ABSTRACT
Background: Various studies have been done to assess the use of serum amylase, serum lipase, and amylase/lipase (A/L) ratio to differentiate between the causes of pancreatitis. Scoring systems (Ranson’s, Glasgow, and APACHE II) have been developed to assess the severity of pancreatitis. Computed tomography (CT), a gold standard for prognosticating pancreatitis, is expensive and affordability is a concern in developing country like ours. Methodology: We did a retrospective study in a tertiary care hospital to assess the use of A/L ratio to predict the severity of acute pancreatitis and to correlate with the presence on necrosis in comparison to that given by CT. Results: The median A/L ratio was higher among those with severe and necrotizing pancreatitis, 0.19 (interquartile range [IQR] 0.124–0.304) and 0.183 (IQR 0.117–0.300), respectively. Cutoff of 0.14 was taken, based on the receiver operating characteristic curves, which could predict severity with 75% sensitivity and 53% specificity and necrosis with 73% sensitivity and 55% specificity. Higher A/L ratio correlated with a decrease in hematocrit (P = 0.08), blood glucose (P = 0.042), and aminotransferase (P = 0.03), which are indicators of clinical severity. Increase in A/L ratio had an increased trend to a longer duration of stay and a higher chance of detecting multiorgan dysfunction syndrome. Conclusions: A/L ratio of >0.14 can be used as a predictor of severity as it indicates the presence or absence of necrosis, which further aids in referral and the need for a CT scan in low-resource settings