| | Effect-modifications by age and sex on the risks of coronary artery disease and revascularization procedures in relation to diabetesReceived 19 January 2006; accepted 18 May 2006. published online 07 July 2006. 1. Introduction  Diabetes becomes epidemic worldwide and there are increasing reports of diabetes in children and adolescents [1], [2]. As the risks of coronary artery disease are higher in diabetic patients [3], [4], [5], [6], [7], the risks associated with younger diabetic subjects are a great concern to our society. Previous studies were limited to the selected population and they usually presented overall risks of coronary artery disease rather than the risks specific to the different age groups of the diabetic patients recruited. Moreover, there are few reports discussing the effects of age and sex on the risks of coronary revascularization procedures in diabetic populations. In Taiwan, the mortality rate associated with diabetes mellitus almost doubled over the past ten years [8], and 19.8% of deaths in diabetic patients were ascribed to cardiovascular disease [9]. There is an increasing trend of childhood diabetes in Taiwan [10] and we need to know the risk of coronary-related complications in all diabetic patients including younger age group. The study presented here aimed to assess the age- and sex-specific effects of diabetes on the risks of coronary-related complications from a nationally representative cohort retrieved from the National Health Insurance (NHI) database of Taiwan. 2. Methods 2.1. Study design and data sources This was a non-concurrent prospective study of coronary artery disease and revascularization procedures among the diabetic population in Taiwan. The NHI program, implemented since 1995, is a universal health program to 96% of total population in Taiwan [11], and 97% of hospitals as well as 90% of clinics all over the island have contract with Bureau of National Health Insurance (BNHI) [12]. The NHI Dataset (NHID) has accumulated eight different registry and eight claims files reported from all contracted hospitals and clinics. Details of both registry and claims files are described elsewhere [13]. To rapidly and effectively respond to current and emerging health issues, the BNHI cooperates with the National Health Research Institute (NHRI) and routinely establishes the NHID to enable health researchers to analyze and improve the health of Taiwan residents. To ensure the accuracy of claim files, the BNHI performed experts’ review on a random sample of every 50 to 100 ambulatory and inpatient claims quarterly [14], and false report of diagnosis would face severe penalty from the BNHI [9], [14]. With the ethnical approval of the NHRI, we used the diabetic ambulatory care expenditures by visits (1997–2002), all inpatient expenditures by admissions (1997–2002) and an updated registry for beneficiaries (1995–2002) in this study. The datasets used in this study can be inter-linked through each individual's unique personal identification number (PIN). 2.2. Selection of diabetic and control groups Diabetic ambulatory care claim records those patients coded with diabetes-related diagnoses either with International Classification of Diseases Ninth Revision (ICD-9) 250 or A-code 181. An individual was classified as a case of diabetes if he/she received an ambulatory care in 1997 for a diabetes-related diagnosis, and had one or more ambulatory visits with the same diagnosis in the subsequent 12 months. In addition, the first and the last outpatient visits within a 12-month period must be more than 30 days to avoid including those accidentally miscoded patients. The final diabetic cohort thus included 500,868 patients altogether. All diabetic patients either with diet control or with hypoglycemic agents were included in this diabetic group if their attending doctors gave a diabetes-related diagnosis. The index date for the patients in the diabetic group was the date of their first outpatient visit in 1997. The age- and sex-matched control group was identified from the registry for beneficiaries, which accumulated information of beneficiaries including PIN, date of birth, gender, geographic area of each member's NHI unit, date of enrollment and withdrawal from NHI each time between March 1995 and December 2002. Due to the missing information on age or sex for 620 diabetic patients, we could select only 500,248 control subjects in this analysis. The index date for subjects in the control group was the first date of enrollment to NHI. If their first dates of enrollment were before January 1, 1997, we set the index date as January 1, 1997, the starting point of this study. Age of each study subject was estimated by the difference in time between the index date and the date of birth. We grouped the area of each member's NHI unit, either the beneficiaries’ residential area or the location of their employment, into four geographic areas or three urbanization statuses according to the National Statistics of Regional Standard Classification [15]. 2.3. Study end-points The inpatient claim includes the records of all hospitalization events and provides various information including PIN, date of birth, gender, date of admission and discharge, a maximum of five leading discharged diagnoses and four operation procedure codes. With the unique PIN, we linked both diabetic and control subjects to the inpatient claim data to identify the first episode of primary or secondary discharged diagnoses or procedure codes of the following coronary artery disease and revascularization procedures as the endpoints of this study: acute myocardial infarction (ICD-9: 410), ischemic heart disease (ICD-9: 410–414), coronary revascularization procedures (ICD-9: 36.0, 36.01, 36.02, 36.05, 36.06, 36.1, 36.10–36.19). All the outcomes were evaluated separately. 