Antibiotic-coated medical devices: with an emphasis on inflatable penile prosthesis

There has been a recent emergence of S. epidermidis strains resistant to the quinolones (particularly cipro-floxacin) and to the glycopeptide antibiotics [24]. In addition, the increase in the frequency of device-related infections associated with methicillin-resistant strains of this organism, has led investigators to search for alternative agents for prevention and treatment of infections involving S. epidermidis.

In vitro, as well as in vivo studies have demonstrated the effectiveness against S. epidermidis of rifampin and minocycline, alone or in combination [25-27]. These agents will be discussed in more detail later in the review. In addition, the synergistic combination of quinupristine/dalfopristin has been shown to have good bactericidal activity against methicillin-resistant S. epidermidis [28,29]. This is a semisynthetic parenteral streptogramin antibiotic which could represent an alternative to glycopeptide antibiotics.

The oxazolidinone antibiotics (U-100592 and U-100766) are another novel class of antimicrobial agents that act by inhibiting protein synthesis. They have been shown to be active against most medically important gram-positive bacteria, however further clinical studies are necessary to demonstrate the safety and efficacy of these agents [30-32].

Nearly synchronous with the advent of the prosthetic medical device, strategies were developed to attempt to minimize the risk of infection. One of the most promising of these is the development of an antimicrobial coating on the surface of the prosthesis to prevent bacterial colonization as well as infection [33].

To be effective, polymer-associated antibiotics need to fulfill several prerequisites. First, they must be effective against the most common organisms found in infections of the particular prosthesis, as well as maintain potency when bound to the polymer. Secondly, the device must provide effective antibacterial concentrations in situ during the period when it is most susceptible to colonization/infection. Finally, the issue of adverse reactions to the antibiotic formulations used needs to be considered. The development of allergic reactions can have disastrous consequences for the patient. Although no antibiotic can presently meet all these criteria, various antibiotics have been looked at. Schierholz et al [34] looked at drug-release rates, bacterial colonization, and morphological features of antibiotic-incorporated polyurethanes in an attempt to determine the antimicrobial activity of these delivery systems. Bacterial colonization was inhibited effectively by preparations showing a slower but more sustained antimicrobial delivery, as seen with gentamicin-base and flucloxacillin. Whereas, polymers loaded with ciprofloxacillin and fosfomycin showed a granular structure of crystallized drug leading to a less predictable rate of release. It was felt that high homogeneity is required for a sustained and prolonged release over time and effective inhibition of bacterial colonization.

Central venous catheters were one of the first devices coated antiseptics and antibiotics in an attempt to minimize the infection rates. They are one of the leading causes of nosocomial infections and primary septiciemia [35]. The most common organisms involved in these cases include coagulase-positive and coagulase-negative staphylococci, particularly slime-producing strains [36].

In an in vitro susceptibility study of 197 catheter-related staphylococcal isolates at the University of Texas M. D. Anderson Cancer Center, minocycline, novobiocin, and rifampin were shown to have antimicrobial activities equivalent to those of vancomycin and other glycopeptide antibiotics [Raad I, Darouiche R. Central venous catheters (CVC) coated with minocycline and rifampin (M/R) for the prevention of catheter-related bacteremia (CRB), abstr. J7, p258. In Program and abstracts of the 35th Interscience Conference on Antimicrobial Agents and Chemotherapy. American Society for Microbiology, Washington, D. C., 1995]. Furthermore, a recent study by Raad et al [25] compared the efficacies of vancomy-cin, clindamycin, novobiocin, and minocycline, alone or in combination with rifampin in an in vitro model of colonization. The models consisted of antibiotic-impregnated cement filling the lumen of catheter segments. They found that the combination of minocycline and rifampin led to the total prevention of colonization by staphylococci, and was more effective than the combination of vancomycin and rifampin. The efficacy of this combination of antibiotics was comparable to that of ceftazidime against P. aeruginosa or S. maltophilia. Moreover, this combination had an in vitro antimicrobial activity against C. albicans that was comparable to that of amphotericin B.

