- Email: [email protected]
Association between postprandial hyperinsulinemia and coronary artery disease among non-diabetic women: a case control study
International Journal of Cardiology 88 (2003) 215–221 www.elsevier.com / locate / ijcard
Association between postprandial hyperinsulinemia and coronary artery disease among non-diabetic women: a case control study Mehmet Baltali*, Mehmet E. Korkmaz, H. Tarik Kiziltan, I. Haldun Muderris, Bulent Ozin, Ruksan Anarat Cardiology, Cardiovascular Surgery, and Biochemistry, Faculty of Medicine, Bas¸kent University, Adana, Turkey Received 29 October 2001; received in revised form 1 July 2002; accepted 20 July 2002
Abstract Background: We planned a case-control study to assess the relation of fasting glucose, fasting insulin, postprandial glucose and postprandial insulin levels with coronary artery disease in nondiabetic women. Methods: Among 968 consecutive nondiabetic women screened, 104 with coronary artery disease (mean age 60, 469) made up the study cohort (group I). One-hundred and four age-matched, nondiabetic women without coronary artery disease who had a similar lipid and blood pressure profile (group II), and 52 healthy, age-matched women served as controls (group III, real control group). Demographics, waist circumference, lipids, fasting glucose postprandial glucose, fasting and postprandial insulin levels were compared among the groups. A separate subgroup analysis were performed in patients with metabolic syndrome. Results: No differences were identified in terms of prevalences of risk factors between group I and group II. Women with coronary artery disease had higher postprandial insulin level than the women in group II and group III. In reverse stepwise logistic regression analysis postprandial hyperinsulinemia was found to be the single independent determinant for coronary artery disease for the entire study group as well as for women with metabolic syndrome. Conclusion: Our data demonstrate that postprandial hyperinsulinemia is independently associated with coronary artery disease, irrespective of fasting glucose, postprandial glucose, and fasting insulin levels in nondiabetic women with clusterings of factors of metabolic syndrome. 2002 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Coronary artery disease; Insulin resistance; Metabolic syndrome
1. Introduction Insulin resistance syndrome consists of a cluster of factors including visceral obesity, hyperinsulinemia, impaired glucose tolerance or diabetes mellitus, hypertriglyceridemia, and hypertension in a single patient [1]. This grouping of well-known metabolic factors, also called the ‘deadly quartet’, is associated with coronary artery disease [2,3]. When this syndrome is expressed, female patients suffer a substantial in*Corresponding author. E-mail address: [email protected] (M. Baltali).
crease in mortality [4]. The association between insulin resistance syndrome and coronary artery disease is complex, and pathways by which the insulin resistant state adversely affects coronary risk factors and the risk for coronary artery disease need to be elucidated. Postprandial lipemia may be a mechanical link between insulin resistance and coronary artery disease [5]. Elevated insulin levels have a negative impact on both risk factor profile and atherosclerotic process [6–8]. A meta-analysis revealed that hyperinsulinemia was a weak but positive indicator of cardiovascular events in middle-aged men [9]. Studies examining insulin level and cor-
0167-5273 / 02 / $ – see front matter 2002 Elsevier Science Ireland Ltd. All rights reserved. doi:10.1016/S0167-5273(02)00399-6
216
M. Baltali et al. / International Journal of Cardiology 88 (2003) 215–221
onary artery disease events in women did not show any association [10,11]. We planned a case-control study to assess the relation of fasting glucose, postprandial glucose, fasting insulin and postprandial insulin levels with presence of coronary artery disease among non-diabetic elderly women in Adana, Turkey.
2. Methods
Individuals with heart failure, malignancy, chronic renal failure, stroke history, thyroid disease, hemorrhagic disorders, rheumatic valve disease, diabetes, smoking history, and hepatic failure were excluded. Patients taking hormone replacement therapy, and medications thought to alter insulin and glucose profile, such as adrenergic beta-blocker, were also excluded. All patients gave informed consent and study protocol was approved by the Bas¸kent University Internal Review Board.
2.1. Study population
2.2. Study design
The study population consisted of 104 non-diabetic women with coronary artery disease (group I), 104 age-matched non-diabetic women without coronary artery disease who had metabolic risk factors including hypertension, obesity, hypertriglyceridemia, and low high-density lipoprotein cholesterol level (group II), and 52 age-matched healthy subjects with no metabolic risk factors (group III). Subjects in the group I and group II were selected from a total of 968 consecutive non-diabetic women evaluated between September 1999 and December 2001 in the outpatient clinic of Bas¸kent University, Adana Medical Center. Group I consisted of patients with angiographic evidence of coronary artery disease, a history of coronary bypass surgery or balloon angioplasty. For every patient in group I, a subject of the same age, with similar body mass index, smoking status, history of hypertension and similar baseline metabolic lipid profile who did not have coronary artery disease was identified from the computer database, and these subjects constituted group II. Subjects in group III (real control group) were selected from individuals in the check-up clinic. These were healthy non-diabetic women who did not have metabolic risk factors including hypertension, obesity, hypertriglyceridemia and low high-density lipoprotein cholesterol levels. One age-matched control (n552) per two cases in group 1 was selected from the database in the check-up clinic. Patients with metabolic syndrome in group I (n5 64), age-matched individuals with metabolic syndrome in group II (n564), and age-matched subjects without metabolic syndrome in group III (n532) were selected and evaluated in a subgroup analysis.
Age, menopausal status, smoking habits, medication, history of coronary artery disease, hypertension, diabetes mellitus, and heredity were recorded for all patients. Physical examination and blood pressure measurements were obtained. Weight, height, waist and hip circumference were measured for each subject; body mass index, obesity, and visceral obesity were determined from these parameters. Waist circumference was measured at the narrowest diameter between costal margin and the iliac crest, and hip circumference was measured at the greatest diameter over the glutei. Bloods samples were drawn after 12-h fasting for the measurement of total cholesterol, HDL-cholesterol, triglyceride, fasting glucose and insulin levels. Plasma glucose and insulin responses were evaluated 120 min after oral (75 g) glucose load. All subjects underwent treadmill exercise test or thallium myocardial perfusion scintigraphy. Patients with ischemic results in non-invasive tests were offered coronary angiograms. Cases with a history of balloon angioplasty or coronary artery bypass grafting, or .70% luminal narrowing in at least one major coronary artery were defined as those having coronary artery disease (n562) and were compared with the control groups.
