OBJECTIVETo use the technique of magnetic resonance imaging (MRI) T2* mapping to diagnose and follow-up of patients with iron overload.

METHODS107 patients who were suspected to have iron overload between 2011.7-2014.3 in Peking Union Medical Colleague Hospital were analyzed retrospectively. Patients had the document of MRI T2* value of liver, heart and pancreas, serum ferritin (SF), transferrin saturation (TS), transfusion amount and other related laboratory tests. T2* values were compared with SF and transfusion amount. T2* values in different organs and their relationship with SF were also evaluated. 10 patients who had been adequately chelated for more than half a year were followed up for their SF and T2* values.

RESULTSThere were 65 males and 42 females with the median age of 51(8-77)-year-old. They were 50 myelodysplastic syndromes (MDS), 36 aplastic anemia, 10 myelofibrosis, 7 hemachromatosis and 4 thalassemia carriers. Liver T2* value was significantly related to SF (r=0.120, P=0.001), but not related to transfusion amount (r=0.019, P=0.175), whereas cardiac MRI T2* was not related either to SF or to transfusion amount. No correlation of the T2* value was found between liver and heart (r=0.015, P=0.235). 70 patients was detected for liver, heart and pancreas T2* simultaneously. Pancreas T2* was compatible to SF (r=0.061，P=0.039) and cardiac T2* (r=0.110, P=0.005), but not correlated to heptic T2* (r=0.047, P= 0.071) or transfusion amount (r=0.000, P=0.960). For the 10 well-chelated patients, during the half year follow-up period, SF changed significantly from (2 566.5±1 152.2) μg/L before chelation to (1 473.4±803.0) μg/L after chelation(P=0.001), while liver T2* remained the same [(6.0±5.1) ms, (6.3±6.0) ms respectively, P=0.629].

CONCLUSIONMRI, although related to SF to some extent, was a valuable additional methods for quantifying iron overload. Iron deposition in different organs might be not related to each other and needed to be evaluated separately. Well-chelation therapy could change SF in half-year follow-up, but T2* change needed longer time to follow-up.