P: 3 Clinical Value of Asterixis in a Large Population of Well-characterised Patients With Cirrhosis and Varying Degree of Hepatic Encephalopathy

Abstract

BACKGROUND: Current guidelines for the diagnosis of overt hepatic encephalopathy (HE) include the clinical sign asterixis, which has also been used as an outcome in clinical trials. Limited information is available on the ability of physicians to detect/grade asterixis, and the relationship between asterixis and other HE indices. The aim of the study was to retrospectively assess the clinical value of asterixis in a large population of well-characterised patients with varying degree of HE. METHODS: Asterixis was sought for/graded in 374 consecutive patients with cirrhosis (59 ± 8 yrs, 280 males; MELD 14 ± 5, Pugh 7.9 ± 2.0) by trainees in Internal Medicine (n = 16), attending the Padova HE clinic between 2011 and 2019. Patients were asked to stretch their arms, extend their wrist and spread their fingers, and observed for 1 min. Asterixis was qualified as absent, rare, frequent or continuous. All underwent neuropsychiatric examination as per local protocols, including electroencephalography (EEG) and a set of neuropsychological tests [Animal Naming Test (ANT); Psychometric Hepatic Encephalopathy Score (PHES); computerised simple (sRT), choice (cRT) and Scan (ScanRT) reaction times]. Laboratory indices (venous ammonia, CRP, sodium) were recorded, together with presence/absence of previous HE episodes and spontaneous/surgical portal-systemic shunts. Patients were followed up for 24 months in relation to the development of HE-related hospitalizations. RESULTS: 280/59/33/2 patients were qualified as having no/rare/frequent/continuous asterixis, thus those with frequent and continuous asterixis were grouped for purposes of subsequent analyses. Associations were observed between the presence/degree of asterixis and the presence of overt HE on the day of study (χ2 = 55, P < 0.001), a history of HE (χ2 = 21, P < 0.001), and the presence of portal-systemic shunts (χ2 = 11.8, P = 0.019). Significant differences in neuropsychiatric HE indices were observed between patients with different degrees of asterixis. Some were prominent between patients with/without the sign, regardless of its frequency (PHES, ScanRT), while for others the relationship was linear (ANT, sRT, cRT, spectral EEG parameters). Significant differences in laboratory indices of liver failure/HE were also observed between patients with different degrees of asterixis. Again, some were prominent between patients with/without the sign (Pugh, CRP), while for others the relationship was linear (MELD, ammonia). Finally, the likelihood of developing HE-related hospitalisations over the follow up period was significantly higher in patients with asterixis (Cox-Mantel P = 0.028). CONCLUSIONS: Asterixis is reliably detected/graded by specialist trainees in a tertiary referral liver centre, and shows significant associations with established neuropsychological, neurophysiological and laboratory HE indices, as well as the likelihood of developing HE-related hospitalisations over time.