2.3. Definitions Following definitions were used: diabetes mellitus, fasting glucose .126 mg / dl, a glucose level .200 mg / dl 120 min after oral glucose load, or usage of antidiabetic drugs; obesity, body mass index .30 kg / m 2 ; hypertension, blood pressure .140 / 90 mmHg on two separate examinations, or usage of
M. Baltali et al. / International Journal of Cardiology 88 (2003) 215–221
antihypertensive agents; metabolic dyslipidemia, a fasting triglyceride level .200 mg / dl and / or a fasting high-density lipoprotein cholesterol level ,40 mg / dl. Metabolic syndrome was defined as having at least three of the following conditions: triglyceride level .150 mg / dl; high-density lipoprotein cholesterol level ,50 mg / dl in women and ,40 mg / dl in men; waist circumference .88 cm in women and .102 cm in men; systolic blood pressure .130 / 85 mmHg or usage of antihypertensive drugs; fasting glucose level .110 mg / dl [12].
2.4. Laboratory measurements Total cholesterol, high density lipoprotein cholesterol, triglyceride, and blood glucose measurements were estimated on an Hitachi 912 autoanalyser (Hitachi, Tokyo, Japan) with standard kits (Boehringer Mannheim). Total cholesterol was measured using CHOD-PAP, glucose using GOD-PAP, and triglycerides using GPO-PAP techniques. High-density lipoprotein cholesterol was estimated by quantitative enzymatic method. Low density lipoprotein cholesterol was calculated using the Friedewald formula. Fasting and postprandial insulin were analyzed on an AxSYM System-Abbott analyzer (Abbott Diagnostics, Park, IL, USA) with 67-5857 / R2 kits using micro-particle enzyme immune assay technique. This assay is highly specific for insulin with essentially no cross-reactivity with proinsulin (,0.005%), C-peptide or 32–33 pro peptides.
2.5. Statistical analysis Data were collated and analyzed with SPSS computer program. The numeric variables were expressed as the mean6S.D., the categorical variables as percentage. The relationship between risk factors among the groups were determined using Student’s t-test for numeric variables and chi-square test for categorical variables. Values for triglyceride levels, fasting and postprandial insulin levels not conforming to a normal distribution were log transformed for comparison and were presented as median with 25th and 75th percentiles. Separate analyses were performed on cases (group I) and controls (group II and group III), and on the
217
subgroup of cases with metabolic syndrome [12]. Reverse stepwise logistic regression analysis was performed to determine the independent association of fasting glucose, fasting insulin, postprandial glucose and postprandial insulin levels with coronary artery disease and metabolic syndrome. A P-value ,0.05 was considered statistically significant.
3. Results
3.1. Clinical and metabolic characteristics of cases As shown in Table 1, 16.3% of the subjects in group I had previous balloon angioplasty, 21.2% had previous coronary bypass surgery, and 71.2% had arteriographically diagnosed coronary artery disease. Patients in group I and group II had similar risk factor profiles. Subjects in group III consisted of healthy patients with a similar age profile, who did not have obesity, hypertriglyceridemia, or low highdensity lipoprotein cholesterol levels. Of individuals in group I and group II, 81% were hypertensive, while incidences for obesity and metabolic dyslipidemia were 52.9 and 35.6%, respectively.
3.2. Comparison of glucose and insulin levels Table 1 compares fasting-postprandial glucose and insulin levels of group I, controls with multiple risk factors, and real controls. Patients in group I had higher postprandial insulin levels than subjects in both control groups (group II and group III). Fasting insulin level did not differ between group I and group II, however patients in these groups had higher fasting insulin levels than those in group III. The difference between group II and group III was not statistically significant. Patients in group II had higher fasting glucose levels than those in group III. Postprandial glucose levels did not differ significantly among the groups. We also compared coronary risk factor profiles, fasting and postprandial glucose and insulin levels in individuals with metabolic syndrome in a subgroup analysis (Table 2). Subjects in group I and group II had higher fasting glucose, fasting insulin and postprandial insulin levels than those in group III, and
218
M. Baltali et al. / International Journal of Cardiology 88 (2003) 215–221
Table 1 Characteristics, risk factor profile, coronary artery disease diagnosis and use of medication
Age (years) Hypertension, n (%) Body mass index (kg / m 2 ) Obesity Waist circumference (cm) Smoking, n (%) Triglyceride (mg / dl)a Total cholesterol (mg / dl) LDL-cholesterol (mg / dl) (n5103) HDL-cholesterol (mg / dl) Metabolic dyslipidemia (%) Fasting insulin a Postprandial insulin a Fasting blood glucose (mg / dl) Postprandial blood glucose (mg / dl)
Group I (n5104)
Group II (n5104)
Group III n552
60.469 84.0 (80.8)* 30.164* 55 (52.9) 99.1610* 17 (16.3) 135.5 (114–173)* 217.4642 138.6637 48.4613* 37 (35.6)* 11.0 (8–14)* 88.1 (63.5–119.9)* ‡ 97.2611 136.1633
60.469 84.0 (80.8)* 30.564* 55 (52.9) 99.869* 17 (16.3) 129.0 (104–179.5)* 214.7639 134.7635 48.9613* 37 (35.6)* 10.0 (7.5–13)* 62.1 (39.3–97.6) 97.6610* 136.6634
59.168 – 26.362 – 87.169 3 (6) 107.5 (82.3–133) 207.1635 127.0633 57.8614 – 8.0 (6–10) 48.5 (38.4–60.8) 93.8612 129.1630
Coronary artery disease diagnosis Arteriography, n (%) CABG history, n (%) PTCA history, n (%)
74 (71.2) 22 (21.2) 17 (16.3)
Use of medication Beta blocker, n (%)b Ca-channel antagonist, n (%) Diuretics, n (%) Nitrate, n (%) Aspirin, n (%)
54 36 10 51 73
(51.9)* ‡ (34.6)* (9.6)* (49.0)* ‡ (70.2)* ‡
– – – 14 41 14 8 17
– – – (13.5)* (39.4)* (13.5)* (7.6)* (16.3)
– – – – 2 (1.9)
HDL, high-density lipoprotein; LDL, low-density lipoprotein. *P,0.05 versus group III. ‡ P,0.05 versus group II. a Median (25th–75th percentile). b Except beta-blockers with alpha-adrenergic activity.
