Correlation between admission laboratory values, early abdominal computed tomography, and severe complications of gallstone pancreatitis

BACKGROUND: Our previous study demonstrated that Balthazar grade D or E pan creatitis on early abdominal computed tomography (CT) scan correlated with severe complications of gallstone pancreatitis (GP). OBJECTIVE: TO compare the efficacy of individual admission laboratory crite ria, multiple criteria scoring systems and CT scan for predicting severe co mplications of GP. METHODS: Consecutively admitted patients with GP underwent selective early CT scanning (<72 hours). Ail patients were prospectively monitored for seve re complications. RESULTS: Of the 66 patients studied, 21 (32%) did not undergo for early CT scanning and underwent cholecystectomy with no complications. Forty-five pa tients (68%) had an early abdominal CT scan. Of the 12 patients with grade E pancreatitis, 6 (50%) developed severe complications versus only 2 of 33 (6%) with grade A to D pancreatitis (P = 0.002). A significant correlation was found between admission white blood cell count <greater than or equal t o>14.5 x 10(9)/L and grade E pancreatitis on early CT scan (P = 0.002). How ever, admission glucose greater than or equal to 150 mg/dL was the best pre dictor of complications (sensitivity 100%, negative predictive value 100%). CONCLUSIONS: Although Balthazar grade E on early CT scan correlates with se vere complications of GP, admission glucose greater than or equal to 150 mg /dL has a better sensitivity and negative predictive value, is quicker to u se, and is more cost-effective as a prognostic indicator. (C) 2001 by Excer pta Medica, Inc.