Tres versiones del "Ogtt Squeezer", un programa para evaluar de un vistazo la información metabólica de la curva de tolerancia a la glucosa / Three versions of the "Ogtt Squeezer", a program to evaluate ataglance the metabolic information of the oral glucose tolerance test

El diagnóstico clínico de resistencia insulínica (RI) es difícil, ya que el Clamp no es aplicable a la clínica. El así llamado "síndrome metabólico", un predictor clínico de la RI, no identifica alrededor de la mitad de los sujetos afectados. Previamente, definimos adecuadamente (Análisis ROC) los niveles de corte diagnóstico de los siguientes predictores bioquímicos: HOMA1, HOMA2, QUICKI e ISI-Composite, a través de analizar datos de 90 sujetos (53 no resistentes y 37 resistentes) que tenían una medición directa de su resistencia insulínica (Test de supresión pancreática, TSP, Test de Reaven) y también, una curva de tolerancia a la glucosa oral (CTG). Los puntos de corte obtenidos exhibieron un mucho mejor desempeño diagnóstico comparados con los puntos de corte convencionales. También encontramos un predictor nuevo, simple, económico y eficiente, el I0*G60. Definimos la "normalidad metabólica" de la CTG usando las medianas de los valores de varios parámetros en 312 sujetos con un G120 dentro de los 2 primeros terciles del grupo de normo-tolerantes a la glucosa (NGT, n=468; G120: 51-110 mg/dL, los con mejor función beta insular). A las medianas de la función beta insular y de la sensibilidad insulínica se les asignó un valor de un 100%. Se calculó el % relativo de función beta insular (%RFBI) y el % relativo de sensibilidad insulínica (%RSI) del resto de la cohorte (n=573) contra estos valores de referencia. El "OGTT Squeezer" se escribió en Excel. Las glicemias y las insulinemias de la CTG fueron las entradas del programa. Las salidas fueron: I0*G60, ISI-OL, QUICKI, and HOMA1 (predictores) y el índice insulinogénico, el índice de disposición, %RFBI y %RSI (parámetros). El programa también caracterizó la tolerancia glucídica de acuerdo a los criterios de la ADA 2003. El formato final del programa, HTML 5, facilita su uso. Desarrollamos tres versiones del programa: completa, abreviada y mínima.

Clinically, diagnosing insulin resistance (IR) is difficult since the Clamp is not applicable to clinical work. The so-called "Metabolic Syndrome", a clinical surrogate of IR, fails to identify around 50% of affected subjects. Previously, we properly defined (ROC Analysis) the diagnostic cut-offs of the following biochemical predictors: HOMA1, HOMA2, QUICKI, and ISI-Composite by analyzing data from 90 subjects (53 non-insulin-resistant and 37 insulin-resistant subjects) who had a direct measurement of insulin resistance (Pancreatic Suppression Test, PST, Reaven's Test), and also, an Oral Glucose Tolerance Test (OGTT). The resulting cut-offs exhibited much better performances compared with the conventional cut-offs. We also found a new, simple, inexpensive and efficient predictor, the I0*G60. We chose to define the "metabolic normalcy" of the OGTT by using the median values of several parameters in 312 NGT subjects with a G120 in the first 2 tertiles of the NGT group (n=468; G120: 51-110 mg/dL, those with the best beta-cell function). The median values of both Beta-Cell Function and Insulin Sensitivity of these subjects were assigned a 100% value. Both % Relative Beta-Cell Function (%RBCF) and % Relative Insulin Sensitivity (%RIS) of everyone else in the cohort (n=573) was calculated against these reference values. The "OGTT Squeezer" was written in Excel. The OGTT's glucose and insulin values served as the inputs of the program. The outputs were: I0*G60, ISI-OL, QUICKI, and HOMA1 (predictors), and Insulinogenic Index, Disposition Index, %RBCF, and %RIS (parameters). Moreover, the program characterized the OGTT according to the ADA 2003 criteria. The HTML 5 format of the program facilitates its use. We developed 3 versions of the program: complete, abbreviated, and minimal versions.

