteries with human vascular cells significantly reduces their thrombogenicity

BACKGROUND

Since autologous veins are unavailable when needed in more than 20% of cases in vascular surgery, the production of personalized biological vascular grafts for implantation has become crucial. Surface modification of decellularized xenogeneic grafts with vascular cells to achieve physiological luminal coverage and eventually thromboresistance is an important prerequisite for implantation. However, ex vivo thrombogenicity testing remains a neglected area in the field of tissue engineering of vascular grafts due to a multifold of reasons.

METHODS

After seeding decellularized bovine carotid arteries with human endothelial progenitor cells and umbilical cord-derived mesenchymal stem cells, luminal endothelial cell coverage (LECC) was correlated with glucose and lactate levels on the cell supernatant. Then a closed loop whole blood perfusion system was designed. Recellularized grafts with a LECC > 50% and decellularized vascular grafts were perfused with human whole blood for 2 h. Hemolysis and complete blood count evaluation was performed on an hourly basis, followed by histological and immunohistochemical analysis.

RESULTS

While whole blood perfusion of decellularized grafts significantly reduced platelet counts, platelet depletion from blood resulting from binding to re-endothelialized grafts was insignificant (p = 0.7284). Moreover, macroscopic evaluation revealed thrombus formation only in the lumen of unseeded grafts and histological characterization revealed lack of CD41 positive platelets in recellularized grafts, thus confirming their thromboresistance.

CONCLUSION

In the present study we were able to demonstrate the effect of surface modification of vascular grafts in their thromboresistance in an ex vivo whole blood perfusion system. To our knowledge, this is the first study to expose engineered vascular grafts to human whole blood, recirculating at high flow rates, immediately after seeding.

Current understanding of intimal hyperplasia and effect of compliance in synthetic small diameter vascular grafts

Despite much effort, synthetic small diameter vascular grafts still face limited success due to vascular wall thickening known as intimal hyperplasia (IH). Compliance mismatch between graft and native vessels has been proposed to be one of a key mechanical factors of synthetic vascular grafts that could contribute to the formation of IH. While many methods have been developed to determine compliance both in vivo and in vitro, the effects of compliance mismatch still remain uncertain. This review aims to explain the biomechanical factors that are responsible for the formation and development of IH and their relationship with compliance mismatch. Furthermore, this review will address the current methods used to measure compliance both in vitro and in vivo. Lastly, current limitations in understanding the connection between the compliance of vascular grafts and the role it plays in the development and progression of IH will be discussed.

Comparison of Conventional and No-Touch Techniques in Harvesting Saphenous Vein for Coronary Artery Bypass Grafting in View of Endothelial Damage

<p><b>Background:</b> Optimization of saphenous vein patency for myocardial revascularization.</p><p><b>Objective:</b> The goal of this study was to present the no-touch technique of saphenous vein preparation. This technique consists of harvesting the vein with a pedicle of surrounding tissue, which protects the vein from distension pressure.</p><p><b>Methods:</b> We performed a prospective, randomized study that compared 2 techniques for harvesting saphenous vein-conventional and no-touchin 40 patients undergoing coronary artery bypass grafting. We carried out a morphologic study of the endothelium with the aid of light and transmission electron microscopy and an immunohistochemical assessment to identify adenosine, inducible nitric oxide synthase (iNOS), and vascular endothelial growth factor (VEGF) in the vein wall.</p><p><b>Results:</b> The integrity of endothelial cell and all vascular layers was maintained better with the no-touch technique than with the conventional procedure. The immunohistochemical assessment revealed that adenosine receptor, iNOS, and VEGF immunoexpression levels were normal or lower in the no-touch group than in the conventional-harvest group, as shown by the staining densities in all layers of the vein wall.</p><p><b>Conclusion:</b> Endothelial integrity and adenosine, iNOS, and VEGF immunoreactivities were better preserved when the no-touch technique was used for vein graft harvesting. The mechanical protection provided by the cushion of surrounding tissue in the no-touch group and the vasorelaxation and thromboresistant activities of nitric oxide may be responsible for the reduction in vasospasms and the improved patency rate.</p>

Glucocorticoids in adult cardiac surgery; old drugs revisited

Glucocorticoids can play a pivotal role in modulating different immune responses. The role of glucocorticoids in cardiac surgery is still controversial as many surgeons are concerned about the potential side effects.