Table 2 Characteristics and risk factor profiles of coronary artery disease cases with metabolic syndrome (group I), age-matched subjects without coronary artery disease and with metabolic syndrome (group II), and age-matched healthy controls without metabolic syndrome (group III)
Age (years) Hypertension, n (%) Body mass index (kg / m 2 ) Waist circumference (cm) Smoking, n (%) Triglyceride (mg / dl)a Total cholesterol (mg / dl) LDL-cholesterol (mg / dl) HDL-cholesterol (mg / dl) Fasting insulin a Postprandial insulin a Fasting blood glucose (mg / dl) Postprandial blood glucose (mg / dl)
Group I (n564)
Group II (n564)
Group III (n532)
58.669 59 (92.2%)* 30.864* 102.568* 10 (15.6%) 157.0 (126.8–198.5) 221.5643 144.8637* 41.869* 12.0 (9.1–14.9)* 99.3 (72.4–124.8)* ‡ 99.669* 139.0634
58.969 57 (89.1%)* 30.964* 101.367* 13 (20.3%) 166.5 (128.8–212.8)* 214.8641 135.0637 42.6610* 10.0 (8.6–14.2)* 66.2 (50.4–104.6)* 99.5611* 144.1634*
58.368 – 25.963 82.866 3 (9.4%) 94.0 (69.3–122.8) 206.3638 127.0633 60.7611 8.0 (6.0–10.4) 46.0 (38.5–61.2) 91.1610 125.8627
HDL, high-density lipoprotein; LDL, low-density lipoprotein. *P,0.05 versus group III. ‡ P,0.05 versus group II. a Median (25th–75th percentile).
M. Baltali et al. / International Journal of Cardiology 88 (2003) 215–221
219
Table 3 Logistic regression analysis Entire study group Determinants of coronary artery disease (subjects in group I and group II) Postprandial insulin, P OR per increase by 1 mIU / ml (95% CI) Determinants of coronary artery disease (subjects in group I and group III) Postprandial insulin, P OR per increase by 1 mIU / ml (95% CI)
0.02 1.05 (1.01–1.09)
Individuals with metabolic syndrome and age-matched controls Determinants of coronary artery disease (subjects selected from group I and group II) Postprandial insulin, P OR per increase by 1 mIU / ml (95% CI)
0.003 1.01 (1.005–1.02)
Determinants of coronary artery disease (subjects selected from group I and group III) Fasting glucose, P OR per increase in mg / dl Postprandial insulin, P OR per increase by 1 mIU / ml (95% CI)
0.04 1.07 (1.003–1.14) 0.0001 1.05 (1.02–1.09)
Determinants of metabolic syndrome (subjects selected from group II and group III) Fasting glucose, P OR per increase in mg / dl Postprandial insulin, P OR per increase by 1 IU / ml (95% CI)
0.007 1.07 (1.02–1.13) 0.005 1.03 (1.01–1.06)
0.002 1.01 (1.004–1.02)
CI, confidence interval; OR, odds ratio.
only postprandial insulin level differed between group I and group II.
3.3. Logistic regression analysis We performed a reverse stepwise logistic regression analyses on non-diabetic women in groups I and II. A separate analysis was performed in groups I and III (Table 3). Variables included in the model were traditional risk factors, fasting glucose, postprandial glucose, fasting insulin and postprandial insulin levels. Hyperinsulinemia was associated independently with coronary artery disease in both analyses. A subgroup analysis was performed on the cases with metabolic syndrome. The analysis on the subjects in groups I and II identified postprandial hyperinsulinemia as the independent determinant of coronary artery disease. The analysis including patients in group II and group III revealed that high fasting glucose and high postprandial insulin levels were associated with metabolic syndrome.
4. Discussion The risk factor profile of our cohort showed that patients had clustering of the components of insulin
resistance syndrome. In non-diabetic women, postprandial hyperinsulinemia was associated with coronary artery disease; fasting glucose level and postprandial hyperinsulinemia were associated with metabolic syndrome. It has been suggested that insulin resistance syndrome is a typical basis for vascular disease in women [5]. This composite is highly associated with coronary artery disease, and subjects without known coronary artery disease and those with the disease suffer a significant increase in mortality when this syndrome is expressed [2–4]. Our data show, that even among non-diabetic women with coronary artery disease, other components of the ‘deadly quartet’ are present. We have found an independent association of postprandial hyperinsulinemia with coronary artery disease in logistic regression analysis in non-diabetic women. Also, in women with metabolic syndrome postprandial hyperinsulinemia was associated with coronary artery disease. The mechanism for this association is complex. However, we believe it can be partly explained by genetic and ethnic variations. Ethnic background modifies the relationship between hyperinsulinemia and coronary artery disease with borderline significance [9]. Coronary artery disease can be associated with
220
M. Baltali et al. / International Journal of Cardiology 88 (2003) 215–221
insulin resistance itself [13]. In vitro studies at the tissue level showed that insulin itself has direct vascular actions that contributes to both vascular protection and vascular injury, depending on the cell type. Insulin resistance on the other hand disturbs the balance between vasoprotective effects of insulin, mediated by nitric oxide, and its atherogenic effects, involving vascular smooth muscle cell growth, migration and plasminogen activator inhibitor-1 production [14–17]. An association between insulin resistance and coronary artery disease independent of insulin levels was also reported [18]. On the other hand, in one recent study, the insulin resistance factor did not predict coronary artery disease events in aged nondiabetic women with the frequent presence of hypertension, hyperlipidemia and increased postprandial insulin level [19]. Taking together, it seems that insulin resistance might play a role in the development of coronary artery disease. We suggest that severe postprandial hyperinsulinemia among non-diabetic women with the components of metabolic syndrome can act as a surrogate marker of insulin resistance.
