Diagnostic Value of Insulin Resistance Indices and Nomogram Models in Identifying Pre-Metabolic Syndrome: A Population-Based Study from Southeast China

Abstract Background Metabolic syndrome (MetS) is a risk factor for cardiovascular diseases and cancer, with its pre-stage (preMetS) also conferring risk. MetS incidence is rising annually, yet the absence of unified diagnostic criteria presents challenges for primary healthcare workers. Insulin resistance (IR) indicators are widely used for MetS screening, but research on their ability to identify preMetS is limited. Objective To estimate preMetS prevalence among adults in Fuzhou City, compare differences and agreement among three diagnostic criteria, and examine the association and diagnostic value of six IR indicators for preMetS. Methods 9155 permanent Fuzhou residents underwent face-to-face questionnaires, physical examinations, and laboratory tests. Direct method calculated age- and sex-standardized prevalence for preMetS and MetS. Agreement among three criteria for preMetS/MetS classification was assessed using consistency tests. Univariate/multivariate logistic regression and restricted cubic spline analyses explored IR-preMetS associations. Diagnostic performance was compared via area under the curve (AUC). A nomogram model incorporating indicators and age was developed to enhance diagnostic capability. Results Among 8997 participants, revised ATP III criteria yielded the highest age-standardized preMetS and MetS prevalence (International Diabetes Federation (IDF) criteria the lowest), irrespective of sex. Revised ATP III showed high agreement with Joint Committee for Developing Chinese Guidelines (JCDCG) (Kappa = 0.700–0.820) but low agreement with IDF (Kappa = 0.316–0.377). The triglyceride-glucose index (TyG) was the optimal screening indicator (AUC = 0.731, 95%CI:0.718–0.744; cutoff = 7.736). In normal-weight individuals, TyG and triglyceride to high-density lipoprotein ratio (TG/HDL) showed comparable performance (P delong > 0.05). The nomogram combining age and TG/HDL demonstrated superior preMetS screening ability (P delong < 0.05) and better goodness-of-fit. Conclusion Revised ATP III and JCDCG showed high consistency for MetS classification, with ATP III being more suitable for large-scale screening. TyG is an effective preMetS identifier in Southeast Chinese adults, while waist-to-height ratio (WHtR) is a valuable non-invasive screening indicator. The "Age + TG/HDL" nomogram model optimizes preMetS diagnostic accuracy.