Kidney involvement in diabetes is a heterogeneous clinical problem. Although diabetic kidney disease (DKD) remains an important cause of chronic kidney impairment, proteinuria or reduced kidney function in a patient with diabetes does not establish diabetic nephropathy. Non-diabetic kidney disease (NDKD) can occur independently or coexist with diabetes-associated injury, creating distinct diagnostic categories with different pathological mechanisms and management implications [1]. Recognising this distinction is especially relevant when the clinical presentation departs from the expected course of diabetic microvascular disease. Kidney biopsy series demonstrate substantial variation in the distribution of these categories. A pooled analysis of 48 studies involving 4,876 patients found wide differences in reported frequencies of diabetic, non-diabetic, and mixed lesions [2]. In a large retrospective biopsy study, Sharma and colleagues identified NDKD alone or superimposed on diabetic nephropathy in a substantial proportion of patients [3]. These findings describe selected populations referred for biopsy; they cannot be interpreted as community prevalence estimates among all individuals with diabetes. Clinical features used to raise suspicion of NDKD include shorter diabetes duration, absent diabetic retinopathy, haematuria, active urinary sediment, abrupt proteinuria, and unexpectedly rapid deterioration of kidney function. Hsieh and colleagues proposed a stepwise approach incorporating clinical features into biopsy assessment [4]. A multicentre study by Bermejo and colleagues also examined factors associated with non-diabetic lesions [5]. However, no single finding provides definitive separation. A systematic review found that ordinary haematuria had limited discriminatory accuracy, whereas dysmorphic erythrocytes were more specific but relatively insensitive [6]. The pathological spectrum includes primary glomerular disorders, infection-associated lesions, vascular disease, and tubulointerstitial injury. Recent Indian experience highlights both changing lesion patterns and the potential importance of disease-specific treatment [7]. Earlier studies from India likewise documented diverse diagnoses among patients with atypical renal presentations [8,9]. Differences in referral practices, biopsy indications, population characteristics, and local disease patterns influence the diagnoses encountered. Consequently, regionally derived evidence is useful for interpreting biopsy yield and identifying the clinical circumstances in which histological clarification is particularly informative. Prospective biopsy evidence has also challenged reliance on conventional clinical predictors. Basu and colleagues demonstrated the coexistence of DKD, NDKD, and mixed disease and emphasised the limitations of clinical discrimination at the individual level [10]. A recent review further situates biopsy findings within personalised diagnostic and therapeutic assessment [11]. Nevertheless, small local series require careful reporting of uncertainty, transparent handling of ambiguous records, and avoidance of unsupported prognostic conclusions. The present study therefore aimed to describe the frequency and histopathological spectrum of NDKD among patients with diabetes whose biopsy reports were analysed at a tertiary care centre in Ananthapuramu and to compare their demographic, clinical, biochemical, and structural kidney characteristics across diagnostic categories.

Study Design and Setting This retrospective, hospital-based observational study was conducted in the Department of Nephrology, Superspeciality Hospital, Government Medical College, Ananthapuramu, Andhra Pradesh, India. The study included kidney biopsy records of patients with diabetes mellitus evaluated between March 2022 and March 2026. Clinical, biochemical, and histopathological information was retrieved from available hospital records and renal biopsy reports. The study aimed to determine the frequency and histopathological spectrum of non-diabetic kidney disease (NDKD) and compare the clinical and pathological characteristics of patients with diabetic kidney disease (DKD), NDKD, and mixed renal lesions. Study Population and Sample Size This retrospective observational study included 53 patients with diabetes mellitus who underwent kidney biopsy for clinically suspected non-diabetic kidney disease (NDKD). The sample size comprised