Povidone Iodine: Properties, Mechanisms of Action, and Role in Infection Control and Staphylococcus aureus Decolonization

Nasal decolonization is an integral part of the strategies used to control and prevent the spread of methicillin-resistant Staphylococcus aureus (MRSA) infections. The two most commonly used agents for decolonization are intranasal mupirocin 2% ointment and chlorhexidine wash, but the increasing emergence of resistance and treatment failure has underscored the need for alternative therapies. This article discusses povidone iodine (PVP-I) as an alternative decolonization agent and is based on literature reviewed during an expert’s workshop on resistance and MRSA decolonization.

Despite active surveillance efforts, advances in the prevention of infection and new antibiotics, MRSA remains a prominent pathogen associated with high rates of mortality (2). Decolonization, the goal of which is to decrease or eliminate bacterial load on the body, is an integral part of the strategies used to control and prevent the spread of MRSA (13). This approach involves eradication of MRSA carriage from the nose may be more clinically relevant, since in vitro studies do not take into account host proteins that can neutralize antiseptic activity (36). In contrast, substances commonly present in test media and diluents (e.g., sulfur-containing amino acids) may negate antiseptic activity and lead to false-negative results (37).
The interpretation of in vitro studies of PVP-I is also complicated by the paradoxical increase in bactericidal activity with dilutions up to a 0.1% strength solution. This is related to the subtle equilibrium between bound and unbound iodine; the concentration of the latter active compound, I 2 , follows a bell-shaped curve as the dilution is increased. The activity of PVP-I against S. aureus correlates to this, and several studies have observed decreased activity at concentrations above and below 0.1% (25)(26)(27)(38)(39)(40)(41).
In vitro studies have confirmed the bactericidal activity of PVP-I 10% solution against clinical isolates of MSSA and MRSA using both suspension tests (28,37,39,(42)(43)(44)(45) and surface test methods (46). In these studies, the action of PVP-I against MSSA and MRSA was rapid, with bactericidal activity typically observed within 15 to 60 s (39,43,45). Comparative in vitro studies have generally showed that, irrespective of exposure time or dilution, 10% PVP-I was more active than chlorhexidine against MRSA and was bactericidal against chlorhexidine-resistant strains (28,39,42,44,46). Furthermore, two in vitro studies also showed that 5% PVP-I cream had bactericidal activity against MSSA, MRSA, and mupirocin-resistant strains of MRSA (40,43). However, the in vitro activity of PVP-I, but not mupirocin, was reduced by the addition of nasal secretions (40).  (23,24). In aqueous solution, a dynamic equilibrium occurs between free iodine and the PVP-I complex (25,26). In an ex vivo model of porcine vaginal mucosa infected with MSSA, 7.5% PVP-I solution significantly reduced bacterial load (measured in log CFU/explant) at 0.25 to 4 h compared to untreated controls, although some regrowth was evident at 24 h (36).
In the same porcine model infected with MRSA, a skin and nasal preparation (SNP) of 5% PVP-I significantly reduced viable bacterial cells after 1 h versus the control (1.09 Ϯ 0.57 versus 5.30 Ϯ 0.06 log 10 CFU/explant, respectively; P Ͻ 0.05), with sustained activity over 12 h (P Ͻ 0.05 versus the control) (47). Interestingly, an ophthalmic iodine preparation lacked this sustained activity, only significantly differing from control at 1 h (2.51 Ϯ 0.20; P Ͻ 0.05). Conversely, mupirocin had a slower onset of action (5.14 Ϯ 0.09 at 1 h, 5.07 Ϯ 0.06 at 6 h; P Ͼ 0.05), with significant bactericidal effects evident only after 12 h ( Fig. 2A) (47). The SNP of 5% PVP-I was also significantly more effective than mupirocin against low-level and high-level mupirocin-resistant MRSA isolates in this porcine model (P Ͻ 0.05) (47). Finally, in a human skin model infected with MRSA, both the SNP and the ophthalmic PVP-I-based 5% preparation were significantly bactericidal after 1, 6, and 12 h compared to controls (P Ͻ 0.05), whereas mupirocin was only significantly bactericidal versus control after 12 h (P Ͻ 0.05) (Fig.  2B) (47).
Potential for resistance. As with antibiotics, resistance to antiseptics (the ability of a bacterial strain to survive the use of an antiseptic that could previously eliminate that strain) can occur in bacteria because of intrinsic properties (e.g., biofilm formation, endospores, and expression of intrinsic mechanisms) or can be acquired via mutation or external genetic material (plasmids or transposons) (24,48,49). However, less is known about fungal or viral mechanisms of resistance to antiseptics (24,48). Although there are no standard criteria for evaluating the capability of an antiseptic agent to induce or select for antibiotic resistance in bacteria, a protocol based on evaluating the change in susceptibility profile to the antiseptic and antibiotics has been proposed (50,51). Based on available reports, no link has been observed between PVP-I and the development of resistance, probably due to its numerous and simultaneous molecular targets (e.g., double bonds, amino groups, and sulfydral groups) (23,47,(52)(53)(54)(55)(56). To date, evidence suggests that PVP-I does not select for resistance among staphylococci (37) and Pseudomonas aeruginosa, Serratia marcescens, Escherichia coli, and Klebsiella aerogenes in vitro (28,57) or