Abstract

Background

Surface-active antiseptics, such as chlorhexidine, are increasingly being used as part of intervention programs to prevent methicillin-resistant Staphylococcus aureus (MRSA) transmission, despite limited evidence and potential for resistance. We report on the effect of an antiseptic protocol on acquisition of both endemic MRSA and an outbreak strain of MRSA sequence type 239 (designated TW).

Methods

Interrupted time-series data on MRSA acquisitions in two 15-bed intensive care units were analyzed using segmented regression models to estimate the effects of sequential introduction of an educational campaign, cohorting, and a chlorhexidine-based antiseptic protocol on transmission of TW and non-TW MRSA strains. Representative TW and non-TW MRSA strains were assessed for carriage of qacA/B genes and antiseptic susceptibility.

Results

The antiseptic protocol was associated with a highly significant, immediate 70% reduction in acquisition of non-TW MRSA strains (estimated model-averaged incidence rate ratio, 0.3; 95% confidence interval, 0.19–0.47) and an increase in acquisition of TW MRSA strains (estimated model-averaged incidence rate ratio, 3.85; 95% confidence interval, 0.80–18.59). There was only weak evidence of an effect of other interventions on MRSA transmission. All TW MRSA strains (21 of 21 isolates) and <5% (1 of 21 isolates) of non-TW MRSA strains tested carried the chlorhexidine resistance loci qacA/B. In vitro chlorhexidine minimum bactericidal concentrations of TW strains were 3-fold higher than those of non-TW MRSA strains, and in vivo, only patients with non-TW MRSA demonstrated a reduction in the number of colonization sites in response to chlorhexidine treatment.

Conclusion

A chlorhexidine-based surface antiseptic protocol can interrupt transmission of MRSA in the intensive care unit, but strains carrying qacA/B genes may be unaffected or potentially spread more rapidly.

The optimum set of interventions required to prevent methicillin-resistant Staphylococcus aureus (MRSA) transmission in an intensive care unit (ICU) is unclear [1]. The mainstay of infection control is prompt identification of MRSA-colonized patients and institution of contact precautions while enforcing compliance with universal hand hygiene practice [2–4]. Guidelines also recommend active surveillance cultures be obtained from high-risk patients, allowing earlier initiation of contact precautions, although the need for surveillance cultures is debated [3, 5]. Isolation or cohorting is also practiced to facilitate contact nursing, although the benefit in the ICU remains unclear [6]. Studies have reported a benefit of using surface antiseptics, such as chlorhexidine [7–11], and guidelines recommended their use for high-risk groups in which basic interventions have failed to reduce rates of MRSA infection to acceptable levels for decolonization [2, 3] and routine cleansing of all patients [2, 12]; however, given the limited evidence of eradication, some reviews recommend against routine decolonization to prevent MRSA transmission [1, 13].

Emerging resistance is a particular concern with the use of antimicrobials and antiseptics as decolonization agents. MRSA can develop clinically significant resistance to mupirocin [14–16], and some MRSA strains carry genes that confer increased minimum bactericidal concentrations (MBCs) to antiseptics [17]. The plasmid-borne qacA/B genes code for multidrug efflux pumps [18, 19] found in 10%–20% of UK [20, 21], 63% of European [18], 80% of Brazilian [22], and 55% of Taiwanese MRSA strains [23], the last being predominantly sequence type (ST) 239. Such strains have a 2-fold to 4-fold increase in MBCs for antiseptics, such as chlorhexidine [20], although because the MBC remains well below the concentrations used to treat patients, the clinical significance of qacA/B carriage remains unclear [21].

We recently reported measures introduced to control MRSA transmission in an ICU where there was high-level endemic MRSA transmission, mostly due to ST-22 and ST-36, and a 2-year outbreak with a novel variant of ST-239 called TW [24]. Here, we assess the effect of the chlorhexidine-based antiseptic protocol introduced during the TW MRSA outbreak using a retrospective interrupted time-series study.

