Helical CT for detecting focal liver lesions in patients with breast carcinoma: comparison of noncontrast phase, hepatic arterial phase, and portal venous phase.

PURPOSE

Our goal was to compare noncontrast phase (NCP), hepatic arterial phase (HAP), and portal venous phase (PVP) helical CT for the detection of focal liver lesions in patients at risk for having metastases from breast carcinoma.

METHOD

Eighty-four consecutive CT scans in 80 women with known or suspected liver metastases from breast carcinoma were prospectively evaluated with triple phase helical CT. After NCP, Isovue 300 was administered at 3 ml/s for 40 s, then 2 ml/s for 30 s, with scan delays of 25 s (HAP) and 76 s (PVP), slice thickness of 7 mm, and pitch of 1:1. Two reviewers evaluated each phase for focal liver lesions in a blinded and random fashion followed by side-by-side review for consensus.

RESULTS

By consensus, 40 CT scans were normal and 44 CT scans had a total of 105 lesions (46 lesions were graded malignant). PVP detected 39 (85%), HAP 27 (59%), and NCP 28 (61%) malignant lesions. Two malignant lesions were seen only on HAP, 3 only on NCP, and 10 only on PVP. The remainder of lesions were seen on more than one phase. PVP was graded best for detecting lesions in 27 (61%), HAP best in 7 (16%), NCP best in 4 (9%), and PVP equivalent to HAP in 6 (14%) of the 44 cases with lesions.

CONCLUSION

In our breast cancer patient population, PVP was superior to NCP and HAP for liver lesion detection. Because no CT scan was converted from negative to positive due to the addition of NCP or HAP, the routine use of these two phases cannot be justified when the clinical concern is the presence or absence of metastases.