Novel laboratory index, based on fasting blood parameters, accurately reflects insulin sensitivity

Simple and reliable measurement of insulin sensitivity may be important
for the prevention of insulin-resistance related diseases. Surrogate
indices of insulin sensitivity are of limited utility in population
without signs of metabolic syndrome. The aim of our study was to provide
simple and accurate index of insulin sensitivity. The study group
comprised 150 young healthy participants. Hyperinsulinemic-euglycemic
clamp was performed. Regression models with different laboratory
parameters were constructed. Validation cohort 1 comprised independent
group of 110 subjects, including individuals with prediabetes and newly
diagnosed type 2 diabetes. Validation cohort 2 comprised 38 obese subjects
before and after diet-induced weight loss. Validation cohort 3 comprised
60 nondiabetic subjects from an independent center. The supervised
principal component model established optimal set of variables correlated
with insulin sensitivity. This model (Fasting Laboratory Assessment of
Insulin Sensitivity, FLAIS) used red blood cell count, alanine
aminotransferase activity, serum C-peptide, SHBG, IGF-binding protein 1
and adiponectin concentrations. FLAIS exhibited strong correlation with
clamp-derived insulin sensitivity. The sensitivity of the model was 90%
and the specificity was 68%. In the validation cohort 1, differences in
FLAIS among the groups paralleled those observed with the clamp, with the
lowest values in prediabetes and diabetes. In the validation cohort 2,
FLAIS reflected the change in insulin sensitivity after weight loss. The
main findings were confirmed in the validation cohort 3. We provide simple
and accurate method of assessing insulin sensitivity, which allows to
identify insulin resistance even in the population without overt metabolic
disturbances.