Hepatobiliary contrast agents and LI-RADS

LI-RADS (Liver Imaging Reporting and Data System) is both a set of standardized terminology and a classification system for imaging findings in liver lesions. The LI-RADS score for a liver lesion is an indication of its relative risk for hepatocellular carcinoma (HCC). The classification system is meant to be used in livers which have risk factors for HCC (e.g. cirrhotic livers).

The original LI-RADS was developed for extracellular gadolinium contrast agents (e.g. Gadavist). Using a hepatobiliary contrast agent (e.g. Eovist/Primovist, MultiHance) changes the evaluation of hepatic lesions.

Five phases are often used for evaluation:

• T1W precontrast

• late arterial (30-35 sec)

• portal venous (~75 sec)

• "transitional" (~3 min)

• hepatobiliary (~20 min with Eovist/Primovist) 

Assessment of hepatobiliary phase

The first important step is to evaluate the quality of the hepatobiliary phase. If the hepatic parenchyma is hyperintense relative to the blood pool, the phase can be considered adequate.

Although the hepatobiliary phase is usually acquired at 15-20 minutes for Eovist/Primovist (or 1 hour for MultiHance), severely cirrhotic livers may require a longer time (e.g. 30 minutes for Eovist/Primovist) to generate adequate images.

Hepatobiliary phase

If the lesion is:

• hyperintense: indeterminate, and it could be either

  • benign: e.g. FNH, dysplastic nodule

  • malignant: e.g. up to 12% of HCCs

  • a hypointense rim around a hyperintense lesion is thought to represent tumor capsule

    • feature favoring malignancy (cannot upgrade lesion past LR4)

    • this feature may be seen in intrahepatic cholangiocarcinoma

• isointense: an ancillary feature that favors benignity

• hypointense: an ancillary feature that favors malignancy (cannot upgrade lesion past LR4)

  • although often seen in HCC, it is not specific and may also represent

    • metastases (LR-M)

    • hepatic hemangioma

    • some dysplastic nodules and siderotic nodules 

    • focal fat deposition

    • confluent fibrosis

  • if isointense on earlier sequences, then ≥LR3

Practical points

• hepatobiliary phase hypointensity is not "washout" 

• "washout" may be problematic to evaluate because the signal intensity of the liver is changing over time, relative to the lesion

  • "washout" during the portal venous phase is reliable

  • "washout" during the transitional stage is less reliable as this may be due to relative parenchymal contrast uptake or true lesion washout ³²

• Eovist/Primovist has been associated with transient tachypnea during imaging in some patients (~15%) ³²

  • this may render the arterial phase suboptimal, making the important evaluation of hypervascularity suboptimal

  • reduced arterial enhancement comparted to extracellular contrast agents ³²

• the capsule around some HCCs may be more difficult to identify

• 10-15% of HCC demonstrate hepatobiliary phase hyperintensity ²

• hepatobiliary contrast reduces the ability to identify portal vein thrombus

• infiltrative HCC may be harder to visualize than with extracellular contrast agents

• Eovist/Primovist is not accurate at delineating the margins of TACE or RF ablation cavities

• some hepatic lesions (such as hepatic hemangiomas) may be more difficult to evaluate with hepatobiliary imaging than with extracellular imaging