Comparison study of four different scoring systems on short-term prognostic value in patients with decompensated liver cirrhosis

ZHENG Sheng,TANG Ying-mei,LIU Hai,YOU Li-ying
DOI: https://doi.org/10.3969/j.issn.1008-1704.2012.06.003
2012-01-01
Abstract:Objective To compare the short-term predicting values for prognosis of patients with decompensated liver cirrhosis through the model for end-stage liver disease(MELD) scoring system,MELD-Na scoring system,MESO exponent scoring system and iMELD scoring system.Methods Two hundred and thirty hospitalized patients with decompensated liver cirrhosis in gastroenterology department of the Third People's Hospital in Yunnan Province from October 2008 to October 2011 were enrolled in the study.MELD,MELD-Na,MESO and iMELD score of each patient were calculated at the time of admission to hospital respectively.Kaplan-Meier method was adopted to compare their survival rates.Receiver characteristic curve(ROC) and area under the curve(AUC) were used to compare the short-term prognostic values of patients with decompensated liver cirrhosis through the four different scoring systems.Results Sixty eight patients of two hundred and thirty patients died within the three months.MELD,MELD-Na,MESO and iMELD scores of the survival group were 22.34±4.36,24.26±5.45,1.62±0.23 and 37.59±6.97 respectively,and those of the death group were 27.76±5.28,30.11±6.19,2.05±0.18 and 46.65±7.01 respectively.The scores of MELD, MELD-Na,MESO and iMELD scoring system in the dead group were higher than those in the survival group.The comparison of the four scoring ystems between the two groups had statistical significance(P = 0.005,0.005,0.000, 0.003).The areas under ROC curves of MELD,MELD-Na,MESO and iMELD scoring systems were 0.852,0.856, 0.857 and 0.847 respectively and their 95%confidence intervals were 0.759-0.897,0.754-0.893,0.760-0.898 and 0.781-0.906.Conclusion MELD,MELD-Na,MESO and iMELD scoring systems all had prognostic values in patients with decompensated liver cirrhosis.However,there was no significantly statistical differences of prognosis values among the four sconing systems.More accurate determination of prognosis still needs combination with clinical practice and experience.
What problem does this paper attempt to address?