Expansion of small-diameter abdominal aortic aneurysms is not reflected by the release of inflammatory mediators IL-6, MMP-9 and CRP in plasma.

OBJECTIVES: The aim of this study was to evaluate a possible correlation between plasma levels of interleukin-6 (IL-6), metalloproteinase-9 (MMP-9) and C-reactive protein (CRP) and the expansion of small abdominal aortic aneurysms (AAAs). DESIGN: Patients were selected from a prospective randomised clinical trial and categorised in two groups, in which one group received active treatment (azithromycin) and the other received placebo. No statistical difference in the expansion rate of AAAs between the groups was found and the two groups were considered as one cohort in the present study. MATERIAL AND METHODS: In this study, 213 patients with AAAs between 35 and 49 mm were followed-up with ultrasound examination every 6th month. Blood samples were taken on two occasions (6 months apart). IL-6 and MMP-9 were analysed on one occasion using Quantikine analysing kits (R&D Systems, Inc., USA). CRP was analysed using sensitive-CRP method. RESULTS: Levels of IL-6, MMP-9 and CRP did not correlate with AAA expansion. Neither was there any correlation between statin medication and changes in MMP-9 levels over the 6-month period. Patients on statins had a lower expansion rate than those not taking statins: 0.16 versus 0.25 cm per year. CONCLUSION: No correlation was found between levels of circulating IL-6, MMP-9, CRP and the expansion of small-diameter AAAs, indicating no clinical use of these markers in AAA surveillance.

Risk factors for complications after carotid endarterectomy--a population-based study.

OBJECTIVES: The overall benefit of carotid endarterectomy (CEA) is dependent on the outcome from the procedure. However, many reports are from selected centres and not population-based. The aim of this study was to assess the 30-day complication rate for a whole country and also to determine independent risk factors for serious complications. MATERIALS AND METHODS: One thousand five hundred and eighteen CEA were retrospectively reviewed, covering principally all the CEAs in Sweden, during a three year period. Indications for surgery were; minor stroke 34%, TIA 34%, amaurosis fugax 18%, asymptomatic 11% and others 3%. Data were collected from the Swedish Vascular Registry (Swedvasc). Combined cohort and case-control methodology was used. RESULTS: Registered complications were; 43 permanent strokes, 32 transient strokes (<30 days), 18 TIA/amaurosis fugax and 22 deaths (seven fatal stokes). In the cohort study, the 30-day permanent stroke and death rate were 4.3% (65/1518). Significant risk factors in multivariate analyses were the indication for surgery (minor stroke vs. other indications) (p=0.02, RR=1.38), diabetes (p=0.02, RR=1.41), cardiac disease (p<0.01, RR 1.43) and operation at a university hospital (p=0.02, RR=1.39). In the case-control study comparing the 65 cases of permanent stroke and/or death with 130 matched controls the only significant risk factor was contralateral occlusion (p<0.01, OR=5.27). One patient (1/130) with a permanent stroke was wrongly reported as a local neurological complication (facial paresis). CONCLUSION: This national audit demonstrated population-based data on complication rates after CEA well comparable with previous randomised trials. The validity of the Swedvasc data was confirmed. Combined cohort and case-control methodology was useful in analysing risk factors for serious perioperative complications.

Local haemodynamic changes during carotid endarterectomy--the influence on cerebral oxygenation.

OBJECTIVES: To characterize carotid bifurcation haemodynamics and cerebral oxygenation during clamping and at reperfusion after carotid endarterectomy (CEA). MATERIALS AND METHODS: Sixty-two patients with a symptomatic high-grade stenosis of the internal carotid artery (ICA), who underwent CEA under general anaesthesia, were studied prospectively. Measurements of stump-pressure, volume flow (transit time flowmetry) and changes in cerebral oxygenation (near-infrared spectroscopy (NIRS)) were performed. Selective shunting was based on stump pressure only. RESULTS: Stump pressure correlated with both ICA flow before clamping (r=0.45; p=0.03) and changes in cerebral oxygenation (rSO2) during clamping (r=0.61; p=0.002), the latter was reversed by shunt placement. ICA flow before clamping also correlated with changes in rSO2 during clamping (r=0.41; p=0.01). CONCLUSION: Measurements with transit time flowmetry and cerebral oximetry are technically easy and help to determine the need for selective shunting during CEA. High ICA flow before clamping in combination with a low stump pressure usually indicates the need for a shunt. Volume flow measurements may also be useful in the quality assessment of the CEA.

