Although the actual number of additional lesions detected during the hepatic arterial phase (HAP) alone with dual-phase helical CT of the liver is generally small, there may be benefits above and beyond that of detecting lesions. The purpose of our study is to determine the overall contribution or added-value of the HAP in patients with suspected hypervascular lesions.

Data were collected prospectively on 260 patients with known or suspected hypervascular malignancy who underwent a total of 284 dual-phase helical CT scans of the liver (150-175 ml of Isovue 300 at 4-5 ml/sec; scan delay for the HAP- 30 seconds, and for the portal venous phase (PVP)- 60-65 seconds). Ten abdominal radiologists evaluated the HAP and PVP in a side-by-side fashion and then were queried as to the contribution of the HAP including the detection and characterization of focal hepatic lesions, the confidence of interpretation and the characterization of perfusion abnormalities. Timing of the HAP (early, appropriate or late) was also assessed. An appropriately timed HAP was defined qualitatively as having: 1) no hepatic venous enhancement, 2) minimal hepatic parenchymal enhancement, and 3) early portal venous enhancement.

In 215 (76%) examinations the HAP added no additional information to the PVP; in 118 (42%) of these there were no lesions and in 97 (34%) the lesions were obvious on the PVP. In 69 (24%) examinations the HAP was contributory; the confidence in lesion detection was increased in 30 (11%), the confidence in lesion exclusion was increased in 20 (7%), the number of lesions detected was increased in 13 (5%), and both the number and confidence was increased in 7 (2%) examinations. The HAP was helpful for characterization of perfusion abnormalities in 13 (5%) examinations. There was considerable variation in how often the various readers felt the HAP made a contribution, ranging from 10 to 42%. The timing of the HAP was judged to be too early in 7%, appropriate in 83%, and too late in 10%. There was no significant difference between timing (early, appropriate, late) and the overall contribution of the HAP.

Although it may not be rational to include the HAP routinely in patients with suspected hypervascular lesions, based on the small number of additional lesions detected, an overall diagnostic contribution averaging 24% may justify this additional phase.