Merck
CN
  • Evaluation of Hepatic Iron Overload Using a Contemporary 0.55 T MRI System.

Evaluation of Hepatic Iron Overload Using a Contemporary 0.55 T MRI System.

Journal of magnetic resonance imaging : JMRI (2021-10-21)
Adrienne E Campbell-Washburn, Christine Mancini, Anna Conrey, Lanelle Edwards, Sujata Shanbhag, John Wood, Hui Xue, Peter Kellman, W Patricia Bandettini, Swee Lay Thein
摘要

MRI T2* and R2* mapping have gained clinical acceptance for noninvasive assessment of iron overload. Lower field MRI may offer increased measurement dynamic range in patients with high iron concentration and may potentially increase MRI accessibility, but it is compromised by lower signal-to-noise ratio that reduces measurement precision. To characterize a high-performance 0.55 T MRI system for evaluating patients with liver iron overload. Prospective. Forty patients with known or suspected iron overload (sickle cell anemia [n = 5], ß-thalassemia [n = 3], and hereditary spherocytosis [n = 2]) and a liver iron phantom. A breath-held multiecho gradient echo sequence at 0.55 T and 1.5 T. Patients were imaged with T2*/R2* mapping 0.55 T and 1.5 T within 24 hours, and 16 patients returned for follow-up exams within 6-16 months, resulting in 56 paired studies. Liver T2* and R2* measurements and standard deviations were compared between 0.55 T and 1.5 T and used to validate a predictive model between field strengths. The model was then used to classify iron overload at 0.55 T. Linear regression and Bland-Altman analysis were used for comparisons, and measurement precision was assessed using the coefficient of variation. A P-value < 0.05 was considered statistically significant. R2* was significantly lower at 0.55 T in our cohort (488 ± 449 s-1 at 1.5 T vs. 178 ± 155 s-1 at 0.55 T, n = 56 studies) and in the patients with severe iron overload (937 ± 369 s-1 at 1.5 T vs. 339 ± 127 s-1 at 0.55 T, n = 23 studies). The coefficient of variation indicated reduced precision at 0.55 T (3.5 ± 2.2% at 1.5 T vs 6.9 ± 3.9% at 0.55 T). The predictive model accurately predicted 1.5 T R2* from 0.55 T R2* (Bland Altman bias = -6.6 ± 20.5%). Using this model, iron overload at 0.55 T was classified as: severe R2* > 185 s-1 , moderate 81 s-1  < R2* < 185 s-1 , and mild 45 s-1  < R2* < 91 s-1 . We demonstrated that 0.55 T provides T2* and R2* maps that can be used for the assessment of liver iron overload in patients. 2 TECHNICAL EFFICACY: Stage 2.