Portal de Periódicos da CAPES

Circulating Markers of Abdominal Aortic Aneurysm Presence and Progression

2008; Lippincott Williams & Wilkins; Volume: 118; Issue: 23 Linguagem: Inglês

10.1161/circulationaha.108.802074

1524-4539

Jonathan Golledge, Philip S. Tsao, Ronald L. Dalman, Paul E. Norman,

Infectious Aortic and Vascular Conditions

HomeCirculationVol. 118, No. 23Circulating Markers of Abdominal Aortic Aneurysm Presence and Progression Free AccessResearch ArticlePDF/EPUBAboutView PDFView EPUBSections ToolsAdd to favoritesDownload citationsTrack citationsPermissions ShareShare onFacebookTwitterLinked InMendeleyReddit Jump toFree AccessResearch ArticlePDF/EPUBCirculating Markers of Abdominal Aortic Aneurysm Presence and Progression Jonathan Golledge, Philip S. Tsao, Ronald L. Dalman and Paul E. Norman Jonathan GolledgeJonathan Golledge From the Vascular Biology Unit, School of Medicine and Dentistry, James Cook University, Townsville, Queensland, Australia (J.G.); Divisions of Vascular Surgery (R.L.D.) and Cardiovascular Medicine (P.S.T.), Stanford University, San Francisco, Calif; and School of Surgery, University of Western Australia, Fremantle Hospital, Fremantle, WA Australia (P.E.N.). , Philip S. TsaoPhilip S. Tsao From the Vascular Biology Unit, School of Medicine and Dentistry, James Cook University, Townsville, Queensland, Australia (J.G.); Divisions of Vascular Surgery (R.L.D.) and Cardiovascular Medicine (P.S.T.), Stanford University, San Francisco, Calif; and School of Surgery, University of Western Australia, Fremantle Hospital, Fremantle, WA Australia (P.E.N.). , Ronald L. DalmanRonald L. Dalman From the Vascular Biology Unit, School of Medicine and Dentistry, James Cook University, Townsville, Queensland, Australia (J.G.); Divisions of Vascular Surgery (R.L.D.) and Cardiovascular Medicine (P.S.T.), Stanford University, San Francisco, Calif; and School of Surgery, University of Western Australia, Fremantle Hospital, Fremantle, WA Australia (P.E.N.). and Paul E. NormanPaul E. Norman From the Vascular Biology Unit, School of Medicine and Dentistry, James Cook University, Townsville, Queensland, Australia (J.G.); Divisions of Vascular Surgery (R.L.D.) and Cardiovascular Medicine (P.S.T.), Stanford University, San Francisco, Calif; and School of Surgery, University of Western Australia, Fremantle Hospital, Fremantle, WA Australia (P.E.N.). Originally published2 Dec 2008https://doi.org/10.1161/CIRCULATIONAHA.108.802074Circulation. 2008;118:2382–2392Over the last decade, abdominal aortic aneurysm (AAA) has increasingly been recognized as an important cause of mortality in older persons. In 1999, for example, AAA was noted to be the 15th leading cause of mortality in the United States.1 Exact estimates of AAA-related fatalities are hampered by the low rate of postmortems when sudden death occurs in elderly subjects; however, recent figures suggest that AAA accounts for ≈15 000 deaths annually in the United States despite the increasing number of elective AAA repairs.2,3 Approximately 25 000 endovascular and open AAA repairs are performed annually in the United States.3Ultrasound screening of men >65 years of age has been demonstrated to reduce AAA-related mortality, and selective screening (of men ≥65 of age who have ever smoked) has been introduced in the United States.4 Most screen-detected AAAs are small ( 500 subjects).18Potential Value of Circulating BiomarkersAssociating concentrations of a range of circulating proteins with AAA has been of interest for a number of reasons in keeping with the currently accepted deficiencies in the management of this condition outlined earlier. Most commonly, investigators have measured circulating concentrations of a marker to assess its possible role in the pathogenesis or progression of AAA. When identified, such biomarkers may thus suggest possible targets for new medical treatments to slow AAA progression.19 With increasing interest in identifying AAAs at an early stage, circulating markers also could play a role in the diagnosis of small AAAs. Furthermore, biomarkers may have a role in predicting subsequent progression of AAA and therefore ultimately tailoring the management of the condition. The complications of AAAs are not completely reflected by aortic diameter alone. For example, some small AAAs rupture and some large AAAs remain stable for prolonged periods.20,21 Circulating markers that accurately reflect aortic wall destruction or inflammatory activity could potentially aid substantially in the identification of appropriate patients for different monitoring protocols