Activation of coagulation and fibrinolytic systems in patients with CLI is not normalized after surgical revascularisation.

OBJECTIVE: To study the activation of coagulation and fibrinolysis before, during and after surgical revascularisation in patients with critical limb ischemia (CLI). DESIGN: Prospective clinical study. MATERIALS AND METHODS: Forty patients with CLI underwent femoro-popliteal or femoro-distal reconstruction and were compared to a control-group. Measurements of prothrombin-fragment 1+2 (F1+2) and thrombin-antithrombin complex (TAT) assessed activation of coagulation. Fibrinolysis was determined by tissue plasminogen activator (tPA), plasminogen activator inhibitor (PAI-1) and fibrin degradation product (D-dimer). The inflammatory mediators: Interleukin 2 receptor (IL-2-rec), Interleukin 6 (IL-6), Interleukin 10 (IL-10) and Monocyte chemoattractant protein 1 (MCP-1) was also analysed. RESULTS: Patients (in 35 of the 40 reconstruction was possible) were operated upon using either vein (n=23) or ePTFE (n=12) grafts. Patients with CLI had a preoperative prothrombotic state as indicated by high TAT-levels and also ongoing fibrinolysis with high levels of t-PA and D-dimer. After reperfusion an ongoing prothrombotic state for the first week was demonstrated. A significant as well as defective fibrinolysis was also seen with increased levels of tPA and D-dimer unopposed by PAI-1 after one week and also after 30 days. Increased levels of inflammatory mediators IL-6, IL-10 and MCP-1 was observed after reperfusion and normalised after 30 days. CONCLUSION: This study demonstrates significant disturbances of both the coagulation and fibrinolytic systems before, during and after revascularisation for CLI. This was accompanied by release of inflammatory mediators. A prothrombotic state and increased fibrinolysis were evident also 30 days after successful revascularisation.

A comparative thrombogenicity study of heparin soaked fluoro-passivated polyester and ePTFE patches in sheep.

OBJECTIVES: to compare heparin soaked fluoro-passivated gelatine sealed polyester and expanded polytetrafluoroethylene (ePTFE) patches in a sheep model of acute platelet accumulation following patch angioplasty. MATERIALS AND METHODS: heparin soaked patches were placed in the carotid arteries of 9 sheep and autologous (111)Indium labelled platelets were infused. The patches were explanted two hours after the injection of labelled platelets. Median specimen radioactivity was calculated as a ratio of radioactivity in explanted and in 4 ml of blood. Explanted patches were also investigated by scanning electron microscopy (SEM). RESULTS: platelet accumulation was significantly greater on ePTFE patches. For both materials platelet accumulation was greater at the distal end compared to the proximal (p<0.05). SEM demonstrated more platelets as well as thicker thrombus layer on ePTFE-patches. CONCLUSION: in sheep carotid arteries, a fluoropassivated gelatine sealed polyester patch appears to result in less platelet accumulation when compared to ePTFE.

Arterio-ureteral fistula--a systematic review.

OBJECTIVE: To review published reports on arterio-ureteral fistula. METHOD: Literature search. RESULTS: Eighty cases were identified. Primary fistulas were mainly seen in combination with aortoiliac aneurysmal disease. Secondary fistulas were seen after pelvic cancer surgery, often with radiation, fibrosis and ureteral stenting or after vascular surgery with synthetic grafting. The dominating symptom is massive haematuria, often with circulatory impairment. The clue to a rapid and correct diagnosis is a high degree of suspicion. Most frequently diagnosis has been obtained through angiography or pyelography. When there is a ureteral stent manipulation it will often provoke bleeding and lead to diagnosis. The fistula must be excluded and a vascular reconstruction made. Most frequently this has been obtained through occlusion of the fistula and an extra-anatomic reconstruction (femoro-femoral crossover). Recently stent-grafting has been successfully used but follow-up is short. CONCLUSION: Arterio-ureteral fistula is rare and should be suspected in patients with complicated pelvic surgery and massive haematuria, especially where rigid ureteral stents have been placed.