all eligible biopsy records available during the study period. One record with an ambiguous histopathological diagnosis was excluded from group-wise comparisons, leaving 52 records for comparative analysis. Inclusion Criteria • Patients with diabetes mellitus who underwent kidney biopsy for one or more of the following indications were included: • Nephrotic-range proteinuria or nephrotic syndrome. • Rapidly progressive renal failure or unexplained acute kidney injury. • Absence of diabetic retinopathy with suspected renal disease. • Presence of diabetic retinopathy with proteinuria exceeding 10 g/24 hours. • Active urinary sediment, including microscopic haematuria. • Sudden onset or rapid progression of proteinuria. • Refractory hypertension with renal dysfunction. • Clinical suspicion of associated systemic disease. Exclusion Criteria Patients with incomplete clinical records, unavailable or uninterpretable kidney biopsy findings, or insufficient documentation of diabetes mellitus were excluded. Study Variables and Data Collection Demographic information, including age and sex, was extracted from patient records. Clinical characteristics comprised duration of diabetes mellitus, presence of hypertension, diabetic retinopathy, and indications for renal biopsy where documented. Biochemical parameters included serum creatinine and urinary protein excretion. Histopathological variables included the primary renal diagnosis, percentage of glomerulosclerosis, and extent of interstitial fibrosis and tubular atrophy (IFTA). Available information was reviewed for completeness and consistency before analysis. Missing observations were not imputed, and each variable was analysed using its corresponding number of available records. Histopathological Classification Renal biopsy findings were classified into three diagnostic categories based on the documented histopathological diagnosis. Pure DKD was defined as diabetic renal lesions without a separately identified non-diabetic renal pathology. Pure NDKD included renal diseases unrelated to diabetic nephropathy in the absence of documented DKD lesions. Mixed kidney disease was defined by the simultaneous presence of DKD and at least one distinct non-diabetic renal lesion. Individual NDKD diagnoses, including hypertensive nephrosclerosis, membranous nephropathy, and other documented renal pathologies, were recorded according to the available biopsy reports. The extent of glomerulosclerosis and IFTA was assessed using the percentages documented in the histopathological records. One diagnostically ambiguous record remained unclassified. Bias Control and Statistical Analysis Measures to minimise information and classification bias included consistent extraction of recorded variables, review of diagnostic coding, application of predefined histopathological categories, and transparent handling of ambiguous observations. Nevertheless, the retrospective design and selective performance of kidney biopsies introduced potential referral and selection biases. Statistical analysis was performed using Python with the pandas and SciPy libraries. Continuous variables were expressed as mean ± standard deviation or median with interquartile range, according to their distribution and analytical relevance. Categorical variables were presented as frequencies and percentages. Comparisons among the three classified diagnostic groups were performed using the Kruskal–Wallis test for continuous variables. Categorical variables were compared using Monte Carlo estimates of the Fisher–Freeman–Halton exact test with 100,000 resamples and a fixed random seed. Wilson 95% confidence intervals were calculated for diagnostic proportions. Spearman rank correlation was used to examine the association between serum creatinine and IFTA. All statistical tests were two-sided, with p < 0.05 considered statistically significant for exploratory analysis. No adjustment for multiple comparisons, post hoc pairwise testing, or multivariable regression analysis was performed. Ethical Considerations The study was approved by the Institutional Ethics Committee, Government Medical College, Ananthapuramu, Andhra Pradesh, India, vide approval number IEC/GMC/ATP/2026/141, dated 11 August 2026. Patient confidentiality was maintained throughout data extraction and analysis. Personal identifiers were excluded from the analytical dataset, and the findings were reported in aggregate form.

