A comparison between insulin resistance scores parameters in identifying patients with metabolic syndrome

Abstract

As a consequence of the epidemic of obesity, prediction of metabolic syndrome (MetS) is relevant because of its subsequent association with type 2 diabetes and cardiovascular disease. Also, MetS is considered a state of insulin resistance (IR). This study aimed to evaluate the capacity of different indirect IR scores to identify patients with MetS, compared to standard criteria of MetS diagnosis. To the best of our knowledge, the cut-off values of these indexes, as indicators of MetS, in the Romanian population, have not been established. We used a non-parametric extension of the induced ROC regression methodology to determine the cut-off values of HOMA-IR, QUICKI, McAuley indexes, taking into account the MetS components instead of using 90th percentile criteria. Although HOMA is more commonly used in practice (probably due to its easy use), McAuley Score seems to have a better specificity in identifying patients with MetS. This is different between men and women; if in women, McAuley index has the highest sensitivity, in men, HOMA and QUICKI indexes have the highest sensitivity and specificity.