University of California, San Francisco Logo

University of California, San Francisco | About UCSF | Search UCSF | UCSF Medical Center

Management Recommendations > Neurosyphilis

Ward 86 Management Recommendations

Management of Neurosyphilis

updated August 2018

Contributors: Mark Jacobson, MD
Sulggi Lee, MD, PhD


The term "neurosyphilis" has become confusing because it is commonly used to refer to two distinct phenomenon: Treponema pallidum infection of the central nervous system (ie, abnormal cerebrospinal fluid [CSF] results) and neurologic disease caused by T pallidum (ie, central nervous system [CNS], ocular, or auditory clinical manifestations of syphilis). This distinction is of paramount clinical importance because the vast majority of people who have CSF abnormalities during early syphilis (ie, the first year after an untreated infection) do NOT require a prolonged course of intravenous (IV) penicillin; a single dose of intramuscular (IM) benzathine penicillin suffices to cure the infection and prevent any CNS complications (see Note below). Thus, our approach to treating "neurosyphilis" depends on the duration of untreated infection and the presence of specific neurologic symptoms and objective findings attributable to T pallidum CNS infection as discussed below.

Diagnosis of Neurosyphilis

The diagnosis of syphilis includes serologic blood testing and clinical examination (see section on Diagnosis and Treatment of Syphilis). Patients with suspected symptomatic neurosyphilis or objective findings consistent with CNS neurologic disease attributable to T pallidum CNS infection should undergo cerebrospinal fluid (CSF) analysis for cell count and differential, protein, and Venereal Disease Research Laboratory (VDRL) testing. Of note, mild to moderate headache is a common symptom of secondary syphilis and, in itself, does not justify CSF analysis in a patient who has clinical findings of secondary syphilis (eg, a typical rash) or an RPR titer ≥1:32 (which is unusual outside the setting of secondary syphilis). There also are specific situations in which CSF analysis is unlikely to change management (see second bullet item under Treatment below).

Evidence of CNS invasion (eg, elevated white blood cell [WBC] count or protein in CSF or a positive CSF-VDRL result) can be detected in up to half of patients with early syphilis. Yet, these abnormal findings appear to have little clinical relevance in the absence of symptoms or objective findings of neurologic disease.(1) Indeed, the CSF-VDRL test has limited sensitivity and specificity. Among patients with confirmed, symptomatic, tertiary neurosyphilis, approximately 25% have a falsely negative CSF-VDRL, whereas, among patients with early syphilis and no clinical evidence of neurologic disease, up to 16% have been reported to have a positive CSF-VDRL result.(1) For cases in which there is a high suspicion of tertiary neurosyphilis, additional tests that can address the poor sensitivity of the CSF-VDRL may be considered. These nonroutine tests include the CSF treponemal antibody absorption test (FTA-ABS), an assay that is more sensitive than CSF-VDRL but lacks specificity, and a titratable CSF T pallidum hemaglutination assay (CSF-TPPA), which was recently shown to have better diagnostic specificity if a titer of ≥1:640 was used as the cutoff value for positivity.(2) A further confounding factor in interpreting CSF results among HIV-infected patients is that the usual cutoff for abnormal CSF pleocytosis (>5 WBC/µL) may not apply. HIV-infected patients who have an absolute CD4+ T-cell count of >200 cells/µL AND are not virologically suppressed on antiretroviral therapy typically have up to 20 WBC/µL in CSF (presumed to be an inflammatory response to uncontrolled HIV replication in the CNS).(3,4)

Treatment of Neurosyphilis

  • CSF abnormalities during early syphilis, without neurologic symptoms or signs of neurologic disease attributable to this infection, occur commonly in early syphilis and only extremely rarely lead to serious neurologic disease after treatment with a single dose of IM benzathine penicillin.(1) Thus, the finding of CSF abnormalities in a patient who has active early syphilis but no neurologic symptoms or findings attributable to CNS disease caused by T pallidum is not an indication for prolonged, high-dose IV penicillin therapy. As noted above, mild to moderate headache is a common symptom of secondary syphilis and, in itself, does not justify CSF analysis in a patient who has clinical findings of secondary syphilis (eg, a typical rash) or an RPR titer ≥1:32 (which is unusual outside the setting of secondary syphilis).

  • Symptomatic neurologic disease occurring during serologically documented secondary syphilis does require a 10- to 14-day course of IV penicillin G (24 million units daily in divided doses), which requires hospitalization in most cases. The most common manifestations of neurologic disease associated with secondary syphilis involve cranial nerves II and VIII and rarely, cranial nerve VII. Uveitis, optic neuritis, and retinitis are the most common ocular complications, and abnormalities should be documented by slit-lamp exam and dilated, indirect ophthalmoscopy. Cranial nerve VIII disease most commonly manifests as new hearing loss, tinnitus, or vertigo, and audiometry should be done to document and monitor hearing loss. The recommendation for IV penicillin is based on expert opinion. There have been no randomized trials addressing treatment for these vision- and auditory-threatening complications of secondary syphilis.

