Computed tomography (CT) scanning is not highly sensitive in the assessment of liver iron content and magnetic resonance imaging (MRI) appears to be more efficient. The aim of this study was to determine the effectiveness of MRI in the evaluation of liver iron content using a standard spin-echo technique. The study included 23 patients with genetic hemochromatosis and 24 non-iron-overloaded patients as controls. A comparison was made of: (a) MRI signal intensity of liver, spleen, paravertebral muscles and subcutaneous adipose tissue using two different spin-echo sequences (SE 500/28; SE 2000/28,56); (b) liver attenuation determined by a single energy CT scan; and (c) a biochemical determination of hepatic iron. There was a significant decrease in liver signal intensity in the genetic hemochromatosis group (256 +/- 201, mean +/- S.D.) compared with the control group (801 +/- 413, p less than 0.001), but there was no correlation with liver iron concentration. However, such a correlation was found and was even more highly significant than in CT when the ratio between the liver and another organ was taken into account. For a lower limit of liver/spleen ratio calculated at 0.46 (mean 2 S.D. in the control group), the specificity (0.96) of MRI was satisfactory, but the sensitivity (0.78) remained insufficient (MRI being unable to detect an iron overload of up to 125 mumol/g). Hopefully, these results might be improved in the near future by using more sensitive sequences such as gradient echo sequences.