Medizinische Universität Graz Austria/Österreich - Forschungsportal - Medical University of Graz
Gewählte Publikation:
Wurm, S.
Simple risk stratification for patients with decompensated cirrhosis upon hospital admission
Humanmedizin; [ Diplomarbeit ] Graz Medical University; 2017. pp. 104
[OPEN ACCESS]
FullText
- Autor*innen der Med Uni Graz:
- Betreuer*innen:
- Spindelböck Walter Johann
- Stadlbauer-Köllner Vanessa
- Altmetrics:
- Abstract:
- Background: Accurate assessment of the disease severity and prognosis of decompensated cirrhosis is crucial to establish an optimal therapeutic management, but it remains challenging to accomplish. Currently, well-established scoring systems for liver impairment, such as the Model of End stage Liver Disease (MELD) score or the Child-Turcotte-Pugh (CTP) score are used to assess the short-term and long-term mortality in patients with cirrhosis. However, there is growing evidence that in patients with decompensated cirrhosis the mortality rate is dependent on superimposed organ failure/s and the presence of systemic inflammatory response syndrome (SIRS) rather than isolated liver failure – which these widely used scoring systems do not assess for. Consequently, there is a growing interest in evaluating predictive parameters reflecting multiple organ failure and SIRS alongside liver impairment, as these will potentially provide a more reliable mortality risk assessment in patients with decompensated cirrhosis. Aims: The aim of the present study is to propose a simple risk stratification system for the short-term as well as for the long-term mortality rate in patients with decompensated cirrhosis. Material and methods: Clinical and laboratory data were collected from the database of the Hospital of the Medical University of Graz. 165 patients diagnosed with cirrhosis and hospitalized due to acute decompensation of their underlying liver disease between January 2008 and December 2011 were retrospectively enrolled. Multivariate and univariate binary logistic regression analysis were carried out to evaluate potential predictive variables for mortality at day 30, day 90 and 1 year. The Youden’s index was used to calculate the best diagnostic cut-off values. Results: In the cohort of this study, the average CTP and MELD scores were respectively 10 ± 2 and 21 ± 8 and the most frequent etiology of cirrhosis was chronic alcohol consumption (89%). Short-term mortality (at day 30 and day 90) was best predicted by albumin levels <2,55g/dL, CRP levels >31mg/L, MELD >23 and the presence of ACLF (any grade). CRP levels >31mg/L and MELD >23 best predicted long-term mortality (1 year). When combining the MELD score or the presence of ACLF (independent of the grade) with the most predictive variables we obtained four 2-variable models for risk stratification that were accurate mortality predictors: the combination of ACLF and albumin (‘ACLF-Albumin model’) was predictive for the thirty-day mortality, while the ninety-day mortality was best predicted by the combination of MELD score and CRP (‘MELD-CRP model’); for the one-year mortality, both the combination of MELD score and CRP (‘MELD-CRP model’) and the combination of ACLF and CRP (‘ACLF-CRP model’) were similarly predictive. Importantly, these models showed improved accuracy when compared with the MELD score or the presence of ACLF (any grade) alone. Conclusion: The newly defined 2-variable models ‘ACLF-Albumin model’, ‘MELD-CRP model’ and ‘ACLF-CRP model’ allow for simple risk stratification of patients with decompensated cirrhosis into groups with low, intermediate and high short-term mortality, as well as long-term mortality rates. Importantly, these models show improved accuracy when compared with the original scores. Applying these models, and depending on the mortality risk, either discharge, ward admission or ICU admission may be recommended. Taking our results into account, establishing scoring systems that combine liver impairment with other predictive parameters reflecting SIRS and extrahepatic organ failure is required for the adequate assessment of disease severity in patients with decompensated cirrhosis.