4.1. Study limitations A major limitation of our study is that this is a case control study on a limited number of patients. A larger study may reveal more conclusive data. Another limitation is that the real control group (group III) had 52 patients instead of 104 patients as in group I and group II. Although case control studies are powerful tools for highlighting differences, their cross-sectional natures do not allow definition of causal relations. Due to the hospital-based nature of this study, the true risk factors are not reflected. 5. Conclusion In women with clustering of factors of metabolic syndrome other than diabetes mellitus, severe postprandial hyperinsulinemia was associated with CAD. If hyperinsulinemia acts as a surrogate marker of insulin resistance, this needs to be elucidated. Acknowledgements The authors would like to acknowledge the help of
the following physicians in providing data of the patients involved in this study: Dr Fatıh Yalc¸in, Dr Semra Topc¸u, Dr Senol Demircan, Dr Fatma Yigit, ¨ ¨ Bas¸kent UniversiDr Pelin Deliaga, Dr Goksel Bork, ty, Adana Medical Center. We would also like to ¨ thank Meltem Kirisci, Mine Gunal, and C ¸ igdem ¨ ¨ for their skillful assistance in data collection. Tuylu
References [1] Grundy SM. Hypertriglyceridemia, atherogenic dyslipidemia, and the metabolic syndrome. Am J Cardiol 1998;81(suppl):18B–25B. [2] Genest Jr. J, Cohn JS. Clustering of cardiovascular risk factors: targeting high-risk individuals. Am J Cardiol 1995;76:8A–20A. [3] Yusuf HR, Giles WH, Croft JB, Anda RF, Casper ML. Impact of multiple risk factor profiles on determining cardiovascular disease risk. Prev Med 1998;27:1–9. [4] Sprecher DL, Pearce GL. How deadly is the ‘deadly quartet’? A post-CABG review. J Am Coll Cardiol 2000;36:1159–65. [5] Jeppesen J, Hollenbeck CB, Zhou MY, Coulston AM, Jones C, Chen Y-DI et al. Relation between insulin resistance, hyperinsulinemia, postheparin plasma lipoprotein lipase activity, and postprandial lipemia. Arterioscler Thromb Vasc Biol 1995;15:320–4. [6] Eschwege E, Richard JL, Thibult N, Ducimetiere P, Warnet JM, Claude JR et al. Coronary heart disease mortality in relation with diabetes, blood glucose and plasma insulin levels: the Paris Prospective Study, 10 years later. Horm Metab Res Suppl 1985;15:41–6. ´ J-P, Lamarche B, Mauriege ` [7] Despres P, Cantini B, Dagenais DR, Moorjani S et al. Hyperinsulinemia as an independent risk factor for ischemic heart disease. N Engl J Med 1996;334:952–7. ¨ ¨ M, Miettinen H, Laakso M, Pyorala ¨ ¨ K. Hyperinsulinemia [8] Pyorala predicts coronary heart disease risk in healthy middle-aged men: the 22-year follow-up results of the Helsinki Policeman Study. Circulation 1998;98:398–404. [9] Ruige JB, Assendelft WJ, Dekker JM, Kostense PJ, Heine RJ, Bouter LM. Insulin and risk of cardiovascular disease: a metaanalysis. Circulation 1998;97:996–1001. [10] Ferrara A, Barrett-Connor EL, Edelstein ASL. Hyperinsulinemia does not increase the risk of fatal cardiovascular disease in elderly men or women without diabetes: the Rancho Bernardo Study, 1984–1991. Am J Epidemiol 1994;140:857–69. [11] Welborn TA, Wearne K. Coronary heart disease incidence and cardiovascular mortality in Busselton with reference to glucose and insulin concentrations. Diabetes Care 1979;2:154–60. [12] Executive Summary of the Third Report of the National Cholesterol Education Program Expert Panel on Detection Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III). J Am Med Assoc 2001;285:2486–97. [13] Stern MP. Diabetes and cardiovascular disease. The ‘common soil’ hypothesis. Diabetes 1995;44:369–74. [14] Wu HY, Jeng YY, Yue CJ, Chyu KY, Hsueh WA, Chan TM. Endothelial-dependent vascular effects of insulin and insulin-like growth factor I in the perfused rat mesenteric artery and aortic ring. Diabetes 1994;43:1027–32. [15] Banskota NK, Taub R, Zellner K, King GL. Insulin, insulin-like growth factor I and platelet-derived growth factor interact additively in the induction of the protooncogene c-myc and cellular proliferation in cultured bovine aortic smooth muscle cells. Mol Endocrinol 1989;3:1183–90.
M. Baltali et al. / International Journal of Cardiology 88 (2003) 215–221 [16] Laakso M, Edelman SV, Brechtel G, Baron AD. Decreased effect of insulin to stimulate sceletal muscle blood flow in obese man: a novel mechanism for insulin resistance. J Clin Invest 1990;85:1844–52. [17] Hsueh WA, Law RE. Cardiovascular risk continuum: implications of insulin resistance and diabetes. Am J Med 1998;105(1A):4S–14S. [18] Revers M, D’Agostino Jr. R, Burke G. Coronary artery disease is
221
associated with low insulin sensitivity independent of insulin levels and cardiovascular risk factors. Diabetes 1996;45(suppl 2):52A, (abstract). ¨ ¨ ¨ ¨ ¨ K, Laakso M, Kuusisto J. [19] Lempiainen P, Mykkanen L, Pyorala Insulin resistance syndrome predicts coronary heart disease events in elderly non-diabetic men. Circulation 1999;100:123–8.
