Efficacy of Concurrent Application of Chlorhexidine Gluconate And

Efﬁcacy of concurrent application of chlorhexidine gluconate and povidone iodine against six nosocomial pathogens

Michele J. Anderson, PhD,a Mary E. Horn, BS,a Ying-Chi Lin, PhD,a PatrickJ.Parks,MD,PhD,a,b and Marnie L. Peterson, PharmD, PhDa Minneapolis and St. Paul, Minnesota

Background: Chlorhexidine gluconate (CHG) and povidone iodine (PI) are rarely used concurrently despite a lack of evidence regarding functional incompatibility of these agents. Methods: CHG and PI, alone and combined, were evaluated against Staphylococcus aureus (methicillin-suseptible S aureus [MSSA] and methicillin-resistant S aureus [MRSA]), Staphylococcus epidermidis (MRSE), Acinetobacter baumannii, Pseudomonas aeruginosa, and Escherichia coli using checkerboard microbroth dilution techniques. Minimum bactericidal concentration (MBC) was the con- $ centration (percent wt/vol) that reduced bacterial burden 5-log10 colony-forming units/mL at 2 hours when compared with bacterial densities in growth controls. Fractional bactericidal concentration indexes (FBCIs) were calculated to determine CHG and PI compatibility. Additionally, tissue plugs from freshly excised porcine vaginal mucosa were infected with S aureus (MSSA), treated for 2 hours with CHG 3%, PI 5%, or CHG 3% and PI 5% combined and then viable bacteria on the tissue plugs enumerated. Results: In broth, CHG demonstrated dose-dependent bactericidal activity, whereas PI activity was all-or-none. All isolates studied were similarly susceptible to CHG (MBCs: 0.0078% 6 0.0019%, 0.0069% 6 0.0026%, 0.0024% 6 0.0005%, 0.0024% 6 0.0005%, 0.0059% 6 0.0%, and 0.0029% 6 0.0%, respectively). The MBCs of PI were identical (0.625%) for all isolates. Overall, FBCI calculations showed indifference. Treatment of MSSA-infected porcine tissue for 2 hours demonstrated that the CHG-PI combination was superior to either antiseptic alone. Conclusion: FBCIs, determined in broth culture, indicate that combining CHG and PI had no negative impact on antisepsis. More- over, data from an ex vivo porcine mucosal infection model suggest a potential beneﬁt when combining the 2 antiseptic agents. Key Words: Chlorhexidine gluconate; povidone iodine; antiseptic; bacteria; fractional bactericidal concentration index. Copyright ª 2010 by the Association for Professionals in Infection Control and Epidemiology, Inc. Published by Elsevier Inc. All rights reserved. (Am J Infect Control 2010;38:826-31.)

