In patients at risk for acquired immunodeficiency syndrome who present with a mass lesion, a dilemma arises as to whether to treat empirically for toxoplasmosis or perform a brain biopsy. We present data that further define the indications for performing brain biopsy vs empiric treatment. We reviewed charts on 59 patients with acquired immunodeficiency syndrome—related disorders and cerebral mass lesions. Thirty-two patients met diagnostic criteria for toxoplasmosis. Bayesian analysis demonstrated that the prior probability of toxoplasmosis was increased by the presence of contrast enhancement on computed tomographic scans (0.68) and toxoplasmosis titers greater than 1:64 (0.81). Features associated with decreasing probabilities of toxoplasmosis included the absence of contrast enhancement on computed tomographic scans (0.29) and toxoplasmosis titers less than or equal to 1:64 (0.14). Ten percent of patients had complications of brain biopsy. Treatment with pyrimethamine and sulfadiazine produced complications in 29% and serious complications in 8% of treated patients. These data favor empiric therapy for patients with typical features of toxoplasmosis and brain biopsy for defined subsets of patients with atypical features.
(Arch Intern Med. 1991;151:1381-1384)