Quantitative T2 MRI for bone marrow iron overload: normal reference values and assessment in thalassemia major patients

• Published in: Radiologia medica Vol. 127; no. 11; pp. 1199 - 1208
• Main Authors: Meloni, Antonella, Pistoia, Laura, Restaino, Gennaro, Missere, Massimiliano, Positano, Vincenzo, Spasiano, Anna, Casini, Tommaso, Cossu, Antonella, Cuccia, Liana, Massa, Antonella, Massei, Francesco, Cademartiri, Filippo
• Format: Journal Article
• Language: English
• Published: Milan Springer Milan 01.11.2022; Springer Nature B.V
• Subjects: Abdominal Radiology; Blood diseases; Bone marrow; Coefficient of variation; Correlation; Diagnostic Radiology; Females; Ferritin; Imaging; Interventional Radiology; Iron; Liver; Mathematical analysis; Medicine; Medicine & Public Health; Neuroradiology; Radiology; Reproducibility; Ultrasound; Vertebrae; Iron overload; Bone marrow; Lower limit of normal; Magnetic resonance imaging; Thalassemia major
• ISSN: 1826-6983; 0033-8362; 1826-6983
• DOI: 10.1007/s11547-022-01554-w

Purpose We evaluated the feasibility and reproducibility of bone marrow T2* values and established the lower limit of normal in a cohort of healthy subjects. We investigated the clinical correlates of bone marrow T2* values in patients with thalassemia major (TM). Material and Methods Thirty healthy subjects and 274 consecutive TM patients (38.96 ± 8.49 years, 151 females) underwent MRI at 1.5T. An axial slice in the upper abdomen was acquired by a T2* gradient-echo multiecho sequence and the T2* value was calculated in a circular region of interest defined in the visible body of the first or second lumbar vertebra. In patients, also liver and heart T2* values were assessed. Results In healthy subjects bone marrow T2* values were independent of age and gender. The lower limit of normal for bone marrow T2* was 13 ms. In both healthy subjects and 30 randomly selected patients, the coefficient of variation for inter-operator-reproducibility was < 10%. TM patients exhibited significantly lower bone marrow T2* values than healthy subjects (7.47 ± 5.18 ms vs. 17.08 ± 1.89 ms; p  < 0.0001). A pathological bone marrow T2* was detected in 82.8% of TM patients. In TM, the female sex was associated with reduced bone marrow T2* values. Bone marrow T2* values were inversely correlated with mean serum ferritin levels (R = -0.431; P  < 0.0001) and hepatic iron load ( R  = − 0.215; P  < 0.0001). A serum ferritin level > 536 ng/ml predicted the presence of a pathological bone marrow T2*. A positive correlation was found between bone marrow and heart T2* values ( R  = 0.143; P  = 0.018). A normal bone marrow T2* showed a negative predictive value of 100% for cardiac iron. Conclusion Bone marrow T2* measurements can be easily obtained using the same sequences acquired for liver iron quantification and may bring new insights into the pathophysiology of iron deposition; hence, they should be incorporated into clinical practice.