Abstract
Abstract
Background:
Surgical site infection (SSI) is the most common hospital-acquired infection in the Netherlands. There is little evidence in regard to differences in the efficacy of pre-operative topical antisepsis with iodine–alcohol as compared with chlorhexidine-alcohol for preventing SSI.
Methods:
We conducted a retrospective analysis at a single center, involving all patients who underwent breast, colon, or vascular surgery in 2010 and 2011, in which pre-operative disinfection of the skin was done with iodine–alcohol in 2010 and with chlorhexidine–alcohol in 2011. Demographic characteristics, surgical parameters, and rates of SSI were compared in the two groups of patients. Subgroup analyses were done for wound classification, wound type, and type of surgery performed. Associations of patient characteristics with SSI were also investigated. Data were analyzed with χ2 tests, Student t-tests, and logistic regression analysis.
Results:
No statistically significant difference was found in the rates of SSI in the two study groups, at 6.1% for the patients who underwent antisepsis with iodine–alcohol and 3.8% for those who underwent disinfection with chlorhexidine–alcohol (p=0.20). After multivariable analysis, an odds ratio (OR) of 0.68 (95% confidence interval [CI] 0.30–1.47) in favor of chlorhexidine–alcohol was found. Male gender, acute surgery, absence of antibiotic prophylaxis, and longer hospital length of stay (LOS) were all associated with SSI after pre-operative topical antisepsis.
Conclusion:
In this single-center study conducted over a course of one year with each of the preparations investigated, no difference in the rate of SSI was found after an instantaneous protocol change from iodine–alcohol to chlorhexidine–alcohol for pre-operative topical antisepsis.
T
Application of antiseptic agents to the skin before surgery is a routine practice and is aimed at reducing the risk of SSI. In The Netherlands, the Dutch National Institute for Public Health and Environment (RIVM) recommends using solutions that contain 60%–90% alcohol (and arbitrarily either chlorhexidine 0.5% or iodine 1%) for such topical pre-operative antisepsis [3]. Although iodine–alcohol has been the gold standard for pre-operative antisepsis for many years, many centers in The Netherlands have recently made a transition to the use of chlorhexidine–alcohol for this purpose. This transition was largely triggered by the results of a randomized controlled trial conducted by Darouiche et al. and reported in the literature in 2010 [4]. In this trial, chlorhexidine–alcohol proved to be superior to aqueous povidine–iodine solution for pre-operative topical antisepsis in clean-contaminated surgery [4]. This trial was soon followed by a meta-analysis and systematic review that confirmed this result [5,6]. Furthermore, it has been proved that chlorhexidine–alcohol can reduce the risk of catheter-related blood stream infections by 50% below that with povidone–iodine in hospitalized patients who receive catheter-site care [7].
Although the results of these studies were convincing, the validity of comparing an alcoholic solution containing chlorhexidine with an aqueous iodine solution is questionable [8]. There is little evidence for a difference in the efficacy of preventing SSI through the use of chlorhexidine–alcohol as compared with its prevention with iodine–alcohol. In an effort to resolve this issue, we conducted an analysis of the rates of SSI following general surgery with pre-operative topical antisepsis with chlorhexidine–alcohol and with iodine–alcohol, respectively, during two consecutive one-year periods of the use of each agent in a large Dutch teaching hospital.
