Objective: To determine the utility of brain thallium-201 single photon emission computerized tomography (Tl-201 SPECT) combined with Toxoplasma serology for the diagnosis of focal CNS lesions in patients with AIDS.
Methods: Sixty-one consecutive HIV-infected patients with focal CNS lesion(s) on head computed tomography (CT) or MRI scan who underwent brain Tl-201 SPECT and serum Toxoplasma serology were evaluated, retrospectively. Thallium-201 uptake ratios were calculated by comparing lesion activity to contralateral scalp activity. Diagnoses were made by a combination of histology, serology, PCR, and empirical response to therapy. Toxoplasma serologies (IgG IFA) were compared in the patients with central nervous system (CNS) toxoplasmosis and those without CNS toxoplasmosis.
Results: Fifty-six patients were evaluable and a definitive diagnosis was made in 38 patients: toxoplasmosis (17), lymphoma (14), PML (three), Aspergillus (one), tuberculoma (one), Cryptococcus (one), varicella-zoster virus (one). Patients with lymphoma had significantly higher lesion/contralateral scalp ratios compared to patients without lymphoma: 1.03 vs. 0.67, P < 0.05. Using a cut-off of 0.90 for the lesion/scalp uptake ratios (based on analysis of ROC curves) the sensitivity and specificity for the diagnosis of lymphoma were 86% and 83%, respectively. Serum Toxoplasma IgG titres were significantly higher in patients diagnosed with toxoplasmosis compared to those with a diagnosis other than toxoplasmosis, 1:5444 vs. 1:15, P < 0.05. Only one patient with confirmed toxoplasmosis had a Toxoplasma serology < 1:256, while no patients without toxoplasmosis (including all lymphoma patients) had serologies > 1:256.
Conclusions: In a series of HIV-infected patients, Tl-201 SPECT was able to accurately differentiate primary brain lymphoma from other causes of focal CNS lesions in most patients; however, both false positive and false negative results occurred. By combining Tl-201 SPECT with serum Toxoplasma IgG, diagnostic accuracy was improved.