Abstract

1537 Background: The evaluation of CNS lesions in AIDS patients is challenging as both CNS lymphoma and non-neoplastic lesions can have similar clinical presentations and imaging findings. Both Tl-201 SPECT and FDG PET imaging have been used to differentiate malignancy from infection. This study investigates the accuracy of FDG PET compared to Tl-201 SPECT in the diagnosis of CNS lymphoma in patients with AIDS-associated CNS lesions. Methods: Nine patients with AIDS who were found to have rim enhancing brain lesions on MRI underwent both Tl-201 SPECT and FDG PET imaging prior to diagnosis. Diagnosis was made by stereotactic brain biopsy in 6 pts and clinical follow-up in the remaining 3. FDG PET of the brain was performed using a dedicated PET/CT instrument (GE Discovery) 1 hour after 444 MBq FDG. Brain SPECT was performed on a dual head gamma camera (GE Hawkeye) with 185 MBq Tl-201. Images were reviewed along with MRI for anatomic correlation by 2 experienced nuclear medicine physicians blinded to the diagnosis. Results: 43 MRI lesions in 9 patients were evaluated. Clinically, 5 pts (18 lesions) were diagnosed with lymphoma (4 biopsy proven, 1 with known history of lymphoma), and 4 pts (25 lesions) were diagnosed with toxoplasmosis (2 with negative biopsies for malignancy, 2 with clinical course consistent with toxoplasmosis). FDG PET had true positive results for lymphoma in 5 of 5 pts, while Tl-201 SPECT was true positive in only 2 of 5. FDG PET had true negative results in 4 of 4 pts with toxoplasmosis, while Tl-201 had false positive results in 2 of 4. FDG PET had positive and negative predictive values of 100%, while Tl-201 SPECT had PPV 50% and NPV 40%. On a lesion basis, PET showed increased uptake of FDG in 8 of 18 brain lesions in patients with lymphoma, while SPECT showed increased uptake of Tl-201 in 4 of 18 lesions. Tl-201 was positive in 2 lesions in patients diagnosed with toxoplasmosis, while FDG had no false positive lesions. Conclusion: In this limited study, FDG PET proved to be superior to Tl-201 SPECT in the differentiation of CNS lymphoma from toxoplasmosis (overall accuracy 100% vs 44%, respectively). With the increasing availability of FDG PET, this test should probably supplant Tl-201 SPECT as the study of choice following conventional imaging (CT, MRI) of CNS lesions in AIDS patients. No significant financial relationships to disclose.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call