Iron Overload of the Kidneys of Patients with Thalassemia and Sickle-Cell Disease: Final Results of a Prospective Study in 224 Patients Using Magnetic Resonance Imaging T2 and R2* Techniques

Iron overload is a significant complication of hemoglobinopathies, despite the adequate iron chelation, which is administered to patients nowadays. Recently, magnetic resonance imaging (MRI) gradient echo (T2*), the reciprocal of T2* (known as R2*) and spin echo (T2) techniques have been developed to quantify tissue iron in the liver and the heart. Patients with thalassemia or sickle-cell disease (SCD) may develop kidney impairment during the course of their disease. However, the significance of iron overload in the kidneys of these patients in the development of renal impairment is poorly understood and, according to our knowledge, it has not been evaluated in a prospective study. Thus, the aim of our study was to evaluate kidney R2* and T2 MRI as surrogates for kidney iron of patients with different hemoglobinopathies and correlate these results with markers of renal impairment, markers of hemolysis and other disease features.

We prospectively studied 224 patients with hemoglobinopathies who are followed and treated in a single center: 109 had thalassemia major (TM), 58 thalassemia intermedia (TI) and 57 SCD (35% of those were double heterozygous for SC/beta-thalassemia; HbS/beta-thal). All patients had MR examinations of the kidneys, liver and heart. For T2* mapping a gradient echo, multi echo FSPGR sequence was used, while for T2 mapping a T2-prepared single shot SSFP technique with multi echoes was performed. The normal reference values of T2 in our lab was <80 msec, while for R2* normal range we used the one previously described by Schein et al (J Magn Reson Imaging 2008;28:698-704), which was 21.3±5.8 Hz with a 95% confidence upper limit of 33.7 Hz. Regarding renal function, patients were divided in the 5 chronic kidney disease (CKD) stages of the KDIGO classification, according to eGFR evaluated by the CKD-EPI formula.

Kidney T2 values were (mean±SD) 109.7±12.1 msec, 101.4±12.3 msec and 100.1±11.3 msec for patients with TM, TI and SCD, respectively; 9/109 (8.2%) TM, 3/57 (5.2%) SCD and 2/58 (3.4%) TI patients had T2 values lower than the normal limit of <80 msec. Of interest, none of the HbS/beta-thal patients had T2 values lower than the normal limit. Kidney R2* values were 17.3±10.8 Hz, 21.5±15.4 Hz and 19.1±13.6, for patients with TM, TI and SCD, respectively. Compared to historic controls, 5/109 (4.5%) TM, 8/58 (13.7%) TI and 4/57 (7%) SCD patients had R2* >upper normal limit of 33.7 Hz. There was no difference between the different groups regarding T2 or R2* values.

eGFR values were 102±21, 111±12 and 98±28 ml/min/1.73m², for patients with TM, TI and SCD, respectively. Patients with TI had higher eGFR than TM and SCD patients (p=0.014 and p=0.001, respectively). Six (5.5%) patients with TM and six (10.5%) with SCD had renal impairment of CKD stage 3-5. No patient with TI presented with stage 3-5 renal impairment. There was no correlation between T2 or R2* values and eGFR or serum creatinine in all patients groups. This may be due to the different T2 or R2* values between the two kidneys of the same patient (p=0.001 and p=0.01, for T2 and R2*, respectively).

Patients with low T2 values (<80 msec) had higher serum ferritin and lower Hb (p=0.001 and p=0.01, respectively). Patients with high R2* values (>33.7 Hz) had higher serum LDH and reticulocyte counts (p=0.012 and p=0.01, respectively). Renal R2* (r=-0.286, p=0.001) but not T2 values correlated also with liver T2* values. In TM patients there was a negative correlation of T2 values with ferritin (r=-0.300, p<0.01) and LDH (r=-0.340, p<0.01), a positive correlation of R2* with ferritin (r=0.4, p<0.001) and LDH (r=0.382, p<0.001) and a negative correlation with Hb (r=-0.278, p=0.01).

Our study provides evidence that T2 and R2* reflects kidney iron deposition in patients with different hemoglobinopathies: from 4.5% of TM patients who regularly receive iron chelation to 13.7% of TI patients. Importantly we found differences in iron overload between the two kidneys of the same patient in all studied group; finding which needs further evaluation and examination of possible correlations with each kidney's function. In TM patients both MRI techniques correlated with serum ferritin and LDH. Further studies, possibly with the performance of renal biopsies, will reveal the best MRI technique for detecting iron overload in the kidneys of patients with hemoglobinopathies as well as their role in the development of renal impairment in some of these patients.