OBJECTIVE: The purpose of this study was to determine the diagnostic accuracy of rim and segmental MRI enhancement of hepatic metastasis of colorectal cancer after administration of mangafodipir trisodium (MnDPDP). SUBJECTS AND METHODS: Sixty-one patients with a potentially resectable hepatic metastasis of colorectal cancer consecutively underwent breath-hold T1-weighted MRI in the axial and coronal planes 30 minutes and 24 hours after administration of MnDPDP. For each lesion, the presence or absence of rim enhancement and segmental enhancement 30 minutes and 24 hours after contrast administration was recorded. These features were evaluated separately for lesions 10 mm in diameter or larger and lesions smaller than 10 mm. The nature of each lesion was determined at histopathologic examination (n = 29) and on follow-up imaging (n = 32). RESULTS: Two hundred thirty lesions were identified at MRI: 210 lesions were metastatic, and 20 were benign. Rim enhancement was observed around 22 of 210 (10%) of the metastatic lesions at 30 minutes and 199 of 210 (95%) of metastatic lesions at 24 hours. Rim enhancement at 24 hours had 94.8% (95% CI, 91.8-97.8%) sensitivity, 90.0% (68.3-98.8%) specificity, 99.0% (97.6-100%) positive predictive value, 62.1% (42.3-79.3%) negative predictive value, and 94.3% (91.4-97.3%) diagnostic accuracy for metastasis. Segmental enhancement was infrequently seen (34/210; 16%) at 24 hours but had 100% (89.7-100%) positive predictive value for metastasis. CONCLUSION: Rim and segmental enhancement at MRI 24 hours after MnDPDP administration enabled accurate characterization of hepatic colorectal metastasis. These features may aid in preoperative mapping of hepatic tumor burden and disease distribution in patients with colorectal cancer.
OBJECTIVE: The purpose of this study was to determine the diagnostic accuracy of rim and segmental MRI enhancement of hepatic metastasis of colorectal cancer after administration of mangafodipir trisodium (MnDPDP). SUBJECTS AND METHODS: Sixty-one patients with a potentially resectable hepatic metastasis of colorectal cancer consecutively underwent breath-hold T1-weighted MRI in the axial and coronal planes 30 minutes and 24 hours after administration of MnDPDP. For each lesion, the presence or absence of rim enhancement and segmental enhancement 30 minutes and 24 hours after contrast administration was recorded. These features were evaluated separately for lesions 10 mm in diameter or larger and lesions smaller than 10 mm. The nature of each lesion was determined at histopathologic examination (n = 29) and on follow-up imaging (n = 32). RESULTS: Two hundred thirty lesions were identified at MRI: 210 lesions were metastatic, and 20 were benign. Rim enhancement was observed around 22 of 210 (10%) of the metastatic lesions at 30 minutes and 199 of 210 (95%) of metastatic lesions at 24 hours. Rim enhancement at 24 hours had 94.8% (95% CI, 91.8-97.8%) sensitivity, 90.0% (68.3-98.8%) specificity, 99.0% (97.6-100%) positive predictive value, 62.1% (42.3-79.3%) negative predictive value, and 94.3% (91.4-97.3%) diagnostic accuracy for metastasis. Segmental enhancement was infrequently seen (34/210; 16%) at 24 hours but had 100% (89.7-100%) positive predictive value for metastasis. CONCLUSION: Rim and segmental enhancement at MRI 24 hours after MnDPDP administration enabled accurate characterization of hepatic colorectal metastasis. These features may aid in preoperative mapping of hepatic tumor burden and disease distribution in patients with colorectal cancer.
Authors: D M Koh; G Brown; A M Riddell; E Scurr; D J Collins; S D Allen; I Chau; D Cunningham; N M deSouza; M O Leach; J E Husband Journal: Eur Radiol Date: 2008-01-12 Impact factor: 5.315