In this review, we looked at the role of glucocorticoid administration in modulating postoperative inflammatory responses, atrial fibrillation (AF) and intimal hyperplasia and whether glucocorticoid use is associated with a significant increase in undesirable postoperative complication.

A parametric numerical investigation on haemodynamics in distal coronary anastomoses

Anastomotic haemodynamics, which plays an important role in the performance of bypass graft, is known to be profoundly affected by the diameter ratio (Phi) and angle (alpha) between the graft and host artery in the peripheral region. We hypothesize that these geometric factors would play similar roles in distal coronary anastomoses and that they could be improved for clinical applications through parametric studies. Anastomotic models covering a range of Phi (1:1, 1.5:1 and 2:1) and alpha (15 degrees , 30 degrees , 45 degrees and 60 degrees ) were investigated numerically in physiological coronary flow conditions. The transient flow patterns, cycle-averaged wall shear stress (WSS), oscillatory shear index (OSI), spatial and temporal WSS gradients (SWSSG and TWSSG) were compared. Results show a stronger influence of Phi than alpha on haemodynamics in distal coronary artery anastomoses. Substantially higher SWSSG and TSSWG occur on the artery floor when Phi=1:1 compared to larger Phi. High levels of OSI occur in critical regions when Phi=1:1 and 2:1. The largest area of high OSI is found in the anastomotic region when alpha=15 degrees , whereas the highest level of SWSSG appears on the artery floor when alpha=60 degrees . The study suggests the clinical relevance of optimizing geometric parameters of coronary anastomoses to improve their haemodynamic performance. We speculate that for a distal coronary anastomosis with a 20:80 proximal-distal flow division ratio maintained in the host artery, Phi=1.5 and alpha=30-45 degrees would enhance its long-term performance.

Vein Graft Harvesting Induces Inflammation and Impairs Vessel Reactivity

BACKGROUND

Saphenous veins are often used for coronary artery bypass grafting (CABG), but loss of patency is a problem. The surgical procedure may contribute to graft injury. Our aim was to study the impact of surgical handling of saphenous veins on graft inflammation and vascular function.

METHODS

Biopsy samples of saphenous veins were taken from 9 patients undergoing elective CABG at the start of vein harvesting (open technique) and after the last proximal anastomosis was sutured. Messenger RNA was extracted and amplified with semiquantitative reverse transcription polymerase chain reaction. Gene expression of proinflammatory cytokines (tumor necrosis factor-alpha, interleukin-1beta), leukocyte adhesion molecules (E-selectin, intercellular adhesion molecule-1), and vasoactive substances (endothelin-1, inducible and endothelial nitric oxide synthase) was investigated. Translocation of nuclear factor-kappaB (NFkappaB) was evaluated with electrophoretic mobility shift assay. Immunostaining for von Willebrand factor was performed to evaluate loss of endothelium, and in vitro vein reactivity to phenylephrine and endothelin-1 was studied.

RESULTS

Gene expression of cytokines and leukocyte adhesion molecules increased after graft harvesting and storage, whereas vasoactive substances did not change. Nuclear translocation of NFkappaB occurred after surgical handling, concurrent with partial loss of endothelium and impaired contractile function.

CONCLUSIONS

Standard surgical handling of vein grafts induces NFkappaB-driven inflammation in the vessel wall and impairs vascular function. This may potentially contribute to both early and late graft occlusion.

Arterial flow induces changes in venous endothelium which are modified by calcium channel blockers

INTRODUCTION

Adaptation of saphenous vein to arterial flow may be critical to the results of bypass. The present paper summarizes work recently presented as a Hunterian Lecture. Work includes in vitro investigation of the response of saphenous vein endothelium to arterial flow and assessment of its clinical importance using a cohort of patients undergoing vein bypass surgery.

METHODS

Freshly excised human saphenous vein segments were placed in an in vitro flow circuit to simulate arterial and venous flow conditions. Changes in the endothelial expression of proteins were assessed using a combination of immunohistochemistry and Western blotting. The role of ion channels in the changes seen induced by arterial flow in the saphenous vein endothelium was assessed by addition of ion channel blocking medication to the medium perfusing vein segments. A cohort of patients undergoing vein bypass surgery were followed to assess graft patency and the influence of prescribed medication on its outcome.

