Insulin resistance, metabolic phenotypes, and the obesity paradox: a triglyceride glucose-body mass index framework in partial kidney diseases

Background Insulin resistance(IR) is a key driver of cardiorenal-metabolic diseases. The Triglyceride Glucose-Body Mass Index (TyG-BMI), a composite marker of lipid-glucose metabolism and overall adiposity, has gained traction as a predictive tool. Its clinical application, however, is complicated by non-linear and sometimes paradoxical associations with clinical outcomes, a phenomenon reminiscent of the”obesity paradox”. This review critically evaluates the performance of TyG-BMI across the spectrum of kidney disease, aiming to reconcile these divergent findings through a metabolic-endocrine framework. Methods We performed a narrative synthesis of observational studies and secondary analyses of large-scale public databases such as National Health and Nutrition Examination Survey(NHANES), China Health and Retirement Longitudinal Study(CHARLS), UK Biobank, Medical Information Mart for Intensive Care IV(MIMIC-IV), identified through PubMed and Google Scholar, to evaluate the association of TyG-BMI with the incidence, progression, and prognosis of acute kidney injury (AKI), diabetic kidney disease (DKD), chronic kidney disease (CKD), and Cardiovascular-Kidney-Metabolic syndrome(CKM). Results TyG-BMI shows significant, context-dependent predictive value. In early-stage or metabolically heterogeneous populations, elevated TyG-BMI consistently predicts higher disease risk. Conversely, in advanced stages (e.g., end-stage renal disease, stage 4 CKM), the association frequently becomes U-shaped, with low TyG-BMI paradoxically linked to excess mortality. This reflects its dual nature: the TyG component captures ‘metabolic toxicity’, while BMI proxies ‘nutritional reserve’. Thus, the ‘obesity paradox’ may represent a metabolic phenotype spectrum—from metabolically healthy obesity (MHO) to metabolically unhealthy lean (MUL)—that the combination of its TyG and BMI components helps differentiate. For disease incidence prediction, TyG-BMI is often outperformed by indices of central obesity (e.g., Triglyceride-Glucose-Waist Circumference Index(TyG-WC), Triglyceride-Glucose–Waist-to-Height Ratio(TyG-WHtR); for mortality in advanced illness, the standalone TyG index may be superior. Conclusion TyG-BMI is a valuable tool for metabolic phenotyping and risk stratification in kidney disease, with its paradoxical associations highlighting the complex interplay between metabolic health and nutritional reserve. Future research must establish context-specific clinical thresholds. TyG-BMI is preferentially useful in settings where waist circumference is unavailable, in early CKD screening, and in select type 2 diabetes mellitus (T2DM) subgroups. In advanced CKD, dialysis, critical illness, and CKM stage 4, alternative indices such as TyG index, TyG-WC, or TyG-WHtR should be prioritized due to the confounding effects of fluid overload and catabolic state.