Pathology of cryptococcal meningoencephalitis: analysis of 27 patients with pathogenetic implications.

In this autopsy series of cryptococcal meningoencephalitis (CME), the authors analyzed neuropathologic lesions in 13 human immunodeficiency virus (HIV) and 14 non-HIV-related cases. Most non-HIV patients did not have immunosuppressive predisposing illness. Analysis of pathological findings revealed significant differences in the inflammatory response to CME in patients with and without HIV infection. None of the acquired immunodeficiency syndrome (AIDS) patients had granulomatous inflammation, whereas most non-HIV-associated cases had granulomas, supporting a role for cell-mediated immunity in CME. Lymphocytic infiltrate in both groups consisted of T cells (CD45RO+). In some non-HIV-associated cases, CME was undiagnosed and untreated. In most HIV-associated cases, CME had an encephalitic component, resulting in grossly or microscopically visible accumulations of fungi within the brain parenchyma, whereas in non-HIV-associated cases, CME was often confined to the subarachnoid space and large perivascular spaces (Virchow-Robin spaces). In non-HIV-associated cases, yeast forms were fewer and showed a more limited distribution. In contrast, many extracellular fungi were present in many cases of HIV-associated CME. The principal reactive cell in CME in AIDS was brain macrophages and microglia, especially those in the perivascular and juxtavascular locations. Reactive astrocytes were limited to large destructive lesions and subpial regions. In several patients with HIV-associated CME, large parenchymal cryptococcomas contained Crytococcus neoformans (CN) with cell wall pigmentation, suggestive of melanin. The authors suggest that in AIDS patients altered immune functions allow CN to accumulate within the brain, predominantly extracellularly, and that deficient macrophage/microglial effector function may be responsible for the altered pathology. In addition, coexisting CNS processes in HIV-associated CME may contribute to the altered pathology. The authors conclude that cryptococcal meningitis is not a disease limited to the cerebrospinal fluid (CSF) space but affects the brain more significantly than suspected. Therapeutic strategies that enhance the effector function of glial cells at the CNS-CSF barrier may be useful for improving the response to therapy.