Alban Zarzavadjian Le Bian1, David Fuks2,3, Filippo Montali4, Manuela Cesaretti5, Renato Costi6,7, Philippe Wind1, Claude Smadja7, Brice Gayet2,3. 1. 1Department of Digestive Surgery and Surgical Oncology, Hôpital Avicenne, Assistance Publique - Hôpitaux de Paris, Bobigny, France. 2. 2Department of Digestive Disease, Oncologic and Metabolic Surgery Institut Mutualiste Montsouris, Paris, France. 3. 3Université Paris Descartes, Paris, France. 4. 4Department of Hepatic, Pancreatic and Biliary Surgery - Hôpital Paul Brousse, Assistance Publique - Hôpitaux de Paris, Villejuif, France. 5. 5HPB Surgery and Liver Transplantation, Nice, France. 6. 6Dipartimento di Scienze Chirurgiche, Università degli Studi di Parma, Parma, Italy. 7. 7Department of Digestive Surgery - Assistance Publique - Hôpitaux de Paris Hôpital Antoine Béclère, Université Paris Sud, Clamart, France.
Abstract
Background: Risk factors for pancreatic fistula (POPF) occurrence after pancreaticoduodectomy (PD) are mostly known. Identifying those that are linked to clinically relevant POPF (Grades B and C) (CR-POPF) is critical, as CR-POPF is associated with more complications and a higher mortality rate. Methods: From 2004 to 2016, 270 consecutive patients who underwent PD in two academic centers were compared retrospectively according to the occurrence of CR-POPF. Results: A series of patients with a median age of 64.5 years (range 30.6-88.7 years) underwent PD. They were allocated to two groups: CR-POPF (Grades B and C)(n = 74; 27.4%) and without clinically relevant POPF (cr-POPF) (no fistula formation or Grade A) (n = 196). Pancreatic ductal adenocarcinoma was the main indication for the procedure (58.5%). Post-operative complications Clavien-Dindo I/II and Clavien-Dindo III/IV and in-hospital death occurred in 109 (40.4%), 67 (24.8%), and 18 (6.7%) patients, respectively. After univariate analysis, CR-POPF was associated with a Body Mass Index (BMI) >25 kg/m2 (p < 0.0001), pancreatic duct diameter <3 mm (p = 0.047), soft pancreas texture to palpation (p = 0.037), and peri-operative transfusion (p < 0.001). After multivariate analysis, high BMI (p = 0.026), transfusion (p < 0.001), length of hospital stay (p < 0.0001), and in-hospital death (p = 0.004) were associated with CR-POPF. Conclusions: In-hospital death and length of hospital stay after PD are related to CR-POPF. A BMI >25 kg/m2 and peri-operative blood transfusion are objective risk factors for CR-POPF.
Background: Risk factors for pancreatic fistula (POPF) occurrence after pancreaticoduodectomy (PD) are mostly known. Identifying those that are linked to clinically relevant POPF (Grades B and C) (CR-POPF) is critical, as CR-POPF is associated with more complications and a higher mortality rate. Methods: From 2004 to 2016, 270 consecutive patients who underwent PD in two academic centers were compared retrospectively according to the occurrence of CR-POPF. Results: A series of patients with a median age of 64.5 years (range 30.6-88.7 years) underwent PD. They were allocated to two groups: CR-POPF (Grades B and C)(n = 74; 27.4%) and without clinically relevant POPF (cr-POPF) (no fistula formation or Grade A) (n = 196). Pancreatic ductal adenocarcinoma was the main indication for the procedure (58.5%). Post-operative complications Clavien-Dindo I/II and Clavien-Dindo III/IV and in-hospital death occurred in 109 (40.4%), 67 (24.8%), and 18 (6.7%) patients, respectively. After univariate analysis, CR-POPF was associated with a Body Mass Index (BMI) >25 kg/m2 (p < 0.0001), pancreatic duct diameter <3 mm (p = 0.047), soft pancreas texture to palpation (p = 0.037), and peri-operative transfusion (p < 0.001). After multivariate analysis, high BMI (p = 0.026), transfusion (p < 0.001), length of hospital stay (p < 0.0001), and in-hospital death (p = 0.004) were associated with CR-POPF. Conclusions: In-hospital death and length of hospital stay after PD are related to CR-POPF. A BMI >25 kg/m2 and peri-operative blood transfusion are objective risk factors for CR-POPF.