Lawrence L. Yeung,* Shaun Grewal,*, Arnold Bullock, H. Henry Lai and Steven B. Brandes

Transcription

1 A Comparison of Chlorhexidine-Alcohol Versus Povidone-Iodine for Eliminating Skin Flora Before Genitourinary Prosthetic Surgery: A Randomized Controlled Trial Lawrence L. Yeung,* Shaun Grewal,*, Arnold Bullock, H. Henry Lai and Steven B. Brandes From the Division of Urology, Department of Surgery, Washington University in St. Louis, St. Louis, Missouri (SG, AB, HHL, SBB), and the Department of Urology, University of Florida, Gainesville, Florida (LLY) Abbreviations and Acronyms AUS artificial urinary sphincter CNS coagulase-negative staphylococci GU genitourinary IPP inflatable penile prosthesis MUS synthetic male urethral sling PB propionibacterium Accepted for publication July 2, 212. Study received institutional review board approval. Supported by a grant from the Sexual Medicine Society of North America. * Nothing to disclose. Correspondence: Department of Surgery, Washington University in St. Louis, St. Louis, Missouri 6313 (telephone: ; FAX: ; grewals@wudosis.wustl. edu). Financial interest and/or other relationship with Pfizer, Astellas, American Medical Systems Inc. and Coloplast. Financial interest and/or other relationship with Pfizer and Astellas. Financial interest and/or other relationship with Astellas, American Medical Systems Inc. and Allergan. Editor s Note: This article is the third of 5 published in this issue for which category 1 CME credits can be earned. Instructions for obtaining credits are given with the questions on pages 398 and 399. Purpose: We defined the relevant skin flora during genitourinary prosthetic surgery, evaluated the safety of chlorhexidine-alcohol for use on the male genitalia and compared chlorhexidine-alcohol to povidone-iodine in decreasing the rate of positive bacterial skin cultures at the surgical skin site before prosthetic device implantation. Materials and Methods: In this single institution, prospective, randomized, controlled study we evaluated consecutive patients undergoing initial genitourinary prosthetic implantation. Patients were randomized to a standard skin preparation with povidone-iodine or chlorhexidine-alcohol. Skin cultures were obtained from the surgical site before and after skin preparation. Results: A total of patients were randomized, with 5 in each arm. Prepreparation cultures were positive in 79% of the patients. Post-preparation cultures were positive in 8% in the chlorhexidine-alcohol group compared to 32% in the povidone-iodine group (p.91). Coagulase-negative staphylococci were the most commonly isolated organisms in post-preparation cultures in the povidone-iodine group (13 of 16 patients) as opposed to propionibacterium in the chlorhexidine-alcohol group (3 of 4 patients). Clinical complications requiring additional operations or device removal occurred in 6 patients (6%) with no significant difference between the 2 groups. No urethral or genital skin complications occurred in either group. Conclusions: Chlorhexidine-alcohol was superior to povidone-iodine in eradicating skin flora at the surgical skin site before genitourinary prosthetic implantation. There does not appear to be any increased risk of urethral or genital skin irritation with the use of chlorhexidine compared to povidone-iodine. Chlorhexidine-alcohol appears to be the optimal agent for skin preparation before genitourinary prosthetic procedures. Key Words: chlorhexidine; penile implantation; urinary sphincter, artificial; suburethral slings; povidone-iodine WITH an aging population and increasing incidence of surgical treatment for prostate cancer, urological prosthetic procedures are being performed with increasing frequency. More than 15, penile prostheses are implanted per year in the United States and up to % of men seek treatment for stress urinary incontinence after radical prostatectomy. 1,2 Clinical infection remains the most dreaded complication of prosthetic sur /13/ / THE JOURNAL OF UROLOGY Vol. 189, , January by AMERICAN UROLOGICAL ASSOCIATION EDUCATION AND RESEARCH, INC. Printed in U.S.A.

