Non-invasive assessment of tissue iron overload in the liver by magnetic resonance imaging.

We investigated the clinical usefulness of a standard magnetic resonance imaging (MRI) system for non-invasive determination of the liver iron concentration in 38 patients with iron overload and 15 normal controls by measurement of the signal intensity ratio between liver and skeletal muscle (SIR). However, SIR was found dependent on the applied repetition time (TR) of the MRI system, which led us to investigate this relationship in autopsy material of liver and muscle tissue specimens with various iron content. Based on these results, adjustment of SIR measurements to a constant value of TR was achieved. By use of this technique we found a close correlation between MRI and chemically determined liver iron concentration (r2 = 0.98) as well as the serum ferritin concentration (r2 = 0.86). The reproducibility was sufficiently good for the use of MRI in the follow-up of iron reductive treatment. The use of iron store parameters in serum was found insufficient as indicators of endpoint for venesection therapy, if 20 mumol Fe/g dry weight was applied as the upper reference limit of the liver iron concentration. It is concluded that MRI based on SIR measurements offers a precise and reproducible non-invasive method for the determination and follow-up of iron overload within a wide range of liver iron concentrations. Our findings may increase the clinical use of MRI in haematological patients with iron overload.