Abstract
Abstract
Purpose
Postoperative pancreatic fistulas are among the most serious complications of pancreatic surgery. Biliary fistulas, though usually less severe, can occur after pancreaticoduodenectomy and, when combined with pancreatic fistulas, increase morbidity and mortality. This study is the first to investigate, whether bile enhances the harmful effects of pancreatic fluid on sealant integrity in a short-term
ex vivo
model using a fibrin-based sealant, thereby contributing to fistula formation.
Methods
We developed and validated an improved ex-vivo setup to simulate the conditions during pancreatic surgery: the physiological incubation bio-simulator 2 (PIBS 2). A defined defect in porcine pleura was sealed with the fibrin sealant TachoSil
®
and exposed to various fluids from human origin (pancreatic fluid, bile fluid and a combination of both) for 60 min in a physiological environment. Electrolyte solution served as control. We evaluated the overall leakage rates and time to leakage. Results were plotted as Kaplan-Meier survival curves and analyzed using the log-rank test (Mantel-Cox) and Fisher’s exact test.
Results
PIBS 2 achieved stable defect closure for the electrolyte solution. Pancreatic fluid demonstrated the highest leakage rates (75%) and significantly impaired sealant stability (
p
p
= 0.63). When combined, bile and pancreatic fluid resulted in a significantly lower leakage rate compared to pancreatic fluid alone (47% vs. 75%,
p
= 0.041; odds ratio 3.31).
Conclusion
This study replicates previous results, showing that pancreatic fluid is highly aggressive towards fibrin-based sealants. In contrast to clinical findings, the addition of bile did not significantly increase this effect in this ex-vivo model, suggesting that morbidity in patients with combined fistulas may not be primarily caused by increased aggressiveness of pancreatic fluid after its combination with bile.