Effect of different degrees of hypertriglyceridemia on severity of acute pancreatitis

Value of BISAP score for predicting severity of hyperlipidemic acute pancreatitis: A meta-analysis

BACKGROUND

Early identification of hyperlipidemic acute pancreatitis (HLAP) patients at risk of developing potentially lethal complications is of great clinical significance. Studies have suggested that the bedside index for in acute pancreatitis (BISAP) score is associated with adverse outcomes.

AIM

To assess the accuracy of BISAP score as a prognostic marker for severity of HLAP.

METHODS

A systematic search of main computerized databases was performed to identify eligible cohort studies on the predictive value of BISAP score for severity of HLAP. The STATA software and Meta-disc software were applied to carry out the meta-analysis.

RESULTS

Ten studies (n = 1591) were included. The overall sensitivity and specificity of BISAP score ≥ 3 for predicting mortality in HLAP were 0.85 (95% confidence interval [CI]: 0.65-0.96) and 0.86 (95%CI: 0.82-0.88), respectively; the area under the curve (AUC) was 0.937, and the diagnostic odds ratio (DOR) was 42.00 (95%CI: 12.86-139.12). Regarding the increase of BISAP score for prediction of severe acute pancreatitis (SAP), the pooled sensitivity was 0.69 (95%CI: 0.61-0.76), and the specificity was 0.82 (95%CI: 0.78-0.85), with AUC and DOR being 0.900 and 18.47 (95%CI: 6.82-42.03), respectively. The pooled sensitivity, specificity, and DOR for prediction of moderately severe acute pancreatitis (MASP) + SAP were 0.54 (95%CI: 0.50-0.63), 0.91 (95%CI: 0.89-0.93), and 15.55 (95%CI: 6.91-34.99), respectively; the AUC was 0.724. BISAP score was superior to APACH Ⅱ and Ranson score in predicting the severity of HLAP.

CONCLUSION

BISAP score is a reliable tool to identify the severity of HLAP, but it has a suboptimal sensitivity for predicting SAP and SAP + MSAP.