2.4. Statistical analysis The follow-up was made from January 1, 1997 to December 31, 2002. The date of encountering each clinical endpoint was the first day of hospitalization. The study subjects who met the following criteria were censored in the survival analysis. First, if they died in the hospital for reasons not relevant to the clinical outcomes of interest, the date of censoring was the date patients died. Second, if the subject who did not encounter in-hospital mortality, the date of censoring was either the date of their last withdrawal from NHI (from the registry of beneficiaries) or the date of termination of follow-up, i.e., December 31, 2002. To evaluate the potential independent effect of age and sex on the risk of coronary-related complications in the diabetic and control groups separately, we conducted Cox proportional hazard regression models with age, sex, geographic area, and urbanization status adjusted simultaneously in the model. To have a better understanding of the association of coronary artery disease and related procedures risks and specific ages, we decided to categorize age into groups with a small interval, i.e., 10-year period. Additionally, previous studies rarely looked into risk of coronary complications for young diabetic adults simply due to a limited number of study subjects. Thus, we decided to use subjects aged <35 as a reference and examined relative risks of coronary complications for diabetic adults aged 35 and more. Adjustment for both geographic areas and urbanization was made due to the fact that there was a clear urban–rural difference in accessibility to medical care in Taiwan [16]. We further performed the trend tests to examine whether age posed any linear gradient relationship with the risk of endpoints of interest. In addition, we assessed the sex-specific effects of diabetes on the risks of coronary-related complications. Moreover, the potential effect-modification by age and sex was evaluated by both statistical testing for the interaction of diabetes with age and sex stratification analysis. All statistical analyses were performed with the SAS version 8.2 (SAS Institute, Cary, NC, USA). A p-value of less than 0.05 was set to declare statistical significance. 3. Results 3.2. Relative hazards of coronary artery diseases Table 2 shows the HRs of coronary artery disease in relation to diabetes, age, and sex. Compared to those aged less than 35 years, the HR increased with age irrespective of diabetic status. For the diabetic group, the highest HR was observed in the subjects aged >84 years for acute myocardial infarction (HR =18.1, 95% CI: 12.7–25.7) and ischemic heart disease (HR=16.7, 95% CI: 14.6–19.0). The positive relationship between age and HR was even more prominent in the control group in which the HRs for the subjects aged >84 years ranged from 88.7 for acute myocardial infarction to 154.3 for ischemic heart disease. The linear trend test disclosed a significant dose–response relationship (p<0.0001) between age and the HRs of all coronary artery disease in both groups. Males had higher HRs of coronary artery disease, and it was more significant in the control group especially for acute myocardial infarction. There were significant interactions of diabetes with both age and sex. Compared to the control group, the overall HRs of coronary artery disease in relation to diabetes were around two (1.97–2.20), and the associated HR increased with decreasing age. The highest HR was consistently observed in the diabetic patients aged <35 years old (Fig. 1). Compared to the control group with the same gender, the relative hazards of the diabetic patients increased consistently in both sexes and the higher HRs were observed in female diabetic patients in all kinds of coronary artery disease (Fig. 2). 4. Discussion In Taiwan, diabetic patients had increased risks of coronary artery disease and revascularization procedures with relative hazards between 1.97 and 2.7. Though the absolute risks of coronary events were higher in men and elderly people, the diabetic patients with younger ages and female patients tended to suffer from higher relative risks of all coronary complications. The absolute risks of coronary artery disease and revascularization procedures significantly increased with age for both the diabetic and their controls. Risks of coronary artery disease increased with age up to >84 years old, but coronary revascularization procedures were less likely to perform beyond 75 years. Arterial stiffening, hypertension and reducing physical activity [17], [18] with age may have contributed to the increased risks of coronary heart disease observed in the older age groups. However, this increasing trend with age was more remarkable in the control group than in the diabetic population. It seems that diabetes “attenuates” the effect of age on the risk of coronary heart disease; once a person develops diabetes, he/she would tend to experience a relatively homogeneous risk of complications so that the role of age becomes less important in determining the risk of these complications in diabetic patients. In our