Purpose. Identification of the possibilities of contrast enhancement computed tomography in evaluated the number of the acinar structures in the pancreatic parenchyma at the preoperative stage to predict the development pancreatic fistula.Material and methods. In 2016–2019, 196 pancreatoduodenectomy were performed. 86 patients were retrospectively selected. Patients were divided into 2 groups: group 1 included 16 observations with the development of clinically significant pancreatic fistula, in 2 – 70 cases without complications. According to preoperative contrast enhancement computed tomography, structure of the pancreas, pancreatic parenchyma thickness, pancreatic duct diameter, the density of the pancreas in the native phase, relative parenchyma enhancement ratio, washout coefficient, pancreas stump volume were evaluated. According histological, the number of acinar and fat cells in the section of the removed pancreas was evaluated.Results. “Soft” structure of the pancreas (r = 0.374, p = 0.000), pancreatic parenchyma thickness (r = 0.549, p = 0.000), the density of the pancreas in the native phase of the scan (r = 0.568, p = 0.000), the values relative parenchyma enhancement ratio (r = 0.63, p = 0.000), pancreas stump volume (r = 0.508, p = 0.000) positively correlated with clinically significant pancreatic fistula and the number of acinar cells. Pancreatic duct diameter (r = −0.339, p = 0.001) negatively correlated with clinically significant pancreatic fistula and the number of acinar cells. Pancreatic fistula risk is 3.09 times higher with the number of acini more than 72.5%, sensitivity 75%, specificity 75.71%. Pancreatic fistula risk is 1.8 times higher with the density of the pancreas in the native phase over 35.5 HU sensitivity 62%, specificity 65%. Pancreatic fistula risk is 2.76 times higher with values parenchyma accumulation coefficient more than 1, sensitivity 75%, specificity 73%.Conclusions. Contrast enhancement computed tomography allows evaluating acinar index in the preoperative period to pick out the high-risk patient group to development of pancreatic fistula.
Prediction of non-functioning pancreatic neuroendocrine tumor grades with fractal analysis of preoperative contrast-enhanced computed tomography images