Raad et al recently evaluated the activities of catheters coated with minocycline/rifampin, and with chlorhexidine gluconate/silver sulfadiazine both in vitro, as well as in a rabbit model [37]. When incubated in serum at 37 , the half-life of the inhibitory activity of catheters coated with minocycline/rifampin and chlorehexidine gluconate/silver sulfadiazine, was 25 days and 3 days respectively. Furthermore, in the rabbit model, catheters coated with the antibiotic combination were significantly more efficacious than the catheters coated with antiseptics in preventing colonization and infection with Staphylococcus aureus (P<0.05). Catheters coated with minocycline/rifampin demonstrated broad-spectrum in vitro inhibitory activity against gram-positive bacteria, gram-negative bacteria, and Candida albicans that was significantly superior to the inhibitory activity of catheters coated with chlorhexidine gluconate/silver sulfadiazine (P<0.01), as determined by size of zone of inhibition .

Finally, this same group conducted a multicenter, randomized clinical trial of 281 patients requiring 298 triple-lumen, polyurethane central venous catheters [27]. They were attempting to determine the efficacy of catheters coated with minocyclin/rifampin in preventing catheter-related colonization and bloodstream infection. In this study, catheters coated with minocycline/rifampin were shown to decrease the risk of catheter-related colonization, 26 % (coated) vs 8 % (control) (P<0.001), as well as bloodstream infections, 0 % (coated) vs 5 %(control) (P<0.01). In addition, they failed to detect any adverse reactions or antimicrobial resistance to the antibiotic combination. It is also interesting to note that although the rate of colonization of Staphylococcus epider-midis was significantly less in the group with coated catheters, the frequency of colonization of catheters with gram-negative bacilli and Candida species was similar in the two groups.

In conclusion, because of the increasing evidence being published about the efficacy of antimicrobial-impregnated central venous catheters, the recent Centers for Disease Control and Prevention guidelines for the prevention and catheter-related infections recommend the use of these catheters for patients with a high rate of infection after full adherence to other infection control measures, such as maximal sterile barrier precaution [38].

Due to the initial success with antibiotic coatings in central venous coatings, a similar approach was adopted in bladder catheters. The presence of a Foley catheter in the urethra provides bacteria with a surface that can be readily colonized. Two points of access exist in the urethral catheter: through the lumen of the catheter and along its external surface. Despite adherence to aseptic technique, approximately 10 % to 30 % of patients with indwelling bladder catheters develop a urinary tract infection (UTI) [39].

In an attempt to prevent bacterial adherence, catheters have been manufactured with silicone and coated with hydrophilic polymers, hydrogels, and Teflon [40-43]. More recently, antimicrobial coatings of urinary catheters were studied. A particularly interesting antibiotic combination used is minocycline and rifampin. This combination is seldom used to treat UTI, the different mechanisms of these two drugs makes the development of resistance less likely, and it has been shown in vitro to have a broad-spectrum of activity against gram-positive bacteria, gram-negative bacteria and Candida [44]. Furthermore, as described earlier, the hand-full of studies that have looked at this combination in central venous catheters appear to be promising.

Dariouch et al conducted a prospective, randomized clinical trial at five medical centers looking the efficacy of bladder catheters impregnated with minocycline and rifampin in reducing catheter-associated bacteriuria [45]. Patients undergoing radical prostatectomy were randomized to receive intraoperatively either regular silicone bladder catheters (controls; n = 68), or silicone bladder catheters impregnated with minocycline and rifampin (n = 56). Catheters remained in place for 2 weeks, and urinary cultures were obtained at 3, 7 and 14 days. Kaplan-Meier curves for freedom from catheter-associated bacteriuria demonstrated that it took significantly longer for patients who received antimicrobial-impregnated catheter to develop bacteriuria than those who received the control catheters (P = 0.006 by the long rank test). In addition, there were significantly lower rates of gram-positive bacteriuria in patients who received the antimicrobial-impregnated catheter versus control catheter (7.1 % vs 38.2 %; P<0.001). However, the two groups had similar rates of gram-negative bacteriuria ( 46.4 % vs 47.1 %; P = 1.0) and candiduria (3.6 % vs 2.9 %; P =1.0). There was a trend towards lower incidence of symptomatic UTIs among patients who received an antimicrobial-impregnated bladder catheter (1.8 % vs 8.8 %), however, this was not statistically significant (P = 0.13).

