Urinary adiponectin as a non-invasive biomarker of renal impairment in type 2 diabetes mellitus: evidence from a cross-sectional study.

Abstract

Background: Early detection of diabetic kidney disease (DKD) is critical, yet current tools such as eGFR and albuminuria often reflect injury only after substantial renal damage. Urinary adiponectin has been proposed as a sensitive marker of glomerular barrier dysfunction. This study evaluated its association with renal outcomes and ability to distinguish kidney involvement stages in type 2 diabetes mellitus (T2DM).

Methods: A total of 385 individuals were evaluated, including 60 healthy volunteers and 325 with T2DM stratified into normo-, micro-, and macroalbuminuria. Clinical, biochemical, and demographic variables were compared across groups. Associations of urinary adiponectin with renal outcomes were tested using correlation analysis and multivariable regression models appropriate for each outcome distribution. For albuminuria, a gamma generalized linear model with log link was applied, whereas estimated glomerular filtration rate (eGFR) was analyzed using a gaussian generalized linear model. Confidence intervals were obtained using profile likelihood or bootstrap approaches. Diagnostic accuracy was tested with ROC curves.

Results: Urinary adiponectin increased progressively from controls to those with advanced albuminuria (p < 0.001). It showed a moderate positive correlation with urine albumin-creatinine ratio (UACR) [ρ = 0.46, p < 0.001] and a weaker inverse association with eGFR (ρ = -0.12, p = 0.02). In multivariable analysis, adiponectin remained independently associated with UACR (β = 0.075, 95% CI: 0.024-0.129, p = 0.002), together with diabetes duration and systolic blood pressure. Its association with eGFR did not reach significance. Diagnostic testing demonstrated modest discrimination for early DKD, with the highest accuracy observed when distinguishing normo- from microalbuminuria (AUC 0.702, 95% CI: 0.635-0.769).

Conclusion: Urinary adiponectin reflects renal impairment in T2DM and demonstrates moderate diagnostic accuracy for identifying early DKD. While not a standalone diagnostic tool, it holds potential as a complementary marker alongside conventional measures.