Nosocomial infection prevention involves a multi- bacteria will develop resistance, whereas antibiotics pronged approach: infection control, antibiotic pro- tend to have specific intracellular targets. Also, resis- phylaxis, and antisepsis.1 The emergence of tance to antiseptics may be overcome by increasing multidrug-resistant organisms underscores the impor- the concentration or time of exposure.3 Chlorhexidine tance of using antiseptic agents as a strategy for infec- gluconate (CHG) and polyvinylpyrrolidone iodine (po- tion prevention.2 In general, antiseptics have a broader vidone iodine; PI) are 2 of the most commonly used spectrum of activity than antibiotics. Antiseptics often antiseptic agents for antimicrobial skin preparation have multiple targets, reducing the likelihood that prior to surgery and central venous catheter (CVC) in- sertion.4 Many studies have compared the individual a From the University of Minnesota, Minneapolis, MN ; and 3M Company, antiseptic activity of these 2 compounds, but few stud- St. Paul, MN.b ies have evaluated the activity of the compounds in Address correspondence to Marnie L. Peterson, PharmD, PhD, Depart- combination. In addition, studies investigating the ment of Experimental and Clinical Pharmacology, College of Pharmacy, University of Minnesota, Room 4-212 McGuire Translational Research combined effects of CHG and PI have applied the anti- Facility, Minneapolis, MN 55455. E-mail: [email protected]. septics sequentially. Langgartner et al concluded that Supported by a grant from the 3M Company, St. Paul, MN (to M.L.P.). skin disinfection with a propanol/CHG solution fol- Previously presented as an abstract at The Society for Healthcare lowed by PI was superior to either regimen alone in Epidemiology of America (SHEA) meeting, 2009, No. 449. preventing CVC colonization, whereas Guzel et al Conflicts of interest: P.J.P. is an employee of 3M Company and provided found that cleaning the skin with 15% CHG followed editorial assistance during manuscript preparation. The remaining by 10% PI was safe and effective for skin antisepsis authors report no conflicts of interest. prior to neurosurgical intervention.5,6 Our study exam- 0196-6553/$36.00 ined the antimicrobial effects of an aqueous solution Copyright ª 2010 by the Association for Professionals in Infection containing both CHG and PI. Therefore, the bacteria Control and Epidemiology, Inc. Published by Elsevier Inc. All rights were exposed to the 2 compounds simultaneously. reserved. CHG is a polycationic bisbiguanide with broad- doi:10.1016/j.ajic.2010.06.022 spectrum antimicrobial activity. CHG damages the

826 www.ajicjournal.org Anderson et al. 827 Vol. 38 No. 10 outer bacterial surface layers, thereby promoting its forming units (CFU)/mL in cation-adjusted Mueller Hinton own uptake and subsequently attacking the cytoplas- broth. The number of bacteria in the culture was verified mic or inner membrane of the organism.3,7 At low using standard saline dilution and plating techniques. concentrations, CHG reduces membrane fluidity, os- The adjusted bacterial suspensions were used to inoculate moregulation, and metabolic capacity.8 At higher con- the 96-well microtiter plates resulting in inoculums of centrations, such as those in commercially available ;106 CFU/mL. preparations, the membrane becomes liquid crystal- MSSA (MN8) was chosen as the model organism in line and loses its integrity.8 CHG has broad-spectrum the studies utilizing the ex vivo porcine model of infec- activity against bacteria and yeast. It has low antiviral tion. A 1-mL aliquot of an overnight culture grown in activity and is not sporicidal but can prevent spore Todd Hewitt broth at 378C was centrifuged to pellet development.3 In addition to its direct effect on mi- the cells and washed in RPMI 1640 media (no supple- crobes, CHG can bind to the outermost layer of the ments). The washed cell pellet was resuspended in ; epidermis and to mucous membranes, which provides RPMI 1640 media to an OD595 of 0.5, which resulted a lingering or persistent antimicrobial effect.9,10 in ;1 3 106 CFU in the 2-mL volume used to infect the PI is a complex of iodide and a solubilizing carrier, tissue plugs. polyvinylpyrrolidone, which acts as a reservoir of ‘‘free’’ iodine (the active component). The iodine is Antiseptics slowly released and delivered to the bacterial cell surface Aqueous CHG 3% wt/vol was prepared from a where it penetrates the cell membrane and inactivates 20% wt/vol stock (Sigma-Aldrich Corporation, St. Louis, key cytosolic proteins, fatty acids, and nucleotides.11 MO) in water and further diluted before addition to mi- Slow release of iodine from the PI complex in solution crotiter plates. CHG was tested over a concentration minimizes iodine toxicity towards mammalian cells. range of 0.0018% to 0.375%. The concentration (per- Iodine has broad-spectrum antibacterial activity, as cent wt/vol) of CHG in commercial products ranges well as activity against fungi, protozoa, viruses, and from 0.12% in oral rinses to 4% in topical skin some bacterial spores.12 cleansers. CHG and PI have different cellular targets and differ- Polyvinylpyrrolidone iodine (PI) 10% wt/vol (Sigma- ent mechanisms of action. These differences may Aldrich Corporation) was prepared from powder, in wa- prove beneficial when using these 2 antiseptics in com- ter. PI was tested over a concentration range of 0.078% bination. CHG damages the outer membrane, which, in to 2.5%. The concentration of PI in commercial prepa- turn, would augment access to the intracellular targets rations is usually 10%, but ophthalmic preparation so- necessary for the bactericidal action of PI. Moreover, lutions contain only 5%. PI’s activity is more immediate, whereas CHG’s activity Higher concentrations were required for antiseptic occurs later and lingers,9,13 implying that these 2 com- activity in the porcine ex vivo mucosal experiments pounds may work cooperatively. In clinical practice, due to the interaction of CHG and PI with mucosal pro- the use of CHG and PI in combination is commonly teins. The concentrations of CHG and PI alone were 3% avoided despite a lack of evidence regarding functional and 5%, respectively. After combining a 6% solution of incompatibility of these agents. The aim of this study CHG with an equal volume of 10% PI, the final concen- was to determine whether the combined activity of trations of CHG and PI in the mixture were the CHG and PI against clinically relevant pathogens is same as the individual concentrations: 3% and 5%, inferior to the activity of either agent alone. respectively.