Comparison of insulin resistance scoring system with indirect methods of estimating insulin resistance in Indian type 2 diabetes subjects

Background: Identification of insulin resistance is very important in management of type 2 diabetes. The euglycemic insulin clamp method, intravenous glucose tolerance tests (IVGTT) and minimal model approximation of glucose (MMAMG) are standard methods of measurement of insulin resistance in research. However, they are impractical in clinical practice and are difficult to perform in population-based research studies. So, a simple scoring system was designed to estimate the insulin resistance. Methods: 200 type 2 diabetes individuals who attended Karnataka Institute of endocrinology and research outpatient department. Fasting plasma glucose, post prandial plasma glucose, fasting insulin, lipid profile, BMI, waist circumference and BP of these subjects were checked. Results: Out of 200 type 2 diabetes subjects 69.5% were males and age group ranging from 26 to 85 years. Duration of diabetes range from 0 to 20 years and 53% of patients had hypertension and 46.5% have hypertriglyceridemia. Insulin resistance calculated by KIER scoring system, HOMA-1, QUICKI, HOMA2 and Fasting Insulin was present in 82%, 63%, 63.5%, 33.5% and 37.5% 0f individuals respectively. KIER scoring system had a statistically significant correlation with HOMA and QUICKY indices. (P value < 0.001) Conclusions: (1) KIER scoring system detects insulin resistance in 82% of type 2 diabetes individuals. (2) HOMA 1 and QUICKI are identical and similarly HOMA 2 and fasting insulin levels are almost identical in estimation of insulin resistance. (3) The KIER scoring system designed is very simple and economical. It takes into consideration the different factors which contribute to insulin resistance.

Prevalence of Insulin Resistance among Cigarette Smokers in Sokoto Metropolis

Introduction: Cigarette smoking (CS) is a well-known risk factor for the development of metabolic diseases, various forms of cancer as well as insulin resistance (IR). IR is considered as an underlying derangement which very commonly aggravates metabolic syndrome. Aim: This study assessed the prevalence of IR in cigarette smokers in Sokoto metropolis using selected surrogate markers. Methodology: This cross sectional study was conducted in Sokoto among 108 subjects. Fasting venous blood samples were collected for plasma glucose, triglycerides and insulin estimation. Plasma glucose and serum triglycerides were analysed using enzymatic methods while insulin was assayed using ELISA method. Homeostasis model of assessment-IR (HOMA-IR), Quantitative insulin sensitivity check index (QUICKI), Mc Auley (McA) and fasting IR index (FIRI) were calculated using standard formula and IR cut-off of >2.5, <0.339, >5.8 and >2.3 respectively were used. Results: Based on the cut off mark, the prevalence of IR for HOMA-IR, QUICKI, McA, FIRI indices were 62(57.4%), 66(61.1%), 39(36.1%) and 60(55.6%) respectively. There was a significant correlation between HOMA-IR and FIRI (p< 0.05, r = 0.999). HOMA-IR also had a significant correlation with McA (p<0.05 r = -0.506). QUICKI had a significant correlation with McA (p<0.05 and r = 0.243). Conclusion: This study established a significantly high prevalence of IR among CS. Importantly, it can be concluded that cigarette smokers may be predisposed to the development of metabolic disease.

Valores de referencia de HOMA-IR y QUICKI durante el embarazo en mujeres mexicanas / HOMA-IR and QUICKI reference values during pregnancy in Mexican women

Resumen: OBJETIVO: determinar los valores de referencia del Homeostatic Model Assessment Insulin Resistance (HOMA-IR) y Quantitative Insulin Sensitivity Check Index (QUICKI) para establecer el diagnóstico de resistencia a la insulina en mujeres mexicanas no embarazadas y embarazadas, por trimestre de gestación. MATERIALES Y MÉTODOS: estudio transversal al que se incluyeron mujeres embarazadas y no embarazadas sin alteraciones concomitantes, mayores de18 años de edad, índice de masa corporal pregestacional entre 18.5-24.9 kg/m2. A todas las participantes se les realizó la curva de tolerancia a la glucosa oral de 75 g-2h. Se excluyeron las mujeres con diabetes gestacional o cualquier alteración pregestacional, índice de masa corporal pregestacional menor de 18.5 o más o menos mayor de 25 kg/m2 y embarazo múltiple. Se calcularon los percentiles 5 y 95 como valores de referencia para definir resistencia a la insulina por HOMA-IR y QUICKI en mujeres sin embarazo y en cada trimestre del embarazo. Resultados: se incluyeron 400 mujeres, agrupadas de la siguiente forma: Grupo de mujeres sin embarazo (SE): n=42, grupo trimestre (T) 1: n=82, grupo T2: n=159 y grupo T3: n=117. Los valores de referencia de HOMA-IR para el percentil 5 y 95 fueron: 0.33-2.6, 0.35-1.6, 0.40-2.9 y 0.38-2.6 y para QUICKI: 0.33-0.46, 0.35-0.46, 0.32-0.45 y 0.33-0.45, para los grupos SE, T1, T2 y T3, respectivamente. Conclusión: el valor de referencia de HOMA-IR para establecer el diagnóstico de resistencia a la insulina en mujeres mexicanas no embarazadas (SE) es ≥ 2.6 y en pacientes embarazadas por trimestre: T1 ≥1.6, T2 ≥2.9 y T3 ≥2.6; respecto de QUICKI, los valores de referencia son SE <0.33, T1 <0.35, T2 <0.32 y T3 <0.33, respectivamente.