Participant characteristics The study contained 53 patient records. Mean age was 48.58 ± 10.82 years, with a range of 22–70 years; 32 patients (60.4%) were male. Mean diabetes duration was 7.59 ± 5.31 years. Hypertension was recorded in 44 patients (83.0%), and retinopathy in 16 (30.2%). Mean serum creatinine was 2.50 ± 1.64 mg/dL, and mean proteinuria was 6.43 ± 4.96 g/24 hours. Baseline characteristics are presented in Table 1. Table 1. Demographic and clinical characteristics of the available biopsy series Characteristic Value (N = 53) Age, years, mean ± SD 48.58 ± 10.82 Male sex, n (%) 32 (60.4) Female sex, n (%) 21 (39.6) Diabetes duration, years, mean ± SD 7.59 ± 5.31 Hypertension, n (%) 44 (83.0) Diabetic retinopathy, n (%) 16 (30.2) Peripheral neuropathy, n (%) 16 (30.2) Serum creatinine, mg/dL, mean ± SD 2.50 ± 1.64 Proteinuria, g/24 hours, mean ± SD 6.43 ± 4.96 Recorded active urinary sediment, n (%) 24 (45.3) Positive RBC field, n (%) 7 (13.2) Dialysis dependence at presentation, n (%) 2 (3.8) Note. SD = standard deviation; RBC = red blood cell. Figure 1: Clinical Characteristics of the Biopsy Series Recorded indications for biopsy Nephrotic presentation was the most frequent recorded indication (31 patients, 58.5%), followed by rapidly progressive renal failure (18, 34.0%). Acute kidney injury and nephritic presentation accounted for two patients each. Table 2. Recorded primary indications for kidney biopsy Indication n % Nephrotic presentation 31 58.5 Rapidly progressive renal failure 18 34.0 Acute kidney injury 2 3.8 Nephritic presentation 2 3.8 Total 53 100.0 Note. Percentages use N = 53; rounding accounts for totals above 100%. Figure 2: Recorded Primary Indications for Kidney Biopsy Diagnostic distribution Explicit coding supported pure DKD in 26 patients (49.1%), pure NDKD in 14 (26.4%), and mixed disease in 12 (22.6%). One record (1.9%) lacked an affirmative group flag and contained conflicting lesion labels, so it remained unclassified. Definite NDKD, either alone or coexisting with DKD, was recorded in 26 patients (49.1%; 95% CI, 36.1–62.1%). Diagnostic proportions and their confidence intervals are shown in Table 3. Group comparisons included the 52 classified records. Table 3. Diagnostic categories and biopsy frequency estimates Category n % of 53 95% CI (%) Pure DKD 26 49.1 36.1–62.1 Pure NDKD 14 26.4 16.4–39.6 Explicit mixed DKD and NDKD 12 22.6 13.5–35.5 Unclassified 1 1.9 0.3–9.9 Any definite NDKD* 26 49.1 36.1–62.1 Note. DKD = diabetic kidney disease; NDKD = non-diabetic kidney disease; CI = confidence interval, calculated using the Wilson method. *Aggregate of pure NDKD and mixed disease; this row overlaps the preceding categories. Figure 3: Histopathological Diagnostic Categories in Patients with Diabetes Clinical comparisons across diagnostic groups Retinopathy occurred in 10 of 26 patients with pure DKD (38.5%), two of 14 with pure NDKD (14.3%), and four of 12 with mixed disease (33.3%). This difference was not statistically significant (p = 0.328). Recorded active sediment occurred in 34.6%, 57.1%, and 58.3%, respectively (p = 0.248). Sex, hypertension, peripheral neuropathy, and recorded biopsy indication also showed no significant overall group differences (Table 4). Table 4. Clinical features in the classified diagnostic groups Feature Pure DKD (n = 26) Pure NDKD (n = 14) Mixed (n = 12) p value Male sex 19 (73.1) 8 (57.1) 4 (33.3) 0.070 Hypertension 23 (88.5) 9 (64.3) 11 (91.7) 0.153 Retinopathy 10 (38.5) 2 (14.3) 4 (33.3) 0.328 Peripheral neuropathy 10 (38.5) 2 (14.3) 3 (25.0) 0.296 Recorded active sediment 9 (34.6) 8 (57.1) 7 (58.3) 0.248 Nephrotic presentation 16 (61.5) 7 (50.0) 7 (58.3) 0.258 Rapidly progressive renal failure 10 (38.5) 4 (28.6) 4 (33.3) — Acute kidney injury 0 (0.0) 2 (14.3) 0 (0.0) — Nephritic presentation 0 (0.0) 1 (7.1) 1 (8.3) — Note. Values are n (% within group). Monte Carlo estimates of Fisher–Freeman–Halton exact p values; 100,000 resamples. The indication p value applies to the full four-category distribution. The unclassified patient was excluded. Biochemical and quantitative pathological findings Median diabetes duration was 10.00 years in pure DKD, 3.50 years in pure NDKD, and 5.00 years in mixed disease; the overall difference was significant (p = 0.002). Serum creatinine and proteinuria did not differ significantly. Glomerulosclerosis and IFTA