The Activity of Antimicrobial Surfaces Varies by Testing Protocol Utilized

Background Contaminated hospital surfaces are an important source of nosocomial infections. A major obstacle in marketing antimicrobial surfaces is a lack of efficacy data based on standardized testing protocols. Aim We compared the efficacy of multiple testing protocols against several “antimicrobial” film surfaces. Methods Four clinical isolates were used: one Escherichia coli , one Klebsiella pneumoniae , and two Staphylococcus aureus strains. Two industry methods (modified ISO 22196 and ASTM E2149), a “dried droplet”, and a “transfer” method were tested against two commercially available antimicrobial films, one film in development, an untreated control, and a positive (silver) control film. At 2 (only ISO) and 24 hours following inoculation, bacteria were collected from film surfaces and enumerated. Results Compared to untreated films in all protocols, there were no significant differences in recovery on either commercial brand at 2 or 24 hours after inoculation. The silver surface demonstrated significant microbicidal activity (mean loss 4.9 Log 10 CFU/ml) in all methods and time points with the exception of 2 hours in the ISO protocol and the transfer method. Using our novel droplet method, no differences between placebo and active surfaces were detected. The surface in development demonstrated variable activity depending on method, organism, and time point. The ISO demonstrated minimal activity at 2 hours but significant activity at 24 hours (mean 4.5 Log 10 CFU/ml difference versus placebo). The ASTEM protocol exhibited significant differences in recovery of staphylococci (mean 5 Log 10 CFU/ml) but not Gram-negative isolates (10 fold decrease). Minimal activity was observed with this film in the transfer method. Conclusions Varying results between protocols suggested that efficacy of antimicrobial surfaces cannot be easily and reproducibly compared. Clinical use should be considered and further development of representative methods is needed.