CROI 2019 Abstract eBook

Abstract eBook

Poster Abstracts

(100%) severe, P=0.02). Individuals with asymptomatic syphilis (early latent or late latent) were more likely than those with primary or secondary disease to have moderate/severe impairment (33/97 (34%) vs. 19/89 (21%), P=0.05). There was no relationship between cognitive impairment and HIV status, a positive toxicology screen, either definition of neurosyphilis, or detection of T. pallidum in CSF. Conclusion: Cognitive impairment was common in this cohort of individuals with syphilis, was not associated with HIV status or neurosyphilis, but was more common in those with high serum RPR titers, and those with latent syphilis. These results suggest that cognitive impairment in individuals with syphilis may be related to bacterial burden and may be seen in those without symptoms or signs of syphilis. 1015 RESPONSE TO SYPHILIS TREATMENT: CDC GUIDELINES IN HIV-INFECTED ADULTS ON cART Melody Ren 1 , Leah Szadkowski 2 , Darrell H. Tan 3 , Sharon Walmsley 4 1 University of Toronto, Toronto, ON, Canada, 2 University Health Network, Toronto, ON, Canada, 3 St. Michael’s Hospital, Toronto, ON, Canada, 4 Toronto General Research Institute, Toronto, ON, Canada Background: Guidelines define an adequate response to syphilis treatment as a four-fold decrease in serum RPR at 6-12 (primary, secondary syphilis) and 12-24 months (early latent, late latent, neurosyphilis). Previous studies reported that 15%–20% of HIV infected persons with primary and secondary syphilis will not achieve the fourfold decline at 1 year after treatment. We assessed if CDC guidelines capture the timeline of serologic response to syphilis treatment in HIV-positive adults in the era of modern ARVs. Methods: We conducted a chart review of 532 HIV-positive adults with positive syphilis serology between 2000 and 2017. Inclusion criteria were: reactive pre-treatment RPR titer; documentation of date and type of syphilis therapy; reversion to a non-reactive RPR or at least 6 months or 1 year of follow-up for early syphilis and late syphilis/neurosyphilis, respectively. Only the first eligible episode was included. Time to four-fold decrease was calculated using Kaplan Meier estimates. Univariable proportional hazards models assessed associations between clinical covariates and time to four-fold decrease. Results: 189 male patients (87%MSM) met inclusion criteria. At syphilis diagnosis, median age (IQR) was 42 (35, 48), median CD4 count (IQR) was 443 (273, 609). 56% had a suppressed viral load (VL). 75%were on ARVs. 12% were primary syphilis, 28% secondary, 12% early latent, 28% late latent, 19% neurosyphilis; stage was undefined for 1%. It was the first syphilis episode for 134 patients (71%), 55 (29.5%) had had previous syphilis. 72% received IM benzathine Penicillin G (27% 1 dose, 45% 2-3 doses), 21% IV Penicillin G, and 5% doxycycline. Median follow-up (IQR) was 2.55 (1.53, 6.14) years. In patients with suppressed VL 42 (97.7%) with primary or secondary syphilis experienced four-fold decrease by 12 months, 31 (91.2%) with early or late latent syphilis and 11 (84.6%) with neurosyphilis by 24 months compared to 21 (87.5%), 26 (89.7%), 18 (100%) in those without VL suppression. Overall, the cumulative incidence of achieving a four-fold decrease at 12 months was 0.94 (95% CI 0.83, 0.99) in patients with suppressed VL and 0.96 (95% CI 0.75, 1) in non-suppressed patients (p=0.56). Age, CD4 count, previous syphilis, current syphilis stage, and number of treatment courses were not associated with time to four-fold decrease. Conclusion: We observed high rates of serologic response to syphilis treatment in HIV infected adults engaged in care that was not impacted by VL suppression with cART. 1016 “TREAT ALL” ADOPTION IMPROVES RAPID TREATMENT INITIATION IN 6 SUB-SAHARAN COUNTRIES Olga Tymejczyk 1 , Ellen Brazier 1 , Constantin T. Yiannoutsos 2 , Peter F. Rebeiro 3 , Kara K. Wools-Kaloustian 2 , Mary-Ann Davies 4 , Elizabeth Zaniewski 5 , Mark Urassa 6 , Jean d’Amour Sinayobye 7 , Nanina Anderegg 5 , Grace Liu 1 , Nathan Ford 8 , Denis Nash 1 , for the IeDEA Consortium 1 City University of New York, New York, NY, USA, 2 Indiana University, Indianapolis, IN, USA, 3 Vanderbilt University, Nashville, TN, USA, 4 University of Cape Town, Cape Town, South Africa, 5 Institute of Social and Preventive Medicine, Bern, Switzerland, 6 National Institute for Medical Research, Mwanza, Tanzania, United Republic of, 7 Rwanda Military Hospital, Kigali, Rwanda, 8 WHO, Geneva, Switzerland Background: Most countries have formally adopted the World Health Organization’s 2015 recommendation of universal HIV treatment (Treat All). Although effects of universal treatment eligibility interventions have been

