IN RECENT YEARS, RATES of syphilis have increased among men who have sex with men (MSM) in cities in the United States, Canada, and Europe,1 especially in human immunodeficiency infection (HIV)-infected patients.2 Similar data have been reported in Brazil.3,4 Syphilis also occurs throughout of the world in heterosexuals who engage in high-risk sexual practices.
Several studies have postulated that Treponema pallidum invades the central nervous system (CNS) early in the course of infection, and HIV-infected patients with syphilis may be more likely to fail treatment to early syphilis and develop neurosyphilis.5 Neurosyphilis presents important diagnostic and therapeutic challenges in HIV-infected patients. Most studies about this coinfection have been performed in developed countries, while few report data from limited-resource settings. Here, we describe the clinical and laboratory features of HIV-infected patients with neurosyphilis in a tertiary referral center from Sao Paulo, Brazil.
Materials and Methods
We retrospectively analyzed data from the clinical records of all HIV-infected patients with confirmed diagnosis of a first episode of neurosyphilis admitted between March 1999 and March 2006 to the Emílio Ribas Institute of Infectious Diseases, a 250-bed tertiary teaching center that serve persons of low socioeconomic level from the Sao Paulo State, Brazil. HIV infection was diagnosed by ELISA and confirmed by Western blotting. The diagnosis of neurosyphilis was based on the presence of a reactive cerebrospinal fluid (CSF) Venereal Disease Research Laboratory (VDRL) test result. In this study, none standardized criteria based on serum VDRL titers or CD4+ cell counts were considered to perform lumbar puncture. Each physician individually decided the indication of lumbar puncture for each patient, including in those neurologically asymptomatic. The syphilis stage was determined according to Center for Disease Control and Prevention (CDC) guidelines.6 In addition, neurosyphilis was categorized as asymptomatic and symptomatic, according to absence or presence of neurologic symptoms or signs recorded in the medical files. Patients with syphilis of unknown duration were included in the group of late latent syphilis. Neurosyphilis was categorized as early and late. Early neurosyphilis include asymptomatic or symptomatic meningitis. Late neurosyphilis include general paresis and tabes dorsalis.
Associations between categorical variables were assessed using Fisher exact test. Associations between continuous variables and categorical variables were assessed using the Mann–Whitney U test. The values of P <0.05 were considered statistically significant. We used the S-Plus 7.0 (Insightful, WA) statistical package for all calculations.
Baseline characteristics of the 27 patients are shown in Table 1. Most of the patients were men, were not previously diagnosed with AIDS, and had late latent syphilis. At the time of neurosyphilis diagnosis, 10 (37%) patients had early syphilis, and 6 of them were neurologically asymptomatic. Ten (37%) patients have been previously treated for nonneurologic syphilis. However, we had not details about these previous episodes. Symptoms and signs included headache (78%), nausea or vomiting (44%), neck stiffness (44%), dizziness (33%), mental confusion (22%), sensory ataxia (11%), facial palsy (11%), and hearing loss (11%). Two patients presented with spinal cord involvement, one with meningomyelitis and the other with tabes dorsalis, representing the only patient with late neurosyphilis. The details of the case with meningomyelitis were previously reported.7
The diagnostic laboratory characteristics of the 27 patients according to the presence of early or late syphilis are shown in Table 2. The median serum VDRL was 1:128. Twenty five (93%) and 23 (85%) patients showed serum VDRL titers ≥1:16 and ≥1:32, respectively. The results of CSF VDRL were qualitative. CSF Treponema pallidum hemagglutination antibody absorption (TPHA) was performed in 23 patients, and all tests were reactive. Serum FTA-ABS was performed in 20 patients, and all tests but one were reactive. However, in the nonreactive case, the CSF TPHA was reactive.
