- Email: [email protected]
Insulin resistance and carotid atherosclerosis in patients with type 2 diabetes
Atherosclerosis 205 (2009) 309–313
Contents lists available at ScienceDirect
Atherosclerosis journal homepage: www.elsevier.com/locate/atherosclerosis
Insulin resistance and carotid atherosclerosis in patients with type 2 diabetes夽 Seok Won Park a,∗ , Soo Kyung Kim a , Yong-Wook Cho a , Dae Jung Kim b , Young-Duk Song c , Young Ju Choi d , Byung Wook Huh d , Sung Hee Choi e , Sun Ha Jee f , Mi Ae Cho g , Eun Jig Lee g , Kap Bum Huh d a
Department of Internal Medicine, Pochon CHA University, 351 Yatapdong, Sungnam 463-712, Republic of Korea Department of Endocrinology and Metabolism, Ajou University School of Medicine, Suwon, Republic of Korea Department of Medicine, Ilsan Hospital, Goyang, Republic of Korea d Huh Diabetes Center and the 21st Century Diabetes and Vascular Research Institute, Seoul, Republic of Korea e Department of Internal Medicine, Seoul National University, Seoul, Republic of Korea f Department of Epidemiology, Graduate School of Public Health, Yonsei University, Seoul, Republic of Korea g Department of Medicine, Yonsei University, Seoul, Republic of Korea b c
a r t i c l e
i n f o
Article history: Received 20 May 2008 Received in revised form 20 November 2008 Accepted 4 December 2008 Available online 13 December 2008 Keywords: Insulin resistance Hyperinsulinemia Carotid atherosclerosis Type 2 diabetes
a b s t r a c t Objective: Insulin resistance may provide a crucial link between type 2 diabetes and cardiovascular disease. However, it is still unclear whether insulin resistance itself or hyperinsulinemia is independently associated with subclinical atherosclerosis. We hypothesized that insulin resistance, but not hyperinsulinemia, would be associated with carotid atherosclerosis in patients with type 2 diabetes. Methods: We examined 2471 patients with type 2 diabetes, consecutively enrolled in Huh Diabetes Center. Insulin sensitivity was directly assessed by a rate constant for plasma glucose disappearance (Kitt) using short insulin tolerance test. Fasting insulin levels were used as a marker of hyperinsulinemia. Both carotid arteries were examined by B-mode ultrasound. Carotid atherosclerosis was defined by having a clearly isolated focal plaque or mean carotid intima-media thickness (IMT) ≥1.1 mm. Results: In multiple regression models, insulin sensitivity index (Kitt) but not hyperinsulinemia was significantly associated with carotid IMT adjusting for known risk factors such as age, sex, BMI, smoking, systolic pressure, HDL and LDL cholesterol. One standard deviation decrease in Kitt was associated with 0.046 mm increase in carotid IMT (p = 0.015). Furthermore, odds ratio for carotid atherosclerosis was 1.43 (95% CI: 1.10, 1.86) in type 2 diabetic patients with insulin resistance (lowest quartile of insulin sensitivity) adjusting for known risk factors. The results were consistent in all subgroups stratified by sex, age, smoking and hypertension. Conclusion: Insulin resistance measured by short insulin tolerance test, but not hyperinsulinemia, is independently associated with carotid atherosclerosis in patients with type 2 diabetes. © 2008 Elsevier Ireland Ltd. All rights reserved.
1. Introduction Insulin resistance (IR) plays a major role for the development of type 2 diabetes [1,2]. IR is also associated with a variety of cardiovascular disease (CVD) risk factors including central obesity, essential hypertension, hypertriglyceridemia, and low HDL cholesterol [3]. The concept of metabolic syndrome comes from the observation of clustering of risk factors within a single individual [4,5]. It has been hypothesized that IR and/or its consequent hyperinsulinemia might provide an underlying pathophysiological mechanism for the
夽 This study was supported by Seoul City R&BD Program (10526), Seoul, Republic of Korea. ∗ Corresponding author. Tel.: +82 31 780 5211; fax: +82 31 780 5208. E-mail address: [email protected] (S.W. Park). 0021-9150/$ – see front matter © 2008 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.atherosclerosis.2008.12.006
clustering of risk factors although ironically neither IR nor hyperinsulinemia is included as a component to define metabolic syndrome [6–8]. CVD is a leading cause of death among individuals with type 2 diabetes [9]. Again, IR has been proposed to play a crucial role in the link between type 2 diabetes and CVD [10–12]. The associations between IR and CVD or its surrogate marker such as intima-media thickness (IMT) have been documented mainly in subjects with varying degrees of glucose tolerance. However, it is still unclear whether IR itself or hyperinsulinemia confers an independent association with IMT beyond known risk factors, particularly in patients with type 2 diabetes. The aim of the study was to investigate whether insulin resistance directly measured by short insulin tolerance test or hyperinsulinemia estimated by fasting insulin level has independent association with carotid atherosclerosis in a large cohort of patients with type 2 diabetes.
310
S.W. Park et al. / Atherosclerosis 205 (2009) 309–313
2. Methods 2.1. Study population We have enrolled 3704 patients with type 2 diabetes, consecutively from January 2003 to June 2006 at the Huh Diabetes Center in Seoul, Korea as a part of Seoul metabolic syndrome cohort. Selection criteria for the study were as follows: age at diagnosis ≥20 years, current age <70 years, BMI ≥18.5 and <40 kg/m2 , no need of insulin therapy for the first 2 years of disease, no history of ketoacidosis, no present glucocorticoid therapy. Among 3220 subjects who met the above criteria, short insulin tolerance test (SITT) was performed in 3030 patients (94.1%) and carotid intima-media thickness was measured in 2946 patients (91.5%). We further excluded patients who were treated with thiazolidinediones (n, 263) because it has been reported to affect insulin resistance as well as IMT. Finally 2471 subjects who had been assessed for both SITT and IMT were included in the study. Clinical characteristics of participants were summarized in Table 1. The protocol was approved by the Ethics Committee of the Yonsei University. Informed consent was obtained from all subjects. All the investigations were performed in accordance with the principles of the Declaration of Helsinki.
2.2. Assessment of insulin sensitivity and hyperinsulinemia Insulin sensitivity was directly assessed by SITT as a rate constant for plasma glucose disappearance (Kitt: %/min) [13–16]. The subjects were attended to the clinic at 08:00 after overnight fasting. Two intravenous cannulas were established, one for the blood sampling and the other one for insulin injection. A heating pad was applied to hand vein for warming to get an arterialized venous sample. After 15 min resting, pre-diluted regular insulin (Humulin® , Eli Lilly, Indianapolis, IN) at a dose of 0.1 U/kg was injected through antecubital vein and blood samples were drawn from contralateral
hand vein through retrograde cannula at 0, 3, 6, 9, 12 and 15 min after insulin injection. Plasma glucose concentrations were determined immediately after sampling using Beckman glucose analyzer II (Beckman Ins., Fullerton, CA) and then the rate constant for the plasma glucose disappearance (Kitt) was calculated from the slope of the fall in log transformed plasma glucose between 3 and 15 min [13]. The Kitt value was used as an index of in vivo insulin sensitivity. The subjects were classified by the quartiles of Kitt; quartile 1 (those with the lowest quartile of Kitt) was the most insulin resistance group whereas quartile 4 was the most sensitive to insulin (Table 1). The intra-participant coefficient of variation for Kitt was 8.8 ± 2.0% [16]. We routinely infused 100 ml of 20% dextrose solution immediate after the test to avoid any potential hypoglycemia. We used fasting insulin concentrations as a marker of hyperinsulinemia. Fasting insulin concentrations were determined by means of immunoradiometric assay using INS-IRMA Kit (BioSource Europe S.A. Nevelles, Belgium), which had sensitivity limit of 6 pmol/l. The intra-assay and inter-assay coefficients of variation were 2.2% and 6.5%, respectively. There was no detectable cross-reactivity with human proinsulin or 32–33 split of proinsulin. 2.3. Assessment of carotid atherosclerosis The carotid arteries were examined bilaterally at the levels of the common carotid artery, the bifurcation, and the internal carotid arteries from transverse and longitudinal orientations by a single sonographer using B-mode ultrasound with 10 MHz linear probe (LOGIQ 7, GE, Milwaukee, WI). Carotid IMT was measured at three points of common carotid artery 1 cm proximal to the bifurcation and the mean value of six measurements from right and left carotid arteries were used. Carotid plaque was defined as an isolated focal lesion protruding into the lumen. Carotid atherosclerosis was defined as having a focal plaque or diffuse thickening of carotid wall (mean carotid IMT ≥1.1 mm).