RESULTS

After arterial flow conditions, the staining area for the endothelial adhesin ICAM-1 and nitric oxide synthase were increased, while that of the anticoagulant protein thrombomodulin was decreased. The concentration of the important stimulant of the clotting cascade tissue factor was unaffected by arterial flow. These changes were modulated by the addition of ion channel blocking drugs to the vein perfusate. In particular, nifedipine abolished the reduction in thrombomodulin, but increased the amount of tissue factor. In a series of 236 patients undergoing primary infra-inguinal vein grafts, prescription of calcium channel blocker was associated with improved primary patency.

CONCLUSIONS

Important changes in the venous endothelium are induced by arterial flow. Ion channel blocking drugs have the potential to modulate these responses.

Urokinase, tissue-type plasminogen activator and plasminogen activator inhibitor-1 expression in severely stenosed and occluded vein grafts with thrombosis

Intimal hyperplasia and subsequent thrombotic occlusions limit the success of vascular reconstructive procedures. Plasminogen activation in situ may be an important factor affecting re-stenosis of the graft. Tissue specimens from eight patients with failing or failed infra-inguinal vein bypasses and three specimens from normal veins were harvested to study urokinase-type plasminogen activator (u-PA), tissue-type plasminogen activator (t-PA) and plasminogen activator inhibitor-1 (PAI-1) by in situ hybridization and immunohistochemistry. The possible presence of thrombi was monitored by platelet and fibrin-specific stainings. In occluded grafts, platelet endothelial cell adhesion molecule (PECAM-1) antibody stained the thrombi but not the endothelial area, indicating the absence of endothelium. Platelet glycoprotein (GP) IIb/IIIa co-localized with PECAM-1 and, furthermore, GP IIb/IIIa staining was positive on the vein walls with thrombi and to some extent in the grafts without thrombi. PAI-1 and u-PA were uniformly upregulated in intimal thickening in grafts without thrombus. In organized thrombi, enhanced u-PA, t-PA and PAI-1 reactivity was detected in the ingrowing subendothelium. In non-occluded grafts with small thrombi, u-PA expression was enriched beneath microthrombi co-localizing with the graft wall injury, while PAI-1 was scattered in the (sub)endothelium. We conclude that fibrinolytic system is upregulated at sites of graft stenosis, and local proteolytic degradation of the graft wall associates with thrombus formation.

Is there a haemodynamic advantage associated with cuffed arterial anastomoses?

The development of intimal hyperplasia at arterial bypass graft anastomoses is a major factor responsible for graft failure. A revised surgical technique, involving the incorporation of a small section of vein (vein cuff) into the distal anastomosis of PTFE grafts, results in an altered distribution of intimal hyperplasia and improved graft patency rates, especially for below-knee grafts. Numerical simulations have been conducted under physiological conditions to identify the flow behaviour in a typical cuffed bypass model and to determine whether the improved performance of the cuffed system can be accounted for by haemodynamic factors. The flow patterns at the cuffed anastomosis are significantly different to those at the conventional end-to-side anastomosis. In the former case, the flow is characterised by an expansive, low momentum recirculation within the cuff. Separation occurs at the graft heel, and at the cuff toe as the blood enters the recipient artery. Wall shear stresses in the vicinity of the cuff heel are low, but high shear stresses and large spatial gradients in the shearing force act on the artery floor during systole. In contrast, a less disturbed flow prevails and the floor shear stress distribution is less adverse in the conventional model. In conclusion, aspects of the anastomotic haemodynamics are worsened when the cuff is employed. The benefits associated with the cuffed grafts may be related primarily to the presence of venous material at the anastomosis. Therefore, caution is advised with regard to the use of PTFE grafts, pre-shaped to resemble a cuffed geometry.