study, men had absolutely higher risks of coronary artery disease than women irrespective of their diabetes status. Commoner unfavorable coronary risk profile such as smoking, higher blood pressure, total cholesterol and lower high density protein in men [19], [20] associated with protective estrogen effect in women [21] might have caused this gender difference in the risk of coronary events. Similarly, men were observed to be more likely to undergo coronary revascularization procedures which were consistent with the result from the Italian study [22]. As for the effects of diabetes on the risk of coronary artery disease complications, this study noted an overall relative hazard of two, which was also similar to the previous reports [3], [7], [23]. The overall risk of coronary revascularization procedures were higher, with HR around 2.7 (95% CI: 2.6–2.8). The age-specific relative hazards estimated, however, greatly vary with different age groups. Generally, the HRs decreased with age and the highest risks were observed in the patients aged <35 years. These results suggest that age plays a significant role as an effect-modifier that moderates the effect of diabetes on all coronary-related complications. Previous studies [4], [24] also reported higher relative risks of cardiovascular events in younger onset diabetic patients but they did not further specify the relative risks according to different age stratifications. Knowledge of relative hazards in different age groups is important in daily diabetic practice. Alarmingly high relative hazards of coronary endpoints in those diabetes aged <35 years may contribute an emerging health problem to our society. O’Connor et al. [25] stated that younger diabetic patients tended to have poor glycemic control, poor health-related behaviors, less frequent clinic visits as well as regular assessment of diabetes-related complications. This could also lead to such a high relative hazards of various coronary complications among young diabetic patients. We have to exert a more aggressive approach to the younger diabetic patients who are apparently vulnerable to all coronary-related complications. We were unable to differentiate between type 1 and type 2 diabetes in the young study subjects recruited in the present analysis. Nonetheless, in Taiwan, type 1 diabetes constitutes only 1.8% of all types of diabetes [26] and the ratio of newly diagnosed type 2 to type 1 diabetes among school children aged 6–18 years is approximately 6:1 [10]. Therefore, the majority of the young diabetic patients in our study tend to be constituted by type 2 diabetic patients. Risk factors for type 2 diabetes of Taiwanese children were reported to be obesity, high blood pressure, hypercholesterolemia and family history of diabetes [10]. Aggregation of such cardiovascular risk factors might partially explain the increased risks of coronary-related endpoints in our young diabetic patients. Interventions designed to effectively reduce these risk factors should be considered and evaluated by future studies in order to prevent the occurrence of young diabetic patients, who suffered very high risks of all coronary complications. In addition to age, sex was also demonstrated to be a significant effect-modifier that modifies the effect of diabetes on all coronary-related complications. Although the Asia-Pacific Cohort Studies [23] did not find any sex difference on the relative risks of major cardiovascular disease in the diabetic patients, our study noted that females were associated a higher risks of coronary-related complications which was consistent with some other studies [3], [5], [6]. The reason for an increased risk associated with diabetic women has been postulated that diabetic women are more vulnerable to have lower high-density lipoprotein cholesterol [27], [28], [29], alteration in low-density lipoprotein size [28], and impaired endothelial dysfunction [30] than diabetic men. The higher relative hazard in female diabetic patients was especially prominent for the coronary revascularization procedures, which warrants clinical notification. Regional differences in the risks of coronary events and related procedures were also noted in our study. The diabetic patients living or working in the rural areas had elevated risks of coronary artery disease but they were less likely to undergo revascularization procedures. Whether this implies insufficient medical resources in these areas or it is simply due to differences in physicians practice in different regions needs further investigation. Our study had several methodological strengths. First, the follow-up was based on a linkage of study subjects’ PIN to the records of inpatient expenditures by admissions. This way of doing left little room for loss to follow-up. Additionally, the control sample was also population-based. Second, one of the potential advantages of using insurance claims data sets for clinical research like ours is that the longitudinal records for a large sample of geographically dispersed patients can be easily obtained [31]. There were also fewer tendencies of low response rate and recall bias of cohort members of this study. Third, such a large number of study subjects also made it possible for us to make stratified analyses according to certain variables of interest such as age and sex. Despite the above strength, we also recognized that the quality of NHID is crucial to the findings from our study. A previous survey of a