CROI 2015 Program and Abstracts

Abstract Listing

Poster Abstracts

830 Decreased TB Recurrence After Introduction of ART in Durban, South Africa Yuri F. van der Heijden 1 ; Farina Karim 2 ; Gary Parker 2 ;Tilagavathy Chinappa 3 ; Grace Mufamadi 3 ; Linda Zako 3 ; Bryan E. Shepherd 1 ;Timothy Sterling 1 ; Alexander Pym 2 1 Vanderbilt University, Nashville, TN, US; 2 KwaZulu Natal Research Institute for TB and HIV (K-RITH), Durban, South Africa; 3 Ethekwini Municipality, Durban, South Africa

Background: Recurrent tuberculosis (TB) threatens TB control. Late recurrence is usually due to reinfection. HIV-infected persons are at increased risk of reinfection, which could be due at least in part to immune compromise. We hypothesized that the introduction of antiretroviral therapy (ART) in a setting with high TB incidence and HIV co-infection would be associated with decreased TB recurrence risk. Methods: We conducted a retrospective cohort study of all adult (> 18 years) TB patients seen at a single, large, urban TB clinic in Durban, South Africa (Prince Cyril Zulu Communicable Diseases Clinic) from January 2000 through December 2012. Follow-up was through December 2013. TB recurrence was defined as a TB episode occurring after treatment completion or cure of a prior TB episode. We compared time to first recurrence among those whose first visit occurred before the introduction of ART into the South African public health system (2000–2005) and after (2006–2012), with attention to early ( ≤ 365 days from first episode treatment start date) and late (>365 days) TB recurrence. Results: There were 71,235 adult TB patients seen at the clinic during the study period, of whom 5,365 (8%) had at least one TB recurrence. There were 573 (11%) early and 4,792 (89%) late TB recurrences (median 0.9 years, IQR 0.8, 0.9 years for early; median 2.8 years, IQR 1.8, 4.4 years for late). Those whose first TB episode occurred in 2006-2012 were significantly less likely to have recurrence at the same time after cure than those whose first TB episode occurred in 2000-2005 using a Cox proportional hazards model adjusting for age at first episode, sex, and race (Hazard Ratio [HR] 0.66; 95% Confidence Interval [CI] 0.62–0.70; p<0.001). HIV status was available for 77% of patients from January 2009 through December 2012. Among 13,180 patients with known HIV status whose initial TB episode occurred during those years, 9,837 (75%) were sero-positive. In a separate Cox proportional hazards model adjusting for the same demographic factors, patients living with HIV were more likely to have a TB recurrence than those who were not HIV positive (HR 1.96; 1.52–2.54; p<0.001). Conclusions: In this large TB cohort, most episodes of TB recurrence occurred late, suggesting exogenous reinfection rather than relapse. TB recurrence risk decreased after ART was implemented in Durban, supporting the role of improved immune status on decreasing the risk of TB reinfection. 831 Incidence of Active Tuberculosis in HIV-Infected Adults and Mortality in Thailand Nicolas Salvadori 1 ; Suwalai Chalermpantmetagul 1 ; Julie Figoni 1 ; SuchartThongpaen 2 ; Ampaipith Nilmanat 3 ; Patinun Chirawatthanaphan 4 ; PramualThaingamsilp 5 ;Tim R. Cressey 1 ; Nicole Ngo-Giang- Huong 1 ; Gonzague Jourdain 1 1 Institut de Recherche Pour le Développement UMI 174-PHPT, Chiang Mai, Thailand; 2 Mahasarakham Hospital, Mahasarakham, Thailand; 3 Hat Yai Hospital, Hat Yai, Thailand; 4 Phaholpolpayuhasaena Hospital, Kanchanaburi, Thailand; 5 Kalasin Hospital, Kalasin, Thailand Background: Thailand is one of the 22 high tuberculosis (TB) burden countries listed by the WHO. We estimated the incidence of active TB in HIV-infected adults, investigated the association between characteristics at antiretroviral therapy (ART) initiation and TB diagnosis, and compared survival rates of adults with and without active TB in a large HIV cohort in Thailand. Methods: ART-naïve adults who enrolled in the PHPT cohort (NCT00433030) between 1999 and 2012 were included in this analysis. Screening for active TB was based on interview, clinical examination, chest X-ray and sputum smear. Incidence was the number of new cases divided by that of person-years of follow-up (PY). Using incidence rate ratios (IRRs) from Poisson regression models, the association between sex, age, education level, BMI, HIV RNA load, CD4 count, period of enrollment and complete blood count parameters at ART initiation and TB diagnosis was assessed. Survival rates were estimated and compared using Kaplan-Meier method and log-rank test. Results: At ART initiation, 1702 adults (82% female) had a median age (IQR) of 31.5 years (27.1-36.8), HIV RNA load 4.8 log 10 copies/mL (4.1-5.2) and CD4 count 144 cells/mm 3 (67- 218). Median follow-up was 6.9 years (2.4-8.3). Overall incidence rate of active TB was 0.98/100 PY (95% CI 0.80-1.19) (99 cases). Incidence rates decreased with ART duration, from 5.4/100 PY within the first 6 months to 0.24/100 PY after 5 years. Median time until TB diagnosis was 7.4 months (1.4-28.4). Male gender (IRR 2.0) and BMI <18.5 kg/m 2 (IRR 4.0) at ART initiation were significantly associated with TB diagnosis (p <0.001), but not age above median (IRR 0.8, p = 0.20). Adjusting for gender and age, TB diagnosis was associated with (all p <0.001): higher (above median) HIV RNA load (IRR 2.2) and neutrophils (IRR 2.2), and lower (belowmedian) lymphocytes (IRR 2.2), CD4 count (IRR 2.7), hemoglobin (IRR 2.8) and hematocrit (IRR 2.7). Of the 99 adults with TB diagnosis, 29 died (median survival time after diagnosis: 2.9 months (IQR 1.1-8.9)). Cumulative survival rates after ART initiation were 87% at 1 year, 73% at 5 years and 67% at 10 years in adults with active TB, versus 97%, 95% and 92% in those without active TB (p <0.001). Conclusions: Active TB is a major cause of death in this HIV cohort in Thailand as in many settings. Most reported predictors are available in many ART programs. They should be carefully considered to accelerate TB diagnosis and treatment in patients initiating ART. 832 Antiretroviral Scale-up and Tuberculosis Mortality in High-Burden Countries Eline Korenromp 2 ; Eran Bendavid 1 ; IsabelYan 1 1 Stanford University, Stanford, CA, US; 2 Futures Institute, Glastonbury, CT, US Background: Antiretroviral therapy (ART) is thought to reduce mortality from active TB. We investigate the extent to which the increasing availability of ART predicts reductions in national TB mortality in countries with high dual HIV-TB burden. Methods: We analyzed annual TB mortality rates in the 41 countries that the World Health Organization (WHO) defines as high HIV-TB burden, between 1996 and 2011. We analyzed two TB mortality outcomes: (1) TB death notifications reported by national TB control programs, adjusted for TB case detection rates, and (2) WHO estimates. National Coverage with ART, the percentage of treatment-eligible HIV+ population on ART, was obtained from UNAIDS. Panel linear regressions with country fixed effects tested the relationship of TB mortality to ART coverage scale-up, controlling for time-varying HIV prevalence (5-year lagged), coverage of TB interventions, gross domestic product per capita, health spending from domestic sources, and urbanization. Results: Across multiple model specifications and lags between ART coverage and TB mortality, we consistently observed that increasing ART coverage was followed by reduced TB mortality. Using death notifications and a 2-year lag between ART scale-up and TB mortality, we find that a 1% increase in ART coverage was associated with a 0.9% faster decline in TB deaths (p=0.002). Using WHO death estimates as outcome, a 1% increase in ART was followed by 1.1% reduced TB deaths (p<0.001). TB mortality was higher at higher HIV prevalence (p<0.001), but not related to coverage of isoniazid preventive therapy, cotrimoxazole preventive therapy, or other covariates.

Poster Abstracts

504

CROI 2015