Patients and Methods
Data acquisition
Data for the study were collected retrospectively from the records of patients who underwent general surgery in 2010 and 2011 at the Rijnland Hospital in Leiderdorp, The Netherlands. In 2010 and prior to this, a preparation of 1% iodine in 70% alcohol was used for topical antisepsis in all patients who underwent surgery at the Rijnland Hospital. In 2011, after an instant change in protocol, a preparation of 0.5% chlorhexidine in 70% alcohol was used for such antisepsis in all surgical patients. The data used for the present study were recorded originally for the Preventie van Ziekenhuisinfecties door Surveillance (PREZIES) network, a Dutch initiative for monitoring and preventing SSI. The criteria of the U.S. Centers for Disease Control and Prevention (CDC) were used to characterize SSI occurring within 30 d post-operatively or, in the case of implant placement, 1 y post-operatively [9]. The data collected and used as variables in the present study were the type of topical antiseptic agent used pre-operatively; patient's gender and age; type of surgical procedure performed; duration of surgery; patient's physical status according to the American Society of Anesthesiologists (ASA) classification; wound classification (i.e., clean, clean-contaminated, contaminated or dirty); incision type (i.e., superficial or deep); implant placement; antibiotic prophylaxis; and whether surgery was elective or acute. A National Nosocomial Infection Surveillance (NNIS) risk index was calculated [10]. Chlorhexidine and iodine were applied in a standardized fashion by extensive painting of the area of the body in which surgery was to be done before placement of a sterile surgical drape.
Except for the difference in topical antiseptic agent, the protocol for preventing SSI did not differ in the two years in which the study was conducted. Patients in each in each category of topical antiseptic treatment were included consecutively for analysis. The types of surgery done on the patients included colon surgery (i.e., colon resection), vascular surgery (i.e., central or peripheral bypass surgery), and breast surgery (i.e., mastectomy with or without sentinel-lymph-node biopsy). The primary objective of the study was to compare the rate of SSI in 2010, in which iodine–alcohol was used for pre-operative topical antisepsis, with that in 2011, in which chlorhexidine–alcohol was used for this purpose. The secondary objective was to compare the rate of opened infected incisions (i.e., dehisced spontaneously or opened by a surgeon), which is an alternative, more stringent definition of SSI and is used by some surgeons [11].
Statistical analysis
Statistical analysis was done with SPSS software version 18.0 (SPSS, Chicago, IL). All patients in the data base were included. Categorical variables were analyzed with χ2 tests and continuous variables were compared through Student t-tests. Logistic regression analysis was done with correction for variables that were significantly associated with SSI. Additionally, sub-group analyses were for types of surgery, wound classifications, and wound types.
Results
Data for a total of 592 patients were analyzed. Of these patients, 278 were disinfected pre-operatively with iodine–alcohol (2010) and 314 were disinfected with chlorhexidine–alcohol (2011). The patients in the two study groups were similar with regard to age, gender, and wound classification (Table 1). No differences were observed in duration of surgery, antibiotic prophylaxis, hospital length of stay (LOS), placement of an implant, ASA score for physical status, or the number of acute surgical procedures done on the patients in each group. Except for one case, all classifications of contaminated or dirty wounds were related to acute surgery. A greater percentage of vascular surgery was done in the chlorhexidine–alcohol group than in the iodine–alcohol group (10.1% vs. 18.8%, respectively, p=0.003). Prosthetic implants were used in all cases of vascular surgery (except for one case) and in 33 cases of breast surgery.
ASA=American Society of Anesthesiologists; NNIS=National Nosocomial Infection Surveillance.
No statistically significant difference was found in the rates of SSI in the iodine-alcohol and chlorhexidine–alcohol groups in the study (6.1% vs 3.8%, respectively, p=0.20) (Table 2). Similarly, no differences were found in the rates of deep or superficial incisional SSIs in the two groups, and no differences in the rates of SSI were observed when the iodine–alcohol group and chlorhexidine–alcohol group were divided into subgroups on the basis of type of surgery or wound classification.
Figures are percentages of the treatment group.
SSI=surgical site infection.
Associations between demographic characteristics and the incidence of SSI are summarized in Table 3. Male gender, acute surgery, absence of antibiotic prophylaxis, and LOS were all associated with a higher incidence of SSI (Table 3). As did the overall analysis, multivariable analysis showed no statistically significant difference in the rates of SSI in the iodine–alcohol and chlorhexidine–alcohol groups in the study (OR 0.68, 95% CI 0.30–1.47) (Table 4). Furthermore, there was no significant difference in the incidence of opened or dehisced wounds in the iodine–alcohol and chlorhexidine–alcohol groups (10 cases [59%] vs. 4 cases [33%], respectively, p=0.39).