among staphylococci after long-term clinical application to catheter exit sites (58). Isolated reports of slow bactericidal activity (27,41) or apparent resistance to PVP-I (41, 59) are available, but they were later attributed to difficulties in determining its in vitro activity and/or the use of culture conditions antagonistic to its action (40). Furthermore, reports of intrinsic contamination of 10% PVP-I solution with Burkholderia (formerly Pseudomonas) cepacia were concerning when initially published in the 1980s and early 1990s (60-66) but were subsequently attributed to confounding factors in the manufacturing process (62). The occurrence of such phenomena has not been widely reproduced with PVP-I, and similar reports of contamination with other antiseptics were also published during this time period (53,67). Finally, unlike other antiseptics, there have been no reports of PVP-I inducing horizontal gene transfer, antibiotic resistance genes, or cross-tolerance and cross-resistance to antibiotics and other antiseptics (20,53).
(ii) Efficacy of intranasal PVP-I decolonization. One large randomized, placebocontrolled study evaluated the effects of intranasal PVP-I on nasal S. aureus colonization in patients undergoing orthopedic surgery (68). In this study, a single preoperative application of 5% PVP-I nasal solution eliminated nasal S. aureus in over two-thirds of patients at 4 h posttreatment, although an off-the-shelf preparation of 10% PVP-I solution was found to be less effective (68). In another randomized, placebo-controlled study, a single application of 10% PVP-I nasal preparation significantly reduced nasal MRSA at 1 and 6 h (P Ͻ 0.05); however, significant activity was not maintained at 12 and 24 h (P Ͼ 0.05) (70). Based on these results, the authors concluded that PVP-I may be effective for the short-term suppression of MRSA during surgery, and they proposed that this may be sufficient to reduce the risk of SSI (70). Of interest, the study demonstrated that repeated dosing with PVP-I every 12 h for 5 days did not enhance its efficacy, since there was no significant reduction in nasal MRSA with PVP-I compared to control (P Ͼ 0.05) (70).
Studies have also investigated the effect of intranasal PVP-I on the rate of SSI. In a retrospective database study, universal preoperative decontamination with intranasal PVP-I plus chlorhexidine wash and oral rinse significantly reduced the 30-day SSI rate versus historical controls in patients undergoing elective orthopedic surgery (1.1% versus 3.8% in controls; P ϭ 0.02) (71). In another retrospective review, the addition of intranasal PVP-I to chlorhexidine wash also significantly reduced the SSI rate in trauma patients undergoing emergency orthopedic surgery compared to chlorhexidine alone (0.2% with chlorhexidine plus intranasal PVP-I versus 1.1% with chlorhexidine alone; P ϭ 0.02) ( Table 2) (75).
Compared to intranasal mupirocin, preoperative intranasal PVP-I was found to have similar efficacy in preventing SSI in patients undergoing orthopedic surgery, with or without screening for MRSA (Table 2) (22,74). In an investigator-initiated, open-label, randomized study, 90-day deep SSI rates caused by any pathogen, including S. aureus, were similar with either intranasal PVP-I applied within 2 h of surgery or intranasal mupirocin given for 5 days before surgery with topical chlorhexidine (22). Likewise, in a retrospective study, the 90-day SSI rate was similar in patients treated with universal intranasal PVP-I 1 h before surgery compared to screening for MRSA followed by intranasal mupirocin for 5 days (with both groups also treated with chlorhexidine washes and a preoperative chlorhexidine wipe) (74). Alongside similar efficacy, patients treated with mupirocin were more likely to report headache, rhinorrhea, congestion, and sore throat than those treated with PVP-I, in which a single case of vasovagal reaction was reported (22). Overall, fewer patients (3.4%) rated their experience of treatment with nasal PVP-I solution treatment as unpleasant/very unpleasant compared to patients treated with intranasal mupirocin ointment (38.8%) (P Ͻ 0.0001) (76). According to a U.S. cost-effectiveness model, universal preoperative decolonization with intranasal PVP-I in patients undergoing orthopedic surgery potentially saved $74.42 per patient compared to preoperative screening and treatment of MRSApositive patients with a 5-day course of intranasal mupirocin (77). Other studies conducted in different settings led to a similar conclusion, i.e., systemic preoperative decolonization with PVP-I was more cost-effective than the standard MRSA screening protocol in orthopedic surgery, with no difference in infection rates (74,78).
(iii) Efficacy of topical PVP-I decolonization. In a prospective study in patients undergoing elective plastic surgery, a preoperative shower with PVP-I significantly decreased presurgical staphylococcal skin colonization versus controls (P Ͻ 0.0019) (69). Furthermore, the efficacy of preoperative skin decontamination with topical 7.5 or 10% PVP-I was similar to that of topical 2% chlorhexidine in the prevention of SSI in patients undergoing spinal or cardiac surgery, respectively (Table 2) (72,73).
(iv) Decolonization of health care professionals. In a study investigating the longer-term use of PVP-I in the ICU neonatal setting, intranasal PVP-I cream was applied to health care personnel three times per working day for 3 months. This procedure dramatically reduced the rate of nasal carriage of MRSA from 13.3% at baseline to 0% (79).

