Unpredictable Effects of Rifampin as an Adjunctive Agent in Elimination of Rifampin-Susceptible and -Resistant Staphylococcus aureus Strains Grown in Biofilms

The use of rifampin as an adjunct in biofilm-associated infections is based on the ability to penetrate into biofilms and a presumed activity against dormant bacteria. Yet, its efficacy remains contradictory, and rifampin-resistant strains frequently emerge during therapy. Therefore, the efficacy against rifampin-susceptible and isogenic rifampin-resistant methicillin-susceptible Staphylococcus aureus (MSSA) strains was evaluated. Biofilms were generated under static conditions using MSSA with various genetic backgrounds. Oxacillin alone or with rifampin at various concentrations was subsequently added, and after 24 h biomass and viable cell counts were determined. Upon rifampin addition, interstrain variations in viable count change, ranging from a tendency toward antagonism to synergy, were observed among all strains tested, irrespective of the genetic background of the strain. Similar variations were observed in changes in biomass. The decrease in viable count upon rifampin addition was negatively correlated to formation of large amounts of biomass, since strains embedded by more biomass showed a diminished reduction in viable count. Rifampin (1 μg/ml) as adjunct to oxacillin achieved greater reductions in biomass produced by most rifampin-susceptible isolates, ranging from 17 to 54%, compared to 4% for oxacillin alone. In contrast, rifampin had no additional value in reduction of biomass of isogenic rifampin-resistant mutants. At subinhibitory concentrations of rifampin (0.008 μg/ml), none of the strains tested yielded an extra reduction in biomass that was ≥40%. In conclusion, the effects of rifampin as adjunct on biomass and viable count were unpredictable, and the use of rifampin against biofilm containing rifampin-resistant strains seems unwarranted.Bacteria embedded in biofilm are often hard to eradicate, which will often result in failure of antibiotic therapy. A biofilm can appear and persist on implanted or indwelling devices. Adherence to bone or natural tissue can also occur. As a consequence of biofilm formation, foreign body infections are a risk factor for persistent bacteremia (12).Staphylococcus aureus biofilm-infected devices often require removal in combination with antimicrobial therapy (6). Even if an infection seems to be cured, a subset of bacteria can survive within the partly unremoved biofilm and the infection can remanifest after several years. These are the so-called low-grade infections (30). The ideal antibiotic should be able to diffuse properly into biofilms, eradicate dormant bacteria, and kill phagocytized S. aureus, since foreign body infections are associated with intracellular S. aureus (16, 18). Rifampin meets these criteria partially (28). Rifampin penetrates biofilms adequately (38) and improves the diffusion of a companion agent through the biofilm (8). It also reduces the adherence of bacteria to surfaces (28). Furthermore, rifampin addition eliminates intrinsic rifampin-resistant Pseudomonas aeruginosa (MIC, 32 μg/ml) from biofilms (11). Therefore, it could be hypothesized that the presence of rifampin leads to biofilm matrix disintegration. However, rifampin cannot be used alone due to the rapid emergence of resistance (37), and it is not as effective as presumed against slow-growing bacteria (21, 38). An alternative approach for S. aureus biofilm-associated infections is the addition of rifampin to a companion drug. In clinical practice, rifampin addition has been occasionally demonstrated as beneficial in terms of clinical or bacteriological cure rates (39), especially for prosthetic device-related infections (10, 18, 22). However, its effect remains poorly defined, since the number of supporting human studies is limited and most of them have been underpowered (10, 18, 22, 29). Moreover, recently performed in vitro studies have demonstrated antagonistic effects after addition of rifampin in experimental foreign body infection and endocarditis models (15, 17).Exposure to combination therapy including rifampin frequently results in the emergence of rifampin-resistant subpopulations, which might hinder clearance of the infection (25). This is probably due to the fact that the pharmacokinetics of the companion drug were different from that of rifampin. Preferably, the companion drug needs to penetrate easily through diverse layers of a biofilm to prevent the development of rifampin resistance. But, although some antibiotics penetrate adequately through biofilms, bacteria may be protected from killing by thickening of the cell wall (38). A high bacterial load is another cause of rapid resistance development (33). To overcome the emergence of resistance it is recommended that rifampin be initiated after the companion drug has been administered for 2 to 5 days, to reduce the high density of the initial burden (1), or at least until a substantial reduction in bacterial load has been achieved (4, 25).Since interstrain variability in response to rifampin addition might be responsible for the contradictory results obtained by various authors, the aim of the present study was to elucidate whether this variability does exist for strains between and within the S. aureus genetic backgrounds associated with multilocus sequence types (MLSTs) CC5, CC8, and CC30. Furthermore, since rifampin-resistant mutants emerge during therapy and rifampin is used for hypothesized destruction of the biofilm matrix, we evaluated whether continuation of exposure to rifampin is useful against biofilms containing rifampin-resistant methicillin-susceptible S. aureus (MSSA). In addition, we examined whether a potential effect was also present at a subinhibitory concentration.