In contrast, insulin sensitivity indices that are solely based on fasting parameters reflect mostly hepatic insulin sensitivity [ 8 ]. Most indices of insulin sensitivity and some for insulin secretion have already been evaluated for correlation with gold standard measurements [ 9 , 10 ].

Certainly, there is an inherent biological and analytical variability in every measurement [ 11 , 12 ]. In addition to the desired feature to be measured, glycemic traits are highly influenced by other confounding factors such as quality of sleep, subclinical conditions or inflammation [ 13 , 14 ].

Glycemic variables obtained during an OGTT are further impacted by the speed of gastric emptying [ 15 ], rate of incretin release [ 16 ] and potential interactions with the gut microbiome. Therefore, test-retest reliability, i.

the intra-individual concordance of subsequent measurements, can strongly vary among these indices. Though, there is still lack of scientific data regarding this variability.

Our work therefore aims to compare the reproducibility for different frequently reported indices for insulin sensitivity and insulin secretion using highly standardized repeated measurements within the same subjects.

We analyzed data from persons who underwent two or more oral glucose tolerance tests OGTT with a time difference of 4 to days. Subjects analyzed in the current study are participants of a number of clinical studies. The research took place in Tuebingen, Baden-Wurttemberg, Germany, between August and March Written informed consent was obtained from all participants.

A vote of approval for all studies has been obtained from the ethics committee of the University Hospital of Tuebingen. For persons who underwent more than two oral glucose tolerance tests, the two tests with the lowest time interval were analyzed. Data of 89 persons in accordance with OGTTs remained.

All participants underwent an OGTT with a standardized 75 g glucose solution Accu-Check Dextro, Roche after overnight fasting. Venous plasma and serum samples were obtained before ingestion at minute 0 and 30, 60, 90 and minutes after the glucose challenge. All blood samples were immediately put on ice and the serum was centrifuged within two hours.

Plasma glucose was determined in an ADVIA autoanalyzer Siemens Healthcare Diagnostics, Erlangen, Germany. Plasma insulin and C-peptide were measured by an immunoassay with the ADVIA Centaur XP Immunoassay System Siemens Healthineers, Eschborn, Germany.

NEFA concentrations were measured enzymatically WAKO Chemicals, Neuss, Germany using the ADVIA analyzer Siemens Healthcare Diagnostics, Eschborn, Germany. C-reactive protein levels were measured via immunoturbidimetry with the wide range reagents using the ADVIA analyzer Siemens, Eschborn, Germany.

We used the following common indices for insulin sensitivity and insulin secretion for our study. The indices used are without any claim to completeness. Calculated indices to measure insulin sensitivity using parameters from fasting state are :. Calculated indices to measure insulin secretion using parameters from fasting state are :.

Calculated indices to measure insulin secretion using parameters from oral glucose tolerance tests are :.

It is important to emphasize that indices based on insulin concentration should be differentiated from indices based on C-peptide concentration. The different indices with the respective necessary measurements are summarized in Table 1. The coefficient of variation is a well-established method for assessing the imprecision of a test.

To also account for the capability of a given test to discriminate different subjects in relation to the robustness of within subject measurements, we also calculated the discriminant ratio DR as proposed Levy et. al [ 32 ] as follows:.

Where MS B and MS W are the between-subject and within subject mean squares respectively and k is the number of repeated measurements within subject.

The discriminant ratio is higher when the differences between subjects are higher and within-subjects are lower. Confidence intervals for discriminant ratios were computed using non-central F distributions and differences between the discriminant ratios in the index groups were tested using Q-statistic, as recommended [ 32 ].

All computations were performed with R version 3. To assess how often different indices are used, we researched the literature via the ISI web of knowledge using the search function title. As key words we used the full title of each original article. The original articles are cited in the upper part see section Methods -Indices.

We looked for the number of citations of the original articles of each investigated index, where possible. All studies found were included. The literature research was conducted on 26th of February Unfortunately, it was not possible to differentiate various HOMA2 indices in our literature research.

Due to difficulties in identifying indices containing areas under curve for C-peptide, insulin and glucose in the literature, it was not possible to assess the citation number for these indices.

