Comparison between two-point and four-point methods for quantification of apparent diffusion coefficient of normal liver parenchyma and focal lesions. Value of normalization with spleen

Abstract

Purpose To compare two quantification techniques of apparent diffusion coefficient (ADC), both in normal liver parenchyma and focal lesions, and to investigate any potential value of normalization. Materials and methods Fifty-six consecutive patients underwent MRI examination of the liver, including a single shot spin-echo echo planar imaging diffusion sequence with four b-values (0, 50, 500 and 1000 s/mm 2). ADC maps were reconstructed based on a two-point method ( b-values: 500 and 1000 s/mm 2) and a four-point method ( b-values: 0, 50, 500 and 1000 s/mm 2). Comparison of absolute ADC measurements of the liver, benign and malignant focal lesions was performed between the two- and four-point techniques. The same analysis was done on normalized ADC values (absolute ADC values divided by spleen ADC values). Results The difference between mean two-point and four-point ADC values of normal liver (absolute: 1.237 × 10 −3, 1.615 × 10 −3 mm 2/s, normalized: 1.40, 1.52, respectively) was statistically significant ( p < 0.0001 and p = 0.0061). Significantly higher absolute ADC values of benign and malignant lesions were recorded with the four-point method (2.860 × 10 −3 and 1.307 × 10 −3 mm 2/s) over the two-point method (2.243 × 10 −3, and 1.011 × 10 −3 mm 2/s) ( p < 0.0001 in both) while the same differences in normalized values were proven statistically non-significant for benign lesions ( p = 0.788) and statistically significant for malignant lesions ( p = 0.015). Both differences in absolute and normalized ADC values of benign versus malignant lesions based on two- and four-point methods were found to be significant ( p < 0.0001). Conclusion ADC quantification of the liver may be performed with a two-point method ( b-values of 500 and 1000 s/mm 2), while normalization of ADC measurements with the spleen is not further improving lesion characterization.