The median CD4+ T cell count was 182 cells/μL. CD4+ T cell counts were <100 cells/μL in 8 (30%) patients; 100 to 200 cells/μL in 8 (30%) patients; and >200 cells/μL in 11 (40%) patients. Five (19%) patients presented with CD4+ T cell counts >350 cells/μL. The median CSF nucleated cell count was 7 cells/μL: 3 cells/μL in asymptomatic patients and 25 cells/μL in symptomatic patients (P <0.2). Thirteen (48%) patients had ≤5 cells/μL in CSF, 10 with asymptomatic neurosyphilis and 3 with symptomatic neurosyphilis. All taps were nontraumatic yet all had CSF TPHA reactive. Comparison of symptomatic and asymptomatic neurosyphilis reveals that CD4+ T cell counts were significantly lower in symptomatic patients (P = 0.03) (Table 3), and counts ≤200 cells/μL were more likely to experience symptomatic neurosyphilis (P = 0.04).
Symptomatic patients had higher protein levels in CSF (P <0.001), and levels of >40 mg/dL were more common in symptomatic patients (P <0.001). Syphilis stage was not associated with symptomatic neurosyphilis (P = 0.4). Interestingly, 6 of 10 patients with early syphilis were neurologically asymptomatic, and 4 of them presented secondary syphilis, manifested by rash.
The most commonly used treatment was intravenous aqueous crystalline penicillin G, 24 million units, given in 6 daily doses. This was used in 15 (56%) cases either 14 days (7 patients) or for 21 days (8 patients). Twelve (44%) patients were treated with ceftriaxone, 2 g once daily for 14 days. No fatal cases were seen during hospitalization. Neurologic sequelae including mental slowing in 2 cases, facial palsy and hearing loss in 1 case, and sensory ataxia in the other were reported.
To the best of our knowledge, this study reports the largest series of neurosyphilis in HIV-infected patients from developing countries. In this population, we described the main clinical and laboratory features, especially the high frequency of cases with serum VDRL ≥1:16.
Neurosyphilis is an important, treatable complication of HIV-infected patients that is prevalent in resource-limited settings, including Brazil.9,10 A recent cross-sectional study reported that neurosyphilis represents 1% of all neurologic diseases in HIV-1 infected patients in Belo Horizonte, Brazil.10 However, given the challenge of diagnosis that neurosyphlis represents, and that this was an in-hospital study, the frequency of neurosyphilis could still be underestimated.
Approximately 90% of patients of our series were men; we believe representing the high association between neurosyphilis and HIV in MSM. One third of the cases included in this study had previous syphilis treatment. Several reports suggested that HIV infection increases the rate of treatment failure for primary syphilis, and consequently, these patients may be more likely to develop neurosyphilis,11,12 probably secondary to neurologic relapse.13 However, reinfection after initial treatment may also have occurred. Approximately two thirds of our patients have not started highly active antiretroviral therapy (HAART) when we diagnosed neurosyphilis. Because HAART has reduced many neurologic complications in HIV, earlier initiation of HAART may also reduce the problem of neurosyphlis.14
In the preantibiotic era, parenchymal or late neurosyphilis (tabes dorsalis or general paresis), were more common than early neurosyphilis (asymptomatic, meningitis, and meningovascular syphilis).16 Currently, late neurosyphilis is rare and early forms are more frequent in HIV-infected patients, as shown.12,13,17–19 HIV-infected patients with access to medical care have more opportunities to be diagnosed before presentation with parenchymal neurosyphilis, and also receive antibiotics that may impact syphilis. In the current series, only one patient had late neurosyphilis, diagnosed as tabes dorsalis. An additional patient had spinal involvement, diagnosed as meningomyelitis, an infrequent form that is secondary to syphilitic meningitis.20
Deciding which HIV-infected patients with syphilis should undergo lumbar puncture remains a controversial area. The CDC and some experts recommend that CSF examination must be performed for HIV-infected patients with late latent syphilis, syphilis of unknown duration, neurologic signs or symptoms, or treatment failure.6,21 However, a recent study showed that 40 of 49 (82%) HIV-positive MSM with symptomatic early neurosyphilis were reported as having early syphilis. In addition, the estimated risk for having symptomatic early neurosyphilis in HIV-positive MSM with early syphilis was very low: 1.7% (40 of 2380).22 If we apply the recommendations of CDC for CSF examination in HIV-infected patients with syphilis, 6 (22%) of our patients should not undergo a lumbar puncture and consequently the neurosyphilis diagnosis would have been delayed and underestimated. Considering that all these 6 patients presented early syphilis and were asymptomatic, our results suggest the benefit of performing lumbar puncture in neurologically asymptomatic HIV-infected patients with early syphilis. Nevertheless, the design of this study does not permit us to directly address this controversial question.
