Fatty pancreas and increased body mass index are risk factors of pancreatic fistula after pancreaticoduodenectomy

Background Pancreatic fistula (PF) after pancreatoduodenectomy (PD) remains a challenging problem. The only commonly accepted risk factor is the soft consistency of the pancreatic remnant. Methods In all, 100 consecutive patients underwent PD. All data, including commonly accepted risk factors for PF and PF defined according to the International Study Group of Pancreatic Fistula, were collected prospectively. On the pancreatic margin, a score of fibrosis and a score of fatty infiltration were assessed by a pathologist blinded to the postoperative course. Results PF occurred in 31% of patients. In univariate analysis, male sex, age greater than 58 years, body mass index (BMI) ≥25 kg/m2, pre-operative high blood pressure, operation for nonintraductal papillary and mucinous neoplasm (IPMN) disease and for ampullary carcinoma, operative time, blood loss, soft consistency of the pancreatic remnant, absence of pancreatic fibrosis, and presence of fatty infiltration of the pancreas were associated with a greater risk of PF. In a multivariate analysis, only BMI ≥25kg/m2, absence of pancreatic fibrosis, and presence of fatty pancreas were significant predictors of PF. A score based on the number of risk factors present divided the patient population into 4 subgroups carrying a risk of PF that ranged from 7% (no risk factor) to 78% (3 risk factors) and from 0% to 81%, taking into account only symptomatic PF (grade B and C). Conclusion The presence of an increased BMI, the presence of fatty pancreas, and the absence of pancreatic fibrosis as risk factors of PF allows a more precise and objective prediction of PF than the consistency of pancreatic remnant alone. A predictive score based on these 3 factors could help to tailor preventive measures. Pancreatic fistula (PF) after pancreatoduodenectomy (PD) remains a challenging problem. The only commonly accepted risk factor is the soft consistency of the pancreatic remnant. In all, 100 consecutive patients underwent PD. All data, including commonly accepted risk factors for PF and PF defined according to the International Study Group of Pancreatic Fistula, were collected prospectively. On the pancreatic margin, a score of fibrosis and a score of fatty infiltration were assessed by a pathologist blinded to the postoperative course. PF occurred in 31% of patients. In univariate analysis, male sex, age greater than 58 years, body mass index (BMI) ≥25 kg/m2, pre-operative high blood pressure, operation for nonintraductal papillary and mucinous neoplasm (IPMN) disease and for ampullary carcinoma, operative time, blood loss, soft consistency of the pancreatic remnant, absence of pancreatic fibrosis, and presence of fatty infiltration of the pancreas were associated with a greater risk of PF. In a multivariate analysis, only BMI ≥25kg/m2, absence of pancreatic fibrosis, and presence of fatty pancreas were significant predictors of PF. A score based on the number of risk factors present divided the patient population into 4 subgroups carrying a risk of PF that ranged from 7% (no risk factor) to 78% (3 risk factors) and from 0% to 81%, taking into account only symptomatic PF (grade B and C). The presence of an increased BMI, the presence of fatty pancreas, and the absence of pancreatic fibrosis as risk factors of PF allows a more precise and objective prediction of PF than the consistency of pancreatic remnant alone. A predictive score based on these 3 factors could help to tailor preventive measures.