P: 70 Quality of Life Assessment May Aid in the Diagnosis of Minimal Hepatic Encephalopathy and Prediction of Overt Hepatic Encephalopathy

Abstract

BACKGROUND: Minimal hepatic encephalopathy (MHE) impinges on quality of life (QoL), is associated with a high risk of overt hepatic encephalopathy (OHE) and is often treatable by simple means. Still, MHE is rarely systematically diagnosed and treated likely because dedicated psychometric tests give the impression of being resource heavy and thus scare off many clinicians. Simple, patient-administered QoL questionnaires e.g. sickness impact profile (SIP), could improve diagnostic rates. This approach was tested in a US-based study introducing SIPCHE score (formula of 4 SIP statements, gender and age). We here aim at externally validate SIPCHE score in a cirrhotic cohort using continuous reaction time (CRT) test and portosystemic hepatic encephalopathy score (PHES) for MHE diagnosis. METHODS: 110 cirrhotic patients without OHE (age 60 years, MELD 11.4, 80% blue-collar) completed cognitive testing and SIP. Abnormal CRT and/or PHES diagnosed MHE. SIP consists of 136 questions inquiring about QoL and standardized QoL scores were compared in MHE and non-MHE patients. The SIPCHE (US cut point >0) was applied and predictive values were calculated. We followed the patients for 2.7 years on average and registered OHE episodes. RESULTS: The SIPCHE was abnormal in 82/110 patients and was in agreement with the psychometric tests in 73/110 cases (66%). The SIPCHE indicated MHE in 58/71 of the patients with MHE according to 2-test psychometry (positive predictive value = 71%, sensitivity = 82%, AUUROC 0.63). The SIPCHE was false positive in 24/39 non-MHE patients (specificity 38%, NPV 53%). A normal SIPCHE did not exclude MHE in our population as 13/28 (46%) with a normal SIPCHE score had MHE according to the CRT and PSE tests. In our cohort using a cut point of >−0.40, in stead of >0, sligthly improved correct classification to 72% of patients. Only 4/28 (14%) with a normal baseline SIPCHE experienced OHE, while 29/82 (35%) with abnormal SIPCHE experienced OHE (P = 0.05). Accordingly, the SIPCHE positive predictive value for a future HE episode is this 87% (likelihood ratio 2.4). CONCLUSIONS: In conclusion, the idea of a patient-reported outcome score as an addition to standard psychometry is appealing. The US derived SIPCHE score is able to identify the majority of patients with MHE and future OHE episodes, but lacks diagnostic specificity. We suggest that development of regional SIPCHE scores could be useful. In future studies using the SIPCHE as a measure of patients-experienced effect of MHE treatment would be of interest.