Table 1 Characteristics of study subjects according to the quartiles of insulin sensitivity (Kitt). Quartile 1
Quartile 2
Quartile 3
Quartile 4
p-Value*
n Age (years) Women (%) Duration (years) BMI (kg/m2 )
616 55.6 ± 9.4 48.1 8.7 ± 7.0 25.3 ± 3.2
619 55.7 ± 8.7 46.7 8.0 ± 7.1 24.9 ± 3.0
617 55.8 ± 8.7 46.4 7.0 ± 6.6 24.7 ± 2.9
619 53.8 ± 8.9 48.9 6.3 ± 6.4 23.9 ± 2.8
<0.001 0.780 <0.001 <0.001
Smoking status (%) Never/current/ex smoker
65.6/14.9/19.5
63.8/17.9/18.3
66.5/14.7/18.8
67.5/13.6/18.9
Current treatment (%) Insulin Sulfonylureas Metformin AGIs ACEi and/or ARB Statins and/or fibrates
23.2 51.9 38.3 14.3 20.0 22.2
11.5 54.4 39.7 9.4 14.4 18.7
10.2 51.2 34.0 9.9 12.0 15.7
8.7 40.9 27.8 7.3 10.0 9.5
<0.001 <0.001 <0.001 0.001 <0.001 <0.001
Fasting plasma glucose (mmol/l) HbA1C (%) Systolic pressure (mm Hg) Diastolic pressure (mm Hg) Total cholesterol (mmol/l) Triglyceride (mmol/l) HDL cholesterol (mmol/l) LDL cholesterol (mmol/l) Fasting insulin (pmol/l) HOMA-IR Kitt (%/min) Mean IMT (mm) Carotid plaque (%)
10.7 ± 3.7 9.3 ± 1.8 139.6 ± 19.1 90.2 ± 11.3 5.3 ± 1.1 2.3 ± 1.6 1.3 ± 0.3 3.1 ± 0.9 71.7 ± 49.4 4.63 ± 2.74 0.9 ± 0.2 0.85 ± 0.18 54.3
8.6 ± 3.1 8.3 ± 1.8 137.7 ± 18.6 89.2 ± 11.2 5.2 ± 1.0 1.8 ± 1.2 1.3 ± 0.4 3.0 ± 0.9 62.3 ± 37.6. 3.27 ± 1.79 1.6 ± 0.2 0.84 ± 0.19 50.9
7.7 ± 2.7 8.1 ± 1.6 136.3 ± 18.2 88.9 ± 11.0 5.0 ± 0.9 1.7 ± 1.1 1.3 ± 0.4 3.0 ± 0.8 57.7 ± 35.0 2.76 ± 1.61 2.2 ± 0.2 0.82 ± 0.18 48.4
7.2 ± 2.4 7.6 ± 1.6 131.6 ± 17.1 86.2 ± 11.0 4.9 ± 0.9 1.4 ± 0.9 1.4 ± 0.4 2.9 ± 0.8 49.9 ± 23.3 2.26 ± 1.23 3.3 ± 0.6 0.80 ± 0.18 39.1
<0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 0.002 <0.001 <0.001 <0.001 <0.001 <0.001
0.527
* Data are expressed as means ± S.D. or proportion (%). p-Values are from one-way ANOVA or 2 test. Quartile 1 corresponds to the most insulin resistant group (lowest Kitt) and quartile 4 corresponds to the most insulin sensitive group (highest Kitt), Kitt (%/min); rate constant for plasma glucose disappearance after injection of regular insulin (0.1 U/kg). AGIs; alpha-glucosidase inhibitors, ACEi: angiotensin converting enzyme inhibiors, ARB: angiotensin II receptor blockers, Statins; HMG-co A reductase inhibitors.
S.W. Park et al. / Atherosclerosis 205 (2009) 309–313 Table 2 Bivariate correlations between clinical variables and mean IMT in patients with type 2 diabetes. Variables
r
p-Value
Age Sex BMI Current smoking Previous smoking Fasting glucose HbA1C Systolic pressure Diastolic pressure Total cholesterol Triglyceride HDL cholesterol LDL cholesterol Fasting insulin HOMA-IR Kitt (%/min)
0.396 0.074 0.040 −0.019 0.044 −0.010 −0.003 0.221 0.062 0.053 −0.006 −0.087 0.088 0.039 0.018 −0.106
<0.001 <0.001 0.047 0.066 <0.001 0.602 0.872 <0.001 0.002 0.009 0.762 <0.001 <0.001 0.051 0.364 <0.001
p-Values from simple linear regression models. IMT: intima-media thickness and Kitt (%/min); rate constant for plasma glucose disappearance after injection of regular insulin (0.1 U/kg).
2.4. Statistical analysis Statistical analyses were performed using SPSS for windows, version 12.0 (SPSS Inc., Chicago, IL). All values were presented as the mean ± S.D. or proportions (%). We used log transformed data for some of the key variables (i.e., fasting insulin, triglycerides, and HOMA-IR), which showed a skewed distribution. The significance of the comparisons of baseline characteristics was determined by ANOVA or 2 analyses. Bivariate correlations between clinical variables and carotid IMT were determined by simple linear regression models. Multiple linear regression models were used to test independent association of IR and IMT adjusting for potential confounders. We selected all the variables which showed potential association (p < 0.20) with IMT in bivariate analyses. We also entered known cardiovascular risk factors such as age, sex, BMI, smoking, blood pressure, and lipids into the model. We selected systolic pressure for blood pressure and LDL and HDL cholesterol for lipids omitting diastolic pressure and total cholesterol because of collinearity problem. Furthermore, the odds ratio for carotid atherosclerosis by the quartiles of insulin sensitivity were determined by multiple logistic regression models in all subjects and various subgroups stratified by age, sex, smoking status and hypertension. A p-value of less than 0.05 was considered statistically significant. 3. Results In patients with type 2 diabetes, those with relative insulin resistance (lower quartiles of Kitt) were a little bit older with longer duration of diabetes than insulin sensitive patients (those with higher quartiles of Kitt). Type 2 diabetic patients with IR also had higher values for metabolic parameters such as weight, BMI, fasting glucose, HbA1C, systolic and diastolic pressure, total cholesterol, triglyceride and LDL cholesterol but lower HDL cholesterol than insulin sensitive patients. Carotid IMT was the thickest in the most insulin resistant group (quartile 1) and decreased as increasing insulin sensitivity (higher quartiles of Kitt) (p < 0.001). In addition, the prevalence of carotid plaque was the highest in the most insulin resistant group (quartile 1) and decreased as increasing insulin sensitivity (towards quartile 4) (p < 0.001) (Table 1). In the bivariate analyses, age, sex, BMI, previous smoking, systolic and diastolic pressure, total cholesterol, HDL and LDL cholesterol and insulin sensitivity (Kitt) were significantly associated with carotid IMT (Table 2). But, neither fasting plasma
311
Table 3 Independent factors associated with mean IMT in patients with type 2 diabetes. Independent variables
ˇ
SE (ˇ)
Standardized ˇ
p-Value
Age (10 years) Gender (male) Systolic pressure (10 mm Hg) LDL cholesterol (mmol/l) HDL cholesterol (mmol/l) Kitt (%/min)
0.079 0.045 0.012 0.017 −0.030 −0.009
0.004 0.009 0.002 0.004 0.010 0.004
0.383 0.122 0.118 0.079 −0.058 −0.046
<0.001 <0.001 <0.001 <0.001 0.003 0.015
Results are from multiple linear regression model. BMI, current smoking, previous smoking and log transformed fasting insulin were not significantly associated with mean IMT (p-values were 0.615, 0.274, 0.303, 0.569, respectively). IMT: intimamedia thickness, Kitt (%/min); rate constant for plasma glucose disappearance after injection of regular insulin (0.1 U/kg).