Potential surrogate markers are necessary to identify the group of patients at the greatest risk of neurosyphilis. Recently, a study reported that neurosyphilis was 6 times more likely to occur in patients with serum rapid plasma reagin (RPR) titer ≥1:32, and 3 times more likely to occur in patients with CD4+ cell count ≤350 cells/μL.23 In the same line, another study concluded that lumbar puncture could be restricted to HIV-infected patients with syphilis who present neurologic manifestations or a serum RPR titer ≥1:32.24 Our results are especially useful in settings when VDRL is the unique available nontreponemal test. Quantitative results of serum VDRL cannot be compared directly with RPR and considering that RPR titers are frequently slightly higher than VDRL titers,6 we define arbitrarily a serum VDRL titer ≥1:16 as similar to serum RPR titer ≥1:32. More than 90% of our patients had VDRL ≥1:16. Furthermore, if we consider these titers to perform lumbar puncture we could correctly identify neurosyphilis in 5 of 6 patients without indication to lumbar puncture using the current CDC guidelines. This finding agrees with the study of Marra et al.,23 which showed the importance of serum RPR titers in the indication of lumbar puncture in HIV-infected patients with syphilis. However, the clinical value of these markers of neurosyphilis was recently questioned because of the absence of longitudinal data evaluating their correlation with clinical outcomes.21 In the current study, the lack of information about the total number of patients with serum VDRL titers and their correlation with CSF VDRL, does not permit a conclusive statement in this regard. A recent review of records from our institution identified 164 HIV-infected patients with syphilis (serum VDRL ≥1:2), between June and December of 2006. Most of them (n = 118, 72%) presented serum VDRL titers <1:16. Twenty three patients underwent lumbar puncture, but none standardized criteria was used, and 16 (70%) patients had serum VDRL titers ≥1:16. The 3 patients with positive CSF VDRL of this series, classified as asymptomatic neurosyphilis, had serum VDRL titers ≥1:16 (J.E. Vidal, personal communication). This survey suggests the importance of serum VDRL titers for neurosyphilis diagnosis, especially in neurologically asymptomatic patients. However, prospective studies with well-defined criteria to perform lumbar puncture are necessary before defining the diagnostic value of serum VDRL titers as a surrogate marker of asymptomatic neurosyphilis in HIV-infected patients.
Some authors question the benefit of treating asymptomatic laboratory-defined neurosyphilis,25 as was reported in the only prospective randomized trial that evaluated this question.26 On the other hand, other authors reinforce the need of active approaches to identify and treat asymptomatic neurosyphilis patients due to the potential serious subsequent complications of neurosyphilis without treatment (e.g., stroke, loss of vision, or hearing).27 These different points of view show the current and unresolved debate about the management of asymptomatic neurosyphilis.
The diagnosis of neurosyphilis in HIV-infected patients has several challenges. A reactive CSF VDRL test is considered diagnostic for neurosyphilis, but this test may be negative in as many as 70% of cases, and the diagnosis may be based solely on evidence of exposure to syphilis with other CSF features.12 The mild pleocytosis frequent in HIV-infected patients, especially in those with severe immunodepression complicates diagnosis further.28 In asymptomatic patients, the median pleocytosis was only 3 cell/μL whereas in symptomatic patients the median pleocytosis was 25 cells/μL. Pleocytosis in symptomatic patients is similar to values of other reports.23,29 The difference of our median pleocytosis from prior reports, probably results from our case definition requiring reactive CSF VDRL without reference to CSF cell count. Interestingly, we observed both asymptomatic and symptomatic patients without CSF pleocytosis, reinforcing the challenge of diagnosis in HIV-infected patients and the need to use other clinical and laboratory markers.