2 GENITOURINARY PROSTHETIC SKIN PREPARATION 137 gery, resulting in additional operations in almost all cases. The primary route of implant infection is thought to be contamination before or during implantation in the majority of cases. 3 The infection rate of penile prostheses has traditionally been cited as 3% to 5%, increasing up to 8% in patients with diabetes, although this rate has decreased in the era of antibiotic coated prosthetics. 4 6 Likewise, infection rates of 5.5% have been reported for artificial urinary sphincters and 3% for male urethral slings. 7,8 Elimination of skin flora at the time of surgical site antisepsis is a logical step in preventing contamination and infection during prosthetic surgery. Povidone-iodine has traditionally been the antiseptic agent of choice for surgery involving the male genitalia. However, studies from other specialties have demonstrated that chlorhexidine-alcohol is superior in eliminating skin flora at the surgical site. 9, We compare the efficacy of chlorhexidine-alcohol (Chlora- Prep, 2% chlorhexidine gluconate and 7% isopropyl alcohol; Enturia, El Paso, Texas) to that of povidoneiodine (Betadine, 7.5% iodine scrub and % iodine paint; Purdue Products, Stamford, Connecticut) in decreasing the rate of positive bacterial skin cultures at the surgical skin site of GU prosthetic device implantation. Furthermore, we evaluated the safety of chlorhexidine-alcohol for use on the male genitalia, and further defined the native flora of the male genitalia and perineum. METHODS A prospective, randomized study of patients undergoing urological prosthetic operations was performed. Institutional review board approval was obtained for the study protocol. The primary aim of the study was to demonstrate that chlorhexidine-alcohol is equivalent to povidone-iodine at eradicating the skin flora before GU prosthetic surgery. Secondary aims of the study were to demonstrate the safety of chlorhexidine-alcohol for use on the male genitalia and to define the native flora at the surgical site during prosthetic surgery. Patients undergoing virgin placement of an IPP, AUS, MUS or testicular implant were included in the study. Patients were excluded from study if they had a history of GU prosthetic implant procedures, prior surgical site infection, or a documented allergy to chlorhexidine, isopropyl alcohol or povidoneiodine. Patients were randomized in a 1:1 ratio to surgical site skin antisepsis with a standard povidone-iodine -minute scrub and paint process or a 2-minute chlorhexidinealcohol scrub. The a priori size effect in our study was determined to be 3% based on data from the orthopedic literature. This resulted in a sample size of 49 patients in each group to detect a 3% difference with a power of.8 (.5). The randomization scheme was computer generated. Allocation concealment was achieved by concealment of the randomization list from the surgeon at all times by the impartial circulating nurse. The randomization group was revealed by the nurse just before preparation of the operating surgical suite. Skin preparation was performed by the circulating nurse or surgeon in all cases. Aerobic and anaerobic cultures were obtained by swabbing the incisional site with a dry, sterile, cotton tipped swab before skin preparation. A second culture was obtained from the surgical site after all drapes and instruments were on the field, just before incision. Specimens were sent immediately to the microbiology laboratory for culture and sensitivity. Patients were examined for local adverse reactions to the skin preparation during the immediate postoperative period and at the postoperative visit at 1 month. All data analyses were performed using SPSS version 15.. Descriptive statistics were created for all variables using means and standard deviations. Data analysis was performed using the intent to treat principle. Percentages were calculated for categorical data. Continuous variables were compared using the 2-tailed t test and categorical variables with the chi-square test. Multivariate logistic regression analysis was used to compare for differences in positive culture and clinical infection rates between the treatment groups. Odds ratios with 95% confidence intervals are presented. Statistical significance was defined as p.5. All data analyses were performed by an independent, unbiased statistician. RESULTS A total of patients were included in the study, with 5 assigned to each group. The baseline characteristics of the patient population are shown in the table. There were 65 patients who underwent placement of an IPP, 14 who underwent AUS placement, 14 who underwent MUS placement, 5 who underwent combination IPP MUS placement, and 2 who underwent placement of a testicular implant (fig. 1). All penile prostheses were implanted through a penoscrotal approach. Mean followup was 9.75 months (range 4 to 15). Mean operating room time was 86.2 minutes ( 32.2, range 45 to 24). Of the cultures 79% were positive before skin preparation and 2% were pos- Baseline patient characteristics Chlorhexidine Povidone-Iodine p Value Mean pt age (SD) 62.2 (13.2) 65.1 (9.).2 No. race: White Black 9 16 No. smoking status: Yes No No. diabetes: Yes No Mean kg/m 2 body mass index (SD) 3.5 (6.3) 31. (5.4).1

