Au, M., Mah, X., Pham, D., Yeoh, S., Parthasarathy, N., Bloom, S., Sawhney, R., and Nicoll, A. (2019) The outpatient Model for End‐Stage Liver Disease Sodium score before intensive care unit (ICU) admission is the best indicator of 12‐month mortality for cirrhotic patients after ICU discharge. Journal of Gastroenterology and Hepatology, 34 (S2). p. 106.

PDF (Published version) - Published Version Restricted to Repository staff only

 

3

Abstract

Background and Aim: Many clinicians use disease scores to assess a patient's degree of decompensation and the suitability of intensive care unit (ICU) admission. However, these scores may be temporarily elevated due to reversible illness and may not reflect the patient's baseline status before acute deterioration. The aim of this prospective study was to examine the rates and predictors of mortality 12 months after ICU discharge, including disease scores both during and before acute deterioration, among patients with cirrhosis in two non‐transplant metropolitan hospitals.

Methods: All patients with cirrhosis admitted to the ICU between April 2016 and March 2019 were recruited at the time of ICU admission. Patient demographics, reasons for ICU admission, and 12‐month mortality rates were collected. Liver‐specific scores (Child–Pugh, Model for End‐Stage Liver Disease Sodium [MELD‐Na], and CLIF Consortium Acute‐on‐Chronic Liver Failure [CLIF‐C‐ACLF]) and ICU prognostic scores (Sequential Organ Failure Assessment [SOFA], Simplified Acute Physiology Score [SAPS] II, and Acute Physiology and Chronic Health Evaluation [APACHE] II) were calculated. We also included pre‐acute deterioration MELD‐Na and Child–Pugh scores based on the results between 1 and 6 months before the deterioration from the clinic or the general practitioner. The primary outcome measure was all‐cause mortality at 12 months after ICU discharge. The ability of the scores to predict the primary outcome was compared using area under receiver operating characteristic curves (AUROCs).

Results: Sixty‐four patients had 12‐month mortality data (71% male; mean age, 57.2 years). The main causes of deterioration requiring ICU admission were gastrointestinal hemorrhage (n = 33); sepsis (n = 11); hypotension (n = 11); hepatic encephalopathy (n = 7); post‐elective surgery (n = 3); cardiac arrhythmia (n = 2); and other (n = 3). The etiologies of cirrhosis were alcohol (n = 56); non‐alcoholic steatohepatitis (n = 9); hepatitis C (n = 7); and other (n = 6). The 12‐month mortality overall was 53% (n = 23), with the majority of deaths from progressive liver failure (n = 16). The best predictor of mortality at 12 months after ICU discharge was the pre‐acute deterioration MELD‐Na score (AUC, 0.815). This was higher than the AUCS for pre‐deterioration Child–Pugh Score (0.638), the MELD‐Na (0.501) and Child–Pugh (0.625) scores at the time of ICU admission, and the ICU prognostic scores (SOFA, 0.808; SAPS II, 0.738; APACHE II, 0.588; and CLIF‐C‐ACLF, 0.658). Among patients who were Child–Pugh class C at the time of ICU admission, those who were already Child–Pugh C before acute deterioration (n = 6) had a mortality rate of 83.3%, but those were Child–Pugh A/B before acute deterioration (n = 24) had a lower mortality rate of 54.2% (P = 0.192).

Conclusion: The pre‐acute deterioration MELD‐Na score is the best predictor of long‐term survival after ICU discharge. A patient's degree of decompensation at baseline should guide decision making about ICU admission, more so than disease scores at the time of ICU referral, as these scores may be temporarily elevated due to reversible illness.

Actions (Repository Staff Only)

Item Control Page