During the last quarter century, the use of implantable penile prostheses has become an increasingly popular and common procedure for men with erectile dysfunction who have failed medical management. However, even adhering to strict sterile operative technique, infection of the prosthesis remains a major cause of morbidity for the patient. The frequency of such infections has been reported to range from 1.7 % to 8.3 % [46-49]. Since attempts to treat established prosthetic infection without explantation have been largely unsuccessful, attention has focused on preventing infections.

Staphylococcus epidermidis is the most common cause of infection of penile prostheses, and is followed in frequency by S. aureus [22]. As with vascular grafts, it is believed that in most cases the bacteria is introduces at the time of surgery. The use of perioperative antibiotic prophylaxis, as well as a preoperative preparation including scrupulous, repeated disinfection of the skin of the genital and perineal region, has been shown to be effective in decreasing the incidence of infection [50]. However, attempts have been made to diminish infection rates further.

As with other prosthetic devices, impregnation of penile prostheses materials with antibiotics has recently been looked at. Haikun et al conducted a preliminary study to evaluate the anti-infective efficacy of antibiotic-coated silicone as a surrogate for the penile prosthesis in a rat model [51]. Rifampin/minocycline, vancomycin and amikacin were used for bonding to silicon strips. These were subsequently dipped in bacterial solutions containing Staphylococcus epidermidis or S. aureus and implanted subcutaneously in adult Sprague-Dawley rats. After seven days, the strips were removed, and the number of bacteria on the strips as well as in the surrounding tissue was determined. In addition, they evaluated the in vitro antibiotic activity of the coated strips against the same organisms.

When contaminated with S. epidermidis, six of the nine rats in the control group yielded strips and tissue producing heavy bacterial growth. None of the six strips coated with rifampin/minocycline yielded bacterial growth, nor did any of the seven strips coated with vancomycin. One of the seven rats that received amikacin -coated strips demonstrated growth. Similar results were obtained using S. aureus as the contaminating organism. None of six, two of seven, and two of six strips were infected in the rifampin/minocyclin, vancomycin, and amikacin groups respectively. Differences were significant for vancomycin treatment (P0.017) and highly significant for rifampin/minocycline (P0.003). They concluded that these graft materials may prove useful in preventing the infection of penile prostheses.

In 1991, the FDA approved an antibiotic-coated surface treatment for the inflatable penile prosthesis (InhibiZone; American Medical Systems, Minnetonka, MN) in an attempt to minimize the infection risk. Mino-cycline and Rifampin is impregnated into the external silicone surfaces of implanted components such as penile prosthesis cylinders, pumps, reservoirs and tubing. The total amounts of Rifampin and Minocycline are low, and are not expected to produce measurable systemic blood levels of either drug. Furthermore, there has been no noticeable effect on the operation, durability or performance characteristics of the device or components. In vitro studies performed by AMS showed mean zones of inhibition when a strip of device was plated on test agar inoculated with S. epidermidis and S. aureus to be 22.6 mm and 17.5 mm respectively. Whereas zones of inhibition for Escherichia coli, Enterococcus faecalis, Candida albicans, and Proteus mirabilis were 6.5 mm, 4.8 mm, 0.1 mm, and 0.6 mm, respectively (InhibiZoneTM, antibiotic surface treatment. Preclinical study. Data on file., American Medical Systems, Inc. 2001.). Only the activity against S. epidermidis and S. aureus was considered significant. Sherertz et al [53] established a quantitative relationship between the size of the zone of inhibition in vitro and the concentration of S. aureus cultured from the tip of coated catheters 7 days after inoculation of the insertion site. Coated catheters with a zone size of 15 mm were highly predictive of in vivo efficacy, as reflected by prevention of colonization of the indwelling catheter. Whether this cut-off can be applied to antibiotic-impregnated penile prostheses is uncertain.

