| | Fibrinogen and high-sensitive C-reactive protein as serologic predictors for perioperative cerebral microembolic lesions after carotid endarterectomyReceived 23 March 2007; accepted 14 May 2007. BackgroundNeurologic deficit caused by cerebral ischemia defines the outcome of carotid endarterectomy (CEA). Although few patients have clinically evident neurologic deficit, diffusion-weighted imaging (DWI) presents a number of cases with ischemic brain lesions. This study should elucidate preoperative risk factors for perioperative microemboli that cause brain infarction. MethodsWe studied 183 patients (58 women, 69.2 ±12.7 years; 125 men, 69.3 ± 8.9 years) with high-degree carotid artery stenosis. DWI was performed before and after CEA to analyze new cerebral ischemia. Blood samples were obtained before operation to measure fibrinogen and C-reactive protein (CRP), and preoperative high-sensitive CRP (hsCRP) was analyzed in 30 consecutive patients. ResultsPostoperative DWI revealed new ipsilateral ischemic lesions in 41 patients (22.4%), and eight (4.4%) showed new neurologic deficit. Preoperative fibrinogen levels were higher in patients with new lesions (397.6 mg/dL ± 104.7 mg/dL) than in those without (324.7 mg/dL ± 74.2 mg/dL, P < .001). Preoperative levels of hsCRP were also higher in patients with new lesions (7.9 mg/dL ± 5.2 mg/dL) than in those without (2.8 mg/dL ± 2.6 mg/dL, P = .004). Significant association was found between fibrinogen and CRP (Spearman ρ = 0.402; P < .001) as well as hsCRP (Spearman ρ = 0.603, P = .003). No association was found between postoperative lesions and CRP (P = .833). ConclusionThe present study demonstrates that preoperative levels of fibrinogen and hsCRP are independent determinants for new periprocedural cerebral ischemic lesions caused by microembolic events. There is still not sufficient evidence to recommend measurement of CRP as a prognostic marker for perioperative cerebral lesion. The main purpose of carotid endarterectomy (CEA) is the prevention of stroke.1 CEA in symptomatic patients is associated with a high incidence for the appearance of diffusion-weighted imaging (DWI) lesions and brain infarction, but the cause for microembolic events remain notional.2 Many studies have shown that raised levels of inflammatory markers increase the risk of coronary heart disease and the risk of ischemic or hemorrhagic stroke.3, 4 Recent studies indicate a strong association between high-sensitive C-reactive protein (hsCRP) levels and subsequent cardiac events after interventional cardiac therapy.5 Elevated fibrinogen is an important predictor of future coronary events in individuals with a history of coronary heart disease.6 The role of plasma fibrinogen as central protein in the coagulation system has been documented by clinical evidence: fibrous plaques are rich in fibrinogen and degradation products, which are involved in mechanisms of endothelial cell injury and platelet aggregation.7 It triggers the formation of progressive atherosclerotic plaques. The relationship between high plasma fibrinogen levels, the thin fibrous cap of atheroma, and a greater incidence of plaque rupture and thrombosis are well known. In patients with elevated fibrinogen levels, high numbers of inflammatory cells are concentrated in the shoulder and cap of plaques.8 Studies on the association between several inflammatory markers and postoperative cerebral lesion after CEA are limited, however. The aim of our study was to assess whether levels of fibrinogen, CRP, and hsCRP are useful as predictive values of postoperative ischemic events in patients after CEA. Material and methods  Patient population We evaluated 183 patients with high-degree carotid artery stenosis (91 asymptomatic, 92 symptomatic patients) intended for CEA. The degree of stenosis was evaluated by color-coded duplex ultrasound imaging following European Carotid Surgery Trial (ECST) criteria. All CEA were performed by experienced vascular surgeons from January 2004 to June 2006. All patients underwent neurologic examination ≤2 days before and after the procedure. The study was performed according to the Guidelines of the World Medical Association Declaration of Helsinki and was approved by the Hospital Review Board of the Interdisciplinary Center for Vascular Diseases. All patients gave informed consent. Magnetic resonance imaging (MRI) was performed using a 1.5 Tesla whole-body imaging system (Magnetom Symphony Quantum gradient, Siemens Medical Systems, Erlangen, Germany) with a dedicated head coil. DWI was performed 1 day before and after CEA. The whole brain DWI was done with an isotropic echo-planar sequence with b-values of 0.500 and 1000 s/mm2, repetition time, 4006 milliseconds; echo time, 83 milliseconds; number of averages, two; slice