Association between postprandial hyperinsulinemia and coronary artery disease among non-diabetic women: a case control study Mehmet Baltali*, Mehmet E. Korkmaz, H. Tarik Kiziltan, I. Haldun Muderris, Bulent Ozin, Ruksan Anarat Cardiology, Cardiovascular Surgery, and Biochemistry, Faculty of Medicine, Bas¸kent University, Adana, Turkey Received 29 October 2001; received in revised form 1 July 2002; accepted 20 July 2002
Abstract Background: We planned a case-control study to assess the relation of fasting glucose, fasting insulin, postprandial glucose and postprandial insulin levels with coronary artery disease in nondiabetic women. Methods: Among 968 consecutive nondiabetic women screened, 104 with coronary artery disease (mean age 60, 469) made up the study cohort (group I). One-hundred and four age-matched, nondiabetic women without coronary artery disease who had a similar lipid and blood pressure profile (group II), and 52 healthy, age-matched women served as controls (group III, real control group). Demographics, waist circumference, lipids, fasting glucose postprandial glucose, fasting and postprandial insulin levels were compared among the groups. A separate subgroup analysis were performed in patients with metabolic syndrome. Results: No differences were identified in terms of prevalences of risk factors between group I and group II. Women with coronary artery disease had higher postprandial insulin level than the women in group II and group III. In reverse stepwise logistic regression analysis postprandial hyperinsulinemia was found to be the single independent determinant for coronary artery disease for the entire study group as well as for women with metabolic syndrome. Conclusion: Our data demonstrate that postprandial hyperinsulinemia is independently associated with coronary artery disease, irrespective of fasting glucose, postprandial glucose, and fasting insulin levels in nondiabetic women with clusterings of factors of metabolic syndrome. 2002 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Coronary artery disease; Insulin resistance; Metabolic syndrome
1. Introduction Insulin resistance syndrome consists of a cluster of factors including visceral obesity, hyperinsulinemia, impaired glucose tolerance or diabetes mellitus, hypertriglyceridemia, and hypertension in a single patient [1]. This grouping of well-known metabolic factors, also called the ‘deadly quartet’, is associated with coronary artery disease [2,3]. When this syndrome is expressed, female patients suffer a substantial in*Corresponding author. E-mail address: [email protected] (M. Baltali).
crease in mortality [4]. The association between insulin resistance syndrome and coronary artery disease is complex, and pathways by which the insulin resistant state adversely affects coronary risk factors and the risk for coronary artery disease need to be elucidated. Postprandial lipemia may be a mechanical link between insulin resistance and coronary artery disease [5]. Elevated insulin levels have a negative impact on both risk factor profile and atherosclerotic process [6–8]. A meta-analysis revealed that hyperinsulinemia was a weak but positive indicator of cardiovascular events in middle-aged men [9]. Studies examining insulin level and cor-
0167-5273 / 02 / $ – see front matter 2002 Elsevier Science Ireland Ltd. All rights reserved. doi:10.1016/S0167-5273(02)00399-6
216
M. Baltali et al. / International Journal of Cardiology 88 (2003) 215–221
onary artery disease events in women did not show any association [10,11]. We planned a case-control study to assess the relation of fasting glucose, postprandial glucose, fasting insulin and postprandial insulin levels with presence of coronary artery disease among non-diabetic elderly women in Adana, Turkey.
2. Methods
Individuals with heart failure, malignancy, chronic renal failure, stroke history, thyroid disease, hemorrhagic disorders, rheumatic valve disease, diabetes, smoking history, and hepatic failure were excluded. Patients taking hormone replacement therapy, and medications thought to alter insulin and glucose profile, such as adrenergic beta-blocker, were also excluded. All patients gave informed consent and study protocol was approved by the Bas¸kent University Internal Review Board.
2.1. Study population
2.2. Study design
The study population consisted of 104 non-diabetic women with coronary artery disease (group I), 104 age-matched non-diabetic women without coronary artery disease who had metabolic risk factors including hypertension, obesity, hypertriglyceridemia, and low high-density lipoprotein cholesterol level (group II), and 52 age-matched healthy subjects with no metabolic risk factors (group III). Subjects in the group I and group II were selected from a total of 968 consecutive non-diabetic women evaluated between September 1999 and December 2001 in the outpatient clinic of Bas¸kent University, Adana Medical Center. Group I consisted of patients with angiographic evidence of coronary artery disease, a history of coronary bypass surgery or balloon angioplasty. For every patient in group I, a subject of the same age, with similar body mass index, smoking status, history of hypertension and similar baseline metabolic lipid profile who did not have coronary artery disease was identified from the computer database, and these subjects constituted group II. Subjects in group III (real control group) were selected from individuals in the check-up clinic. These were healthy non-diabetic women who did not have metabolic risk factors including hypertension, obesity, hypertriglyceridemia and low high-density lipoprotein cholesterol levels. One age-matched control (n552) per two cases in group 1 was selected from the database in the check-up clinic. Patients with metabolic syndrome in group I (n5 64), age-matched individuals with metabolic syndrome in group II (n564), and age-matched subjects without metabolic syndrome in group III (n532) were selected and evaluated in a subgroup analysis.
Age, menopausal status, smoking habits, medication, history of coronary artery disease, hypertension, diabetes mellitus, and heredity were recorded for all patients. Physical examination and blood pressure measurements were obtained. Weight, height, waist and hip circumference were measured for each subject; body mass index, obesity, and visceral obesity were determined from these parameters. Waist circumference was measured at the narrowest diameter between costal margin and the iliac crest, and hip circumference was measured at the greatest diameter over the glutei. Bloods samples were drawn after 12-h fasting for the measurement of total cholesterol, HDL-cholesterol, triglyceride, fasting glucose and insulin levels. Plasma glucose and insulin responses were evaluated 120 min after oral (75 g) glucose load. All subjects underwent treadmill exercise test or thallium myocardial perfusion scintigraphy. Patients with ischemic results in non-invasive tests were offered coronary angiograms. Cases with a history of balloon angioplasty or coronary artery bypass grafting, or .70% luminal narrowing in at least one major coronary artery were defined as those having coronary artery disease (n562) and were compared with the control groups.
