CROI 2019 Abstract eBook

Abstract eBook

Poster Abstracts

criteria in syphilis and HIV coinfected patients and report the frequency of neurosyphilis in these patients. Methods: Retrospective cohort study carried out in all HIV-infected patients under outpatient follow-up in an academic center at São Paulo, Brazil, between 2000 and 2016. We identified all incident syphilis cases, defined as a first positive treponemal test or a ≥4-fold increase in consecutive VDRL titers. We considered lumbar puncture performed < 6 months after incident syphilis. We report the proportion of patients meeting CDC criteria for LP undergoing cerebrospinal fluid testing, and the frequency of confirmed (positive VDRL) or probable (abnormal leukocyte counts: >5 cells/ml among patients with T CD4+ counts <200 and suppressed viral loads; >20 cells/ml otherwise) neurosyphilis. Results: The initial sample comprised 3448 persons living with HIV. Incident syphilis was detected in 669 patients. Of those, 459 met CDC criteria for CSF collection, and 147 (32%, 95%CI 28-37) were referred to LP. Confirmed or probable neurosyphilis was observed in 18 cases (12%, 95%CI 7-19). Of 312 patients not referred to LP despite CDC criteria, 10 (3%, 95%CI 2-6) collected CSF within 6 to 12 months and 3 (30%, 95%CI 7-65) had abnormal results compatible with neurosyphilis. Of those with abnormal results, 13 (72%) had a positive VDRL in the cerebrospinal fluid. Conclusion: Adherence to CDC LP criteria for syphilis and HIV coinfected patients was low, despite follow-up in an academic center. In this subset of patients, the frequency of neurosyphilis was 12% for LP performed in the first 6 months and 30% among those submitted to LP within 6 to 12 months.

high frequency HL. In univariate analysis, odds of any HL were higher in HIV+, those with reactive CSF-VDRL, CSF pleocytosis, detection of Tp in blood or CSF, injection drug use (IDU), older age and higher RPR titers (Table). In multivariate analysis, odds of any HL remained higher in those with CSF pleocytosis, Tp detection in blood, IDU and older age (Table). In multivariate analysis, odds of low-mid frequency HL were higher in those with Tp detection in CSF, IDU and older age, and odds of high frequency HL were higher in those with CSF pleocytosis, Tp detection in blood, and older age (Table). Syphilis stage, current ARV use, CD4+ T-cells and plasma HIV RNA were not associated with any category of hearing loss. Conclusion: HL is common in individuals with syphilis and increases with age. While low-mid frequency HL is more likely in those with Tp detection in CSF, high frequency HL is more likely with CSF inflammation. Low-mid and high frequency HL due to otosyphilis may be due to different pathological mechanisms, and, as such, may respond differently to treatment.

Poster Abstracts

1014 COGNITIVE IMPAIRMENT IN INDIVIDUALS WITH SYPHILIS

Kevin Robertson 1 , Haley Mendoza 2 , Clare Maxwell 2 , Shelia Dunaway 2 , Arielle P. Davis 2 , Sharon Sahi 2 , Lauren Tantalo 2 , Jana Zahlan 2 , Sheryl Storey 2 , Christina Marra 2 1 University of North Carolina at Chapel Hill, Chapel Hill, NC, USA, 2 University of Washington, Seattle, WA, USA Background: Few studies have examined both cognitive function and cerebrospinal fluid (CSF) abnormalities in individuals with syphilis. Methods: 186 individuals with syphilis underwent self-administered cognitive assessment with the Cogstate battery, and 132 (71%) underwent lumbar puncture. Cognitive function was categorized as unimpaired; mild, moderate or severely impaired; or unimpaired/mild or moderate/severe based on normative Cogstate data. Serum rapid plasma reagin (RPR) test titer, detection of T. pallidum in CSF by reverse transcriptase PCR and presence of recreational drugs in urine were determined in a research laboratory, and CSF white blood cell (WBC) enumeration and CSF Venereal Disease Research Laboratory (VDRL) reactivity were determined in a clinical laboratory. Neurosyphilis was defined as a reactive CSF-VDRL alone, or reactive CSF-VDRL or CSF WBCs>10/ul. Results are described as medians (interquartile range [IQR]) or percents, and differences determined by chi square or Fisher exact test. Results: Participants were primarily men (98%), age 35 (28-46) with at least a high school education (91%); 62%were HIV infected with 90% on antiretrovirals. 82% had early syphilis, with RPR 1:64 (1:32-1:256). Urine toxicology was positive for stimulants in 19%, cannabinoids in 21% and both in 9%. Overall, 124 participants were cognitively impaired: 72 (39%) mild, 33 (18%) moderate, 19 (10%) severe. Among those with any cognitive impairment, the proportion of individuals with serum RPR titer ≥1:32 increased with increasing level of impairment (52/72 (72%) mild, 26/33 (79%) moderate, 19/19

1013 HEARING LOSS IN UNSELECTED INDIVIDUALS WITH SYPHILIS

Meena Ramchandani , Jamie Litvack, Clare Maxwell, Lauren Tantalo, Sharon Sahi, Shelia Dunaway, Christina Marra University of Washington, Seattle, WA, USA Background: Little is known about the incidence and pathophysiology of otosyphilis. Methods: Unselected individuals enrolled in a study of cerebrospinal spinal fluid (CSF) abnormalities in syphilis underwent screening audiometry, standardized medical history, lumbar puncture and venipuncture. Serum rapid plasma reagin titers (RPR) and detection of T. pallidum (Tp) in blood by PCR and CSF by RT-PCR were determined in a research laboratory. CSF white blood cell count, CSF-Venereal Disease Research Laboratory (VDRL) reactivity, CD4+ T-cells and plasma HIV RNA were determined in a clinical laboratory. Relationships between hearing loss (HL, unilateral or bilateral), defined as low-mid (500, 1000, 2000 Hz average) or high frequency (4000, 6000, 8000 Hz average) pure tone averages ≥26 dB, or either (any loss), and other variables were determined by logistic regression. For multivariate models, all variables significant at P≤0.10 were included. Those with P-values >0.05 in multivariate models were sequentially removed. Results: 362 individuals without pre-existing HL were evaluated: 99%men, mean age 41 (SD 11), 82% HIV+ (70% on antiretrovirals [ARVs]), median serum RPR titer 1:64 (IQR, 1:16-1:128), 51% treated for uncomplicated syphilis before study visit. 186 (51%) had any HL; 83 (23%) low-mid and 168 (46%)

CROI 2019 397