Quantitative measurement of hepatic portal perfusion by multidetector row CT with compensation for respiratory misregistration

Our purpose was to determine whether hepatic portal perfusion assessed by multidetector row CT using compensation for respiratory misregistration can predict the severity of chronic liver disease. We carried out dynamic CT in 43 patients (chronic hepatitis: n=9; cirrhosis: n=24; normal liver: n=10). In this series, 20 patients had liver tumours. The CT protocol was designed to avoid respiratory artefacts and included two interscan breathing periods during the study. To compensate for respiratory misregistration, image sets in the same z-axis position were acquired from four-slice data on each scan, and the portal perfusion calculations were made according to the maximum slope method. Portal perfusion was compared with and without compensation for respiratory misregistration, and the different types of hepatic disease. In the liver tumour patients in particular, portal perfusion was compared with the degree of hepatic fibrosis in the liver sections. Portal perfusion in the patients without compensation for respiratory misregistration (1.10 ml min(-1)ml(-1)) was higher than that of those with compensation (0.99 ml min(-1)ml(-1); p=0.036). Hepatic portal perfusion of patients with chronic hepatitis (0.97 ml min(-1)ml(-1)) and liver cirrhosis (0.88 ml min(-1)ml(-1)) was less than that of patients with normal liver (1.32 ml min(-1)ml(-1); p=0.03, 0.001). Moderate correlation was seen between portal perfusion and the percentage of fibrosis in patients with liver tumours (r=0.55). Hepatic portal perfusion obtained by multidetector row dynamic CT using compensation for respiratory misregistration has the potential to improve non-invasive assessment of the degree of chronic liver disease.