Evaluating renal iron overload in diabetes mellitus by blood oxygen level-dependent magnetic resonance imaging: a longitudinal experimental study

Weiwei Geng, Liang Pan, Liwen Shen, Yuanyuan Sha, Jun Sun, Shengnan Yu, Jianguo Qiu, Wei XingBMC Medical Imaging 2022 · DOI

TL;DR

This study investigates the use of blood oxygen level-dependent (BOLD) magnetic resonance imaging (MRI) to non-invasively assess renal iron overload in diabetic rabbits. The findings suggest that BOLD MRI can effectively track iron accumulation and its interaction with diabetes, offering a promising tool for understanding and managing diabetic nephropathy.

What Problem Does This Paper Address?

Diabetic nephropathy, a serious complication of diabetes, is strongly linked to iron overload in the kidneys. Currently, assessing renal iron levels non-invasively is challenging, hindering effective management and intervention strategies. This paper aims to bridge this gap by exploring the feasibility of BOLD MRI, a technique sensitive to iron, for evaluating renal iron overload in a diabetic rabbit model. The researchers hypothesize that BOLD MRI can not only detect iron deposition but also indirectly reflect kidney injury caused by this overload, providing a more comprehensive picture of the disease progression.

Methods

The study employed a longitudinal experimental design using a rabbit model. Four groups of rabbits were established: control (C), iron overload (I), diabetes (D), and diabetes with iron overload (DI). Diabetes was induced using alloxan, and iron overload was achieved with iron-dextran injections. BOLD MRI scans were performed on a subset of rabbits (n=10 per group) at baseline (week 0) and at weeks 4, 8, and 12 post-modeling. The MRI protocol involved acquiring T2-weighted images and BOLD MRI sequences to generate R2 maps. Regions of interest (ROIs) were drawn in the renal cortex (CR2) and outer medulla (MR2), and the ratio of MR2 to CR2* (MCR) was calculated. Histopathological examinations (hematoxylin and eosin, Prussian blue, and Masson's trichromatic staining) were conducted on harvested kidney tissues to assess kidney injury, iron deposition, and interstitial fibrosis, respectively. Statistical analysis, including repeated-measures ANOVA and Pearson correlation, was used to compare MRI parameters and correlate them with histopathological findings.

Datasets

Not discussed in the paper. The study utilized an animal model and generated its own imaging and histopathological data.

Key Insights and Ideas

  • Interplay of Diabetes and Iron Overload: The study highlights a significant interaction between diabetes and iron overload in exacerbating kidney injury. The combined DI group showed more severe kidney injury and iron overload compared to the individual I and D groups.
  • BOLD MRI as a Sensitive Tool: BOLD MRI, specifically the R2 relaxation rate, proved to be sensitive to iron accumulation in the kidneys. Changes in CR2 and MR2* values over time reflected the progression of iron deposition and injury.
  • Differential Iron Deposition: The study observed distinct patterns of iron deposition. In the iron-overload group (I), iron primarily accumulated in glomeruli, while in the diabetes group (D), it was found in proximal tubular cells. The DI group exhibited iron deposition in both locations.
  • Indirect Reflection of Kidney Injury: While BOLD MRI directly measures iron-related changes, the study suggests that MR2* values, particularly in the outer medulla, might indirectly reflect kidney injury by indicating reduced oxygenation levels, even in the absence of direct iron deposition in that specific region.
  • MCR as a Combined Metric: The MCR (MR2/CR2) metric provided a way to integrate the dynamic trends of both cortex and outer medulla, revealing different patterns between the groups, potentially reflecting distinct metabolic processes and injury mechanisms.

Key Results

  • Increased R2* in Diseased Groups: At week 4, CR2 values were significantly higher in the I, D, and DI groups compared to the control group. The DI group showed even higher CR2 than the D group. By week 12, CR2 and MR2 values were significantly elevated in the I, D, and DI groups compared to the control, with the DI group consistently showing higher values.
  • Decreased MCR: MCR values significantly decreased in the I, D, and DI groups from week 0 to week 4, indicating increased iron deposition or altered tissue properties.
  • Histopathological Confirmation: Histopathology confirmed iron deposition in glomeruli and proximal tubules, with the DI group showing the most severe iron accumulation and kidney injury (edema, necrosis, fibrosis) by week 12.
  • Correlation with Injury: Pearson correlation analysis revealed moderate to strong correlations between renal R2* values and histopathological findings of iron overload, kidney injury, and interstitial fibrosis across the groups.

Limitations and Future Work

The authors acknowledge several limitations:

  • Small Sample Size: The study had a relatively small sample size, which may limit the generalizability of the findings.
  • Manual ROI Selection: The ROIs for measuring R2* values were manually drawn, introducing potential inter-observer variability. The authors suggest exploring radiomics methods for future analysis.
  • Perfusion Effects: Renal R2 is influenced by physiological factors like blood volume fraction. While the study considered different reperfusion time points, the interactive effect of perfusion and iron deposition on R2 warrants further investigation.

Future work could focus on larger cohort studies, automated ROI segmentation or radiomics approaches, and a deeper exploration of the interplay between renal perfusion and iron overload in BOLD MRI signal changes.

Summary updated

BibTeX
@article{geng2022evaluating,
  author = {Weiwei Geng and Liang Pan and Liwen Shen and Yuanyuan Sha and Jun Sun and Shengnan Yu and Jianguo Qiu and Wei Xing},
  title = {{Evaluating renal iron overload in diabetes mellitus by blood oxygen level-dependent magnetic resonance imaging: a longitudinal experimental study}},
  journal = {BMC Medical Imaging},
  year = {2022},
  doi = {10.1186/s12880-022-00939-7},
}
Figure