Use of In Vivo-Generated Biofilms from Hemodialysis Catheters To Test the Efficacy of a Novel Antimicrobial Catheter Lock for Biofilm Eradication In Vitro

This biofilm study was conducted to assess the in vitro activityof tetrasodium EDTA on catheters that had been routinely removedfrom hemodialysis patients at Leeds Teaching Hospitals Trustdue to maturation of fistula. Catheters were screened by cultureof through-catheter flush, and isolates were identified by standardmethodologies; 20 isolates were found to be biofilm positive.Initial biofilm cell count levels averaged above 10⁵ CFU/1-cmcatheter section. Bacteria identified in the biofilms were gram-negative(1 isolate), gram-positive (11 isolates), or mixed species (8isolates). After a 24-h lock, 40 mg of tetrasodium EDTA perml was effective at eradicating the total biofilm viable countin almost all cases. The efficacy of tetrasodium EDTA as a catheterlock potentially shows that this agent could substantially reducecatheter-related infections and be used to treat patients withlimited access.

INTRODUCTION

Top

Introduction

Catheters have become essential in the management of critical-carepatients, yet the inside of a catheter is a major source ofinfection due to biofilms (1, 3). It has been found that atleast 70% of all nosocomial bloodstream infections occur inpatients with central venous catheters (7). Traditionally, cathetershave been locked with normal saline or heparin solutions, sometimesin combination, to provide anticoagulant activity. Normal salineis generally used to lock short-term peripheral intravenouscatheters, but saline has no anticoagulant or antimicrobialactivity. Heparin solutions are generally used to lock vascularcatheters. Heparin has anticoagulant activity, but it does notfunction as an antimicrobial and therefore does not provideprotection from catheter-related infections. Recently, antibioticlock solutions, with or without an anticoagulant (such as vancomycin-heparin),have been used for the prevention and management of catheter-relatedbacteremia (2, 8, 15). However, there are concerns, particularlywith the use of a vancomycin flush solution, because of thepotential for the development of resistant organisms (5, 9).

EDTA is a calcium and iron chelator with anticoagulant activity.However, as a disodium salt it has been documented to have limitedantistaphylococcal and anti-Candida activities (6, 12, 13).In contrast to this was a study by Raad et al., who demonstratedthat minocycline and EDTA have highly synergistic activitiesin the decontamination of catheter surfaces when they are combinedin a solution (10). It was found in that study that minocycline-EDTAwas active against staphylococci, gram-negative bacilli, andCandida organisms that colonize polymers. This suggests a possiblerole of EDTA in catheter lock procedures.

We undertook the present study by using in vivo renal linesin an in vitro technique to test the efficacy of tetrasodiumEDTA on catheters removed from patients in order to determineits potential as a catheter lock solution to treat preformedbiofilms and, ultimately, catheter-related infections.

MATERIALS AND METHODS

Top

Materials and Methods

Experimental design. The study was conducted at the Leeds Teaching Hospitals Trust,Leeds, United Kingdom. Renal dual-lumen circle C advantage hemodialysiscatheters (Horizon Medical Products, Manchester, United Kingdom),collected after routine removal from patients, were evaluatedfor biofilm prevalence. Twenty catheters were found to be positivefor evidence of biofilms, and these were utilized for evaluatingthe effectiveness of tetrasodium EDTA on biofilms.

Screening protocol. The inner lumens of all catheters, aseptically removed fromthe patient, were slowly flushed with 1 ml of sterile phosphate-buffedsaline to remove the high numbers of planktonic bacteria andconfirm the presence of biofilm. Duplicate 10-µl aliquotsof the flushed lumen samples were spread plated onto fresh bloodagar plates (Oxoid, Basingstoke, United Kingdom) and incubatedat 37°C for 48 h. The samples were vortexed for 1 min priorto plating out. After flushing (screening), all catheters weresterilized on the outer surface with alcohol wipes and thenwere stored at 4°C, in sterile specimen bags, until theresults of the screen were available. All external surfacesof catheters were found to be negative for the presence of contaminatingbacteria after cleaning and sterilization. Sections of the innerlumens of catheters were viewed under scanning electron microscopy(SEM) to confirm biofilm presence. Noncolonized catheter lineswere discarded. All isolated cultures were identified by usingGram staining, microscopy, catalase, coagulase, methicillinstrips, and API 20E and API 20NE kits (BioMerieux, Marcy l'Etoile,France).

