Toshinori Hirai MD, Presenter: Nothing to Disclose

Hideo Nakamura MD, Abstract Co-Author: Nothing to Disclose

Mika Kitajima MD, Abstract Co-Author: Nothing to Disclose

Yoshinori Shigematsu, Abstract Co-Author: Nothing to Disclose

Hiroyuki Uetani, Abstract Co-Author: Nothing to Disclose

Yasuyuki Yamashita MD, Abstract Co-Author: Nothing to Disclose

To evaluate which quantitative indices derived from perfusion magnetic resonance imaging (MRI) or positron emission tomography (PET) yield the best diagnostic value for differentiating between tumor recurrence (TR) and radiation necrosis (RN).

We retrospectively evaluated 3-T dynamic susceptibility contrast perfusion MRI and fluorodeoxyglucose (FDG)- and methionine (MET)-PET studies performed on 16 consecutive patients with TR (n = 8) or RN (n = 8). A definitive diagnosis was made on the basis of pathologic examination for TR and on the basis of pathologic examination or clinical follow-up study for RN. Indices characterizing solid areas of the largest lesions were determined; they included the ratio of the lesional maximum relative cerebral blood volume (rCBV) to the contralateral normal-appearing white matter (max rCBV ratio) and the ratio of the maximum standardized uptake values (SUVmax) of the lesion to the uptake in the contralateral normal-appearing gray matter (L/Nmax) on FDG- and MET-PET images. These indices were compared in the two groups using an independent t-test. Receiver operating characteristic (ROC) curves and the area under the curve (AUC) were calculated to evaluate diagnostic performance; we also performed pairwise comparison of the ROC curves.

The max rCBV ratio was higher in TR than RN (3.29 ± 1.77 vs. 1.49 ± 0.82; p < 0.05). Although there were no significant differences, L/Nmax on FDG-PET and MET-PET was higher in TR than RN (1.27 ± 0.57 vs. 0.89 ± 0.28 and 2.64 ± 0.85 vs. 2.08 ± 0.55, respectively). In ROC analysis, the best diagnostic performance was obtained for the max rCBV ratio (sensitivity 100%; specificity 80%; AUC 0.90; cut-off 1.67) followed by L/Nmax on FDG-PET (sensitivity 62.5%; specificity 85.7%; AUC 0.732; cut-off 1.01). Pairwise comparison of the three ROC curves revealed no significant differences.

In patients with treated glioma the max rCBV ratio tends to yield the best diagnostic performance for differentiating between TR and RN, although there were no significant differences among the three indices derived from perfusion MRI and PET.

Among perfusion MRI and PET studies, perfusion MRI might be most useful for distinguishing TR from RN in patients with treated glioma.

Hirai, T, Nakamura, H, Kitajima, M, Shigematsu, Y, Uetani, H, Yamashita, Y, Differentiation of Tumor Recurrence from Radiation Necrosis in Treated Gliomas: Which Indices from Perfusion MRI or PET Images Yield Optimal Diagnostic Values?.  Radiological Society of North America 2011 Scientific Assembly and Annual Meeting, November 26 - December 2, 2011 ,Chicago IL. http://archive.rsna.org/2011/11034658.html