Value of dual-energy CT in differentiating focal fatty infiltration of the liver from low-density masses

V. Raptopoulos, A. Karellas, J. Bernstein, F. R. Reale, C. Constantinou, J. K. Zawacki

Research output: Contribution to journalArticlepeer-review

100 Scopus citations


Focal (irregular, partial) fatty infiltration of the liver may simulate neoplastic or other hypodense masses on CT. On the basis of previous observations of the phenomenon that differences in X-ray attenuation diminish with increasing energy of X-rays used, we performed a preliminary study to determine if dual-energy CT could be used to discriminate between fatty infiltration and hypodense liver masses. Dual-energy CT at 140 and 80 kVp was performed in 14 patients undergoing liver biopsy and in seven control subjects with presumedly normal liver. Attenuation measurements were taken, and the changes in attenuation between 140 and 80 kVp were calculated. The mean changes in attenuation were 3.5 H for normal liver (n = 7), 2.5 H for hypodense liver masses (n = 6), 13 H for fatty liver (n = 5), 0.3 H for fatty liver combined with hemochromatosis or hemosiderosis (n = 3), and 2 H for the spleen (n = 18). The change in attenuation increased as the fat content in the liver increased. Analysis of variance showed a statistically significant difference (p < .001) between fatty liver and the other groups. A difference greater than 10 H was unique to fatty infiltration. These results suggest that dual-energy CT may help to differentiate focal fatty infiltration of the liver from low-density neoplastic or other lesions, but only if the iron content of the liver is not increased.

Original languageEnglish (US)
Pages (from-to)721-725
Number of pages5
JournalAmerican Journal of Roentgenology
Issue number4
StatePublished - 1991

ASJC Scopus subject areas

  • Radiology Nuclear Medicine and imaging


Dive into the research topics of 'Value of dual-energy CT in differentiating focal fatty infiltration of the liver from low-density masses'. Together they form a unique fingerprint.

Cite this