OBJECTIVE: The purpose of this study was to evaluate the usefulness of double arterial phase CT for the detection of small hypervascular hepatocellular carcinomas, using an automated bolus-tracking technique to initiate the hepatic arterial phase CT. MATERIALS AND METHODS: Double arterial and late phase contrast-enhanced helical CT scans were obtained on 287 consecutive patients suspected of having hepatocellular carcinoma. These included 56 patients with 90 small (< or 3 cm) hepatocellular carcinomas and 50 patients with no hepatocellular carcinomas. CT scans of these patients were interpreted by three reviewers. The first arterial phase scan was initiated automatically 10 sec after the bolus-tracking program detected the threshold enhancement of 50 H in the abdominal aorta. Three reviewers interpreted the late phase CT scans in combination with the first, second, or both hepatic arterial phases. Measures of the reviewers' detection of hepatocellular carcinoma included analysis of interobserver variation, sensitivity, specificity, and area under receiver operating characteristic curve (A(z)). RESULTS: The time elapsed from bolus initiation to threshold aortic enhancement ranged from 10 to 24 sec (mean, 13 sec), resulting in initiation of the first arterial phase CT scan from 20 to 34 sec (mean, 23 sec). The combination of late phase CT and both first and second arterial phase images showed significantly better performance than the combination of the late phase and either the first or second arterial phases, although the difference was most evident in comparison with the combination of second arterial and late phases. CONCLUSION: An automated bolus-tracking program can be used to optimize the timing of hepatic arterial phase CT. Multiphasic CT performed using this technique is useful in detection of small hepatocellular carcinoma.
OBJECTIVE: The purpose of this study was to evaluate the usefulness of double arterial phase CT for the detection of small hypervascular hepatocellular carcinomas, using an automated bolus-tracking technique to initiate the hepatic arterial phase CT. MATERIALS AND METHODS: Double arterial and late phase contrast-enhanced helical CT scans were obtained on 287 consecutive patients suspected of having hepatocellular carcinoma. These included 56 patients with 90 small (< or 3 cm) hepatocellular carcinomas and 50 patients with no hepatocellular carcinomas. CT scans of these patients were interpreted by three reviewers. The first arterial phase scan was initiated automatically 10 sec after the bolus-tracking program detected the threshold enhancement of 50 H in the abdominal aorta. Three reviewers interpreted the late phase CT scans in combination with the first, second, or both hepatic arterial phases. Measures of the reviewers' detection of hepatocellular carcinoma included analysis of interobserver variation, sensitivity, specificity, and area under receiver operating characteristic curve (A(z)). RESULTS: The time elapsed from bolus initiation to threshold aortic enhancement ranged from 10 to 24 sec (mean, 13 sec), resulting in initiation of the first arterial phase CT scan from 20 to 34 sec (mean, 23 sec). The combination of late phase CT and both first and second arterial phase images showed significantly better performance than the combination of the late phase and either the first or second arterial phases, although the difference was most evident in comparison with the combination of second arterial and late phases. CONCLUSION: An automated bolus-tracking program can be used to optimize the timing of hepatic arterial phase CT. Multiphasic CT performed using this technique is useful in detection of small hepatocellular carcinoma.