3F1 Ocular Syphilis and HIV Coinfection: An Opportunity to Identify Previously Undetected HIV

Thursday, September 22, 2016: 10:45 AM
Room 208/209
Anna Cope, PhD, MPH1, Nicole Dzialowy, MSc2, Sara Oliver, MD, MSPH3, Jessica Rinsky, PhD, MPH2, Mara Larson, MPH2, Erika Samoff, PhD MPH2 and Victoria Mobley, MD MPH2, 1Epidemiology and Statistics Branch, Division of STD Prevention, Centers for Disease Control and Prevention, Atlanta, GA, 2Communicable Disease Branch, North Carolina Division of Public Health, Raleigh, NC, 3Division of Viral Disease, National Center for Immunization and Respiratory Diseases, Centers for Disease Control and Prevention, Atlanta, GA

Background:  Reports of ocular syphilis (OS) have increased in North Carolina since 2014 and the association with HIV is an important clinical concern. To better understand this association, we compared OS prevalence by HIV status and timing of HIV diagnosis among syphilis patients. 

Methods:  We reviewed new syphilis case reports during 2014-2015. OS was defined as a confirmed syphilis case with ocular symptoms. Syphilis cases with HIV co-infection were stratified into previously-diagnosed (>30 days before syphilis diagnosis) and concurrently-diagnosed (±30 days from syphilis diagnosis) cases. We used prevalence ratios (PR) and 95% confidence intervals (CI) to compare OS prevalence by HIV status. We compared the timing of HIV diagnosis by age, syphilis stage and closest HIV viral load (VL) and CD4 count (within 1 year) to the syphilis diagnosis date.

Results:  Of the 4232 confirmed syphilis cases (63 OS; 4169 non-OS), 1694 (40%; 35 OS; 1659 non-OS) were HIV co-infected. OS was more prevalent in HIV-positive versus HIV-negative and unknown-status patients (PR: 1.9; 95% CI: 1.1, 3.1) and if HIV-positive, among concurrently-diagnosed versus previously-diagnosed patients (PR: 2.5; 95% CI: 1.2, 5.0). At syphilis diagnosis, concurrently-diagnosed OS patients had a higher median VL (65,780 versus 44,024 copies/ml, p=0.08), were older (median: 48 versus 27 years; p=0.0004) and were more likely to be classified as late latent syphilis cases (82% (9/11) versus 37% (93/253), p=0.004) compared to non-OS patients. OS was associated with a lower median CD4 count in both concurrently (OS=260 versus non-OS=347 cells/μl, p=0.04) and previously-diagnosed patients (OS=388 versus non-OS=511 cells/μl, p=0.03).

Conclusions:  OS was common among concurrently-diagnosed patients, particularly those with late latent syphilis. The association between older age, lower CD4 counts and OS prevalence suggests immune status may play a role in OS development and highlights the importance of HIV screening in all syphilis patients, especially those with OS.