Microdilution assay MATERIALS AND METHODS Bacteria The checkerboard method was used to determine the minimum bactericidal concentration (MBC) of Clinical isolates of methicillin-sensitive and methicillin- each antiseptic agent, both alone and in combination. resistant Staphylococcus aureus (MSSA and MRSA, respec- We deﬁned the MBC as the concentration that reduced tively), methicillin-resistant Staphylococcus epidermidis the bacterial burden by $5-log10 CFU/mL when com- (MRSE), multidrug-resistant Acinetobacter baumannii (re- pared with the bacterial burden in the growth control sistant to cefazolin, cefuroxime, and ceftriaxone), Pseudo- wells. This was an adaptation of the French standard monas aeruginosa,andastrainofEscherichia coli (resistant T72300, which deﬁnes an antiseptic as bactericidal if 8 to cefotaxime) were grown in Todd Hewitt broth at 37 C the challenge inoculum is reduced by 5 log10 following overnight. For the checkerboard microdilution assays, exposure for 1 minute.14 We selected a 2-hour end the overnight cultures were adjusted to an OD595 (optical point to assess sustained bacterial killing based on density) of ;1.0 or bacterial density of ;107 colony- the approximate time elapsed from preoperative skin 828 Anderson et al. American Journal of Infection Control December 2010

A. baumannii preparation to conclusion of routine surgical proce- A 8 dures.15,16 CHG and PI were 2-fold serially diluted E. coli across and down microtiter plates containing cation- MSSA 6 adjusted Mueller Hinton broth. The bacterial suspen- MRSA sion that was prepared from the overnight culture S. epidermidis was used to inoculate the microtiter plates. The desired 4 P. aeruginosa inoculum was ;1 3106 CFU/mL. The plates were incu- bated at 378C for 2 hours, and then an antiseptic neu- CFU/mL (log10) 2 tralizing solution containing Triton X-100, Tween-80, lecithin, and sodium thiosulfate was added to each 0 0 well. After neutralization, the number of viable bacteria 0.001 0.01 CHG (% w/v) in each well was determined by saline dilution and plating techniques using a WASP2 spiral plater B 8 (Microbiology International, Frederick, MD). After over- A. baumannii night incubation at 378C, the colonies on the plates E. coli 6 MSSA were counted using an automated colony counter (Pro- MRSA toCOL; Microbiology International). 4 S. epidermidis Fractional bactericidal concentration index P. aeruginosa

