Berson and Yalow 1 defined insulin resistance (IR) as a

Transcription

1 A Study to Evaluate Surrogate Markers of Insulin Resistance in Forty Euglycemic Healthy Subjects AK Gupta*, SK Jain** Original Article Abstract Objective : Insulin resistance (IR) is increasingly being recognized as an important pathophysiological determinant of not only diabetes but also a number of other clinical states. Methods for directly estimating IR like euglycemic clamp technique and Insulin suppression test (IST) are experimentally demanding and impractical tool for large scale epidemiological studies. We evaluated several surrogate markers of IR in 40 healthy subjects. Methods : Study included 40 euglycemic normal healthy north Indian subjects (33 males, 7 females) of mean ± SD age 38.9 ± 8.6 yrs, BMI 20.5 ± 3.6 kg/m 2 and WHR 0.87 ± All subjects were tested for fasting and postprandial (2 hr post 75 gm glucose) glucose and insulin. Then all the subjects underwent IST by modified Harano s method (simultaneous infusion of 20% 6mg/kg/min and plain human 50 mu/kg/hr). Metabolic clearance rate for glucose (MCR = glucose infusion rate/ steady state plasma glucose, ml/kg/min) was calculated for min of infusion. Correlation of MCR with various surrogate markers which included fasting glucose (FG), fasting insulin (FI), fasting glucose/ insulin ratio (FGIR), 120 min glucose (PPG), 120 min insulin (PPI), 120 min glucose/ insulin ratio (PPGIR), homeostatic model assessment of insulin resistance (HOMA - IR), quantitative insulin sensitivity check index (QUICKI), fasting glucose insulin product (FIGP), insulin sensitivity index (ISI), fasting insulin resistance index (FIRI), insulin ratio (IRa) and insulinogenic index (II) were evaluated. Results : MCR was found to be significantly (p <0.05) correlated with FI (r= ), PPI (r= ), PPG (r= ), PPGIR (r= 0.356), HOMA-IR (r= ), FIGP (r= ), and FIRI (r= ). Conclusions : Among the surrogate markers which were significantly correlated to MCR, there was no significant superiority of one marker over the other. We suggest that measuring insulin levels alone in a single fasting sample can serve as a simple, cheap and convenient indirect qualitative index of IR. INTRODUCTION Berson and Yalow 1 defined insulin resistance (IR) as a state (of a cell, tissue, system, or body) in which greater than normal amount of insulin is required to elicit a quantitatively normal response. Insulin is a key regulator of glucose homeostasis and insulin resistance is determined by both genetic and environmental factors and plays an important pathophysiological role in diabetes. 2 The abnormalities initially associated with insulin resistance in nondiabetic individuals included hyperinsulinemia, borderline glucose tolerance, dyslipidemia (in the form of increased triglycerides and decreased HDL cholesterol) and hypertension. 3 The pathophysiological states associated with insulin resistance *Senior Resident; **Professor of Medicine, Department of Medicine, Lady Hardinge Medical College and Associated Hospitals, New Delhi. Received : ; Revised : ; Accepted : have continued to increase and now include enhanced postprandial lipemia, 4 small dense LDL particle, 5 high uric acid levels, 6 enhanced renal sodium retention and decreased urinary uric acid clearance, 6 higher resting heart rate, 7 dysfibrinolysis, 8 and polycystic ovarian syndrome. 9 Thus, the concept of insulin resistance has evolved from its role in pathogenesis of type 2 diabetes to attain a greater role. The hyperinsulinemic euglycemic clamp technique is the gold standard method for measuring insulin sensitivity. 10 The clamp technique is experimentally demanding and costly and has proven to be impractical tool particularly for conducting large scale crosssectional or longitudinal studies. Insulin suppression test (IST) by modified Harano s method used in this study is another method for assessing insulin resistance. 11 It has excellent correlation with euglycemic clamp technique (r= 0.83). 11 This method is much easier to perform, reliable and well tolerated and requires much simpler experimental setup. However, its application to a large number of subjects JAPI VOL. 52 JULY