percentages differed across the groups (p = 0.012 and p = 0.010, respectively), with descriptively higher medians in pure DKD (Table 5). Overall tests do not establish specific pairwise differences. Table 5. Continuous variables across classified diagnostic groups Variable Pure DKD (n = 26) Pure NDKD (n = 14) Mixed (n = 12) p value Age, years 49.00 (44.25–54.50) 53.00 (47.50–62.00) 46.00 (37.50–48.25) 0.103 Diabetes duration, years 10.00 (6.00–10.00) 3.50 (0.62–7.25) 5.00 (5.00–7.25) 0.002 Creatinine, mg/dL 1.85 (1.23–3.86) 1.50 (1.20–3.42) 2.05 (1.27–3.12) 0.812 Proteinuria, g/24 hours 5.49 (2.65–8.97) 2.80 (2.32–7.17) 5.05 (3.08–8.16) 0.363 Glomerulosclerosis, % 45.50 (29.62–65.75) 13.50 (1.75–31.50) 35.00 (20.75–44.75) 0.012 IFTA, % 50.00 (26.25–60.00) 20.00 (2.50–53.75) 25.00 (17.50–42.50) 0.010 Note. Values are median (interquartile range). p values are from overall Kruskal–Wallis tests. IFTA = interstitial fibrosis and tubular atrophy. Histopathological spectrum of non diabetic lesions Among pure NDKD records, hypertensive nephrosclerosis and membranous nephropathy were equally frequent, each occurring in three patients (21.4%). Explicit mixed lesions most often included focal segmental glomerulosclerosis (four patients, 33.3%), followed by IgA nephropathy and immune-complex/C3-related labels (two each, 16.7%). The lesion distribution is presented in Table 6. Twenty-four of the 26 pure-DKD records described nodular diabetic nephropathy; one described early and one advanced diabetic nephropathy. Table 6. Distribution of non diabetic lesion spectrum Lesion Pure NDKD (n = 14) Mixed (n = 12) Hypertensive nephrosclerosis 3 (21.4) 0 (0.0) Membranous nephropathy 3 (21.4) 1 (8.3) FSGS tip variant 0 (0.0) 2 (16.7) FSGS not otherwise specified 1 (7.1) 2 (16.7) IgA nephropathy 1 (7.1) 2 (16.7) Immune-complex/C3-related GN labels 0 (0.0) 2 (16.7) Proliferative or infection-related GN 2 (14.3) 1 (8.3) Acute tubular necrosis 1 (7.1) 0 (0.0) Acute pyelonephritis 0 (0.0) 1 (8.3) Acute tubulointerstitial nephritis 0 (0.0) 1 (8.3) Chronic interstitial nephritis 1 (7.1) 0 (0.0) Granulomatous interstitial nephritis 1 (7.1) 0 (0.0) Focal global glomerulosclerosis 1 (7.1) 0 (0.0) Total 14 (100.0) 12 (100.0) Note. Values are n (% within group). FSGS = focal segmental glomerulosclerosis; GN = glomerulonephritis. Original lesion labels were harmonised only at the stated descriptive level. Immune-complex and C3 labels require original-report confirmation. Interstitial injury and its association with kidney function The recorded IFTA grades were absent in six patients (11.3%), mild in 19 (35.8%), moderate in 14 (26.4%), and marked in 14 (26.4%). Their distribution differed across the classified groups (p = 0.021; Table 7). Serum creatinine correlated positively with the recorded IFTA percentage across all 53 records (Spearman ρ = 0.482; p < 0.001). Table 7. Distribution of interstitial fibrosis and tubular atrophy grades IFTA grade Overall (N = 53) Pure DKD (n = 26) Pure NDKD (n = 14) Mixed (n = 12) Absent 6 (11.3) 0 (0.0) 4 (28.6) 2 (16.7) Mild 19 (35.8) 7 (26.9) 5 (35.7) 6 (50.0) Moderate 14 (26.4) 10 (38.5) 1 (7.1) 3 (25.0) Marked 14 (26.4) 9 (34.6) 4 (28.6) 1 (8.3) Note. Values are n (%). Overall counts include the unclassified patient, who had mild IFTA. Recorded severity labels were used without deriving thresholds from IFTA percentages. Overall categorical comparison among classified groups: p = 0.021. Figure :4 Distribution of Interstitial Fibrosis and Tubular Atrophy(IFTA) Grades

This biopsy series showed definite NDKD, either isolated or combined with DKD, in 26 of 53 patients. Its frequency supports the clinical relevance of alternative kidney pathology in selected patients with diabetes, but does not estimate population prevalence. The broad confidence interval reflects limited precision. Heterogeneous biopsy frequencies in the pooled analysis by Fiorentino and colleagues reinforce the influence of patient selection [2]. Sharma and colleagues similarly demonstrated substantial non-diabetic pathology in a larger referral series [3]. One unresolved record in the present dataset also illustrates why diagnostic coding requires verification before final submission. Diabetes duration differed across the three classified groups, with longer duration in pure DKD and shorter duration in pure NDKD. This pattern is consistent with the