Published online Aug 7, 2020. doi: 10.3748/wjg.v26.i29.4316
Peer-review started: March 4, 2020
First decision: April 12, 2020
Revised: June 2, 2020
Accepted: July 1, 2020
Article in press: July 1, 2020
Published online: August 7, 2020
Spontaneous bacterial peritonitis (SBP) is a detrimental infection of the ascitic fluid in liver cirrhosis patients, with high mortality and morbidity. Early diagnosis and timely antibiotic administration have successfully decreased the mortality rate to 20%-25%. Early diagnosis of asymptomatic SBP remains a great challenge in the clinic.
Currently, SBP cases are diagnosed based only on clinical symptoms, leading to possible antibiotic abuse. SBP is regulated by a variety of risk factors, including decreased activity of the reticuloendothelial system, advanced liver dysfunction, medications, and genetic factors. A multivariate predictive model may be effective for early screening of asymptomatic SBP.
The present retrospective cohort study aimed to establish an effective predictive model for early screening of asymptomatic SBP in liver cirrhosis patients with ascites. Early diagnosis of asymptomatic SBP will improve antibiotic management strategies and reduce SBP-associated mortality.
Liver cirrhosis patients with ascites who had no typical SBP symptoms were included in the current study, and divided into the case (positive cultures) and control (negative cultures) groups according to microbiological results. The demographic features, clinical information, disease activity, hematological and ascites factors were compared between the case and control groups to identify potential indicators of asymptomatic SBP. The multiple linear stepwise regression method of the logistic regression model was adopted to construct the multivariate predictive model. The diagnostic performance of the model was estimated by the receiver operating characteristic curve.
Patients in the case group were more likely to have advanced disease stages, cirrhosis related-complications, worsened hematology and ascites, and higher mortality. Based on multivariate analysis, the predictive model was as follows: y (P) = 0.018 + 0.312 × MELD (model of end-stage liver disease) + 0.263 × PMN (ascites polymorphonuclear) + 0.184 × N (blood neutrophil percentage) + 0.233 × HCC (hepatocellular carcinoma) + 0.189 × renal dysfunction. The area under curve value of the established model was 0.872, revealing its high diagnostic potential. The diagnostic sensitivity was 73.5% (72/98), the specificity was 86.7% (85/98), and the diagnostic efficacy was 80.1%.
The multivariate predictive model based on model of end-stage liver disease, polymorphonuclear, blood neutrophil percentage, hepatocellular carcinoma, and renal dysfunction exerts high diagnostic efficacy which may improve the early diagnosis of asymptomatic SBP.