In Vivo Fluorescence Imaging of Staphylococcus Aureus Biologic Mesh Infection – a 30 Day Analysis

BACKGROUND: Infection in the setting of hernia repair poses a difficult challenge, and biologic mesh might perform favorably in this setting. The specific response of different biologic mesh constructs to infection is unclear. Using fluorescent-labeled bacteria and innovative in vivo imaging quantitative bacterial analysis, we evaluated the response of two biologic mesh prosthetics to staphylococcus aureus infection.

METHODS: Twenty-four rats underwent creation of a chronic hernia. They were randomly assigned to underlay bridge repair with Permacol (cross linked porcine dermis) (n=12) or Surgisis (noncrosslinked porcine submucosa) (n=12). Half in each group were repaired in sterile fashion (clean cases; n=6) and half were inoculated with 104 CFU/ml of a clinical strain of green fluorescent protein (GFP) labeledStaphylococcus aureus (SA) (clean contaminated cases; n=6). Animals were allowed to survive 30-days, euthanized and the explanted abdominal wall underwent immediate in vivo fluorescent imaging. A 6mm punch biopsy of both tissue and mesh was obtained after imaging. This was followed by homogenizing, serial diluting, plating in blood agar dishes, incubation for 24 hours at 37° C and CFU/gm estimates. Validation of our imaging was performed by in vitro fluorescence imaging of serial concentrations of GFP SA (103 – 108 CFU/ml) to determine the minimal level of GFP signal detection. Fisher’s exact statistic was performed.

RESULTS: All animals survived 30 days following repair. CLEAN CASES: Culture results for both clean groups did not reveal evidence of infection (0 CFU/gm). CLEAN CONTAMINATED CASES: One of six (17%) Permacol animals had residual GFP SA infection (8 x 102 CFU/gm; n=1) while the remainder had no GFP SA (0 CFU/gm; n=5). Three of six (50%) Surgisis animals had residual GFP SA infection (1.3 x 102 CFU/gm; n=3) while the others had no detectable GFP SA infection (p-value = 0.55). IMAGING: In vivo imaging corroborated our low bacterial counts and was unable to detect presence of GFP SA or a difference in signal between clean contaminated cases of Permacol or Surgisis (0 RFU signal/group; n=12). IN VITRO VALIDATION: There was a positive linear correlation (R2 = 0.9121; Y=715X+186) between imaging device signal and bacterial counts above a concentration of 105 CFU/ml. This was determined to be the lower limit of GFP SA detection with our in vivo fluorescence-imaging device.

CONCLUSIONS: This study confirms that biologic mesh can tolerate clean contaminated situations. In these settings most animals were able to clear the bacterial contamination as demonstrated by low quantitative cultures and corroborated with in vivo fluorescent imaging. Further analysis as to the exact mechanism of bacterial clearance and biomechanical alterations of the mesh in infected fields is ongoing.

Session: Podium Presentation

Program Number: S065

« Return to SAGES 2010 abstract archive