Table 4 The odds ratio (OR) for carotid atherosclerosis in patients with type 2 diabetes according to the quartiles of insulin sensitivity (Kitt). Kitt
OR
95% CI
p
Adjusted ORa
95% CI
p
Quartile 1 Quartile 2 Quartile 3 Quartile 4
1.72 1.80 1.57 1
1.37, 2.15 1.44, 2.26 1.26, 1.97
<0.001 <0.001 <0.001
1.43 1.30 1.28 1
1.10, 1.86 1.01, 1.68 0.99, 1.64
0.007 0.042 0.057
Results are from multiple logistic regression model. Quartile 1 corresponds to the most insulin resistant group (lowest Kitt) and quartile 4 corresponds to the most insulin sensitive group (highest Kitt), Kitt (%/min); rate constant for plasma glucose disappearance after injection of regular insulin (0.1 U/kg). a Adjusted for age, sex, BMI, systolic pressure, HDL, LDL cholesterol, current and previous smoking and log transformed fasting insulin.
glucose nor HbA1C was associated with carotid IMT. Fasting insulin showed a tendency to be associated with IMT (p = 0.051). In multiple regression models, the association between insulin sensitivity (Kitt) and carotid IMT remained significant after controlling for age, sex, BMI, smoking status, systolic pressure, HDL, LDL cholesterol and fasting insulin levels. One standard deviation decrease in Kitt was associated with 0.046 mm increase in carotid IMT adjusting for above risk factors (p = 0.015) (Table 3). Further adjustment for various medications including sulfonylureas, metformin, alphaglucosidase inhibitors, angiotensin converting enzyme inhibitors, angiotensin II receptor blockers, HMG-co A reductase inhibitors, and fibrates did not eliminate the association of Kitt and carotid IMT (standardized ˇ: −0.041, p = 0.035). The odds ratio (OR) for carotid atherosclerosis was 1.72 (95% CI: 1.37, 2.15) in the most insulin resistant patients (quartile 1), 1.80 (1.44, 2.26) in quartile 2 and 1.57 (1.26, 1.97) in quartile 3 compared to the most insulin sensitive patients (quartile 4). Adjustments for age, sex, BMI, systolic pressure, HDL, LDL cholesterol and fasting insulin levels attenuated the ORs but the association between IR and carotid atherosclerosis remained significant (Table 4). In the subgroup analyses, the increased OR for carotid atherosclerosis were found in the type 2 diabetic patients with insulin resistance in all subgroups stratified by sex, age, smoking status and hypertension (Fig. 1). 4. Discussion In the present study, we demonstrated a significant association between insulin sensitivity index (Kitt) measured by SITT and subclinical atherosclerosis of carotid artery defined by either focal plaque or diffuse thickening of carotid IMT in a large cohort of patients with type 2 diabetes. This association remained significant after adjustment for known risk factors such as age, sex, BMI, smoking, systolic pressure, LDL and HDL cholesterol. Furthermore, the increased odds of carotid atherosclerosis in subjects with insulin resistance were found in all subgroups regardless of sex, age, smok-
312
S.W. Park et al. / Atherosclerosis 205 (2009) 309–313
Fig. 1. Odds ratio (OR with 95% CI) for carotid atherosclerosis in various subgroups of type 2 diabetic patients: those with insulin resistance (quartile 1) versus insulin sensitive group (quartile 4).
ing status and hypertension. These findings suggest that insulin resistance directly assessed by SITT is independently associated with subclinical atherosclerosis in patients with type 2 diabetes. Our results are consistent with recent findings from Watarai et al. [17] and Cardellini et al. [18] who showed that insulin sensitivity measured by euglycemic hyperinsulinemic clamp was associated with carotid IMT in patients with type 2 diabetes and in nondiabetic normotensive offspring of type 2 diabetic patients. Many [17–22,31,32] but not all [23–26] of previous studies have reported an independent association between insulin resistance and either clinical or subclinical CVD. The differences in study population in terms of varying degrees of glucose tolerance and the methods used to assess insulin sensitivity seem to result in inconsistent results between studies. Particularly, the results were inconsistent between studies when the study was cross-sectional and the surrogate of CVD such as carotid IMT was used as an outcome [17,18,21,22,24,25]. In the present study, we used SITT to assess in vivo insulin sensitivity in patients with type 2 diabetes. The plasma glucose disappearance rate (Kitt) from 3 to 15 min of SITT has been shown to be highly correlated with insulin sensitivity indices from euglycemic hyperinsulinemic clamp technique [13–15]. In our own reliability study, the correlation coefficient between Kitt and euglycemic clamp was 0.78 (p < 0.001) and intraparticipant coefficient of variation for Kitt was 8.8 ± 2.0% [16]. We found that hyperinsulinemia estimated by fasting insulin levels showed no association with carotid atherosclerosis in our study. Although we do not have direct evidence to explain this result, fasting insulin level may be insensitive index in patients with type 2 diabetes. It can be argued that fasting insulin level does not necessarily reflect body’s exposure to insulin. However, measurement of plasma insulin concentrations throughout the day is not practical and fasting insulin level has been used as a marker of hyperinsulinemia in previous studies [18,26]. It is also possible that fasting insulin is associated with IMT only in non-diabetic subjects but not in diabetic patients because diabetes is basically a status of beta cell failure which leads to relative insulin deficiency. We also found that only Kitt but not insulin resistance index from homeostasis model assessment (HOMA-IR) was an independent factor associated with carotid IMT in subjects with type 2 diabetes. Many of previous studies used HOMA-IR as a marker of insulin resistance because it is simple and easily calculated from fasting plasma glucose and insulin concentrations. However, there are also limitations of using HOMA-IR as a marker of insulin resistance particularly in patients with significant hyperglycemia [27–29]. In our previous study [29], we found that the correlation between HOMA-IR and
euglycemic hyperinsulinemic clamp was relatively poor in subjects with lower BMI, poor beta cell function, and higher fasting plasma glucose concentrations such as lean type 2 diabetic patients. In fact, such characteristics are very common in patients with type 2 diabetes in Korea as well as in many other Asian countries [30]. Our results suggest that the association between IR and carotid IMT could be affected by the methods of assessing IR, particularly in subjects with type 2 diabetes. There have been concerns about the risk of hypoglycemia limiting the clinical use of SITT in patients with type 2 diabetes. However, in our cohort of 2471 patients with type 2 diabetes, only 19 patients (0.8%) developed hypoglycemia defined as plasma glucose <2.8 mmol/l (50 mg/dl). Most of the hypoglycemic events were asymptomatic and further potential adverse events could be avoided by routine administration of 100 ml of 20% dextrose solution immediate after 15 min sampling. No serious event such as loss of consciousness, cardiovascular disease, or death was reported in relation to SITT in our study. There are several limitations for the study. First, because this is a cross-sectional study, the present results reflect only an association with prevalent carotid atherosclerosis, but not causal relationship. We could not completely rule out confounding effects of various medications though we excluded diabetic subjects treated with thiazolidinediones. Second, insulin resistance found in patients with type 2 diabetes might be secondary to chronic hyperglycemia rather than primary defect. We could not exclude this possibility. However, if the assessment of insulin resistance (Kitt) was influenced by the level of glycemia it might influence the result toward a null (no association) because neither fasting plasma glucose nor HbA1C was associated with carotid IMT in our study. Moreover, we performed a sensitivity analysis excluding subjects with severe hyperglycemia (fasting plasma glucose ≥10 mmol/l) and the results were essentially the same (data not shown). Lastly, we did not measure inflammatory cytokines such as C-reactive protein, TNF-␣, or IL-6 which have been known to be associated with IR and CVD. It is possible that these or other unmeasured inflammatory cytokines may mediate the link between IR and carotid atherosclerosis. In conclusion, we have demonstrated an independent association of insulin resistance measured by SITT and carotid atherosclerosis in a large cohort of patients with type 2 diabetes. We also found that fasting insulin level was not associated with carotid IMT, suggesting that insulin resistance itself, but not hyperinsulinemia might be an underlying mechanism of subclinical atherosclerosis in patients with type 2 diabetes. The Kitt derived from SITT may be used as a quantitative marker of IR and subclinical CVD in patients with type 2 diabetes. Acknowledgments This study was supported by Seoul City R&BD Program (10526), Seoul, Republic of Korea. We thank Jung Mi Park for her technical assistance in ultrasound examination of carotid arteries. References [1] Weyer C, Bogardus C, Mott DM, Pratley RE. The natural history of insulin secretory dysfunction and insulin resistance in the pathogenesis of type 2 diabetes mellitus. J Clin Invest 1999;104:787–94. [2] Bonora E, Kiechl S, Willeit J, et al. Population-based incidence rates and risk factors for type 2 diabetes in white individuals: the Bruneck Study. Diabetes 2004;53:1782–9. [3] Reaven GM. Role of insulin resistance in human disease. Diabetes 1988;37:1595–607. [4] DeFronzo RA, Ferrannini E. Insulin resistance: a multifaceted syndrome responsible for NIDDM, obesity, hypertension, dyslipidemia, and atherosclerotic cardiovascular disease. Diabetes Care 1991;14:173–94. [5] Bonora E, Kiechl S, Willeit J, et al. Metabolic syndrome: epidemiology and more extensive phenotypic description. Int J Obes Relat Metab Disord 2003;27:1283–9.