SSI=surgical site infection.
CI=confidence interval; OR=odds ratio.
Discussion
Several studies have compared the efficacies of the preoperative topical use of chlorhexidine–alcohol and povidone–iodine for preventing SSI [4–6]. However, not much is known about the antiseptic effect of chlorhexidine–alcohol in comparison with that of iodine–alcohol. In our study, no difference in the prevention of SSI was found for topically applied chlorhexidine–alcohol and iodine–alcohol. Although the differences in efficacy of the two antiseptic preparations were statistically non-significant, both univariate and multivariable analyses showed a trend toward superiority for chlorhexidine–alcohol over iodine–alcohol in preventing SSI. These findings in favor of chlorhexidine–alcohol are in accord with previous findings in studies that compared its pre-operative topical application with that of povidone–iodine in aqueous solution for preventing SSI. However, in evaluating the results of these studies, the antiseptic quality of alcohol itself should not be neglected [12–14]. It should also be kept in mind that in the studies in which chlorhexidine–alcohol was compared with povidone–iodine in aqueous solution, a combination of two antiseptically active compounds was being compared with only a single active compound.
A randomized controlled trial conducted by Veiga et al. showed no significant difference in the rates of SSI with the use of chlorhexidine–alcohol and iodine–alcohol for pre-operative topical antisepsis, even though there were statistically significantly fewer positive skin cultures in the chlorhexidine–alcohol group [15]. Other studies, by Saltzman et al. and Ostrander et al., both of which focused on orthopedic procedures, also showed fewer positive cultures following the pre-operative topical use of chlorhexidine–alcohol than with the use of 2% chlorhexidine gluconate–70% isopropyl alcohol, 0.7% iodophore–74% isopropyl alcohol, a povidone–0.75% iodine scrub, a povidone–1% iodine paint, 0.7% iodine–74% isopropyl alcohol, or 3% chloroxylenol [16,17]. On the other hand, a prospective study by Swenson et al. suggested that iodophor-based solutions in alcohol were superior to chlorhexidine–alcohol in preventing SSI [18]. A recent systematic review concluded that it remains unclear which antiseptic is superior to others, in accord with the finding in the current study [19].
The present study has several limitations. First of all, it was a single-center study with a small number of cases of SSI, and local variations in patient populations might have influenced the results. Second, in the center from which the study data were taken, there is increasing focus on the prevention of SSI. Attention to other means of preventing SSI (e.g., reduction of the number of door movements in the operating room, a new standard for changing gloves before closing the abdomen in colorectal surgery, and changes in surgical staff) might have confounded the results of the study. Lastly, although the trial conducted by Darouiche et al. included only clean-contaminated wounds, our study included clean (n=460) and clean-contaminated (n=106) wounds as well as contaminated (n=19) and dirty (n=5) wounds. With respect to this, we divided our population into subgroups so that it was possible to observe differences in the rates of SSI according to different wound classifications. It should be noted that most of the operations done on the patients in our study were elective operations, which were nearly always clean or clean-contaminated.
In summary, the present study, with approximately 300 patients in each study group, found that when a hospital center changed its pre-operative topical antiseptic protocol for general surgery from one involving the use of iodine–alcohol to one in which chlorhexidine–alcohol was used for this purpose, no difference in SSI was found after a one-year period of the use of each antiseptic preparation. Given the limitations of a single-center retrospective study, a multi-center randomized trial is required to provide sufficient evidence for the best topical antiseptic preparation for preventing SSI. We therefore plan to conduct a prospective, multi-center comparative trial (the SKINFECT trial) of various topical antiseptic agents for preventing SSI, to begin in 2013.
Footnotes
Acknowledgments
The authors thank Gerda Lelieveld for collecting the data used in the study.
Author Disclosure Statement
None of the authors has competing interests relevant to this manuscript.