STATE OF THE ART AND PERSPECTIVES
Intranasal mupirocin and topical chlorhexidine are currently the preferred agents for the decolonization of S. aureus. However, the widespread use of mupirocin has led to resistance and treatment failures, and antibiotic resistance and the horizontal transfer of mobile antibiotic resistance elements following low-level exposure to chlorhexidine have also been reported (20,80), highlighting the need for alternative treatment strategies.
Decolonization is most effective among patient populations who are at risk of infection for only a short period (13). As documented in the WHO guidelines (1), the strongest evidence for decolonization is among surgical patients to prevent postoperative SSIs, particularly those undergoing cardiac and orthopedic surgery (13). Overall, the clinical data published to date on intranasal PVP-I support its short-term use prior to orthopedic surgery. Its successful use in trauma patients undergoing emergency surgery within 24 h of admission is also notable, since the use of a 5-day regimen of mupirocin is not feasible in this setting (75). Recent data indicate that a single preoperative application of intranasal PVP-I may be enough for the short-term suppression of MRSA during the perioperative period (70). Therefore, repeated dosing with intranasal PVP-I after surgery in order to enhance or prolong its antimicrobial activity may not be necessary, although studies are required to investigate this further. A large, randomized, controlled clinical trial is currently recruiting patients in order to compare • Very limited occurrence of the induction of iodine resistance (37,57,58) Preoperative intranasal PVP-I in orthopedic surgery: • Active against S. aureus regardless of the presence of antibiotic or antiseptic resistance (28,39,40,(42)(43)(44)46) • Does not induce crossresistance to antibiotics (57,