Methods

Clinical setting and infection control practice

The St Thomas' Hospital site has two 15-bed ICUs on adjacent floors. All patients admitted to the ICUs from 1 January 2002 through 20 April 2006 were included in this study. Throughout, alcohol gel was at every bed space and contact precautions were in place. MRSA screening swab samples (of the nose, perineum, and axilla) were obtained from all patients at admission to the ICU and every Monday morning before the first application of antiseptics. After 1 November 2005, axilla swabs were stopped, and throat and rectal swab samples were added to the screening [25]. Other samples were obtained when the infection was clinically suspected. In response to high rates of MRSA transmission, the infection control committee instigated sequential additional measures, as previously reported [24]. An educational campaign initiated on 15 July 2003 focused on reenforcing hand hygiene and barrier nursing, covert audit of hand hygiene and barrier nursing practice on ward rounds (66%–77% compliance on 3 documented occasions), and monthly feedback on rates of MRSA infection (intervention A). Beginning 15 October 2003, MRSA-colonized patients were nursed in side rooms or in pairs (intervention B). Beginning 26 April 2004, a surface antiseptic protocol was introduced: patients with known MRSA infection had 1% (weight/volume) chlorhexidine gluconate (Hibitane; Derma) applied to the nostrils, around the mouth, and at tracheostomy sites 4 times daily; had 1% chlorhexidine acetate powder (CX Antiseptic Dusting Powder; Adams Health) applied to groin, axillae, and skinfolds daily; and were washed daily with 4% chlorhexidine (Hibiscrub; SSL International) applied by a wet cloth. Patients who had test results that were negative for MRSA had the same protocol apart from Hibitane use twice daily and 2% (weight/volume) triclosan (Aquasept; Medlock Medical) instead of Hibiscrub (intervention C). Mupirocin was not used because of resistance concerns [11, 26, 27]. Retrospective review revealed that 12% of MRSA-positive patients harbored a mupirocin-resistant strain.

Data collection and definitions

Clinical, demographic, microbiological, antimicrobial treatment, intervention, bed occupancy, and staffing level data were extracted from intensive care (CareVue; Philips), microbiology (MC&S; GSTT), and electronic patient administration systems (iSoft) to form the anonymized Guy's and St Thomas' Staphylococcal Transmission and Antimicrobial Record database, with approval from the hospital ethics committee [28]. Integrity of data extraction was validated for completeness and accuracy by abstracting 2% of data and comparing it manually with the original source data. The following data were used for this study: age, sex, specialty, admission Acute Physiology and Chronic Health Evaluation (APACHE) II score, length of ICU stay, ICU mortality rate, dates of isolation of MRSA from any sample and its antibiotic resistance profile, dates of ventilation and renal replacement, and staffing levels. A patient was defined as importing MRSA if MRSA was cultured from any sample taken within the first 48 h after ICU admission and acquiring MRSA if MRSA was isolated for the first time from any sample taken after 48 h in the ICU.

Laboratory techniques

Samples were processed using standard laboratory techniques. MRSA was identified in screens using tube-coagulase and disk-diffusion testing with methicillin disks [24, 25]. The antimicrobial resistance pattern of the first patient isolate defined them as having either TW (resistant to methicillin, erythromycin, ciprofloxacin, gentamicin, neomycin, tetracycline, and trimethoprim) or non-TW MRSA (all other resistance patterns) [24, 25]. Restriction-modification analysis [29] showed that all available TW isolates (n = 34) were CC8/239, consistent with previous microarray analysis [24], and 169 of 174 non-TW isolates were CC22 or CC30, consistent with local and national epidemiology data (data not shown). A total of 21 TW and 21 non-TW isolates were randomly selected to determine carriage of the qacA/B genes from all cluster acquisitions occurring before introduction of the antiseptic protocol. A cluster acquisition was defined as >1 acquisition of a strain with the same resistance pattern for >1 month. DNA was extracted using the ChargeSwitch DNA kit (Invitrogen) according to the manufacturer's protocol. Single primer pair sequences were used to detect qacA/B, as described elsewhere [18]. Polymerase chain reactions (PCRs) were performed in 50 µL of Platinum Blue Supermix (Invitrogen) containing 25 ng of DNA, according to the manufacturer's protocol. The PCR was performed at 96°C for 10 min, then 35 cycles of 94°C for 30 s, 52°C for 30 s, 74°C for 30 s, and final elongation at 72°C for 10 min. Products were analyzed by 2% agarose gel electrophoresis. Five qacA/B PCR-positive TW and 5 qacA/B PCR-negative non-TW isolates were selected for determination of biocide MBCs using a modified Clinical and Laboratory Standards Institute protocol [30]. Briefly, 5 × 105 colony-forming units (CFUs) from an overnight culture were incubated at 37°C for 24 h in serial dilutions of heat-and filter-sterilized chlorhexidine gluconate 20% (weight/volume) or 2% triclosan (weight/volume) (Sigma) in Mueller-Hinton broth. After 24 h, broths were subcultured onto Mueller-Hinton agar for colony counting. The MBCs were calculated as the concentration producing a 99.9% kill. Broths were reincubated and plated in the same fashion after an additional 24 h to confirm lack of growth. Each experiment was performed in triplicate and repeated twice.

