Predictive value of serum cholinesterase in the mortality of acute pancreatitis: A retrospective cohort study

Background Severe acute pancreatitis has a high mortality of 20%-40%, but there is a lack of optimal prognostic biomarker for the severity of acute pancreatitis (AP) or mortality. This study is designed to investigate the relationship between serum cholinesterase (ChE) level and poor outcomes of AP. Methods A total of 1904 AP patients were screened in the study, and we finally got 692 patients eligible for analysis. Patients were divided into 2 groups based on serum ChE. The primary outcome was mortality, and multivariable logistic regression analysis for mortality was completed. Additionally, we used receiver operating characteristic (ROC) curve analysis to clarify the predictive value of serum ChE for mortality and organ failure. Results Three hundred and seventy eight patients and 314 patients were included in the ChE-low and ChE-normal group, respectively. Patients in the ChE-low group were older (46.68 +/- 12.70 vs. 43.56 +/- 12.13 years old, p = .001) and had a lower percentage of man (62.4% vs. 71.0%, p = .017) when compared to the ChE-normal group. Mortality was significantly different in two groups (10.3% vs. 0.0%, p < .001). Moreover, organ failure also differed significantly in two groups (46.6% vs. 8.6%, p < .001). Decreased ChE level was independently associated with mortality in acute pancreatitis (odds ratio: 0.440; 95% confidence interval, 0.231, 0.838, p = .013). The area under the curve of serum ChE was 0.875 and 0.803 for mortality and organ failure, respectively. Conclusions Lower level of serum ChE was independently associated with the severity and mortality of AP.