2 is also problematic because of necessity of frequent blood sampling and analysis. Thus there is a pressing need to evolve indirect or surrogate markers of insulin resistance which are more applicable for large population-based epidemiological investigations. Several such markers have been proposed The utility and significance of these surrogate estimates of insulin resistance depend on the degree to which they correlate with the direct estimate of this variable. Due to extreme importance of this issue, we attempted to define the degree of correlation between a quantitative measure of insulin resistance i.e. metabolic clearance rate for glucose (MCR) determined by modified Harano s method 11 and various surrogate markers in 40 euglycemic subjects. MATERIALS AND METHODS Study was conducted in 40 normal and healthy north Indian subjects. There were 33 males and seven females subjects with mean ± SD age of 38.9 ± 8.6 years (range 22-57), BMI of 20.5 ± 3.0 kg/m 2 (range ) and WHR of 0.87 ± 0.05 (range ). The study was conducted at the Department of Medicine, Lady Hardinge Medical College and Dr. RML Hospital and Radioimmunoassay Lab, INMAS, Delhi. All subjects had a normal medical history, and routine clinical laboratory tests. Written informed consent of each subject was taken before undertaking the procedure. All subjects were tested for plasma glucose and serum insulin which were measured before (fasting) and 120 min after the ingestion of 75 g oral glucose challenge. On a separate admission, all subjects underwent IST. After an overnight fast insulin (regular, Human Actrapid, Novo 50 mu/kg/hr, and glucose as 20% 6 mg/kg/min were infused through an I/V catheter placed on one arm by a syringe infusion pump (Model Pilot A2, Fresenius, Germany) and a volumetric infusion pump (Cure Mate Model SM 2100, Jong Sang Techno Co. Ltd, Korea) respectively. Blood samples were collected at every five minutes interval between 120 to 150 min of infusion through another I/V catheter placed on the other arm. Insulin levels were measured by radioimmunoassay (Coat-A-Count Insulin kits of Diagnostic Products Corporation, US) and glucose by autoanalyser (glucose oxidase method). The mean concentration of glucose in the seven samples ( min) was calculated representing steady state plasma glucose (SSPG). Metabolic clearance rate for glucose (MCR) was calculated as rate of glucose infusion/sspg. During the procedure each subject was closely observed for signs and symptoms of hypoglycemia and other complications of intravenous infusion. Various surrogate markers of insulin resistance as reported by different authors were calculated and then correlated with MCR (Table 1). All data are expressed as mean ± SEM or mean ± SD, and all analysis were performed using the SPSS 10.0 package for windows. Pearson s correlation (r) between MCR and all surrogate measures of insulin resistance were calculated. P <0.05 was considered to indicate statistical significance. Table 1 : Various surrogate markers calculated 1. Fasting glucose (FG) mg/dl 2. Fasting insulin (FI) µu/ml (18,19) 3. Fasting glucose/insulin ratio [FGIR = FG/FI mg/10-4 U (13,19)] min glucose (PPG) mg/dl min insulin (PPI) µu/ml (19) min glucose/insulin ratio [PPGIR = PPG/PPI mg/10-4 U (13)] 7. Homeostatic model assessment of insulin resistance [HOMA-IR = FI x FG/22.5 (µu/mol/l 3 (12)]. 8. Quantitative insulin sensitivity check index [QUICKI = 1/Log(FI)+ Log (FG) (16)]. 9. Fasting insulin glucose product (FIGP) (17). 10. Insulin sensitivity index [ISI = 10 4 /FIxFG (18)] 11. Fasting insulin resistance index [FIRI = FGxFI/25 (14)]. 