diagnostic associations described by Hsieh and colleagues and the prospective study by Basu and colleagues [4,10]. However, the present comparisons were unadjusted and cannot establish independent predictive value. Duration is also subject to uncertainty because the date of diagnosis does not necessarily identify the onset of diabetes. The finding therefore supports contextual clinical interpretation rather than a stand-alone biopsy threshold. Retinopathy was less frequent in pure NDKD, but the group difference was not statistically significant. Absence of retinopathy also occurred in 16 patients with pure DKD. Thus, neither its presence nor its absence provided definitive classification. This interpretation is compatible with the limitations of clinical prediction highlighted in multicentre and prospective investigations [5,10]. Active sediment likewise failed to distinguish groups significantly. Its binary recording, without a standardised microscopy definition, limits comparison with studies using quantitative sediment thresholds. The limited accuracy of ordinary haematuria reported by Jiang and colleagues further cautions against treating urinary findings as decisive [6]. Serum creatinine and proteinuria showed substantial overlap across categories and no significant overall differences. Histological injury nevertheless varied: glomerulosclerosis and IFTA percentages differed across groups, with descriptively higher values in pure DKD. The positive creatinine–IFTA correlation indicates an association between functional impairment and chronic structural damage at biopsy, without proving future progression. The pathological framework described by Tervaert and colleagues supports separate assessment of glomerular and tubulointerstitial compartments [12]. Outcome claims concerning treatment response, kidney survival, or dialysis progression cannot be drawn because follow-up information was unavailable. Hypertensive nephrosclerosis and membranous nephropathy were equally frequent among pure NDKD records, whereas focal segmental glomerulosclerosis predominated among explicit mixed lesions. Indian and Thai studies describe different lesion distributions, reflecting regional and referral heterogeneity [7–9,14]. These differences argue for local pathological assessment rather than assumption of a universal dominant lesion. Recent synthesis emphasises the relevance of distinguishing diagnoses when considering individualised care [11]. The present findings are most applicable to comparable tertiary nephrology services selecting patients for biopsy. Larger prospective studies with verified pathology, standardised sediment assessment, and longitudinal outcomes are needed to evaluate clinical predictors and treatment implications. Limitations This study was limited by its retrospective, single-centre design and relatively small sample size of 53 patients. Selection bias was present because renal biopsies were performed in selected diabetic patients with clinical indications rather than routinely. The absence of multivariable adjustment limited assessment of independent associations. These factors restrict generalisability to the broader diabetic population and prevent causal interpretation of the observed findings.

The present study identified non-diabetic kidney disease, either independently or in combination with diabetic kidney lesions, in 26 of 53 patients (49.1%) whose renal biopsy reports were analysed at a tertiary care centre. Diabetes duration differed significantly among the diagnostic groups, whereas diabetic retinopathy, serum creatinine, and proteinuria showed no statistically significant differences. Significant variations in glomerulosclerosis and interstitial fibrosis/tubular atrophy highlighted the histopathological heterogeneity of kidney disease in patients with diabetes mellitus. These findings emphasise the diagnostic importance of renal biopsy in appropriately selected diabetic patients, particularly those presenting with atypical clinical features or unexplained renal dysfunction. Histopathological evaluation enables differentiation between diabetic and non-diabetic renal lesions, providing information relevant to individualised clinical management. Larger prospective multicentre studies are warranted to validate these observations and identify reliable clinical predictors of non-diabetic kidney disease.