CFU/mL (log10) 2 Fractional bactericidal concentration indexes (FBCIs) were calculated to determine the compatibility of CHG 0 and PI when used in combination, as have been de- 0.0 0.2 0.4 0.6 scribed previously.17,18 The concentration of PI in the PI (% w/v) $ presence of CHG that produced a 5-log10 decrease in Fig 1. Dose reponse of CHG and PI. (A) CHG. (B) bacterial density was divided by the concentration of PI. Results presented are representative of $ 3 PI that produced the same effect when used alone. independent experiments. CFU, colony-forming The same calculation was performed for CHG. The units. sum of these 2 fractions is the FBCI: antibiotic-free RPMI 1640 media 3 times. The plugs ðMBC PIin combinationÞ=ðMBC PIaloneÞ1 were then placed mucosal side up on a 0.4-mm cell cul- ðMBC CHG Þ=ðMBC CHG Þ in combination alone ture insert in 6-well plates containing fresh serum and antibiotic-free RPMI 1640. Bacterial suspension (2 mL) containing approximately 1 3106 CFU of the MSSA iso- The results were categorized as follows: FBCI , 0.5 late was pipetted onto the outer mucosal surface of (S 5 synergistic); 0.5 $ FBCI , 1.0 (PS 5 partially syn- each tissue plug. Inoculated tissue plugs were incu- ergistic); FBCI 5 1.0 (A 5 additive); 1.0 . FBCI # 4.0 bated at 378Cin7%CO for 2 hours prior to treatment (I 5 indifferent); and FBCI . 4.0 (AN 5 antagonistic). 2 with antiseptics to allow the bacteria to recover. Unin- The FBCI was determined for each organism a minimum oculated tissue plugs were included as controls. After of 3 times. 2 hours incubation of infected or control tissue plugs, m Ex vivo porcine vaginal mucosa culture 10 L of each antiseptic agent alone or a mixture of both was applied to the mucosal surface using a pipet. An ex vivo porcine vaginal mucosal infection model Tissue plugs were returned to the incubator (378C, 7% was utilized to determine the efﬁcacy of CHG and PI CO2) for a 2-hour treatment period. After incubation, alone and in combination in the presence of biologic tissue plugs were transferred to 5-mL centrifuge tubes tissues and proteins. Specimens of normal porcine vag- containing 250 mL of the same solution that was used inal mucosa were excised from animals at slaughter to neutralize the activity of CHG and PI in the broth and transported to the laboratory in RPMI 1640 media studies (Triton X-100, Tween-80, lecithin, and sodium supplemented with 10% fetal calf serum, penicillin (50 thiosulfate). Each plug was homogenized for 30 sec- IU/mL), streptomycin (50 mg/mL), and amphotericin B onds using an Omni International tissue homogenizer (2.5 mg/mL). Tissue was utilized within 3 hours of exci- (Marietta, GA) at the highest setting to release bacteria sion. Tissue plugs of uniform size were obtained from into solution. Tissue homogenates that were either the porcine vagina using a 5-mm biopsy punch. Excess neat or serially diluted in phosphate-buffered saline muscle tissue was trimmed away with a scalpel. Tissue were applied to sheep blood agar plates using a spiral plugs were washed in fetal calf serum-supplemented plater (Microbiology International). The following day, www.ajicjournal.org Anderson et al. 829 Vol. 38 No. 10

Table 1. MBCs and FBCIs of PI and CHG alone and in combination against 6 nosocomial pathogens*

y Organism MBC of PI alone MBC of PI combined MBC of CHG alone MBC of CHG combined FBCI Result

MSSA 0.625 0.625 0.0078 0.0078 2.00 I MRSA 0.625 0.625 0.0069 0.0092 2.33 I Staphylococcus epidermidis 0.625 0.556 0.0024 0.0023 1.85 I Acinetobacter baumannii 0.625 0.519 0.0024 0.0017 1.49 I Pseudomonas aeruginosa 0.625 0.519 0.0059 0.0049 1.66 I Escherichia coli 0.625 0.625 0.0029 0.0053 2.83 I