Abstract: OBJECTIVE: To determine the reference values of Homeostasis Model Assessment Insulin Resistance, (HOMA-IR) and Quantitative Insulin Sensitivity Check Index (QUICKI) to define insulin resistance (IR) in women without pregnancy (WP) and each trimester of pregnancy among Mexican women. METHODS: Cross-sectional study, women without pregnancy and pregnant women age >18 years, without pathologies, pre-pregnancy body mass index (BMI) between 18.5-24.9 kg/m2 were included. All participants underwent CTOG 75gr-2h to rule out diabetes. We excluded women with gestational diabetes or any pre-pregnancy pathology, pre-pregnancy BMI <18.5 or ≥25 kg/m2 and multiple pregnancy. Percentiles 5 and 95 were calculated as reference values to define RI by HOMA-IR and QUICKI in women without pregnancy and each trimester of pregnancy. RESULTS: A total of 400 women were included, which were grouped as follows: Group of women without pregnancy (SE): n = 42, quarter Group (T) 1: n = 82, T2 Group: n = 159 and T3 group: n = 117. The reference values of HOMA-IR for the 5th and 95th percentile were: 0.33-2.6, 0.35-1.6, 0.40-2.9 and 0.38-2.6 and QUICKI: (0.33 to 0.46, 0.35 to 0.46, 0.32 to 0.45 and 0.33- 0.45, for groups SE, T1, T2 and T3, respectively. CONCLUSION: The reference value of HOMA-IR to define RI in Mexican women should be ≥2.6 and the T1 ≥1.6, T2 pregnancy: ≥2.9 and T3 ≥2.6 and QUICKI in women <0.33, T1 <0.35, T2 <0.32 and T3 <0.33.

Insulin Resistance: Detection By Fasting Insulin In North Indian Obese Adults.

Background&objectives:Several studies have shown that obesity is closely related to Insulin resistance(IR). Insulin resistance has been suggested as a primary cause for metabolic syndrome. Identifying such individuals would help to prevent progression of comorbidities associated with IR. Hence present study was planned to assess the importance of Fasting Insulin(FI) as a measure of IR and to analyze its correlation with other indirect methods for the assessment of IR . Methods: Study was conducted in fifty obese and overweight subjects.Body Mass Index of all subjects was calculated . Blood glucose, and FI were assayed after twelve hours of fasting. Homeostasis model assessment (HOMA) and Quantitative insulin sensitivity check indices (QUICKI) were calculated. Results: Present study showed that 90% of subjects had IR by HOMA and QUICKI.Correlation of FI with HOMA and QUICKI was statistically significant (P < 0.05). FI test had significant sensitivity and specificity when compared with HOMA and QUICKI indices. Validity of FI was further analyzed by Cohen’s kappa test and had good agreement (κ =0.67). Conclusion: FI was sensitive and also specific as HOMA and QUICKI in assessment of IR in obese. Thus, FI can be used as a simple test and feasable tool to detect IR in obese subjects. [AnshuKhatri NJIRM 2016; 7(5):1-4]

Valores de insulina basal y post carga de glucosa oral, homa-ir y quicki, en niños y adolescentes de la ciudad de Mérida, Venezuela: Influencia del sexo y estadio puberal (estudio credefar)

