Cerebral tuberculosis may resemble toxoplasmosis on magnetic resonance imaging (MRI), even in patients without pulmonary symptoms or cerebrospinal fluid abnormalities.
Researchers described a 30-year-old female patient born in Madagascar who had immigrated to Germany 14 months prior to presentation. Routine MRI to monitor an asymptomatic pituitary adenoma diagnosed 2 years earlier, revealed multiple rim-enhancing lesions involving both cerebral hemispheres and supratentorial and infratentorial regions. The findings suggested cerebral toxoplasmosis or neurocysticercosis. The lesions were assessed using neurologic examination, laboratory investigations, cerebrospinal fluid (CSF) analysis, chest computed tomography (CT), and microbiological testing.
At presentation, the patient had no fever, pulmonary symptoms, or neurologic abnormalities. Laboratory investigations revealed an erythrocyte sedimentation rate of 84 mm/h and a C-reactive protein concentration of 60 mg/L, despite a normal white blood cell count. The CD4-positive T-lymphocyte count was slightly reduced. CSF analysis showed a normal cell count and no other notable abnormalities, while testing for human immunodeficiency virus and parasitic infections was negative.
Chest CT revealed multiple bilateral pulmonary nodules measuring 1 to 3 mm, consistent with miliary tuberculosis. Although sputum cultures remained negative, polymerase chain reaction (PCR) testing detected DNA from the Mycobacterium tuberculosis complex in a sputum sample. CSF PCR testing for tuberculosis was negative.
Based on the findings, the researchers revised the diagnosis to probable parenchymal central nervous system (CNS) tuberculosis associated with miliary pulmonary tuberculosis. During the second week of hospitalization, the patient developed headaches, coughing, and word-finding difficulties.
Treatment was initiated with rifampicin, isoniazid, pyrazinamide, and ethambutol, alongside adjunctive corticosteroids for CNS involvement. Prednisolone was administered at 60 mg daily for 4 weeks, followed by a gradual taper over the subsequent 8 weeks. Severe gastrointestinal intolerancerequired replacement of pyrazinamide and ethambutol with moxifloxacin and linezolid.
After 12 months of antituberculosis therapy, serial MRI demonstrated progressive regression of the cerebral lesions and complete clinical and radiologic remission was documented 12 months posttreatment. MRI findings were normal at follow-up, and no paradoxical reactions occurred during treatment or follow-up.
Cerebral tuberculosis, toxoplasmosis, and neurocysticercosis can share several imaging characteristics, including multifocal contrast-enhancing lesions and surrounding edema. Cerebral tuberculomas may exhibit ring or nodular enhancement, while the radiologic appearance of neurocysticercosis varies according to the stage of infection. Overlapping characteristics can make differentiation difficult using conventional MRI alone.
The absence of CSF abnormalities presented an additional diagnostic challenge. Unlike tuberculous meningitis, isolated parenchymal tuberculosis may not produce inflammatory changes in CSF because the infection remains confined to brain tissue. Normal CSF findings do not exclude cerebral tuberculosis when other clinical and imaging findings are compatible.
The researchershighlighted the importance of epidemiologic background. Although the patient originated from Madagascar, where tuberculosis incidence is high, she had lived in Germany for over 1 year without recent international travel. Infectious diseases more prevalent in Germany were initially considered more likely. The researchers acknowledged that earlier consideration of tuberculosis and targeted microbiological testing might have facilitated diagnosis. An interferon-gamma release assay could also have increased clinical suspicion, although it cannot distinguish active from latent tuberculosis or confirm CNS involvement.
The report had several limitations. It involved 1 patient, preventing conclusions about the frequency of diagnostic misclassification or the comparative effectiveness of treatment approaches. Cerebral tuberculosis was classified as probable because direct microbiological confirmation from CSF or brain tissue was unavailable. The diagnosis instead relied on compatible neuroimaging, pulmonary microbiological evidence, exclusion of alternative infections, and regression of cerebral lesions during treatment.
"Our intention is not to draw epidemiologic conclusions from a single case but to raise awareness that cerebral [tuberculosis] may occur in the absence of meningeal symptoms or CSF abnormalities and closely mimic more common CNS infections such as cerebral toxoplasmosis," wrote lead study author Daniel Martens, of the Department of Neurology at Saarland University in Germany, and colleagues.
The study authors declared no conflicts of interest.
