Double arterial phase multi-detector row helical CT for detection of hypervascular hepatocellular carcinoma

From: Takamichi Murakami, MD, PhD,* Tonsok Kim, MD,* Masatoshi Hori, MD, PhD,* Michael P. Federle, MD Department of Radiology, Osaka University Graduate School of Medicine 2–2 Yamadaoka, Suita-city, Osaka 565-0871, Japan* Department of Radiology, University of Pittsburgh Medical Center, Pa e-mail: murakami@radiol.med.osaka-u.ac.jp Editor: Optimal timing for acquisition of the arterial dominant set of computed tomographic (CT) images can be influenced by numerous variables, such as patient size and cardiovascular status. The fast imaging ability of multi–detector row helical CT to scan the entire liver in 10 seconds or less gives us the opportunity to acquire two separate sets of CT images in the liver within the time period generally regarded as the hepatic arterial dominant phase (double arterial phase). The double arterial phase images were expected to cover optimal timing of the arterial dominant phase. In our previous study (1), we reported that the double arterial phase, including the early and late arterial phases, could improve the detection of hepatocellular carcinoma (HCC). However, Dr Laghi and colleagues (2) reported that the use of an early arterial phase did not result in any improvement. There were a few differences in scanning parameters between their study and ours. The amount of contrast medium used for test injection was 20 mL in their study and 15 mL in our study, and scanning time of the double arterial phase was 24 seconds in their study and 26 seconds in our study. These may be some causes of the discrepancy. Another possible explanation for this discrepancy was suspected by Dr Laghi and colleagues to be the different mean sizes of HCC nodules. We suspected that the main possible reason for this discrepancy depended on patient size in each study—in other words, total amount of contrast medium used for patients. In both studies, the volume of contrast agent delivered was 2 mL per kilogram of body weight. The volume of contrast agent administered in our study ranged from 82 to 182 mL (mean, 114.2 mL). This was less than that in their study, which ranged from 120 to 168 mL (mean, 128 mL), because the body weight of our patients was less than that of patients in their study. Time to aortic peak enhancement was reported to be proportional to the injection duration or injection rate at a given injection duration; therefore, time to peak enhancement is shorter with shorter injection duration (3). The magnitude of peak contrast enhancement increases linearly with the total mass (concentration of contrast medium times volume of iodine injected at a given injection rate) (3,4). The possibility existed that aortic peak enhancement in smaller patients in our study, such as those who weighed less than 50 kg (injection duration less than 20 seconds at an injection rate of 5 mL/sec, although scanning time of the double arterial phase was 26 seconds), was reached sooner, and aortic enhancement decreased more rapidly in the late arterial phase. Time to peak arterial enhancement of HCC nodules in such smaller patients was also shorter than that in larger patients, and hypervascular HCC nodules in such smaller patients might show peak enhancement on the early arterial phase images. Our recent data (unpublished data, 2002), obtained by using multi–detector row helical CT with 0.5-second scanning (which can be used to obtain double arterial phase images in about 20 seconds), showed that late arterial phase CT images have significantly superior sensitivity to that of early arterial phase CT images. Our data also showed that there is no significant difference between the late arterial and double arterial phase images for detection of HCC nodules