S.W. Park et al. / Atherosclerosis 205 (2009) 309–313 [6] Expert panel on detection, evaluation, and treatment of high blood cholesterol in adults: executive summary of the third report of the national cholesterol education program (NCEP) expert panel on detection, evaluation, and treatment of high blood cholesterol in adults (Adult Treatment Panel III). JAMA 2001;285:2486–97. [7] World Health Organization. Guidelines for the prevention, management and care of diabetes mellitus. Geneva: World Health Org.; 2006 (EMRO Technical Publications Series 32). [8] Alberti KGMM, Zimmet PZ, Shaw JE. The metabolic syndrome—a new worldwide definition from the International Diabetes Federation Consensus. Lancet 2005;366:1059–62. [9] Geiss LS, Herman WH, Smith PJ. Mortality in non-insulin dependent diabetes mellitus. In: Bethesda MD, editor. Diabetes in America. 2nd ed. National Institutes of Health; 1995 (NIH publ. nos. 95–1468). [10] Stern MP. Diabetes and cardiovascular disease. The common soil hypothesis. Diabetes 1995;44:369–74. [11] Lebovitz HE. Insulin resistance—a common link between type 2 diabetes and cardiovascular disease. Diabetes Obes Metab 2006;8:237–49. [12] Hanley AJ, Williams K, Stern MP, Haffner SM. Homeostasis model assessment of insulin resistance in relation to the incidence of cardiovascular disease. Diabetes Care 2002;25:1177–84. [13] Bonora E, Moghetti P, Zancanaro C, et al. Estimates of in vivo insulin action in man: comparison of insulin tolerance tests with euglycemic and hyperglycemic glucose clamp studies. J Clin Endocrinol Metab 1989;68:374–8. [14] Young RP, Critchley JA, Anderson PJ, Lau MS, Lee KK, Chan JC. The short insulin tolerance test: feasibility study using venous sampling. Diabet Med 1996;13:429–33. [15] Alberti KGGM, Daly ME, Robibson A, Marshall SM, Maathers JC. The short insulin tolerance test is safe and reproducible. Diabet Med 1999;16:352–3. [16] Park SW, Yun YS, Ahn CW, et al. Short insulin tolerance test for the determination of in vivo insulin sensitivity: a comparison with euglycemic clamp test. Kor Diabetes J 1998;22:199–208. [17] Watarai T, Yamasaki Y, Ikeda M, et al. Insulin resistance contributes to carotid arterial wall thickness in patients with non-insulin-dependent-diabetes mellitus. Endocr J 1999;46:629–38. [18] Cardellini M, Marini MA, Frontoni S, et al. Carotid artery intima-media thickness is associated with insulin-mediated glucose disposal in nondiabetic normotensive offspring of type 2 diabetic patients. Am J Physiol Endocrinol Metab 2007;292:E347–352. [19] Bonora E, Kiechl S, Willeit J, et al. Insulin resistance as estimated by homeostasis model assessment predicts incident symptomatic cardiovascular disease
[20]
[21]
[22]
[23]
[24]
[25]
[26]
[27] [28]
[29]
[30]
[31] [32]
313
in Caucasian subjects from the general population. Diabetes Care 2007;30: 318–24. Yip J, Facchini FS, Reaven GM. Resistance to insulin-mediated glucose disposal as a predictor of cardiovascular disease. J Clin Endocrinol Metab 1998;83:2773–6. Raajala U, Laakso M, Paivansalo M, Pelkonen O, Suramo I, KeinanenKiukaanniemi S. Low insulin sensitivity measured by both quantitative insulin sensitivity check index and homeostasis model assessment method as a risk factor of increased intima-media thickness of the carotid artery. J Clin Endocrinol Metab 2002;87:5092–7. Bonora E, Tessari R, Micciolo R, et al. Intima-medial thickness of the carotid artery in nondiabetic and NIDDM patients. Relationship with insulin resistance. Diabetes Care 1997;24:627–31. Rutter MK, Meigs JB, Sullivan LM, D’Agostino RB, Wilson PW. Insulin resistance, the metabolic syndrome, and incident cardiovascular events in the Framingham offspring study. Diabetes 2005;54:3252–7. Ishizaka N, Ishizaka Y, Takahashi E, et al. Association between insulin resistance and carotid arteriosclerosis in subjects with normal fasting glucose and normal glucose tolerance. Arterioscler Throm Vasc Biol 2003;23:295–301. Larsson H, Berglund G, Ahren B. Insulin sensitivity, insulin secretion, and glucose tolerance versus intima-media thickness in nondiabetic postmenopausal women. J Clin Endocrinol Metab 2003;88:4791–7. Ferrara A, Barrett-Connor EL, Edelstein SL. Hyperinsulinemia does not increase the risk of fatal cardiovascular disease in elderly men or women without diabetes: the Rancho Bernardo Study. Am J Epidemiol 1994;140:857–69. Wallace TM, Levy JC, Matthews DR. Use and abuse of HOMA modeling. Diabetes Care 2004;27:1487–95. Muniyappa R, Lee S, Chen H, Quon MJ. Current approaches for assessing insulin sensitivity and resistance in vivo: advantages, limitations, and appropriate usage. Am J Physiol Endocrinol Metab 2008;294:E15–26. Kang ES, Yun YS, Park SW, et al. Limitation of the validity of the homeostasis model assessment as an index of insulin resistance in Korea. Metabolism 2005;54:206–11. Kim DJ, Song KE, Park JW, Cho HK, Lee KW, Huh KB. Clinical characteristics of Korean type 2 diabetic patients in 2005. Diabetes Res Clin Pract 2007;77S:S252–7. Howard G, O’Leary DH, Zaccaro S, et al. Insulin sensitivity and atherosclerosis. Circulation 1996;93:1809–17. Haffner SM, D’Agostino R, Mykkänen L, et al. Proinsulin and insulin concentrations in relation to carotid wall thickness: insulin resistance atherosclerosis study. Stroke 1998;29:1498–503.