Objetivo: Obtener valores de insulina basal, post carga de glucosa oral, HOMA-IR y QUICKI, y establecer su relación con el sexo y el estadio puberal en niños y adolescentes sanos del Municipio Libertador de la ciudad de Mérida, Venezuela. Materiales y métodos: Participaron 922 sujetos de 9 a 18 años de instituciones públicas y privadas. Se recogieron datos antropométricos. Se determinaron glucemia e insulina (Quimioluminiscencia) en ayunas y 2 horas post carga de glucosa oral. Se calcularon los índices HOMA-IR y QUICKI. Se realizó la distribución percentilar de las variables por sexo y estadio puberal. Resultados: El 51,6% eran de sexo femenino y el 48,4% masculino; el 52,7% de instituciones públicas y el 47,3% de privadas. El 20,4% era prepúber, el 18,3% se encontró en el estadio II de Tanner, el 11,8% en el III, el 16% en el IV y el 33,4% en estadio puberal V. Los valores de insulina y HOMA-IR fueron mayores en el sexo femenino. Los valores más altos se observaron en los estadios II, III y IV. La insulina 2 horas postcarga mostró sus máximos valores en los estadios IV y V. En nuestra población, se proponen valores altos (>pc95) de insulina en ayunas aquellos mayores de 9 mU/mL en el prepúber y de 12 mU/mL en el púber; de insulina 2 horas postcarga mayores de 35 mU/mL en el prepúber y de 65 mU/mL en el púber; de HOMA-IR mayores de 2 en el prepúber y de 2,5 en el púber. Se consideran disminuidos aquellos valores de QUICKI menores de 0,31 (

Objective: To obtain values of fasting insulin levels and 2 hours post glucose oral test, values of HOMA-IR and QUICKI and its association with sex and pubertal stages in healthy children and adolescents from the Libertador Municipality of Mérida, Venezuela. Material and methods: We evaluated 922 students between 9 to 17,9 years from publics and privates educational institutions. Anthropometric variables were taken. Fasting glucose and insulin and 2 hours post glucose oral test were measured, and the HOMA and QUICKI indexes were calculated. The percentile distribution of the studied variables according to sex and pubertal stages was performed. Results: The 51.6% were female and 48.4% male; 52.7% were from public and 47.3% from private institution. According to Tanner Stages: 20.4% were Tanner I, 18.3% Tanner II, 11.8% Tanner III, 16% Tanner IV and 33.4% were Tanner V. Values of insulin levels and HOMA were higher in female sex. The higher levels were observed in stages II, III and IV. Insulin 2 hours post oral glucose test levels showed the highest values in Tanner stages IV and V. To our population, we propose more than 9 mU/mL as a high value (>pc95) of fasting insulin in the pre-pubertal stage and 12 mU/mL in pubertal stage; insulin 2 hours post glucose oral test higher than 35 mU/mL in the pre-pubertal and 65 mU/mL for the pubertal stage; HOMA-IR higher than 2 in the pre-pubertal and 2.5 in the pubertal stage. QUICKI levels under 0.31 (

TG/HDL-C Ratio: A Surrogate Marker of Insulin Resistance in Patients with Metabolic Syndrome.

Insulin resistance (IR) is hallmark of metabolic syndrome. It is important to identify IR as it is the early stage before development of diabetes mellitus. The standard method to measure insulin resistance is the euglycemic clamp technique, which is laborious. Hence, a number of surrogate measures like homeostasis model assessment of insulin resistance (HOMA-IR), quantitative insulin sensitivity check index (QUICKI) and triglyceride/high density lipoprotein cholesterol (TG/HDL-C) ratio have been developed. Both of the former involve calculations, while TG/HDL ratio may be readily available for clinicians. Therefore, this study was undertaken to assess whether TG/HDL-C ratio serves as a better predictive marker of IR. Objectives: The aim of the present study was to evaluate the triglyceride/HDL-C ratio as a surrogate marker of IR in metabolic syndrome patients. Materials and methods: Total 110 patients were recruited in the study after obtaining informed written consent. They were divided into two groups. Group I included healthy controls (n = 50) and subjects with metabolic syndrome (MS) (n = 60) as per NCEP ATP III criteria were included in group II. Anthropometric measurements and biochemical analysis was performed in all subjects. Results: There was statistically significant difference in anthropometric, glycemic and lipid parameters in control and study group (p < 0.0001). The regression model between HOMA-IR and TG/HDL-C ratio showed was positive correlation (r = 0.29, p = 0.01) while between QUICKI and TG/HDL-C ratio showed negative correlation (r = –0.37, p = 0.002). Conclusion: We report in our study that TG/HDL-C can be adopted in routine laboratory practice as a surrogate marker for prediction of insulin resistance. So that patients with metabolic syndrome may be beneficial at an early stage.

