Clinical Predictors of Non-Diabetic Glomerular Disease in Type 2 Diabetic Patients With Nephrotic-Range Proteinuria.

AIM: Nephrotic-range proteinuria in patients with Type 2 diabetes mellitus (T2DM) may result from diabetic nephropathy or non-diabetic glomerular disease (NDGD). Kidney biopsy risks and limited accessibility complicate patient selection. We aimed to identify predictors of NDGD and develop a score to guide biopsy decisions. METHODS: We conducted a single-center retrospective study of T2DM patients undergoing native kidney biopsy for nephrotic-range proteinuria at a tertiary hospital in Thailand between 2014 and 2022. Biopsy findings were classified according to the presence or absence of NDGD. Independent predictors were identified using multivariable logistic regression, and a weighted clinical prediction score was derived. Discrimination was evaluated using the area under the receiver operating characteristic curve (AUROC). RESULTS: Among 289 patients, 142 (49.1%) had NDGD with or without concomitant DN. The most frequent diagnosis was IgA nephropathy. Independent predictors of NDGD included older age, shorter duration of T2DM, absence of diabetic retinopathy, absence of prior hypertension, higher estimated glomerular filtration rate, and hematuria. Higher haemoglobin was retained in the prediction model despite its borderline statistical association. The prediction score demonstrated good discrimination (AUROC, 0.85; 95% confidence interval, 0.80-0.89). At a threshold of ≥ 3 points, the score had 96.5% sensitivity and 40.8% specificity, whereas a threshold of ≥ 7 points had 66.9% sensitivity and 84.4% specificity. CONCLUSION: NDGD is common among T2DM patients with nephrotic-range proteinuria. The proposed clinical prediction score showed good discrimination in this cohort but requires external validation before it can be used to support biopsy decision-making in clinical practice.