thickness, 4 to 6 mm; 128 × 128 matrix size, and 220 × 220 mm2 field of view. Sagittal, coronal, and transverse views were obtained. All MRI results were reviewed by two experienced neuroradiologists without knowledge of vascular status and side of operation and blinded to each other’s findings. An acute DWI lesion (Fig 1) was only diagnosed if an increased signal intensity was visible on two planes, if a corresponding decreased signal intensity was detected in the apparent diffusion coefficient image, and if both neuroradiologists agreed on their findings. Clinical history and physical examination were taken with special attention to cardiovascular risk factors, as summarized in Table I. Routine laboratory values, urinalysis results, and chest radiographs were used to exclude any inflammatory or infectious diseases. Severe cardiac disease, coagulopathy, or any operation ≤6 months was an exclusion criterion. Blood samples for measurement of fibrinogen, CRP, and hsCRP were taken from an antecubital vein a day before operation and were immediately analyzed. Plasma fibrinogen activity was measured quantitatively by the Clauss method, where fibrinogen is converted from a soluble protein to an insoluble polymer by the action of thrombin, resulting in the formation of a fibrin clot. The thrombin clotting time is inversely proportional to the fibrinogen concentration of the plasma. Thus, measuring the clotting time of dilute plasma when excess thrombin is added compared with a standardized fibrinogen preparation results in quantitative determination of fibrinogen concentration (Dade Behring, Schwalbach, Germany). Measurement of CRP was performed on serum using an ultrasensitive assay based on a particle-enhanced turbidimetric immunoassay technique (Dade Behring). The assay was performed according to manufacturer instructions. Blood samples were mixed with the antiserum solution. The CRP reacts specifically with latex particles coated with antihuman CRP antibody to yield insoluble aggregates. The increase in turbidity that accompanies aggregation is proportional to the CRP concentration and was determined using an automated nephelometer (BNII System, Dade Behring). In 30 consecutive asymptomatic and symptomatic patients, we also collected blood samples before operation for hsCRP enzyme-linked immunosorbent assay (ELISA) analysis. The samples were centrifuged at 1600 rpm for 10 minute. The obtained serum was stored at −70°C until laboratory testing. According to the results of Pai et al,9 who investigated the stability of plasma markers in different time ranges, the range from sample collection until freezing was held to a maximum of 2 hours. The hsCRP ELISA is based on the principle of solid-hase enzyme-linked immunosorbent assay (LifeDiagnostics, West Chester, Pa), with an analytical sensitivity of <0.01 mg/dL and intraassay and interassay coefficients of variation of 4.2% and 4.1%, respectively. Statistical analysis Data were analyzed with SPSS 13.0 (SPSS Inc, Chicago Ill). Values of continuous variables are expressed as mean ± standard deviation. Groups were compared by the Mann-Whitney Utest or matched signed rank test. Correlations between continuous variables were calculated using Spearman rank correlation coefficient. To investigate the relationship between binary outcome and measured covariates, the logistic regression model was used. Calculations of odds ratios with 95% confidence intervals (CI) were performed and illustrated with forest plots. Accuracy for prediction was additional assessed by receiver operating curve (ROC) analyses; thereby, area under the curve (AUC) and optimal cutoff values of explanatory variables were calculated. All tests were considered significant at P = .05. Results Clinical characteristics of the patients are given in Table I. Both groups were similar with regard to all analyzed risk factors, medications, and degree of stenosis measured as peak systolic velocity (PSV). Logistic regression analysis showed no significant effect on the likelihood of occurring new cerebral lesions (Fig 2). Preoperative DWI was performed in all patients, and 13 (14.3%) of the 91 preoperatively asymptomatic patients and 28 (30.4%) of the 92 preoperatively symptomatic patients showed new postoperative cerebral ischemic lesions. All lesions occurred ipsilateral to the operated side. New neurologic deficits developed in eight patients, two (2.2%) within the prior asymptomatic group and six (6.5%) within the symptomatic group (P = .242). We saw two transient ischemic attacks (TIAs) in the asymptomatic group, and five TIAs and one permanent minor stroke in the symptomatic group. Mean serum levels of CRP (0.79 ± 0.71 mg/L vs 0.84 ± 1.0 mg/L) and fibrinogen (331.2 ± 84.5 mg/dL vs 349.9 ± 90.99 mg/dL) relating