Is Non-Contrast CT Adequate for the Evaluation of Hepatic Metastasis in Patients Who Cannot Receive Iodinated Contrast Media?

Objective To evaluate the appropriateness of follow-up with only non-enhanced CT (NECT) in patients with gastrointestinal cancer. Subjects and Methods This retrospective study included 323 patients with colorectal and gastric cancer who underwent two consecutive CT examinations (CT1 and CT2), including non-contrast and portal venous phase CT images, with an interval of 1 year. Patients were divided into 2 groups: Group A included patients with no hepatic metastasis on CT1 and with or without newly developed metastasis on CT2 to evaluate the diagnostic performance of NECT for detecting newly developed hepatic metastasis; Group B included patients with known hepatic metastasis both on CT1 and CT2 to evaluate the accuracy of NECT for the assessment of hepatic metastasis based on RECIST criteria (version 1.1). Contrast-enhanced CT (CECT) images were considered as reference standards. Results Group A included 172 patients (M:F = 107: 65; mean age, 62.6 years). Among them, 57 patients had 95 metastases (mean size, 2.2 +/- 1.3 cm). Per patient and per lesion sensitivity for diagnosing newly developed hepatic metastasis was 56.1-66.7% and 52.6-56.8%, respectively. In terms of small metastases (<1.5 cm), per lesion sensitivity was significantly decreased to 28.1-34.4% (P<0.05). Metastasis size measurements were significantly smaller on NECT (P<0.001) compared with reference standards. In Group B, the accuracy of response evaluation based on RECIST criteria was 65.6-72.2%. Conclusions NECT showed inadequate diagnostic performances in both detecting newly developed hepatic metastasis and evaluating the response of hepatic metastasis based on RECIST criteria.