Work

Neurosyphilis During Acute HIV Infection: A CNS Immunologic and Virologic Characterization

Year: 2019

Type: article

Abstract: To the Editors: BACKGROUND Syphilis is a re-emerging sexually transmitted infection caused by the spirochete Treponema pallidum. Syphilis and HIV-1 infection are known to impact each other's course, and both invade the central nervous system (CNS) within days after infection.1,2 During chronic HIV-1 infection, incident syphilis leads to decline in CD4+ T lymphocytes and elevated HIV-1 RNA levels in blood.3 Advanced HIV-1 infection may increase the risk of neurosyphilis4 and syphilis treatment failure.5 Although a simultaneous diagnosis of HIV-1 and syphilis infection is not uncommon, neurosyphilis diagnosed at the time of acute HIV-1 infection (AHI) and its corresponding features in cerebrospinal fluid (CSF) and brain imaging have not been described. Here, we report an individual who presented with meningeal syphilis at AHI. CASE PRESENTATION A 21-year-old Thai man presented for HIV screening at an anonymous testing center in Bangkok where routine evaluation for AHI is performed.1 He was diagnosed with Fiebig IV AHI (HIV-1 RNA+, anti-HIV IgM+, and Western Blot indeterminate) and consented to enroll into the RV254/SEARCH010 AHI cohort (NCT00796146). He reported multiple sexual encounters with male partners including anal sex without use of condoms within 60 days before HIV testing. Acute retroviral syndrome (ARS), with symptoms including fever, headache, malaise, pharyngitis, and oral ulcer, started approximately a week before the HIV diagnosis. His medical history was unremarkable except for a single episode of childhood seizure of unknown origin. He worked as a sex worker. He was a chronic smoker and daily drinker with occasional use of methamphetamine, but he denied intravenous drug use. Five months before the AHI, his blood rapid plasma reagin was 1:32, and he received a single intramuscular injection of penicillin-G. Cardiovascular, respiratory, abdominal, and neurological examinations were unremarkable. There was bilateral tender inguinal lymphadenopathy without penile lesion. Initial laboratory tests showed normal complete blood count, kidney function, and clotting profile. Liver enzymes (alanine aminotransferase, aspartate aminotransferase, and gamma-glutamyl transferase) were mildly elevated. The plasma HIV-1 RNA level was 5.04 Log10 copies/mL; CD4+ and CD8+ T-lymphocyte counts were 236 and 370 cells/µT, respectively. Sexually transmitted infection screening revealed anal chlamydia and gonorrhea, hepatitis C (HCV) infection with HCV RNA 4.47 log10 copies/mL; rapid plasma reagin titer in the local laboratory rose from 1:32 five months ago to 1:128 while plasma venereal disease research laboratory (VDRL) titer was 1:64 (Table 1). Given the possibility of syphilis treatment failure and the fact that late syphilis could not be ruled out based on the available history, 3 once-weekly intramuscular penicillin-G injections (2.4 MIU) were given. Oral azithromycin and intramuscular ceftriaxone were prescribed for chlamydia and gonorrhea infection.TABLE 1.: Laboratory Parameters at Acute HIV Infection and 24-Week Follow-upHe underwent lumbar puncture for research purposes, followed by immediate initiation of antiretroviral therapy (ART) consisting of dolutegravir, abacavir, and lamivudine. CSF examination revealed a lymphocytic pleocytosis of 105 cells/mm3 with 99% mononucleated cells. CSF protein and glucose levels were normal, and CSF-VDRL was nonreactive. Magnetic resonance imaging (MRI) of the brain (Fig. 1 top) demonstrated hyperintense signals along the sulci of the bilateral occipital and high frontal regions on axial T2-FLAIR (fluid-attenuated inversion recovery) images with fat suppression, suspicious of underlying leptomeningitis. A stored CSF sample was positive for T. pallidum hemagglutination assay (TPHA). Consistent with the diagnostic criteria of neurosyphilis in the US CDC guideline for HIV-infected individuals, he was treated with intravenous penicillin-G injection, 3 mega-units every 4 hours for 14 days.FIGURE 1.: MRI of the brain [axial T2-weighted fluid-attenuated inversion recovery (FLAIR) with fat suppression] at acute HIV infection (top) and 24 weeks (bottom). Arrows: hyperintense signals along the sulci of the bilateral occipital and high frontal regions, suggestive of underlying leptomeningitis.His acute retroviral syndrome symptoms resolved soon after ART, before penicillin treatment. The plasma HIV-1 RNA was once suppressed (<20 copies/mL) at 12 weeks with a CD4+ T-lymphocyte count of 565 cells/µT. Subsequently, he stopped taking ART at 20 weeks for fear of disclosure after moving in with a new partner and presented to us at 24 weeks having an HIV-1 RNA level of 5.44 log10 copies/mL and CD4+ T-lymphocyte count of 443 cells/µs. His blood VDRL titer dropped from 1:64 at baseline to 1:8. Repeat neurological examination was unremarkable. Repeat lumbar puncture showed incomplete improvement of CSF pleocytosis of 10 cells/mm3 with 100% mononuclear cells. CSF HIV-1 RNA was 3.84 log10 copies/mL. Repeat brain MRI using the same scanner and scanning parameters demonstrated mostly resolved leptomeningeal abnormalities (Fig. 1 bottom) on axial T2-FLAIR images with fat suppression. Given his