Microbiome Dysbiosis with Enterococcus Presence in the Upper Gastrointestinal Tract is A Risk Factor for Mortality in Patients Undergoing Surgery for Pancreatic Cancer.

Abstract

Recent retrospective studies suggest a role for distinct microbiota in the perioperative morbidity and mortality of pancreatic head resections. We aimed to prospectively investigate the microbial colonization of critical operative sites of pancreatic head resections to identify microbial stratification factors for surgical and long-term oncologic outcomes. Prospective biomarker study applying 16S rRNA sequencing and microbial culturing to samples collected from various sites of the GI tract and surgical sites of patients during pancreatic head resections at a German single high-volume pancreatic center. A total of 101 patients were included (38 non-cancer, 63 cancer patients [50 PDAC patients]) in the study. In a first data analysis series, 16S rRNA sequencing data were utilized from 96 patients to assess associations of microbiome profiles with clinical parameters and outcomes. In general, microbiome composition varied according to sampling site, cancer, age or preoperative ERCP intervention, notably for the bile microbiome. In the PDAC subcohort, compositional variance of the bile or periampullary microbiome was significantly associated with postoperative complications such as ICU admission; on a taxonomic level we observed Enterococcus spp. to be significantly more abundant in patients developing deep or organ-space surgical site infections (SSI). Elevated Enterococcus relative abundances in the upper GI tract, in turn, were associated with 6-months mortality rates. In a second step, we focused on microbiological cultures collected from bile aspirates during surgery and investigated associations with perioperative complications and long-term survival. Notably, Enterococcus spp. were among the most prevalent pathobiont isolates observed in cancer patient bile specimens that were associated with severe SSIs, and thereby elevated mortality rates up to 24 months. Clinically relevant postoperative pancreatic fistulas or severe SSI were found as other major variables determining short-term mortality in this cancer patient cohort. In the context of adverse microbiological factors, a preoperative ERCP was also observed to segregate long-term survival, and it appeared to interact with the presence of Enterococcus spp. as highest mortality rates were observed in PDAC patients with both preoperative ERCP and presence of E. faecalis in bile aspirates. The presence of Enterococcus spp. in bile ducts of PDAC patients undergoing pancreatic surgery represents a significant risk factor for perioperative infections and, thereby, elevated postoperative and long-term mortality. This finding supports previous data on the use of the antibiotic drug piperacillin-tazobactam as appropriate perioperative antibiotic prophylaxis for preventing adverse outcomes after pancreatoduodenectomy.