Helical CT versus CT arterial portography in the detection of hepatic metastasis of colorectal carcinoma.

Abstract

The study is a prospective comparison of helical CT with nonhelical CT arterial portography (CTAP) in the detection of liver metastases from colorectal carcinoma, using surgical and histologic findings as the gold standard. Thirty-five patients with colorectal carcinoma and suspected liver metastases were prospectively examined with helical CT and CTAP before surgery. In nine cases, surgery was not performed. In the remaining 26 patients, imaging results were correlated with surgical and pathologic findings. Three radiologists prospectively assessed metastatic involvement with both techniques. The results were compared with pathologic and surgical findings on a lesion-by-lesion basis. In a second phase, three radiologists not directly concerned in the design of the study independently assessed metastatic involvement of the liver as revealed on helical CT and CTAP on a segment-by-segment basis with a five-level scale of confidence. Results were analyzed by receiver operating characteristic methods. The results of the histologic study disclosed 50 metastatic lesions. Helical CT had a sensitivity of 76% (38/50) and a positive predictive value of 90%. CTAP had a sensitivity of 74% (37/50) and positive predictive value of 69%. Receiver operating characteristic analysis revealed a greater area under the curve (Az index), 0.96, for helical CT than for CTAP (0.86). Differences were statistically significant (p < .001). Helical CT is superior to nonhelical CTAP in the detection of hepatic metastases from colorectal carcinoma.