3 138 GENITOURINARY PROSTHETIC SKIN PREPARATION chlorhexidine-alcohol povidone-iodine 8 chlorhexidinealcohol 15 6 povidone-iodine IPP AUS MUS MUS + IPP Tes cular Prosthesis CNS SA PB FMS Figure 1. Distribution of patients Figure 3. Post-preparation culture results. SA, S. aureus. FMS, few mixed organisms. itive after skin preparation. Figure 2 displays the culture results before skin preparation. A total of 16 (32%) cultures were positive in the povidone-iodine group after skin preparation compared to 4 (8%) in the chlorhexidine-alcohol group (p.91, OR 5.3; CI 1.51, 18.5). Multivariate analyses were performed to evaluate the effect of baseline risk factors on post-preparation culture and clinical complications. Main effects included in the multivariate regression model were diabetes, smoking status, age, body mass index and race. The Hosmer-Lemeshow test was used to test the goodness of fit. Diabetes was found to be a significant risk factor for positive skin culture after skin preparation in our patient population (p.4, OR 3.38, CI 1.4, 11.). Furthermore, smoking status was found to be a significant risk factor for clinical complications (p.46, OR 4.92, CI 1.3, 23.26). The most commonly isolated organisms after skin preparation with povidone-iodine were CNS (13 patients), followed by mixed gram-positive organisms (2 patients) and Staphylococcus aureus (1 patient). Number of pa ents No Growth Mixed Org. Coag Neg Staph Pre Prep Cultures Group B Strep Staph Aureus Figure 2. Culture results before skin preparation. Org., organisms. Coag Neg Staph, CNS. Strep, streptococci. E. Faecalis, Enterococcus faecalis. Pseudomonas Bacillus Sp Aspergillus Niger Propionobacterium E. Faecalis Coag Neg Staph + Bacillus PB was the only organism isolated after skin preparation in the chlorhexidine-alcohol group (fig. 3). There was no significant difference in the incidence of clinical complications between the 2 groups. The overall incidence of clinical complications was 7%. A superficial wound infection developed in 1 patient and was successfully managed with oral antibiotics. Removal of the penile prosthesis was required in 5 patients, and urethral erosion of a MUS occurred in 1 patient and necessitated sling removal. Subset analyses were performed for the group of patients undergoing a penoscrotal incision (IPP) vs a perineal incision (MUS or AUS). A significant difference was noted in the post-preparation positive culture rate for povidone-iodine vs chlorhexidine-alcohol in the IPP subgroup (22.2% vs 8.8%, p.2). No significant difference was noted in the positive culture rate in the AUS/MUS subgroup (3.7% vs 6.7%, p.13). No urethral (dysuria) or genital complications (skin irritation) were noted in either group during the perioperative and postoperative period or at 1 and 6-month followup. DISCUSSION Device infection remains the most dreaded complication of GU prosthetic surgery. A device infection can lead to additional operations, which can result in significant patient morbidity. 3 Presumably the majority of infections are caused by contamination of the prosthetic by microorganisms at the time of implantation, resulting in biofilm formation. Biofilm, a layer that is formed around the implant surface, is composed of extracellular polymers and matrix produced by microorganisms that become attached to the surface of the implant. 11 This biofilm layer suppresses the host immune response and allows the organisms to remain viable with reduced

4 GENITOURINARY PROSTHETIC SKIN PREPARATION 139 growth rates, which can promote resistance to antibiotics. 12,13 Therefore, surgical site antisepsis is one of the most important steps in preventing prosthetic colonization, which can result in biofilm formation and lead to clinical infections. Chlorhexidine-alcohol has a rapid onset of action due to the alcohol component of the solution causing denaturation of protein and disruption of cell membranes. It also has antimicrobial persistence out to 48 hours because of the binding of chlorhexidine to the skin. 14,15 The appropriate application method is a 2-minute scrubbing followed by a 3-minute drying period to allow the alcohol to evaporate to prevent inadvertent ignition by electrocautery. In comparison, povidone-iodine functions by damaging proteins and DNA via free iodine that is released from solution after approximately 2 minutes of surface contact. 16 The recommended scrub time for povidone-iodine is 5 minutes, followed by a painting process, and then allowing the preparation to dry. This 2-step process typically takes approximately minutes and is the length of time advocated by many experts in GU prosthetic implantation. 