A specialized hepatic encephalopathy testing clinic improves rational decision making for he therapy and can detect alternative causes for cognitive impairment in cirrhosis

Background: Cognitive impairment in cirrhosis is commonly presumed to be hepatic encephalopathy (HE) & is reflexively treated (Figure 1A) with either lactulose (difficult to tolerate) or rifaximin (expensive). However, many patients symptoms may not be HE related. Most patients are not routinely tested for minimal HE (MHE) and on-demand MHE testing/interpretation in clinical settings needs to be studied. Methods: An on-demand standard of care HE testing clinic (Figure 1A) was established, separate from original clinic visit for cognitive issues. Mini-mental status exam (MMSE out of 30, >25=no dementia), psychometric hepatic encephalopathy score (PHES) & EncephalApp Stroop was performed by a trained medical assistant and interpreted by a hepatologist. Chart/medication review, recommendations and time spent (testing/interpretation/medical decisions) were recorded. Results: There were 282 patients were evaluated from 2012-2022, majority (84%) due to cognitive complaints by pts/families. Four had MMSE<25, which were referred for dementia evaluation without further tests. No-MHE: 111 (39%) had normal cognitive performance (Figure 1B). These patients (Table 1) were younger, with less prior HE, depression, ascites, and lower MELD-Na vs who tested impaired. Most (N=84) were reassured & did not need lactulose; remaining were referred to other specialties if requested. Cognitively impaired: 47(17%) continued current therapy, either no therapy (refused therapy) or same HE regimen, 56 (20%) were initiated on lactulose, & 27 (10%) on rifaximin. 37 patients had major contributor(s) unrelated to cirrhosis for their cognitive impairment, such as pain medications, obstructive sleep apnea, dementia, or neuromodulator therapy and were referred to primary care doctors, neurologists, or pain management. Time needed: Medical assistant took 34±12 min/pt. Hepatologist took 12±5 min to interpret & complete the recommendations, which were billed for. Conclusion: A dedicated US-based HE testing clinic run by a trained medical assistant and supervised by an attending reduced reflexive HE therapy initiation in majority of patients. With specialized testing that <40 minutes to perform & interpret, and was billable, 39% patients showed normal cognition & were spared reflexive lactulose. Fourteen percent of needed referral for other neurocognitive issues & only 30% needed HE therapy change, or initiation. Dedicated HE testing clinics may be effective in streamlining HE management.Figure 1.: A: Study Flow. B: Results of Specialized Testing. Table 1. - Comparison of cognitive impairment vs no cognitive impairment Cognitive Impairment (n=172) No Cognitive Impairment (n=110) P Value Age 60.7±6.9 57±8.9 0.005 Male Gender 165 105 0.95 Caucasian Race 106 67 0.83 Etiology (Alcohol/HCV/NASH/Mixed/Others) 41/51/25/44/11 20/47/15/19/10 0.15 Prior ascites 69 26 0.003 Prior HE 66 8 <0.0001 History of varices 62 30 0.008 Prior SBP 14 0 <0.0001 Anxiety 17 8 0.44 Depression 48 19 0.03 Chronic pain 41 29 0.65 Non-selective beta-blockers 69 29 0.65 Labs Platelet count 125±71.2 149±74.6 0.006 Creatinine 1.09±0.43 1.05±0.32 0.42 INR 1.34±0.41 1.29±0.60 0.38 Bilirubin 2.04±2.76 1.32±1.64 0.007 Sodium 137.9±3.8 138.6±2.8 0.09 Albumin 3.22±0.70 3.48±0.68 0.002 MELD-Na 13.51±6.31 10.92±3.79 0.03 Dedicated testing MMSE (of 30) 28.33±1.71 29.08±0.99 <0.0001 PHES (5 to -15, high=good) -6.65±3.95 -1.08±0.99 <0.0001 Stroop Off+On Time (low=good) 228.3±88.5 175.7±42.6 <0.0001