PDF
ePub
Citation
Print
Abstract
Purpose
We wanted to evaluate the value of a lymph node specific MR contrast agent, Gadofluorine M, for the differentiation of hyperplastic and metastatic lymph nodes.
Materials and Methods
This study included thirty-one rabbits. In ten rabbits, an injection of egg yolk or feces of rat into the calf muscles induced hyperplasia of the lymph node. In sixteen rabbits, metastasis of the lymph node was induced by implantation of VX2 tumor. Five rabbits were normal control models. We acquired the T1-, T2-weighted and SPGR coronal imaging before enhancement with 1.5 T MR. After injection of Gadofluorine M (5 micromol/mL, total amount: 4 mL) interstitially into the interdigital skin fold of the hind limb, we acquired the SPGR coronal imaging at 15, 30, 60 and 90 minutes. We calculated the signal-to-noise ratios on the sequential images, and we recorded the number, size and location of the popliteal and iliac lymph nodes. Three readers assessed the state of the lymph nodes according to the pattern of enhancement: they were deemed hyperplastic nodes when totally enhanced and as metastatic nodes when there was no or partial enhancement. We also compared the imaging patterns with the histopathological results.
Results
Among the 26 hyperplasia- or metastasis-induced rabbits, two rabbits were excluded because of failure to be enhanced. Histopathologic evaluation of the 24 rabbits detected one hundred seventeen lymph nodes: forty-six lymph nodes in nine hyperplasia-induced rabbits and seventy-one (metastasis in twenty-eight) lymph nodes in fifteen metastasis-induced rabbits. Out of one hundred two lymph nodes that were larger than 5 mm in size, MR enabled us to detect one hundred one lymph nodes (99.1%). The means of sensitivity, specificity, and the positive and negative predictive values for the diagnosis of lymph node metastasis by three readers were 97.6% (82/84), 98.2% (215/219), 95.3% (82/86), and 99.1% (215/217), respectively (p < 0.05).
XML Download