We analyzed data of OGTTs from 89 persons 51 females, 38 males. For each of them, data from two OGTTs with a median of 86 days IQR 64— between the two measurements were analyzed. Participants had a median age of 42 years IQR 29—57 and a median body mass index BMI of In Table 3 the mean and the standard deviation of the coefficient of variation as well as the discriminant ratio for each calculated index arranged by insulin secretion or insulin sensitivity as well as by the sampling method fasting or OGTT are shown.

Moreover, the mean per group is shown here. For the indices derived from fasting blood that assess insulin sensitivity , the smallest mean coefficient of variation was found for revised QUICKI at 2.

With 4. For indices that use measurements from oral glucose tolerance tests to measure insulin sensitivity , the smallest coefficient of variation was found for OGIS with 6.

With a coefficient of variation of The highest discriminant ratio was found for NEFA-ISI 2. With 1. For the indices for insulin secretion that use fasting variables , the lowest coefficient of variation was found for HOMA2-B C-Peptide with Comparing the indices using OGTT-based variables to assess insulin secretion , the smallest coefficient of variation at The different coefficients of variation of AUCs are shown in Table 3.

The largest coefficient of variation for measuring insulin secretion from OGTT-derived variables was demonstrated for IGI with We have also performed a sensitivity analysis of the coefficients of variation after excluding female premenopausal participants see S2 and S3 Tables. Taking both, the coefficient of variation and the discriminant ratio into account, CGR and HOMA2-B C-Peptide seem to be superior to HOMA2-B Insulin in the group assessing insulin secretion using fasting values.

For indices assessing insulin sensitivity using fasting variables, HOMA2-S Insulin shows the highest coefficient of variation and the lowest discriminant ratio.

The best performing index for discrimination between different subjects in relation to its variability is the NEFA-ISI with a low coefficient of variation.

Fig 1 shows a boxplot of the coefficients of variation of the different indices arranged by the fore mentioned order. Fig 2 shows the discriminant ratios and the corresponding confidence interval for each index.

The disposition index had been omitted due to a very low point estimate 0. The discriminant ratios and their confidence intervals, as well as the p-values of the tests of differences within groups were computed as proposed by Levy at al [ 32 ].

A groupwise aggregation of different coefficients of variation showed a mean of An overview of the mean discriminant ratios for each group shows a mean discriminant ratio of 1. In the insulin sensitivity OGTT group we found a discriminant ratio of 1.

Our literature research yielded inter alia citations of the original paper of Matsuda, citations of the original paper of QUICKI, of the original paper of HOMA2 and citations of the original paper of DI.

The original paper of revised QUICKI was cited times, that of NEFA-ISI 11 times. An overview of the number of citations per year calculated by the total number of citations divided by the years since first publication in relation to the coefficients of variation for each group is given in S1 Fig.

With a coefficient of variation range between 2. The intra-individual coefficient of variation of indices using fasting variables is smaller than that of indices using variables measured during an OGTT for both insulin sensitivity and insulin secretion see Table 3.

Schousboe et al. already described a larger coefficient of variation in indices measuring insulin sensitivity and secretion than in 2h post-load glucose and in fasting blood glucose [ 34 ].

The difference may be due to the greater number of measurements for the OGTT based variables generating more variance and due to variable passage and resorption of the glucose solution after ingestion [ 34 , 35 ].

Change of insulin sensitivity through the menstrual cycle is controversially discussed as a potential further source of intraindividual variation [ 36 , 37 ]. To address this, we excluded female premenopausal participants in an additional analysis see S2 and S3 Tables. In addition to coefficient of variation, we also computed the discriminant ratio of all indices.

The discriminant ratio as recommended by Levy et. Al [ 32 ] is an excellent instrument which not only incorporates the imprecision of a test but quantifies this in relation to the capability to distinguish between different subjects.

Revised QUICKI [ 21 ] and the original QUICKI [ 20 ] showed the smallest coefficients of variation among the indices estimating insulin sensitivity from fasting variables as well as compared to all evaluated indices. Of note, revised QUICKI has been shown to have the highest correlation with gold standard measurements in a meta-analysis [ 9 ].

