Aim. To improve the results of pancreatic resections through decrease of postoperative pancreatitis incidence. Material and methods. It was analyzed 207 patients who underwent pancreatic surgery for pancreatic tumor (n=137) or chronic pancreatitis (n=70). 22 risk factors of postoperative pancreatitis were analyzed in 112 patients retrospectively. In prospective study of 95 patients the efficacy of lornoxicam to prevent postoperative pancreatitis was assessed. 68 parameters of immune state were studied to estimate effect of lornoxicam. Results. Significant factors were mellow pancreatic parenchyma, tumoral disease, pancreatic duct diameter over 3 mm, pancreatric duct index over 0.2, body mass index over 27 kg/m2. Likelihood of postoperative pancreatitis was 40%, 63%, 74.3% and 88.9% if 2, 3, 4 and 5 factors were combined respectively. Preventive use of lornoxicam reduced significantly incidence of postoperative pancreatitis (p=0.042). Incidence of pancreatic fistula and arrosive bleeding was decreased insignificantly due to small number of observations. Conclusion. Assessment of significant risk factors and use of medical prevention are available to decrease likelihood of postoperative pancreatitis.