Although the incidence of short-term vascular catheter-related infections (CRI) is decreasing, CRI remains a leading cause of healthcare-associated infections (HAI) in intensive care units (ICUs) and the most preventable HAI [1]. During the first week of catheterization, colonization of the catheter from skin pathogens and subsequent infection is the main source for CRI [1]; hence, optimal skin preparation prior to catheter placement and during catheter use is paramount for prevention.

To identify the best skin antiseptic solution, numerous studies have been conducted comparing mainly povidone iodine (PI) to chlorhexidine (CHG), either in aqueous or alcoholic formulations. Key findings from these studies are twofold: first, aqueous solutions of 10 % PI are less effective than CHG-based antiseptics [2] and second, the combination of 2 % CHG–alcohol is superior to 5 % PI–alcohol [3, 4]. Alcohol is bactericidal within 15–30 s; however, because its efficacy disappears once evaporated, it cannot be used alone for skin preparation. The hallmark of CHG is slower onset and outstanding durability of bactericidal effect after application. In addition, unlike PI, CHG remains active in the presence of proteins such as those in blood, saliva, and other body fluids. Finally, new studies show that enhanced cleansing of the skin that is not grossly contaminated is unnecessary, regardless of the antiseptic solution used [3, 4].

These findings inform clinical recommendations (Table 1) [5]: first, application of sterile 2 % CHG–alcohol to disinfect the skin prior to insertion of a vascular catheter represents standard of care. Second, no cleansing of the skin with soap or detergent is necessary when the skin is not obviously contaminated. Third, use of aqueous or alcoholic PI as a first-line agent for skin antisepsis can no longer be recommended. These findings are consistent with other reports from skin disinfection before abdominal surgery [6] and skin preparation before cesarean section [7].

Table 1 Practical recommendations for skin antisepsis before insertion of a short-term vascular catheter
Full size table

Despite these data and supporting literature, areas of uncertainty persist. First, the optimal alcohol and CHG concentration is still not well understood. Existing studies mostly use 70 % isopropyl alcohol, but often do not report if the 70 % was as volume (v/v) or by gram percent (g/%). This is problematic because 70 % v/v alcohol equals 55 % g/%, a concentration much lower than expected for clinical efficacy. Second, studies with CHG–alcohol compare either 0.25, 0.5, or 2 % CHG for antiseptic skin preparation; thus, the optimal CHG concentration remains unclear. This question is important because of the tradeoff between efficacy and cutaneous toxicities associated with higher concentration of CHG. In one study of patients undergoing coronary artery bypass grafting comparing 0.5 % CHG to 2 % CHG diluted in the same alcohol, lower numbers of microorganisms on the skin 2 min after skin antisepsis and after wound closure were reported with the use of the highest CHG concentration [8]. Similarly, two non-randomized studies and one small randomized study also suggest 2 % CHG solution may offer enhanced bactericidal activity and further protection against CRI [3, 9, 10]. Third, how best to apply the antiseptic application remains controversial. Hands-free application of antiseptic using applicators or sterile gauzes handled with a pincer may allow for increasing antiseptic diffusion into the deeper layers of skin while keeping the hands of caregivers away to reduce risk of contamination. Relatedly, use of single-use vials containing sterilized antiseptic may reduce the risk of potential application of contaminated solutions from multi-use bottles [1], but may increase costs. However, a recent large randomized multicenter trial involving more than 5000 arterial or central venous catheters in severely ill patients showed that compared to application of 5 % PI–alcohol from multiple-use bottles with sterile gauzes, use of hands-free applicators containing sterile 2 % CHG–alcohol was, in fact, cost-effective [4]. By reducing the incidence of CRI by more than fivefold, the extra cost for preventing a single episode of CRI with 2 % CHG–alcohol use was estimated at €227 and compared favorably with the cost of one CRI estimated in similar patients at €19,583 [4].

Finally, despite the well-documented efficacy of CHG, several side effects remain problematic and may limit clinical utility. First, cutaneous skin reactions ranging from dermatitis to irritation are a concern that appears to be related to the concentration. Fortunately, in most cases, this does not require any specific intervention though some cases of severe dermatitis have been reported [4]. Second, allergy is a rare but serious concern especially among certain patient populations [1]. Whether this is mediated by a genetic susceptibility, CHG formulation, or an interaction between both is not known. Third, CHG resistance remains a feared adverse event. To date, in vitro data suggest that there has been a decrease of susceptibility rather than true resistance since—as with almost all disinfectants—the concentration of CHG in clinical use still far exceeds the minimum bactericidal concentration of CHG. In addition, such strains are very rare and currently of little clinical relevance [11]. However, as use of CHG in the form of skin disinfectants, topical wipes, baths, and impregnated devices and dressings grows, physicians must remain mindful of the risk of selection of resistant strains. Alternatives such as octenidine dihydrochloride [12, 13], 4 % CHG [14], or new hypochloride solutions [15] may thus be of value in the near future. Though not available widely, octenidine solutions may be of particular value in this regard.

With the integration of sterile 2 % CHG–alcohol into current available bundles for preventing CRI, getting to a zero risk of CRI seems within reach. Future work aimed at studying how best to implement such bundles to ensure high adherence requires more research, since all bundles are futile if not included in the daily clinical work.