Contents lists available at ScienceDirect
Atherosclerosis journal homepage: www.elsevier.com/locate/atherosclerosis
Insulin resistance and carotid atherosclerosis in patients with type 2 diabetes夽 Seok Won Park a,∗ , Soo Kyung Kim a , Yong-Wook Cho a , Dae Jung Kim b , Young-Duk Song c , Young Ju Choi d , Byung Wook Huh d , Sung Hee Choi e , Sun Ha Jee f , Mi Ae Cho g , Eun Jig Lee g , Kap Bum Huh d a
Department of Internal Medicine, Pochon CHA University, 351 Yatapdong, Sungnam 463-712, Republic of Korea Department of Endocrinology and Metabolism, Ajou University School of Medicine, Suwon, Republic of Korea Department of Medicine, Ilsan Hospital, Goyang, Republic of Korea d Huh Diabetes Center and the 21st Century Diabetes and Vascular Research Institute, Seoul, Republic of Korea e Department of Internal Medicine, Seoul National University, Seoul, Republic of Korea f Department of Epidemiology, Graduate School of Public Health, Yonsei University, Seoul, Republic of Korea g Department of Medicine, Yonsei University, Seoul, Republic of Korea b c
a r t i c l e
i n f o
Article history: Received 20 May 2008 Received in revised form 20 November 2008 Accepted 4 December 2008 Available online 13 December 2008 Keywords: Insulin resistance Hyperinsulinemia Carotid atherosclerosis Type 2 diabetes
a b s t r a c t Objective: Insulin resistance may provide a crucial link between type 2 diabetes and cardiovascular disease. However, it is still unclear whether insulin resistance itself or hyperinsulinemia is independently associated with subclinical atherosclerosis. We hypothesized that insulin resistance, but not hyperinsulinemia, would be associated with carotid atherosclerosis in patients with type 2 diabetes. Methods: We examined 2471 patients with type 2 diabetes, consecutively enrolled in Huh Diabetes Center. Insulin sensitivity was directly assessed by a rate constant for plasma glucose disappearance (Kitt) using short insulin tolerance test. Fasting insulin levels were used as a marker of hyperinsulinemia. Both carotid arteries were examined by B-mode ultrasound. Carotid atherosclerosis was defined by having a clearly isolated focal plaque or mean carotid intima-media thickness (IMT) ≥1.1 mm. Results: In multiple regression models, insulin sensitivity index (Kitt) but not hyperinsulinemia was significantly associated with carotid IMT adjusting for known risk factors such as age, sex, BMI, smoking, systolic pressure, HDL and LDL cholesterol. One standard deviation decrease in Kitt was associated with 0.046 mm increase in carotid IMT (p = 0.015). Furthermore, odds ratio for carotid atherosclerosis was 1.43 (95% CI: 1.10, 1.86) in type 2 diabetic patients with insulin resistance (lowest quartile of insulin sensitivity) adjusting for known risk factors. The results were consistent in all subgroups stratified by sex, age, smoking and hypertension. Conclusion: Insulin resistance measured by short insulin tolerance test, but not hyperinsulinemia, is independently associated with carotid atherosclerosis in patients with type 2 diabetes. © 2008 Elsevier Ireland Ltd. All rights reserved.
1. Introduction Insulin resistance (IR) plays a major role for the development of type 2 diabetes [1,2]. IR is also associated with a variety of cardiovascular disease (CVD) risk factors including central obesity, essential hypertension, hypertriglyceridemia, and low HDL cholesterol [3]. The concept of metabolic syndrome comes from the observation of clustering of risk factors within a single individual [4,5]. It has been hypothesized that IR and/or its consequent hyperinsulinemia might provide an underlying pathophysiological mechanism for the
夽 This study was supported by Seoul City R&BD Program (10526), Seoul, Republic of Korea. ∗ Corresponding author. Tel.: +82 31 780 5211; fax: +82 31 780 5208. E-mail address: [email protected] (S.W. Park). 0021-9150/$ – see front matter © 2008 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.atherosclerosis.2008.12.006
clustering of risk factors although ironically neither IR nor hyperinsulinemia is included as a component to define metabolic syndrome [6–8]. CVD is a leading cause of death among individuals with type 2 diabetes [9]. Again, IR has been proposed to play a crucial role in the link between type 2 diabetes and CVD [10–12]. The associations between IR and CVD or its surrogate marker such as intima-media thickness (IMT) have been documented mainly in subjects with varying degrees of glucose tolerance. However, it is still unclear whether IR itself or hyperinsulinemia confers an independent association with IMT beyond known risk factors, particularly in patients with type 2 diabetes. The aim of the study was to investigate whether insulin resistance directly measured by short insulin tolerance test or hyperinsulinemia estimated by fasting insulin level has independent association with carotid atherosclerosis in a large cohort of patients with type 2 diabetes.
310
S.W. Park et al. / Atherosclerosis 205 (2009) 309–313
2. Methods 2.1. Study population We have enrolled 3704 patients with type 2 diabetes, consecutively from January 2003 to June 2006 at the Huh Diabetes Center in Seoul, Korea as a part of Seoul metabolic syndrome cohort. Selection criteria for the study were as follows: age at diagnosis ≥20 years, current age <70 years, BMI ≥18.5 and <40 kg/m2 , no need of insulin therapy for the first 2 years of disease, no history of ketoacidosis, no present glucocorticoid therapy. Among 3220 subjects who met the above criteria, short insulin tolerance test (SITT) was performed in 3030 patients (94.1%) and carotid intima-media thickness was measured in 2946 patients (91.5%). We further excluded patients who were treated with thiazolidinediones (n, 263) because it has been reported to affect insulin resistance as well as IMT. Finally 2471 subjects who had been assessed for both SITT and IMT were included in the study. Clinical characteristics of participants were summarized in Table 1. The protocol was approved by the Ethics Committee of the Yonsei University. Informed consent was obtained from all subjects. All the investigations were performed in accordance with the principles of the Declaration of Helsinki.
2.2. Assessment of insulin sensitivity and hyperinsulinemia Insulin sensitivity was directly assessed by SITT as a rate constant for plasma glucose disappearance (Kitt: %/min) [13–16]. The subjects were attended to the clinic at 08:00 after overnight fasting. Two intravenous cannulas were established, one for the blood sampling and the other one for insulin injection. A heating pad was applied to hand vein for warming to get an arterialized venous sample. After 15 min resting, pre-diluted regular insulin (Humulin® , Eli Lilly, Indianapolis, IN) at a dose of 0.1 U/kg was injected through antecubital vein and blood samples were drawn from contralateral
hand vein through retrograde cannula at 0, 3, 6, 9, 12 and 15 min after insulin injection. Plasma glucose concentrations were determined immediately after sampling using Beckman glucose analyzer II (Beckman Ins., Fullerton, CA) and then the rate constant for the plasma glucose disappearance (Kitt) was calculated from the slope of the fall in log transformed plasma glucose between 3 and 15 min [13]. The Kitt value was used as an index of in vivo insulin sensitivity. The subjects were classified by the quartiles of Kitt; quartile 1 (those with the lowest quartile of Kitt) was the most insulin resistance group whereas quartile 4 was the most sensitive to insulin (Table 1). The intra-participant coefficient of variation for Kitt was 8.8 ± 2.0% [16]. We routinely infused 100 ml of 20% dextrose solution immediate after the test to avoid any potential hypoglycemia. We used fasting insulin concentrations as a marker of hyperinsulinemia. Fasting insulin concentrations were determined by means of immunoradiometric assay using INS-IRMA Kit (BioSource Europe S.A. Nevelles, Belgium), which had sensitivity limit of 6 pmol/l. The intra-assay and inter-assay coefficients of variation were 2.2% and 6.5%, respectively. There was no detectable cross-reactivity with human proinsulin or 32–33 split of proinsulin. 2.3. Assessment of carotid atherosclerosis The carotid arteries were examined bilaterally at the levels of the common carotid artery, the bifurcation, and the internal carotid arteries from transverse and longitudinal orientations by a single sonographer using B-mode ultrasound with 10 MHz linear probe (LOGIQ 7, GE, Milwaukee, WI). Carotid IMT was measured at three points of common carotid artery 1 cm proximal to the bifurcation and the mean value of six measurements from right and left carotid arteries were used. Carotid plaque was defined as an isolated focal lesion protruding into the lumen. Carotid atherosclerosis was defined as having a focal plaque or diffuse thickening of carotid wall (mean carotid IMT ≥1.1 mm).
