192 THE LB MODEL: A NEW PROGNOSTIC TOOL FOR PREDICTION OF SHORT TERM SURVIVAL IN PATIENTS WITH END STAGE LIVER DISEASE

Abstract

Results: At listing, no significant differences were noted between the 2 groups (HIV+ vs HIV−) for sex ratio, viral etiology, number of nodule(s) (1.6±0.9 vs 2.3±2.4, p = 0.8), maximal diameter of nodule(s) (27±8mm vs 25±12mm, p = 0.9), and AFP level. Rate of patients in each group that exceeded Milan criteria were similar (2/20 (10%) vs 11/61 (18%)). HIV+ patients were younger (49±5 years vs 56±5 years, p = 0.001). The rate of patients treated by chemoembolization (mean number of cure: 1.5±0.5 vs 1.4±1, p = 0.86) was similar but the rate of radiofrequency was higher in HIV+ patients (8/20 (40%) vs 11/61 (18%), p = 0.04). Drop-out (DO) of HIV+ patients were significantly higher (7/20 (35%) vs 6/61 (9%), p = 0.007). Tumoral progression (n = 5) and HIV evolution (n = 2) were responsible of DO in HIV+. With a similar waiting-time (7±8 months vs 4±5 months, p = 0.13), 13 HIV+ and 55 HIV− were transplanted with one post-operative death in each group. On the explanted liver, no significant differences were noted concerning the number of nodule(s) (2.3±1.6 vs 2.6±3.2, p = 0.70), with a maximal diameter of (27±13mm vs 28±14mm, p = 0.82), presence of satellite nodules (4/13 vs 21/49, p = 0.39), vascular invasion (6/13 vs 22/48, p = 0.98) but Edmonson grade was higher in HIV+ (3.1±1.1 vs 2.5±1.1, p = 0.04). After a mean follow-up of 16±18 months and 24±16 months for HIV+ and HIV− patients (p = 0.11), respectively, a tumoral recurrence was noted in 4/13 (30%) HIV+ patients at 2, 3, 11 and 37 months vs 2/49 (4%) in HIV− patients at 20 and 28 months after LT (p< 0.001). Conclusion: In our series, because of high drop-out (35%) and recurrence after LT (30%) in HIV+ patients, a better selection in this kind indication must be performed.