Risk of death from acute pancreatitis. Role of early, simple "routine" data

CONCLUSIONS. The analysis of all the data available in 192 patients at 24 h from admission shows that only serum glucose above 250 mg/dL (13.88 mmol/L) and serum creatinine above 2 mg/dL (176.8 mumol/L) are prognostic factors of death (P < 0.0001). When, however, pathological chest X-rays are also considered in a subset of 149 patients, these and serum creatinine are prognostic factors of death with odd ratios of 2.9 (95% CL 1.3-6.3) and 9.4 (95% CL 2.2-40.7), respectively (P < 0.0001). BACKGROUND. In patients suffering from acute pancreatitis, neither Ranson scores nor Glasgow criteria evaluation at 24 h yield a sufficiently reliable prognosis of the risk of death from the first acute attack. METHODS. After excluding posttraumatic, postsurgical, and post-ERCP acute pancreatitis, we selected 192 consecutive patients admitted in the first instance to our center for a first attack, distinguishing between patients who died and patients who survived. We used Cox's model to analyze the prognostic weight of variables available within 24 h of admission (sex, age, alcohol intake, smoking habits, 17 biochemical tests, body mass index, chest X-rays, body temperature, and shock status). RESULTS. Seventeen (8.8%) patients died; mortality showed a decreasing trend over the period of years considered and was correlated, among other things, with necrotizing type of pancreatitis, idiopathic etiology, and shock status on admission