Central venous catheterization is an integral part of the management of critically ill patients. However, it is associated with many complications, notably infection. Microbial colonization, a prerequisite of infection, frequently occurs on the internal and external surfaces of catheters resulting in sepsis ranging from cellulitis at the insertion site to septicaemia . Methods to reduce the incidence of catheter colonization and subsequent infection have focused on patient management and polymer improvements. Catheter development has included anti-adhesive coatings and more recently the incorporation of antibiotics and antiseptics into the polymers [2,3]. The use of catheters impregnated with antibiotics used to treat infections may however, be limited by the emergence of antimicrobial resistant micro-organisms, an increasing clinical problem [4,5]. These include the recent emergence of glycopeptide-resistant Staphylococcus aureus[6,7]. In comparison, antiseptic-impregnated central venous catheters are less likely to induce resistance. Many polymers incorporating antiseptics have been developed [8,9]; however, relatively few are available for clinical use [10,11]. A central venous catheter coated with chlorhexidine and silver sulphadiazine on its external surface only has been studied extensively, with varying results [10,12-15]. Of concern are the reported adverse reactions, albeit only in Japan, associated with this central venous catheter .
More recently a central venous catheter which is coated on both the internal and external surfaces with the quaternary ammonium compound, benzalkonium chloride has been produced (Becton Dickinson, Swindon, UK). Benzalkonium chloride has a broad spectrum of antimicrobial activity and is used widely in healthcare both as an antiseptic and preservative. The benzalkonium chloride catheters are coated with a hydrophilic hydromer layer  which consists of polyvinylpyrrolidone molecules bound to the polyurethane surface and cross-linked to form a network throughout the layer. The benzalkonium chloride molecules are entrapped within this network in an anhydrous state but diffuse into the surroundings when the polyvinylpyrrolidone comes into contact with an aqueous solution. This mechanism ensured that the benzalkonium chloride is continuously present on both the internal and external surface of the catheters. In vitro studies demonstrated that the antiseptic central venous catheter inhibited the adherence of a range of bacterial species [18,19]. The benzalkonium chloride in the catheter did not result in any toxicity in animal models during in vivo studies (Becton Dickinson data on file). Similarly, in a toxicology study in 30 patients (Elliott, unpublished data) no adverse side-effects were reported. In this preliminary clinical evaluation the new benzalkonium chloride catheter was assessed in an open, comparative, controlled randomized study for the prevention of microbial colonization.
Ethical committee approval was obtained prior to commencement of the study. Patients over 18 years of age who were admitted to the Queen Elizabeth Hospital, Birmingham, UK and who required a central venous catheter as part of their clinical management, and who had no known allergy or sensitivity to benzalkonium chloride were recruited to the study. The patients were all admitted for routine surgical procedures including coronary artery bypass grafting. Following informed consent each patient was randomly assigned to either the benzalkonium chloride catheter group or the control group. The randomization system consisted of sequentially sealed envelopes each of which contained information stating the specified study catheter the patient should receive. Two hundred and thirty-five patients were enrolled; 117 in the benzalkonium chloride group and 118 in the control group. Patients received only one study catheter. The central venous catheters were inserted as part of the patient's immediate preoperative preparation. All patients received prophylactic antibiotics that included a β-lactam (flucloxacillin or a cephalosporin). Metronidazole was also given if the patient underwent an abdominal operation. The antibiotics were given for 2 days for cardiac or abdominal surgery and for 5 days if the patient underwent a solid organ transplant.
Both catheters were polyurethane, non-cuffed, triple lumen, 7 Fr and 20 cm long. The control central venous catheter (Becton Dickinson) was a standard catheter whereas the antiseptic central venous catheter (Becton Dickinson) had a hydrophilic coating containing benzalkonium chloride.
