An in vitro analysis of microbial transmission during EUS-guided FNA and the utility of sterilization agents

Background: The risk of infection and potential microbial transmission with EUS-guided FNA (EUS-FNA) of cystic lesions remains unknown. Objective: We developed an in vitro model to study the incidence of transmucosal microbial transmission during EUS-FNA of cystic lesions and to evaluate the in vitro efficacy of bacteriocidal agent washings of mucosa before FNA under experimental conditions. Design: Conical tubes, 15 mL, filled with aerobic blood culture bottle media were prepared. Then sterile sections of bovine tripe were fastened over the top of the conical tubes in a sterile fashion (conical tube-tripe unit). FNA was performed with 22-gauge FNA needles. A series of 6 experiments were performed. Ten conical tube-tripe units underwent FNA once through the tripe into the blood culture media to ensure sterility. The surface of 10 conical tube-tripe units were inoculated with 50 mu L of a 1.5 x 10(8) 1:1 mixture of Escbericbia coli (E coli) and Enterococcus sp, and FNA was performed one time into the blood culture media to ensure contamination (controls). The surface of 40 conical tube-tripe units were inoculated with 50 mu L of a 1.5 x 108 1:1 mixture of E coli and Enterococcus sp Each of 4 sets of 10 conical tube-tripe units underwent experimental scenarios that consisted of washings with either I mL of 0.5% povidone iodine, chlorhexidine, absolute ethanol, or sterile water. FNA was performed once through the tripe into the blood culture media after washing the surface of the tripe. After each conical tube-tripe unit under-went FNA one time, 1 ml, blood culture media was obtained and mixed on pour plate agar media and was incubated along with the conical tubes. Microbial evaluation of the conical tubes that contained the blood culture media and pour plates was performed after 48 hours of incubation. Setting: Gastroenterology and Microbiology Departments of Scott White Memorial Hospital and Clinic in Temple, Texas. Interventions: EUS-FNA of cystic lesions. Main Outcome Measurements: Microbial contamination during EUS-FNA of an in vitro cystic environment. Results: A control without E coli and Enterococcus sp was with 0% contamination. A control group with E coli and Enterococcus sp was with 100% contamination; sterile water washings, 100% contamination (P = 1.00); iodine washings, 20% contamination (P <.001); chlorhexidine washings, 80% contamination (P =.47); and absolute ethanol washings, 90% contamination (P = 1.00). Results were compared with our control group by statistical tests of proportions by using the Fisher exact test. Conclusions: EUS-FNA of sterile cystic lesions resulted in transmucosal microbial contamination. However, our model demonstrated that iodine sterilization of a contaminated mucosal surface produced a very highly statistically significant (P <.001) reduction in the transmission of infectious agents into a sterile environment. This in vitro model could translate into clinical practice by providing evidence that microbial transmission by FNA occurred. The utility of povidone iodine washings could alter procedure methods and patient care.