2.3. Definitions Following definitions were used: diabetes mellitus, fasting glucose .126 mg / dl, a glucose level .200 mg / dl 120 min after oral glucose load, or usage of antidiabetic drugs; obesity, body mass index .30 kg / m 2 ; hypertension, blood pressure .140 / 90 mmHg on two separate examinations, or usage of
M. Baltali et al. / International Journal of Cardiology 88 (2003) 215–221
antihypertensive agents; metabolic dyslipidemia, a fasting triglyceride level .200 mg / dl and / or a fasting high-density lipoprotein cholesterol level ,40 mg / dl. Metabolic syndrome was defined as having at least three of the following conditions: triglyceride level .150 mg / dl; high-density lipoprotein cholesterol level ,50 mg / dl in women and ,40 mg / dl in men; waist circumference .88 cm in women and .102 cm in men; systolic blood pressure .130 / 85 mmHg or usage of antihypertensive drugs; fasting glucose level .110 mg / dl [12].
2.4. Laboratory measurements Total cholesterol, high density lipoprotein cholesterol, triglyceride, and blood glucose measurements were estimated on an Hitachi 912 autoanalyser (Hitachi, Tokyo, Japan) with standard kits (Boehringer Mannheim). Total cholesterol was measured using CHOD-PAP, glucose using GOD-PAP, and triglycerides using GPO-PAP techniques. High-density lipoprotein cholesterol was estimated by quantitative enzymatic method. Low density lipoprotein cholesterol was calculated using the Friedewald formula. Fasting and postprandial insulin were analyzed on an AxSYM System-Abbott analyzer (Abbott Diagnostics, Park, IL, USA) with 67-5857 / R2 kits using micro-particle enzyme immune assay technique. This assay is highly specific for insulin with essentially no cross-reactivity with proinsulin (,0.005%), C-peptide or 32–33 pro peptides.
2.5. Statistical analysis Data were collated and analyzed with SPSS computer program. The numeric variables were expressed as the mean6S.D., the categorical variables as percentage. The relationship between risk factors among the groups were determined using Student’s t-test for numeric variables and chi-square test for categorical variables. Values for triglyceride levels, fasting and postprandial insulin levels not conforming to a normal distribution were log transformed for comparison and were presented as median with 25th and 75th percentiles. Separate analyses were performed on cases (group I) and controls (group II and group III), and on the
217
subgroup of cases with metabolic syndrome [12]. Reverse stepwise logistic regression analysis was performed to determine the independent association of fasting glucose, fasting insulin, postprandial glucose and postprandial insulin levels with coronary artery disease and metabolic syndrome. A P-value ,0.05 was considered statistically significant.
3. Results
3.1. Clinical and metabolic characteristics of cases As shown in Table 1, 16.3% of the subjects in group I had previous balloon angioplasty, 21.2% had previous coronary bypass surgery, and 71.2% had arteriographically diagnosed coronary artery disease. Patients in group I and group II had similar risk factor profiles. Subjects in group III consisted of healthy patients with a similar age profile, who did not have obesity, hypertriglyceridemia, or low highdensity lipoprotein cholesterol levels. Of individuals in group I and group II, 81% were hypertensive, while incidences for obesity and metabolic dyslipidemia were 52.9 and 35.6%, respectively.
3.2. Comparison of glucose and insulin levels Table 1 compares fasting-postprandial glucose and insulin levels of group I, controls with multiple risk factors, and real controls. Patients in group I had higher postprandial insulin levels than subjects in both control groups (group II and group III). Fasting insulin level did not differ between group I and group II, however patients in these groups had higher fasting insulin levels than those in group III. The difference between group II and group III was not statistically significant. Patients in group II had higher fasting glucose levels than those in group III. Postprandial glucose levels did not differ significantly among the groups. We also compared coronary risk factor profiles, fasting and postprandial glucose and insulin levels in individuals with metabolic syndrome in a subgroup analysis (Table 2). Subjects in group I and group II had higher fasting glucose, fasting insulin and postprandial insulin levels than those in group III, and
218
M. Baltali et al. / International Journal of Cardiology 88 (2003) 215–221
Table 1 Characteristics, risk factor profile, coronary artery disease diagnosis and use of medication
Age (years) Hypertension, n (%) Body mass index (kg / m 2 ) Obesity Waist circumference (cm) Smoking, n (%) Triglyceride (mg / dl)a Total cholesterol (mg / dl) LDL-cholesterol (mg / dl) (n5103) HDL-cholesterol (mg / dl) Metabolic dyslipidemia (%) Fasting insulin a Postprandial insulin a Fasting blood glucose (mg / dl) Postprandial blood glucose (mg / dl)
Group I (n5104)
Group II (n5104)
Group III n552
60.469 84.0 (80.8)* 30.164* 55 (52.9) 99.1610* 17 (16.3) 135.5 (114–173)* 217.4642 138.6637 48.4613* 37 (35.6)* 11.0 (8–14)* 88.1 (63.5–119.9)* ‡ 97.2611 136.1633
60.469 84.0 (80.8)* 30.564* 55 (52.9) 99.869* 17 (16.3) 129.0 (104–179.5)* 214.7639 134.7635 48.9613* 37 (35.6)* 10.0 (7.5–13)* 62.1 (39.3–97.6) 97.6610* 136.6634
59.168 – 26.362 – 87.169 3 (6) 107.5 (82.3–133) 207.1635 127.0633 57.8614 – 8.0 (6–10) 48.5 (38.4–60.8) 93.8612 129.1630
Coronary artery disease diagnosis Arteriography, n (%) CABG history, n (%) PTCA history, n (%)
74 (71.2) 22 (21.2) 17 (16.3)
Use of medication Beta blocker, n (%)b Ca-channel antagonist, n (%) Diuretics, n (%) Nitrate, n (%) Aspirin, n (%)
54 36 10 51 73
(51.9)* ‡ (34.6)* (9.6)* (49.0)* ‡ (70.2)* ‡
– – – 14 41 14 8 17
– – – (13.5)* (39.4)* (13.5)* (7.6)* (16.3)
– – – – 2 (1.9)
HDL, high-density lipoprotein; LDL, low-density lipoprotein. *P,0.05 versus group III. ‡ P,0.05 versus group II. a Median (25th–75th percentile). b Except beta-blockers with alpha-adrenergic activity.
