PTU-038 End of life care planning in patients with end-stage liver disease: clinical practice remains variable

Introduction End-stage liver disease (ESLD) is a terminal diagnosis without transplantation, and anticipatory discussions regarding end of life (EoL) care are appropriate when poor prognostic factors are present. Aim To assess the point at which decisions regarding resuscitation and EoL care were made with patients who died of ESLD and to analyse factors associated with delayed discussion. Methods We identified inpatients with proven ESLD under the care of a hepatology team over a 24 month period (Jan 2017-Dec 2018) who died during or shortly after their last admission. Data was collected relating to clinical course, prognostic indicators, EoL care planning considerations, dates of DNACPR decision, palliative care referral and interval to death. Results 19 patients were identified (12 male: 7 female); mean age at death was 61. 11/19 patients had alcoholic liver disease (ALD), 5 had ALD/Hepatitis C. 12 patients died on the ward, 3 in ICU, 2 at home and 2 at a nursing home. Child-Pugh Scores (CPS) ranged from B-6 to C-13 and average MELD score was 23 (range 8–38). None were eligible for transplantation: 12 due to active alcohol use, 3 due to co-morbidities, 1 malignancy, 2 unknown. Median number of admissions in the year preceding death was 2 (range 0–5). Predominant symptoms prior to death were respiratory distress, confusion and pain. Average interval between admission and death was 24 days. 13/19 patients were referred for inpatient palliative care input. Although all patients had a DNACPR notice in place, the average DNACPR-to-death interval was just 20 days (range 0–139 days). EoL decisions were made ‘early’ (DNACPR-to-death time >21 days) in patients (n=6) who had gradual disease evolution and/or a long period of contact with the service. In those with ‘short’ DNACPR-to-death times (n=13), 9 had been hospitalised 2 or more times in the year prior to death (this being a known marker of poor prognosis), and 2 had been admitted 4 or 5 times. 5/19 patients had a documented Amber Care Bundle referral (prognostic uncertainty tool). 14 patients had a documented ceiling of care discussion; of these 12 were for ward-based care only and 2 were for Level 3 escalation. Conclusions Patients with end stage liver disease continue to be engaged in EoL and treatment escalation discussions relatively late, despite clear indicators of poor prognosis (including recurrent admissions and non-transplantable status) within the previous year. Those well known to specialist teams who deteriorate gradually have a greater chance of expressing their preferences. Increased awareness of poor prognostic features is required in the secondary care setting.