In addition to fasting glucose and insulin, revised QUICKI uses the fasting level of NEFA. NEFA are not measured routinely in most metabolic studies and require precise pre-analytics in order to avoid in vitro lipolysis [ 38 ].

Generally, insulin concentrations in the bloodstream have a large biological variability due to the hormones short serum half-life, its pulsatile secretion [ 39 ], a marked first-pass effect before reaching systemic circulation.

In addition to the biological variance, there is a considerable analytical variability of laboratory insulin measurement approaches [ 10 ]. Of note, HOMA2-S Insulin [ 22 ] had higher coefficients of variation than QUICKI, despite calculating insulin sensitivity from the same variables.

The lowest coefficient of variation to measure insulin sensitivity using dynamic parameters was found for OGIS [ 24 ]. OGIS has shown good agreement with the hyperinsulinemic euglycemic clamp and a better correlation in comparison to other indices [ 24 , 40 ].

Another advantage is that blood sampling is only required at three different time-points during the OGTT minutes 0, 90 and However, a limitation of OGIS is the complex function, that can, however, be circumvented by web-based calculators and available Excel-sheets [ 25 ].

Both the Belfiore [ 26 ] and the NEFA-ISI [ 27 ] use NEFA levels for the assessment of insulin sensitivity. They exhibit somewhat larger coefficients of variation than OGIS. NEFA-ISI only comprises insulin and NEFA levels at different time-points and it is the only insulin sensitivity index that does not utilize glucose concentrations at all.

This makes it more accurate to measure insulin sensitivity in special situations like pregnancy, a state with physiologically lower glucose levels that often hinders comparison of pregnant with non-pregnant women. The Matsuda index [ 7 ] correlates well with insulin sensitivity assessed from euglycemic hyperinsulinemic clamp [ 41 ].

It is calculated from fasting glucose and insulin levels and the mean of glucose and insulin during OGTT [ 7 ]. Matsuda, computed from 5-point OGTT data, has shown an intraindividual coefficient of variation of The higher number of measured points may explain the higher coefficient of variation as well as the fore mentioned higher variability of insulin and glucose levels.

When comparing the computed coefficients of variations and the discriminant ratios between groups, fasting-based indices showed lower coefficients of variation This might indicate that the indices using fasting values might be more stable during repetitive measurements, but have less power to discriminate metabolic differences between subjects.

Stumvoll-ISI and Stumvoll-MCR incorporate demographic data as age, sex and BMI in addition to glucose and insulin levels obtained during an OGTT.

These are reliable indices [ 29 ]. Unfortunately, in our group of subjects we found several negative indices in persons with high insulin levels minutes after glucose load so that we excluded these indices from these analyses. The lowest coefficient of variation in the measurement of insulin secretion from fasting parameters was shown for HOMA2-B C-Peptide.

This is most likely due to the above mentioned difficulties in insulin measurements. C-peptide and insulin are secreted in equimolar amounts based on their cleavage out of proinsulin.

It is a more stable parameter than insulin and has insignificant clearance by the liver and a longer half-life than insulin [ 42 ]. Therefore C-peptide levels better approximate pancreatic insulin secretion than insulin levels [ 43 ], and so the use of C-peptide to compute HOMA2-B has clear advantages.

Our data show that this ratio has a comparable coefficient of variation to the other indices of this category. The OGTT had been described as an acceptable compromise to assess ß-cell function [ 44 ]. This is well in line with the noted superiority of C-peptide over direct insulin measurements.

Prediabetes and diabetes occur when the pancreas doesn't make enough insulin to maintain normal blood glucose levels. Insulin resistance occurs when cells in your body do not respond well to insulin. Insulin is the key that allows glucose to move from the blood into cells where it is used for energy.

With insulin resistance, it takes more insulin to complete this process. To make up for insulin resistance, the pancreas secretes more insulin. This helps maintain normal blood glucose levels.

It does this by making more insulin to maintain normal blood glucose levels. Over time, however, your pancreas may no longer be able to produce enough insulin to overcome insulin resistance. This can lead to higher-than-normal blood glucose levels.