12. Insulin ratio [IRa = PPI/FI (18)] 13. Insulinogenic index [II = (PPI-FI)/(PPG-FG) (18)]. Table 2 : Measured values of MCR, fasting and 120 min glucose and insulin MCR (ml/kg/min) 8.6 ± 0.3 ( ) FG (mg/dl) 87.0 ± 1.6 ( ) FI ((µu/ml) 10.1 ± 1.5 ( ) PPG (mg/dl) ± 3.8 ( ) PPI (µu/ml) 25.2 ± 2.5 ( ) Data are mean ± SEM (range) Table 3 : Correlations (r) of MCR with various surrogate markers of insulin resistance Pearson s correlation (r value) p value Fasting glucose (FG) [(-) to 0.462] Fasting insulin (FI) * [(-) to (-) 0.066] Fasting glucose/insulin ratio (FGIR) [(-) to 0.575] 120 min glucose * (PPG) [(-) to (-) 0.030] 120 min insulin * (PPI) [(-) to (-) 0.134] 120 min glucose/insulin * ratio (PPGIR) [0.077 to 0.635] Homeostatic model assessment * of insulin resistance [(-) to (-) 0.067] (HOMA-IR) Quantitative Insulin sensitivity check index (QUICKI) [(-) to 0.461] Fasting insulin glucose * product (FIGP) [(-) to (-) 0.067] Insulin sensitivity index (ISI) [(-) to 0.575) Fasting insulin resistance * index (FIRI) [(-) to (-) 0.067] Insulin ratio (IR) [(-) to 0.210] Insulinogenic index (II) [(-) to 0.048] *p <0.05. Confidence interval given in parentheses JAPI VOL. 52 JULY 2004

3 (A) (B) (C) (D) (E) Fig. 1 : Relationship of MCR glucose with (A) 120 min glucose, (B) fasting insulin, (C) 120 min insulin, (D) 120 min glucose/insulin ratio, and (E) HOMA-IR. RESULTS Values of MCR and various surrogate markers of insulin resistance for the 40 subjects are shown in Table 2. Pearson s correlation (r) of MCR with these surrogate markers of insulin resistance calculated are shown in Table 3. These results demonstrated that simple Pearson s correlation coefficients between MCR, and fasting insulin, 120 min insulin, 120 min glucose, 120 min glucose/insulin ratio, HOMA-IR, FIGP and FIRI were statistically significant (p <0.05) (Fig. 1). The 120 min insulin was found to be most closely related to MCR (p= 0.01). No significant correlations of MCR were achieved with fasting glucose, fasting glucose/insulin ratio, QUICKI, ISI, insulin ratio, and insulinogenic index. Out of the various surrogate measures which are variations of a formula that involves fasting insulin and fasting glucose i.e. HOMA-IR, FGIR, QUICKI, FIGP, FIRI and ISI, only HOMA-IR, FIGP and FIRI achieved statistically significant (p <0.05) correlations with MCR glucose. The r value was similar for all of these variables (r= ). Also all of these measures, based on JAPI VOL. 52 JULY

4 fasting glucose and insulin were found to be highly significantly correlated with each other (p < 0.001). Large majority of the healthy subjects had FI less than 10 µu/ ml However, five subjects had extremely high FI more than 20 µu/ ml. Furthermore, these five subjects had higher levels of PPI (Fig. 1C) and HOMA-IR (Fig. 1E), and lower values of both fasting and PP glucose/ insulin ratio (Fig. 1D), when compared to rest of the group. DISCUSSION We studied correlations of presently available surrogate markers of IR with MCR and also compared these markers with each others. Metabolic clearance rate for glucose (MCR) was found to be statistically significantly correlated with fasting insulin, 120 min insulin, 120 min glucose, 120 min glucose/insulin ratio, HOMA-IR, FIGP and FIRI (p <0.05). The highest degree of correlation of MCR was achieved with 120 min insulin (r= ) closely followed by fasting insulin (r= 0.347). Thus both fasting and post-glucose load insulin were found to be significantly related to direct estimate of insulin resistance (MCR) with no significant difference in their correlations with MCR. Since it is very convenient to take a single fasting sample, fasting insulin as a surrogate marker of insulin resistance can serve as convenient, simple and cheap index of insulin resistance particularly suited for large scale population based studies. Fasting insulin is one of the most commonly studied surrogate marker of insulin resistance. Yeni - Komshian et al 19 studied 490 healthy non-diabetic volunteers. They derived steady state plasma glucose (SSPG) from insulin suppression test and correlated it with various surrogate measures. They reported significant