TEXT

Top Abstract Text References

Catheter-related infections are common in hospitalized patients in the United States and Europe (2). The case fatality rate for catheter-related bacteremia (CRB) has been estimated to be 10 to 20% (5). Bacteria may colonize the medical device and migrate along it inside the lumen or outside at the surface, resulting in CRB (14). To reduce CRB, the primary approach involved replacing the device at certain intervals, but frequent replacement leads to an increase in medical costs. In addition, repetitive attempts of cannulation required for device replacement can enhance the risk of bacterial colonization and subsequent infection. Antiseptic treatment of indwelling or percutaneous devices, such as urinary catheters and central venous catheters, has been developed to reduce the incidence of CRB and to reduce medical costs when replacement of a device becomes necessary less often. Various antiseptics, such as silver (3, 4, 19), benzalkonium chloride (21), iodine (8), Irgasan (11-13), and chlorhexidine (3, 15), and antibiotics, such as dicloxacillin (20), teicoplanin (7), minocycline with rifampin (17), and mupirocin (6), have been tested for this purpose. Silver sulfadiazine and chlorhexidine in combination with silver sulfadiazine are among the most common agents (3). There are several methods to impregnate a device with silver in order to achieve a gradual release (18, 19). Spire Services Technology Systems (SSTS [Milan, Italy]) has developed an ion implantation technique to silver-coat the device polymer as an "actively sterile" surface (SPI-ARGENT II). It is a vacuum process in which ions are generated from a source, accelerated, and injected into the subsurface region. According to the company, the silver-impregnated polymer is highly bactericidal to Staphylococcus aureus, Pseudomonas aeruginosa, Escherichia coli, and Candida albicans. We studied the microbicidal activity of this new polymer in vitro in comparison to that of the untreated polymer.

The microorganisms included S. aureus ATCC 6538, P. aeruginosa ATCC 15442, E. coli ATCC 11229, and C. albicans ATCC 10231. Bacterial strains were passaged three times on tryptose agar plates (product no. CM 233; Oxoid, Basingstoke, England) supplemented with 5% sheep blood, and C. albicans was passaged three times on Sabouraud-glucose agar (product no. CM 41; Oxoid) before experiments were carried out. The device polymer with and without silver ion implantation was purchased from SSTS. A piece of ~15 by 15 cm was available for all experiments.

Silver was neutralized with 5% horse serum (product no. SR 35; Oxoid) (22). The efficacy of horse serum to neutralize silver was shown in three parallel experiments before the survival experiments were carried out. The end point was survival or no survival of the microorganisms in the presence of silver. The experiment was controlled for survival in distilled water. Bacteria were incubated in silver-containing distilled water at 37°C for 48 h, and C. albicans organisms were incubated at 30°C for 72 h. S. aureus, P. aeruginosa, E. coli, and C. albicans were not able to survive silver at a concentration of 0.0125 ppm (survival controls without silver were positive). In the presence of horse serum (5%), however, S. aureus, P. aeruginosa, and E. coli were able to survive silver at a concentration of 0.8 ppm, and C. albicans was able to survive even at a silver concentration of 1.6 ppm.

Pieces 1 cm² in size were cut from nonantiseptic-impregnated and silver-impregnated polymer with sterile scissors. They were inoculated either with five colonies of each microorganism by using a sterile swab (dry inoculum) or with 50 µl of a suspension with sterile saline (liquid inoculum). The inoculated pieces were incubated in a humid chamber at 37°C. Recovery of the organisms was carried out immediately after the inoculation (time 0) and after 1 h, 24 h, and 7 days. Pieces were placed into glass tubes containing tryptic soy broth and horse serum (5%) to neutralize the silver. A serial dilution was performed. Fractions of 100 µl were distributed from appropriate dilutions on tryptic soy agar or Sabouraud agar with a spiralometer (Whitley automatic spiral-plater; Don Whitley Scientific, Ltd., Shipley, England). The remaining broth was filtered with cellulose nitrate membranes (pore size, 0.2 µm). Tryptic soy agar plates were incubated at 37°C for 48 h, and Sabouraud agar plates were incubated at 30°C for 72 h. The number of colonies was determined, and the total number of surviving organisms was calculated (recovery rate). Six parallel experiments were carried out with all four microorganisms (dry inoculum) and with P. aeruginosa and S. aureus (liquid inoculum) at each reaction time for both the ion-implanted polymer and the control polymer.

The log₁₀ number of organisms recovered from each type of polymer was calculated for each of six parallel experiments (one colony was assumed when zero colonies were detected after filtration). Data are presented as means ± standard deviations for each organism and reaction time. The Mann-Whitney U test was used to compare the number of recovered organisms between the two polymers at a specific time (1). A P value of <0.05 was regarded as significant.

