Predictors of acute pancreatitis with low elevation of serum amylase

Background and aims: Serum amylase is a traditional measure used to establish the diagnosis of acute pancreatitis (AP). The current study aimed to assess the predictors and clinical outcome of AP with low serum amylase. Methods: All patients were divided into two groups, based on their serum amylase level within the first 2 days after hospital admission: group 1 (amylase >= 300 U/L) and group 2 (amylase <300 U/L). Clinical outcomes were compared between the two groups before and after 1: 1 propensity score matching. Clinical and biochemical parameters were collected and evaluated as potential predictors of AP with low serum amylase. Results: A total of 464 patients were enrolled. After propensity score matching according to age, gender, time interval before admission, hematocrit, blood urea nitrogen and creatinine, 108 matched pairs of patients were selected. There was no significant statistical difference between group 2 and group 1 with respect to severity of AP, median days of stay in hospital and death. Multivariate analysis indicated that biliary etiology (odds ratio [OR]: 0.499; 95% confidence interval [CI]: 0.265-0.942; P=0.003), low-density lipoprotein cholesterol (LDL-C) (OR: 1.009; 95% CI: 1.002-1.017; P=0.017) and triglyceride levels (OR: 1.001; 95% CI: 1.0001-1.001; P=0.015) were independently associated with development of AP along with low serum amylase. Conclusion: Serum amylase level was not related to the severity of AP, median hospital stay (days) and death. Biliary etiology, LDL-C and triglyceride levels were independently associated with the development of AP with lower elevation of serum amylase.