17 Traditionally, chlorhexidine-alcohol has been considered contraindicated for use on the genitalia due to concerns of urethral irritation from the desiccating effects of the alcohol component of the preparation. Additionally, concerns over the flammability of alcohol on the genitalia have led to resistance to the use of alcohol based preparations on this region of the body. 18 As a result, many operating room nurses and surgeons advocate the use of aqueous based solutions instead, such as povidone-iodine. However, operating room fires due to the flammability of alcohol are exceedingly rare. As a result, the National Fire Protection Association, which sets national fire safety standards, lifted its ban on the use of alcohol based surgical preparation solutions in the operating room in Ignition of the alcohol can be prevented by allowing for proper drying and preventing the pooling of the alcohol based agent. Literature from other specialties has documented the superiority of chlorhexidine compared to povidone-iodine for surgical site antisepsis. Saltzman et al published a series showing a decreased skin contamination rate (7% vs 31%, respectively) on postpreparation culture swabs before orthopedic shoulder surgery. 9 In addition, Darouiche et al revealed a statistically significant decrease in wound infection rates (9.5% vs 16.1%, respectively) during clean contaminated surgery in a randomized trial comparing chlorhexidine to povidone-iodine for surgical site antisepsis. Our data show a lower positive culture rate with chlorhexidine-alcohol compared to povidone-iodine with a skin colonization rate of 8% vs 32%, respectively, after antisepsis. On the basis of our findings, 5 patients would need to undergo skin preparation with chlorhexidine-alcohol instead of povidone-iodine to prevent 1 positive skin culture. In a subset analysis chlorhexidine appeared to be superior for penoscrotal and perineal incisions, with the lack of statistical significance in the AUS/MUS subgroup presumably explained by the small number of positive cultures in this group (5). Additionally, chlorhexidine-alcohol appears to be superior to povidone-iodine at eliminating CNS, the most commonly isolated organisms from prosthetic biofilm. 2 PB was the predominant organism cultured from the surgical site after skin preparation with chlorhexidine. It has infectious potential, and has been described in infections of joint prostheses, shunts and prosthetic heart valves. 21 However, no reports to date have described the involvement of Propionibacterium in infections of urological implants. Further research is needed to define the role of this organism in GU prosthetic infections as well as the optimal method of eliminating this potentially pathogenic skin flora at the time of skin antisepsis. Diabetes status appeared to be a risk factor for positive skin culture after surgical site antisepsis in our patient cohort and the reason for this finding is unclear. However, it may be related to higher baseline bacterial colony counts or differences in the native skin flora in this population. This finding could, in part, explain the higher rates of clinical infection in patients with diabetes demonstrated in larger cohorts. 4,5 In addition, smoking status was a risk factor for clinical complications in our study. A population based analysis of National Surgical Quality Improvement Program data has shown smoking to be a risk factor for wound infections in breast cancer surgery. 22 To our knowledge, this is the first study to demonstrate smoking as a risk factor for clinical complications in prosthetic surgery. We hypothesize that this finding is related to the associated peripheral vascular disease and impaired wound healing in this patient population. Patients who smoke should be counseled regarding the potential increased risk of complications before prosthetic implantation. In addition to superior effectiveness in the elimination of skin flora, surgical site antisepsis with chlorhexidine-alcohol is more cost-effective in terms of operating room costs because of the decreased skin preparation time required for adequate antisepsis. At our institution the hospital cost for chlorhexidine-alcohol is almost 4% more than for povidone-iodine ($3.64 for povidone-iodine, $5.94 for chlorhexidine-alcohol). However, assuming operating room fees of $62 per minute and anesthesia fees of $4 per minute, the total cost of a standard - minute povidone-iodine paint and scrub would be