When this happens, you have prediabetes. Unfortunately, prediabetes has no symptoms. As a result, most people will not know they have it unless they are screened. Screening can entail:. This is because there are so many individual factors that affect people.

Insulin resistance left unmanaged can also lead to diabetes as well other metabolic complications. These include:.

Certain risk factors increase your risk for insulin resistance and prediabetes. These should be discussed with your doctor so you can be screened and monitored properly. You can prevent and even reverse insulin resistance as well as prediabetes by adopting healthy lifestyle habits.

The key is to start these changes early on. Once you have diabetes, it can be very difficult to reverse, notes Dr. Diabetes medications are not prescribed until you have been diagnosed with diabetes. There are many classes of medications that work differently to achieve normal glucose levels.

Some of these function by increasing your body's sensitivity to insulin. Blount says. If your condition progresses to diabetes, you may experience the following symptoms. Insulin resistance has also been assessed qualitatively if one or more insulin values exceed an upper limit of normal at appropriate intervals.

Researchers have compared various methods for assessing insulin sensitivity in type 2 diabetics using the OGTT and found good correlations between AUCinsulin, insulin level at minutes I , and the steady state plasma glucose concentrations derived from a modified ITT.

As mentioned before, the search for uncomplicated and inexpensive quantitative tools to evaluate insulin sensitivity has led to development of fasting state homeostatic assessments. These tests are based on fasting glucose and fasting insulin, and use straightforward mathematical calculations to assess insulin sensitivity and beta cell function.

Several homeostatic approaches have been developed in recent years, each with its merits and deficiencies. One of the weaknesses of these models is that they assume the relationship between glucose and insulin is linear when in fact it's parabolic.

Fasting insulin I0 : Fasting serum insulin is an inexpensive assay, and does not require any mathematical calculations. At least one researcher has advocated averaging two or three readings to account for day-to-day variability.

Although I0 is less variable than other fasting procedures in normoglycemic patients, clinicians must still interpret results cautiously. Remember that insulin sensitivity is the ability of the hormone to reduce serum glucose.

If fasting glucose is high—for example, in a patient with impaired glucose tolerance—that may indicate a diminished effect from circulating insulin or in severe cases of insulin resistance, diminished quantity of the hormone.

Hence I0 should not be used in glucose-intolerant or diabetic patients. The ratio of glucose to insulin is easily calculated, with lower values depicting higher degrees of insulin resistance. Homeostatic model assessment HOMA : HOMA has been widely employed in clinical research to assess insulin sensitivity.

The constant should be replaced by The HOMA value correlates well with clamp techniques and has been frequently used to assess changes in insulin sensitivity after treatment. Quantitative insulin sensitivity check index QUICKI : Like HOMA, QUICKI can be applied to normoglycemic and hyperglycemic patients.

It is derived by calculating the inverse of the sum of logarithmically expressed values of fasting glucose and insulin:. Many investigators believe that QUICKI is superior to HOMA as a way of determining insulin sensitivity, although the two values correlate well.

As the SI decreases, QUICKI values increase. McCauley et al. An ISI of 6. The authors present two formulae for estimating ISI; one uses I0, BMI, and TG, and the other uses only I0 and TG. In comparisons with the euglycemic insulin clamp technique the first formula with BMI has a specificity of 0.

The second forumula without BMI has a specificity of 0. Home Departments Family Medicine Research RCMAR Insulin Resistance. Family Medicine. Medical Student Education. Rural Clerkship.

Because no IV access is needed, OGTT is better suited for assessment of large populations than the other techniques we outlined. A modified OGTT that uses a or g glucose load and measures glucose and insulin at various intervals over 2 to 4 hours has been used in clinical studies.

Like other minimal approaches to diagnosis, OGTT provides information on beta cell secretion and peripheral insulin action, and various mathematical equations have been used to provide an SI value.

Insulin resistance has also been assessed qualitatively if one or more insulin values exceed an upper limit of normal at appropriate intervals. Researchers have compared various methods for assessing insulin sensitivity in type 2 diabetics using the OGTT and found good correlations between AUCinsulin, insulin level at minutes I , and the steady state plasma glucose concentrations derived from a modified ITT.

