Tuberculous meningitis and cryptococcal meningitis co-infection in hospitalised patients in Durban, Kwa Zulu Natal (KZN), South Africa

Background: Cryptococcal meningitis (CCM) and tuberculous meningitis (TBM) co-infection has been previously described, although the diagnosis is rare. This may be due to the paucibacillary nature of TBM and the difficulty in diagnosing these conditions due to the overlap of symptoms and cerebrospinal fluid (CSF) findings. This study is aimed to determine the frequency and outcome of TBM and CCM co-infection in hospitalised patients.
Methods: A retrospective review of routine laboratory and clinical records of patients with TBM and CCM co-infection at four regional hospitals in a single metropolitan district in Durban, South Africa was conducted between 1^st January 2005 and 31^st December 2009.
CSF Tuberculosis(TB) data at Inkosi Albert Luthuli Central Hospital’s TB culture laboratory was the starting point for identifying CSF TB culture-positive cases. CSF microbiology laboratory data at each identified hospital study site were then reviewed for this retrospective analysis.
All adult patients whose CSF yielded positive TB cultures were included in this study. These positive Mycobacterium tuberculosis (MTB) samples were then matched with CSF samples positive for cryptococcosis, confirmed by cryptococcal latex agglutination test (CLAT), India ink test or culture. A chart review of dual-infected cases was then conducted.
Results: 418 patients were identified based on the gold standard of the time, namely positive CSF TB culture extracted from laboratory data at each study site. Nucleic acid amplification testing was not available during the study period. A total of 15 patients had dual infection, with 10 patients confirmed to be human immunodeficiency virus (HIV) infected. The HIV statuses of the other 5 patients were unknown but they were clinically immunocompromised. The prevalence of dual infection was 15/418 cases (3.6%). Twelve of the 15 patients were diagnosed with TBM posthumously with an in-hospital mortality of 80% (12/15). Despite the age of the data set, the study remains valid and relevant due to the rarity of TBM and CCM co-infection.
Conclusion: In this study, we found that dual infection of the meninges with Mycobacterium tuberculosis and Cryptococcus neoformans in HIV-infected patients is rare. However, the diagnosis may often be missed or not considered initially. Underdiagnosis and possible delay in diagnosis of dual infection may result in patients not being treated appropriately and adequately, leading to increased morbidity and mortality.