Abstract

Hypothesis: Intravenous fluid therapy is established in skin burn and acute pancreatitis. Surface area and depth decide the fluid volume in burn. Acute pancreatitis is a chemical burn of pancreas and retroperitoneum, can the volume of fluid therapy determine the severity?

Introduction: Acute pancreatitis can be mild to Severe. Currently available methods to predict severe pancreatitis are complex, cumbersome and inaccurate. A simple way is required.

Materials and Methods: Patients presenting clinically as acute pancreatitis with elevated serum amylase/lipase thrice than normal were managed by intravenous fluid similar to skin burnt injury, so as to have urine output of 0.5 - 1 mL per Kg. Based on the difference between the intake and output, they were predicted as necrotising/non-necrotising at 24 hrs. Confirmation was done on CT scan after 72 hrs or at autopsy.

Results: There was 63 patients with 9 females and 54 males, age ranging from 19 - 59 years. 38 were predicted mild and 25 as severe based on the fluid deficit ranged from 0.4 lit to 9.9L after 24 hrs. the prediction was accurate in 84.126% (severe 85.185% and non-severe 83.333%) cases and inaccurate in 15.873% of cases. And on ROC the area under the curve came out to be 88% at a confidence interval of 95% with a sensitivity and specificity of 88.416 and 82.857% respectively. Patients with fluid deficit > 8.7L died.

Conclusion: Fluid deficit at 24 hrs can predict severe AP 85.2%. Fluid deficit > 8.7 L at 24 hours were fatal.

Keywords: Acute Pancreatitis; Predicting Severe; 24 Hrs Fluid Deficit; Non-Necrotising; Sterile Necrosis; Infected Necrosis