A model predicting short-term mortality in patients with advanced liver cirrhosis and concomitant infection

Ying Li, Roongruedee Chaiteerakij, Jung Hyun Kwon, Jeong Won Jang, Hae Lim Lee, Stephen Cha, Xi Wei Ding, Charat Thongprayoon, Fu Shuang Ha, Cai Yun Nie, Qian Zhang, Zhen Yang, Nasra H. Giama, Lewis R. Roberts, Tao Han

Research output: Contribution to journalArticlepeer-review

2 Scopus citations


Infection is a common cause of death in patients with advanced cirrhosis. We aimed to develop a predictive model in Child–Turcotte–Pugh (CTP) class C cirrhotics hospitalized with infection for optimizing treatment and improving outcomes. Clinical information was retrospectively abstracted from 244 patients at Tianjin Third Central Hospital, China (cohort 1). Factors associated with mortality were determined using logistic regression. The model for predicting 90-day mortality was then constructed by decision tree analysis. The model was further validated in 91 patients at Mayo Clinic, Rochester, MN (cohort 2) and 82 patients at Seoul St. Mary’s Hospital, Korea (cohort 3). The predictive performance of the model was compared with that of the CTP, model for end-stage liver disease (MELD), MELD-Na, Chronic Liver Failure–Sequential Organ Failure Assessment, and the North American consortium for the Study of End-stage Liver Disease (NACSELD) models. The 3-month mortality was 58%, 58%, and 54% in cohort 1, 2, and 3, respectively. In cohort 1, respiratory failure, renal failure, international normalized ratio, total bilirubin, and neutrophil percentage were determinants of 3-month mortality, with odds ratios of 16.6, 3.3, 2.0, 1.1, and 1.03, respectively (P < .05). These parameters were incorporated into the decision tree model, yielding area under receiver operating characteristic (AUROC) of 0.804. The model had excellent reproducibility in the U.S. (AUROC 0.808) and Korea cohort (AUROC 0.809). The proposed model has the highest AUROC and best Youden index of 0.488 and greatest overall correctness of 75%, compared with other models evaluated. The proposed model reliably predicts survival of advanced cirrhotics with infection in both Asian and U.S. populations.

Original languageEnglish (US)
Article numbere12758
JournalMedicine (United States)
Issue number41
StatePublished - Oct 1 2018


  • 3-month mortality
  • Cirrhosis
  • Infection
  • MELD score

ASJC Scopus subject areas

  • General Medicine


Dive into the research topics of 'A model predicting short-term mortality in patients with advanced liver cirrhosis and concomitant infection'. Together they form a unique fingerprint.

Cite this