Allocation of patients with liver cirrhosis and organ failure to intensive care: Systematic review and a proposal for clinical practice

Table 1 Included trials/studies

Author, year, location, study design	Main target of the study	Patient population	Scores	Allocation system	In hospital mortality	ICU mortality
Kavli et al[7], 2012, Denmark, cohort study	This study investigated the severity of organ failure, and the frequency and outcome of withholding therapy in patients with advanced alcoholic liver cirrhosis admitted to a Scandinavian ICU	87 adult patients with clinical or histological diagnosis of liver cirrhosis admitted to ICU at a University hospital in Denmark, within a 3 years period from January 2007-January 2010	APACHE II, SAPS II, and SOFA were better at predicting mortality than Child-Pugh score	No specific allocation system is proposed	Only ICU data	With 3 or more organ failures the ICU mortality was > 90%
Shawcross et al[1], 2012, United Kingdom, cohort study	The aim of this study was to prospectively study the resource allocation and cost of a large cohort of patients with cirrhosis and one or more extrahepatic organ failure(s)	563 patients were admitted to the Liver ICU at King’s College Hospital, between 2000 and 2007	The median (IQR) for all patients admitted and surviving for > 8 h on day 1 (n = 548) was Child-Pugh score 12 (11-13), MELD 25 (14-34), APACHE II 22 (16-28) and SOFA 11 (8-13)	No specific allocation system is proposed	Overall hospital mortality of 59% (330/563)	256/563 (51%) patients died whilst in the Liver ICU
				Patients with cirrhosis admitted to ICU require high levels of organ support but ICU admission is not necessarily futile
Ginès et al[9], 2012, Spain, review	This review focuses on the diagnostic approach and treatment strategies cur-rently recommended in the critical care management of patients with cirrhosis	None	MELD and Child-Pugh scores have important limitations in the establishment of prognosis in critically ill cirrhotic patients	Encephalopatic cirrhotic patients (grade 3 or 4 hepatic encephalopathy) require ICU admission and intubation	Hospital mortality rates in patients with 1, 2 or 3 organ/system failures were 48%, 65%, and 70%, respectively	ICU and 6-mo mortality rates of 41% and 62%, respectively
				Patients with a low MELD score (< 15), should be immediately considered for ICU. Contrary, in patients with end-stage cirrhosis (MELD > 30), 3 or more organ failures, and no perspective of transplantation, aggressive management is questionable		59% of cirrhotic patients placed on mechanical ventilation died during their stay in the ICU
Berry et al[10], 2013, United Kingdom, review	This review focuses on patients with cirrhosis, especially survival analysis and prognostic models		Child-Pugh score does not perform as well as general critical illness scoring systems	No specific allocation system is proposed	Greater than 60%	ICU mortality of up to 65%, rising to 90% with sepsis, if more than 1 d of respiratory support and renal support were required
			The MELD score performs better than the Child-Pugh score, yet the SOFA score is superior to both Child-Pugh and MELD score	Early aggressive approach to organ support is justified
Saliba et al[11], 2013, France, review	This review focuses on prognostic scores and admission to ICU for critically ill cirrhotic patients	None	Suggests that ICU scores (SOFA, APACHE II, SAPS II) predict the outcome of cirrhotic patients admitted to the ICU better than liver scores (MELD and Child-Pugh)	No specific allocation system is proposed	Only ICU data	Ranges between 34%-69%
				The persistence after ICU admission of three or more organ failures and the need for three or more organ supports, may lead to consider a limitation in life sustaining treatments, as a fatal outcome is almost constant

Included studies of the present review. The table describes the included studies. ICU: Intensive care unit; SOFA score: Sequential Organ Failure Assessment score; MELD: Model of End-stage Liver Disease.