In vitro and in vivo evaluation of the antimicrobial effectiveness of non-medicated hydrophobic wound dressings.

There is an increasing use of non-medicated wound dressing with claims of irreversible bacterial binding. Most of the data are from in vitro models which lack clinical relevance. This study employed a range of in vitro experiments to address this gap and we complemented our experimental designs with in vivo observations using dressings obtained from patients with diabetes-related foot ulcers. A hydrophobic wound dressing was compared with a control silicone dressing in vitro. Test dressings were placed on top of a Pseudomonas aeruginosa challenge suspension with increasing concentrations of suspension inoculum in addition to supplementation with phosphate buffered saline (PBS) or increased protein content (IPC). Next, we used the challenge suspensions obtained at the end of the first experiment, where bacterial loads from the suspensions were enumerated following test dressing exposure. Further, the time-dependent bacterial attachment was investigated over 1 and 24 h. Lastly, test dressings were exposed to a challenge suspension with IPC, with or without the addition of the bacteriostatic agent Deferiprone to assess the impacts of limiting bacterial growth in the experimental design. Lastly, two different wound dressings with claims of bacterial binding were obtained from patients with chronic diabetes-related foot ulcers after 72 h of application and observed using scanning electron microscope (SEM). Bacteria were enumerated from each dressing after a 1-h exposure time. There was no statistical difference in bacterial attachment between both test dressings when using different suspension inoculum concentrations or test mediums. Bacterial attachment to the two test dressings was significantly lower (p < 0.0001) when IPC was used instead of PBS. In the challenge suspension with PBS, only the hydrophobic dressing achieved a statistically significant reduction in bacterial loads (0.5 ± 0.05 log colony forming units; p = 0.001). In the presence of IPC, there was no significant reduction in bacterial loads for either test dressing. When bacterial growth was arrested, attachment to the test dressings did not increase over time, suggesting that the number of bacteria on the test dressings increases over time due to bacterial growth. SEM identified widespread adsorption of host fouling across the test dressings which occurred prior to microbial binding. Therein, microbial attachment occurred predominantly to host fouling and not directly to the dressings. Bacterial binding is not unique to dialkylcarbamoyl chloride (DACC) dressings and under clinically relevant in vitro conditions and in vivo observations, we demonstrate (in addition to previously published work) that the bacterial binding capabilities are not effective at reducing the number of bacteria in laboratory models or human wounds.