correlations of SSPG with fasting insulin, 120 min insulin, insulin area under the curve, fasting glucose/ insulin ratio and HOMA-IR. Although they described total insulin response during an OGTT as the best surrogate measure, they also suggested that determination of fasting insulin can provide a qualitative estimate of insulin resistance. Similarly Markku Laakso 15 measured insulin response to an oral glucose load and quantitated insulin resistance using the euglycemic hyperinsulinemic clamp technique to evaluate the correlation between fasting and postprandial insulin levels, and the degree of insulin resistance in individuals with varying degrees of glucose tolerance. They reported that although correlations of direct estimates of insulin resistance with fasting or post-glucose load insulin levels were consistent in subjects with normoglycemia, in subjects with impaired glucose tolerance and diabetes, only the fasting insulin correlated significantly with insulin resistance. They suggested that in population studies, only the fasting insulin level should be used as a marker of insulin resistance. Similarly Hanson RL et al 18 achieved high degree of correlation of fasting and post-glucose load insulin with direct measure of insulin resistance as determined by hyperinsulinemia euglycemic clamp. Also in our study, fasting insulin was found to be highly significantly (p <0.001) correlated with other markers e.g. post glucose load insulin (r= 0.564), fasting glucose/ insulin ratio (r= 0.728), 120 min glucose/insulin ratio (r= 0.419), HOMA-IR (r= 0.988), QUICKI (r= 0.712), ISI (r= 0.728), FIGP (r=0.988), FIRI (r=0.988) and IRa (r= 0.428, p <0.01). No significant correlation was observed between MCR and fasting glucose/ insulin ratio although significant correlation was achieved with 120 min post glucose load glucose/insulin ratio (p <0.05). Yeni - Komshian et al 19 and Legro RS et al 13 described significant correlations with both fasting and post-glucose load glucose/insulin ratio though the correlations with the latter were much significant. We also observed statistically significant correlations (p <0.05) of MCR with different markers based on fasting glucose and insulin e.g. HOMA-IR, FIGP and FIRI. Correlation coefficients (r value) were exactly similar for all of them (r= ). It can be explained on the basis that all of these markers are based on product of fasting glucose and insulin, and thus are no different or superior to each other. This is in variance with what has been reported by different authors. 14,17 In fact HOMA-IR, FIGP and FIRI were found to be highly correlated (r= 1.0) with each other. Katz et al 16 defined a quantitative insulin sensitivity index (QUICKI) which is also based on fasting glucose and insulin but transforms the data by taking both the logarithm and reciprocal of the glucose insulin product. They claimed that QUICKI holds good correlation with direct measures of insulin sensitivity derived from two different methods i.e., euglycemic hyperinsulinemic clamp and frequently sampled intravenous glucose tolerance test (FSIVGTT). However, this variable correlated poorly and insignificantly with MCR in our study. We could not demonstrate any significant correlation with measures like insulin sensitivity index, insulin ratio and insulinogenic index which is in variance with the work of Hanson RL et al. 16 Also in an interesting observation, five subjects had extremely high FI ( >20 (µu/ ml ) associated with higher PPI, higher HOMA-IR, and low levels of both fasting and PP glucose/insulin ratio as compared to rest of the subjects. These subjects are otherwise normal and healthy. They may belong to syndrome X. Other manifestations of the syndrome like hypertension, diabetes, coronary artery disease, dyslipidemia which although are not apparent at present may appear later in the life. In summary, we evaluated almost all of the presently reported surrogate markers of insulin resistance in a single study and could demonstrate significant correlations of quite a few of them with direct measure of insulin resistance i.e., MCR derived from modified Harano s method. We conclude that simply measuring fasting insulin levels may serve as a simple, cheap and reliable qualitative marker of insulin resistance, and measuring other variables do not offer any significant advantage. REFERENCES 1. Berson SA, Yalow RS. In: Ellenberg M, Rifkin H (eds) Diabetes Mellitus : Theory and Practice, New York: McGraw-Hill, 1970; DeFronzo RA, Bonadanna RC, Ferrannini F. Pathogenesis of JAPI VOL. 52 JULY 2004

