Covert Hepatic Encephalopathy Diagnosis With The New Stroop

Covert Hepatic Encephalopathy Diagnosis With The New Stroop App (Encephal. App) Using Age-Based Cut-off Values Jasmohan S. 1 Bajaj , Douglas M. 1 Heuman , Leroy Richard K. R. Todd Stravitz Arun J. Velimir A. Nicole Melanie White 1, Ariel Unser 1, Pamela Monteith 1, Puneet Puri 1, James Wade 2 1 Gastroenterology, Hepatology and Nutrition, 2 Psychiatry, 3 Biostatistics Virginia Commonwealth University and Mc. Guire VAMC, Richmond, BACKGROUND • Diagnosis of covert/minimal hepatic encephalopathy (CHE) is difficult therefore a high sensitivity/specificity screening test is required. • The older Stroop App (downloaded from i. Tunes) has been validated for diagnosis of CHE but it needed updates to make it CHE-specific • Therefore the new Encephal. App_Stroop was created to: • Streamline timing: removes unnecessary time wastage between stimuli and therefore to reduces total testing time), • Improves and systematizes result transmission: results are emailed in the form of Excel sheets automatically in order to prevent copying errors. • Added an i. Pad-ready version • Further population norms were also required for this app AIM To validate Encephal. App for the CHE diagnosis using agebased cut-offs and to evaluate the i. Pad version performance compared to the i. Pod/i. Phone version. METHODS • Cirrhotics and healthy controls underwent cognitive testing using standard batteries (Number connection-A/B, (NC-A/B), digit symbol (DS), block design (BD) • Performance on ≥ 2 tests impaired two standard deviations beyond controls was considered CHE • Subjects were then tested on the new App. • The App has two parts; • Stroop Off in which the subject needs to press the correct color in which # signs are presented (Figure 1 A). • Stroop On in which the subject needs press the correct color in which words meaning red, green or blue are presented in discordant colors i. e. if “red” is presented in green color, the correct answer is green (Figure 1 B). 1 A 3 Thacker , 1 Sterling , 1 B METHODS • Time to complete 5 Off runs(Off. Time) and On(On. Time) were recorded; a lower time indicates good performance. • Number of trials to achieve 5 correct runs was also recorded. • Off. Time+On. Time >2 standard deviations above controls was considered impaired on Encephal. App. • Sensitivity and specificity were performed for Encephal. App using standard tests are gold standard • A group of subjects then underwent the Encephal. App on i. Pad or i. Pod in random order and their results were compared. RESULTS • 100 controls (age 43± 14 yrs, education 15± 2 yrs, 50% >45 yrs age) and 132 cirrhotics (MELD 12± 5, years of education 14± 2) were included. • 37% of cirrhotics had prior overt HE currently controlled on rifaximin or lactulose at time of testing. • All patients underwent Encephal. App & standard testing at the same sitting. • Controls performed significantly better than cirrhotics on • Standard tests (NC-A 23 vs 43 s, NC-B 59 vs 114 s, DS 78 vs 53, BD 42 vs 22, all p<0. 0001) and • Encephal. App Off. Time+On. Time (132 vs 206 s p<0. 0001). • We found evidence of CHE in 40% of pts on standard tests. 1 Sanyal , 1 Luketic , 1 Noble , RESULTS • Off. Time+On. Time was correlated with age (r=0. 6, p<0. 0001) but not education in controls, therefore their App results were analyzed separately for those > or <45 yrs. • Off. Time+On. Time(±SD) in controls <45 yrs was 119± 13 sec & 146± 22 sec in those>45 yrs. Therefore 2 standard deviations impaired beyond controls was >145 sec in pts <45 yrs &>190 sec in >45 yrs on Off. Time+On. Time. • Cirrhosis group: Off. Time+On. Time was significantly higher in CHE cirrhotic patients compared to those without regardless of whether those with prior overt HE were included or not (Figures 2 A and B) and was correlated with MELD (r=0. 5, p=0. 004) but not age/education. Number of trials to achieve 5 correct Off (5 vs 6) or On (5 vs 6) were not different between CHE/no CHE. • CHE diagnosis: Using the age-adjusted cutoffs, 52% of patients had impaired App performance. CHE was more in those with prior OHE (94% vs no OHE 30%, p=0. 0001). Using standard tests as reference sensitivity of these age-variable cut-offs was 90% while specificity was 78%. 2 A 2 B i. Pad vs i. Pod/i. Phone: 17 subjects (8 controls and 9 cirrhotics) underwent sequential testing with these two modes of administration in random order (9 had i. Pad first and 8 had i. Pod first). There were no differences in Encephal. App results and the Off. Time+On. Time were significantly correlated (Figure 3) R=0. 94, p<0. 0001 3 CONCLUSIONS: The new age-based cut-offs for Encephal. App_Stroop are highly sensitive for covert HE screening and could be used to guide future testing in cirrhosis regardless of mode of administration (i. Pad or i. Phone/i. Pod)