Oral polyphosphate as non-antibiotic approach to prevent anastomotic leak
Author(s):
SANJIV HYOJU, University of Chicago; Robbin Klabbers, Radboud University Nijmegen,Netherlands; Melissa Arron, Radboud University Nijmegen,Netherlands; Monika Krezalek, University of chicago; Olga Zaborina, University of Chicago; John Alverdy, University of Chicago
Background: Anastomotic leak is the most feared and disabling complication following intestinal surgery. Previous work from our lab demonstrated that collagenolytic bacteria present on anastomotic tissues can contribute to the pathogenesis of anastomotic leak. We have demonstrated that phosphate is a universal “cue” that regulates bacterial virulence: phosphate deficiency can activate bacterial virulence whereas local phosphate abundance suppresses virulence.
Hypothesis: We hypothesize that inorganic hexamer polyphosphate (PPi-6) can deliver phosphate to the colon and prevent collagenolytic bacteria from causing an anastomotic leak.
Methods: We developed a mouse model of anastomotic leak using a high collagenase producing bacterium Serratia marcencens isolated from the stool of a critically ill patient. First, in vitro experiments were performed to determine if PPi-6 can suppress S. marcencens collagenase production. Next, groups of mice were randomly assigned to receive either 1% PPi-6 in the drinking water or water alone. Mice were sacrificed at postoperative day (POD) 10 and the anastomotic healing score was calculated (0=normal healing, 1=loose adhesions, 2=dense adhesions/inflammation, 3=abscess with dense adhesions/inflammation and 4=gross leak).
Results: In vitro studies demonstrated that 1% PPi-6 completely suppressed S. marcencens collagenase production without affecting its growth (n=15, p<0.001, 3 independent experiments). Mouse experiments demonstrated that mice drinking water only had a poor healing score of 3 ± 0.21 compared to a good healing score of 1 ± 0.21 for mice drinking PPi-6 (p<0.0001, t-test). Among the 10 control mice drinking water only, 20% developed anastomotic leakage (score=4) characterized by a visible anastomotic disruption and peritoneal contamination while 40% developed transmural abscesses with dense adhesions (score= 3). Mice drinking PPi-6 displayed significantly improved anastomotic healing with only 2 of 10 mice developing dense adhesions (score=2), while 6 mice had only mild adhesions (score=1). The remaining 2 mice demonstrated normal anastomotic healing (score=0). Histologic and microbiologic analyses are pending.
Conclusions: Oral polyphosphate protects mice from S. marcescens-induced anastomotic complications perhaps owing to its ability to locally suppress collagenase, an enzyme we previously showed is required for microbe-mediated anastomotic leak. Further work is ongoing to elucidate the mechanisms of PPi-6 protection against anastomotic leak.