Association Between Serum Homocysteine and Chronic Kidney Disease Severity in Vietnamese Outpatients With Type 2 Diabetes: A Cross-Sectional Study.

INTRODUCTION: CKD is a major complication of T2DM and a leading cause of end-stage kidney disease. Because urinary albumin-to-creatinine ratio (UACR) and eGFR have recognized limitations, additional biomarkers may help characterize kidney dysfunction. HCY is frequently elevated in CKD and may reflect pathways of renal injury. We evaluated the association of HCY with kidney dysfunction and its cross-sectional discrimination of CKD status in Vietnamese adults with T2DM. METHODS: A descriptive cross-sectional study was conducted among adult outpatients with T2DM recruited at UMC HCMC (Campus 2) from December 2023 to December 2025 using convenience sampling. CKD was defined as eGFR < 60 mL/min/1.73 m2 and/or UACR > 30 mg/g. HCY and other laboratory parameters were measured using standardized protocols; eGFR was calculated using the 2021 CKD-EPI creatinine-cystatin C equation. Correlation, logistic regression, and ROC analyses were performed. Albuminuria alone was examined in a sensitivity analysis, and AUCs were compared using DeLong's test. RESULTS: Among 201 participants, 164 (81.6%) met the study-defined CKD criteria. HCY was higher in participants with CKD and increased across worsening eGFR and albuminuria categories. HCY correlated more strongly with creatinine and cystatin C (r = 0.67 for both) and eGFR (r = -0.59) than with UACR (r = 0.26). Higher HCY remained associated with CKD after adjustment for age and HbA1c (adjusted OR = 1.27, 95% CI 1.14-1.42). The HCY-only AUC was 0.79; the Youden-derived threshold was 11.35 μmol/L (sensitivity 0.76; specificity 0.78). For albuminuria alone, the association persisted (OR = 1.08, 95% CI 1.03-1.14), but discrimination was modest (AUC = 0.63). The multivariable AUC was 0.812 and did not differ significantly from the HCY-only AUC (DeLong p = 0.325). CONCLUSIONS: Higher HCY was associated with CKD and kidney dysfunction, with a stronger relationship to filtration markers than to albuminuria. These cross-sectional findings do not establish screening or prognostic utility. The 11.35 μmol/L threshold is exploratory and requires external validation.