Further in vitro testing by Brock et al (Brock G, Bochinski D. InhibiZone treatment: the first antibiotic treatment impregnated into the tissue-contacting surface of an inflatable penile prosthesis. Abstract presented at the AUA annual meeting, 2001.), indicated that the majority of both drugs elute in the first 24 hours. This is intended to eliminate any organisms that have contaminated the device during implantation. It was also felt that since penile prostheses are permanent implants, it would be desirable that the drugs not persist at high levels for a long time.

In addition, AMS conducted a limited animal infection model using 11 rabbits (InhibiZoneTM, antibiotic surface treatment. Preclinical study. Data on file., American Medical Systems, Inc. 2001.). These animals were implanted with either test samples consisting of portions of an InhibiZoneTM treated AMS 700 pump, or the control samples consisting of portions of a standard AMS 700 pump without antibiotic coating. Five rabbits were implanted subcutaneously with 6 test samples each, and five rabbits were implanted subcutaneously with 6 control samples each. One rabbit received three test samples and three control samples. All samples were soaked in 103-104 CFU solution of Staphylococcus aureus for 8 hours. After 2 days, all samples were removed and observed for bacterial growth. Although detailed results are not available, they claim that the number of treated samples infected was statistically lower than the number of untreated samples that were infected. The length of time the samples were left in vivo, the sample size, the fact that absolute numbers of infected devices in the groups were not given, and that bacterial growth as opposed to clinical evidence of infection was looked at, makes this study of limited value.

However, there is increasing evidence in recent clinical studies that the InhibiZone coating on penile prostheses may decrease the subsequent device infection rates. Initial results presented by Wilson et al (Wilson SK, Delk JR, Henry GD. Early results show antibiotic coating (InhibiZone) decreases infection in AMS 700 CX penile prosthesis. Abstract presented at the South Central Section AUA meeting, 2002.) using penile prostheses coated with InhibiZone demonstrated 2 infections (2.4 %) in 84 patients with short follow-up (6 weeks to 8 months). Both patients had risk factors for infection; one with diabetes mellitus and one undergoing dual implant with concurrent placement of an artificial urinary sphincter. In a small study looking at 10 high-risk patients (5 with diabetes and 5 undergoing prosthesis revision) there were no prosthesis infections with follow-up ranging from 11 to 34 weeks (Ryan RT, Jones LA. Early experience with antibiotic impregnated inflatable penile prosthesis (IPP) in a high-risk patient population. Abstract presented at the South Central Section AUA meeting, 2002.).

Carson conducted a large review of revision cases from American Medical Systems Patient Information Forms (PIF) to compare the infection rate in InhibiZone treated and Non-InhibiZone revision implants (Carson CC. Initial success with AMS 700 series inflatable penile prosthesis with InhibiZone antibiotic surface treatment: a retrospective review of revision cases incidence and comparative results versus non-treated devices. Abstract presented at the AUA annual meeting, 2004.). Data was reviewed from 5 310 (60.7 %) Non-InhibiZone implants and 3 444 (39.3 %) InhibiZone implants. He found a
44 % decrease in infection rate in InhibiZone implants (1.4 % versus 2.4 %).

Therefore, preliminary experiments seem to suggest that antibiotic-coated silicone implant material, particularly rifampin/minocycline, may prove useful in the prevention of infection of penile prostheses. However, a large, prospective multi-center cohort would be required to confirm their true value in prosthetic surgery.

