CROI 2018 Abstract eBook

Abstract eBook

Poster Abstracts

Methods: In a nationally-representative cross-sectional survey conducted from October 2015–September 2016, samples were collected from 2561 persons aged ≥15 years on ART ≥12 months from 50 health facilities in Kenya. HIV-1 RNA quantification was performed on 2497 of the collected DBS samples. Dried blood spots were prepared fromwhole blood on Whatman 903 DBS filters for analysis using the Abbott m2000rt system. We assessed VL suppression (VL less than 1000 copies/ml) and factors associated with non-suppression using logistic regression. Missing covariates were multiply imputed and analyses weighted to account for complex survey design. Results: Of 2497 patients with samples analyzed for VL, 744 (29.8%) were men. Median age was 41.3 years (inter-quartile range [IQR] 34.5–49.0) and 596 (23.2%) were youth aged 15-24 years. Median time on ART was 5.3 years (IQR 2.9–7.9); 2474 (99.3%) were on NNRTI-based regimens at ART initiation. Overall viral suppression was 83.5% (95% confidence interval [CI] 78.7–88.3) with no difference between women and men (p=0.952), nor based on ART regimen. (p=0.206). Suppression was lower among adolescents aged 15–19 years (55.6% [95% CI 42.8–68.4]) and highest among persons aged ≥55 years (87.6% [95% CI 80.2–95.1]). Students had lower suppression rates, 53.9% (95% CI 35.5–72.4), compared to employed persons at 88.0% (95% CI 85.3–90.7). In multivariable logistic regression, younger age (15–24 years) was independently associated with non-suppression (adjusted odds ratio (AOR) =2.18, 95% CI 1.29–3.70), as was being a student versus employed (AOR=2.08, 95% CI 1.04–4.13) and non- disclosure to spouse, however, there was no difference in viral suppression by time on ART, baseline CD4, WHO stage, or BMI. Conclusion: Overall viral suppression among ART patients in Kenya is approaching 90%, but youth and adolescents show substantially lower suppression rates. Identifying and implementing effective interventions to help young people and students achieve viral suppression should be prioritized. 849 STUNTING AND GROWTH OF HIV-INFECTED ADOLESCENTS: HOW TO INTERPRET PROGRAMMATIC DATA? Julie Jesson 1 , Michael Schomaker 2 , Karen Malateste 3 , Dewi K. Wati 4 , Azar Kariminia 5 , Mariam Sylla 6 , Kouakou Kouadio 7 , Shobna Sawry 8 , Michael J. Vinikoor 9 , Samuel Ayaya 10 , Rachel Vreeman 11 , Mary-Ann Davies 2 , Valériane Leroy 1 1 INSERM, Toulouse, France, 2 University of Cape Town, Cape Town, South Africa, 3 INSERM, Bordeaux, France, 4 Sanglah Hospital, Bali, Indonesia, 5 Kirby Institute, Sydney, NSW, Australia, 6 Hôpital Gabriel Touré, Bamako, Mali, 7 CIRBA, Abidjan, Côte d’Ivoire, 8 University of the Witwatersrand, Johannesburg, South Africa, 9 Centre for Infectious Disease Research in Zambia, Lusaka, Zambia, 10 Moi University, Eldoret, Kenya, 11 Indiana University, Indianapolis, IN, USA Background: Among the increasing cohort of HIV-infected adolescents worldwide, little is known about their growth. We aimed to estimate the prevalence of stunting and describe growth of HIV-infected adolescents in the International Epidemiologic Databases to Evaluate AIDS (IeDEA) consortium of programmatic cohorts, where data come from patient cohorts in the course of routine HIV care. Methods: Data from sub-Saharan Africa, the Asia-Pacific, and the Caribbean and South America collected between 2003 and 2016 were used. All HIV- infected patients with at least one height measurement available while aged 10-19 years were included. If HIV exposure type was not reported, perinatal infection was assumed as having entered care before 15 years of age, and behavioral infection after 15 years. Factors associated with stunting (Height- for-Age Z-score [HAZ] <-2 SD, WHO Child Growth Standards) at age 10, 15 and 18 years were studied using logistic regression. Growth HAZ curves were stratified by sex, and stunting was described at entering or leaving (death or drop-out) care between 10 and 19 years. Results: Overall, 50,434 adolescents met the inclusion criteria; 59% from Southern Africa, 20% from East Africa. Median age at first visit was 11.2 years (Interquartile Range [IQR] 7.1-16.1), 60%were female, 95% received antiretroviral therapy (ART), and 70%were perinatally infected. Prevalence of stunting at 10, 15 and 18 years was 36% (N=19078), 45% (N=14292) and 27% (N=11332), respectively. Factors associated with stunting included late age at ART initiation (e.g. at 15y: adjusted Odds Ratio [aOR] 10-15y vs 0-5y=2.5 [2.1- 3.0]); and low CD4 count at current age (e.g. at 10y: aOR <350 vs >500 cells/mL =1.9 [1.7-2.1]). At 18y, those perinatally infected were more likely stunted than those behaviorally infected (aOR=1.3 [1.0-1.5]). Growth trajectories between 10 and 19 years differed by sex; HAZ decreased between 10 and 15 years old, then increased (Figure 1). Those entering or leaving care between 10 and 15 years

