PTU-111 Determining Ceiling Of Care In Decompensated Cirrhosis – Right Decisions, Right People, Right Time

Abstract

Na fell in 34/60 patients (57%) and was less likely if baseline hyponatraemia existed (38% VS 74% p = 0.004). A fall of ≥5mmol/l occurred in 23%.Median time to nadir Na was 3 days and time to recovery to pre-treatment Na was 6.5 days. No complications of hyponatremia were observed.Patients with VB were more likely (vs HRS patients) to have any fall in Na or a ≥5mmol/l reduction (68% vs 47% p = 0.1 and 32% vs 16% p = 0.12 respectively) but failed to reach significance.Mortality was 22% overall and a fall in Na was actually associated with reduced mortality -9% vs 34% (p = 0.01). Conclusion Serum Na falls in >50% receiving terlipressin and a fall ≥5mmol/l noted in 23%.However, no significant complications occurred and a fall in serum Na was actually associated with improved mortality. Patients with VB treated with terlipressin trended towards a greater likelihood of Na reduction versus those with HRS. Disclosure of Interest None Declared. PTU-111 DETERMINING CEILING OF CARE IN DECOMPENSATEDCIRRHOSIS -RIGHT DECISIONS, RIGHT PEOPLE, RIGHT TIME Introduction Decisions to initiate intensive care measures in patients with decompensated liver cirrhosis are often controversial, with mortality approaching 90% in cirrhotics with 3 organ failure. The 2013 NCEPOD report 'Measuring the Units', which examined alcoholic liver disease-related deaths, nonetheless found that 31% of those who stood to benefit from higher level care did not receive it. We studied escalation of care decisions and subsequent outcomes in cirrhotic patients with organ failure. Methods Consecutive patients with a diagnosis of cirrhosis admitted over a 90 day period in 2013 to the Bristol Royal Infirmary were studied. Severity of liver disease was assessed using ChildsPugh and UKELD. Organ failure was defined using SOFA (Sequential Organ Failure Assessment) criteria. Care escalation/ withdrawal decisions were assessed in respect to timing, seniority and expertise of decision maker. Outcome measures of ICU admission, mortality and instigation of palliative care were recorded. Results 42 admissions for 37 patients (ages 16-79, 79% male, 81% related to alcohol, 22% Childs A, 54% Childs B, 24% Childs C) were scrutinised. 30% had suffered variceal haemorrhage on, or during, admission. Of 17 patients admitted in organ failure, ICU admission was requested on 8 occasions (6 by a hepatologist, 1 during out of hours admission, 1 following out of hours deterioration). Escalation plans had been discussed with ICU prior to the point of clinical deterioration in 50%. 3 patients were accepted to ICU for mechanical ventilation, of which none survived. 1 patient was accepted in principle but improved clinically. 4 patients were declined ICU admission on grounds of poor prognosis, all of whom had alcoholic cirrhosis. Of this group all required non-invasive ventilation, with 75% surviving to discharge. Across the entire cohort 55% of hepatologist led "for full escalation if required" decisions were agreed in principle with ICU. 33% of ICU decisions to withdraw care were disc...