P: 19 Predictive Value of Induced Hyperammonaemia and Neuropsychiatric Profiling in Relation to the Occurrence of Post-tips Hepatic Encephalopathy (HE)

Abstract

Listen

Translate article

BACKGROUND: Hepatic encephalopathy (HE) occurs in 20%–50% of patients after transjugular intrahepatic portosystemic shunt (TIPS). While indices such as an older age, a history of overt HE and severe liver failure have been associated with post-TIPS HE, it remains difficult to identify patients at risk. The aim of this study was to evaluate the predictive value of a large set of clinical, laboratory and neuropsychiatric parameters, both at baseline and after the induction of hyperammonaemia, in relation to the development of post-TIPS HE for 12 months in a group of well-characterised TIPS candidates. METHODS: 18 TIPS candidates (58 ± 8 yrs, MELD 11 ± 3; 13 refractory ascites, 1 hydrothorax, 4 uncontrolled bleeding) underwent neurophysiological [Electroencephalography (EEG)], neuropsychological [Psychometric Hepatic Encephalopathy Score (PHES) and Scan test], capillary ammonia and subjective sleepiness assessment both at baseline and after the induction of hyperammonaemia by an oral amino acid challenge (AAC). RESULTS: At baseline, 3 (17%) patients had abnormal EEG, 1 (5%) abnormal PHES, and 6 (33%) abnormal Scan performance; fasting capillary ammonia concentrations were 199 ± 88 μg/dL and subjective sleepiness 2.5 ± 1.2 (1–9 scale). Post-AAC, 3 (17%) patients had abnormal EEG, none abnormal PHES, and 3 (17%) abnormal Scan performance. Post-AAC, patients exhibited the expected increase in ammonia [F(4, 56) = 2.7213, P = 0.038] and subjective sleepiness [F(4, 52) = 5.4002, P = 0.001]. Six months after TIPS, 3 patients had had an HE-related hospitalization. Compared to their counterparts who had not, they showed significantly lower, pre-AAC fasting ammonia concentrations (96 ± 93 vs. 225 ± 58 μg/dL, P = 0.017). They also showed worse PHES and Scan performance at baseline, as well as worse Scan performance and slower EEG post-AAC. Twelve months post-TIPS, 5 patients had had an HE-related hospitalization, and comparisons between the two groups confirmed the findings at 6 months. CONCLUSIONS: TIPS candidates, who are by definition at low risk of HE, seem to encompass two different populations: i) those who are well and have not had overt HE because their ammonia levels are near-normal, ii) those who are well and have not had overt HE despite hyperammonaemia, either because of habituation or a personal inclination not to exhibit the HE phenotype. Fasting ammonia may be a promising and easily obtained parameter for future validation and inclusion in models for the prediction of post-TIPS HE.