Sommer WH, Zech CJ, Bamberg F, Auernhammer CJ, Helck A, Paprottka PM, Notohamiprodjo M, Reiser MF, Herrmann KA. Fluid-fluid level in hepatic metastases: a characteristic sign of metastases of neuroendocrine origin.
Eur J Radiol 2011;
81:2127-32. [PMID:
21978774 DOI:
10.1016/j.ejrad.2011.09.012]
[Citation(s) in RCA: 18] [Impact Index Per Article: 1.4] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 06/27/2011] [Revised: 09/07/2011] [Accepted: 09/10/2011] [Indexed: 11/28/2022]
Abstract
OBJECTIVES
The aim of this study is to determine MRI characteristics which indicate liver metastases of neuroendocrine tumors (NET) rather than metastases of other origin (non-NET).
METHODS
Sixty-nine patients with histopathologically proven liver metastases from NET and 69 patients with known liver metastases of other origin underwent MRI of the liver using a 1.5 T MR-scanner. Two board certified radiologists assessed presence of fluid-fluid-levels, number and distribution pattern, signal intensity (SI) characteristics, lesion homogeneity, presence of central necrosis and intratumoral hemorrhage in T2w and T1w non-contrast imaging. A multivariate logistic regression analysis was performed to determine the independent association of image findings and occurrence of NET.
RESULTS
Fluid-fluid-levels were identified in 19/69 of patients with NET-metastases, and in none of the patients in the control group (p<0.0001). Hyperintense SI in T1w imaging, markedly hyperintense SI in T2w imaging, a disseminated distribution pattern and intratumoral hemorrhage were indicative of NET metastases (p<0.05). After statistical adjustment for all significant MRI findings, fluid-fluid-levels (OR: 17.6, 95% CI: 1.9-166.5), strongly hyperintense SI in T2w (OR: 4.7, 95% CI: 1.8-12.7) and a disseminated distribution pattern (OR: 2.9, 95% CI: 1.1-7.4) were independent predictors for NET metastases.
CONCLUSIONS
The presence of fluid-fluid-levels is highly indicative of NET liver metastases and can be used as an independent predictor to distinguish them from metastases of other origin.
Collapse