From the Literature Journal of Neurosurgery 97: 794, October 2002 Issue.
Histopathological validation of a three-dimensional magnetic resonance spectroscopy index as a predictor of tumor presence.
Mcknight et al. from the Departments of Radiology, Neurosurgery, and Radiation Oncology at UCSF.
In this article the authors examined the queston of whether the choline-NAA index (CNI) determined on MR spectroscopy could be used to detect metabolically active tumor in T2 hyperintense yet non-enhancing regions in patients with intrinsic brain tumors.
The authors first used histologic criteria to generate a proposed cut-off for CNI to predict tumor versus non-tumorous tissue. They found that in non-tumorous biopsies that the CNI was 1.7 while in Grade II, Grade III and Grade IV astrocytoma, this ratio was over 5.3. Using a CNI of 2.5, tumor could be distinguished from non-tumorous tissue with 90% specificity and 86% specificity.
Using this index, the authors then examined what percentage of MR defined lesion by T2 or T1 with constrast had a CNI greater than 2.5. More than 1/3 of cases had CNI values consistent with tumor in the T2 intense region surrounding the contrast enhanced zone.
The authors also assessed heterogeneity of the CNI in the T2 hyperintense regions yet nonenhancing regions of Grade II, III, and IV astrocytomas. In each of these grade tumors, >40% of the total T2 hyperintense, non enhancing volume had a CNI greater than 2.5.
Commentary: The non-invasive detection of metabolically active tumor continues to be actively pursued as these authors describe. One important implication of the current study is that it may be possible to target the most metabolically active regions of non-enhancing tumors as defined by a CNI ratio of >2.5. These so called “MR defined hotspots” may be useful to improve the diagnostic yield in the biopsy of non-enhancing lesions. ~Bob Carter, M.D., ~Fred Barker, M.D.