MRI Diffusion Metrics Track Kidney Function in Diabetic CKD

Researchers at Cairo University Hospitals reported that renal apparent diffusion coefficient (ADC) values derived from diffusion-weighted magnetic resonance imaging (DWI-MRI) correlate strongly with kidney function and disease stage in patients with diabetic chronic kidney disease (CKD). The cross-sectional comparative study, published in the Egyptian Journal of Radiology and Nuclear Medicine, evaluated 80 diabetic patients to determine if non-invasive MRI metrics could differentiate early-stage kidney impairment from more advanced disease. The findings suggest that ADC measurements serve as a supportive imaging tool for assessing renal parenchymal dysfunction, particularly in the early and moderate stages of diabetic kidney disease, although their ability to distinguish between advanced stages is limited.

The study aimed to evaluate the role of renal ADC values in diabetic CKD, their relationship with renal function parameters, and their utility in disease staging. Diabetic kidney disease is a major health burden, gradually damaging glomeruli and renal tubules and leading to a decline in kidney function. Standard markers such as serum creatinine and estimated glomerular filtration rate (e-GFR) are often insufficient for detecting early stages of kidney impairment. Conventional imaging modalities like ultrasound and computed tomography provide anatomical information but offer limited functional insight, with drawbacks including operator dependence and radiation exposure. DWI-MRI offers a non-invasive alternative by assessing tissue cellularity and microstructural changes through the random Brownian movement of water molecules.

The research team enrolled 80 diabetic patients aged 20 years or older, diagnosed with type 1 or type 2 diabetes and presenting with CKD stages G1 through G5 according to the KDIGO classification. The participants were divided into two groups: an early-stage group comprising 40 patients with CKD stage G1 and a moderate-to-advanced group comprising 40 patients with CKD stages G2 through G5. No healthy control group was included, as the primary objective was to evaluate ADC variations across different stages of diabetic CKD rather than to compare diseased kidneys with normal renal parenchyma. Patients with other known etiologies of CKD, contraindications to MRI, acute kidney injury, or multiple renal masses were excluded from the study.

All participants underwent renal DWI-MRI on a 1.5-T Philips Achieva MR scanner using a single-shot spin-echo echo-planar imaging sequence. The imaging protocol utilized b values of 0, 50, and 800 s/mm2 to acquire axial diffusion-weighted images. ADC maps were automatically produced by the scanner software for quantitative analysis. An experienced radiologist, unaware of the patients’ clinical and laboratory findings, interpreted the images. ADC measurements were recorded from the upper, middle, and lower poles of both kidneys using circular regions of interest measuring approximately 70–100 mm2. The mean ADC value for each kidney was calculated from these three regions, and the average value from both kidneys was employed for statistical analysis. Separate cortical and medullary measurements were not attempted due to insufficient corticomedullary differentiation in many patients with advanced diabetic CKD.

The results demonstrated that renal ADC values were significantly lower in patients with moderate-to-advanced diabetic CKD (stages G2–G5) than in those with early-stage disease (stage G1). The median ADC value for the moderate-to-advanced group was 1.6 × 10−3 mm2/s (interquartile range 1.5–1.8), compared to a median of 2.1 × 10−3 mm2/s (interquartile range 2.0–2.2) in the early-stage group. This difference was statistically significant with a p-value of less than 0.001. Receiver operating characteristic (ROC) curve analysis indicated excellent diagnostic performance for differentiating early-stage from moderate-to-advanced diabetic CKD, with an area under the curve (AUC) of 0.97 (95% confidence interval 0.91–0.99). An optimal ADC cutoff value of ≤ 1.8 × 10−3 mm2/s yielded 100% sensitivity and 85% specificity for this distinction.

Additional ROC analyses explored the diagnostic utility of ADC values for distinguishing between specific adjacent stages. A cutoff value of ≤ 1.9 × 10−3 mm2/s differentiated stage G1 from stage G2 with 100% sensitivity and 85% specificity. Similarly, a cutoff value of ≤ 1.7 × 10−3 mm2/s distinguished stage G2 from stage G3 with 84.6% sensitivity and 100% specificity. These findings suggest that ADC measurements can facilitate disease stratification in routine practice, particularly for identifying early transitions in kidney function. However, the researchers noted that these cutoff values should be interpreted cautiously until validated in larger multicenter cohorts representing diverse patient populations and MRI systems.

The study also examined the association between renal ADC values and conventional renal function parameters. ADC values showed a strong positive correlation with e-GFR (r = 0.9, p < 0.001) and a strong negative correlation with serum creatinine (r = −0.91, p < 0.001). Patients with moderate-to-advanced diabetic CKD had significantly higher median serum creatinine levels (1.9 mg/dl) and lower median e-GFR values (37 mL/min/1.73 m2) compared to those with early-stage disease (0.7 mg/dl and 104.5 mL/min/1.73 m2, respectively). These correlations support the association between reduced water diffusion in the renal parenchyma and impaired renal function. The progressive decline in ADC values with advancing CKD stage reflects biological changes such as increased interstitial fibrosis, tubular atrophy, and vascular alterations that restrict water movement.