    Patients with obvious ocular or auditory manifestations of syphilis related to secondary syphilis (eg, because the classic rash of secondary syphilis is present or a high rapid plasma reagin [RPR] titer is documented) do not necessarily need CSF analysis in order to confirm the diagnosis of neurosyphilis. For example, a patient with a high serum RPR titer and optic neuritis should be considered to have immediately sight-threatening neurosyphilis and should receive a full course of IV penicillin even if the CSF results are completely normal. For many cases of ocular and cranial nerve VIII syphilis, the results of a lumbar puncture will not change management decision making unless other CNS diseases need to be excluded (eg, in patients who have a low CD4+ cell count or those who do not improve rapidly with treatment of syphilis). On the other hand, having a baseline CSF analysis may be useful for comparison if a patient's symptoms do not resolve and a repeat lumbar puncture will need to be performed.

    Of note, outbreaks of ocular syphilis, in particular uveitis (including posterior uveitis), were reported by the Centers for Disease Control and Prevention in 2014 among HIV-infected men in Seattle and San Francisco. Subsequent epidemiologic survelliance revealed an increase in ocular syphilis cases among men who have sex with men (MSM) in many areas of the United States.(5) Clinicians should be particularly aware of this complication in diagnosing and treating cases of secondary syphilis, as such patients could have a falsely negative RPR titer due to the prozone phenomenon (see section on Diagnosis and Treatment of Syphilis).

    Rarely, patients with secondary syphilis have symptoms and signs of meningitis (eg, severe headache, neck stiffness) that lead to hospitalization and treatment with IV penicillin. It is unknown whether IV treatment should be continued for the full 10- to 14-day course in patients who improve earlier.

  • Asymptomatic T pallidum CNS infection in late syphilis (ie, untreated infection >1 year) is quite rare but is believed to require IV penicillin treatment in order to prevent catastrophic complications such as stroke. For the patient with late, latent syphilis without clinical evidence of neurosyphilis (eg, neurologic, visual, or auditory symptoms; signs of cognitive difficulties; myelopathy; peripheral neuropathy), we consider performing CSF analysis to evaluate the possibility of asymptomatic but active late neurosyphilis if: 1) the serum RPR titer is >1:16 and failed to drop 4-fold ≥6 months after receiving 3 weekly IM injections of benzathine penicillin; AND 2) the patient's sexual history suggests no interim reinfection.

    Conversely, for the patient with asymptomatic, late latent syphilis and a baseline RPR titer ≤1:16 that failed to drop 4-fold, we have a low suspicion of neurosyphilis and do not recommend CSF analysis. Our rationale is that the likely explanation is that the patient is serofast (see section on Diagnosis and Treatment of Syphilis) rather than failing treatment owing to an occult focus of CNS infection. Such patients should be monitored by repeat RPR and neurologic evaluation.

    Given that tertiary neurosyphilis has become a very rare diagnosis in recent years and most cases of ocular and cranial nerve secondary syphilis can be diagnosed and treated appropriately without CSF analysis, the case of an asymptomatic patient with a persistent RPR titer >1:16 despite 3 weekly IM injections of benzathine penicillin (and in whom re-infection can be confidently excluded) is precisely the most common clinical situation that does merit lumbar puncture to rule out neurosyphilis. However, clinicians should be aware, as noted above, that untreated HIV in patients whose absolute CD4 T-cell count remains >200 cells/µL may account for a mild CSF pleocytosis (presumed to be due to an inflammatory response to CNS HIV infection).

  • Symptomatic neurologic disease in late syphilis (ie, tertiary neurosyphilis) requires IV penicillin treatment to prevent further progression or catastrophic sequela. This recommendation is based on expert opinion. There have been no randomized trials addressing treatment for tertiary neurosyphilis. Meningovascular syphilis is the most common form of tertiary syphilis. This condition involves intracerebral arteritis and typically presents with severe headache or stroke. Diagnosis requires neurological examination and imaging of the brain (preferably magnetic resonance imaging [MRI]) and intracranial vessels, as well as CSF analysis. Management requires consultation with a neurologist.

  • Patients with early latent syphilis who are incidentally found to have CSF abnormalities, including a positive CSF VDRL result, also should be treated with a single dose of benzathine penicillin but should undergo follow-up lumbar puncture: a) immediately if any ocular or neurologic symptoms develop, or b) 6 months after treatment of early latent syphilis to assure that these CSF abnormalities have resolved and are not due to another pathologic process.