Significant differences between the number of organisms on both polymers were only found for S. aureus after 24 h (dry inoculum, P = 0.037 [Table 1]; liquid inoculum, P = 0.004 [Table 2]) and C. albicans after 24 h (dry inoculum, P = 0.004 [Table 1]). E. coli and P. aeruginosa did not yield significant differences between the polymers at any reaction time (Tables 1 and 2).

Various methods are available to impregnate device polymer with silver. SSTS has developed an ion implantation technique and claims that the polymer (SPI-ARGENT II) is microbicidal to S. aureus, E. coli, P. aeruginosa, and C. albicans. We studied the survival of the same microorganisms at incubation times of 1 h, 24 h, and 7 days. No significant difference in the number of surviving organisms was found between the polymer with and without silver at any time for any microorganism, except for C. albicans after 24 h (dry inoculum, colony count on silver-impregnated polymer significantly higher) and S. aureus after 24 h (dry inoculum, colony count on silver-impregnated polymer significantly higher; liquid inoculum, colony count on the control polymer significantly higher). This finding is in contrast to the information offered by the manufacturer of this new polymer, who claims that the surface produces 100% killing of bacteria in both short-term tests (30 to 45 min) and long-term studies (42 days) by the shake flask method.

Various methods are available to study the microbicidal activity of silver ions in a polymer. The manufacturer was able to demonstrate the antimicrobial effectiveness by the shake flask method as follows. Treated and control samples are directly inoculated with an organism suspended in a buffered solution. Samples are agitated continuously for 1 h, serially diluted, and counted by the plate count method. According to the manufacturer, the surface is bactericidal on contact and minimally leaching. That is why we have chosen a method which allows assessment of microbicidal activity on contact by closely resembling the bacterial or fungal colonization of a device. In our method, bacteria and fungi are inoculated onto the silver ion-implanted polymer surface, and their survival is studied in comparison to that with the nontreated polymer. According to our results, we cannot exclude that it is the polymer itself and not the ion implantation which may have a microbicidal effect, if there is any effect at all. Lower inocula may yield very different results and allow for differentiation. Due to the limited availability of the polymer, further experiments with a lower inoculum were not possible.

Neutralization of an antiseptic is very important in in vitro testing, because residual activity of the antiseptic may result in false high-reduction factors. We have used horse serum (5%) to neutralize residual silver. In preliminary experiments, we were able to demonstrate that horse serum can neutralize silver at a concentration of up to 0.8 ppm. Dilution of microorganisms in tryptic soy broth instead of horse serum in the presence of silver yielded the same result. It is therefore possible that the horse serum effect is not neutralization but an effect on the bacteria. The manufacturer of SPI-ARGENT II does not provide any information about whether residual silver has been neutralized in the shake flask method or which neutralizing agents have been used. This is a possible explanation for the discrepancy between our results and the information provided by the manufacturer. Other investigators have used a combination of 5% Tween, 2% lecithin, 0.6% sodium oleate, 0.5% sodium thiosulfate, 0.1% proteose peptone, and 0.1% tryptone to neutralize silver sulfadiazine and chlorhexidine (3). Whatever combination of neutralizing agents is used, effective neutralization of residual antiseptics is a condition for in vitro testing of antiseptic materials (19). It will also play a crucial role in the microbiology laboratory when antiseptic-coated catheters are to be investigated for bacterial or fungal colonization (23).

In addition to the antimicrobial effect of a polymer, the adhesive effect of the polymer surface for bacteria and fungi is of importance (9). SPI-ARGENT II silicone rubber is minimally adhesive according to the manufacturer, with a critical surface tension of 27 dynes per cm (control silicone rubber, 14 dynes per cm). Material with a surface tension of between 20 and 30 dynes per cm is regarded as naturally thromboresistant and bacteriostatic. The survival of E. coli was surprisingly poor on both polymers in comparison to that of the other microorganisms. It is possible that this finding is due to a serum bactericidal effect. Repeating the experiments with another method for neutralization could help to explain this finding.

We did not investigate the adhesive effect of the SPI-ARGENT II, but we noticed that liquid application of the microorganisms did not result in complete wetting of the surface, indicating the presence of some surface tension on both polymers. Further experiments will be necessary to study the adhesive effect of the polymer. Antiseptic treatment of device material should not only have an in vitro microbicidal effect but should also significantly reduce the incidence of catheter-associated nosocomial infections. Antimicrobial coating of central venous lines has been shown to decrease the rate of catheter-associated infections (10). Coating of the outer surface may not be sufficient, because the internal lumen becomes more important with time of catheterization as a risk factor for CRB (16). The impact of an antiseptic-treated device on the clinical outcome will decide its use in the future.

Our data provide evidence that implantation of silver ion into the polymer (SPI-ARGENT II) does not influence the survival of various microorganisms compared with that with the control polymer. Its value for clinical use appears to be doubtful.