unique presentation, his CSF and peripheral blood mononuclear cell (PBMC) samples at baseline and 24 weeks were sent for flow cytometry to examine for neurosyphilis-associated CSF B-lymphocyte pleocytosis.6 In general, the CSF CD19+ B/CD3+ T-lymphocyte ratio (B/T ratio) is less than 0.1 as T-lymphocyte activation is the predominant immune response at AHI.7 At baseline, 57% of lymphocytes in CSF were CD19+ B lymphocytes, with a B/T ratio of 1.42. Such B-lymphocyte expansion was compartmentalized in CSF but not in the patient's PBMC sample, in which only 15% of lymphocytes were CD19+ B lymphocytes and the B/T ratio was 0.25. At 24 weeks, the frequency of CD19+ B lymphocytes in CSF dropped from 57% to 4%, with a corresponding B/T ratio of 0.06 (Table 1). Together with the improvement in imaging, plasma VDRL, and CSF pleocytosis, the findings supported the resolution of neurosyphilis. DISCUSSION We describe MRI findings and an unusual pattern of proportions of CSF cellular subsets that we posit are due to a combination of AHI and neurosyphilis. Both HIV-1 and T. pallidum can invade the CNS and cause CSF pleocytosis2; however, AHI monoinfection seldom causes CSF pleocytosis greater than 20 cells/µL (unpublished observation). Our group has examined CSF samples and brain imaging from more than 100 individuals with untreated AHI, but this is our first encounter of a grossly elevated CSF white cell count (WCC) accompanied by leptomeningeal abnormalities on neuroimaging. Leptomeningitis is a common manifestation of neurosyphilis, presenting as stand-alone meningitis, or in combination with vascular and cranial nerve complications. Our patient lacked overt symptoms but had CSF and neuroimaging abnormalities consistent with meningeal syphilis. Elevated CSF protein levels and CSF pleocytosis occur in up to 70% of neurosyphilis cases. CSF pleocytosis is more pronounced in HIV-infected individuals who present with neurosyphilis.8 In particular, CSF pleocytosis in HIV-1 infection is likely linked to CSF HIV-1 RNA levels rather than the severity of the immunocompromised state or to CD4+ T-lymphocyte count.9 Current recommendations suggests a cutoff of CSF WCC >5 cells/mm3 to be used as diagnostic criteria for neurosyphilis in HIV-infected individuals on ART with undetectable plasma HIV-1 RNA, whereas a cutoff of CSF WCC >20 cells/mm3 is recommended for viremic individuals.10 CSF-VDRL is currently considered the gold standard for diagnosis of neurosyphilis. CSF-VDRL is highly specific, but its sensitivity is reported to be between 30% and 70%.10 By contrast, CSF Treponemal specific tests such as CSF-TPHA and CSF-TPPA (T. pallidum particle agglutination) are sensitive but less specific.10 Recent studies suggest the implementation of a titer cutoff in CSF-TPHA and CSF-TPPA tests to improve their specificity.11,12 CSF CXCL-13 (C-X-C Motif Chemokine Ligand 13), a B-cell chemoattractant, has also been proposed to improve the diagnostic yield as the CSF usually contains a high concentration of B lymphocytes in neurosyphilis.6 Our participant's virologic profile was not typical of AHI monoinfection. His CSF HIV-1 RNA level was high with respect to that in plasma. His plasma-CSF HIV-1 RNA ratio (PCratio), mathematically equivalent to log10 plasma HIV-1 RNA minus log10 CSF HIV-1 RNA, was 0.81 log10 copies/mL. In general, the PCratio is around 2 log10 within first year of HIV-1 infection,1,13 and around 1 log10 in chronic infection. Given the lack of a well-established CNS HIV-1 reservoir at AHI, a reduced PCratio suggests an increased HIV-1 CNS transmigration from the systemic circulation. We previously reported a reduced PCratio (<1 log10) in around 7% of individuals at AHI.13 Reduced PCratio was not associated with systemic syphilis, but the effect of neurosyphilis on CSF HIV-1 RNA was unclear. We speculate that the “improvement” of PCratio from 0.81 log10 copies/mL at baseline to 1.60 log10 copies/mL after neurosyphilis treatment was due to eradication of intrathecal T. pallidum, which facilitates not only viral replication in chronic HIV-1 infection but also CNS invasion in AHI. There are growing concerns about the effect of syphilis on HIV-1 infection. Although individuals with neurosyphilis showed higher levels of CSF chemokines and systemic monocyte activation than systemic syphilis alone,14 whether neurosyphilis, or previous systemic syphilis, links to a lower cognitive performance is inconclusive.14,15 Without clear biomarkers for the duration of syphilis infection, we are unable to determine whether our participant acquired the syphilis concurrently with or before the HIV-1 transmission. However, our patient's presentation may indicate a facilitating role of neurosyphilis in accelerating HIV-1 CNS invasion at AHI, which may hasten the establishment of CNS compartmentalization and eventually the development of neurocognitive complications in HIV-1 infection. CONCLUSIONS Our case illustrates that neurosyphilis at AHI may closely resemble the radiological and CSF patterns seen in chronic HIV-1 infection. Future research should examine the role of syphilis, and particularly neurosyphilis, in early HIV-1 infection to determine whether it alters disease progression.

Source: JAIDS Journal of Acquired Immune Deficiency Syndromes

DOI: 10.1097/qai.0000000000002114