1.Anders HJ, Huber TB, Isermann B, Schiffer M. CKD in diabetes: diabetic kidney disease versus nondiabetic kidney disease. Nat Rev Nephrol. 2018;14(6):361–377. doi:10.1038/s41581-018-0001-y. PMID: 29654297. 2.Fiorentino M, Bolignano D, Tesar V, Pisano A, Van Biesen W, Tripepi G, et al. Renal biopsy in patients with diabetes: a pooled meta-analysis of 48 studies. Nephrol Dial Transplant. 2017;32(1):97–110. doi:10.1093/ndt/gfw070. PMID: 27190327. 3.Sharma SG, Bomback AS, Radhakrishnan J, Herlitz LC, Stokes MB, Markowitz GS, et al. The modern spectrum of renal biopsy findings in patients with diabetes. Clin J Am Soc Nephrol. 2013;8(10):1718–1724. doi:10.2215/CJN.02510213. PMID: 23886566. 4.Hsieh JT, Chang FP, Yang AH, Tarng DC, Yang CY. Timing of kidney biopsy in type 2 diabetic patients: a stepwise approach. BMC Nephrol. 2020;21(1):131. doi:10.1186/s12882-020-01794-w. PMID: 32293326. 5.Bermejo S, González E, López-Revuelta K, Ibernon M, López D, Martín-Gómez A, et al. Risk factors for non-diabetic renal disease in diabetic patients. Clin Kidney J. 2020;13(3):380–388. doi:10.1093/ckj/sfz177. PMID: 32699618. 6.Jiang S, Wang Y, Zhang Z, Dai P, Yang Y, Li W. Accuracy of hematuria for predicting non-diabetic renal disease in patients with diabetes and kidney disease: a systematic review and meta-analysis. Diabetes Res Clin Pract. 2018;143:288–300. doi:10.1016/j.diabres.2018.07.027. PMID: 30059756. 7.Prasad N, Veeranki V, Bhadauria D, Kushwaha R, Meyyappan J, Kaul A, et al. Non-Diabetic Kidney Disease in Type 2 Diabetes Mellitus: A Changing Spectrum with Therapeutic Ascendancy. J Clin Med. 2023;12(4):1705. doi:10.3390/jcm12041705. PMID: 36836240. 8. Das U, Dakshinamurty KV, Prayaga A, Uppin MS. Nondiabetic kidney disease in type 2 diabetic patients: a single center experience. Indian J Nephrol. 2012;22(5):358–362. doi:10.4103/0971-4065.103912. PMID: 23326046. 9. Kanodia KV, Vanikar AV, Nigam L, Patel RD, Suthar KS, Patel H. Clinicopathological study of nondiabetic renal disease in type 2 diabetic patients: a single center experience from India. Saudi J Kidney Dis Transpl. 2017;28(6):1330–1337. doi:10.4103/1319-2442.220877. PMID: 29265044. 10.Basu M, Pulai S, Neogi S, Banerjee M, Bhattacharyya NP, Sengupta S, et al. Prevalence of non-diabetic kidney disease and inability of clinical predictors to differentiate it from diabetic kidney disease: results from a prospectively performed renal biopsy study. BMJ Open Diabetes Res Care. 2022;10(6):e003058. doi:10.1136/bmjdrc-2022-003058. PMID: 36517108. 11.Sreedharan S, M K M, K B V, Raju N. Non-diabetic Kidney Disease in Type 2 Diabetes: From Kidney Biopsy to Precision Medicine. Cureus. 2026;18(1):e100716. doi:10.7759/cureus.100716. PMID: 41646563. 12.Tervaert TWC, Mooyaart AL, Amann K, Cohen AH, Cook HT, Drachenberg CB, et al. Pathologic classification of diabetic nephropathy. J Am Soc Nephrol. 2010;21(4):556–563. doi:10.1681/ASN.2010010010. PMID: 20167701. 13.Vandenbroucke JP, von Elm E, Altman DG, Gøtzsche PC, Mulrow CD, Pocock SJ, et al. Strengthening the Reporting of Observational Studies in Epidemiology (STROBE): explanation and elaboration. Int J Surg. 2014;12(12):1500–1524. doi:10.1016/j.ijsu.2014.07.014. PMID: 25046751. 14.Kritmetapak K, Anutrakulchai S, Pongchaiyakul C, Puapairoj A. Clinical and pathological characteristics of non-diabetic renal disease in type 2 diabetes patients. Clin Kidney J. 2018;11(3):342–347. doi:10.1093/ckj/sfx111. PMID: 29942497.