Table 1 Characteristics of study subjects according to the quartiles of insulin sensitivity (Kitt). Quartile 1
Quartile 2
Quartile 3
Quartile 4
p-Value*
n Age (years) Women (%) Duration (years) BMI (kg/m2 )
616 55.6 ± 9.4 48.1 8.7 ± 7.0 25.3 ± 3.2
619 55.7 ± 8.7 46.7 8.0 ± 7.1 24.9 ± 3.0
617 55.8 ± 8.7 46.4 7.0 ± 6.6 24.7 ± 2.9
619 53.8 ± 8.9 48.9 6.3 ± 6.4 23.9 ± 2.8
<0.001 0.780 <0.001 <0.001
Smoking status (%) Never/current/ex smoker
65.6/14.9/19.5
63.8/17.9/18.3
66.5/14.7/18.8
67.5/13.6/18.9
Current treatment (%) Insulin Sulfonylureas Metformin AGIs ACEi and/or ARB Statins and/or fibrates
23.2 51.9 38.3 14.3 20.0 22.2
11.5 54.4 39.7 9.4 14.4 18.7
10.2 51.2 34.0 9.9 12.0 15.7
8.7 40.9 27.8 7.3 10.0 9.5
<0.001 <0.001 <0.001 0.001 <0.001 <0.001
Fasting plasma glucose (mmol/l) HbA1C (%) Systolic pressure (mm Hg) Diastolic pressure (mm Hg) Total cholesterol (mmol/l) Triglyceride (mmol/l) HDL cholesterol (mmol/l) LDL cholesterol (mmol/l) Fasting insulin (pmol/l) HOMA-IR Kitt (%/min) Mean IMT (mm) Carotid plaque (%)
10.7 ± 3.7 9.3 ± 1.8 139.6 ± 19.1 90.2 ± 11.3 5.3 ± 1.1 2.3 ± 1.6 1.3 ± 0.3 3.1 ± 0.9 71.7 ± 49.4 4.63 ± 2.74 0.9 ± 0.2 0.85 ± 0.18 54.3
8.6 ± 3.1 8.3 ± 1.8 137.7 ± 18.6 89.2 ± 11.2 5.2 ± 1.0 1.8 ± 1.2 1.3 ± 0.4 3.0 ± 0.9 62.3 ± 37.6. 3.27 ± 1.79 1.6 ± 0.2 0.84 ± 0.19 50.9
7.7 ± 2.7 8.1 ± 1.6 136.3 ± 18.2 88.9 ± 11.0 5.0 ± 0.9 1.7 ± 1.1 1.3 ± 0.4 3.0 ± 0.8 57.7 ± 35.0 2.76 ± 1.61 2.2 ± 0.2 0.82 ± 0.18 48.4
7.2 ± 2.4 7.6 ± 1.6 131.6 ± 17.1 86.2 ± 11.0 4.9 ± 0.9 1.4 ± 0.9 1.4 ± 0.4 2.9 ± 0.8 49.9 ± 23.3 2.26 ± 1.23 3.3 ± 0.6 0.80 ± 0.18 39.1
<0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 0.002 <0.001 <0.001 <0.001 <0.001 <0.001
0.527
* Data are expressed as means ± S.D. or proportion (%). p-Values are from one-way ANOVA or 2 test. Quartile 1 corresponds to the most insulin resistant group (lowest Kitt) and quartile 4 corresponds to the most insulin sensitive group (highest Kitt), Kitt (%/min); rate constant for plasma glucose disappearance after injection of regular insulin (0.1 U/kg). AGIs; alpha-glucosidase inhibitors, ACEi: angiotensin converting enzyme inhibiors, ARB: angiotensin II receptor blockers, Statins; HMG-co A reductase inhibitors.
S.W. Park et al. / Atherosclerosis 205 (2009) 309–313 Table 2 Bivariate correlations between clinical variables and mean IMT in patients with type 2 diabetes. Variables
r
p-Value
Age Sex BMI Current smoking Previous smoking Fasting glucose HbA1C Systolic pressure Diastolic pressure Total cholesterol Triglyceride HDL cholesterol LDL cholesterol Fasting insulin HOMA-IR Kitt (%/min)
0.396 0.074 0.040 −0.019 0.044 −0.010 −0.003 0.221 0.062 0.053 −0.006 −0.087 0.088 0.039 0.018 −0.106
<0.001 <0.001 0.047 0.066 <0.001 0.602 0.872 <0.001 0.002 0.009 0.762 <0.001 <0.001 0.051 0.364 <0.001
p-Values from simple linear regression models. IMT: intima-media thickness and Kitt (%/min); rate constant for plasma glucose disappearance after injection of regular insulin (0.1 U/kg).
2.4. Statistical analysis Statistical analyses were performed using SPSS for windows, version 12.0 (SPSS Inc., Chicago, IL). All values were presented as the mean ± S.D. or proportions (%). We used log transformed data for some of the key variables (i.e., fasting insulin, triglycerides, and HOMA-IR), which showed a skewed distribution. The significance of the comparisons of baseline characteristics was determined by ANOVA or 2 analyses. Bivariate correlations between clinical variables and carotid IMT were determined by simple linear regression models. Multiple linear regression models were used to test independent association of IR and IMT adjusting for potential confounders. We selected all the variables which showed potential association (p < 0.20) with IMT in bivariate analyses. We also entered known cardiovascular risk factors such as age, sex, BMI, smoking, blood pressure, and lipids into the model. We selected systolic pressure for blood pressure and LDL and HDL cholesterol for lipids omitting diastolic pressure and total cholesterol because of collinearity problem. Furthermore, the odds ratio for carotid atherosclerosis by the quartiles of insulin sensitivity were determined by multiple logistic regression models in all subjects and various subgroups stratified by age, sex, smoking status and hypertension. A p-value of less than 0.05 was considered statistically significant. 3. Results In patients with type 2 diabetes, those with relative insulin resistance (lower quartiles of Kitt) were a little bit older with longer duration of diabetes than insulin sensitive patients (those with higher quartiles of Kitt). Type 2 diabetic patients with IR also had higher values for metabolic parameters such as weight, BMI, fasting glucose, HbA1C, systolic and diastolic pressure, total cholesterol, triglyceride and LDL cholesterol but lower HDL cholesterol than insulin sensitive patients. Carotid IMT was the thickest in the most insulin resistant group (quartile 1) and decreased as increasing insulin sensitivity (higher quartiles of Kitt) (p < 0.001). In addition, the prevalence of carotid plaque was the highest in the most insulin resistant group (quartile 1) and decreased as increasing insulin sensitivity (towards quartile 4) (p < 0.001) (Table 1). In the bivariate analyses, age, sex, BMI, previous smoking, systolic and diastolic pressure, total cholesterol, HDL and LDL cholesterol and insulin sensitivity (Kitt) were significantly associated with carotid IMT (Table 2). But, neither fasting plasma
311
Table 3 Independent factors associated with mean IMT in patients with type 2 diabetes. Independent variables
ˇ
SE (ˇ)
Standardized ˇ
p-Value
Age (10 years) Gender (male) Systolic pressure (10 mm Hg) LDL cholesterol (mmol/l) HDL cholesterol (mmol/l) Kitt (%/min)
0.079 0.045 0.012 0.017 −0.030 −0.009
0.004 0.009 0.002 0.004 0.010 0.004
0.383 0.122 0.118 0.079 −0.058 −0.046
<0.001 <0.001 <0.001 <0.001 0.003 0.015
Results are from multiple linear regression model. BMI, current smoking, previous smoking and log transformed fasting insulin were not significantly associated with mean IMT (p-values were 0.615, 0.274, 0.303, 0.569, respectively). IMT: intimamedia thickness, Kitt (%/min); rate constant for plasma glucose disappearance after injection of regular insulin (0.1 U/kg).