Statistical analysis

TW and non-TW MRSA acquisition data were analyzed using separate segmented regression models. Specifically, the weekly number of acquisitions was analyzed using Poisson models with a log link function accounting for extra-Poisson variation and variations in exposure caused by changes in the weekly number of ICU admissions. Independent variables were week number (to model secular log-linear trends), week number after interventions A–C (to estimate changes in trends), which continued to the end of the study in each case, and indicator variables for the 3 interventions (to estimate sudden changes). The number of TW or non-TW cases during the previous week was included as an independent variable to account for autocorrelation. Parameter estimates were obtained using Bayesian model averaging (with equal prior probabilities for possible models) to account for model uncertainty [31]. Analysis was performed using the BMA package in R, version 2.6, averaging over all models with posterior probabilities greater than one-twentieth of the most likely model [32]. Numerical data are presented as mean values (± standard deviations [SDs]) and median values, unless otherwise indicated. Numerical variables were compared using Student's t test or Mann-Whitney U test. All simple analyses were performed using Stata software, version 10 (StataCorp).

Results

Clinical and epidemiologic characteristics

A total of 4570 patients were admitted to the ICU for a total of 44,505 days. Of these patients, 517 (11%) were admitted with MRSA infection, and 347 of 3073 susceptible patients staying >48 h acquired MRSA during their stay. Figure 1 shows the weekly number of ICU admissions and acquisitions of TW and non-TW strains. The educational campaign (intervention A), cohorting (intervention B), and antiseptic protocol (intervention C) are indicated with vertical dotted lines. Introduction of the antiseptic protocol appeared to be associated with an immediate reduction in acquisition of non-TW, but not TW, strains, although there was no apparent effect after the educational campaign or cohorting. This association was investigated further. Demographic, clinical, and other factors with the potential to influence MRSA acquisitions were compared before and after introduction of the antiseptic protocol (Table 1). No significant reduction was found in the proportion of patients admitted with MRSA. In addition, we found no obvious decrease in patient disease severity that might prompt fewer staff hand contacts, as measured by admission APACHE II score, ventilation, renal replacement, and length of stay. The daily nursing staff to bed occupancy ratio, which has been associated with MRSA acquisitions when reduced [33], deteriorated after introduction of the antiseptic protocol, although it remained >1. We explored whether the reduction in patients undergoing surgery might have contributed to the reduction in transmission by reducing the number of skin breaches available for acquisition. We saw no reduction in skin breach sampling after introduction of the antiseptic protocol and therefore considered this to be an unlikely explanation for the reduction in MRSA transmission.

Figure 1

Weekly total number of admissions and number of patients admitted with or acquiring methicillin-resistant Staphylococcus aureus (MRSA) in the intensive care unit during the study. Time of introducing interventions A, B, and C are shown. The expected number of outbreak strains of MRSA sequence type 239 (TW) and non-TW MRSA acquisitions generated using the best-fitting segmented regression models (model 1 in Table 2) are represented by a broken line. *The TW MRSA outbreak was first recognized in January 2004.

Figure 1

Weekly total number of admissions and number of patients admitted with or acquiring methicillin-resistant Staphylococcus aureus (MRSA) in the intensive care unit during the study. Time of introducing interventions A, B, and C are shown. The expected number of outbreak strains of MRSA sequence type 239 (TW) and non-TW MRSA acquisitions generated using the best-fitting segmented regression models (model 1 in Table 2) are represented by a broken line. *The TW MRSA outbreak was first recognized in January 2004.

Table 1

Clinical and Microbiological Characteristics of Patients Admitted to the Intensive Care Unit (ICU) before and after Introduction of the Antiseptic Protocol

Characteristic Pre-antiseptic protocol (n = 2480) Postantiseptic protocol (n = 2490) 
Age, mean years (±SD) 60 ± 17 59 ± 18 .09 
Male sex 1507 (61) 1308 (63) .21 
Specialty    
General medical 1444 (58) 1316 (62) .001 
General surgical 518 (21) 416 (20)  
Cardiothoracic surgery 518 (21) 358 (17)  
Admission APACHE II score, mean value (± SD) 16.2 ± 8.6 15.9 ± 8.3 .19 
Length of ICU stay,a median days (IQR) 4 (2–11) 4 (2–10) .68 
ICU mortality 581 (23) 442 (21) .07 
Admitted to ICU with MRSA 291 (12) 226 (11) .37 
Daily staff-to-bed occupancy rate, median value (IQR) 1.07 (1.02–1.1) 1.04 (1.02–1.09) .001 
Renal replacement 584 (24) 496 (24) .91 
Mechanical ventilation 2027 (82) 1704 (82) .89 
Characteristic Pre-antiseptic protocol (n = 2480) Postantiseptic protocol (n = 2490) 
Age, mean years (±SD) 60 ± 17 59 ± 18 .09 
Male sex 1507 (61) 1308 (63) .21 
Specialty    
General medical 1444 (58) 1316 (62) .001 
General surgical 518 (21) 416 (20)  
Cardiothoracic surgery 518 (21) 358 (17)  
Admission APACHE II score, mean value (± SD) 16.2 ± 8.6 15.9 ± 8.3 .19 
Length of ICU stay,a median days (IQR) 4 (2–11) 4 (2–10) .68 
ICU mortality 581 (23) 442 (21) .07 
Admitted to ICU with MRSA 291 (12) 226 (11) .37 
Daily staff-to-bed occupancy rate, median value (IQR) 1.07 (1.02–1.1) 1.04 (1.02–1.09) .001 
Renal replacement 584 (24) 496 (24) .91 
Mechanical ventilation 2027 (82) 1704 (82) .89 

