Establishing a clinical prediction model of neurosyphilis via a lumbar-puncture-free nomogram

Neurosyphilis (NS) can cause a range of central nervous system (CNS) damage, from asymptomatic states to severe mental disorders. While lumbar puncture is a reliable diagnostic method for NS, it is often poorly accepted due to its invasive nature, particularly by patients with mild symptoms. This study aims to develop a prediction model for the early diagnosis of NS that without requiring lumbar puncture.

Clinical data, including imaging, routine blood tests, immune markers, HIV status and cerebrospinal fluid (CSF) examination, were collected from the Shanghai Public Health Clinical Center from 2021 to 2023,. The dataset was randomly split into training and validation sets in an 8:2 ratio. Univariate analysis, the least absolute shrinkage and selection operator (LASSO) and multivariate logistic regression were used to identify significant predictors.

Out of 1078 suspected patients, 702 confirmed syphilis patients with positive Treponema pallidum specific antibody (anti-TP) and positive Treponema Pallidum Particle Agglutination (TRUST) in serum were selected. Among them, 246 patients were diagnosed as NS based on the positive anti-TP in CSF, while 456 patients were classified as non-neurosyphilis (NNS). In the NS group, cerebral ischemia/infarction, ataxia, decreased vision, mental/behavioral disorders, memory impairment, high levels of anti-TP and TRUST (>1:16), elevated red blood cell (RBC) and platelet (PLT) counts, and shorter prothrombin time (PT) were identified. A nomogram was established based on independent prognostic factors. The receiver operating characteristic (ROC) curves and calibration curves showed high predictive accuracy, and clinical decision curve analysis (DCA) indicated good clinical applicability.

The developed nomogram offers a reliable and non-invasive method for early identification of NS in syphilis patients, regardless of their HIV or immune status.