5 14 GENITOURINARY PROSTHETIC SKIN PREPARATION $525.7 more expensive than a 2-minute chlorhexidine preparation. 23 Even if the recommended 3-minute drying time recommended for chlorhexidine-alcohol is accounted for (which typically occurs during the draping process and adds no additional waiting period), skin preparation with chlorhexidine-alcohol is still $327.7 more cost-effective than povidoneiodine. Our study has several limitations. While we were able to demonstrate a significant difference in decreasing the rate of positive bacterial skin cultures at the surgical skin site with chlorhexidine-alcohol, we could not detect a significant difference in clinical infection rates because the study was not powered to do so. A study would require the enrollment of 1,141 patients to show a reduction in clinical infection rates from 4% to 2%. In addition, as a single center study, our results may not have external validity to other institutions where there may be differences in native skin flora, surgical techniques and operating room practices. We also analyzed qualitative rather than quantitative culture results. It is possible that quantitative analyses of culture specimens after skin preparation may show a correlation with clinical infection rates in a larger study, which we were not able to perform. Lastly, our sample represents a heterogeneous patient cohort, with possible differences in flora between the penoscrotal and perineal incisions. Nonetheless, effective elimination of skin flora before incision is a logical step in minimizing the risk of infection during prosthetic implantation. Given the lack of urethral or genital skin irritation and the cost benefit with the use of chlorhexidinealcohol, this agent appears to be the superior choice for surgical site antisepsis before GU prosthetic procedures. CONCLUSIONS Chlorhexidine-alcohol appears to be superior to povidone-iodine in eliminating skin flora at the surgical site before GU prosthetic procedures. Chlorhexidine-alcohol has persistent antimicrobial effects for up to 48 hours after application, is safe for use on the male genitalia and is more cost-effective than povidone-iodine. Although larger studies are needed to determine a difference in clinical outcome, chlorhexidine-alcohol appears to be the optimal agent for skin preparation before urological prosthetic procedures. ACKNOWLEDGMENTS Liu Yang assisted with statistical analysis. REFERENCES 1. Kojic EM and Darouiche RO: Candida infections of medical devices. Clin Microbiol Rev 24; 17: Stanford JL, Feng Z, Hamilton AS et al: Urinary and sexual function after radical prostatectomy for clincally localized prostate cancer: the Prostate Cancer Outcomes Study. JAMA 2; 283: Wilson SK and Costerton JW: Biofilm and penile prosthesis infections in the era of coated implants: a review. J Sex Med 212; 9: Jarow JP: Risk factors for penile prosthetic infection. J Urol 1996; 156: Wilson SK, Carson CC, Cleves MA et al: Quantifying risk of penile prosthesis infection with elevated glycosylated hemoglobin. J Urol 1998; 159: Wilson SK, Henry GD, Delk JR et al: Prevention of infection in revision of penile prosthesis using antibiotic coated prosthesis and Mulcahy salvage protocol. J Urol, suppl., 23; 169: 325, abstract Lai HH, Hsu EI, Teh BS et al: 13 Years of experience with artificial urinary sphincter implantation at Baylor College of Medicine. J Urol 27; 177: Guimarães M, Oliveira R, Pinto R et al: Intermediate-term results, up to 4 years, of a boneanchored male perineal sling for treating male stress urinary incontinence after prostate surgery. BJU Int 29; 3: Saltzman MD, Nuber GW, Gryzlo SM et al: Efficacy of surgical preparation solutions in shoulder surgery. J Bone Joint Surg Am 29; 91: Darouiche RO, Wall MJ, Itani KM et al: Chlorhexidine-alcohol versus povidone-iodine for surgical site antisepsis. N Engl J Med 2; 362: Jefferson KK: What drives bacteria to produce a biofilm? FEMS Microbiol Lett 24; 236: Von Eiff C, Neilman C and Peters G: New aspects in the molecular basis of polymer-associated infections due to Staphylococcus. Eur J Clin Microbiol Infect Dis 1999; 18: Costerton W, Veeh R, Shirtliff M et al: The application of biofilm science to the study and control of chronic bacterial infections. J Clin Invest 23; 112: Data on file. CareFusion, Inc., San Diego, California. 15. Hemani ML and Lepor H: Skin preparation for the prevention of surgical siteinfection: which agent is best? Rev Urol 29; 11: Trostle SS and Hartmann FA: Surgical infection. In: Equine Surgery, 2nd ed. Edited by JA Auer. Philadelphia: WB Saunders 1992; p Henry GD, Kansal NS, Callaway M et al: Centers of excellence concept and penile prostheses: an outcome analysis. J Urol 29; 181: Armstrong EP, Patrick KL and Erstad BL: Comparison of preoperative skin preparation products. Pharmacotherapy 21; 21: Segal CG: Infection control: start with skin. Nurs Manage 26; 37: Henry GD, Wilson SK, Delk JR 2nd et al: Revision washout decreases penile prosthesis infection in revision surgery: a multicenter study. J Urol 25; 173: Perry A and Lambert P: Propionibacterium acnes: infection beyond the skin. Expert Rev Anti Infect Ther 211; 9: de Blacum C, Ogunleye A and Momoh A: High body mass index and smoking predict morbidity in breast cancer surgery: a multivariate analysis of 26,998 patients from national surgical quality improvement database. Ann Surg 212; 255: Macario A: What does one minute of operating room time cost? J Clin Anesth 2; 22: 233.