Article Text

Cerebrospinal fluid HIV viral load may be detectable, despite serum viral load being undetectable, in patients diagnosed with syphilis
  1. Eavan Gabrielle Muldoon1,
  2. Suzie Coughlan2,
  3. Fiona Mulcahy1
  1. 1Department of Genitourinary Medicine and Infectious Diseases, St James's Hospital, Dublin, Ireland
  2. 2UCD National Virus Reference Laboratory, University College Dublin, Belfield, Dublin 4
  1. Correspondence to Dr Eavan Gabrielle Muldoon, Department of Genitourinary Medicine and Infectious Diseases, St James's Hospital, James's Street, Dublin 8, Ireland; eavan{at}

Statistics from

With an increasing proportion of male patients co-infected with syphilis and HIV, the management of these patients has become more relevant. HIV and syphilis both enter the central nervous system early in the course of infection, and have been shown to be present at all stages of the disease. Syphilis is reported to increase the serum HIV viral load (VL) in HIV co-infected patients;1 higher CSF HIV VLs have been demonstrated in patients with neurosyphilis compared with those without.2

HIV-positive patients newly diagnosed with syphilis (defined by seroreactive syphilis serology without prior history or a fourfold rise in rapid plasma reagin) undergoing LP, were prospectively recruited, and their informed consent obtained. CSF HIV VL was performed on CSF samples. The study was approved by the hospital Ethics Committee.

Thirty-seven patients underwent lumbar puncture; mean age 35 years (22–56 years). 9/37 (21.6%) were diagnosed with neurosyphilis. HIV VL was performed on 27/37 (72.9%) samples; the remaining samples had insufficient CSF remaining. 25/27 (92.6%) had detectable CSF HIV VL. Mean detectable CSF HIV VL 10 416 copies/ml (5–193 000 copies/ml). Mean CSF HIV VL in patients with neurosyphilis was 41 125 copies/ml (95% CI 64 451 to 146 702 copies/ml) significantly higher than those without neurosyphilis 2494 (95% CI 541 to 4447 copies/ml) p=0.03. Serum HIV VL was available for 32/37 (86.5%) patients. Patients with detectable serum HIV VL were more likely to have detectable CSF VL (p=0.04). Six patients had serum HIV VL <50 copies/ml and detectable HIV CSF VL; three patients had HIV VL CSF values <50 copies/ml, while three patients CSF HIV VL values >50 copies/ml, with none of these three patients being diagnosed with neurosyphilis.

The observation that three patients with undetectable serum HIV VL had detectable CSF HIV VL is interesting. Given the data demonstrating higher CSF HIV VL in patients with neurosyphilis,2 it could be that syphilis infection up-regulates the viral replication of HIV within the CSF. This may have important implications for HIV evolution within CSF and serum different strains have been demonstrated in both compartments in the same individual.3 As it has been reported that CSF HIV VL has neurological prognostic implications,4 this up-regulation may have long-term consequences for HIV patients diagnosed with syphilis.


View Abstract


  • Funding Funding provided by research bursary awarded by the Infectious Diseases Society of Ireland.

  • Competing interests None.

  • Ethics approval St James's Hospital Research Ethics Committee 2008/05/13.

  • Provenance and peer review Not commissioned; internally peer reviewed.

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.