Open Access Open Badges Research article

Factors influencing cerebrospinal fluid and plasma HIV-1 RNA detection rate in patients with and without opportunistic neurological disease during the HAART era

Paulo P Christo13*, Dirceu B Greco2, Agdemir W Aleixo2 and Jose A Livramento3

Author Affiliations

1 Eduardo de Menezes Hospital, FHEMIG, Belo Horizonte, MG, Brasil

2 Department of Infectious Diseases, Federal University of Minas Gerais, Brasil

3 Department of Neurology, University of São Paulo, Brasil

For all author emails, please log on.

BMC Infectious Diseases 2007, 7:147  doi:10.1186/1471-2334-7-147

Published: 21 December 2007



In the central nervous system, HIV replication can occur relatively independent of systemic infection, and intrathecal replication of HIV-1 has been observed in patients with HIV-related and opportunistic neurological diseases. The clinical usefulness of HIV-1 RNA detection in the cerebrospinal fluid (CSF) of patients with opportunistic neurological diseases, or the effect of opportunistic diseases on CSF HIV levels in patients under HAART has not been well defined. We quantified CSF and plasma viral load in HIV-infected patients with and without different active opportunistic neurological diseases, determined the characteristics that led to a higher detection rate of HIV RNA in CSF, and compared these two compartments.


A prospective study was conducted on 90 HIV-infected patients submitted to lumbar puncture as part of a work-up for suspected neurological disease. Seventy-one patients had active neurological diseases while the remaining 19 did not.


HIV-1 RNA was quantified in 90 CSF and 70 plasma samples. The HIV-1 RNA detection rate in CSF was higher in patients with neurological diseases, in those with a CD4 count lower than 200 cells/mm3, and in those not receiving antiretroviral therapy, as well as in patients with detectable plasma HIV-1 RNA. Median viral load was lower in CSF than in plasma in the total population, in patients without neurological diseases, and in patients with toxoplasmic encephalitis, while no significant difference between the two compartments was observed for patients with cryptococcal meningitis and HIV-associated dementia. CSF viral load was lower in patients with cryptococcal meningitis and neurotoxoplasmosis under HAART than in those not receiving HAART.


Detection of HIV-1 RNA in CSF was more frequent in patients with neurological disease, a CD4 count lower than 200 cells/mm3 and detectable plasma HIV-1. Median HIV-1 RNA levels were generally lower in CSF than in plasma but some patients showed higher CSF levels, and no difference between these two compartments was observed in patients with cryptococcal meningitis and HIV-associated dementia, suggesting the presence of intrathecal viral replication in these patients. HAART played a role in the control of CSF HIV levels even in patients with cryptococcal meningitis and neurotoxoplasmosis in whom viral replication is potentially higher.