Índices de sensibilidad insulínica (homa y quicki): en escolares y adolescentes sanos en Valera: estado Trujillo-Venezuela / Insuline sensibility index among healthy children and adolescents in Valera: Trujillo state-Venezuela

La resistencia a la insulina contribuye a la fisiopatología de la diabetes tipo 2 y es la antesala de la obesidad, el síndrome metabólico y muchas enfermedades cardiovasculares, de allí la importancia de su detección temprana. Evaluar el grado de sensibilidad insulinica mediante los índices HOMA y QUICKI y la asociación de la insulinosensibilidad basal con algunas variables biológicas (edad, sexo, estado nutricional). Se realizó un estudio descriptivo de corte transversal en escolares y adolescentes entre 6 y 18 años, entre marzo y julio de 2005. Se calculó el IMC y se emplearon las curvas percentiles de FUNDACREDESA para su categorización. Se extranjeron 5mL de sangre para detectar los valores de glucosa e insulina. La sensibilidad insulínica basal se calculó mediante los índices HOMA [glicemia en ayunas (mmol/l) x insulina en ayunas (mU/l)]/22,5 y QUICKI (1/(Log glicemia ayuno (mg/dl)+Log insulina ayuno (µU/ml). Se calcularon los estadísticos descriptivos y las diferencias fueron estudiadas mediante la prueba Chi², considerando significativo a todo valor de p<0,05. 269 niños tenían peso normal (84,5 por ciento), 29 presentaron sobrepeso (9,11 por ciento) y 20 eran obesos (6,28 por ciento). No se encontró asociación significativa entre la sensibilidad insulínica y el estado nutricional. Hubo diferencia significativa en las glicemias en ayunas entre escolares y adolescentes eutróficos y con sobrepeso (p<0.001). No hubo insulino resistencia en escolares ni adolescentes. El índice HOMA alcanzó un valor cercano a 1, y el índice de QUICKI se mantuvo alrededor de 0,40 independientemente del estado nutricional.

Insulin resistance contributes to the physiopathology of diabetes and is the previous step to obesity, metabolic syndrome and many cardiovascular disease, therefore the importance of its early detection. To evaluate the degree of insulin sensitivity by means of the indexes HOMA and QUICKI and the association with some biological variables (age, gender, nutritional state). The study is descriptive of transverse cut type and included children and adolescents between 6 and 18 years. Body Mass Index was calculated and the categorization was performed by Fundacredesa's charts. Glucose and insulin were measured in blood. Basal insulin sensitivity was calculated by means of the indexes HOMA, [blood fasting sugar (mmol/1) x fasting insulin (mU/1)]/22,5 and QUICKI (1/(Log fasting blood sugar (mg/dl) + Log fasting blood insulin (µU/ml). Descriptive statistic were calculated and differences were studied by means of the test X², considering as significant, values under 0.05. There were 269 children with normal weight (84,5%), 29 were overweight (9,11%) and obese (6,28%). There was no significant association between insulin sensitivity and nutritional state. There were significant differences in fasting blood sugar between normal and overweight children and adolescents (p<0,001). There were neither children nor adolescents with insulin resistance the HOMA index reached a value of 1 and the QUICKI index was near 0,40 independently of the nutritional state.

Homair, quicki y leptina en adolescentes deportivas

Objetivos : Evaluar la resistencia y la sensibilidad a la insulina a través de los índices HOMAIR y QUICKI en adolescentes deportistas y establecer la relación entre estos índices, la grasa corporal, la leptina y el perfil lipídico. Métodos: Se estudiaron 52 deportistas: 39 varones y 13 hembras en edades comprendidas entre 14 y 17 años, en estadio IV de Tanner. Se tomaron las medidas antropométricas: talla, peso y se calculó el índice de masa corporal (IMC). Se tomó muestra de sangre para glucemia, insulina, leptina, triglicéridos (Tg), colesterol total (CT), colesterol LDL y colesterol HDL. Se calcularon los índices HOMAIR y QUICKI. Resultados: Las concentraciones de glucemia e insulina no mostraron diferencias entre los dos grupos. El valor promedio del HOMAIR fue de 1,11±0,52 en los varones y de 1,02±0,43 en las hembras; el promedio del QUICKI fue de 0,384±2,68 en los varones y de 0,388±2,46 en las hembras. Los valores de leptina fueron significativamente mas altos (p<0,001) en el sexo femenino. Los niveles de lípidos sanguíneos se encontraron dentro del rango normal para la edad, con discreto aumento de triglicéridos en el grupo masculino y del colesterol en el grupo femenino pero sin significancia estadística. En ambos sexos el IMC se correlacionó positivamente con la concentración de leptina y negativamente con la sensibilidad a la insulina. Los triglicéridos se relacionaron positivamente con el IMC en el sexo masculino. Conclusiones: Los valores de HOMAIR y de QUICKI en adolescentes deportistas son similares a los obtenidos en adultos con IMC normal. Se corrobora el dimorfismo sexual en la concentración de leptina. La relación negativa entre el índice QUICKI y el IMC sugiere una mayor sensibilidad de dicho índice en relación con adiposidad corporal.

