Use of magnetic resonance imaging to detect neoplastic meningitis: limited use in leukemia and lymphoma but convincing results in solid tumors

Background: An early diagnosis of meningitis is important to improve patients' survival. Data about a direct comparison of cerebrospinal fluid cytology (CSF-cytology) and MRI are very limited. Therefore, the aim of this study was to compare these two diagnostic modalities in diagnosing meningitis in patients with hematopoietic and solid malignancies.

Methods: In 68 patients suspicious for neoplastic meningitis, cytology and MRI (1.5 T) was performed. The meningeal, pial or intraparenchymal hyperintense signal or contrast enhancement was correlated to the final CNS diagnosis and to cytology.

Results: 44 patients (64.7%) had neoplastic meningitis, 21 patients (30.9%) had non-neoplastic meningitis. The sensitivity to diagnose meningeal disease was 49.2% for MRI and 95.4% for cytology (p<0.001). In patients with neoplastic meningitis, sensitivity was 45.5% for MRI and 93.2% for cytology (p<0.001). In patients with infectious meningitis, sensitivity was 57.1% for MRI and 100% for cytology (p=0.0013). In patients with solid tumors, the sensitivity was 84.6% for both diagnostic methods. The sensitivity for MRI was low in patients with leukemia (20.0%) and lymphoma (37.5%). The positive predictive value (PPV) for MRI to differentiate infectious from neoplastic meningitis was high in patients with infectious meningitis (75.0%), in patients with lymphoma (83.3%), and in patients with solid tumors (72.7%). Ppv was low in patients with leukemia (33.3%).

Conclusion: Diagnostic value of MRI for diagnosing meningitis is especially limited in patients with hematopoietic malignancies. MRI better detected leptomeningeal involvement caused by solid tumors than by leukemia or lymphoma. The ppv to specify neoplastic meningitis depends on tumor subtype.