CROI 2017 Abstract e-Book

Abstract eBook

Poster and Themed Discussion Abstracts

1 Univ of Pennsylvania, Philadelphia, PA, USA, 2 Boston Univ, Boston, MA, USA, 3 Vanderbilt Univ, Nashville, TN, USA, 4 Harbor–UCLA Med Cntr, Torrance, CA, USA, 5 Yale Univ, New Haven, CT, USA, 6 Univ of Pittsburgh, Pittsburgh, PA, USA, 7 Baylor Coll of Med, Houston, TX, USA, 8 VA Connecticut Hlthcare System, West Haven, CT, USA, 9 Philadelphia FIGHT Community Hlth Cntrs, Philadelphia, PA, USA Background: Hepatic steatosis is highly prevalent in Western countries, but its association with HIV infection is poorly defined. Factors associated with hepatic steatosis have not been completely examined in HIV. We compared the prevalence of hepatic steatosis between HIV+ and uninfected persons and evaluated risk factors for steatosis among persons with HIV. Methods: We performed a cross-sectional study among 171 HIV+ and 97 uninfected patients with no prior cardiovascular disease diagnosis in the Veterans Aging Cohort Study. Using non-contrast abdominal computed tomography (CT), hepatic steatosis was defined by: 1) liver-to-spleen (L/S) attenuation ratio <1 (indicates hepatic steatosis affecting >5% of liver parenchyma), and 2) liver attenuation of <40 Hounsfield Units (HU; indicates steatosis affecting ≥30% of liver parenchyma). Logistic regression was used to determine adjusted odds ratios (ORs) of steatosis associated with HIV, controlling for body mass index (BMI), diabetes, and history of alcohol abuse. Among HIV+ persons, linear regression was performed to evaluate associations between risk factors of interest (nadir CD4 count, HIV viremia, hepatitis C virus [HCV] coinfection, and hepatic fibrosis by FIB-4) and severity of steatosis by liver attenuation (in HU), after adjustment for BMI, alcohol abuse, and diabetes. Results: The overall prevalence of hepatic steatosis was similar between HIV+ and uninfected persons when defined by L/S attenuation ratio <1 (13 [7.6%] versus 8 [8.2%]; p=0.85); however, steatosis defined by liver attenuation <40 HU was less prevalent in HIV+ patients (13 [7.6%] versus 15 [15.5%]; p=0.04). After adjustment for BMI, diabetes, and alcohol abuse, HIV was not associated with hepatic steatosis (defined by L/S attenuation ratio <1) compared to uninfected individuals (OR, 1.00; 95% CI, 0.40-2.60). Among the HIV+ group, higher BMI (p=0.01), HCV coinfection (p=0.008), and higher FIB-4 (p<0.001), but not lower nadir CD4 count (p=0.14) or HIV viremia (p=0.87), were associated with greater severity of steatosis. Conclusion: The prevalence of hepatic steatosis, defined by abdominal CT, was similar between HIV+ and uninfected persons, though the severity of steatosis was greater among uninfected persons. HCV coinfection, higher BMI, and more advanced hepatic fibrosis were associated with greater severity of steatosis in HIV+ persons. Future studies should compare the impact of steatosis on liver outcomes among HIV+ patients. 703 LIVER STEATOSIS AND FIBROSIS IN AT-RISK EUROPEAN HIV-MONOINFECTED PATIENTS Maud Lemoine 1 , Lambert Assoumou 2 , Patrick Ingiliz 3 , Pierre-Marie Girard 4 , Stephane de Wit 5 , Marc-Antoine Valantin 6 , Anja Huefner 7 , Jacqueline Capeau 8 , Dominique Costagliola 9 , Georg Behrens 10 1 Imperial Coll London, London, UK, 2 Inst Pierre Louis d’Epidémiologie et de Santé Publique, Paris, France, 3 Cntr for Infectiology, Berlin, Germany, 4 Hôpital Saint-Antoine, Paris, France, 5 Univ Libre de Bruxelles, Brussels, Belgium, 6 AP–HP, Paris, France, 7 Universitätsklinikum Hamburg, Hamburg, Germany, 8 Univ Pierre & Marie Curie, Paris, France, 9 Medizinische Hochschule Hannover, Hannover, Germany Background: Nonalcoholic fatty liver disease (NAFLD) has emerged as a new concern in HIV-infected patients. The ECHAM (European Cohort on HIV, Ageing and Metabolic liver disease) Study Group aimed to assess the prevalence of NAFLD and its complications i.e nonalcoholic steato-hepatitis (NASH), fibrosis and cirrhosis in at-risk HIV-monoinfected ART-treated individuals. Methods: This cross-sectional study conducted in seven European centers enrolled HIV-infected individuals with persistently elevated transaminases (>1.5 ULN) and/or metabolic syndrome (MS), and/or lipodystrophy without other causes of liver disease (i.e HCV or HBV coinfections or excessive alcohol intake). All patients underwent complete non-invasive metabolic and liver assessments including hepatic MRI, Fibroscan®/CAP and Fibrotest®. A liver biopsy was indicated in case of