and intervention (both drug based when developed and surgery) rather than the use of aortic diameter alone. Given these potential values of circulating biomarkers in the management of AAA, we carried out a literature review to summarize the present evidence for an association between different markers and AAA presence or progression.Literature SearchGiven the potentially large number of markers that may have been investigated, we carried out a staged search strategy. First, we performed a general search to identify biomarkers that have been investigated for association with AAA. The search terms biomarker and abdominal aortic aneurysm (articles identified, 185), plasma and abdominal aortic aneurysm (limits, human; articles identified, 245), and serum and abdominal aortic aneurysm (limits, human; articles identified, 332) were used to search the PUBMED database. These terms were selected because we wanted to concentrate on factors measurable in serum and plasma, rather than cell-associated markers, because we thought these markers would be more easily applicable to measurement in the average diagnostic laboratory. Publications with at least 50 patients (cases and controls combined) were selected for further examination. Markers for which ≥2 studies were identified from this initial search were assessed in more individualized searches using focused terms. Thus, searches for fibrinogen and abdominal aortic aneurysm, for example, were carried out, along with other identified markers. In general, 3 types of studies were identified from these searches. The most common study type involved a comparison of the circulating biomarker concentrations in cases with AAA and controls who were aged matched and healthy or had other cardiovascular disease. Other studies related the biomarker concentration to AAA expansion. Finally, a small number of studies related biomarkers to clinical outcomes such as the likelihood of presentation with aortic rupture or a successful aortic surgery. In the following sections, we describe the findings of these studies, concentrating most on association studies with AAA presence because they make up the commonest study type.Circulating Biomarkers and Their Association With AAA Presence in Case-Control StudiesRationale for Biomarker SelectionA number of approaches are possible in the identification of possible biomarkers for AAA. Most studies have chosen biomarkers on the basis of the present understanding of the pathogenesis of AAA. Examination of aortic biopsies from patients undergoing open repair demonstrates medial destruction associated with a paucity of vascular smooth muscle cells, an accumulation of macrophages and lymphocytes, elastin fragmentation, high concentrations of proteolytic enzymes and cytokines, and laminated thrombus.19,22–25 On the basis of these pathological features, investigators have examined circulating lipids, thrombosis-related proteins, markers of extracellular matrix turnover and degradation, and proinflammatory and related cytokines.A potential alternative approach to biomarker selection is based on screening patients with AAA compared with control subjects. A range of proteomic screening techniques are now available, including those based on gel electrophoresis, mass spectrometry, and multiplex antibody arrays.26 These and related techniques have been used to compare aortic wall, but not blood, samples from patients with AAA and control subjects.27–30 These studies have suggested the upregulation of proinflammatory cytokines such as interleukin (IL)-1β, IL-6, IL-8, and tumor necrosis factor-α27,30; chemokines such as chemokine CC motif ligand 227; matrix-degrading enzymes, particularly matrix metalloproteinase (MMP)-928,30; and expression of genes involved in immune function in general.29 With the results of such analyses, it would be possible to select potential circulating biomarkers for AAA. It should be noted, however, that all these tissue biomarker studies have involved very small numbers (n≤10) of patients with AAA and similar numbers of control subjects. The selection of suitably matched controls also is problematic in these studies, especially because the availability of normal-aged aorta is often limited to postmortem samples, which are usually collected in a very different way from operative AAA biopsies.Circulating Extracellular Matrix MarkersThree markers of extracellular matrix remodeling have been associated with AAA in case-control studies: the carboxyterminal propeptide of type I procollagen, the aminoterminal propeptide of type III procollagen, and tenascin-X.31–36 