Objective : Evaluate the insulin resistance and insulin sensitivity indexes in sport-players adolescents and, examine the relationship between these two indexes with body mass index, lipids and leptin serum levels. Methods: We studied 52 Tanner IV stage sport-players adolescents: 39 males and 13 females with ages between 14 and 17 years. Height, weight and body mass index (BMI) were registered. Fasting glucose, insulin, leptin, triglycerides (Tg), total cholesterol CT, LDL-C, and HDL-C levels were measured. HOMA IR and QUICKI indexes were calculated. Results : No significant differences in glycemia and insulin serum levels were found. HOMA IR mean value was 1.11 ± 0.52 for males and 1.02 ± 0.43 for girls. QUICKI mean value was 0.384 ± 2.68 for boys, and 0.388 ± 2.46 for girls. Leptin serum levels were significantly higher in girls group (p<.001). Plasma triglycerides levels were non significantly higher in males, similar tendency was observed in total cholesterol in female group. In both genders the IMC index was positively correlated with leptin serum concentration, and negatively with to insulin sensitivity index. Triglycerides were positively correlated with IMC index in boys group. Conclusions: HOMA IR and the QUICKI index values in sport-players adolescents seem to be similar to those reported in adults with normal BMI. Leptin sexual dimorphism was clearly observed. The negative correlation between the QUICKI index and IMC might suggests this index is a good insulin sensitivity marker related to ponderosity.

Indices of Insulin Resistance in Children with Simple Obesity / 대한소아내분비학회지

PURPOSE:We performed this study to compare correlation between the indices of insulin resistance using fasting insulin and glucose level and body mass index (BMI), and to determine the clinical usefulness of glucose/insulin ratio (G/I ratio), which is easily available in clinical base. METHODS:Total 119 children with simple obesity, whose BMI is over 95th percentile, were evaluated. We calculated G/I ratio, logInsulin, HOMA-IR, logHOMA-IR, and QUICKI and evaluated their relationship to BMI. RESULTS:Children with high-degree obesity had higher insulin resistance than children with mild to moderate-degree obesity (logInsulin, 1.13+/-.23 vs 1.27+/-.29; logHOMA-IR, 0.46+/-.24 vs 0.61+/-.30; QUICKI, 0.33+/-.03, 0.31+/-.03)(P<0.01), and pubertal children had higher insulin resistance than prepubertal children (G/I ratio, 7.39+/-.07 vs 4.85+/-.29; logInsulin, 1.14+/-.27 vs 1.31+/-.22; logHOMA-IR, 0.47+/-.28 vs 0.65+/-.22; QUICKI, 0.33+/-.03 vs 0.31+/-.02) (P<0.001). BMI had correlation coefficient as -0.436 for QUICKI, -0.432 for G/I ratio, 0.430 for logInsulin, and 0.425 for logHOMA-IR (P=0.000). G/I ratio was well correlated with QUICKI (r=0.901, P=0.000), logHOMA-IR (r=-0.865, P=0.000), and logInsulin (r=0.899, P=0.000). The changes of BMI were correlated with changes of G/I ratio (r=-0.547, P<0.01), QUICKI (r=-0.464, P=0.01), and logHOMA-IR (r=0.429, P<0.05). CONCLUSION: This study revealed that the degree of BMI had statistically significant correlation with insulin resistance, which can be reflected by G/I ratio, logHOMA-IR and QUICKI. G/I ratio was well correlated with logHOMA-IR and QUICKI, which suggests that G/I ratio could be used as an bedside index of insulin resistance. The changes of G/I ratio were more correlated with changes of BMI than those of logHOMA-IR and QUICKI.