BACKGROUND

Diagnosis of neurosyphilis is challenging since the clinical presentation is diverse and nonspecific, and classical cerebrospinal fluid tests (nontreponemal test, cell count, protein concentration) demonstrate low diagnostic accuracy.

OBJECTIVE

To develop and validate a diagnostic model to complement existing neurosyphilis diagnostic procedures based on biomarkers associated with intrathecal humoral immune response.

MATERIAL AND METHODS

A set of cerebrospinal fluid tests, including the concentration of chemokine CXCL13, indicators of intrathecal immunoglobulin synthesis and permeability of the blood-cerebrospinal fluid barrier, and the concentration of immunoglobulins (G/M/A) was measured in 237 syphilis-positive individuals who were HIV-negative and had no signs of tick-borne infections. CSF samples were categorized for intrathecal humoral immune response to the following types: 1) “induced”, 2) “active”, 3) persistent”, 4) “indeterminate”, or 5) “absent”, depending on the increase in CXCL13 concentration (1, 2), evidence of intrathecal synthesis (2, 3), increase in immunoglobulin levels (4) or the absence of deviations of all parameters studied (5).

RESULTS

Classical tests were significantly more sensitive for the “active” type in untreated patients (100%; 95% CI 89-100%) than in previously treated (42%; 95% CI 22-63%). The “induced” type in secondary syphilis was also poorly recognized (14%; 95% CI 0.3-58%). After intravenous administration of high-dose benzylpenicillin, the severity of cerebrospinal fluid abnormalities decreased significantly, but the “active” type persisted in 67% (95% CI 42-85%) for 0.7-2.8 years after therapy, which was detected by classical tests only in 33% (95% CI 14-57%) of cases. In the “persistent”, “indeterminate”, or “absent” types, only anecdotal and insignificant deviations of the tests were observed.

CONCLUSION

Our study proposes an immunobiological diagnostic model that provides a theoretical basis for evaluating tests for neurosyphilis diagnosis.