total of 1350 residents all over Taiwan for their self-report of having diabetes or not; the questionnaire included the following five items: I1) Have doctors ever told you that you have diabetes, high blood sugar, or sugar in your urine (exclusive of pregnancy)? (2) Have you ever taken an oral hypoglycemic agent? (3) Have you ever used an insulin injection? (4) Have you had an examination for blood sugar during the year 2000? (5) What was/were the result(s) of the blood sugar test(s)? The possible answers were positive, negative and uncertain. Patients were classified as diabetic if they gave any affirmative response to the first three items of the questionnaire. In addition, subjects with a negative or uncertain response to these three questions, but who had hypoglycemic agents in pharmacy claims data were also classified as diabetic. The information obtained from questionnaires was then compared against the NHID. The validation study found an overall rate of concordance of 74.6% [32]. Such findings are similar to those reported in other countries [33], [34]. The erroneous information of diabetes would lead to exposure misclassification associated with ascertainment of diabetic patients. Additionally, the study participants of the control groups were selected from those who had no diabetic diagnosis in 1997. Such inclusion criteria might lead to recruitment of some diabetic patients with no ambulatory care visits. Nonetheless, the aforementioned exposure misclassification bias was likely to be non-differential, which would tend to underestimate rather than overestimate the true relative hazard [35]. Besides, we had used at least two diabetes-related diagnoses with the first and the last visits more than 30 days apart, which would largely reduce the likelihood of exposure misclassification. Another validity study of claim data performed in a medical center showed there were high sensitivity and specificity of diabetes and acute myocardial infarction diagnoses [36]. Any erroneous information of disease diagnoses is again likely to lead to non-differential disease misclassification that will attenuate the real hazard ratios. Moreover, the diagnoses of coronary complications were retrieved from inpatient claim files which are believed to be more valid than the ambulatory care claim data [37]. In addition to the issue of data quality, we also noted several other limitations in our study. Firstly, as we previously described, we were unable to differentiate between type 1 and type 2 diabetes in our study. Secondly, under the regulation by the Personal Information Protection Act, the NHID include only scrambled identification numbers and does not allow external linkage to other nationwide registries including Taiwan's mortality registry, which might be able to provide an endpoint more valid than what was used in the current analysis. Third, we did not know the duration of diabetes for each study patient. Nor did we have idea about the associated risk factors (such as smoking and sedative lifestyle) and co-morbidities of the study populations, which might entails certain degrees of confounding in our analyses. Thirdly, since the diabetic cohort included in this study were prevalent cases of diabetes who might have a prior history of coronary complications. Lastly, very few clinical events in female control subjects <35 years also prevented us from performing further age and sex stratification analysis. In conclusion, the NHI database which nearly covers the entire population of Taiwan, may be a useful tool for surveillance of coronary-related complications in diabetic population. Coronary events, which are a tremendous threat to all diabetic patients especially in a younger age group, will be a major health problem to our society. This information gives a hint to the necessity of aggressive diabetic care system to high-risk groups to prevent premature morbidity and mortality from coronary-related complications. Acknowledgements This study was partially supported by a contract with the National Scientific Council (NSC-93-2320-B-030-006). The interpretation and conclusions contained herein do not represent those of Bureau of National Health Insurance, Department of Health or National Health Research Institutes. References [1]. [1]Alberti G, Zimmet P, Shaw J, Bloomgarden Z, Kaufman F, Silink M. Type 2 diabetes in the young: the evolving epidemic. The International Diabetes Federation Consensus Workshop. Diabetes Care. 2004;27:1798–1811. MEDLINE |
CrossRef
[2]. [2]Pinhas-Hamiel O, Zeitler P. The global spread of type 2 diabetes mellitus in children and adolescents. J Pediatr. 2005;146:693–700. Full Text |
Full-Text PDF (159 KB)
|
CrossRef
[3]. [3]Kannel WB, McGee DL. Diabetes and glucose tolerance as risk factors for cardiovascular disease: the Framingham Study. Diabetes Care. 1979;2:120–126. MEDLINE [4]. [4]Manson JE, Colditz GA, Stampfer MJ, Willett WC, Krolewski AS, Rosner B, et al. A prospective study of maturity-onset diabetes mellitus and risk of coronary heart disease and stroke in women. Arch Intern Med. 1991;151:1141–1147. MEDLINE [5]. [5]Kuusisto J, Mykkänen L, Pyörälä K, Laakso M. NIDDM and its metabolic control predict coronary heart disease in elderly subjects. Diabetes. 1994;43:960–967. MEDLINE [6]. [6]Laakso M, Rönnemaa T, Lehto S, Puukka P, Kallio V, Pyörälä K. Does NIDDM increase the risk for coronary heart disease similarly in both low- and high-risk populations?. Diabetologia. 1995;38:487–493.
