Background: NOTES via the gastrointestinal tract raises the specter of intra-peritoneal infection. Various anti-microbial techniques have been employed in animal and human survival studies, including saline lavage, intravenous and topical antibiotics, and povidone-iodine, although there is a paucity of data regarding their general effectiveness.
Aim: To assess the effectiveness of existing sterilization techniques for NOTES by quantifying and speciating colony-forming units (CFUs) before and after treatment.
Design: Ex vivo animal studies; bacteriological study.
Methods: Stomachs and distal colons were harvested en bloc from ten fasted adult white pigs following euthanasia. Half received cefazolin 1 g intravenously prior to killing. Multiple tissue samples were obtained from each resected organ. Each tissue sample was then assigned to one of five treatment arms: (1) normal saline, (2) Betadine, (3) cefazolin/metronidazole suspension, (4) chlorhexidine, (5) no treatment. Fifteen samples were used per arm. After treatment, the mucosal surface of each sample was swabbed and inoculated in normal saline, followed by serial dilutions, which were then plated onto sheep's blood agar plates and incubated under aerobic and anaerobic conditions. CFUs were quantified and speciated.
Results: Median bacterial density was estimated to be 8.0 × 10(5) CFUs/ml (stomach) and 1.9 × 10(6) CFUs/ml (colon). The predominant organisms were Escherichia coli (stomach) and both E. coli and Enterococcus sp. (colon). Saline and antibiotic suspension lavages caused a 1-log reduction in stomach and colon. Betadine/chlorhexidine lavage resulted in a 4-log reduction. Intravenous antibiotics alone resulted in a 4-log reduction. Combining intravenous antibiotics and either Betadine or chlorhexidine decreased counts to the 10(1) level. By Kruskal-Wallis method, differences were statistically significant (p = 0.001).
Conclusions: The use of intravenous antibiotics in addition to topical Betadine or chlorhexidine effectively reduced microbial burden in both gastric and colonic mucosa in this porcine model to the 10(1) level.