MBC, minimum bactericidal concentration deﬁned as the concentration that reduced the bacterial burden by "greater than or equal to" 5-log10 CFU/mL when compared with the bacterial burden of the growth control. *Average MBCs of PI and CHG alone, average MBCs of PI when combined with CHG, average MBCs of sub-MBC concentrations of CHG when combined with sub-MBC concentrations of PI, and the calculated FBCIs of the combination. y FBCI 5 (MBC PIin combination)/(MBC PIalone) 1 (MBC CHGin combination)/(MBC CHGalone); FBCI , 0.5 (S 5 synergistic); 0.5 $ FBCI , 1 (PS 5 partially synergistic); FBCI 5 1 (A 5 additive); 1.0 . FBCI # 4.0 (I 5 indifferent); and FBCI . 4.0 (AN 5 antagonistic).

$ colonies were enumerated using an automated colony necessary to produce a 5-log10 decrease in the num- counter (ProtoCOL; Microbiology International). ber of viable bacteria was signiﬁcantly reduced when that agent was combined with sub-MBC concentrations RESULTS of the second agent. For the MRSA isolate, when sub- MBCs of PI (,0.625%) were combined with a sub- This study determined the antiseptic activity of CHG MBC of CHG (,0.0069%), the CHG curve was not and PI alone and when applied simultaneously against affected, ie, no shift (Fig 2A). When sub-MBCs of CHG clinically relevant nosocomial pathogens: MSSA, were combined with a sub-MBC of PI, the bacterial MRSA, MRSE, A baumannii, P aeruginosa,andE coli. load was reduced, but the concentration of PI required , $ In general, low concentrations of CHG ( 0.00293%, to produce 5-log10 decrease in the number of viable MRSA; ,0.000365, A baumanii) were determined to bacteria remained at 0.625% (Fig 2B). The data for have no effect on CFU/mL over the 2-hour time frame A baumannii showed a slight shift to the left when of the microbroth experiments. These results implied sub-MBCs of PI (,0.625%) were combined with CHG that CHG was bacteriostatic at low concentrations (Fig (,0.0024%). However, this shift was not great enough 1A), which was consistent with previous reports.7 At to suggest additivity or synergism (Fig 2C). When sub- higher concentrations (.0.008%), CHG was bacteri- MBCs of CHG were combined with PI at a sub-MBC, $ cidal ( 5-log10 decrease in CFU/mL) for all organisms the number of bacteria recovered decreased. The higher tested. Conversely, the activity of PI was all-or-none concentrations of CHG (0.002930% and 0.005860%) (Fig 1B). Once the bactericidal concentration threshold were above the MBC of CHG alone (0.0024%) and, pre- $ (0.625%) was reached, there was a 5-log10 decrease dictably, produced a $5-log10 decrease in the number of in the number of viable bacteria for all studied viable bacteria (Fig 2D). organisms. Table 1 lists the FBCIs for each of the bacteria, which All study organisms were similarly susceptible to CHG. were calculated to determine the overall effect of com- MBCs of CHG alone were 0.0078% 6 0.0019%, bining CHG and PI, ie, antagonistic, synergistic, addi- 0.0069% 6 0.0026%, 0.0024% 6 0.0005%, 0.0024% 6 tive, or indifferent. Overall, using the microbroth 0.0005%, 0.0059% 6 0.0%, and 0.0029% 6 0.0% for dilution assay, the combination of CHG and PI had no MSSA, MRSA, MRSE, A baumannii, P aeruginosa, and effect on antisepsis; FBCIs 5 indifferent for all orga- E coli, respectively (Ta b l e 1 ). The MBC of PI alone was nisms (1.0 . FBCI # 4.0). In 3 independent experi- identical (0.625%) in 3 or more independent experiments, the average FBCIs for MSSA, MRSA, MRSE, ments and across the different species tested (Table 1). A baumannii, P aeruginosa, and E coli were 2.00, 2.33, As reported previously, CHG was a great deal more po- 1.85, 1.49, 1.66, and 2.83, respectively. The average tent than PI.13,19,20 FBCI for E coli was the highest at 2.83 but still within Bactericidal activities of CHG and PI combined the range of indifference for 2 out of 3 experiments. against MRSA, a representative gram-positive organism, Finally, treatment of MSSA-infected porcine mucosal and A baumannii, a representative gram-negative orga- tissue with either 5% PI or 3% CHG individually for nism, are illustrated in Fig 2. Synergy was determined if 2 hours resulted in numerically lower bacterial counts the kill-curve generated using data from the 2-agent compared with untreated control (Fig 3). When PI and combinations shifted to the left of the curve that was CHG were combined in solution before applying to the generated using single agent data, ie, the concentration infected tissue, the resulting decrease in bacterial bur- of the single agent (depicted on the x-axis) that was den (.3-log10) was signiﬁcantly different from control 830 Anderson et al. American Journal of Infection Control December 2010