Finally in 2002, another prosthesis manufacturer (Mentor, Santa Barbara, CA, USA) developed a hydrophilic penile prosthesis substrate coating that is designed to help inhibit bacterial adherence. Recent publications have reported on the efficacy of antibiotic soaked, hydrophilic-coated substrates, particularly coated central venous catheters [54] and hydrocephalic shunts [55]. Hellstrom et al [56] investigated whether this hydrophilic coating (ResistTM) applied to inflatable penile prostheses can prolong the effect of intraoperative antibiotics. The activity of antibiotic-soaked Bioflex?/FONT> (penile prosthetic substrate material) discs with and without Resist was examined by measuring the zone of inhibition following in vivo exposure in four groups of rabbits. All disks were soaked in gentamicin and bacitracin. The Resist coating was found to be especially effective against Staphylo-coccuc epidermidis, and statistically significant improvements were observed for the coating over the uncoated substrate up to 3 days following implantation.

Clinical data on the hydrophilic-coated inflatable penile prosthesis is limited, however Wolter et al (Wolter C, Rajpurkar AD, Kendirci M, Dhabuwala CB, Hellstrom WJ. Hydrophilic-coated inflatable penile prosthesis: one-year experience. Abstract presented at the AUA annual meeting, 2004.) recently presented their one-year experience with the device. The infection rate in 2 357 coated penile prostheses was 1.06 % compared to 2.07 % in 482 uncoated penile prostheses implanted over the same time period. All data was collected from Mentor's internal database. Although preliminary data using this device also shows promise, long-term follow-up and prospective studies are not yet available.

Coating medical devices with antibiotics rather than antiseptic agents such as chlorhexidine has raised some concern about the potential for the emergence of antibiotic-resistant bacterial strains. Some have argued that unlike the heavily contaminated areas such as the gastrointestinal tract, devices such as penile prostheses are rarely heavily colonized or exposed to high-grade bacteremia, and are therefore less likely to induce resistance [25].

In vitro [57, 58] and in vivo [59, 60] studies have demonstrated that minocycline protects against the emergence of staphylococcal strains that are resistant to rifampin. Cross-resistance to other beta-lactam or glycopeptide antibiotics has not been observed with either rifampin or minocycline [61]. Furthermore, as stated earlier, the prospective randomized clinical study by Raad et al [27] involving central venous catheters coated with minocycline and rifampin failed to demonstrate organisms resistant to these antibiotics.

On the other hand, a recent study by Tambe et al [62] has provided some insight into the relative risk of emergence of resistant strains of Staphylococcus epidermidis and the efficacy of antimicrobial central venous catheters against antibiotic-resistant bacteria. The risk of development of resistance in this organism to number of antibiotics and antiseptics impregnated in central venous catheters was evaluated. The microbial culture was passaged 10 - 20 times through subinhibitory concentrations of different antimicrobials, singly and in combination, and the MIC of each antimicrobial before and after passage was compared. Interestingly, they found a 10- to 16-fold increase in the MIC of the combination of minocycline and rifampicin, while no significant increase was seen in the MIC of minocycline alone. On the other hand, there was no substantial change in the susceptibility to the antiseptic chlorhexidine when used alone or in combination with silver sulphadiazine. Furthermore, zone of inhibition tests of catheters impregnated with minocycline and rifampicin showed that their activity against rifampicin-resistant strains was lower than against the susceptible strain. However, the activity of the antiseptic (chlorhexidine and silver sulphadia-zine) catheters against the rifampicin-resistant and -susceptible strains was similar.

Although the data on antimicrobial-coated medical devices are promising, the evidence of effectiveness in clinical practice is still preliminary. Prospective trials will be needed to confirm the decreased infection rates seen thus far in retrospective studies. Further study will help determine whether these innovations are accepted as the standard of care for penile prosthesis surgery. However, these coated devices are not without potential risks. Two of the most important concerns include the emergence of antibiotic-resistant organisms, and hypersensitivity reactions. As long-term results become available, the importance of these concerns will become clear.