Table 2 Characteristics and risk factor profiles of coronary artery disease cases with metabolic syndrome (group I), age-matched subjects without coronary artery disease and with metabolic syndrome (group II), and age-matched healthy controls without metabolic syndrome (group III)
Age (years) Hypertension, n (%) Body mass index (kg / m 2 ) Waist circumference (cm) Smoking, n (%) Triglyceride (mg / dl)a Total cholesterol (mg / dl) LDL-cholesterol (mg / dl) HDL-cholesterol (mg / dl) Fasting insulin a Postprandial insulin a Fasting blood glucose (mg / dl) Postprandial blood glucose (mg / dl)
Group I (n564)
Group II (n564)
Group III (n532)
58.669 59 (92.2%)* 30.864* 102.568* 10 (15.6%) 157.0 (126.8–198.5) 221.5643 144.8637* 41.869* 12.0 (9.1–14.9)* 99.3 (72.4–124.8)* ‡ 99.669* 139.0634
58.969 57 (89.1%)* 30.964* 101.367* 13 (20.3%) 166.5 (128.8–212.8)* 214.8641 135.0637 42.6610* 10.0 (8.6–14.2)* 66.2 (50.4–104.6)* 99.5611* 144.1634*
58.368 – 25.963 82.866 3 (9.4%) 94.0 (69.3–122.8) 206.3638 127.0633 60.7611 8.0 (6.0–10.4) 46.0 (38.5–61.2) 91.1610 125.8627
HDL, high-density lipoprotein; LDL, low-density lipoprotein. *P,0.05 versus group III. ‡ P,0.05 versus group II. a Median (25th–75th percentile).
M. Baltali et al. / International Journal of Cardiology 88 (2003) 215–221
219
Table 3 Logistic regression analysis Entire study group Determinants of coronary artery disease (subjects in group I and group II) Postprandial insulin, P OR per increase by 1 mIU / ml (95% CI) Determinants of coronary artery disease (subjects in group I and group III) Postprandial insulin, P OR per increase by 1 mIU / ml (95% CI)
0.02 1.05 (1.01–1.09)
Individuals with metabolic syndrome and age-matched controls Determinants of coronary artery disease (subjects selected from group I and group II) Postprandial insulin, P OR per increase by 1 mIU / ml (95% CI)
0.003 1.01 (1.005–1.02)
Determinants of coronary artery disease (subjects selected from group I and group III) Fasting glucose, P OR per increase in mg / dl Postprandial insulin, P OR per increase by 1 mIU / ml (95% CI)
0.04 1.07 (1.003–1.14) 0.0001 1.05 (1.02–1.09)
Determinants of metabolic syndrome (subjects selected from group II and group III) Fasting glucose, P OR per increase in mg / dl Postprandial insulin, P OR per increase by 1 IU / ml (95% CI)
0.007 1.07 (1.02–1.13) 0.005 1.03 (1.01–1.06)
0.002 1.01 (1.004–1.02)
CI, confidence interval; OR, odds ratio.
only postprandial insulin level differed between group I and group II.
3.3. Logistic regression analysis We performed a reverse stepwise logistic regression analyses on non-diabetic women in groups I and II. A separate analysis was performed in groups I and III (Table 3). Variables included in the model were traditional risk factors, fasting glucose, postprandial glucose, fasting insulin and postprandial insulin levels. Hyperinsulinemia was associated independently with coronary artery disease in both analyses. A subgroup analysis was performed on the cases with metabolic syndrome. The analysis on the subjects in groups I and II identified postprandial hyperinsulinemia as the independent determinant of coronary artery disease. The analysis including patients in group II and group III revealed that high fasting glucose and high postprandial insulin levels were associated with metabolic syndrome.
4. Discussion The risk factor profile of our cohort showed that patients had clustering of the components of insulin
resistance syndrome. In non-diabetic women, postprandial hyperinsulinemia was associated with coronary artery disease; fasting glucose level and postprandial hyperinsulinemia were associated with metabolic syndrome. It has been suggested that insulin resistance syndrome is a typical basis for vascular disease in women [5]. This composite is highly associated with coronary artery disease, and subjects without known coronary artery disease and those with the disease suffer a significant increase in mortality when this syndrome is expressed [2–4]. Our data show, that even among non-diabetic women with coronary artery disease, other components of the ‘deadly quartet’ are present. We have found an independent association of postprandial hyperinsulinemia with coronary artery disease in logistic regression analysis in non-diabetic women. Also, in women with metabolic syndrome postprandial hyperinsulinemia was associated with coronary artery disease. The mechanism for this association is complex. However, we believe it can be partly explained by genetic and ethnic variations. Ethnic background modifies the relationship between hyperinsulinemia and coronary artery disease with borderline significance [9]. Coronary artery disease can be associated with
220
M. Baltali et al. / International Journal of Cardiology 88 (2003) 215–221
insulin resistance itself [13]. In vitro studies at the tissue level showed that insulin itself has direct vascular actions that contributes to both vascular protection and vascular injury, depending on the cell type. Insulin resistance on the other hand disturbs the balance between vasoprotective effects of insulin, mediated by nitric oxide, and its atherogenic effects, involving vascular smooth muscle cell growth, migration and plasminogen activator inhibitor-1 production [14–17]. An association between insulin resistance and coronary artery disease independent of insulin levels was also reported [18]. On the other hand, in one recent study, the insulin resistance factor did not predict coronary artery disease events in aged nondiabetic women with the frequent presence of hypertension, hyperlipidemia and increased postprandial insulin level [19]. Taking together, it seems that insulin resistance might play a role in the development of coronary artery disease. We suggest that severe postprandial hyperinsulinemia among non-diabetic women with the components of metabolic syndrome can act as a surrogate marker of insulin resistance.