Central venous catheter insertion and maintenance
All central venous catheters were inserted in anaesthetic rooms, under strict aseptic technique by anaesthetists using a Seldinger technique. Immediately prior to cannulation the insertion site was sprayed with chlorhexidine 2.5% in industrial methylated spirit 70% (Hydrex DS Derma Spray, DuPuy Healthcare, Leeds, UK), which was allowed to dry for up to 2 min. After successful cannulation the catheter insertion site was again cleaned and a nonocclusive, transparent dressing applied (Tegaderm™, 3M, Borken, Germany). The dressings were changed every 72 h, or earlier if there was evidence of blood or moisture collection. At each dressing change the insertion site was cleaned with gauze moistened with normal saline and then sprayed with chlorhexidine gluconate 2.5% in industrial methylated spirit.
Serum sodium and serum potassium concentrations in blood
Serum sodium and potassium concentrations determined for blood samples taken via heparin benzalkonium chloride-bonded umbilical catheters have resulted in elevated values . In the current study, to determine any effect of benzalkonium chloride coated catheters on electrolyte measurements, immediately following catheter insertion two blood samples were taken: one via the distal lumen of the central venous catheter and the other from a separate peripheral venepuncture. Serum sodium and potassium concentrations of these samples were determined by a direct reading ion selective electrode analyser (Chiron Diagnostics Ltd, Halstead, Essex, UK).
Patients were observed daily for symptoms and signs of local and systemic catheter-related infection  as well as for evidence of catheter-related adverse events. Catheter-related adverse events were defined as any undesirable clinical occurrence that was attributable to the study central venous catheter. If catheter-related infection was suspected, blood cultures were obtained via the central venous catheter and also from a separate peripheral venepuncture.
Determination of the number of micro-organisms at the central venous catheter skin insertion site
The 2 cm2 area surrounding the central venous catheter skin insertion site was swabbed prior to skin cleaning and catheter removal, as previously described .
The decision to remove the study central venous catheter was made independently of the research team. Immediately prior to removal the insertion site was cleaned with sterile normal saline followed by chlorhexidine gluconate 2.5% in industrial methylated spirit which was allowed to dry for 2 min. On removal, the 6-cm distal tip and the 3-cm subcutaneous portion (intradermal segment, Fig. 1) were aseptically obtained for microbiological examination. All catheter segments were stored at 4°C until processed.
Determination of the number of micro-organisms present on the central venous catheter external and internal surfaces and the hubs
The number of micro-organisms colonizing the external catheter surfaces was determined by the roll plate method . Microbial colonization of the internal catheter surface was assessed by flushing 1 mL of Stuart's transport medium containing lecithin (2%) and Tween 80 (3%) through the lumen . The lecithin and Tween 80 neutralized any benzalkonium chloride present. Each central venous catheter hub was also sampled with a swab, which were cultured on blood agar for the presence of micro-organisms as previously described .
Determination of antibacterial activity of benzalkonium chloride retained within the central venous catheter
After removal, in vitro antimicrobial activity of the catheters was determined by assessing bacterial adherence to both the internal and external surfaces using a dynamic model  and by zones of inhibition against Staphylococcus epidermidis NCTC 11047 . Separate sections of the catheters were used for each determination.
Antimicrobial activity of benzalkonium chloride
The minimum inhibitory concentration and the minimum bactericidal concentration of benzalkonium chloride were also determined, by standard methods, against the micro-organisms isolated from the central venous catheter skin insertion site, the catheter hubs and the catheter segments from patients who received an antiseptic catheter.
Unadjusted odds ratios were calculated and tested using the Fisher's exact test to assess the effect of catheter type on the incidence of colonization and lumen blockage. A Wilcoxon rank-sum test was used to evaluate the effect of catheter type on maximum erythema and oedema scores. An analysis of variance (ANOVA) was employed to assess the relationship between the number of colonizing units and the zone of inhibition.
Data from 11 patients from the benzalkonium chloride catheter group and 20 patients from the control group were nonevaluable as the central venous catheters were unavailable for microbiological analysis. A total of 204 patients were evaluable, 106 in the benzalkonium chloride group and 98 in the control group. The characteristics of the two patient groups were similar (Table 1). Indications for central venous catheterization included: administration of fluids, drugs and the monitoring of central venous pressure.
None of the patients in either group experienced hypersensitivity reactions to the catheters. Mild erythema (< 1 mm) at the skin insertion site was present in 27 patients (26%) in the benzalkonium chloride group and 28 patients (28%) in the control group (P=0.5).
