LAUSR

Background and aim The management of patients with decompensated liver disease is multifaceted. Hepatologists and Allied Health Care professionals are involved in treating the medical complications. Care often involves discussions about prognosis given the associated mortality. A validated prognostic screening tool was developed by Hudson et al to identify those at high risk of death given the unpredictable trajectory in cirrhosis. Having adopted this tool, we reviewed the outcomes of patients discussed at our weekly decompensated liver disease multidisciplinary (MDT). Methods We prospectively collected data on admissions to the Hepatology ward between September 2019 and March 2020. Outcomes were defined by the key performance indicators from Improving Quality in Liver Services (IQILS) and the presence of the 5 evidence-based factors associated with poor prognosis. These include 1) Childs-Pugh C 2) more than 2 liver-related admissions in 6 months 3) current alcohol consumption 4) Unsuitable for transplant work up 5) WHO performance status 3–4. Those with over 2 factors have a poorer prognosis so this triggered discussions around ceiling of treatment and end of life care. Results During this period, 55 individual patients were admitted to the hepatology ward with decompensated liver disease 44 (80%) had alcohol related liver disease (ArLD) and were referred to the alcohol liaison service. Twenty-seven (49%) patients died within 6 months of admission, 14 (52%) were male, mean age 60 years and 20 (74%) had ArLD. Seventeen (63%) had ascites with a median MELD score of 21. The median number of days from admission to death was 35 (3–256). Twenty-one (78%) of those who died had over 2 factors associated with poor prognosis compared with 12/28 (43%) who survived. Nineteen (70%) had Childs Pugh C cirrhosis, 15 (56%) with over 2 admissions within 6 months, 17 (63%) had current alcohol consumption, 26 (96%) were deemed unsuitable for a liver transplant and 13 (48%) had performance status 3 or 4. Among the 27 patients who died, there were 38 (mean 1.4) readmissions before death. 10 paracentesis were performed in 4 patients in the Ambulatory Care Unit, 4 Palliative Care referrals were made and 5 died at home. Conclusions This data confirms the utility of the prognostic tool in identifying patients at high risk of death. Utilisation of such prognostic models can change the focus of patient care particularly around ceilings of treatment, access to ambulatory facilities, decisions around end of life care and involvement of palliative care.