Diagnostic yield of MR-guided liver biopsies compared with CT- and US-guided liver biopsies

PURPOSE: To compare diagnostic yield and complication rates of magnetic resonance (MR)-guided versus computed tomography (CT)- and ultrasound (US)-guided liver biopsies. MATERIALS AND METHODS: MR-, CT-, and US-guided liver biopsies performed between 9/96 and 9/98 were compared. Sixty patients (21 men and 39 women, mean age 60 years) underwent MR-guided biopsy of liver lesions. Thirty patients (16 men and 14 women, mean age 59 years) underwent CT-guided biopsy. Eighteen patients (seven men and 11 women, mean age 50 years) underwent US-guided biopsy. MR procedures were performed in an open-configuration 0.5-T Signa SP MR unit. Lesion localization used standard T1 and T2 sequences, whereas biopsies were performed with multiplanar spoiled gradient recalled echo and fast gradient recalled echo sequences. A coaxial system with an MR-compatible 18-gauge stabilizing needle and a 21-gauge aspiration needle was used to obtain all samples. In CT and US procedures, a 19-gauge stabilizing needle and a 21-gauge aspiration or a 20-gauge core biopsy needle were used. A cytotechnologist was present to determine the adequacy of samples. RESULTS: MR had a diagnostic yield of 61%. CT and US had diagnostic yields of 67% and 61%, respectively. No serious complications were reported for MR and US procedures. Two CT biopsies resulted in postprocedural hemorrhage. One patient required surgical exploration and died. CONCLUSIONS: MR-guided biopsy of liver lesions with use of a 0.5-T open-configuration magnet is safe and accurate when compared with CT and US. No statistical difference was observed between the diagnostic yield of biopsies performed with MR, CT, and US guidance. MR enabled biopsy of a number of lesions in the hepatic dome and lesions with low contrast, which would normally be difficult to sample safely with use of CT or US.