Monitoring of sodium and potassium concentrations
Serum sodium and potassium concentrations were measured in 62 (59%) patients in the benzalkonium chloride group and in 45 (46%) patients in the control group. There was no significant difference in values between the samples obtained (P=0.5).
Microbial colonization at the central venous catheter skin insertion site and catheter hubs
Micro-organisms were isolated from the skin insertion site of 62/103 (60%) of the benzalkonium chloride group and 60/89 (67%) of the control patients immediately prior to catheter removal (P=0.883). One or more catheter hubs were colonized in 47/104 (45%) of the benzalkonium chloride group and 44/95 (46%) in the control patients (P=0.88).
Microbial colonization of the external and internal catheter surfaces
The number of central venous catheters colonized on the external and internal surfaces is given in Table 2. When central venous catheters were in situ for < 5 days microbial colonization of the external surface of the distal segment (11 out of 78; 14%) was significantly lower in the benzalkonium chloride group compared with the control (21 out of 72; 29%) (P=0.029). Similarly, there was a significant reduction at the external surface of the subcutaneous segments of the BZC catheter (P=0.016). Benzalkonium chloride catheters in place for=5 days also showed a reduction in microbial colonization at the distal segment with only 9 (32%) catheters being colonized in comparison to 14 (54%) in the control group.
One (0.9%) patient in the benzalkonium chloride catheter group and two (2.2%) patients in the control group developed a localized infection at the catheter insertion site. Similarly, the incidence of systemic infection was low. There were no systemic catheter-related infections in the benzalkonium chloride catheter group. Two patients (2.2%) in the control group were clinically diagnosed as having a systemic catheter-related infection related to coagulase-negative staphylococci, which were isolated from the internal and external surface of the central venous catheter in one patient and from external catheter segments in the other patient. The clinical signs of sepsis included low-grade pyrexia and evidence of local infection at the site of catheter insertion. Blood cultures were negative in both patients.
Antimicrobial activity retained by the central venous catheter on removal
Following removal 35% of distal tips from benzalkonium chloride catheters elicited a zone of inhibition (mean 3.9 mm), whereas only 5% of the control catheters produced zones (mean 0.8 mm) (P= < 0.01). Zones were present in 46% of the benzalkonium chloride subcutaneous catheter segments (mean 4.7 mm) in comparison to 5 (5%) of the subcutaneous sections of control catheters (mean 0.8 mm) (P= < 0.01). There was no correlation between the zone of inhibition and the number of micro-organisms recovered from the benzalkonium chloride catheters' external surfaces.
Retained antimicrobial activity in 16 (15%) of the benzalkonium chloride group and 10 (10%) of the control group, was also assessed by the ability of Staphylococcus epidermidis to adhere to the central venous catheter in vitro. There was a reduction in the number of micro-organisms adhering to the external surface in the benzalkonium chloride group (mean number of organisms 190±163) in comparison to the controls (mean number of organisms 446±87) (P= < 0.01). This reduction was also demonstrated at the catheter internal surface (P= < 0.001).
Activity of benzalkonium chloride against isolated micro-organisms
The coagulase-negative staphylococci were the predominant micro-organisms isolated from all catheter sites in this study. The antimicrobial activity of benzalkonium chloride against micro-organisms isolated from the surface of the catheter was determined. The range of micro-organisms tested include the coagulase-negative staphylococci which accounted for > 95% of all isolates. Other micro-organisms studied were Staphylococcus aureus and Gram-negative aerobic bacilli.
The coagulase-negative staphylococci and Staphylococcus aureus had a mean minimum inhibitory concentration to benzalkonium chloride of 2 mg L−1 (range 0.125-8 mg L−1, n=231) and minimum bactericidal concentration of 4 mg L−1 (range 0.5-32 mg L−1). The mean minimum inhibitory concentration for Gram-negative micro-organisms including Pseudomonas sp., Xanthomonas sp. and Acinetobacter sp. was 8 mg L−1 (range 0.5-8 mg L−1, n=21) and minimum bactericidal concentration of 16 mg L−1 (range 1-32 mg L−1).