Treatment of the penicillin-allergic patient

Patients with neurosyphilis who require high-dose IV penicillin but are allergic ideally should undergo penicillin desensitization. If desensitization is not feasible, ceftriaxone 2 g IV once daily for 10 days is a reasonable alternative. This ceftriaxone regimen was compared with IV penicillin 24 million units/day in divided doses in a small, open-label, randomized trial conducted in HIV-infected patients with neurosyphilis, defined as a serum RPR >1:16 plus CSF abnormalities (positive VDRL result, CSF white blood cell count of >20 cells/µL or CSF protein >50 mg/dL).(6) There were no differences in clinical or CSF laboratory outcomes. Four-fold declines in serum RPR actually occurred more frequently in the ceftriaxone group than in the standard IV penicillin group. Limitations of the study include the fact that the trial involved only 30 evaluable subjects, and only 10 in each arm were symptomatic. Also, the possibility of reinfection was not excluded, which could have biased these results.

Treatment Monitoring

It is not necessary to repeat CSF VDRL titers for patients who receive IV penicillin therapy if the serum RPR decreases appropriately after treatment (ie, a 4-fold decrease in serum titer or reversion of the test result to nonreactive within 6 months of treatment).(7) A repeat CSF analysis is not required, but may be useful in patients whose symptoms do not resolve and have a comparison baseline CSF analysis performed prior to treatment.

Note: The issue of whether single-dose IM benzathine penicillin is adequate therapy for early syphilis was examined in a randomized trial conducted in the 1990s that enrolled 541 subjects with newly diagnosed early syphilis, of whom 101 were HIV infected and 52% had secondary syphilis.(1) Subjects were randomized to receive a standard single dose of IM benzathine penicillin or a 10-day course of high-dosage amoxicillin (2,000 mg 3 times daily) plus probenecid (which inhibits renal excretion of amoxicillin). This latter regimen achieves treponemicidal antibiotic levels in the CSF approximating those achieved by 24 million units per day of IV penicillin administered in divided doses. Clinical and serologic success rates were identical in the two treatment arms. In addition, a subset of patients underwent lumbar puncture, and half of those had some CSF abnormality attributable to syphilis, including a positive CSF VDRL test result in 16% of HIV-infected patients and 7% of HIV-uninfected patients. Twenty-five percent of both HIV-infected and HIV-uninfected patients had a positive CSF PCR test result for T pallidum. Serologically defined treatment failure was no more likely to occur among patients who had T pallidum detected in pretreatment samples of CSF than those who did not, regardless of treatment regimen. The authors concluded that asymptomatic neurosyphilis during the early stages of disease was common, though not more common among HIV-infected patients than those without HIV infection, that it did not predict failure of standard single-dose IM treatment for early syphilis, and that most testing of CSF before treatment of early syphilis is not useful in clinical decision-making. These data are supported by a recent observational study evaluating the prevalence of asymptomatic neurosyphilis among HIV-infected MSM a median 8 months after treatment with a single dose of IM benzathine penicillin or 14-21 days of oral doxycycline for early syphilis.(8) Of 59 patients who consented to lumbar puncture and had full CSF results, only 1 patient had CSF evidence consistent with asymptomatic late neurosyphilis.


  1. Rolfs RT, Joesoef MR, Hendershot EF, et al. A randomized trial of enhanced therapy for early syphilis in patients with and without human immunodeficiency virus infection. The Syphilis and HIV Study Group. N Engl J Med. 1997 Jul 31;337(5):307-14.
  2. Marra CM, Maxwell CL, Dunaway SB, et al. Cerebrospinal fluid treponema pallidum particle agglutination assay for neurosyphilis diagnosis. J Clin Microbiol. 2017 Jun;55(6):1865-70.
  3. Marshall DW, Brey RL, Cahill WT, et al. Spectrum of cerebrospinal fluid findings in various stages of human immunodeficiency virus infection. Arch Neurol. 1988 Sep;45(9):954-8.
  4. Marra CM, Maxwell CL, Collier AC, et al. Interpreting cerebrospinal fluid pleocytosis in HIV in the era of potent antiretroviral therapy. BMC Infect Dis. 2007 May 2;7:37.
  5. Oliver SE, Aubin M, Atwell L, et al. Ocular syphilis -- eight jurisdictions, United States, 2014-2015. MMWR Morb Mortal Wkly Rep. 2016 Nov 4;65(43):1185-8.
  6. Marra CM, Boutin P, McArthur JC, et al. A pilot study evaluating ceftriaxone and penicillin G as treatment agents for neurosyphilis in human immunodeficiency virus-infected individuals. Clin Infect Dis. 2000 Mar;30(3):540-4.
  7. Marra CM, Maxwell CL, Tantalo LC, et al. Normalization of serum rapid plasma reagin titer predicts normalization of cerebrospinal fluid and clinical abnormalities after treatment of neurosyphilis. Clin Infect Dis. 2008 Oct 1;47(7):893-9.
  8. Tomkins A, Ahmad S, Cousins DE, et al. Screening for asymptomatic neurosyphilis in HIV patients after treatment of early syphilis: an observational study. Sex Transm Infect. 2018 Aug;94(5):337-339.