Re: Further Evidence Supporting the Accuracy of Quantitative Magnetic Resonance Imaging for Evaluating Iron Load in Dialysis Patients.

To the Editor:

In his editorial accompanying our article, Daniel Coyne raises important issues regarding the validity of magnetic resonance imaging (MRI) for quantifying iron load in dialysis patients.1, 2 We are disappointed that he did not analyze our article devoted to this topic, published in January 2017. There is indeed a need to validate these MRI techniques in dialysis patients, notably by comparison with liver biopsy. However, liver biopsy is an invasive and risky procedure, especially in frail patients with end-stage renal disease, and such studies therefore raise ethical concerns.

In a pilot study, on the advice of ethicists, we compared the classic Scheuer score and Deugnier and Turlin histological classification of iron overload (Perls staining of hemosiderin deposits) with signal-intensity-ratio MRI values obtained with the Rennes University algorithm in 11 hemodialysis patients in whom liver biopsy was formally indicated for their medical follow-up. For Scheuer’s histological classification, the Wilcoxon matched-pairs test showed no significant difference in the ranking of iron overload by histology and MRI (summary of ranks = 1.5; P = 1) (Figure 1). The MRI and Scheuer histological classifications were strongly correlated (rho = 0.866, P = 0.0035, Spearman coefficient), as were the absolute liver iron concentrations on MRI (rho = 0.860, P = 0.0013, Spearman coefficient). The absolute liver iron concentrations on MRI also correlated strongly with the Deugnier-Turlin histological score (rho = 0.841, P = 0.0033, Spearman coefficient). We think these recent findings in the field of dialysis-related iron overload warrant the attention of the broad readership of Kidney International Reports.

Figure 1

Scatterplot of ranks of the liver magnetic resonance imaging (MRI) and histologic (Perls-Scheuer) classifications in 11 hemodialysis patients. To allow a formal comparison between the MRI scale according to Rennes University (4 categories) and its Perls counterpart according to Scheuer, we combined categories 0 and 1 of the Scheuer classification, which relate to normal liver iron; this category is referred to as category 1. According to Rostoker et al.3