suspected significant fibrosis (≥F2) based on Fibroscan® (>7kPa) and/or Fibrotest® (≥ 0.49). Results: Between March 2014 and November 2015, 461 individuals were screened, 442 met the inclusion criteria and 402 had full liver assessment and were further analyzed. Patients were mainly males (85%) with a median age of 55 years (IQR 50-61). The median time since HIV diagnosis was 19 years (14-24), ART duration 16 years (12-19). Nadir of CD4 count was 184 (84 – 266)/mm3. HIV viral load was<50cp/mL in 97% of cases with a median CD4 count of 630/mm3 (510-832). Of the 402 patients, 269 (67%) had a MS and 218 (54%) insulin resistance defined by HOMA index ≥2.5. Median ALT, AST and GGT levels were 34 (24-50), 29 (23-37) and 48 (29-81) IU/L, respectively. Hepatic MRI identified 257 (64%) patients with significant steatosis defined by a fat fraction >5%. Using non-invasive markers of fibrosis, 140 (34%) were classified as suspected significant liver fibrosis including 12.5%with cirrhosis. However, the concordance between Fibroscan® and Fibrotest® for the diagnosis of fibrosis was poor (kappa coefficient 13%). Of 140 patients eligible for liver biopsy, 49 accepted the procedure; their demographic and clinical characteristics were similar to those who refused biopsy. Histological analysis reported steatosis in 76%, NASH in 43%, significant fibrosis (≥F2) in 30% and cirrhosis in 2 (4%) patients. Conclusion: Nonalcoholic HIV-monoinfected patients on ART with metabolic disorders are at high risk of liver steatosis and fibrosis. However, non-invasive markers of fibrosis should be interpreted with caution in this population. (Study registered on clinicialtrial.org, NCT02093754) 704 CHANGES IN LIVER FIBROSIS AND STEATOSIS IN HIV MONOINFECTED PATIENTS OVER 24 MONTHS Raphael Mohr , Christoph Boesecke, Carolynne Schwarze-Zander, Jan-Christian Wasmuth, Jonel Trebicka, Jürgen K. Rockstroh Univ Hosp Bonn, Bonn, Germany Background: While cross-sectional studies recently have identified factors associated with hepatic fibrosis and steatosis in HIV mono-infected patients, scare data are available on their evolution. The aim of this study was to assess the changes in liver stiffness and hepatic fat accumulation over time and to evaluate possible predictive factors for hepatic fibrosis and/or steatosis development. Methods: 344 HIV mono-infected patients underwent annual transient elastography (TE) with simultaneous determination of the controlled attenuation parameter (CAP) over a period of 24 months. The associations between LS/HS changes and demographics, metabolic data, virologic factors and antiretroviral therapy were analyzed. Results: The prevalence of significant liver fibrosis (TE > 7.1 kPa) did not change during the study, being 10% (median LS 5.2 kPa) at enrolment and 12% (median LS 5.4 kPa) after 24 months (p = 0.868). The only factor associated with an increase of LS was a lower CD4+ level (p = 0.037). Moreover, longer cART duration protected (HR, 0.777; 95% CI, 0.712-0.847), whereas persistent HIV viral replication promoted the development of significant liver fibrosis (HR 2.430; 95% CI, 1.095-5.396). At baseline 41% of the patients had significant steatosis (CAP > 238 dB/m) vs. 54% of the patients after 24 months (p = 0.023). Over the follow-up a significant increase in overall CAP values was observed (median CAP 226 dB/m vs. 244 dB/m; p = 0.002). Factors associated with increased CAP values were higher HbA1c levels (p = 0.015), higher BMI (p < 0.001), and higher triglycerides (p = 0.012). Longer cART-naïve periods were associated with less hepatic steatosis progression (HR 0.884; 95% CI, 0.832-0.940). Importantly, antiretroviral drugs had no impact on severity of steatosis and/or fibrosis. Conclusion: Even though prevalence of liver fibrosis in HIV patients did not change over time, our data suggest that the HI virus might directly influence hepatic fibrosis and control of HIV replication blunts fibrosis. By contrast, prevalence of steatosis significantly increased during the follow-up of 24 months. Hepatic steatosis might be triggered by metabolic factors and should be a target of treatment in the future. 705 CO-TRIMOXAZOLE PROPHYLAXIS DECREASES TUBERCULOSIS RISK AMONG ASIAN PATIENTS WITH HIV Wen-Wei Ku 1 , Awachana Jiamsakul 2 , Kedar Joshi 3 , Mark Kristoffer Ungos Pasayan 4 , Alvina Widhani 5 , Romanee Chaiwarith 6 , Sasisopin Kiertiburanakul 7 , Kiat Ruxrungtham 8 , Ly Penh Sun 9 , Wingwai Wong 1

Poster and Themed Discussion Abstracts

CROI 2017 308