Incidence of pancreatic anastomotic failure and delayed gastric emptying after pancreatoduodenectomy in 507 consecutive patients: Use of a web-based calculator to improve homogeneity of definition

Background. Many definitions are used in the literature for pancreatic anastomotic failure (PAF) and delayed gastric emptying (DGE) after pancreatoduodenectomy (PD). To promote homogeneity, published reports after 2005 have used the International Study Group on Pancreatic Surgery (ISGPS) consensus definition for PAT, and DGE; however; subsequent authors have had to interpret or modify the ISGPS classification to make it useable. The solution might be to create a web-based calculator; test it for ambiguity and reliability with a large number of cases, and then make it available to the public. Methods. Using 507 consecutive patients undergoing PD and 14 questions, we created a web-based calculator based on the ISPGS classification to assess the incidence and grade of clinical impact (none, moderate, or major deviation) for PAP and DGE. As the calculator's formulas were tested, ambiguous terms were identified and resolved. Results. The incidence for cases with clinical impact from PAF was 10% and from DGE it was 12%. Multivariate analysis identified 4 factors predictive for PAF male sex, body mass index (BMI) > 30 kg/m(2), soft gland texture, and main pancreatic duct size <= 3 nun. Predictive factors for DGE included 2 factors: not using a surgical microscope, and simultaneous PAF. Conclusion. A web-based calculator was developed to promote homogeneity of method for grading of PAF and DGE after PD. Anyone with access to the web can now compare their results to the current study. (Surgery 2010;147:503-15.)