As mentioned before, the search for uncomplicated and inexpensive quantitative tools to evaluate insulin sensitivity has led to development of fasting state homeostatic assessments.

These tests are based on fasting glucose and fasting insulin, and use straightforward mathematical calculations to assess insulin sensitivity and beta cell function.

Several homeostatic approaches have been developed in recent years, each with its merits and deficiencies. One of the weaknesses of these models is that they assume the relationship between glucose and insulin is linear when in fact it's parabolic.

Fasting insulin I0 : Fasting serum insulin is an inexpensive assay, and does not require any mathematical calculations. At least one researcher has advocated averaging two or three readings to account for day-to-day variability.

Although I0 is less variable than other fasting procedures in normoglycemic patients, clinicians must still interpret results cautiously. Remember that insulin sensitivity is the ability of the hormone to reduce serum glucose. If fasting glucose is high—for example, in a patient with impaired glucose tolerance—that may indicate a diminished effect from circulating insulin or in severe cases of insulin resistance, diminished quantity of the hormone.

Hence I0 should not be used in glucose-intolerant or diabetic patients. The ratio of glucose to insulin is easily calculated, with lower values depicting higher degrees of insulin resistance.

Homeostatic model assessment HOMA : HOMA has been widely employed in clinical research to assess insulin sensitivity. The constant should be replaced by The HOMA value correlates well with clamp techniques and has been frequently used to assess changes in insulin sensitivity after treatment.

Quantitative insulin sensitivity check index QUICKI : Like HOMA, QUICKI can be applied to normoglycemic and hyperglycemic patients. It is derived by calculating the inverse of the sum of logarithmically expressed values of fasting glucose and insulin:. Many investigators believe that QUICKI is superior to HOMA as a way of determining insulin sensitivity, although the two values correlate well.

As the SI decreases, QUICKI values increase. McCauley et al. An ISI of 6. The authors present two formulae for estimating ISI; one uses I0, BMI, and TG, and the other uses only I0 and TG. In comparisons with the euglycemic insulin clamp technique the first formula with BMI has a specificity of 0.

The second forumula without BMI has a specificity of 0. Home Departments Family Medicine Research RCMAR Insulin Resistance. Estimation of insulin resistance in Maxican adults by the [ 13 C] glucose breath test corrected for total endogenous CO2 production.

Mizrahi, M. Assessment of insulin resistance by a 13 C glucose breath test: a new tool for early diagnosis and follow-up of high-risk patients. Nutrition Journal 9, 1—9 Gutt, M. Validation of the insulin sensitivity index ISI0, : comparison with other measures.

Diabetes Res Clin Pract 47, — Qureshi, K. Comparative evaluation of whole body and hepatic insulin resistance using indices from oral glucose tolerance test in morbidly obese subjects with nonalcoholic fatty liver disease.

Journal of Obesity , 1—7 Dix D. The oral glucose tolerance test: A comparison of the time points on the basis of limit values, normal dispersion and reproducibility. Acta Diabetol. Schwartz, M. Diabetes: insulin resistance and obesity.

Wallace, T. Use and abuse of HOMA modelling. Matsuda, M. Insulin sensitivity index obtained from oral glucose tolerance testing: comparison with the euglycemic insulin camp. Diabetes Care 22, — Shapiro, E. Insulin secretion and clearance: comparison after oral and intravenous glucose.

Diabetes 36, — Grzybowski, M. Statistical methodology: III. receiver operating characteristic ROC curves. Academic Emergency Medicine 4, — American Diabetes Association. Standards of medical care in diabetes Diabetes care 37, Suppl 1:S14—S80 Barker, S.

Ghosh, C. Oxygen isotope of breath CO2 linking to erythrocytes carbonic anhydrase activity: a biomarker for pre-diabetes and type 2 diabetes. Article CAS PubMed PubMed Central Google Scholar. Banik, G. Swets, J. Measuring the accuracy of diagnostic systems.

Article CAS ADS MathSciNet Google Scholar. Download references. The author C. Ghosh acknowledges S. Bose Centre for PhD fellowship. We are also extremely grateful to all volunteers for participating in this study. Department of Chemical, Biological and Macromolecular Sciences, S.