to asymptomatic and symptomatic patients showed no significant difference (Table II). Significant association was found between levels of fibrinogen and CRP (Spearman rho 0.402; P < .001) as well as hsCRP (Spearman rho 0.603; P = .003) whereas no association was found between postoperative lesions and CRP (P = .833). Focusing on cerebral events, the group with new lesions had significantly higher levels of fibrinogen (397.6 ± 104.7 mg/dL vs 324.1 ± 75.1 mg/dL, P = .001) and hsCRP (7.9 ± 5.2 mg/dL vs 2.8 ± 2.6 mg/dL, P = .004). CRP (1.0 ± 1.48 mg/L vs 0.8 ± 0.66 mg/L, P = .625) showed no significant difference (Fig 3, Fig 4, Fig 5). In logistic regression analysis, hsCRP (P = .048) and fibrinogen (P < .001) showed a significant effect on the likelihood of new cerebral lesions occurring, but CRP was not significant (P = .100). Accuracy for prediction assessed by ROC analysis showed an AUC of 0.879 for hsCRP, with an optimal cutoff value of 4.16 (sensitivity, 0.75; specificity, 0.77), and an AUC of 0.884 for fibrinogen with a cut-off value of 364 (sensitivity, 0.88; specificity, 0.71). The odds ratio for a new lesion was 4.49 (95% CI, 2.12 to 9.49) by categorized fibrinogen and 11.0 (95% CI, 1.42 to 85.2) by hsCRP, noting that fewer measurements resulted in a large CI for the estimated effect of hsCRP. The odds ratio achieved by CRP with a cutoff value 0.75 was 1.75 (95% CI, 0.81 to 3.80), thus corresponding to a weak but not significant trend (Fig 6). The positive and negative predictive value was 63.6% and 90.9% for fibrinogen and 85.7% and 86.7% for hsCRP. Discussion In earlier studies, we observed a significant correlation between the number and volume of cerebral lesions in DWI and the occurrence of brain infarction in follow-up MRI. These findings indicate that the severity of new postoperative neurologic events correlates with DWI lesion size.2 Assuming that a loss of brain tissue impairs the result of CEA, we elucidated the influence of pre-existing activity of atherosclerosis represented by acute-phase proteins. Inflammatory processes play a pivotal role in the pathogenesis of atheroma development to ultimate rupture of unstable atherosclerotic plaques.5 Atherosclerosis is typically associated with low-grade vascular inflammation that can be measured. Dosa et al5 evaluated plasma fibrinogen and serum hsCRP during CEA. The study indicated that removal of atherosclerotic plaques from the carotid arteries markedly decreased the production of acute-phase proteins owing to the decrease of inflammatory burden or the removal of the advanced plaques able to produce these proteins.5 Increased hsCRP levels may be related to the presence of macrophages and T-lymphocytes in carotid plaque, which is associated with instability that leads to ischemic events. Kondo et al10 demonstrated that plasma hsCRP is a marker of carotid atherosclerosis activity. Rerkasem et al11 investigated levels of inflammatory markers between patients with symptomatic carotid stenosis and those who were asymptomatic. Plasma hsCRP was elevated in symptomatic compared with asymptomatic patients. Their study showed that hsCRP was of prognostic value in a number of cardiovascular conditions.11 In contradiction to these results, we could detect patients in both asymptomatic and symptomatic groups with a higher expression of acute-phase reactants and new postoperative cerebral lesions. Our results also contradict those of Choi et a1,12 who found that a higher hsCRP level was associated with a higher risk score of coronary heart disease but not with carotid atherosclerosis. We also found a significant interaction between serum levels and activity of carotid atherosclerosis expressed in postoperative ischemic lesions.12 Actually, acute coronary syndromes are thought to result from plaque rupture that is induced by the inflammatory process in the atherosclerotic tissue. Patients with non-ST elevation acute coronary syndromes who showed no event ≤6 months were characterized by a decrease in hsCRP levels from baseline to follow-up. Most events in the observation period of 3 years occurred in patients with follow-up hsCRP-levels >60% of the initial level. It was therefore hypothesized that a repeated measurement of hsCRP levels in CAD patients could help to discriminate those at high risk of further events.13 Thus determination of circulating hsCRP levels may be a useful additional marker of risk in patients with high-grade carotid stenosis.14 Zwaka et al15 showed hsCRP-dependent mediation of low-density lipoprotein uptake by macrophages, and Pasceri et al16 showed that hsCRP influences atherogenesis and induces adhesion molecule expression in human endothelial cells. Furthermore, Hashimoto et al17 found that hsCRP is an independent predictor of the rate