Optimization of imaging diagnosis of 1-2 cm hepatocellular carcinoma: An analysis of diagnostic performance and resource utilization

Background & Aims: To determine the optimal imaging scan or combinations in terms of diagnostic performance and resource utilization for 1-2 cm nodules found on surveillance for hepatocellular carcinoma. Methods: Eighty-four cirrhotic patients with 101, 1-2 cm nodules (34 malignant, 67 non-malignant) prospectively underwent standardized contrast-enhanced ultrasound, CT, and MRI scans. Sensitivity/specificity and potential imaging/biopsy utilization of individual imaging modalities and two-modality combinations performed at the same time (coincidental) or in sequence were measured. Final diagnosis was determined by biopsy (23), growth (10), recurrence (1), or stability in size for 18 months (67). Results: For single imaging scans, sensitivities/specificities ranged between 53-62% and 91-100%. When two scans were combined requiring both to be positive, sensitivities/specificities ranged between 29-41% and 99-100%. When two scans were combined sequentially, requiring only one to be positive, sensitivities/specificities ranged between 74-89% and 91-99%. When comparing combination of two positive tests (MRI and CT) to MRI alone, there was a significant drop in sensitivity (41% vs. 62%, p = 0.02), no change in specificity (both 100%), with twice as many scans performed, and 9% rise in potential biopsies or 7% rise in follow-up scans. When comparing the combination of MRI then CT (if MRI negative) to MRI alone, there was an insignificant rise in sensitivity (74% vs. 62%, p = 0.13), drop in specificity (97% vs. 100%), with 77% more scans performed, and 6% drop in potential biopsies or 7% drop in potential follow-up scans. Conclusions: Single imaging scans have similar specificity to two coincidental positive scans with much less resource utilization. Sequential imaging provides the best sensitivity but with diminished specificity. (C) 2010 European Association for the Study of the Liver. Published by Elsevier B.V. All rights reserved.