Release of activin and follistatin during cardiovascular procedures is largely due to heparin administration.

Recent evidence has suggested that activin A complexed to its binding protein, follistatin, may be present on the surface of cells through their interaction with heparan sulfate proteoglycans. As heparin is used routinely in many cardiovascular procedures for its anticoagulation properties, it may also cause the release of heparin-binding growth factors, including activin and follistatin, from the vascular endothelium. We examined the effect of two cardiovascular procedures and the use of heparin directly on the circulating concentrations of activin A and follistatin. A rapid and robust release of activin A and follistatin occurred in the circulation of patients undergoing abdominal aortic aneurysm repair or carotid endarterectomy at the time of vessel clamping and administration of heparin (5000 IU). This release pattern was dissimilar to that of the inflammatory marker, interleukin-1beta. However, administering heparin (2500 IU) to coronary angiography patients produced a similar activin and follistatin response, whereas placebo-treated angiography patients had no response. These findings illustrate that the routine use of heparin in surgical procedures elicits a rapid and robust release of activin and follistatin. This has direct clinical relevance by potentially activating heparin-binding growth factors that are important in injury, hyperplasia, and restenosis of vessels.

Maintaining carotid flow by shunting during carotid endarterectomy diminishes the inflammatory response mediating ischaemic brain injury.

OBJECTIVES: to assess whether shunting during carotid reconstruction affects the release of inflammatory mediators from the ipsilateral hemisphere. MATERIALS AND METHODS: a catheter was placed in the ipsilateral jugular bulb during carotid endarterectomy (CEA) in 20 patients. Eight patients with ICBP (internal carotid backpressure) <40 mmHg received a shunt during CEA and 12 patients with ICBP >40 mmHg were operated upon without a shunt. Four patients with a carotid body tumour were used as controls. Blood was taken from the catheter as well as from the radial artery; before clamping, 5, 15, 30 min after clamping and 5 min after declamping. The oxygen extraction (AVO(2)) was calculated. Plasma concentrations of interleukin-1beta (IL-1beta), phospholipase A(2)(PLA(2)), thromboxane B(2)(TXB(2)), 6-keto-prostaglandin F1alpha (6-keto-PGF1alpha) and prostaglandin E(2)(PGE(2)) were measured by enzyme-linked immunosorbent assay (ELISA) technique. RESULTS: all patients had a normal postoperative course except for one patient in the no-shunt group, who suffered a stroke 1 h later due to occlusion of the endarterectomy site. The AVO(2)extraction increased during clamping in patients operated upon without a shunt (p <0.05). This increase was partly recovered to pre-clamp levels 5 min after reperfusion. The extraction remained stable in the non-shunted patients and the control group. The increased extraction in the non-shunted group correlated with increased levels of IL-1beta during clamping ( p <0.05) and reperfusion ( p <0.01). PLA(2)also increased during reperfusion in the non-shunted group ( p <0.05). An increased ratio between TXB(2)and 6-keto-PGF1alpha was noted during clamping ( p <0.05) and further increased during reperfusion. The levels of PGE(2)remained stable in both CEA groups. The PLA(2)levels, as well as TXB(2), 6-keto-PGF1alpha and PGE(2)levels, remained unchanged during the procedure in the control group. CONCLUSIONS: there is a metabolic response to carotid cross-clamping when no shunt is used. However, the clinical significance of this is unclear, since there were no intraoperative strokes.

Immunophenotypic analysis of the aortic aneurysm wall suggests that vascular dendritic cells are involved in immune responses.