4.1. Study limitations A major limitation of our study is that this is a case control study on a limited number of patients. A larger study may reveal more conclusive data. Another limitation is that the real control group (group III) had 52 patients instead of 104 patients as in group I and group II. Although case control studies are powerful tools for highlighting differences, their cross-sectional natures do not allow definition of causal relations. Due to the hospital-based nature of this study, the true risk factors are not reflected. 5. Conclusion In women with clustering of factors of metabolic syndrome other than diabetes mellitus, severe postprandial hyperinsulinemia was associated with CAD. If hyperinsulinemia acts as a surrogate marker of insulin resistance, this needs to be elucidated. Acknowledgements The authors would like to acknowledge the help of
the following physicians in providing data of the patients involved in this study: Dr Fatıh Yalc¸in, Dr Semra Topc¸u, Dr Senol Demircan, Dr Fatma Yigit, ¨ ¨ Bas¸kent UniversiDr Pelin Deliaga, Dr Goksel Bork, ty, Adana Medical Center. We would also like to ¨ thank Meltem Kirisci, Mine Gunal, and C ¸ igdem ¨ ¨ for their skillful assistance in data collection. Tuylu
References [1] Grundy SM. Hypertriglyceridemia, atherogenic dyslipidemia, and the metabolic syndrome. Am J Cardiol 1998;81(suppl):18B–25B. [2] Genest Jr. J, Cohn JS. Clustering of cardiovascular risk factors: targeting high-risk individuals. Am J Cardiol 1995;76:8A–20A. [3] Yusuf HR, Giles WH, Croft JB, Anda RF, Casper ML. Impact of multiple risk factor profiles on determining cardiovascular disease risk. Prev Med 1998;27:1–9. [4] Sprecher DL, Pearce GL. How deadly is the ‘deadly quartet’? A post-CABG review. J Am Coll Cardiol 2000;36:1159–65. [5] Jeppesen J, Hollenbeck CB, Zhou MY, Coulston AM, Jones C, Chen Y-DI et al. Relation between insulin resistance, hyperinsulinemia, postheparin plasma lipoprotein lipase activity, and postprandial lipemia. Arterioscler Thromb Vasc Biol 1995;15:320–4. [6] Eschwege E, Richard JL, Thibult N, Ducimetiere P, Warnet JM, Claude JR et al. Coronary heart disease mortality in relation with diabetes, blood glucose and plasma insulin levels: the Paris Prospective Study, 10 years later. Horm Metab Res Suppl 1985;15:41–6. ´ J-P, Lamarche B, Mauriege ` [7] Despres P, Cantini B, Dagenais DR, Moorjani S et al. Hyperinsulinemia as an independent risk factor for ischemic heart disease. N Engl J Med 1996;334:952–7. ¨ ¨ M, Miettinen H, Laakso M, Pyorala ¨ ¨ K. Hyperinsulinemia [8] Pyorala predicts coronary heart disease risk in healthy middle-aged men: the 22-year follow-up results of the Helsinki Policeman Study. Circulation 1998;98:398–404. [9] Ruige JB, Assendelft WJ, Dekker JM, Kostense PJ, Heine RJ, Bouter LM. Insulin and risk of cardiovascular disease: a metaanalysis. Circulation 1998;97:996–1001. [10] Ferrara A, Barrett-Connor EL, Edelstein ASL. Hyperinsulinemia does not increase the risk of fatal cardiovascular disease in elderly men or women without diabetes: the Rancho Bernardo Study, 1984–1991. Am J Epidemiol 1994;140:857–69. [11] Welborn TA, Wearne K. Coronary heart disease incidence and cardiovascular mortality in Busselton with reference to glucose and insulin concentrations. Diabetes Care 1979;2:154–60. [12] Executive Summary of the Third Report of the National Cholesterol Education Program Expert Panel on Detection Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III). J Am Med Assoc 2001;285:2486–97. [13] Stern MP. Diabetes and cardiovascular disease. The ‘common soil’ hypothesis. Diabetes 1995;44:369–74. [14] Wu HY, Jeng YY, Yue CJ, Chyu KY, Hsueh WA, Chan TM. Endothelial-dependent vascular effects of insulin and insulin-like growth factor I in the perfused rat mesenteric artery and aortic ring. Diabetes 1994;43:1027–32. [15] Banskota NK, Taub R, Zellner K, King GL. Insulin, insulin-like growth factor I and platelet-derived growth factor interact additively in the induction of the protooncogene c-myc and cellular proliferation in cultured bovine aortic smooth muscle cells. Mol Endocrinol 1989;3:1183–90.
M. Baltali et al. / International Journal of Cardiology 88 (2003) 215–221 [16] Laakso M, Edelman SV, Brechtel G, Baron AD. Decreased effect of insulin to stimulate sceletal muscle blood flow in obese man: a novel mechanism for insulin resistance. J Clin Invest 1990;85:1844–52. [17] Hsueh WA, Law RE. Cardiovascular risk continuum: implications of insulin resistance and diabetes. Am J Med 1998;105(1A):4S–14S. [18] Revers M, D’Agostino Jr. R, Burke G. Coronary artery disease is
221
associated with low insulin sensitivity independent of insulin levels and cardiovascular risk factors. Diabetes 1996;45(suppl 2):52A, (abstract). ¨ ¨ ¨ ¨ ¨ K, Laakso M, Kuusisto J. [19] Lempiainen P, Mykkanen L, Pyorala Insulin resistance syndrome predicts coronary heart disease events in elderly non-diabetic men. Circulation 1999;100:123–8.