Post hoc analysis revealed that renal ADC values decreased progressively with advancing CKD stage, from a median of 2.1 × 10−3 mm2/s in stage G1 to 1.5 × 10−3 mm2/s in stage G5. Significant differences were observed between stages G1 and G2 (p < 0.001) and between stages G2 and G3 (p = 0.029). However, no statistically significant differences were noted among stages G3 and G4 (p = 0.144) or between stages G4 and G5 (p = 0.485). This pattern indicates that renal ADC measurements are more sensitive for detecting early and moderate deterioration in renal function than for discriminating among advanced CKD stages. The overlap in ADC values in advanced stages likely reflects extensive fibrosis and chronic microvascular changes that produce comparable diffusion characteristics despite further functional decline.

Demographic variables such as patient age and sex did not show a statistically significant association with renal ADC measurements in this cohort. This suggests that the observed changes in ADC values were primarily connected with the severity of diabetic CKD rather than demographic characteristics. The study’s cross-sectional design prevented the evaluation of longitudinal changes in renal ADC values and their potential role in predicting disease progression or therapeutic response. Additionally, the absence of histopathological confirmation limited the ability to directly associate ADC values with the severity of renal fibrosis. Inter-observer reproducibility was not assessed, as measurements were collected by a single experienced radiologist using a standardized protocol.

The researchers concluded that diffusion-weighted MRI provides clinically meaningful information regarding renal parenchymal integrity in diabetic CKD. Renal ADC measurements are not intended to replace conventional biochemical markers but can be used as a complementary quantitative imaging biomarker. This approach is particularly useful during the early stages of the disease when therapeutic intervention is most likely to affect long-term kidney outcomes. The study highlights the potential of non-invasive imaging techniques to enhance the assessment of diabetic kidney disease, offering a tool that complements standard laboratory tests. Future multicenter prospective studies involving larger patient numbers, standardized DWI acquisition protocols, and longitudinal follow-up are advised to further assess the clinical role of renal ADC measurements in diabetic CKD.

The study was conducted between October 2023 and March 2025, following approval from the institutional Ethics Committee of Cairo University Hospitals. All enrolled participants provided informed written consent. The research team calculated the minimum sample size based on an expected area under the ROC curve of 0.95, derived from previously reported performance of renal ADC in CKD. To ensure statistical power for subgroup analyses across various CKD stages, the investigators expanded the cohort to 80 patients. This sample size calculation aimed to maintain a 95% confidence level with an absolute precision of 0.05, ensuring reliable comparisons between the early-stage and moderate-to-advanced disease groups.

Statistical analysis was performed using SPSS Statistics version 24.0. The researchers assessed the normality of continuous variables using both Kolmogorov–Smirnov and Shapiro–Wilk tests. Normally distributed data were presented as mean and standard deviation, while non-normally distributed data were reported as median and interquartile range. Comparisons between the two main groups utilized the independent-samples t test for normal data and the Mann–Whitney U test for non-normal variables. Differences among multiple CKD stages were evaluated using the Kruskal–Wallis test, followed by post hoc analysis where appropriate. Categorical variables were assessed using the Chi-square test, and continuous variables were analyzed using Spearman’s rank correlation coefficient. Statistical significance was defined as a two-tailed p value less than 0.05.

The authors noted that previous studies investigating the relationship between declined renal ADC values and renal function decline have yielded considerable variability in published findings. These discrepancies were attributed to differences in study populations, imaging protocols, and disease classification methods. Relatively limited investigations have evaluated ADC measurements across the complete KDIGO spectrum of diabetic CKD or established clinically useful cutoff values for distinguishing disease severity. The current research intended to address this gap by estimating the diagnostic performance of renal ADC measurements and evaluating their association with conventional renal function parameters across the full range of diabetic CKD stages.

Recent developments in renal MRI have given diffusion-weighted imaging significant traction in the work-up of diabetic kidney disease. Studies using multi-parametric approaches suggest that subtle renal abnormalities can be detected at an early stage, even when overt structural lesions have not yet appeared. These quantitative techniques are proving useful as add-on tools for assessing renal status. The current study demonstrates that conventional renal ADC measurements alone provide excellent diagnostic performance while remaining simple to obtain and readily applicable in routine clinical practice, offering a practical alternative to more complex quantitative approaches.

Subject of Research: Evaluation of renal apparent diffusion coefficient as a biomarker for diabetic chronic kidney disease staging and function

Article Title: Renal apparent diffusion coefficient in diabetic chronic kidney disease: correlation with renal function and CKD stage

Article References: Ali, N. M., Barsoum, N. R., Thabet, K. K., & Shokralla, S. Y. (2026). Renal apparent diffusion coefficient in diabetic chronic kidney disease: correlation with renal function and CKD stage. Egyptian Journal of Radiology and Nuclear Medicine, 57(1), Article 198. https://doi.org/10.1186/s43055-026-01840-x

Image Credits: AI Generated

DOI: 10.1186/s43055-026-01840-x

Keywords: diabetic kidney disease, diffusion-weighted MRI, apparent diffusion coefficient, chronic kidney disease, renal function, non-invasive imaging, KDIGO staging, biomarker, nephrology, radiology, Renal, apparent