5 NIDDM. A balanced overview. Diabetes Care 1992;15: Reaven GM. Role of insulin resistance in human disease. Diabetes 1988;37: Jeppesen J, Hollenbeck CB, Zhou M-Y, et al. Relationship between insulin resistance, hyperinsulinemia, post heparin plasma lipoprotein lipase activity, and postprandial lipemia. Arterioscler Thromb Vasc Biol 1995;15: Reaven GM, Chen Y-DI, Jeppesen J, Maheux P, Krauss RM. Insulin resistance and hyperinsulinemia in individuals with small, dense, low density lipoprotein particles. J Clin Invest 1993;92: Facchini F, Chen YD-I, Hollenbeck C, Reaven GM. Relationship between resistance to insulin mediated glucose uptake, urinary uric acid clearance, and plasma uric acid concentration. JAMA 1991;266: Facchini FS, Riccardo A, Stooh s A, Reaven GM. Enhanced sympathetic nervous system activity : the linchpin between insulin resistance, hyperinsulinemia, and heart rate. Am J Hypertens 1996;9: Juhan-Vague I, Thompson SG, Jespersen J, on behalf of the ECAT Angina Pectoris Study Group; Involvement of the hemostatic system in the insulin resistance syndrome. Arterioscler Thromb 1993;13: Barbicri RI, Ryan KJ. Hyperandrogenism, insulin resistance and acanthosis nigricans syndrome : a common endocrinopathy with distinct pathophysiology features. Am J Obstet Gynecol 1983;147: DeFronzo RA, Tobin JD, Andres R. Glucose clamp technique : a method for quantifying insulin secretion and resistance. Am J Physiol 1979;273:E Heine RJ, Home PD, Poncher M, et al. A comparison of 3 methods for assessing insulin sensitivity in subjects with normal and abnormal glucose tolerance. Diabetes Res 1985;2: Mathews DR, Hosker JP, Rudenski AS, Naylor BA, Treacher BA, Turner RC. Homeostasis model assessment : Insulin resistance and β-cell function from fasting plasma glucose and insulin concentrations in man. Diabetologia 1985;28: Legro RS, Fine Good D, Dunaif A. A fasting glucose to insulin ratio is a useful measure of insulin sensitivity in women with polycystic ovary syndrome. J Clin Endocrinol Metab 1998;83: Duncan MH, Singh BM, Wise PH, Carter G, Alaghband ZJ. A simple measure of insulin resistance. Lancet 1995;346: Laakso M. How good a marker is insulin level for insulin resistance? Am J Epidemiol 1993;137: Katz A, Nambi SS, Mather K, et al. Quantitative insulin sensitivity check index: A simple, accurate method for assessing insulin sensitivity in humans. J Clin Endocrinol Metab 2000;85: Thomas GN, Critchley J, Tomlinson B, Anderson PJ, Lee ZS, Chan JC. Obesity, independent of insulin resistance, is a major determinant of blood pressure in normoglycemic Hong Kong Chinese. Metabolism 2000;49: Hanson RL, Pratley RE, Bogardus C, et al. Evaluation of simple indices of insulin sensitivity and insulin secretion for use in epidemiologic studies. Am J Epidemiol 2000;151: Yeni - Kom Shian H, Carantoni M, Abbasi F, Reaven GM. Relationship between several surrogate estimates of insulin resistance and quatification of insulin mediated glucose disposal in 490 healthy non-diabetic volunteers. Diabetes Care 2000;23: Announcement XXIV Annual Conference of Association of Physicians of India, Orissa Chapter will be held on 13th and 14th November, 2004 at IMA House - Berhamper, Orissa. For further details, please contact : Dr. LM Meher, Organising Secretary, Conference Secretariat of Medicine Dept. MKCG Medical College, Berhampur, Orissa Tel ; lkmeher@yahoo.com JAPI VOL. 52 JULY