Diabetic kidney disease (DKD) persists as the leading cause of chronic kidney disease (CKD) and often leads to end-stage renal disease (ESRD). Worldwide, 30-50% of patients with diabetes are affected by DKD, while DKD contributes to about half of ESRD. Previously, DKD had been defined based on overt proteinuria-that is, a urine albumin-to-creatinine ratio (UACR) above 300 mg/g-after a stage of microalbuminuria (UACR 30-300 mg/g). However, emerging data suggest that a significant number of patients develop renal functional decline without albuminuria, suggesting that DKD can occur in the absence of protein excretion. This phenotype of normoalbuminuric or non-proteinuric DKD (NA-DKD or NP-DKD) is emerging as an important clinical entity. It is characterized by a gradual decline in renal function, commonly with an annual reduction in estimated glomerular filtration rate (eGFR) > 3 mL/min/1.73 m[2] or an eGFR < 60 mL/min/1.73 m[2], while the UACR remains < 30 mg/g. Growing rates of NP-DKD expose limitations inherent in traditional models of DKD pathogenesis and underscore the need for diagnostic and therapeutic paradigms that are not reliant on albuminuria-only criteria. Here, we present a comprehensive review of the NP-DKD to guide a more inclusive model of DKD pathogenesis, its diagnosis, and therapy.

Background/Objectives: Vitamin B12 deficiency is a common but often underdiagnosed complication in patients with type 2 diabetes (T2D) undergoing long-term metformin therapy. Accurate early prediction could enable targeted screening and timely intervention. This study aimed to develop and interpret a machine learning model for predicting vitamin B12 deficiency in metformin-treated patients with T2D, using eXtreme Gradient Boosting (XGBoost). Methods: A retrospective cross-sectional study was conducted at a single endocrinology centre (La Rabta University Hospital, Tunis, Tunisia). Patients with T2D treated with metformin for at least three years were included (n = 257); those with conditions independently affecting vitamin B12 metabolism were excluded. Vitamin B12 deficiency was defined as a serum B12 level below 150 pmol/L or a borderline level (150-221 pmol/L) with concurrent hyperhomocysteinemia (>15 μmol/L). XGBoost was selected after comparison with Logistic Regression (L2), Random Forest, and Support Vector Machine on the same 5-fold stratified cross-validated pipeline. Hyperparameters were optimized via Bayesian search (100 iterations × 5-fold stratified cross-validation), with the Matthews correlation coefficient (MCC) as the primary optimization metric to account for class imbalance. Model interpretability was achieved using SHapley Additive exPlanations (SHAP). Discrimination and calibration were assessed on an independent test set using bootstrap 95% confidence intervals (2000 resamples). Results: Of 257 patients, 95 (37.0%) presented with vitamin B12 deficiency. On the independent test set (n = 52), the optimized XGBoost model achieved an ROC-AUC of 0.671 [95% CI: 0.514-0.818], sensitivity of 0.737 [95% CI: 0.533-0.938], specificity of 0.545 [95% CI: 0.375-0.710], MCC of 0.273 [95% CI: 0.018-0.517], and a Brier Score of 0.259. SHAP analysis identified HbA1c, microalbuminuria, autonomic neuropathy, BMI, DN4 score, and fasting glucose as the most influential predictors. Nonlinear SHAP interaction plots revealed an increased predicted risk in patients with low HbA1c combined with a high cumulative metformin dose. Conclusions: The XGBoost-SHAP framework provided interpretable predictions of vitamin B12 deficiency in patients with T2D on metformin, identifying key clinical profiles for targeted screening. External multi-centre validation is required before clinical deployment.

Hyperglycemia, glucose fluctuations, and thyroid dysfunction contribute to the progression of diabetic nephropathy. This study aimed to investigate the associations of glycemia, glucose variability, and thyroid hormones with albuminuria in patients with type 2 diabetes mellitus (T2DM). In total, 451 T2DM patients were included. The data of continuous glucose monitoring system (CGMS)-generated indices, thyroid hormones, and the ratios of urine creatinine to urine microalbumin (URCA) were collected. There were 152 patients with normal URCA (< 30 mg/g), 230 patients with microalbuminuria (30 mg/g ≤ UACR < 300 mg/g), and 69 patients with macroalbuminuria (UACR ≥ 300 mg/g). Free triiodothyronine (FT3) was negatively correlated with hemoglobin A1C (HbA1C), glucose, and glucose variability in T2DM patients (most p < 0.05). HbA1C (p = 0.002), the standard deviation (SD) of glucose (p < 0.001), the coefficient of variation (CV) of glucose (p < 0.001), the time below range (TBR) (p = 0.002), and the mean amplitude of glycemic excursions (MAGE) (p = 0.033) were positively associated with albuminuria. Moreover, FT3 was negatively associated with macroalbuminuria (p = 0.003). According to multivariate logistic regression analyses, HbA1C, SD, and CV of glucose, the TBR, and the MAGE were independently associated with a greater risk of albuminuria after adjustment for demographics, duration of T2DM, biochemical indices, and medications for T2DM and hypertension (all p < 0.05). Moreover, the associations of HbA1C, SD, and CV of glucose, TBR, and MAGE with albuminuria were partially mediated by FT3, FT3, and TSH, but the weight of mediation was low. In conclusion, glycemia and glucose variability derived from the CGMS are correlated with a lower level of FT3, and they are positively associated with albuminuria in patients with T2DM.