Defining Global Benchmarks for Total Pancreatectomy: A Multicenter Study from 22 International Expert Centers

Abstract Background Total pancreatectomy (TP) is the preferred treatment for patients with advanced pancreatic cancer or multifocal pancreatic tumors. Furthermore, in patients with concomitant vascular resections, TP may be performed to avoid the risk of pancreatic fistula and thus improve the perioperative risk profile. Aims To enable conclusive comparisons with other types of pancreatic resections and among different centers, international benchmark values are urgently needed. The aim of this international multicenter study was to identify the first benchmark values for TP from international expert centers. Methods This multicenter study analyzed outcomes from patients undergoing primary TP for malignant or benign lesions from 22 international expert centers. Patients were included from January 2017 up to November 2023 with a minimum follow-up of 1 year in each patient. Fifteen benchmark values were identified and were compared to a cohort of TP with vascular resection and the published multi-national benchmark values for pancreatoduodenectomy (PD). Benchmark patients were without significant comorbidities, age <80 years, cardiac disease, chronic pulmonary disease and renal failure. Results 553 (74%) of a total of 749 patients qualified as benchmark cases. The proportion of benchmark patients varied between 50%-100% per center. For example, benchmark cut-offs showed relevant 6 month- postoperative mortality (<6.8%) and major morbidity (≤44.2%). Especially benchmark values for pancreas specific morbidity such as delayed gastric emptying (≤41.2%) and postoperative hemorrhage (≤26.2%) were high. Benchmark cutoffs were further defined for operative time (≤510min), CCI at 6 months (≤27.3), and hospital stay (≤18 days). For ductal adenocarcinoma benchmark cutoffs for number of lymph nodes were ≥22 with an R0 resection rate of > 69.1%. Compared to the published PD, benchmark values for TP for overall- and pancreas specific complications were markedly higher. Furthermore, the cohort of patients with TP and concomitant vascular resections displayed several outcomes outside the TP benchmarks, for example a 90-mortality rate of 8% and major morbidity rate of 54%. Conclusion This benchmark analysis sets the first global reference values for TP, indicating significantly higher postoperative morbidity and mortality as compared to PD. The inferior outcomes were especially observed in TP with vascular resections. These reference values serve for quality control of pancreatic surgery in different centers, countries or surgical techniques.