CROI 2018 Abstract eBook

Abstract eBook

Poster Abstracts

647 TRANSPLANT FREE SURVIVAL IN HIV ASSOCIATED NON-CIRRHOTIC PORTAL HYPERTENSION Kathryn Childs 1 , Lauren Hookham 1 , Corinna Psomadakis 2 , Rosa Miquel 1 , Ruth Byrne 1 , Mary Cannon 1 , Kosh Agarwal 1 , Mark Nelson 2 , Chris Taylor 1 , Abid Suddle 1 1 King’s College Hospital, London, UK, 2 Chelsea and Westminster Hospital, London, UK Background: In the last decade there have been numerous reports of patients with HIV presenting with non-cirrhotic portal hypertension (NCPH). We aimed to describe the natural history and clinical outcome in patients from 2 centres who together represent the largest published cohort. Methods: This is an observational cohort study. Demographics, laboratory, radiological and histological data were obtained on patients at 2 London centres with non-cirrhotic portal hypertension. This was defined by the presence of portal hypertension and either a biopsy or Fibroscan which excluded cirrhosis AND the absence of any other form of liver disease. Results: 44 patients were identified. From the time of diagnosis of NCPH the median follow up (FU) time was 6.6 yrs (3.3, 8.5) with a total FU period of 310 patient yrs. Characteristics shown in table 1. All patients had been exposed to didanosine (DDI). 28 patients underwent biopsy, 10 showed features of nodular regenerative hyperplasia or portal obliterative venopathy but many showed non-specific inflammation or steatosis. No patient had greater than F3 (Ishak) fibrosis, 24/28 were F0-F2. 27/44 patients developed portal vein thrombosis, 16 were anticoagulated. During 9.5 patient years of FU under anticoagulation, no major bleeding events occurred. At 5 years of NCPH follow up, 5 patients had died (3 liver cause), 1 had undergone orthoptic liver transplantation (OLT). After 10 years of FU, 7 had died (4 liver death), 5 had undergone OLT of whom 1 died. This gives a 5 year all-cause death and transplant free survival rate of 86% and a 10 year survival rate of 73%. 5 year liver death and transplant free survival rate of 91% and a 10 year survival rate of 79.6% 5 patients were transplanted. The indication for OLT was encephalopathy in 3 cases, recurrent ascites in 1 case and synthetic failure in 1 case. In the last case, the diagnosis of NRH was made on histology of the explanted liver. 1 patient died shortly post-transplant of gut ischaemia related to superior mesenteric vein thrombosis. Total follow up post OLT is 8.6 years. Conclusion: We report a large cohort with a long duration of follow up for HIV associated NCPH. This is a serious condition with 27% of patients having died or undergone transplantation at 10 years. Anticoagulation was safe in this group. Many presented with subtle derangements of liver function, it is vital that HIV care providers maintain a high degree of clinical suspicion so patients can receive appropriate investigation and management.

HIV+ than HIV- patients (5.8% vs 1.9%, p<0.05). In HIV+ patients, the Baveno VI guidelines had a sensitivity 0.82, specificity 0.62, positive predictive value 0.26, negative predictive value 0.95 for the diagnosis of esophageal varices as compared to universal EGD, which were similar to HIV- patients. In multivariable analysis, after adjustment for age, gender, BMI and anti-HCV positivity, being HIV+ was the strongest factor associated with failing to screen when indicated by Baveno VI guidelines (aOR=10.0, 95% CI 7.5-13.5; p<0.0001). Conclusion: Despite the Baveno VI guidelines performing well in HIV+ patients, they are significantly less likely to receive standard of care screening for esophageal varices than HIV- patients, placing them at higher risk of fatal complications from hemorrhage. 646 MEDIUM-TERM EFFECTS OF DAA THERAPY ON HVPG IN PATIENTS WITH HCV-ASSOCIATED CIRRHOSIS Cristina Diez 1 , Juan Berenguer 1 , Luis Ibañez 1 , Elba L. Herrera 2 , Leire Pérez Latorre 1 , María V. Catalina 1 , Victor Hontanon 3 , Diego Rincón 1 , Teresa Aldámiz- Echevarría 1 , Javier Martínez 4 , José M. Bellón 1 , José L. Calleja 2 , Agustín Albillos 4 , Juan González-García 3 , Rafael Bañares 1 1 General University Hospital Gregorio Marañón, Madrid, Spain, 2 Hospital Puerta de Hierro, Madrid, Spain, 3 La Paz University Hospital, Madrid, Spain, 4 Hospital Ramon y Cajal, Madrid, Spain Background: In patients with liver cirrhosis (LC), hepatic venous pressure gradient (HVPG) is the most accurate predictor of liver-related outcomes. Little is known about the effects of therapy (Rx) with DAA on HVPG in patients with HCV-related LC. We assessed changes in HVPG following SVR after DAA -Rx in HCV- monoinfected patients (HIV-) and HIV/HCV- coinfected patients (HIV+) with LC. Methods: Prospective study (4 centers) with patients initiating all-oral DAA Rx between during Jan-Dec 2015. HVPG was measured at baseline and 48 wk after completion of Rx. Inclusion criteria: i) Compensated LC (c-LC) or decompensated LC (d-LC) defined as prior clinical decompensation or Child-Pugh score > 6; ii) clinically significant portal hypertension (CSPH) defined as an HVPG ≥ 10 mmHg; iii) achievement of SVR; iv) no Rx with non-selective betaß-blockers initiated during the study period. Main endpoint: reduction in HVPG to < 10 mmHg. Secondary outcome: decrease in HVPG ≥ 20% or reduction in HVPG to ≤ 12 mm Hg (goal associated with reduced liver complications in patients with variceal bleeding undergoing pharmacological treatment of portal hypertension). Results: Of 44 patients with LC and paired HVPG measurements, 34 met the 4 inclusion criteria. Main characteristics: median age 53 yr; 22 males, 21 HIV+, 18 previously treated, 17 with d-LC; median MELD score 8; median liver stiffness (LS) 34 kPa. HVPG decreased from 16.5 (IQR 14.4-18.0) at baseline to 14.0 (IQR 10.0 -– 15.0) mmHg at wk 48; (P<.001) (Figure); with a median decrease of 3.1 (IQR 2.0-4.9) mm Hg. No significant differences in the decrease in HVPG decline were observed between HIV- and HIV+. The main outcome endpoint was achieved by 6 patients (18%) and was more frequent in c-LC than in d-LC (35.3 % vs 0 %; P=0.018). The secondary endpoint was achieved by 16 patients (47%); 10 with c-LC (58.8%) and 6 with d-LC (35.3%); P=.303. Of note, HVPG increased from baseline in 4 patients (11.8%). The spearman rho correlation coefficient between changes in HVPG and changes in LS was 0.237, P=.191. Conclusion: Our findings suggest that, in the medium term, SVR after DAA Rx in patients with LC and CSPH is associated with a decrease in HVPG that is sufficient to reduce the risk of liver complications. However, the frequent persistence of CSPH despite SVR, especially in patients with more advanced disease, indicates a persistent risk of decompensation. The correlation between change in LS and reduction in HVPG was very weak.

Poster Abstracts

CROI 2018 240