Multiphase multislice spiral CT for liver assessment: optimization in cirrhotic patients

The aim of our study was to optimize a multiphase study protocol with double arterial phase acquisition in a patient population with cirrhosis using a multislice spiral CT scanner.

Thirteen patients (10 males, 3 females, mean age 58 years) with known cirrhosis were selected for the study. All examinations were performed with a multislice spiral CT scanner (Somatom Plus 4 Volume Zoom; Siemens, Erlangen, Germany). Images were acquired using the following parameters: slice collimation, 2.5 mm; slice thickness, 3.0 mm; table feed, 10.8 mm/sec; mAs, 165; kVp, 120. Four scans of the hepatic parenchyma were obtained after the administration of contrast material. The first pass (early arterial phase) was acquired in a cranio-caudal direction; the second pass (late arterial phase) was acquired in a caudo-cranial direction. Early and late arterial phases were obtained during a single breath-hold of 24 sec. The third pass (portal-venous phase) was acquired with a 60-sec delay time after contrast material injection. The fourth pass (equilibrium phase) was obtained with a 180-sec delay time. Optimal delay time to start CT acquisition was assessed by means of injecting a 20-ml minibolus of contrast material and by performing serial dynamic scans every two sec at the level of the hepatic hilum. The time of peak aortic enhancement was used as the start time for the early arterial phase. Attenuation values of aorta, portal vein, and liver parenchyma were calculated in all the acquisitions. CT data from the early arterial phase were used to produce three-dimensional angiographic images of the hepatic and mesenteric circulation.

The enhancement of liver parenchyma progressively increased from pre-contrast phase to portal-venous and equilibrium phases. The highest difference in attenuation values between aorta and hepatic parenchyma was observed during the second acquisition (early arterial phase, 247.78+/-106.29 HU) rather than during the third acquisition (late arterial phase, 185.72+/-109.23 HU); this difference was statistically significant (p<0.01).

Results from our study emphasize the potential of multiphase acquisition in the evaluation of cirrhotic patients; in particular, the use of an early arterial phase is useful for studying the hepatic and mesenteric vascular anatomy, whereas the late arterial and the portal-venous phases are of paramount importance for adequate evaluation of liver parenchyma and focal lesions. Further studies are needed to evaluate whether the benefits deriving from double arterial phase acquisition would justify the increase in cost and patient radiation exposure.