Table 4 The odds ratio (OR) for carotid atherosclerosis in patients with type 2 diabetes according to the quartiles of insulin sensitivity (Kitt). Kitt
OR
95% CI
p
Adjusted ORa
95% CI
p
Quartile 1 Quartile 2 Quartile 3 Quartile 4
1.72 1.80 1.57 1
1.37, 2.15 1.44, 2.26 1.26, 1.97
<0.001 <0.001 <0.001
1.43 1.30 1.28 1
1.10, 1.86 1.01, 1.68 0.99, 1.64
0.007 0.042 0.057
Results are from multiple logistic regression model. Quartile 1 corresponds to the most insulin resistant group (lowest Kitt) and quartile 4 corresponds to the most insulin sensitive group (highest Kitt), Kitt (%/min); rate constant for plasma glucose disappearance after injection of regular insulin (0.1 U/kg). a Adjusted for age, sex, BMI, systolic pressure, HDL, LDL cholesterol, current and previous smoking and log transformed fasting insulin.
glucose nor HbA1C was associated with carotid IMT. Fasting insulin showed a tendency to be associated with IMT (p = 0.051). In multiple regression models, the association between insulin sensitivity (Kitt) and carotid IMT remained significant after controlling for age, sex, BMI, smoking status, systolic pressure, HDL, LDL cholesterol and fasting insulin levels. One standard deviation decrease in Kitt was associated with 0.046 mm increase in carotid IMT adjusting for above risk factors (p = 0.015) (Table 3). Further adjustment for various medications including sulfonylureas, metformin, alphaglucosidase inhibitors, angiotensin converting enzyme inhibitors, angiotensin II receptor blockers, HMG-co A reductase inhibitors, and fibrates did not eliminate the association of Kitt and carotid IMT (standardized ˇ: −0.041, p = 0.035). The odds ratio (OR) for carotid atherosclerosis was 1.72 (95% CI: 1.37, 2.15) in the most insulin resistant patients (quartile 1), 1.80 (1.44, 2.26) in quartile 2 and 1.57 (1.26, 1.97) in quartile 3 compared to the most insulin sensitive patients (quartile 4). Adjustments for age, sex, BMI, systolic pressure, HDL, LDL cholesterol and fasting insulin levels attenuated the ORs but the association between IR and carotid atherosclerosis remained significant (Table 4). In the subgroup analyses, the increased OR for carotid atherosclerosis were found in the type 2 diabetic patients with insulin resistance in all subgroups stratified by sex, age, smoking status and hypertension (Fig. 1). 4. Discussion In the present study, we demonstrated a significant association between insulin sensitivity index (Kitt) measured by SITT and subclinical atherosclerosis of carotid artery defined by either focal plaque or diffuse thickening of carotid IMT in a large cohort of patients with type 2 diabetes. This association remained significant after adjustment for known risk factors such as age, sex, BMI, smoking, systolic pressure, LDL and HDL cholesterol. Furthermore, the increased odds of carotid atherosclerosis in subjects with insulin resistance were found in all subgroups regardless of sex, age, smok-
312
S.W. Park et al. / Atherosclerosis 205 (2009) 309–313
Fig. 1. Odds ratio (OR with 95% CI) for carotid atherosclerosis in various subgroups of type 2 diabetic patients: those with insulin resistance (quartile 1) versus insulin sensitive group (quartile 4).
ing status and hypertension. These findings suggest that insulin resistance directly assessed by SITT is independently associated with subclinical atherosclerosis in patients with type 2 diabetes. Our results are consistent with recent findings from Watarai et al. [17] and Cardellini et al. [18] who showed that insulin sensitivity measured by euglycemic hyperinsulinemic clamp was associated with carotid IMT in patients with type 2 diabetes and in nondiabetic normotensive offspring of type 2 diabetic patients. Many [17–22,31,32] but not all [23–26] of previous studies have reported an independent association between insulin resistance and either clinical or subclinical CVD. The differences in study population in terms of varying degrees of glucose tolerance and the methods used to assess insulin sensitivity seem to result in inconsistent results between studies. Particularly, the results were inconsistent between studies when the study was cross-sectional and the surrogate of CVD such as carotid IMT was used as an outcome [17,18,21,22,24,25]. In the present study, we used SITT to assess in vivo insulin sensitivity in patients with type 2 diabetes. The plasma glucose disappearance rate (Kitt) from 3 to 15 min of SITT has been shown to be highly correlated with insulin sensitivity indices from euglycemic hyperinsulinemic clamp technique [13–15]. In our own reliability study, the correlation coefficient between Kitt and euglycemic clamp was 0.78 (p < 0.001) and intraparticipant coefficient of variation for Kitt was 8.8 ± 2.0% [16]. We found that hyperinsulinemia estimated by fasting insulin levels showed no association with carotid atherosclerosis in our study. Although we do not have direct evidence to explain this result, fasting insulin level may be insensitive index in patients with type 2 diabetes. It can be argued that fasting insulin level does not necessarily reflect body’s exposure to insulin. However, measurement of plasma insulin concentrations throughout the day is not practical and fasting insulin level has been used as a marker of hyperinsulinemia in previous studies [18,26]. It is also possible that fasting insulin is associated with IMT only in non-diabetic subjects but not in diabetic patients because diabetes is basically a status of beta cell failure which leads to relative insulin deficiency. We also found that only Kitt but not insulin resistance index from homeostasis model assessment (HOMA-IR) was an independent factor associated with carotid IMT in subjects with type 2 diabetes. Many of previous studies used HOMA-IR as a marker of insulin resistance because it is simple and easily calculated from fasting plasma glucose and insulin concentrations. However, there are also limitations of using HOMA-IR as a marker of insulin resistance particularly in patients with significant hyperglycemia [27–29]. In our previous study [29], we found that the correlation between HOMA-IR and
euglycemic hyperinsulinemic clamp was relatively poor in subjects with lower BMI, poor beta cell function, and higher fasting plasma glucose concentrations such as lean type 2 diabetic patients. In fact, such characteristics are very common in patients with type 2 diabetes in Korea as well as in many other Asian countries [30]. Our results suggest that the association between IR and carotid IMT could be affected by the methods of assessing IR, particularly in subjects with type 2 diabetes. There have been concerns about the risk of hypoglycemia limiting the clinical use of SITT in patients with type 2 diabetes. However, in our cohort of 2471 patients with type 2 diabetes, only 19 patients (0.8%) developed hypoglycemia defined as plasma glucose <2.8 mmol/l (50 mg/dl). Most of the hypoglycemic events were asymptomatic and further potential adverse events could be avoided by routine administration of 100 ml of 20% dextrose solution immediate after 15 min sampling. No serious event such as loss of consciousness, cardiovascular disease, or death was reported in relation to SITT in our study. There are several limitations for the study. First, because this is a cross-sectional study, the present results reflect only an association with prevalent carotid atherosclerosis, but not causal relationship. We could not completely rule out confounding effects of various medications though we excluded diabetic subjects treated with thiazolidinediones. Second, insulin resistance found in patients with type 2 diabetes might be secondary to chronic hyperglycemia rather than primary defect. We could not exclude this possibility. However, if the assessment of insulin resistance (Kitt) was influenced by the level of glycemia it might influence the result toward a null (no association) because neither fasting plasma glucose nor HbA1C was associated with carotid IMT in our study. Moreover, we performed a sensitivity analysis excluding subjects with severe hyperglycemia (fasting plasma glucose ≥10 mmol/l) and the results were essentially the same (data not shown). Lastly, we did not measure inflammatory cytokines such as C-reactive protein, TNF-␣, or IL-6 which have been known to be associated with IR and CVD. It is possible that these or other unmeasured inflammatory cytokines may mediate the link between IR and carotid atherosclerosis. In conclusion, we have demonstrated an independent association of insulin resistance measured by SITT and carotid atherosclerosis in a large cohort of patients with type 2 diabetes. We also found that fasting insulin level was not associated with carotid IMT, suggesting that insulin resistance itself, but not hyperinsulinemia might be an underlying mechanism of subclinical atherosclerosis in patients with type 2 diabetes. The Kitt derived from SITT may be used as a quantitative marker of IR and subclinical CVD in patients with type 2 diabetes. Acknowledgments This study was supported by Seoul City R&BD Program (10526), Seoul, Republic of Korea. We thank Jung Mi Park for her technical assistance in ultrasound examination of carotid arteries. References [1] Weyer C, Bogardus C, Mott DM, Pratley RE. The natural history of insulin secretory dysfunction and insulin resistance in the pathogenesis of type 2 diabetes mellitus. J Clin Invest 1999;104:787–94. [2] Bonora E, Kiechl S, Willeit J, et al. Population-based incidence rates and risk factors for type 2 diabetes in white individuals: the Bruneck Study. Diabetes 2004;53:1782–9. [3] Reaven GM. Role of insulin resistance in human disease. Diabetes 1988;37:1595–607. [4] DeFronzo RA, Ferrannini E. Insulin resistance: a multifaceted syndrome responsible for NIDDM, obesity, hypertension, dyslipidemia, and atherosclerotic cardiovascular disease. Diabetes Care 1991;14:173–94. [5] Bonora E, Kiechl S, Willeit J, et al. Metabolic syndrome: epidemiology and more extensive phenotypic description. Int J Obes Relat Metab Disord 2003;27:1283–9.