Both type I collagen and type III collagen are important components of the aortic media, and fragmentation and synthesis of new type I and III collagen are typically found in biopsies of AAAs.37 Parts from both the carboxyterminal and aminoterminal ends of the precursor molecule are split off and released during collagen synthesis. Hence, it might be assumed that type I and III procollagen fragments would be found at increased concentrations in the circulation of patients with AAA. Measurement of these fragments is not straightforward with investigators using radioimmunoassays.31–35Circulating concentrations of the aminoterminal propeptide of type III procollagen were reported to be significantly increased in patients with AAA compared with control subjects in 2 studies (Table 1).31,33 However, the most recent and the largest studies showed no association of serum type III procollagen peptides with AAA.34,35 The circulating concentration of the carboxyterminal propeptide of type I procollagen has been compared between patients with AAA and control subjects in 3 small studies of only 190 cases.31–33 None of these studies demonstrated higher concentrations of this peptide in patients with AAA. Nakamura and colleagues32 reported lower concentrations in the plasma of patients with AAA (n=17) compared with healthy control subjects (n=22) but not those with peripheral artery disease (n=14). Table 1. Association of Circulating Biomarkers and AAAMarker/ReferenceCases, nControls, nConcentration, CasesConcentration, ControlsPIIINP indicates aminoterminal propeptide of type III procollagen; tPA, tissue plasminogen activator. Numbers are mean±SD, mean (95% CI), or median (interquartile range or range) as appropriate.*Measured in plasma.†Healthy controls.‡Measured in serum.§Significantly higher concentration in cases (P<0.05).∥Controls with atherosclerosis and no AAA.PIIINP, μg/L 34*9583†324±127312±101 35*201246†3.73 (3.53–3.93)3.65 (3.49–3.81) 31‡8620†4.6±1.5§2.8±0.9 33*8790∥3.47 (3.2–3.7)§2.73 (2.5–3.0)MMP-9, ng/mL 41‡539†/17∥622.0±400.2§280.8±165.5/284.3±151.4 34‡9583†353±252455±499 42*2212∥61±4951±28 43*4510†30.9±17.1§8.9±2.5 44*255†/15∥99.4±17.4§36.1±7.7/54.7±10.5 45*228†/9∥85.6±11.64§13.16±1.94/25.75±4.16Fibrinogen, g/L 49*110110∥2.89 (2.45–3.40)§2.53 (2.1–3.07) 50*8998†3.5 (2.9–4.1)§3.1 (2.7–3.6) 51*3637†/31∥3.80 (3.22–4.20)2.95 (2.64–3.22)/4.05 (3.30–4.39) 52* Men2632699†3.72±0.91§3.32±0.88 Women743350†3.77±0.68§3.43±0.80 53*2320†3.6 (1.9–6.3)§2.2 (1.3–3.0) 54*2142†/42∥3.6±1.23.3±0.9/3.8±0.6 55*40200†3.05 (2.61–3.57)§2.62 (2.21–3.08) 56*22244†5.65 (4.81–6.68)5.40 (4.78–6.31)D-dimer, ng/mL 59*4041†625 (460–1437)§86 (38–176) 50*8998†441.5 (198.8–771.0)§93.0 (57.8–158.8) 60*1810†421±400§238±180 61*3625†7700±6700§1000±1200 55*40200†142 (84–210)§83 (68–129) 62*4130∥8400±10 800§2100±2700tPA, μg/L 59*4041†10.5 (8.6–14.8)11.1 (8.5–15.2) 72*42100†13.6±4.7§11.4±4.3 50*8998†7.9 (6.0–11.1)8.6 (6.8–11.5) 53*2320†10.2 (4.9–24.2‡)9.4 (2.9–19.0‡) 55*40200†8.5 (7.0–10.5)7.9 (5.8–10.1)Deficiency of the extracellular matrix protein tenascin-X has been implicated in some cases of Ehlers-Danlos syndrome, a condition prone to aortic dissection and aneurysm formation.38 Zweers and colleagues36 reported higher serum concentrations of tenascin-X in 87 patients with AAA compared with 86 control subjects. Unlike most studies, these investigators adjusted their analyses for potential confounding factors and reported that tenascin-X serum concentrations in the highest quartile were associated with a 5-fold increased risk of AAA (odds ratio, 5.3; 95% CI, 2.0 to 13.8).36 At present, this association has not been confirmed in another cohort.Matrix-Degrading EnzymesFragmentation of the extracellular matrix of the aortic media is perhaps the most specific histological hallmark of AAA.22 Matrix-degrading enzymes implicated in AAA include the MMP and cathepsin groups.22,30,39,40 Surprisingly, relatively few and small studies have compared the circulating concentrations of these enzymes or their inhibitors in patients with AAA and control subjects.34,41–46 Circulating MMP-9 concentrations have been investigated most frequently, but the findings have not been completely consistent (Table 1).34,41–45 Four of 6 studies reported higher concentrations of circulating MMP-9 in patients with AAA compared with healthy control subjects or subjects with atherosclerosis but not AAA.41,43–45 Two studies, including the largest, found no association of MMP-9 with AAA.34,42 Concentrations of MMP-9 are several-fold higher in serum than plasma (because of platelet degranulation), and this