An in vitro study assessing the effect of mesh morphology and suture fixation on bacterial adherence

Prosthetic infections, although relatively uncommon in hernia surgery, are a source of considerable morbidity and cost. The aims of this experimental study were to assess the influence of the morphological properties of the mesh on bacterial adherence in vitro. The morphological properties assessed were the polymer type, filament type, filament diameter, mesh weight, mean pore size, and the addition of silver chlorhexidine and titanium coatings. In addition, the study assessed the effect on bacterial adherence of adding a commonly used suture to the mesh and compared adherence rates to self-gripping mesh that does not require suture fixation.

Eight commercially sourced flat hernia meshes with different material characteristics were included in the study. These were Prolene(®) (Ethicon(®)), DualMesh(®) (Gore(®)), DualMesh(®) Plus (Gore(®)), Parietex™ ProGrip (Covidien™), TiMesh(®) Light (GfE Medical), Bard(®) Soft Mesh (Bard(®)), Vypro(®) (Ethicon(®)), and Omyra(®) (Braun(®)). Individual meshes were inoculated with Staphylococcus epidermidis and Staphylococcus aureus with a bacterial inoculum of 10(2) bacteria. To assess the effect of suture material on bacterial adhesion, a sterile piece of commonly used monofilament suture material (2.0 Prolene(®), ZB370 Ethicon(®)) was sutured to selected meshes (chosen to represent different commonly used polymers and/or the presence of an antibacterial coating). Inoculated meshes were incubated for 18 h in tryptone soy broth and then analysed using scanning electron microscopy. A previously validated method for enumeration of bacteria using automated stage movement electron microscopy was used for direct bacterial counting. The final fraction of the bacteria adherent to the mesh was compared between the meshes and for each morphological variable. One-way analysis of variance (ANOVA) was performed on the bacterial counts. Tukey's test was used to determine the difference between the different biomaterials in the event the ANOVA was significant.

Properties that significantly increased the mean bacterial adherence were the expanded polytetrafluoroethylene polymer (P < 0.001); multifilament meshes (P < 0.001); increased filament diameter (P < 0.001); increased mesh weight (P < 0.001); and smaller mean pore size (P < 0.001). In contrast, mesh coating with antibacterial silver chlorhexidine significantly reduced bacterial adhesion (S. epidermidis mean bacterial count 140.7 ± 19.1 SE with DualMesh(®) vs. 2.3 ± 1.2 SE with DualMesh(®) Plus, P < 0.001; S. aureus mean bacterial count 371.7 ± 22.7 SE with DualMesh(®) vs. 19.3 ± 4.7 SE with DualMesh(®) Plus, P = 0.002). The addition of 2.0 Prolene suture material significantly increased the mean number of adherent bacteria independent of the mesh polymer or mesh coating (P = 0.04 to <0.001).

The present study demonstrates the significant influence of the prosthetic load on bacterial adherence. In patients at increased risk of infection, low prosthetic load materials, i.e., lightweight meshes with large pores, may be beneficial. Furthermore self-fixing meshes, which avoid increasing the prosthetic load and antibacterial impregnated meshes, may have an advantage in this setting.