CrossRef
[7]. [7]Abbott RD, Curb JD, Rodriguez BL, Masaki KH, Yano K, Schatz IJ, et al. Age-related changes in risk factor effects on the incidence of coronary heart disease. Ann Epidemiol. 2002;12:173–181. Abstract | Full Text |
Full-Text PDF (155 KB)
|
CrossRef
[8]. [8]Chang CJ, Lu FH, Yang YC, Wu JS, Wu TJ, Chen MS, et al. Epidemiologic study of type 2 diabetes in Taiwan. Diabetes Res Clin Pract. 2000;50(Suppl. 2):S49–S59. Abstract | Full Text |
Full-Text PDF (99 KB)
|
CrossRef
[9]. [9]Tseng CH. Mortality and causes of death in a national sample of diabetic patients in Taiwan. Diabetes Care. 2004;27:1605–1609. MEDLINE |
CrossRef
[10]. [10]Wei JN, Sung FC, Lin CC, Lin RS, Chiang CC, Chuang LM. National surveillance for type 2 diabetes mellitus in Taiwanese children. JAMA. 2003;290:1345–1350.
CrossRef
[11]. [11]Lu JF, Hsiao WC. Does universal health insurance make health care unaffordable? Lessons from Taiwan. Health Aff. 2003;22:77–88. [12]. [12]Chiang TL. Taiwan's 1995 health care reform. Health Policy. 1997;39:225–239. Abstract |
Full-Text PDF (1082 KB)
|
CrossRef
[13]. [13]National Research Health Research Database Website. Available at: http://www.nhri.org.tw/nhird/date_04.htm. [14]. [14]Bureau of National Health Insurance Website. Available at: http://www.nhi.gov.tw/information/bulletin_file/421_0890036465-19.doc. [15]. [15]Directorate-General Budget, Accounting and Statistics. National Statistics of Regional Standard Classification data from Directorate-General Budget, Accounting and Statistics, ROC, 1993. [16]. [16]Tan HF, Tseng HF, Chang CK, Lin W, Hsiao SH. Accessibility assessment of the Health Care Improvement Program in rural Taiwan. J Rural Health. 2005;21:372–377.
CrossRef
[17]. [17]Gerstenblith G, Lakatta EG. Aging and the cardiovascular system. In: Willerson JT, Cohn JN editor. Cardiovascular Medicine. Philadelphia: Churchill Livingstone; 2000;p. 2015–2024. [18]. [18]Llyod-Jones DM, Kannel WB. Coronary risk factors: an overview. In: Willerson JT, Cohn JN editor. Cardiovascular medicine. Philadelphia: Churchill Livingstone; 2000;p. 2193–2215. [19]. [19]Njølstad I, Arnesen E, Lund-Larsen PG. Smoking, serum lipids, blood pressure, and sex differences in myocardial infarction. A 12-year follow-up of the Finnmark Study. Circulation. 1996;93:450–456. MEDLINE [20]. [20]Jousilahti P, Vartiainen E, Tuomilehto J, Puska P. Sex, age, cardiovascular risk factors, and coronary heart disease. A prospective follow-up study of 14,786 middle-aged men and women in Finland. Circulation. 1999;99:1165–1172. MEDLINE [21]. [21]Shahar E, Folsom AR, Salomaa VV, Stinson VL, McGovern PG, Shimakawa T, et al. Relation of hormone replacement therapy to measures of plasma fibrinolytic activity. Circulation. 1996;93:1970–1975. MEDLINE [22]. [22]The DAI Study Group. The prevalence of coronary heart disease in type 2 diabetic patients in Italy: the DAI Study. Diabet Med. 2004;21:738–745. MEDLINE |
CrossRef
[23]. [23]Asia Pacific Cohort Studies Collaboration. The effects of diabetes on the risk of major cardiovascular diseases and death in the Asia-Pacific Region. Diabetes Care. 2003;26:360–366. MEDLINE |
CrossRef
[24]. [24]Hillier TA, Pedula KL. Complications in young adults with early-onset type 2 diabetes. Losing the relative protection of youth. Diabetes Care. 2003;26:2999–3005. MEDLINE |
CrossRef