may have contributed to the disparate findings.47 The studies reporting no association between MMP-9 and AAA involved assessments of plasma in 1 instance and serum in the other example, suggesting that the inconsistent findings are not explained by differences in sampling alone.34,42 All studies assessing plasma MMP-9 concentrations, however, did report higher mean concentrations in subjects with AAA, although this difference reached significance in only 3 of 4 studies.42–45 Single studies have investigated the association of circulating concentrations of MMP-1, MMP-2, MMP-3, tissue inhibitor of MMP-1, and α1-antitrypsin with AAA.32,34,43,46 These studies reported higher concentrations of MMP-3 and tissue inhibitor of MMP-1 in patients with AAA, but the sample sizes (n=55 and 53) and the absence of replication make any firm conclusions impossible.32,43Proteins Associated With ThrombosisMost AAAs contain significant quantities of intraluminal thrombus, and the volume is correlated with the severity of aortic dilatation.25 The content of such thrombus includes a number of proteases and has been implicated in AAA progression.47,48 In vitro aortic thrombus is capable of releasing a range of thrombus-associated products.48 Proteins involved in, stimulated by, or associated with thrombosis have been the biomarkers most commonly assessed in AAA. Markers evaluated include fibrinogen, D-dimer, homocysteine, tissue plasminogen activator, von Willebrand factor, soluble thrombomodulin, plasminogen activator inhibitor-1, activated protein C–protein C inhibitor complexes, plasmin-antiplasmin complexes, P-selectin, thrombin–antithrombin III complex, and fibrinogen degradation products.49–72 The association between plasma fibrinogen and AAA has been studied extensively, with 5 of 10 investigations reporting higher concentrations in patients with AAA (Table 1).49–58 The studies showing no association generally included very small numbers of patients,51,54,57,58 except for 1 investigation carried out exclusively in smokers.56 Unlike other studies listed in Table 1, controls in the study by Franks et al56 did not have an AAA excluded by imaging. Plasma fibrinogen is increased by smoking; thus, the association of fibrinogen with AAA could simply reflect the well-established link between smoking and AAA.19,73 The association of fibrinogen with AAA was present after adjustment for other risk factors in 2 studies, supporting the independent link between fibrinogen and AAA.52,55 The odds ratios for an increase in fibrinogen of 1 SD were 1.5 (95% CI, 1.1 to 2.2) and 1.4 (95% CI, 1.2 to 1.7). One of these studies, a population screening investigation, reported an independent association between fibrinogen and AAA in men but not women.52Plasma concentrations of D-dimer reflect the extent of fibrin turnover in the circulation because this antigen is present in several products from the degradation of cross-linked fibrin by plasmin.74 The circulating concentrations of D-dimer have been compared in patients with AAA and control subjects in 7 studies.50,55,57,59–62 Six of the 7 studies, including a total of 264 cases and 404 controls, demonstrated significantly higher D-dimer concentrations in subjects with AAA (Table 1). The 2 largest studies also reported an independent association between plasma D-dimer concentrations and AAA after adjustment for other risk factors.50,55 The odds ratios for an increase in D-dimer of 1 SD were 4.6 (95% CI, 2.3 to 9.5) and 3.8 (95% CI, 1.8 to 7.8). Only 1 study incorporating just 23 cases and 16 controls reported no significant difference in plasma D-dimer concentrations, although the actual values were not reported.57The circulating concentration of the complex between activated protein C and protein C inhibitor is related to thrombin generation and has been associated with AAA in 1 study.63 Kölbel and colleagues64 reported median plasma concentrations of activated protein C–protein C inhibitor complexes to be 0.45 (95% CI, 0.24 to 1.47) and 0.22 (95% CI, 0.15 to 0.48; P<0.01) in 78 patients with AAA and 73 with carotid artery disease, respectively (P<0.001). The investigators later extended this finding to a larger group of patients with AAA (total, 232) and reported a correlation between activated protein C–protein C inhibitor levels and aortic diameter (r=0.22). This marker has not yet been reported in other cohorts.Studies comparing the plasma concentration of homocysteine in subjects with AAA and control subjects have recently been reviewed.65 This review identified 5 case-control studies, with all investigators reporting