Lock protocol. Prior to a lock treatment test with 40 mg of tetrasodium EDTA(Sigma-Aldrich, Dorset, United Kingdom) per ml, the catheterswhich were positive for sessile bacteria were locked with 5ml of nutrient broth in each lumen, administered with a 5-mlsyringe, and incubated overnight at 37°C. Following incubation,each catheter lumen was slowly flushed with 5 ml of phosphate-bufferedsaline (as described above), and two 1-cm pieces were cut fromthe distal end with a sterile scalpel blade. One piece, neutralizedin 1 ml 1 of M calcium chloride (Sigma-Aldrich), was used forculture and bacterial identification, and the other cathetersection was stored in 4% formal saline prior to SEM analysis.The catheters were then locked with 40 mg of tetrasodium EDTAlock fluid per ml for up to 24 h. After 3 h, two 1-cm sectionswere cut from each catheter, using a sterile scalpel blade,in a biological safety cabinet to avoid exogenous contamination.One catheter section was neutralized in 1 ml of 1 M calciumchloride for culturing, and the other catheter section was placedin 4% formal saline for SEM analysis. This same procedure wascarried out at 6 and 24 h.

Biofilm removal. Following treatment and the appropriate incubation periods,the catheter sections that were placed in 1 ml of 1 M calciumchloride were sonicated in a sonicating water bath (DP 200-00;Ultrawave, Cardiff, United Kingdom) for 15 min and plated outonto fresh blood agar for colony counts. This procedure wasrepeated for all lock times. Fifteen minutes of sonication wasfound to be the most effective time for maximum biofilm removal(data not shown).

After a 24-h lock, both lumens of the catheters were brushedwith an endoluminal brush (model RC4060/5; FAS Medical International)to confirm the absence of viable organisms. Each brush was addedto 1 ml of CaCl (1 M), sonicated in a water bath for 15 min,plated out onto fresh blood agar, and incubated overnight at37°C. SEM was used to confirm biofilm presence on all cathetersafter brushing.

SEM. Catheter sections were cut into small sections and fixed byplacing them in 4% formal saline overnight at room temperature.The samples were then dehydrated in a graded series of ethanolsolutions (20, 40, 60, and 80%) for 15 min each at room temperature,immersed in acetone for 4 h at room temperature, and dehydratedovernight by using a critical-point dryer. Finished specimenswere mounted on aluminum stubs with carbon paint, sputter coatedwith 25-nm-diameter gold particles (Polaron E 5300), and examinedwith a CamScan series III SEM (Obducat Camscan Ltd., Cambridge,United Kingdom).

RESULTS AND DISCUSSION

Top

Results and Discussion

Effectiveness of tetrasodium EDTA on monoculture biofilms. Biofilm, confirmed by viable cell counts and SEM, was clearlyevident on all catheters removed from patients. The biofilmsdetected had viable cell counts ranging from log 4 to log 6CFU per ml, as shown in Fig. 1. With these particular patients,the predominant species of bacteria isolated from the cathetersincluded methicillin-resistant Staphylococcus aureus (MRSA),coagulase-negative Staphylococcus, Enterobacter, and Enterococcusspp., as shown in Table 1. After 3 h, the effectiveness of tetrasodiumEDTA was found to significantly (P < 0.05) reduce the sessileviable population by over 30% in the majority of cases. After6 h, the biofilm had been reduced by over 80% of it originalviable cell count, and finally, after 24 h, tetrasodium EDTA,in the majority of cases, had effectively eradicated the biofilm.