OBJECTIVE: Inflammatory infiltrates similar to those in atherosclerotic plaques are prominent in the abdominal aortic aneurysm wall. Antigen presenting vascular dendritic cells are present in both early and advanced atherosclerotic lesions but their possible participation in abdominal aortic aneurysms has not been previously examined. This study reports the presence of vascular dendritic cells in abdominal aortic aneurysms and their participation in immune responses. METHODS: Samples of the anterior wall were collected from 18 atherosclerotic infrarenal abdominal aortic aneurysms ranging in diameter from 5-8 cm. All the patients were operated upon electively and no ruptured or rapidly expanding abdominal aortic aneurysms were included. Specimens were immediately frozen or fixed in 10% buffered formalin. Vascular dendritic cells were identified with anti-CD1a or with S-100. T cells and T cell subpopulations were identified with anti-CD3, anti-CD4 and anti-CD8. B cells were studied with anti-CD20. Analyses were carried out in sets of consecutive parallel sections immunostained with these antibodies and double immunostaining included different combinations of antigens such as CD1a/CD3, S-100/CD4, S-100/CD8. RESULTS: Most inflammatory infiltrates were found in the adventitia. These infiltrates contained B cells (CD20+) and T cells (CD3+) with their CD4+ and CD3+ cell subpopulations. In the aneurysm wall, CD1a+/S-100+ cells exhibiting dendritic appearance were detected and double immunostaining demonstrated that these vascular dendritic cells contained different lymphocyte populations including CD3+, CD4+, CD8+ and CD20+ cells. In some inflammatory infiltrates, B cells (CD20+) represented the predominant cell population (60-80%). Double immunostaining demonstrated that, in these infiltrates, vascular dendritic cells contacted CD20+ cells. CONCLUSIONS: Vascular dendritic cells are involved in immune reactions in the aneurysm wall, and this process mostly occurs in the adventitia. Vascular dendritic cells contact both T cells and B cells, suggesting that these vascular dendritic cells differ from other dendritic cells, subtypes of which associate with T cells (Langerhans cells, interdigitating cells) and B cells (follicular dendritic cells).

Tumor necrosis factor-alpha and interleukin-6 release from white blood cells induced by different graft materials in vitro are affected by pentoxifylline and iloprost.

Inflammatory mediators such as cytokines produced by white blood cells (WBCs) at the site of implantation are important for the biocompatibility of vascular grafts. The aim of the present study was to demonstrate the tumor necrosis factor-alpha (TNF-alpha) and interleukin-6 (IL-6) release from WBCs incubated with expanded polytetrafluoroethylene (ePTFE) or woven Dacron grafts. In a second series the effects of pentoxifylline (PTX) and iloprost (ILO), both known to inhibit white blood cell function, on this release were determined. Woven Dacron grafts induced significantly higher release of both TNF-alpha and IL-6 compared to ePTFE. TNF-alpha was detectable first after 2 h, whereas IL-6 was seen after 4 h. Maximum values were reached at 6 and 12 h, respectively. The addition of an endotoxin gave more pronounced patterns of cytokine release not influenced by time. Preincubation with both PTX and ILO at final concentrations of 100 and 10 micrograms/mL, respectively, reduced significantly the TNF-alpha release without differences between the two graft materials, whereas the effect on the IL-6 release varied and was graft material-dependent. In conclusion, graft material-dependent induction of TNF-alpha and IL-6 from WBCs was demonstrated. PTX and ILO influenced the cytokine release. It might be suggested that graft material-induced cytokine production could contribute to intimal hyperplasia in vivo. The present findings encourage further studies regarding graft material-induced WBC alterations and the role of pharmacologic agents influencing this function.

Endovascular abdominal aortic aneurysm repair induces significant alterations in surface adhesion molecule expression on donor white blood cells exposed to patient plasma.

OBJECTIVE: To determine the response of white blood cells in endovascular aortic aneurysm repair. MATERIALS AND METHODS: Seven patients treated with an endoluminal procedure (AAA-E) and seven patients undergoing conventional surgery (AAA-C) were included (all males, aged 52-80 years). A panel of monoclonal antibodies against CD11a, CD11b, CD11c, CD18 and L-selectin was used. To determine the surface receptors on both circulating and sequestered white blood cells, plasma from the patients and cells from healthy donors were combined for flow-cytometry. RESULTS: The expression of CD11a adhesion molecules only showed slight variations regarding granulocytes, but was more pronounced on monocytes, however, without significant differences between the two patient groups, CD11b, CD11c and CD18 molecules on both granulocytes and monocytes were significantly upregulated 60 min after the endovascular procedure compared to conventional aneurysm repair, and L-selectin molecules were by this time correspondingly cleaved off. CONCLUSION: Endovascular aneurysm repair differed significantly from conventional aneurysm surgery with peak adhesion molecule expression 60 min after balloon deflation, probably caused by release of tumour necrosis factor-alpha (TNF-alpha).