higher circulating homocysteine in subjects with AAA.66–70 One of these studies, which was much larger (438 cases and 438 controls) than all the others ( 19 μmol/L in men and 15 μmol/L in women) and AAA after adjustment for other risk factors (odds ratio, 7.8; 95% CI, 4.6 to 13.2).66 A more recent study failed to show any independent association between homocysteine and AAA after adjustment for serum creatinine.71 These authors instead found a negative association between circulating vitamin B6 concentrations and AAA.71 Given the varying policies on folate and B vitamin supplementation throughout the world, any association between homocysteine and AAA is likely to vary in different populations.Tissue plasminogen activator activates plasmin, an important fibrinolytic enzyme, which has been implicated in AAA development in experimental models.75 Plasma tissue plasminogen activator has been compared in subjects with and without AAA in 5 studies.50,53,55,59,72 Only one of the studies reported significantly higher concentrations in patients with AAA (Table 1). Overexpression of plasminogen activator inhibitor-1, an inhibitor of plasmin activation, inhibits AAA formation in mice.75 The association of AAA and plasma plasminogen activator inhibitor-1 concentrations was assessed in 3 studies.59,66,72 The largest of these studies reported mean plasma plasminogen activator inhibitor-1 concentrations of 28.6±21.6 and 17.8±12.6 mg/dL in 431 patients with AAA and 431 healthy control subjects, respectively.66 Plasminogen activator inhibitor-1 concentrations >42.6 mg/dL were independently associated with AAA after adjustment for other risk factors (odds ratio, 3.2; 95% CI, 1.7 to 6.1). Two small studies found no association between plasminogen activator inhibitor-1 and AAA but included a total of only 82 cases and 141 controls.59,72 The fibrinolytic effects of plasmin are inactivated in the circulation by antiplasmin, leading to the formation of plasmin-antiplasmin complexes. Fowkes and colleagues50 reported higher median plasma concentrations of plasmin-antiplasmin complex in 89 patients with AAA (596 μg/L; interquartile range, 432 to 878 μg/L) compared with 98 healthy control subjects (384 μg/L; interquartile range, 274 to 486μg/L; P=0.01). The investigators did not comment on whether this association was maintained in models adjusted for other risk factors.Endothelial von Willebrand factor mediates platelet adhesion, and circulating concentrations of von Willebrand factor are a marker of endothelial injury and dysfunction.76 None of the 4 studies identified comparing the concentrations of circulating von Willebrand factor in cases and controls have reported any significant association with AAA.50,54,55,63 Thrombomodulin is an endothelium-bound protein that plays an important role in the protein C anticoagulant pathway. Circulating levels of thrombomodulin also have been correlated to endothelial dysfunction.77 Two small studies involving a total of 79 cases and 144 controls investigated the association of circulating thrombomodulin with AAA and reported disparate findings.54,70 One study reported an association of thrombomodulin with AAA, but this association was not apparent in the other study. One small study reported an association between circulating concentrations of the activated platelet marker P-selectin and AAA, although this finding has not yet been replicated.54A number of other thrombosis-associated products, including thrombin–antithrombin III complexes and fibrinogen degradation products, have been assessed in small studies. Two studies involving a total of 77 cases and 55 controls reported increased concentrations of fibrinogen degradation products in those with AAA.61,62 One of these studies also reported higher thrombin–antithrombin III complex concentrations in 36 patients with AAA and 25 healthy control subjects.61LipidsCirculating lipids are normally considered a risk factor for, rather than a biomarker of, atherothrombosis. Low-density lipoprotein (LDL) and high-density lipoprotein (HDL) are independent risk factors for cardiovascular events such as myocardial infarction and stroke, so it would be expected that these lipids would be the primary ones assessed in case-control studies for AAA.78 We identified 10 studies that compared HDL between patients with and without AAA, including 3 population screening studies.52,54,79–86 The 8 studies in which average HDL concentrations were stated are listed in Table 2. Five of these studies, including the 2 largest studies, reported lower concentrations of HDL in subjects with AAA compared