Introduction Hospital admissions for cirrhosis and related complications are rising and patients are getting younger. Hence, physicians are increasingly faced with making difficult referrals to intensive care for patients with multiple organ dysfunction. We examined the attitudes of a mixed cohort of physicians and intensivists, including trainees, to compare critical care admission decisions for a range of medical diagnoses including cirrhosis.

Methods A web survey containing eight clinical scenarios, including one describing a 45 year old man with severely decompensated ALD (bilirubin 410 umol/L), sepsis and renal failure (prior to resuscitation) was advertised via email to trusts in the south of England. Respondents were asked to rate the degree with which they would advocate for ICU admission on a scale on 1–10 (1 = would not consider ICU, 10 = insist on ICU). All cases had similar SOFA scores (10–11). Other cases included pneumonia, chronic airways disease, GI bleeding with loss of output, relapsed myeloma, post operative aspiration, ruptured AAA, and CKD requiring renal replacement. Opinions on the level of organ support to be offered, or alternatively the ceilings of de-escalated care were further explored.

Results Of 144 respondents, 23% were consultant physicians, 22% consultant anaesthetists and 22% specialist trainees. Mean advocacy score for ALD was 7.2, which ranked 4 out of 8 scenarios. COPD scored lowest, with a mean score of 4.9, acute on chronic kidney disease highest with 8.5. 55% would strongly advocate for escalation (score 8, 9 or 10). Of the 21 who did not favour escalation to ICU (score 1–5), “unlikely to survive ICU admission” (80%) and “end stage organ disease” (85%) were the most frequently cited reasons, and 6 cited “lifestyle decision”. 9 recommended making the patient DNACPR and 3 would institute palliative care measures. Of the majority who would consider escalation, 69% recommended “No limits on care – full escalation”. In a separate question 34% of all respondents said they “frequently” (12%) or “sometimes” (21%) considered resource utilisation or cost when making individual clinical decisions on escalation of care.

Conclusion Most respondents favoured escalation of care to some degree, however a significant minority interpreted the same clinical information with a degree of prognostic pessimism. Continued education regarding early opportunities to improve prognosis in decompensated liver disease is required.

Disclosure of Interest None Declared.