58)
• Significantly reduced 30-day SSI rate in combination with topical CHG (1.1%) vs controls (3.8%); P ϭ 0.02 (71) • Similar efficacy (6/842) to a 5-day course of intranasal mupirocin (14/855) in reducing 90-day deep SSI rate following surgery (fraction of surgeries with presence of deep SSI) (22) Chlorhexidine Broad spectrum, including activity against Gram-positive bacteria and some Gram-negative bacteria and limited activity against fungi (e.g., yeasts) and enveloped viruses (81) • Biocidal against MRSA within 2 to 30 min (45) • Several reports of the induction of CHG resistance but a low incidence in some studies (81,(84)(85)(86)(87)(88)(89)(90) Preoperative topical CHG in orthopedic surgery: • Dual CHG and mupirocinresistant MRSA rare but has been reported to cause decolonization failure (82) • More common in MRSA than MSSA (91) • 2% CHG no-rinse cloth reduced SSI rates from 3.19% to 1.59% when introduced into decolonization protocol in orthopedic surgery (93) • • Several reports of resistance (95)(96)(97)(98)(99) Intranasal mupirocin in orthopedic surgery: • No significant difference in SSI rate between nasal mupirocin (3.8%) and placebo (4.7%) (100) • High-level resistance (MIC Ն512 g/ml) associated with treatment failure (95) Intranasal mupirocin in other surgery (gynecologic, neurologic, or cardiothoracic surgery): • the efficacy and safety of alcohol-based solutions of 5% PVP-I and 2% chlorhexidine in reducing SSI after cardiac surgery (102). Such a study should help to address the paucity of data regarding preoperative decolonization with topical PVP-I versus chlorhexidine for the prevention of SSI in patients undergoing cardiac surgery. Moving forward, studies investigating PVP-I in other patient groups who are at an elevated risk of infection for short periods (e.g., ICU patients) would be of interest.
The management of MRSA colonization continues to evolve, and decolonization should not be considered in isolation. Other notable preventative measures include screening, contact isolation, environmental disinfection, and good hand hygiene practices. Usually, a collection of interventions works better than just one, and combined interventions can reduce infection rates by 40 to 60% (2). With this in mind, further research is required to define the best approaches for persistent carriers, as well as the efficacy of different decolonization strategies and protocols in both surgical and nonsurgical patients (1,2,13).
Despite the wealth of evidence obtained from different in vitro and ex vivo settings supporting the use of various concentrations of PVP-I (5, 7.5, and 10%), the selection of the most appropriate concentration of PVP-I for clinical use should be made on a case-by-case basis. One factor to consider that may guide selection is whether PVP-I is available in aqueous or alcoholic solution. For example, based on our clinical experience, we advocate 10% PVP-I in aqueous solution for use on mucous membranes, whereas we recommend 5% PVP-I in alcoholic solution for use on healthy skin before an invasive or surgical procedure. The final selection should be made by the treating physician, with the ultimate goal to select the concentration of PVP-I which will significantly reduce the bacterial load of the skin without causing issues with skin toxicity.

FUTURE AREAS OF RESEARCH WITH POVIDONE IODINE
Studies investigating the longer term effects of PVP-I (e.g., prevention of recolonization postsurgery and use in long-term-care facilities) are needed. We also recognized the paucity of information related to bacterial resistance to PVP-I. For this reason, studies examining the potential for bacterial resistance to PVP-I are also recommended, as well as studies to confirm the absence of an association between exposure to PVP-I and the selection of antibiotic resistance among recent clinical isolates.
Evidence suggests that PVP-I in combination with chlorhexidine may prove to be more effective in preoperative antisepsis than when either agent is used alone, a finding indicative of a possible synergistic effect between the two agents (103-105). Indeed, given the different mechanisms of action of PVP-I and chlorhexidine, there is good reason to believe that the disruptive action of chlorhexidine on the bacterial cell membrane may facilitate intracellular entry of PVP-I, thereby potentiating its antimicrobial efficacy (104). A synergistic effect with the use of two or more antimicrobials would provide the opportunity to reduce the dose of each respective antimicrobial, helping to further minimize possible adverse effects without sacrificing antimicrobial activity. Overall, there is a general absence of data relating to the possible synergistic actions of PVP-I in combination with other antiseptics or antibiotics, and this is an area of research that warrants further investigation. Some studies suggest that PVP-I may have a shorter time to bacterial regrowth than other antiseptic agents (36,68,70,106,107). Although this is unlikely to be an issue for surgical patients, clinical studies are needed to understand the long-term dynamics of PVP-I. In the case of S. aureus, the bacterium is capable of invading nasal epithelial cells, which appears to protect it from host defense mechanisms (7). A better understanding of the role of the resulting intracellular reservoir of S. aureus during nasal colonization may lead to improved decolonization procedures. This is necessary since both mupirocin and chlorhexidine exhibit weak activity against intracellular S. aureus (108), and there are currently no data available for PVP-I.

CONCLUSIONS
Based on current evidence, PVP-I may be a useful preoperative decolonizing agent for the prevention of S. aureus infections, including MRSA and mupirocin-resistant strains. The broad spectrum of activity of PVP-I, encompassing viruses and fungi, and its reported activity against biofilm formation distinguish it from other antiseptics. However, compared to the current literature, additional experimental and clinical data are required to further evaluate the use of PVP-I in this setting.

ACKNOWLEDGMENTS
Medical writing assistance in the preparation of this review article was provided by Jane Murphy and Jack Lassados (CircleScience, an Ashfield Company, part of UDG Healthcare plc). Writing/editorial support was funded by Meda Pharma S.p.A., a Mylan company. D