PURPOSE: Clinically relevant postoperative pancreatic fistulas (CR-POPF) occurring after distal pancreatectomy often cause intra-abdominal infections. We monitored the presence of bacterial contamination in the ascitic fluid after distal pancreatectomy to clarify the bacterial origin of intra-abdominal infections associated with CR-POPF. METHODS: In 176 patients who underwent distal pancreatectomy, ascitic fluid bacterial cultures were performed on postoperative days (POD) 1-4 and when the drainage fluid became turbid. The association between postoperative ascitic bacterial contamination and CR-POPF incidence was investigated. RESULTS: CR-POPF occurred in 18 cases (10.2%). Among the patients with CR-POPF, bacterial contamination was detected in 0% on POD 1, in 38.9% on POD 4, and in 72.2% on the day (median, day 9.5) when the drainage fluid became turbid. A univariate analysis revealed a significant difference in ascitic bacterial contamination on POD 4 (p < 0.001) and amylase level on POD 3-4 (p < 0.001). A multivariate analysis revealed the amylase level and ascitic bacterial contamination on POD 4 to be independent risk factors. CONCLUSIONS: In the CR-POPF group, ascitic bacterial contamination was not observed in the early postoperative stage, but the bacterial contamination rate increased after pancreatic juice leakage occurred. Therefore, CR-POPF-related infections in distal pancreatectomy may be caused by a retrograde infection of pancreatic juice.
PURPOSE: Clinically relevant postoperative pancreatic fistulas (CR-POPF) occurring after distal pancreatectomy often cause intra-abdominal infections. We monitored the presence of bacterial contamination in the ascitic fluid after distal pancreatectomy to clarify the bacterial origin of intra-abdominal infections associated with CR-POPF. METHODS: In 176 patients who underwent distal pancreatectomy, ascitic fluid bacterial cultures were performed on postoperative days (POD) 1-4 and when the drainage fluid became turbid. The association between postoperative ascitic bacterial contamination and CR-POPF incidence was investigated. RESULTS: CR-POPF occurred in 18 cases (10.2%). Among the patients with CR-POPF, bacterial contamination was detected in 0% on POD 1, in 38.9% on POD 4, and in 72.2% on the day (median, day 9.5) when the drainage fluid became turbid. A univariate analysis revealed a significant difference in ascitic bacterial contamination on POD 4 (p < 0.001) and amylase level on POD 3-4 (p < 0.001). A multivariate analysis revealed the amylase level and ascitic bacterial contamination on POD 4 to be independent risk factors. CONCLUSIONS: In the CR-POPF group, ascitic bacterial contamination was not observed in the early postoperative stage, but the bacterial contamination rate increased after pancreatic juice leakage occurred. Therefore, CR-POPF-related infections in distal pancreatectomy may be caused by a retrograde infection of pancreatic juice.