Characterization of intracranial space-occupying lesions by ⁹⁹(m)Tc-Tetrofosmin SPECT

Abstract

Differentiating neoplastic from non-neoplastic intracranial lesions is of paramount importance for patient management. Benign lesions can have many of the features of malignant brain tumors on both computed tomography (CT) and conventional magnetic resonance imaging (MRI). Herewith, we set out to investigate the role of (99m)Tc-Tetrofosmin (⁹⁹(m)Tc-TF) brain SPECT in the differentiation of neoplastic from non-neoplastic intracranial lesions. We prospectively studied, between September 2004 and September 2009, patients with intracranial lesions suspected of being tumors on CT/MRI that were operated on. All patients with suspected tumor on CT/MRI underwent ⁹⁹(m)Tc-TF brain SPECT within a week before surgery and CT/MRI studies. Radiotracer accumulation in intracranial lesions was assessed visually and then a semiquantitative method of image analysis was applied, by calculating the lesion-to-normal (L/N) uptake ratio. We compared the L/N ratios between low-grade gliomas and high-grade gliomas, low and high-grade gliomas and intra-axial non-neoplastic lesions, low and high-grade gliomas and metastases, and typical versus anaplastic meningiomas Ninety patients suffered from neoplastic lesions and 16 harboured non-neoplastic pathologies. There was a significant difference between low-grade gliomas and high-grade gliomas (P = 0.0019). ROC analysis provided 2.8 as the optimum cutoff value thresholding discrimination between these two groups, with 91.3% sensitivity and 83.3% specificity. When comparing gliomas (low and high-grade) with intra axial non-neoplastic lesions the difference was statistically significant (P < 0.0001). There was no statistically significant difference between gliomas (low and high-grade) and metastases. Regarding meningiomas, there was a statistically significant difference between typical and anaplastic meningiomas (P = 0.0002). ROC analysis provided 9.6 as the optimum cutoff value thresholding discrimination between these two groups, with 96% sensitivity and 100% specificity. (99m)Tc-TF brain SPECT may differentiate neoplastic from non-neoplastic intracranial pathologies and could prove useful for pre-surgical evaluation of intracranial lesions.