Comparison of liver iron concentration calculated from R2* at 1.5 T and 3 T

Area of Science:

• Medical Imaging
• Biophysics

Background:

• R2* measurements via magnetic resonance imaging (MRI) are used to estimate liver iron concentration (LIC).
• While both 1.5 Tesla (T) and 3 T scanners are utilized, the conversion of 3 T R2* values to LIC is less established.
• This variability necessitates a comparison to ensure accurate LIC assessment across different field strengths.

Purpose of the Study:

• To compare liver iron concentration (LIC) estimations derived from 3 T R2* and 1.5 T R2* MRI measurements.
• To assess the interchangeability of LIC values obtained from 3 T and 1.5 T MRI scanners.
• To evaluate the impact of different conversion methods on the agreement between 3 T and 1.5 T R2*-based LIC estimations.

Main Methods:

• Thirty participants underwent MRI scans at both 1.5 T and 3 T field strengths.
• R2* values were measured at each field strength and converted to LIC using distinct calibration curves.
• Statistical analyses included linear regression, Pearson's correlation coefficient (rho), Intraclass Correlation Coefficients (ICCs), and Bland Altman plots to assess agreement and bias.

Main Results:

• All comparisons between 1.5 T and 3 T LIC estimations yielded a Pearson's rho of 0.99 (p < 0.001).
• Intraclass Correlation Coefficients (ICCs) ranged from 0.83 (p = 0.005) to 0.96 (p < 0.001), indicating strong agreement.
• Observed biases between the two field strengths were minimal, with a magnitude less than 0.2 mg/g dry weight.

Conclusions:

• The agreement and bias between 3 T and 1.5 T R2*-based LIC estimations are dependent on the chosen conversion methodology.
• Excellent agreement and small biases were observed in specific instances, suggesting potential interchangeability.
• Caution is advised regarding the selection of conversion methods when using 3 T R2* LIC estimations alongside or in place of 1.5 T values.