This is the first clinical assessment of a polyurethane catheter coated on both the internal and external surfaces with a quaternary ammonium compound, benzalkonium chloride. The antiseptic significantly reduced the microbial colonization of both catheter surfaces in comparison to the unprotected control devices. Microbial adherence to central venous catheters is a likely important prerequisite to the development of these infections  and devices with protection on both surfaces are required, as is offered by this benzalkonium chloride coating. Other currently available antiseptic central venous catheters include a silver-sulphadiazine and chlorhexidine device that has been shown to reduce colonization and sepsis, demonstrating further the relationship between these two end points. However, this catheter is only currently coated on the external surface . Other antimicrobial catheters that contain antibiotics used to treat infections, including a combination of doxycycline and rifampicin, have also been developed . Although there were no reports in clinical trials of associated antimicrobial resistance with this antimicrobial catheter this remains a possibility. The application of antiseptic for the protection of catheters overcomes this potential problem of antibiotic resistance.
The benzalkonium chloride catheters were well tolerated, confirming earlier developmental toxicology studies. Benzalkonium chloride has been associated with elevated serum sodium and potassium concentrations when blood samples were taken via heparinbonded umbilical catheters which contained the antiseptic . However, in this present study there was no difference between the serum sodium and potassium values obtained either via the benzalkonium chloride catheter and those taken simultaneously from a separate venepuncture site.
One of the main problems associated with antimicrobial polymers is retention of activity. Previous studies have demonstrated that antimicrobial polymers may rapidly lose their activity, often within days of placement within a patient [8,9]. With the benzalkonium chloride catheters, activity was detected in devices that had been in situ for up to 12 days. The methods used to detect this activity were zones of inhibition and an in vitro colonization model. The in vitro model was shown to be the most sensitive method for determining continued antimicrobial activity. Although the in vitro results demonstrated retained antimicrobial activity for > 5 days, the in vivo findings suggested that colonization increases after this time. This may reflect differences in the in vitro and in vivo situations, including factors such as protein deposition on the catheter surface or slime production by the micro-organisms. Of note was the small number of control catheters, which produced a zone of inhibition. This was probably related to the antibiotics administered to the patients around the time of catheter removal.
The results of this preliminary study demonstrate that by the application of benzalkonium chloride uniquely incorporated into both the internal and external surfaces of a catheter by cross-linking a network of polyvinylpyrrolidone a reduction of microbial colonization can be achieved. This reduction in colonization may potentially decrease the incidence of subsequent catheter-related infection. The colonized benzalkonium chloride catheters were in situ for a longer time as compared to uncolonized devices. A similar trend was noted for the subcutaneous segment of the control catheter. However, this relationship was not evident for the distal tip segment of the control catheters. This could have reflected the duration of catheterization which may not have allowed adequate time for micro-organisms to reach and colonize the distal catheter surface.
In the present study less than 1% of patients developed a catheter-related infection, which is a relatively low infection rate compared to those reported in other studies. This may reflect the strict aseptic schedule followed in our hospital as well as the relatively short period for which the catheters were in situ. The perioperative antibiotic prophylaxis used are unlikely to have accounted for this low infection rate as the majority of the micro-organisms which cause catheter-related sepsis are resistant to these antimicrobials. The current results therefore suggest that antiseptic-coated catheters should be primarily reserved for patients where the incidence of catheter-related infection is comparatively high. Further studies need to be conducted in such high-risk patients to determine the efficacy of this antimicrobial catheter in not only reducing the rate of microbial colonization but to also investigate more fully the relationship between colonized catheters and infection.
We would like to thank Max Derrick for technical advice, Quintiles (UK) for the statistical evaluation, and all the staff at the Queen Elizabeth Hospital who helped us with the study. The work was supported by a grant from Becton-Dickinson, Swindon, UK.
1 Elliott TSJ, Faroqui MH. Infections and intravascular devices. Br J Hosp Med
2 Tebbs SE, Elliott TSJ. Modification of central venous
catheter polymers to prevent in vitro
microbial colonisation. Eur J Clin Microbiol Infect Dis
3 Elliott TSJ. Role of antimicrobial central venous
catheters for the prevention of associated infections. J Antimicrob Chemother
4 House of Lords Select Committee on Science and Technology. Resistance to antibiotics and other antimicrobial agents. London: HMSO, 1998.