More than 80% of our patients had CD4+ T cell counts <350 of cells/μL, suggesting that as with other opportunistic neurologic complications, neurosyphilis becomes more likely with increasing immunosuppression due to HIV.5 Indeed, we found that a lower CD4+ cell count was associated with symptomatic neurosyphilis. This association was observed with CD4+ T cell counts ≤200 cells/μL but not with a cut-off of 350 cells/μL. This result suggests that clinical expression of neurosyphilis seem to depend on more severe immunodepression, and highlights the importance of cellular immunity in the modulation of this disease. Nevertheless, we believe neurosyphilis can be diagnosed in any stage of HIV-infection, and a high index of suspicion should be maintained to reduce symptomatic neurosyphilis and late complications.
We also identified that CSF protein concentration was significantly associated to symptomatic neurosyphilis, suggesting that the blood–brain barrier is compromised in these patients. Although the difference in pleocytosis in CSF did not reach statistical significance, there was a trend toward higher nucleated cell counts in symptomatic neurosyphilis. These results suggest a correlation between symptomatic neurosyphilis and meningeal reaction. Serum VDRL titers, a potential marker of patients with syphilis who develop neurosyphilis, were not different between asymptomatic and symptomatic patients.
This retrospective study has several limitations. These include the absence of a standardized history and neurologic examination that would result in misclassification of patients with regard to symptomatic or asymptomatic condition, the lack of a systematic ophthalmic, auditory, and imaging evaluation, and the lack of a complete neurologic evaluation at discharge and an adequate clinical and laboratory follow-up. In addition, we included only patients with positive CSF VDRL, and therefore our results are applicable to patients who met these strict diagnosis criteria (e.g., not applicable to HIV-infected patients with syphilis and CSF pleocytosis but negative CSF VDRL). Finally, the lack of the total number of syphilis cases to use as a denominator for the number of HIV-infected persons with syphilis makes difficult to conclude the true value of serum VDRL titers as marker of neurosyphilis.
In conclusion, one-third of the patients included in this study had early syphilis, and most patients were young men with asymptomatic and early neurosyphilis. Factors associated with symptomatic neurosyphilis were lower CD4+ cell counts and higher protein levels on CSF. Most patients had serum VDRL titers ≥1:16, regardless of syphilis stage.
1. Ciesielski CA. Sexually transmitted diseases in men who have sex with men: An epidemiologic review. Curr Infect Dis Rep 2003; 5:145–152.
2. Blocker ME, Levine WC, St Louis ME. HIV prevalence in patients with syphilis, United States. Sex Transm Dis 2000; 27:53–59.
3. Sutmoller F, Penna TL, de Souza CT, et al; Oswaldo Cruz Foundation STD/HIV Prevention Group. Human immunodeficiency virus incidence and risk behavior in the “Projeto Rio”: Results of the first 5 years of the Rio de Janeiro open cohort of homosexual and bisexual men, 1994–98. Int J Infect Dis 2002; 6:259–265.
4. Rodrigues EH, Abath FG. Sexually transmitted diseases in patients infected with HIV/AIDS in the State of Pernambuco, Brazil. Rev Soc Bras Med Trop 2000; 33:47–52.
5. Marra CM. Neurosyphilis. Curr Neurol Neurosci Rep 2004; 4:435–440.
6. Centers for Disease Control and Prevention. Sexually transmitted diseases treatment guidelines, 2006. MMWR Morb Mortal Wkly Rep 2006; 55:1–94.
7. Vidal-Bermudez JE, Bonasser-Filho F, Schiavon-Nogueira R. Syphilitic meningomyelitis in a patient with AIDS. Rev Neurol 2004; 38:998–999.
