Abstract
271
Objectives Assess the accuracy of the FDG-PET/MRI fused images compared to MRI spectroscopy in the evaluation of the recurrent brain tumors in patients previously treated with radiation and/or surgery.
Methods This is a retrospective study of 18 sets of PET/MRI and MRI spectroscopy in 14 patients with brain tumors previously treated with radiation and/or surgery: 13 WHO III and IV primary and 1 metastatic brain tumors. A 30 min 3D PET was acquired after IV injection of 185 MBq of FDG and 30 min uptake period. Multisequence MRI images were acquired on a 1.5T or 3T magnet before and after gadolinium. MR spectroscopies were performed using single voxel and multivoxel techniques. The axial PET images were fused to the axial T1 post gad images. PET and MRI spectroscopy images were interpreted by an experienced nuclear medicine and neuroradiologist, respectively. Patients were followed for a minimum of 3 months. The “gold” standard for determining the presence or absence of tumor recurrence included: biopsy, follow up studies (PET and/or MRI) and clinical follow up (seizures, headaches, mental status changes or death). The sensitivity, specificity, and accuracy of fused PET/MR images and of MRI spectroscopy were determined.
Results Fused FDG-PET/MRI images had sensitivity of 100%, specificity of 75% and accuracy of 94%. A false positive PET/MRI study occurred where normal gray matter FDG uptake was confused for recurrence. MRI spectroscopy had sensitivity of 71%, specificity of 25% and the accuracy of 61%. The p value comparing sensitivities is 0.0124 and for specificities is not significant due to the limited number of the true negative patients.
Conclusions Our results show that FDG-PET/MRI fused images may be more sensitive, specific and accurate than MRI spectroscopy for detecting the recurrence of high grade brain tumors. The FDG-PET/MRI fusion or hybrid scanners may also provide better diagnostic value by guiding the biopsy or be used as treatment planning for radiation