5 Elliott TSJ, Lambert PA. Antibacterial resistance in the intensive care unit: mechanisms and management. Br Med Bull
6 Smith TL, Pearson ML, Wilcox KR et al.
Emergence of vancomycin resistance in Staphylococcus aureus.
Glycopeptide-Intermediate Staphylococcus aureus
Working Group. N Engl J Med
7 Sieradzki K, Roberts RB, Haber SW, Tomasz A. The development of vancomycin resistance in a patient with methicillin-resistant Staphylococcus aureus
infection. N Engl J Med
8 Kingston D, Seal DV, Hill ID. Self-disinfecting plastics for intravenous catheters and prosthetic inserts. J Hyg
9 Jansen B, Jansen S, Peter G, Pulverer G. In vitro
efficacy of a central venous
catheter ('Hydrocath') loaded with teicoplanin to prevent bacterial colonization. J Hosp Infect
10 Maki DG, Stolz SM, Wheeler S, Mermel LA. Prevention of central venous
catheter-related bloodstream infection by use of an antiseptic-impregnated catheter. A randomized, controlled trial. Ann Intern Med
11 Goldschmidt H, Hahn U, Salwender H-J et al.
Prevention of catheter-related infection by silver coated central venous
catheters in oncological patients. Zentrablatt Fur Bakteriologie
12 Logghe C, Van Ossel C, D'Hoore W, Ezzedine H, Wauters G, Haxhe JJ. Evaluation of chlorhexidine and silver sulfadiazine impregnated central venous
catheters for the prevention of bloodstream infection in leukaemic patients: a randomised controlled trial. J Hosp Infect
13 Heard SO, Wagle M, Vijayakumar E et al.
Influence of triple-lumen central venous
catheters coated with chlorhexidine and silver sulfadiazine on the incidence of catheter-related bacteraemia. Arch Intern Med
14 Haxhe JJ, D'Hoore W. A meta-analysis dealing with the effectiveness of chlorhexidine and silver-sulfadiazine impregnated central venous
catheters. J Hosp Infect
15 Darouiche RO, Raad II, Heard SO et al.
A comparison of two antimicrobial impregnated central venous
catheters. Catheter Study Group. N Engl J Med
16 Food and Drink Administration (FDA). Public Health Notice. Potential hypersensitivity reactions to chlorhexidine impregnated medical devices. March, 1998.
17 Jansen B. Current approaches to the prevention of catheter-related infections. In: Seifert H, Jansen B, Farr BM, eds. Catheter-Related Infections.
New York: Dekker, 1997: 411-446.
18 Tebbs SE, Elliott TSJ. A novel antimicrobial central venous
catheter impregnated with benzalkonium chloride. J Antimicrob Chemother
19 Tebbs SE, Elliott TSJ. Intravascular catheters impregnated with benzalkonium chloride. J Antimicrob Chemother
20 Gaylord MS, Pittman PA, Bartness J, Tuinman AA, Lorch V. Release of benzalkonium chloride from a heparin-bonded umbilical catheter with a resultant fictitious hypernatraemia and hyperkalaemia. Paediatrics
21 Tebbs SE, Elliott TSJ. Prevention of central venous
catheter-related infection. J Hosp Infect
22 Elliott TSJ, Moss HA, Tebbs SE et al.
A novel approach to investigate a source of microbial contamination of central venous
catheters. Eur J Clin Microbiol Infect Dis
23 Maki DG, Weise CE, Sarafin HW. A semi-quantitative culture method for identifying intravenous catheter related infection. N Engl J Med
24 Cleri DJ, Corrado ML, Seligman SJ. Quantitative culture of intravenous catheters and other intravascular inserts. J Infect Dis
25 Raad I, Darouiche R, Dupuis J et al. Central venous
catheters coated with minocycline and rifampin for the prevention of catheter-related colonization and bloodstream infections. A randomized, double-blind trial. The Texas Medical Center Catheter Study Group. Ann Intern Med