of increase of carotid atherosclerosis.17 Tanne et al18 examined the association between CRP levels and the subsequent risk of incident ischemic stroke among 2979 patients with stable coronary heart disease and showed that the risk of stroke per 1000 person-years increased according to CRP levels. These findings demonstrate the risk prediction for incident ischemic stroke conferred by CRP levels in patients.18 Our findings contradicted these results: we could not detect significant difference in CRP levels that affected postoperative cerebral lesions. The analysis of the EUROSTROKE project indicates that increased fibrinogen is a powerful predictor of stroke.3 However, an increase of 100 mg/dL in a patient’s fibrinogen level within the range tested (between 250 mg/dL and 562 mg/dL) was associated with a significantly increased risk of heart disease and stroke. Ma et al19 found that those with high fibrinogen levels, >343 mg/dL, had a twofold increase in the risk of myocardial infarction (MI). A multivariate analysis by Coppola et al20 revealed that fibrinogen plasma levels in patients after acute MI were the only independent predictor of mortality in a 42-month follow-up after acute MI. Our results, together with other observations from recent studies, suggest that fibrinogen evaluation may be useful in identifying patients at higher risk of atherosclerosis-associated events.20 Finally, fibrinogen may also be more than a marker because it binds to platelets and contributes to platelet aggregation and fibrin formation. Further studies should elucidate the connection between serum levels and plaque morphology. Elevated plasma levels of fibrinogen and hsCRP have been shown to predict future risk of plaque rupture and ischemic stroke, particularly in men and in young and middle-aged individuals.4 Patients with high fibrinogen levels after ischemic stroke had higher mortality rates than patients with lower fibrinogen levels.21 Although we also found an association between hsCRP and fibrinogen, we could not detect any influence of sex on the laboratory values or new postoperative DWI events. HsCRP has been reported to reflect inflammation related to the pathology and biology of ischemic stroke.22 Elevated levels of fibrinogen and hsCRP in patients with carotid artery stenosis vs controls imply that these acute-phase reactants are markers of plaque stability. This is supported by the fact that carotid artery stenosis is a predictor of widespread atherosclerosis23 and that atherosclerosis is an inflammatory process. In atherosclerosis, macrophages produce cytokines such as interleukin-1 and 6 and tumor necrosis factor, all of which stimulate hepatocytes to produce fibrinogen and hsCRP. Furthermore, our findings are in accordance with those by Levenson et al24 that fibrinogen was elevated in subjects with silent atherosclerosis, particularly in those with disease in several arterial beds.24 Finally, within the Copenhagen City Heart Study, patients with ischemic heart disease and thus atherosclerosis had elevated fibrinogen.25 Conclusion Our data demonstrate that elevated levels of the inflammatory markers fibrinogen and hsCRP are associated with increased risk of new cerebral ischemic lesions after CEA. Our observation also suggests the possibility that the tested inflammatory markers may provide a method of identifying people for whom a specific antiinflammatory therapy before operation is necessary, a hypothesis requiring direct testing in randomized clinical trials. Author contributions Conception and design: HP Analysis and interpretation: HP, LT Data collection: HP, PH Writing the article: HP, PJ Critical revision of the article: HP, PJ Final approval of the article: HP Statistical analysis: ST Obtained funding: EH Overall responsibility: HP, EH We thank Felicitas Altmayr for her technical support and Bernhard Holzmann, MD, from the Department of Surgery, Technical University of Munich, for providing the laboratory setting. References 1. 1Beneficial effect of carotid endarterectomy in symptomatic patients with high-grade carotid stenosis (North American Symptomatic Carotid Endarterectomy Trial Collaborators). N Engl J Med. 1991;325:445–453. MEDLINE 2. 2Wolf O, Heider P, Heinz M, Poppert H, Schmidt-Thieme T, Sander D, et al. Frequency, clinical significance and course of cerebral ischemic events after carotid endarterectomy evaluated by serial diffusion weighted imaging. Eur J Vasc Endovasc Surg. 2004;27:167–171. Abstract | Full Text |
Full-Text PDF (587 KB)
|
CrossRef
3. 3Bots ML, Elwood PC, Salonen JT, de Freire C, Sivenius J, Di Carlo A, et al. Level of fibrinogen and risk of fatal and non-fatal stroke (EUROSTROKE: a collaborative study among research centres in Europe). J Epidemiol Community Health. 2002;56:i14–i18.