MRSA-CHG A. baumannii - CHG A 8 C 8 PI % PI % 0 6 0 6 0.078 0.078 0.156 0.156 4 0.313 4 0.313 0.625 0.625 2 2 CFU/mL (log10) CFU/mL (log10)

0 0 0 0.001 0.01 0.1 0 0.001 0.01 0.1 % CHG % CHG

MRSA-PI A. baumannii - PI B 8 CHG % D 8 CHG % 0 0 6 0.00073 6 0.000183 0.00146 0.000365 0.00293 0.000730 4 0.00586 4 0.001460 0.01172 0.002930 2 2 0.005860 CFU/mL (log10) CFU/mL (log10)

0 0 0 0.1 1 0 0.1 1 % PI % PI Fig 2. Antibacterial effects of combinations of CHG and PI. (A) MRSA: effect of adding PI to CHG; and (B) MRSA: effect of adding CHG to PI. (C) Acinetobacter baumannii: effect of adding PI to CHG; and (D) A baumannii: effect of adding CHG to PI. Results presented are the average of 3 independent experiments. CFU, colony-forming units.

(P , .05, analysis of variance followed by Bonferroni 2-hour end point to more closely approximate in-use post-test for multiple comparisons, n 5 3). conditions when antiseptics are employed to decon- taminate skin prior to surgery and to evaluate the pro- 15,16 DISCUSSION longed effect during surgery. Second, we defined the MBC as the concentration (percent wt/vol) that re- The aim of this study was to evaluate the potential duced the bacterial burden by $5-log10 CFU/mL interaction between CHG and PI and the effect the com- when compared with the bacterial burden in the bination had on antimicrobial activity using checker- growth control wells when typically MBC is defined 9,23,24 board microdilution assays and infected porcine as 99.9% or 3-log10 kill of the original inoculum. mucosal tissue explants. The efficacy of these antisep- FBCI was calculated using the MBC data obtained using tics used individually is well established. However, to broth microdilution checkerboard techniques. FBCI, our knowledge, this is the first report evaluating the also referred to as the interaction index or SFBC, math- efficacy of these 2 antiseptic agents used concurrently. ematically expresses the interaction between 2 antimi- PI has rapid bactericidal activity, but activity is dimin- crobial agents and is an accepted method for ished shortly after contact with organic matter present determining synergism.25 Finally, we chose to validate in skin.21,22 CHG is not as fast acting but exhibits sus- the in vitro findings using an ex vivo porcine mucosal tained antimicrobial activity and is not readily neutral- model because in vitro broth techniques have been ized by organic matter.7 We surmised that these 2 shown to overestimate the antibacterial efficacy of agents would compliment one another by creating an biocides.3,21,22 initial and rapid kill of resident flora followed by a sus- The results from the checkerboard microdilution as- tained antibacterial effect, which could prove quite says performed using the 6 nosocomial organisms indi- beneficial in the clinical setting. cate that there is no functional incompatibility when Our strategy for determining the efficacy of PI and combining CHG and PI. In fact, at sub-MBC concentra- CHG when used concurrently against nosocomial path- tions, CHG may enhance the antiseptic activity of PI ogens differed from previously published studies. First, alone. Furthermore, our ex vivo full-thickness mucosal we chose a well-established in vitro checkerboard mi- model of infection suggests that there may be a benefit crobroth dilution method but determined the efficacy to combining the 2 antiseptic agents against MSSA. It is at 2 hours instead of the 18 to 24 hours used commonly well-known that CHG and PI are useful skin antiseptics, when determining minimum inhibitory concentrations but simultaneous application of these 2 agents for use (MICs) and MBCs in other studies.9,23,24 We chose a in clinical situations is rare despite a lack of evidence to www.ajicjournal.org Anderson et al. 831 Vol. 38 No. 10