S.W. Park et al. / Atherosclerosis 205 (2009) 309–313 [6] Expert panel on detection, evaluation, and treatment of high blood cholesterol in adults: executive summary of the third report of the national cholesterol education program (NCEP) expert panel on detection, evaluation, and treatment of high blood cholesterol in adults (Adult Treatment Panel III). JAMA 2001;285:2486–97. [7] World Health Organization. Guidelines for the prevention, management and care of diabetes mellitus. Geneva: World Health Org.; 2006 (EMRO Technical Publications Series 32). [8] Alberti KGMM, Zimmet PZ, Shaw JE. The metabolic syndrome—a new worldwide definition from the International Diabetes Federation Consensus. Lancet 2005;366:1059–62. [9] Geiss LS, Herman WH, Smith PJ. Mortality in non-insulin dependent diabetes mellitus. In: Bethesda MD, editor. Diabetes in America. 2nd ed. National Institutes of Health; 1995 (NIH publ. nos. 95–1468). [10] Stern MP. Diabetes and cardiovascular disease. The common soil hypothesis. Diabetes 1995;44:369–74. [11] Lebovitz HE. Insulin resistance—a common link between type 2 diabetes and cardiovascular disease. Diabetes Obes Metab 2006;8:237–49. [12] Hanley AJ, Williams K, Stern MP, Haffner SM. Homeostasis model assessment of insulin resistance in relation to the incidence of cardiovascular disease. Diabetes Care 2002;25:1177–84. [13] Bonora E, Moghetti P, Zancanaro C, et al. Estimates of in vivo insulin action in man: comparison of insulin tolerance tests with euglycemic and hyperglycemic glucose clamp studies. J Clin Endocrinol Metab 1989;68:374–8. [14] Young RP, Critchley JA, Anderson PJ, Lau MS, Lee KK, Chan JC. The short insulin tolerance test: feasibility study using venous sampling. Diabet Med 1996;13:429–33. [15] Alberti KGGM, Daly ME, Robibson A, Marshall SM, Maathers JC. The short insulin tolerance test is safe and reproducible. Diabet Med 1999;16:352–3. [16] Park SW, Yun YS, Ahn CW, et al. Short insulin tolerance test for the determination of in vivo insulin sensitivity: a comparison with euglycemic clamp test. Kor Diabetes J 1998;22:199–208. [17] Watarai T, Yamasaki Y, Ikeda M, et al. Insulin resistance contributes to carotid arterial wall thickness in patients with non-insulin-dependent-diabetes mellitus. Endocr J 1999;46:629–38. [18] Cardellini M, Marini MA, Frontoni S, et al. Carotid artery intima-media thickness is associated with insulin-mediated glucose disposal in nondiabetic normotensive offspring of type 2 diabetic patients. Am J Physiol Endocrinol Metab 2007;292:E347–352. [19] Bonora E, Kiechl S, Willeit J, et al. Insulin resistance as estimated by homeostasis model assessment predicts incident symptomatic cardiovascular disease
[20]
[21]
[22]
[23]
[24]
[25]
[26]
[27] [28]
[29]
[30]
[31] [32]
313
in Caucasian subjects from the general population. Diabetes Care 2007;30: 318–24. Yip J, Facchini FS, Reaven GM. Resistance to insulin-mediated glucose disposal as a predictor of cardiovascular disease. J Clin Endocrinol Metab 1998;83:2773–6. Raajala U, Laakso M, Paivansalo M, Pelkonen O, Suramo I, KeinanenKiukaanniemi S. Low insulin sensitivity measured by both quantitative insulin sensitivity check index and homeostasis model assessment method as a risk factor of increased intima-media thickness of the carotid artery. J Clin Endocrinol Metab 2002;87:5092–7. Bonora E, Tessari R, Micciolo R, et al. Intima-medial thickness of the carotid artery in nondiabetic and NIDDM patients. Relationship with insulin resistance. Diabetes Care 1997;24:627–31. Rutter MK, Meigs JB, Sullivan LM, D’Agostino RB, Wilson PW. Insulin resistance, the metabolic syndrome, and incident cardiovascular events in the Framingham offspring study. Diabetes 2005;54:3252–7. Ishizaka N, Ishizaka Y, Takahashi E, et al. Association between insulin resistance and carotid arteriosclerosis in subjects with normal fasting glucose and normal glucose tolerance. Arterioscler Throm Vasc Biol 2003;23:295–301. Larsson H, Berglund G, Ahren B. Insulin sensitivity, insulin secretion, and glucose tolerance versus intima-media thickness in nondiabetic postmenopausal women. J Clin Endocrinol Metab 2003;88:4791–7. Ferrara A, Barrett-Connor EL, Edelstein SL. Hyperinsulinemia does not increase the risk of fatal cardiovascular disease in elderly men or women without diabetes: the Rancho Bernardo Study. Am J Epidemiol 1994;140:857–69. Wallace TM, Levy JC, Matthews DR. Use and abuse of HOMA modeling. Diabetes Care 2004;27:1487–95. Muniyappa R, Lee S, Chen H, Quon MJ. Current approaches for assessing insulin sensitivity and resistance in vivo: advantages, limitations, and appropriate usage. Am J Physiol Endocrinol Metab 2008;294:E15–26. Kang ES, Yun YS, Park SW, et al. Limitation of the validity of the homeostasis model assessment as an index of insulin resistance in Korea. Metabolism 2005;54:206–11. Kim DJ, Song KE, Park JW, Cho HK, Lee KW, Huh KB. Clinical characteristics of Korean type 2 diabetic patients in 2005. Diabetes Res Clin Pract 2007;77S:S252–7. Howard G, O’Leary DH, Zaccaro S, et al. Insulin sensitivity and atherosclerosis. Circulation 1996;93:1809–17. Haffner SM, D’Agostino R, Mykkänen L, et al. Proinsulin and insulin concentrations in relation to carotid wall thickness: insulin resistance atherosclerosis study. Stroke 1998;29:1498–503.