Prosthetic joint infection (PJI) is a serious complication following joint replacement. Antiseptic solutions are often used for intraoperative wound irrigation particularly in cases of revision for PJI. Antiseptic irrigation is intended to eradicate residual bacteria which may be either free floating or in residual biofilm although there is no clear clinical efficacy for its use. Also, reviewing the scientific literature there is discordance in in vitro results where some studies questions antiseptic efficacy whilst others suggest that even at low concentration antiseptic agents are effective at eradicating bacterial biofilms.

The aim of this in vitro study was to establish the efficacy of undiluted antiseptic agents at eradication of a typical PJI forming biofilm and determine the importance of an antiseptic neutralisation step in this assessment.

Mature Staphylococcus epidermidis biofilms grown on TiAl6V4 discs were submerged in chlorohexidine (CHL) gluconate 4%, povidone-iodine (PI) 10% or phosphate-buffered saline (PBS) control solution. The discs were then rinsed, the biofilm bacteria suspended in solution using sonication and vortexing, and the viable count (CFU/ml) of the bacterial suspensions determined. The rinse/suspension solution was either (a) PBS or (b) Dey-Engley neutralization broth (NB).

When PBS was used to rinse/suspend the biofilm a highly significant, 7.5 and 4.1, mean log reduction in biofilm vitality was observed from the control, for CHL 4% and PI 10%, respectively. However, when NB was the rinse/suspension solution the apparent antiseptic biofilm eradication efficacy was replaced with a statistically significant but clinically irrelevant less the one log-reduction in biofilm vitality.

Clinical antiseptic agents are ineffective at eradicating S. epidermidis biofilm in an in vitro PJI model and absence of a neutralisation step gives the false impression of efficacy. Antiseptics alone are an ineffective treatment for biofilm related PJI and no substitute for meticulous debridement.

Prosthetic joint infection (PJI) is an important cause of arthroplasty failure. There is no method to disclose the presence or map the distribution of the in vivo biofilm on infected arthroplasty despite the recognition that such a tool would aid intraoperative decision making and improve novel implant design. The aim of this study was to test the efficacy of four dyes to disclose bacterial biofilm in an in vitro setting.

Four dyes with known affinity to bacterial biofilm were assessed to determine their efficacy to disclose biofilms in an in vitro model of PJI. Three dyes (Methylene Blue, Indocyanine Green and Rose Bengal) have established clinical utility and the other, Thioflavin T, is known to fluoresce in the presence of amyloid a known biofilm constituent. The efficacy of the dyes to discriminate between biofilms of different mass and vitality (high, low or the non-inoculated control) was determined after three minutes exposure of the biofilm to the dyes by calculating the amount of dye bound to the biofilm via sonication and spectrophotometry, quantification of the dye through standardised photographic imaging of the stained biofilm and the calculation of inter-observer agreement. Each experiment was performed in triplicate for each dye and repeated three times.

For each of the disclosure dyes assessed there was significant difference demonstrated between the amount of dye bound to the high and low mass biofilms (p<0.05) as well as in the amount of dye quantified in photographic and fluorescent image assessment between biofilms of differing mass (p<0.01). There was excellent agreement between three observers, for each disclosure dye, in determining the biofilm mass of each stained disc (Kappa>0.91).

This study demonstrates the efficacy of biofilm disclosure dyes in an in vitro PJI model which could one day be used to disclose and map the clinical biofilm in vivo.