examined in large trials and using modeled data, there are few rigorous assessments of the real-world impact of Treat All on antiretroviral treatment (ART) uptake across different contexts. Methods: We used longitudinal data for 814,603 patients enrolling in HIV care during 2004-2018 in six sub-Saharan African countries participating in the International epidemiology Databases to Evaluate AIDS (IeDEA) consortium (Burundi, Kenya, Malawi, Rwanda, Uganda, and Zambia). Using a quasi- experimental regression discontinuity design, we assessed the change in the proportion of individuals initiating treatment within 30 days of enrollment in HIV care (rapid ART initiation) after country-level adoption of Treat All policies. A modified multivariable Poisson model was used to identify factors associated with failure to initiate ART rapidly among persons enrolling in HIV care under Treat All. Results: In all countries, national adoption of Treat All was associated with large increases in rapid ART initiation. The greatest increase in rapid ART initiation immediately after Treat All policy adoption was observed in Rwanda, from 44.4% to 78.9% of patients (34.5 percentage points (pp); 95% CI: 27.2-41.7 pp), Kenya (25.7pp, 95% CI: 21.8 to 29.5pp), and Burundi (17.7pp, 95% CI: 6.5 to 28.9pp), while the rate of rapid ART initiation accelerated sharply following Treat All policy adoption in Malawi, Uganda, and Zambia. Under Treat All, younger patients (16-24 years) and men were at increased risk of not rapidly initiating ART (compared to older patients and women, respectively). However, rapid ART initiation following enrollment increased for all groups as more time elapsed since Treat All adoption. Conclusion: Adoption of Treat All policies had a strong effect on increasing rates of rapid ART initiation and increases followed different trajectories across the six countries. Adoption and implementation of Treat All policies should be accelerated, with particular care to identify and address possible inequities in access to treatment by subgroups at higher risk of not rapidly initiating treatment following diagnosis and care enrollment.

Poster Abstracts

1017 OUTCOMES OF “TEST AND START” AT SCALE WITHIN HAITI’S NATIONAL ART PROGRAM Nancy Puttkammer 1 , Kesner Francois 2 , Yrvel Desir 3 , Gracia Desforges 2 , Nadjy Joseph 3 , Nathaelf Hyppolite 4 , Jean Guy Honore 5 , Jean Solon Valles 6 , Ermane Robin 2 , Patrice Joseph 6 1 University of Washington, Seattle, WA, USA, 2 Ministry of Public Health and Population, Port-au-Prince, Haiti, 3 NASTAD, Port-au-Prince, Haiti, 4 Centre Haïtien pour le Renforcement du Système de Santé, Port-au-Prince, Haiti, 5 I-TECH, Port-au- Prince, Haiti, 6 CDC, Port-au-Prince, Haiti Background: In July 2016, Haiti’s Ministry of Health endorsed the universal “test and start” (T&S) strategy, offering HIV antiretroviral therapy (ART) to all patients upon diagnosis with HIV, regardless of health status. The outcomes of Haiti’s nationally scaled-up T&S initiative have not previously been described. This study’s aims were to: 1) describe trends in timeliness of ART initiation before and after July 2016; and 2) explore the association between rapid initiation and retention on ART. Methods: Our retrospective cohort study included data from 148,680 patients who received a first HIV diagnosis at one of 94 hospitals and clinics in Haiti from 2004 through March 2018. Data were drawn from a large electronic medical record system as well as Haiti’s national HIV/AIDS case based surveillance data system. We studied trends in linkage to care and ART initiation using descriptive

CROI 2019 398