8. Deleted in proof.
9. Trujillo JR, Jaramillo-Rangel G, Ortega-Martinez M, et al. International NeuroAIDS: Prospects of HIV-1 associated neurological complications. Cell Res 2005; 15:962–969.
10. de Oliveira JF, Greco DB, Oliveira GC, et al. Neurological disease in HIV-infected patients in the era of highly active antiretroviral treatment: A Brazilian experience. Rev Soc Med Trop 2006; 39:146–151.
11. Malone JL, Wallace MR, Hendrick BB, et al. Syphilis and neurosyphilis in a human immunodeficiency virus type-1 seropositive population: Evidence for frequent serologic relapse after therapy. Am J Med 1995; 99:55–63.
12. Marra MC. Neurosyphilis. In: Scheld WM, Whitley RJ, Marra CM, eds. Infectious of the Central Nervous System. Philadelphia: Lippincott, Williams & Wilkins, 2004:649–658.
13. Musher DM. Syphilis, neurosyphilis, penicillin, and AIDS. J Infect Dis 1991; 163:1201–1206.
14. Vidal JE, Hernandez AV, Penalva de Oliveira AC, et al. Cerebral toxoplasmosis in HIV-positive patients in Brazil: Clinical features and predictors of treatment response in the HAART era. AIDS Patient Care STDs 2005; 19:840–848.
15. Deleted in proof.
16. Merrit HH, Adams RD, Solomon HC. Neurosyphilis. New York: Oxford, 1946.
17. Berger JR. Neurosyphilis in human immunodeficiency virus type 1-seropositive individuals: A prospective study. Arch Neurol 1991; 48:700–722.
18. Katz DA, Berger JR, Duncan RC. Neurosyphilis: A comparative study of the effects of infection with human immunodeficiency virus [published erratum in Arch Neurol 1993; 50:614]. Arch Neurol 1993; 50:243–249.
19. de Souza MC, Nitrini R. Effects of human immunodeficiency virus infection on the manifestations of neurosyphilis. Neurology 1997; 49:893–894.
20. Berger JR. Spinal cord syphilis associated with human immunodeficiency virus infection: a treatable myelopathy. Am J Med 1992; 92:101–103.
21. Zetola NM, Klausner JD. Syphilis and HIV infection: An update. Clin Infect Dis 2007; 44:1222–1228.
22. Centers for Disease Control and Prevention. Symptomatic early neurosyphilis among HIV-positive men who have sex with men—Four cities, United States, January 2002–June 2004. MMWR Morb Mortal Wkly Rep 2007; 56:625–628.
23. Marra CM, Maxwell CL, Smith SL, et al. Cerebrospinal fluid abnormalities in patients with syphilis: association with clinical and laboratory features. J Infect Dis 2004; 189:369–376.
24. Libois A, de Wit S, Poll B, et al. HIV and syphilis: When to perform a lumbar puncture. Sex Transm Dis 2007; 34:141–144.
25. Zetola NM, Engelman J, Jensen TP, et al. Syphilis in the United States: An update for clinicians with an emphasis on HIV infection. Mayo Clin Proc 2007; 82:1091–1102.
26. Rolfs RT, Joesoef MR, Hendershot EF, et al; Syphilis and HIV Study Group. A randomized trial of enhanced therapy for early syphilis in patients with or without human immunodeficiency virus infection. N Engl J Med 1997; 337:307–314.
27. Marra C. Déjà vu all over again: when to perform a lumbar puncture in HIV-infected patients with syphilis. Sex Transm Dis 2007; 34:145–146.
28. Marshall DW, Brey RL, Cahill WT, et al. Spectrum of cerebrospinal fluid findings in various stages of human immunodeficiency virus infection. Arch Neurol 1988; 45:954–958.
29. Marra CM, Maxwell CL, Tantalo L, et al. Normalization of cerebrospinal fluid abnormalities after neurosyphilis therapy: Does HIV status matter?. Clin Infect Dis 2004; 38:1001–1006.