were slightly more stunted than those entering or leaving care between 15 and 19 years (entering: 49% vs 22%, leaving: 46% vs 28%). Conclusion: Stunting is a major concern among HIV-infected adolescents worldwide. Our growth curves were generated from a mixed population, with perinatally infected adolescents often having advanced HIV and stunted, compared to the mostly healthier behaviourally infected. This heterogeneity is relevant for future nutritional programs.

Poster Abstracts

850 PREVALENCE OF GLOMERULAR DYSFUNCTION AMONG PERINATALLY HIV-INFECTED THAI ADOLESCENTS Tavitiya Sudjaritruk 1 , Suparat Kanjanavanit 2 , Linda Aurpibul 3 , Tawalchaya Chotecharoentanan 3 , Saowalak Sarachai 3 , Kamolrawee Sintupat 3 , Ratchaneekorn Khampan 3 1 Chiang Mai University, Chiang Mai, Thailand, 2 Nakornping Hosp, Chiang Mai, Thailand, 3 Research Institute for Health Sciences, Chiang Mai University, Chiang Mai, Thailand Background: This study aimed to assess the prevalence and associated factors of glomerular dysfunction among perinatally HIV-infected Thai adolescents. Methods: This is a multicenter, prospective cohort study. We enrolled (1) HIV-infected adolescents (aged 10-25 years) who received antiretroviral treatment (ART) for ≥1 year, and (2) age- and sex-matched healthy controls (ratio 2:1). HIV-infected adolescents were further classified according to their current status of tenofovir disoproxil fumarate use (TDF vs. non-TDF users). Glomerular function parameters, including serum creatinine, and random urine creatinine and protein levels were measured at baseline and 3 months. Estimated glomerular filtration rate (eGFR) and urine protein-to-creatinine ratio (UPCR) were calculated. Glomerular dysfunction was defined as having either (1) impaired eGFR: eGFR<60mL/min/1.73m2 for 2 times, or decline ≥30% from baseline; or (2) proteinuria: UPCR ≥0.2mg/mg for 2 times. Median values are provided with interquartile ranges. Logistic regression analysis was performed to identify associated factors of glomerular dysfunction among our HIV-infected adolescents. Results: Between December 2016 and June 2017, 140 HIV-infected adolescents and 70 healthy controls were enrolled, half (50%) were female. The median age and body mass index (BMI) were 18 (15-21) years, and 20 (18-22) kg/m2, respectively. Thirty-seven adolescents (18%) were wasted (BMI<5th percentile or <18.5kg/m2 for participants aged<18 or ≥18 years, respectively). Among HIV-infected adolescents, 70 (50%) were TDF users, of whom 35 (50%) were concurrently receiving protease inhibitor (PI)-based regimens. Glomerular dysfunction was identified in 15 TDF users (21%; 95%CI: 13-33%), 6 non-TDF users (9%; 95%CI: 3-18%), and 1 healthy control (1%: 95%CI: 0.1-8%) (P<0.01) (Table1). Among TDF users, the prevalence of glomerular dysfunction was not different between individuals on PI-based vs. non-nucleoside reverse transcriptase inhibitor-based regimens (8 [23%] vs. 7 [20%]; P=0.77). In the multivariable analysis among HIV-infected adolescents, TDF use (adjusted odds ratio[aOR]:4.2; 95%CI: 1.2-15.2), age <18 years (aOR:11.1; 95%CI: 2.3-53.3), and wasting (aOR: 7.2: 95%CI: 1.9-27.3) were significantly associated with glomerular dysfunction.

CROI 2018 321