6 a N=3 5. Langgartner J, Linde HJ, Lehn N, Reng M, Scholmerich J, Gluck T. p<0.05 Combined skin disinfection with chlorhexidine/propanol and aqueous a,b povidone-iodine reduces bacterial colonisation of central venous cath- 4 a,b eters. Intensive Care Med 2004;30:1081-8. 6. Guzel A, Ozekinci T, Ozkan U, Celik Y, Ceviz A, Belen D. Evaluation of b the skin ﬂora after chlorhexidine and povidone-iodine preparation in 2 neurosurgical practice. Surg Neurol 2009;71:207-10. 7. Jones CG. Chlorhexidine: is it still the gold standard? Periodontol

CFU/g tissue (log10) 2000 1997;15:55-62. 0 8. Gilbert P, Moore LE. Cationic antiseptics: diversity of action under a common epithet. J Appl Microbiol 2005;99:703-15. d ) 5% (3%) 9. Fuursted K, Hjort A, Knudsen L. Evaluation of bactericidal activity and I ( treate P G (3%) n H lag of regrowth (postantibiotic effect) of five antiseptics on nine bac- u C + CHG terial pathogens. J Antimicrob Chemother 1997;40:221-6. (5%) I 10. Lorian V, editor. Antibiotics in laboratory medicine. Fifth edition Phil- P adelphia: Lippincott Williams and Wilkins; 2005. Treatment (2 h exposure) 11. Durani P, Leaper D. Povidone-iodine: use in hand disinfection, skin Fig 3. CHG and PI do not antagonize each other preparation and antiseptic irrigation. Int Wound J 2008;5:376-87. when used to treat MSSA infected ex vivo porcine 12. Rutala WA. APIC guideline for selection and use of disinfectants. 1994, 1995, and 1996 APIC Guidelines Committee. Association for Profes- vaginal mucosal tissue. Treatments were applied for sionals in Infection Control and Epidemiology, Inc. Am J Infect Control 2 hours, and those not sharing a letter are 1996;24:313-42. significantly different, ie, ‘‘a’’ is different from CHG 13. Chaiyakunapruk N, Veenstra DL, Lipsky BA, Saint S. Chlorhexidine 1 PI, ‘‘b’’ is different from control, and ‘‘a,b’’ are not compared with povidone-iodine solution for vascular catheter-site different from any treatment; P , .05, n 5 3, care: a meta-analysis. Ann Intern Med 2002;136:792-801. 14. Hill RLR, Casewell MW. The in vitro activity of povidone-iodine cream representative of 3 individual experiments. against Staphylococcus aureus and its bioavailability in nasal secretions. J Hosp Infect 2000;45:198-205. support incompatibility. Recently, there have been 2 re- 15. Mangram AJ, Horan TC, Pearson ML, Silver LC, Jarvis WR. Guideline ports in the literature in which sequential use of CHG for prevention of surgical site infection, 1999. Hospital Infection Con- trol Practices Advisory Committee. Infect Control Hosp Epidemiol and PI was superior to use of either agent alone. Lang- 5 1999;20:250-78. gartner et al demonstrated that bacterial colonization 16. Darouiche RO, Wall MJ Jr, Itani KM, Otterson MF, Webb AL, Carrick of CVCs was significantly reduced when skin was disin- MM, et al. Chlorhexidine-alcohol versus povidone-iodine for surgical- fected with a propanol/CHG preparation for 1 minute site antisepsis. N Engl J Med 2010;362:18-26. followed by PI disinfection for 1 minute prior to CVC in- 17. Tascini C, Gemignani G, Ferranti S, Tagliaferri