Objective: Hepatic metastases from neuroendocrine tumors are often markedly hyperintense on unenhanced T2-weighted MR images, making their appearance similar to that of cavernous hemangiomas. In contrast, cavernous hemangiomas show characteristic enhancement on dynamic gadolinium chelate-enhanced gradient-recalled echo MR images. The purpose of this study was to determine the relative merits of dynamic gadolinium chelate-enhanced gradient-recalled echo MR imaging versus MR imaging with unenhanced spin-echo pulse sequences for distinguishing between hepatic metastases from neuroendocrine tumors and cavernous hemangiomas.
Materials and methods: The unenhanced spin-echo and dynamic gradient-recalled echo MR images obtained after IV administration of a gadolinium chelate in 28 patients (14 patients with pathologically proven hepatic metastases from neuroendocrine tumors and 14 patients with hepatic cavernous hemangiomas) were reviewed blindly and independently by three interpreters. Unenhanced spin-echo and dynamic gadolinium chelate-enhanced gradient-recalled echo MR images were compared for accuracy in characterizing liver lesions.
Results: The most intense enhancement of hepatic metastases from neuroendocrine tumors was observed on early dynamic gadolinium chelate-enhanced gradient-recalled echo MR images; enhancement was peripheral in four patients, global and heterogeneous in seven patients, and global and homogeneous in three patients. On late dynamic gadolinium chelate-enhanced gradient-recalled echo MR images, enhancement of hepatic metastases from neuroendocrine tumors was predominantly peripheral in five patients, global and heterogeneous in five patients, and global and homogeneous in four patients. Differentiation between cavernous hemangiomas and hepatic metastases from neuroendocrine tumors was impossible in five cases with unenhanced spin-echo MR imaging alone, in five cases with dynamic gadolinium chelate-enhanced gradient-recalled echo MR imaging alone, and in no case with the combination of unenhanced spin-echo MR imaging and dynamic gadolinium chelate-enhanced gradient-recalled echo MR imaging. In comparison with unenhanced spin-echo MR imaging alone or dynamic gadolinium chelate-enhanced gradient-recalled echo MR imaging alone, the combination of unenhanced spin-echo MR imaging and dynamic gadolinium chelate-enhanced gradient-recalled echo MR imaging allowed significantly (p < .001) clearer differentiation between hepatic metastases from neuroendocrine tumors and cavernous hemangiomas.
Conclusions: Early enhancement and heterogeneity on dynamic gadolinium chelate-enhanced gradient-recalled echo MR images are the most common features of hepatic metastases from neuroendocrine tumors. The combination of unenhanced spin-echo and dynamic gadolinium chelate-enhanced gradient-recalled echo MR images allows more accurate characterization of hepatic metastases from neuroendocrine tumors and clearer differentiation from cavernous hemangiomas.