with healthy control subjects. Additionally, Alcorn and colleagues,85 in a population study of 4741 subjects, reported a negative association between HDL and AAA, whereas in a smaller study of only 200 subjects, Blanchard et al86 reported no association between HDL and AAA. It may be important that neither of the 2 small studies comparing HDL concentrations in subjects with AAA and control subjects with atherothrombosis reported lower concentrations in those with AAA.54,84 This finding may indicate the lack of specificity of HDL to AAA as opposed to atherothrombosis; other larger studies incorporating subjects with peripheral artery disease are required. We identified 7 studies that compared LDL concentrations in patients with and without AAA. In general, these studies found a less consistent association of LDL with AAA than seen for HDL (Table 2). Three studies reported higher concentrations in subjects with AAA.81,82,85 The increasing use of statins in subjects as a result of concurrent atherothrombosis makes it unlikely that LDL is a practical marker of AAA. Table 2. Association of Serum HDL and LDL With AAAReferenceCasesControlsHDL, Cases, mmol/LHDL, Controls, mmol/LLDL, Cases, mmol/LLDL, Controls, mmol/LNS indicates not stated. To convert mmol/L to mg/dL, multiply by 38.67. Numbers are median (interquartile range) or mean±SD as appropriate.*Healthy controls.†Significantly lower concentration in cases (P<0.05).‡Significantly higher concentration in cases (P<0.05).§Only mean, no SD, is stated.∥Controls with atherosclerosis and no AAA.793026*0.83±0.18†1.21±0.182.84±0.772.49±0.548035140*1.2 (1.1–1.2)†1.3 (1.3–1.4)3.5 (3.2–3.7)3.3 (3.2–3.5)81206252*1.13 (1.07–1.20)1.16 (1.11–1.22)4.05 (3.92–4.19)‡3.71 (3.59–3.83)52 Men2632699*1.28±0.37†1.42±0.37NSNS Women743350*1.46±0.42†1.68±0.438225266*1.27§1.39§4.24§‡3.52§542142*/42∥1.1±0.5†1.5±0.3/1.3±0.33.9±0.73.4±1.1/4.0±1.183691460*1.32±0.31†1.57±0.403.96±1.063.74±1.018411457∥1.22 (1.07–1.50)‡1.08 (0.95–1.44)4.50 (3.70–5.40)4.88 (3.83–5.70)Some of the studies discussed above also assessed the relationship between AAA and triglycerides or apolipoprotein A and B. A number of studies,52,79,83 but not all,46,81,84 have shown slightly higher concentrations of serum triglycerides in patients with AAA. This association has not generally been maintained after adjustment for other risk factors such as atherothrombosis.52 Three studies that reported apolipoprotein concentrations in relation to AAA were identified.56,82,83 Two of these studies reported higher concentrations of apolipoprotein B in subjects with AAA, but no adjusted analyses were presented.82,83 A greater number of studies assessing the association of lipoprotein(a) with AAA were identified (Table 3). These studies demonstrate higher concentrations of lipoprotein(a) in subjects with AAA compared with healthy control subjects but no difference compared with patients with atherothrombosis.56,66,83,87–89Table 3. Association of Lipoprotein(a) and IL-6 With AAAMarker/ReferenceCases, nControls, nConcentration, CasesConcentration, ControlsLp(a) indicates lipoprotein(a). To convert nmol/L to mg/dL, divide by 35.7. To convert pg/mL to pmol/L, multiply by 0.037. Numbers are median (interquartile range) or mean±SD as appropriate.*Measured in plasma.†Healthy controls.‡Controls with atherosclerosis and no AAA.§Significantly higher concentration in cases (P<0.05).∥Measured in serum.¶Only mean or median is stated; no SD or interquartile range is given.Lp(a), nmol/L 87*425230†/262‡29.9 (<2–583)17.0 (<2–285)§/28.8 (<2–539) 66∥438438†728 (7–9646)§352 (32–2481) 88∥7543†674 (<342–1445)§<342 (<342–582) 89*29274†382§¶228¶ 56*22244†750 (417–1267)742 (262–2099) 83∥691460†453±511414±453IL-6, pg/mL 92*2715‡4.94±0.48§2.65±0.51 50*8998†2.8 (2.0–4.2)§1.8 (1.3–2.7) 90*7430†64.2±157.3§6.7±5.1 91∥5038†/42‡0.24§0.014/0.086Markers of InflammationA large number of circulating markers of inflammation have been assessed for association with AAA; however, this has mostly involved measuring the different biomarkers in single populations usually involving small numbers of subjects. Examples of markers assessed are IL-1β, IL-2, IL-6, IL-8, tumor necrosis factor-α, interferon-γ, C-reactive protein, osteopontin, resistin, leptin, adiponectin, sCD28, sCD86, sCTLA-4, sVCAM-1, sICAM-1, endothelin 1/2, and antibodies to Chlamydia pneumoniae.50,54,90–98 The most consistently associated with AAA is IL-6.50,90–92 Increased concentrations of IL-6 have been demonstrated in biopsies of AAA compared with control biopsies.99 Plasma concentrations of IL-6 have been shown to increase distal