[25]. [25]O’Connor PJ, Desai JR, Solberg LI, Rush WA, Bishop DB. Variation in diabetes care by age: opportunities for customization of care. BMC Fam Pract. 2003;4:16. MEDLINE |
CrossRef
[26]. [26]Chuang LM, Tsai ST, Huang BY, Tai TYThe DIABCARE (Taiwan) Study Group. The current state of diabetes management in Taiwan. Diabetes Res Clin Pract. 2001;54(Suppl. 1):S55–S65. [27]. [27]Barrett-Connor EL, Cohn BA, Wingard DL, Edelstein SL. Why is diabetes mellitus a stronger risk factor for fatal ischemic heart disease in women than in men? The Rancho Bernardo Study. JAMA. 1991;265:627–631. MEDLINE [28]. [28]Howard BV, Cowan LD, Go O, Welty TK, Robbins DC, Lee ET. Adverse effects of diabetes on multiple cardiovascular disease risk factors in women. The Strong Heart Study. Diabetes Care. 1998;21:1258–1265. MEDLINE |
CrossRef
[29]. [29]Lee WL, Cheung AM, Cape D, Zinman B. Impact of diabetes on coronary heart disease in women and men: a meta-analysis of prospective studies. Diabetes Care. 2000;23:962–968. MEDLINE |
CrossRef
[30]. [30]Steinberg HO, Paradisi G, Cronin J, Crowde K, Hempfling A, Hook G, et al. Type II diabetes abrogates sex differences in endothelial function in premenopausal women. Circulation. 2000;101:2040–2046. [31]. [31]Jollis JG, Ancukiewicz M, DeLong ER, Pryor DB, Muhlbaier LH, Mark DB. Discordance of databases designed for claims payment versus clinical information systems. Implications for outcomes research. Ann Intern Med. 1993;119:844–850. MEDLINE [32]. [32]Lin CC, Lai MS, Syu CY, Chang SC, Tseng FY. Accuracy of diabetes diagnosis in health insurance claims data in Taiwan. J Formos Med Assoc. 2005;104:157–163. MEDLINE [33]. [33]Fisher ES, Whaley FS, Krushat WM, Malenka DJ, Fleming C, Baron JA, et al. The accuracy of Medicare's hospital claims data: progress has been made, but problems remain. Am J Public Health. 1992;82:243–248. MEDLINE |
CrossRef
[34]. [34]Hux JE, Ivis F, Flintoft V, Bica A. Diabetes in Ontario: determination of prevalence and incidence using a validated administrative data algorithm. Diabetes Care. 2002;25:512–516. MEDLINE |
CrossRef
[35]. [35]Gordis L. More on causal inferences: bias, confounding, and interaction. Epidemiology. Philadelphia: W.B. Saunders; 2000;. [36]. [36]P. Chou, Validation of comorbidity derived from health insurance data, (2003–2004) Science and Technology Policy Research and Information Center, National Applied Research Laboratories Website. Available at: http://grbsearch.stic.gov.tw/tiff/92/rrpg92070589.pdf. [37]. [37]Lai HT, Yaung CL, Fan BY. Study on the quality and its related factors of international classification of disease coding system in national health insurance. Chin J Public Health. 1998;17:337–347. a Department of Endocrinology, Far Eastern Memorial Hospital, Taipei Hsien, Taiwan b Department of Public Health, College of Medicine, Fu Jen Catholic University, 510, Chung Cheng Rd., Hsinchuang, Taipei Hsien 242, Taiwan  Corresponding author. Tel.: +886 2 29053432; fax: +886 2 29052098.
PII: S0168-8227(06)00223-3 doi:10.1016/j.diabres.2006.05.020 © 2006 Elsevier Ireland Ltd. All rights reserved. | |
|
FULL TEXT
00223-3/journalimage?img=PIIS0168822706002233.gr1.lrg.gif&fig=fig1&kwhquery=&issn=0168-8227&ishighres=false&allhighres=false&free=yes','journalimage');)
00223-3/journalimage?img=PIIS0168822706002233.gr2.lrg.gif&fig=fig2&kwhquery=&issn=0168-8227&ishighres=false&allhighres=false&free=yes','journalimage');)