Initial Inoculation Concentration Does Not Affect Final Bacterial Colonization of In vitro Vascular Conduits

Background: Despite improved peri-operative care, prosthetic graft infections continue to cause substantial morbidity and mortality. Contemporary graft infection models have tested a conduit's infectability using varying concentrations without standardization. Using a static assay in vitro model, we sought to evaluate the impact of inoculation concentration on vascular conduit attachment. Methods: The 2-hour and 24-hour attachment of Staphylococcus aureus TCH1516 and Pseudomonas aeruginosa PA01-UW were determined on polytetrafluoroethylene (PTFE), Dacron ® , nitinol, cobalt chromium, and Viabahn ® (W.L. Gore and Associates, Newark, DE) endoprotheses. Individually and in combination, concentrations at 10 4 , 10 5 , 10 6 , 10 7 , and 10 8 were tested on 2-mm sections of each graft. After each time interval, the prosthetics were rinsed to remove non-attached bacteria, sonicated to release the attached bacteria, spiral plated, and then analyzed for the attached concentration. Results: After two hours, the higher initial inoculation concentration translated into a higher attachment percentage, but the mean attachment percentage was only 14.8% in the 10 8 group. Pseudomonas aeruginosa had the greatest mean attachment across all material and concentration groups. The sequence of attachment on the conduits followed a constant order: Dacron, PTFE, cobalt, nitinol, and Viabahn with no difference between Dacron and PTFE. Although there were still differences at the 24-hour mark, the median attachment at each concentration was greater than the highest initial concentration (10 8 ). Conclusions: Initial attachment percentage is poor consistently regardless of inoculation concentration, however, S taphylococcus aureus and Pseudomonas aeruginosa are still able to achieve full attachment after 24 hours. A concentration of less than 10 7 should be used in vascular graft infection models to ensure adequate bacterial attachment.