Combination of Tigecycline and N-Acetylcysteine Reduces Biofilm-Embedded Bacteria on Vascular Catheters

Abstract

To assess the efficacy of an antibiofilm/antimicrobial agent combination, we incubated catheter segments colonized with one of six studied bacterial organisms in N-acetylcysteine, tigecycline, N-acetylcysteine-tigecycline, or saline. Segments were washed, sonicated, and cultured. N-acetylcysteine-tigecycline significantly decreased all viable biofilm-associated bacteria and was synergistic for methicillin-resistant Staphylococcus aureus and Staphylococcus epidermidis.

Vascular catheter-associated bacteremia has a substantial impact on morbidity, mortality, duration of stay, and overall cost of health care (4, 16, 23). Infections due to biofilm-embedded bacteria are difficult to eradicate without the removal of the infected device (10, 12). However, other access sites may not be available and patients may not wish to undergo further procedures. In instances where removal of an infected tunneled vascular catheter is not feasible or desirable, a trial of an antibiotic lock solution (plus systemic antibiotics) can be considered for the treatment of uncomplicated bacteremia (12). Since this practice has not been uniformly successful (7, 17, 19), there is a need to assess other strategies, such as the use of antibiofilm/antimicrobial catheter lock solution, to salvage infected vascular catheters.

N-acetylcysteine (NAC) decreases biofilm formation by a variety of bacteria (9, 15, 20) and reduces the production of extracellular polysaccharide matrix (14) while promoting the disruption of mature biofilm (9, 20). NAC is widely used in medical practice via inhalation and oral and intravenous routes (11, 13, 27), and it has an excellent safety profile (6). Tigecycline is active against a range of multiresistant organisms and is bactericidal against biofilm-associated Staphylococcus epidermidis at a lower minimal bactericidal concentration than that of vancomycin and daptomycin (8).

We hypothesized that the combination of NAC and tigecycline is synergistic in the treatment of catheter-associated biofilm as they both act on different components of the biofilm.

We used six microorganisms that had been isolated from patients with catheter-associated bacteremia. These were methicillin-resistant Staphylococcus aureus (MRSA), methicillin-sensitive S. aureus (MSSA), methicillin-resistant S. epidermidis (MRSE), vancomycin-resistant enterococcus (VRE), Klebsiella pneumoniae, and Acinetobacter baumannii. We purchased 20% N-acetylcysteine (Hospira, Inc., Lake Forest, IL) and lyophilized tigecycline (Wyeth Pharmaceuticals, Inc., Philadelphia, PA). We used 80 mg/ml of NAC based on preliminary in vitro data that showed a dose-response relationship on planktonic bacteria (unpublished data) and 1 mg/ml of tigecycline (this dose is 1,000-fold higher than its MIC for the organisms tested in the planktonic phase).

Four-centimeter segments of 7-French, triple-lumen, central venous catheters (Cook, Inc., Bloomington, IN) were incubated in bacterial suspensions that contained 10⁵ CFU/ml of bacteria in Trypticase soy broth to allow biofilm formation. After incubation at 37°C for 24 h, segments were removed and excess broth was shaken off. One catheter segment was rinsed and cultured to obtain a baseline value, and the remaining segments were suspended for 4 or 12 h at 37°C in one of the following treatment solutions: N-acetylcysteine, tigecycline, NAC-tigecycline combination, or normal saline (NS) as control.

Catheter segments were rinsed three times with NS to remove planktonic bacteria. The distal and proximal 0.5-cm ends were cut, and the remaining segment was divided into three 1-cm sections. These sections were individually sonicated for 10 min and vortexed for 30 s in 1 ml of phosphate-buffered saline. One-hundred-microliter aliquots of the original sonication fluid and successive dilutions were inoculated onto blood agar plates (limit of detection, 10 CFU). Bacterial colonies were counted after incubation for 24 h. The median colony count of the three 1-cm sections from the same 4-cm segment was considered the representative value for that segment. Each set of experiments was repeated five times. The CFU per centimeter values of catheters for different groups were compared by the Mann-Whitney U test (STATA, version 9; StataCorp, College Station, TX), and a P value of <0.05 indicated significance.

We recovered 10⁴ to 10⁵ CFU/cm of bacteria from catheter-associated biofilm at 24 h for all organisms except for VRE, which consistently had 10³ CFU/cm (Fig. 1). NAC had an independent antimicrobial effect on biofilm-associated MRSA, MRSE, and K. pneumoniae (P was <0.01 compared to the control at 12 h). For other organisms, the effect of NAC alone was similar to that of the control. Tigecycline alone significantly reduced the CFU of all tested bacteria, except for VRE, compared to the control (P < 0.05).

FIG. 1.

The effect of normal saline control, NAC, tigecycline, and NAC-tigecycline combination on bacterial colony counts of different organisms. Each bar represents mean CFU per centimeter of catheter at different time points along with respective standard errors of the means. The y axis is on a logarithmic scale. The horizontal lines denote the lower limit of detection (LLD) of 10 CFU/cm. *, P < 0.05 compared to the control at the same time. §, P < 0.05 compared to the combination at the same time. (Microsoft Excel was used to create the graph.)

The NAC-tigecycline combination consistently decreased viable biofilm-associated bacteria relative to the control. This combination was synergistic for MRSA at 4 h (P < 0.01) and MRSE at 12 h (P < 0.05). We were unable to detect any bacteria in the cases of MRSA, MRSE, MSSA, and K. pneumoniae after incubation with the combination.

Utilizing high doses of antimicrobials to eradicate biofilm has had limited success in the clinical setting (7, 19). This may be due to inadequate penetration of antibiotics, higher MICs of antibiotics for biofilm-associated bacteria, reduced growth rates, and local alterations in the biofilm environment that impair the activity of antibiotics (2, 3, 24). Since the antimicrobial susceptibility of biofilm-associated S. epidermidis, MRSA, and MSSA is enhanced in disrupted biofilm (5), it is conceivable that an antibiofilm/antimicrobial agent combination would be synergistic. By degrading the extracellular polysaccharide matrix of biofilm (9, 14), it is possible that NAC may have made the biofilm-associated bacteria more susceptible to tigecycline, although we did not specifically test this hypothesis. Our NAC-tigecycline combination was synergistic for S. aureus and S. epidermidis, the two organisms that are most commonly associated with vascular catheter-related infections (4, 17).

A similar 4- to-5 log₁₀ decrease in viable cells dislodged from catheters has been described after incubation for 24 h with other lock solutions, such as minocycline-rifampin, ciprofloxacin-rifampin (22), taurolidine-citrate (21), and in some cases, minocycline-EDTA (18). The NAC-tigecycline combination compared favorably with vancomycin and linezolid, which led to decreases of only 1 log₁₀ and 3 log₁₀, respectively, in biofilm-embedded S. epidermidis after incubation for 24 h (1), of 2 log₁₀ in S. aureus after incubation for 24 h with each drug (26), and of 2 log₁₀ in VRE after incubation for 24 h with linezolid (26). Although most previous studies evaluating catheter lock solutions have used a continuous dwell time of anywhere from 24 to 336 h to achieve sterilization of the biofilm (1, 18, 25), we used a shorter dwell time of 12 h, which is more clinically feasible.

The promising in vitro results prompted us to initiate a single-arm pilot clinical trial evaluating the use of the NAC-tigecycline combination as a catheter lock solution for the treatment of catheter-associated bacteremia arising from tunneled hemodialysis catheters. Should the clinical trial demonstrate the efficacy and safety of this innovative catheter lock solution, we plan to explore the use of the NAC-tigecycline combination in a randomized fashion.