Inflammatory response to aorto-bifemoral graft surgery.

OBJECTIVE: To evaluate the systemic release of major cytokines and complemental activation during elective aorto-bifemoral bypass surgery and the influence on the postoperative course. DESIGN: Prospective randomised study. SETTING: University hospital, Sweden. PATIENTS: Fourteen consecutive patients with aorto-iliac occlusive disease were randomised to receive either a bifurcated e PTFE graft with stretch properties or a collagen coated Dacron graft. MAIN OUTCOME MEASURES: Immunologic parameters were assessed and included the cytokines TNF alpha (tumor necrosis factor), IL-6, IL-8, S-IL-2R and complemental factor C5a. Furthermore, acute plasma proteins, including C-reactive protein, and the different white blood cell fractions were determined. Sampling was performed frequently during surgery and postoperatively up to one month. RESULTS: An increase of serum-TNF alpha levels was observed early after declamping. This response preceded an increase of IL-6 levels and of C-reactive protein. No release of IL-8 was identified. A significant correlation between TNF alpha, IL-6 and C-reactive protein was observed (p < 0.001). A positive correlation was also observed for the degree of surgical trauma (blood loss). No significant differences between the two graft materials were encountered. The complemental system was also involved in the acute reactions and a marked increase of C5a levels was noted. A decrease of S-IL-2R levels as well as lymphocyte concentrations was also observed postoperatively and was interpreted as a downregulation of the immune-system in the immediate postoperative course. CONCLUSIONS: The results demonstrate an early and generalized inflammatory response during and after aortic surgery with involvement of different cytokines as well as the complemental system. TNF alpha appears to play a central role in the release of other cytokines, but IL-6 seemed to correlate best with later development of nonvascular complications. Few differences were found between the different grafts and the response seems to be influenced by other factors such as the surgical procedure and ischaemia/reperfusion injury.

Surface adhesion molecule expression on human blood cells induced by vascular graft materials in vitro.

The expression of surface adhesion molecules on granulocytes, monocytes (CD11a, CD11b, CD11c, CD18, L-selectin), and platelets (P-selectin, gpIIb-IIIa) was determined after incubation with different graft surfaces [expanded polytetrafluoroethylene (ePTFE) or woven Dacron]. Woven Dacron grafts upregulated the CD11b and CD11c surface antigens on both granulocytes and monocytes. Both graft materials demonstrated increased expression of CD11a and CD18 adhesion molecules on white blood cells at 30 min, followed by a downregulation. Maximum L-selectin expression was seen at 120 min on granulocytes and at 90 min on monocytes without differences between the graft materials. A rapid downregulation of gpIIb-IIIa complexes on platelets was noticed, while no expression of platelet P-selectin molecules was observed. In conclusion, both graft materials induced alteration of the white blood cell adhesion molecule expression, but the intensity and time course were dependent on the cell type and the graft material, suggesting that different mechanisms might be implicated. The expression of platelet surface antigens was less clearly influenced. The clinical significance of an enhanced cell surface antigen receptor expression caused by woven Dacron (CD11b, CD11c) has to be further studied. However, determination of adhesion molecule expression might offer possibilities to predict biocompatibility.

NADPH-diaphorase expression of endothelial cells during four weeks' healing of a stretch-ePTFE graft: an experimental porcine study.