CrossRef
4. 4Kofoed SC, Wittrup HH, Sillesen H, Nordestgaard BG. Fibrinogen predicts ischaemic stroke and advanced atherosclerosis but not echolucent, rupture-prone carotid plaques: the Copenhagen City Heart Study. Eur Heart J. 2003;24:567–576.
CrossRef
5. 5Dosa E, Rugonfalvi-Kiss S, Prohaszka Z, Szabo A, Karadi I, Selmeci L, et al. Marked decrease in the levels of two inflammatory markers, hs-C-reactive protein and fibrinogen in patients with severe carotid atherosclerosis after eversion carotid endarterectomy. Inflamm Res. 2004;53:631–635. MEDLINE |
CrossRef
6. 6Maresca G, Di Blasio A, Marchioli R, Di Minno G. Measuring plasma fibrinogen to predict stroke and myocardial infarction: an update. Arterioscler Thromb Vasc Biol. 1999;19:1368–1377. MEDLINE 7. 7Bini A, Kudryk BJ. Fibrin and its derivatives in the normal and diseased vessel wall. Ann N Y Acad Sci. 1992;667:112–126. MEDLINE |
CrossRef
8. 8Mauriello A, Sangiorgi G, Palmieri G, Virmani R, Holmes DR, Schwartz RS, et al. Hyperfibrinogenemia is associated with specific histocytological composition and complications of atherosclerotic carotid plaques in patients affected by transient ischemic attacks. Circulation. 2000;101:744–750. 9. 9Pai JK, Curhan GC, Cannuscio CC, Rifai N, Ridker PM, Rimm EB. Stability of novel plasma markers associated with cardiovascular disease: processing within 36 hours of specimen collection. Clin Chem. 2002;48:1781–1784. MEDLINE 10. 10Kondo K, Kitagawa K, Nagai Y, Yamagami H, Hashimoto H, Hougaku H, et al. Associations of soluble intercellular adhesion molecule-1 with carotid atherosclerosis progression. Atherosclerosis. 2005;179:155–160. Abstract | Full Text |
Full-Text PDF (101 KB)
|
CrossRef
11. 11Rerkasem K, Shearman CP, Williams JA, Morris GE, Phillips MJ, Calder PC, et al. C-reactive protein is elevated in symptomatic compared with asymptomatic patients with carotid artery disease. Eur J Vasc Endovasc Surg. 2002;23:505–509. Abstract |
Full-Text PDF (106 KB)
|
CrossRef
12. 12Choi H, Cho DH, Shin HH, Park JB. Association of high sensitivity C-reactive protein with coronary heart disease prediction, but not with carotid atherosclerosis, in patients with hypertension. Circ J. 2004;68:297–303. MEDLINE |
CrossRef
13. 13Jahn J, Hellmann I, Maass M, Giannitsis E, Dalhoff K, Katus HA. Time-dependent changes of hs-CRP serum concentration in patients with non-ST elevation acute coronary syndrome. Herz. 2004;29:795–801. MEDLINE |
CrossRef
14. 14Alvarez GB, Ruiz C, Chacon P, Sabin JA, Matas M. High-sensitivity C-reactive protein in high-grade carotid stenosis: risk marker for unstable carotid plaque. J Vasc Surg. 2003;38:1018–1024. Abstract | Full Text |
Full-Text PDF (106 KB)
|
CrossRef
15. 15Zwaka TP, Hombach V, Torzewski J. C-reactive protein-mediated low density lipoprotein uptake by macrophages: implications for atherosclerosis. Circulation. 2001;103:1194–1197. 16. 16Pasceri V, Willerson JT, Yeh ET. Direct proinflammatory effect of C-reactive protein on human endothelial cells. Circulation. 2000;102:2165–2168. 17. 17Hashimoto H, Kitagawa K, Hougaku H, Shimizu Y, Sakaguchi M, Nagai Y, et al. C-reactive protein is an independent predictor of the rate of increase in early carotid atherosclerosis. Circulation. 2001;104:63–67.