E, Leonildi A, Lucarini A, et al. Microbiological activity and clinical efficacy of a colistin and rif- sertion. The authors conclude that CHG and PI may 5 ampin combination in multidrug-resistant Pseudomonas aeruginosa in- have synergistic disinfectant activity. In another study, fections. J Chemother 2004;16:282-7. skin flora was evaluated after cleaning the skin with 18. Biancofiore G, Tascini C, Bisa M, Gemignani G, Bindi ML, Leonildi A, CHG and PI for preparation prior to neurosurgery.6 et al. Colistin, meropenem and rifampin in a combination therapy for Guzel et al showed that a small amount of bacteria per- multi-drug-resistant Acinetobacter baumannii multifocal infection: a case report. Minerva Anestesiol 2007;73:181-5. sisted after the CHG preparation but that, when CHG 19. Maki DG, Ringer M, Alvarado CJ. Prospective randomized trial of was followed by PI application, the cultures showed povidone-iodine, alcohol, and chlorhexidine for prevention of infec- no growth. Together, our data and these recent reports tion associated with central venous and arterial catheters. Lancet suggest that there may be clinical benefits to using a 1991;338:339-43. combination of these agents to prepare skin prior to 20. Lio PA, Kaye ET. Topical antibacterial agents. Infect Dis Clin North Am 2004;18:717-33. surgery. 21. Messager S, Goddard PA, Dettmar PW, Maillard JY. Determination of the antibacterial efficacy of several antiseptics tested on skin by an ex The authors thank Laurie Baeker Hovde for her assistance writing this manuscript. vivo test. J Med Microbiol 2001;50:284-92. 22. Maillard JY, Messager S, Veillon R. Antimicrobial efficacy of biocides References tested on skin using an ex vivo test. J Hosp Infect 1998;40:313-23. 1. Traore O, Fayard SF, Laveran H. An in vitro evaluation of the activity of 23. Karpanen TJ, Worthington T, Hendry ER, Conway BR, Lambert PA. povidone-iodine against nosocomial bacterial strains. J Hosp Infect Antimicrobial efficacy of chlorhexidine digluconate alone and in com- 1996;34:217-22. bination with eucalyptus oil, tea tree oil and thymol against planktonic 2. Jones RN. Resistance patterns among nosocomial pathogens: trends and biofilm cultures of Staphylococcus epidermidis. J Antimicrob Che- over the past few years. Chest 2001;119:S397-404. mother 2008;62:1031-6. 3. McDonnell G, Russell AD. Antiseptics and disinfectants: activity, 24. DeBaun B. Evaluation of the antimicrobial properties of an alcohol- action, and resistance. Clin Microbiol Rev 1999;12:147-79. free 2% chlorhexidine gluconate solution. AORN J 2008;87:925-33. 4. Valles J, Fernandez I, Alcaraz D, Chacon E, Cazorla A, Canals M, et al. 25. Hallander HO, Dornbusch K, Gezelius L, Jacobson K, Karlsson I. Syn- Prospective randomized trial of 3 antiseptic solutions for prevention ergism between aminoglycosides and cephalosporins with antipseudo- of catheter colonization in an intensive care unit for adult patients. monal activity: interaction index and killing curve method. Antimicrob Infect Control Hosp Epidemiol 2008;29:847-53. Agents Chemother 1982;22:743-52.