Endothelial Nitric Oxide Synthase (eNOS) mediates the conversion of L-argine to NO and citrulline, which requires Nicotinamide Adenine Dinucleotide Phosphate (NADPH) as an essential cofactor. Evidence has been given that NOS accounts for the NADPH-diaphorase activity. The aim of the present study was to identify the histochemical and immunocytochemical appearance of NADPH-diaphorase and von Willebrand factor (factor VIII) respectively, in endothelial cells (ECs) during healing of stretch-expanded polytetrafluoroethylene (ePTFE) arterial grafts. On six Swedish domestic pigs an iliac bypass was bilaterally performed using 6 mm stretch-ePTFE grafts. The animals were allowed to survive one, two or four weeks. After explanation the grafts were prepared for NADPH-diaphorase histochemistry and factor VIII immunohistochemistry. Positive staining for the two identification markers was demonstrated after two weeks, whereas a more intense staining was seen after four weeks at the proximal and distal anastomoses, indicating maturation by time. No stained cells were observed at the mid-region of the grafts at any time. The cells differed from normal ECs, the former being less intense which may reflect immature ECs and probably a decreased expression of NO. In conclusion, the present study suggests tha NADPH-d histochemistry can be used to identify ECs. Whether a lower expression of NO compared to normal cells also means a reduced function, capable of causing adverse events has to be further evaluated.

Dilatation of aorto-bifemoral knitted Dacron grafts after a mean implantation of 5 years.

OBJECTIVE: To evaluate the dilatation-rate of a reverse locknitted Dracon graft. EXPERIMENTAL DESIGN: Comparative study, a follow-up after 5 years. SETTING: Department of Surgery, University Hospital. PATIENTS: During 1987 to 1990, 105 patients underwent elective aorto-bifemoral bypass surgery for aorto-iliac occlusive disease. Nineteen patients received a reverse locknitted Dacron graft. During follow-up in 1993 ten patients were available for evaluation of graft-diameter. For comparison 14 patients who received a wrap knitted Dacron graft (12 collagen impregnated and two uncoated) operated during the same time period were randomly chosen. INTERVENTION: CT-scan evaluation. RESULTS: A significantly lower increase (a mean aortic body dilatation of 6.8%) of the graft-size was found for this reverse locknitted Dacron (p < 0.001) compared to two wrap knitted Dacron grafts (collagen impregnated and uncoated knitted Dacron) (31.0% and 37.8% respectively). At the femoral level mean dilatation rates of respectively, 2.5%; 23.8% and 31.7% were found. Mural thrombus was observed in three grafts. No false aneurysms, graft degradation or graft occlusions were observed. CONCLUSION: It was concluded that reverse locknitted Dacron dilates to a minimum, however, further studies are needed to evaluate the relation of graft-composition/structure and dilatation.

Aortobifemoral surgery induces complement activation and release of interleukin-6 but not tumour necrosis factor-alpha.

The aim of the present study was to determine the inflammatory response by an extended analysis of complement in 16 patients undergoing aortobifemoral bypass surgery. The patients were randomized to receive either a bifurcated expanded polytetrafluoroethylene graft (n = 8; group I) or a collagen-impregnated knitted Dacron graft (n = 8; group II) to determine whether differences in graft surface properties might influence the inflammatory response during and after the procedure. The following components of complement: C1q, C4, C3, C3d, C5a and terminal complement complexes were all analysed. C-reactive protein and interleukin-6 were also determined to assess the acute phase response. The complement data were corrected for haemodilution, which was assessed from alpha 2-macroglobulin concentrations. A significant decrease of C1q (P < 0.0001) and an increase in C5a (P < 0.0005) was observed in both groups. C4 and C3 levels showed slight fluctuations in group I, whereas in group II these proteins increased significantly (P < 0.05, P < 0.005, respectively) between 2 and 7 days after surgery. Terminal complement complexes remained unchanged in both groups. Interleukin-6 levels peaked at 12-24 h and the C-reactive protein at 24-72 h. Higher interleukin-6 levels (P < 0.05) were found in group II 6 h after surgery compared with group I; no release of tumour necrosis factor-alpha was identified. An early inflammatory response was found in all patients. The patterns of the complement proteins varied with a C1q depletion and a C5a increase, interpreted as complement activation. Whether the variations between the two graft groups represent any differences in graft surface properties has to be further elucidated.