View larger version (29K):
[in this window]
[in a new window]
FIG. 1. In vivo renal lines in an in vitro technique to test the efficacy of tetrasodium EDTA (40 mg/ml) in removing monospecies bacterial biofilm. Colony counts over time for catheters 1 to 12 (Table 1) are shown.

View this table:
[in this window]
[in a new window]
TABLE 1. Microbiology of biofilms removed from hemodialysis catheters

Effectiveness of tetrasodium EDTA on mixed-species biofilm. From Fig. 2 it is evident that the biofilm viable cell countranged from log 5 to log 6 CFU per ml in catheters removed frompatients. After a lock of 3 h, tetrasodium EDTA showed littleeffect on mixed-species biofilms, with a large increase in sessilecounts in the biofilm containing Pseudomonas and Streptococcus(Fig. 2; Table 1). However, after 6 h, the effect of tetrasodiumEDTA was clearly apparent, with a reduction of viable cell countsof over 50% of the original biofilm. After 6 h, the biofilmviable cell count decreased further, with no viable cell countin the majority of cases after a 24-h lock. Clearly, in allcases the action of tetrasodium EDTA had significantly (P <0.05) reduced the biofilm viable cell count.

View larger version (32K):
[in this window]
[in a new window]
FIG. 2. In vivo renal lines in an in vitro technique to test the efficacy of tetrasodium EDTA (40 mg/ml) in removing mixed-species bacterial biofilm. Colony counts over time for catheters 13 to 20 (Table 1) are shown.

Catheter-related infections due to biofilms remain a significantclinical problem in many institutions, and a lock solution,which can prevent or control the growth of biofilms in a catheterlumen, is imperative for the prevention of catheter-relatedbloodstream infections. Many studies have examined the preventionof endoluminal catheter-related infection based on catheterflushing, but these studies have shown conflicting results.For example Schwartz et al., using a solution of heparin andvancomycin to flush central venous catheters compared its efficacywith that of heparin (15). It was found in this study that dailyflushing with heparin and vancomycin significantly decreasedthe frequency of catheter-related bacteremia attributed to luminalcolonization.

Most catheters are colonized with organisms embedded in a biofilm,but infection appears to depend on whether the organisms onthe catheter surface, particularly those in a planktonic free-floatingphase, exceed a certain quantitative threshold. In the presentstudy, which involved 20 catheters removed from hospitalizedpatients, a concentration of 40 mg of tetrasodium EDTA per mlwas found to eradicate biofilm; this must significantly aidin the reduction of catheter-related infections. TetrasodiumEDTA was found to have a broad-spectrum inhibitory activityagainst MRSA and gram-negative bacilli.

Many studies have shown that a combination of minocycline andEDTA is uniquely useful in disrupting the biofilm and synergisticallyeradicating organisms from the biofilm environment (4, 11, 14).In a more recent study the combination of minocycline and EDTAflush solution was shown to be highly efficacious in preventingcatheter-related colonization, bacteremia, septic phlebitis,and endocarditis in an animal model (10).

The multifocal mechanisms of action of tetrasodium EDTA in inhibitingbiofilm formation primarily involve its metal-chelating activity.In this present study, which involved 20 catheters removed fromhospitalized patients, a tetrasodium EDTA concentration of <40mg/ml was found to eradicate biofilm, based on viable cell counts,in the majority of cases. Tetrasodium EDTA was found to havebroad-spectrum activity against both gram-positive and gram-negativeorganisms in in vivo-generated biofilms.

ACKNOWLEDGMENTS

We acknowledge the generous financial support of Aseptica Inc.,Seattle, Wash.

FOOTNOTES

* Corresponding author. Mailing address: Department of Microbiology and Anti-Infectives, Global Development Centre, ConvaTec (Bristol-Myers and Squibb), Deeside Industrial Park, Flintshire CH52NU, United Kingdom. Phone: 44 (0) 1244584485. Fax: 44 (0) 1244584311. E-mail: steven.percival{at}bms.com.

REFERENCES

Top