Bose National Centre for Basic Sciences, JD Block, Sector III, Kolkata, , Salt Lake, India. Department of Medicine, Vivekananda Institute of Medical Sciences, 99 Sarat Bose Road, Kolkata, , India. Department of Medicine, Raipur Institute of Medical Sciences, Raipur, , Chhattisgarh, India. You can also search for this author in PubMed Google Scholar.

provided the funding; M. and C. designed the whole study and provided the conception; M. and S. supervised the overall study; C. collected and analysed the samples; All authors drafted the manuscript with critical revision. This work is licensed under a Creative Commons Attribution 4. Reprints and permissions.

Sci Rep 5 , Download citation. Received : 13 March Accepted : 11 June Published : 07 July Anyone you share the following link with will be able to read this content:. Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative. Proceedings of the National Academy of Sciences, India Section A: Physical Sciences By submitting a comment you agree to abide by our Terms and Community Guidelines.

If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate. Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Skip to main content Thank you for visiting nature. nature scientific reports articles article. Download PDF. Subjects Diagnostic markers Type 2 diabetes. Abstract New strategies for an accurate and early detection of insulin resistance are important to delay or prevent the acute onset of type 2 diabetes T2D.

Introduction Type 2 diabetes mellitus T2D , the most common deleterious metabolic disease at present all over the world, is usually preceded by the combined effects of pancreatic β-cell dysfunction and insulin resistance 1 , 2.

Figure 1. Full size image. Figure 2. N Engl J Med. American Diabetes Association. Executive summary: standards of medical care in diabetes— Bossuyt PM, Reitsma JB, Bruns DE, Gatsonis CA, Glasziou PP, Irwig L, et al.

STARD an updated list of essential items for reporting diagnostic accuracy studies. Gayoso-Diz P, Otero-González A, Rodriguez-Alvarez MX, Gude F, García F, Francisco AD, et al.

Insulin resistance HOMA-IR cut-off values and the metabolic syndrome in a general adult population: effect of gender and age: EPIRCE cross-sectional study.

BMC Endocr Disord. Naing L. In: Universiti Sains Malaysia. Accessed 07 Mar Delong ER, Delong DM, Clarke-Pearson DL. Comparing the areas under two or more correlated receiver operating characteristic curves: a nonparametric approach. Ruopp MD, Perkins NJ, Whitcomb BW, Schisterman EF.

Youden index and optimal cut-point estimated from observations affected by a lower limit of detection. Biom J. Kramer CK, Zinman B, Gross JL, Canani LH, Rodrigues TC, Azevedo MJ, et al. Coronary artery calcium score prediction of all cause mortality and cardiovascular events in people with type 2 diabetes: systematic review and meta-analysis.

Armstrong RA. When to use the Bonferroni correction. Ophthalmic Physiol Opt. Prediction of type 2 diabetes using simple measures of insulin resistance: combined results from the San Antonio Heart Study, the Mexico City Diabetes Study, and the Insulin Resistance Atherosclerosis Study.

Lorenzo C, Haffner SM, Stančáková A, Laakso M. Relation of direct and surrogate measures of insulin resistance to cardiovascular risk factors in nondiabetic finnish offspring of type 2 diabetic individuals.

J Clin Endocrinol Metab. Ruscica M, Baragetti A, Catapano AL, Norata GD. Translating the biology of adipokines in atherosclerosis and cardiovascular diseases: gaps and open questions. Nutr Metab Cardiovasc Dis. Frankenberg ADV, Nascimento FV, Gatelli LE, Nedel BL, Garcia SP, Oliveira CV, et al.

Major components of metabolic syndrome and adiponectin levels: a cross-sectional study. Diabetol Metab Syndr. Frankenberg ADV, Reis AF, Gerchman F.

Relationships between adiponectin levels, the metabolic syndrome, and type 2 diabetes: a literature review. Arch Endocrinol Metab. Shulman GI. Ectopic fat in insulin resistance, dyslipidemia, and cardiometabolic disease. Download references. LPA reviewed the literature relating to the various indices of insulin resistance and wrote the manuscript.

BLN, TCP and LdAM coassisted in researching the data and reviewed the manuscript. Fernando Gerchman discussed the results with all others, and edited the manuscript.