CrossRef
18. 18Tanne D, Benderly M, Goldbourt U, Haim M, Tenenbaum A, Fisman EZ, et al. C-reactive protein as a predictor of incident ischemic stroke among patients with preexisting cardiovascular disease. Stroke. 2006;37:1720–1724.
CrossRef
19. 19Ma J, Hennekens CH, Ridker PM, Stampfer MJ. A prospective study of fibrinogen and risk of myocardial infarction in the Physicians’ Health Study. J Am Coll Cardiol. 1999;33:1347–1352. Abstract | Full Text |
Full-Text PDF (216 KB)
|
CrossRef
20. 20Coppola G, Rizzo M, Abrignani MG, Corrado E, Di Girolamo A, Braschi A, et al. Fibrinogen as a predictor of mortality after acute myocardial infarction: a forty-two-month follow-up study. Ital Heart J. 2005;6:315–322. 21. 21Gonzalez-Conejero R, Fernandez-Cadenas I, Iniesta JA, Marti-Fabregas J, Obach V, Alvarez-Sabin J, et al. Role of fibrinogen levels and factor XIII V34L polymorphism in thrombolytic therapy in stroke patients. Stroke. 2006;37:2288–2293.
CrossRef
22. 22Di Napoli M, Papa F, Bocola V. Prognostic influence of increased C-reactive protein and fibrinogen levels in ischemic stroke. Stroke. 2001;32:133–138. 23. 23Ross R. Atherosclerosis–an inflammatory disease. N Engl J Med. 1999;340:115–126. MEDLINE |
CrossRef
24. 24Levenson J, Giral P, Razavian M, Gariepy J, Simon A. Fibrinogen and silent atherosclerosis in subjects with cardiovascular risk factors. Arterioscler Thromb Vasc Biol. 1995;15:1263–1268. MEDLINE 25. 25Sethi AA, Tybjaerg-Hansen A, Gronholdt ML, Steffensen R, Schnohr P, Nordestgaard BG. Angiotensinogen mutations and risk for ischemic heart disease, myocardial infarction, and ischemic cerebrovascular disease (Six case-control studies from the Copenhagen City Heart Study). Ann Intern Med. 2001;134:941–954. MEDLINE a Department of Vascular Surgery, Rechts der Isar Medical Center, Technical University of Munich, Munich, Germany b Department of Neurology, Rechts der Isar Medical Center, Technical University of Munich, Munich, Germany c Department of Neuroradiology, Rechts der Isar Medical Center, Technical University of Munich, Munich, Germany d Institute of Epidemiology and Medical Statistics, Rechts der Isar Medical Center, Technical University of Munich, Munich, Germany.  Reprint requests: Dr Peter Heider, Rechts der Isar Medical Center, Technical University of Munich, Ismaninger Str 22, D-81675 Munich, Germany.
Competition of interest: none. Financial support for this study was provided by the Commission of Clinical Research, Rechts der Isar Medical Center, Technical University of Munich (Kommission für Klinische Forschung, KKF-Nr.: 8744652). PII: S0741-5214(07)00963-9 doi:10.1016/j.jvs.2007.05.035 © 2007 The Society for Vascular Surgery. Published by Elsevier Inc. All rights reserved. | |
|
FULL TEXT00963-9/journalimage?img=PIIS0741521407009639.gr1.lrg.gif&fig=fig1&kwhquery=&issn=0741-5214&ishighres=false&allhighres=false&free=yes','journalimage');)
00963-9/journalimage?img=PIIS0741521407009639.gr2.lrg.gif&fig=fig2&kwhquery=&issn=0741-5214&ishighres=false&allhighres=false&free=yes','journalimage');)
00963-9/journalimage?img=PIIS0741521407009639.gr3.lrg.gif&fig=fig3&kwhquery=&issn=0741-5214&ishighres=false&allhighres=false&free=yes','journalimage');)
00963-9/journalimage?img=PIIS0741521407009639.gr4.lrg.gif&fig=fig4&kwhquery=&issn=0741-5214&ishighres=false&allhighres=false&free=yes','journalimage');)
00963-9/journalimage?img=PIIS0741521407009639.gr5.lrg.gif&fig=fig5&kwhquery=&issn=0741-5214&ishighres=false&allhighres=false&free=yes','journalimage');)
00963-9/journalimage?img=PIIS0741521407009639.gr6.lrg.gif&fig=fig6&kwhquery=&issn=0741-5214&ishighres=false&allhighres=false&free=yes','journalimage');)