Numerical investigation of the haemodynamics at a patched arterial bypass anastomosis

Intimal hyperplasia at arterial bypass graft anastomoses is a major factor responsible for graft failure. A revised surgical technique, incorporating a Taylor vein patch into the distal anastomosis of PTFE grafts, results in a decrease in intimal hyperplasia and improved patency rates. Numerical simulations of pulsatile, non-Newtonian blood flow through life-like femorodistal bypass models have been performed to determine whether haemodynamic benefits arise from the modified geometry of the Taylor anastomosis. In a conventional bypass, the distal anastomotic flow exhibited considerable spatial and temporal variations. Steep spatial gradients in the shearing force acted along the floor during systole. The effect of the Taylor geometry was to reduce gradually the momentum of the blood approaching the junction. Thus, flow disturbances were abated, undesirable flow separation at the toe was diminished, and a less adverse floor shear stress distribution prevailed in that case. Intimal thickening should be alleviated at the toe in the Taylor model where separation is reduced, and where the thrombogenic graft surface is replaced with a vein patch. Intimal hyperplasia on the floor may be inhibited in the Taylor model due to more favourable shear stresses. The improved flow through the patched anastomosis should contribute to its enhanced performance.

Computational and experimental simulations of the haemodynamics at cuffed arterial bypass graft anastomoses

The development of intimal hyperplasia at arterial bypass graft anastomoses is a major factor responsible for graft failure. A revised surgical technique, involving the incorporation of a small section of vein (vein cuff) into the distal anastomosis of polytetrafluoroethylene (PTFE) grafts, alters the distribution of intimal hyperplasia and improves graft performance. Numerical and in vitro flow visualization experiments have been conducted to identify the flow behaviour in the cuffed bypass model and to determine whether the improved performance of the cuffed system can be accounted for by haemodynamic factors. The flowfield at the cuffed anastomosis is characterized by an expansive recirculation. Separation occurs at the graft heel, and at the cuff toe as the blood enters the recipient artery. Wall shear stresses in the vicinity of the cuff heel are low, but high shear stresses and large spatial gradients in the shearing force act for a time on the artery floor. In the conventional model, a less disturbed flow prevails while the gradients of shear stress on the floor are smaller. Aspects of the anastomotic haemodynamics are worsened when the cuff is employed. The superior patency rates of cuffed bypasses may not be explained purely on the basis of local haemodynamic factors.

Arterial flow induces changes in saphenous vein endothelium proteins transduced by cation channels

OBJECTIVES

expression of leukocyte adhesins and proteins controlling thrombosis is likely to be an important determinant of graft patency early following vein bypass. We have previously demonstrated rapid increase in endothelial ICAM-1 and nitric oxide synthase (eNOS) concentrations in human saphenous vein exposed to arterial flow. The aim of this study was to investigate whether ion-channel-blocking drugs could alter these flow-induced changes.

METHODS

human saphenous vein segments, freshly excised from patients, were placed in a validated in vitro circuit using flow conditions shown to simulate arterial or venous circulations for 90 min, in the presence or absence of ion-channel blockers. The concentrations of ICAM-1, VCAM-1, eNOS and tissue factor (TF) were assessed by quantitative immunohistochemistry in vein exposed to flow and compared with that in freshly excised vein from the same patient. The endothelial protein concentration was calculated as the mean area of staining as percentage of that for the control protein CD31, using computer-aided image analysis.

RESULTS

after arterial flow conditions the area ratio of ICAM-1 increased from 21.4+/-1.4 to 44.6+/-2.0%, of eNOS increased from 50.0+/-5.6 to 70.1+/-5.0%, of VCAM-1 decreased from 16.6+/-3.4 to 3.6+/-1.0%, whereas TF staining area ratio was unchanged. Inclusion of the non-selective K(+)channel blocker, tetraethylammonium, in the arterial perfusion solution abolished all these arterial flow-induced changes. Inclusion of the K(+)ATP channel blocker, glibenclamide, selectively abolished the arterial flow-induced changes in ICAM-1 and VCAM-1. Inclusion of the calcium channel blocker, nifedipine, abolished the arterial flow-induced changes in eNOS and VCAM-1 but increased the TF staining area ratio from 3.0+/-0.4 to 8.5+/-0.7%, p=0.01. Inclusion of the stretch-activated cation-channel blocker, gadolinium, enhanced the arterial flow-induced increase in eNOS, but prevented the arterial flow-induced increase in ICAM-1.

CONCLUSIONS

perfusion of veins under arterial flow conditions with gadolinium was associated with low endothelial concentrations of ICAM-1, VCAM-1 and TF, but high levels of eNOS. Such a concentration of endothelial proteins may be advantageous in newly implanted vein grafts. In contrast, nifedipine could have adverse effects by promoting increase in TF concentration.