All authors read and approved the final manuscript. Ralph Anthony DeFronzo for his critical review and helpful feedback of this study. Informed consent was obtained from all individual participants included in the study.

This work was supported by the Rio Grande do Sul Research Foundation FAPERGS; Grant Number Luciana P. Antoniolli received a scientific initiation scholarship from the Rio Grande do Sul Research Foundation and the Brazilian National Research Council. Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Scientific Initiation Program, Faculty of Medicine, Federal University of Rio Grande do Sul, Rua Ramiro Barcelos, , Second floor, Porto Alegre, RS, , Brazil. Metabolism Unit, Endocrinology Division, Hospital de Clínicas de Porto Alegre, Rua Ramiro Barcelos, , Building 12, Fourth floor, Porto Alegre, RS, , Brazil.

You can also search for this author in PubMed Google Scholar. Correspondence to Luciana Pavan Antoniolli. Development and validation method of the equations for insulin resistance indices. This file contains a table that describes the development and validation method of each insulin resistance index.

It also contains comments regarding their performance and applications in previous studies. The first table shows the AUC values of insulin resistance indices to identify metabolic syndrome and their statistical comparison to the reciprocal of Gutt AUC, which had the best AUC.

The second table shows the sensitivity, specificity, likelihood ratios and positive and negative predictive values of selected equations in identifying presence of metabolic syndrome.

Performance of insulin resistance indices to identify metabolic syndrome in a stratified analysis by age, BMI and waist circumference.

The tables show the performance of insulin resistance indices in subgroups stratified by age, BMI and waist circumference.

Number of metabolic syndrome criteria by median of insulin sensitivity and resistance indices. The tables show a comparison between groups divided by median of selected insulin sensitivity and resistance indices for number of metabolic syndrome criteria.

Open Access This article is distributed under the terms of the Creative Commons Attribution 4. Reprints and permissions. Antoniolli, L. et al. Accuracy of insulin resistance indices for metabolic syndrome: a cross-sectional study in adults. Diabetol Metab Syndr 10 , 65 Download citation.

Received : 18 January Accepted : 11 August Published : 20 August Anyone you share the following link with will be able to read this content:. Sorry, a shareable link is not currently available for this article. Provided by the Springer Nature SharedIt content-sharing initiative.

Skip to main content. Search all BMC articles Search. Download PDF. Research Open access Published: 20 August Accuracy of insulin resistance indices for metabolic syndrome: a cross-sectional study in adults Luciana Pavan Antoniolli ORCID: orcid. Abstract Background This study aimed to determine the ability of commonly used insulin resistance indices to identify the metabolic syndrome.

Methods people referred for outpatient care at the Metabolism Unit of Hospital de Clínicas de Porto Alegre were evaluated with anthropometric, blood pressure, lipid profile, and adiponectin measurements.

Results The study population was comprised of people Conclusions A number of commonly employed indices of insulin resistance are capable of identifying individuals with the metabolic syndrome. Background Insulin resistance is a condition in which a greater than normal amount of insulin is required to obtain a quantitatively normal metabolic response.

Methods Subjects The patient population initially consisted of consecutive participants who did not have a previous diagnosis of metabolic syndrome and who were referred for outpatient care in the Metabolism Unit of Hospital de Clínicas de Porto Alegre. Methods Participants underwent a standard evaluation, which included medical history, physical examination, and anthropometric measurements.

Accuracy of insulin resistance indices in the diagnosis of metabolic syndrome ROC analysis showed that the reciprocal of Gutt and OGIS indices yielded an AUC above 0. Full size image. Discussion In the present study, we evaluated the ability of all published indices of insulin resistance to identify the metabolic syndrome.

Abbreviations AUC: area under the curve FIRI: fasting insulin resistance index HbA1c: glycosylated hemoglobin HOMA: homeostasis model assessment Ins 0min : fasting insulin ISI 0min : fasting insulin sensitivity index ISI min : 2 h-insulin sensitivity index OGIS: oral glucose insulin sensitivity index QUICKI: quantitative insulin sensitivity check index ROC: receiver operating characteristic.

References Berson S, Yalow R.