OBJECTIVE: Our goal was to compare short tau inversion recovery MRI (STIR) to CT during arterial portography (CTAP) in the detection of hepatic lesions. MATERIALS AND METHODS: Over a 24 month period, 40 patients evaluated for possible hepatic resection underwent T1, T2, and STIR MRI and CTAP. These studies were randomly ordered and interpreted by two independent readers. The total number of lesions and number of lesions within each hepatic segment were analyzed using McNemar's test. RESULTS: Readers 1 and 2 detected 95 and 91 lesions by CTAP and 78 and 83 by STIR, respectively, which was not significantly different (p = 0.1, reader 1; p = 0.4, reader 2). For both readers CTAP and STIR MRI detected more lesions than either T2 or T1 MRI (p < 0.01). In the medial segment, readers 1 and 2 detected 18 and 17 lesions by CTAP and 7 and 9 by STIR, respectively (p = 0.01, reader 1; p = 0.035, reader 2). CONCLUSION: Short tau inversion recovery MRI and CTAP are similar in total lesion detection except in the medial segment where lesions were more frequently identified by CTAP than STIR MRI.