NOTE. Data are no. (%) of patients, unless otherwise stated. APACHE II, Acute Physiology and Chronic Health Evaluation II; IQR, interquartile range; MRSA, methicillin-resistant Staphylococcus aureus; SD, standard deviation.

a

Median values (IQR) are presented where data are not normally distributed.

Estimating the impact of interventions on transmission of MRSA

Segmented regression analysis was used to assess step changes and trends associated with the interventions. This provided strong evidence of a large (∼3-fold) instantaneous reduction in the risk of acquiring non-TW strains after introduction of the antiseptic protocol. This result was identified in the full model (all covariates), in the 3 best models (Table 2), and when model selection uncertainty was accounted for (Table 3). In contrast, there was little evidence that other interventions had any effect on non-TW strains in the full model, the best single models (Table 2), or when model uncertainty was accounted for (Table 3). However, the short interval between the educational campaign and cohorting means that the uncertainty is substantial, and model 2 (Table 2) indicates it is possible, although relatively unlikely, that the educational campaign caused a large reduction in transmission of non-TW MRSA, whereas cohorting had the opposite effect. The posterior model probability indicates that this model was 5 times less likely than model 1, which attributed all changes to the antiseptic protocol (Table 2).

Table 2

Estimated Parameters from the Poisson Segmented Regression Applied to the Non-TW and TW Methicillin-Resistant Staphylococcus aureus (MRSA) Data

 Full model Model 1 Model 2 Model 3 
Variable IRR (95% CI) IRR (95% CI) IRR (95% CI) IRR (95% CI) 
Non-TW MRSA         
Intervention A 0.32 (0.05–1.31) .16   0.45 (0.21–0.85) .03   
Intervention B 2.39 (0.72–10.30) .19   2.52 (1.27–5.66) .01   
Intervention C 0.37 (0.15–0.89) .03 0.31 (0.22–0.43) <.001 0.26 (0.17–0.40) <.001 0.33 (0.23–0.47) <.001 
Trend before interventions 1.00 (0.99–1.01) .96       
Change in trend after intervention A 1.06 (0.88–1.28) .55       
Change in trend after intervention B 0.93 (0.76–1.12) .43       
Change in trend after intervention C 1.02 (0.98–1.06) .33       
Non-TW MRSA in previous week 1.04 (0.97–1.12) .28     1.05 (0.98–1.13) .18 
Posterior model P valuea <.01  0.54  0.11  0.09  
TW MRSA         
Intervention A 2.01 (0.77–5.02) .14     2.12 (1.09–4.26) .03 
Intervention B 0.73 (0.20–2.71) .64       
Intervention C 3.23 (1.14–10.23) .04 4.37 (1.76–11.37) .002 1.70 (1.86–16.43) .002 6.69 (2.43–19.05) <.001 
Trend before interventions 1.05 (1.03–1.08) <.001 1.06 (1.04–1.08) <.001 1.07 (1.05–1.09) <.001 1.05 (1.03–1.08) <.001 
Change in trend after intervention A 0.88 (0.78–0.98) .03 0.91 (0.88–0.94) <.001 0.89 (0.86–0.91) <.001 0.89 (0.86–0.92) <.001 
Change in trend after intervention B 1.07 (0.95–1.22) .26       
Change in trend after intervention C 0.93 (0.87–0.99) .03 0.95 (0.91–0.99) .02     
TW MRSA in previous week 1.10 (0.90–1.32) .34       
Posterior model P valuea <.01  0.24  0.14  0.13  
 Full model Model 1 Model 2 Model 3 
Variable IRR (95% CI) IRR (95% CI) IRR (95% CI) IRR (95% CI) 
Non-TW MRSA         
Intervention A 0.32 (0.05–1.31) .16   0.45 (0.21–0.85) .03   
Intervention B 2.39 (0.72–10.30) .19   2.52 (1.27–5.66) .01   
Intervention C 0.37 (0.15–0.89) .03 0.31 (0.22–0.43) <.001 0.26 (0.17–0.40) <.001 0.33 (0.23–0.47) <.001 
Trend before interventions 1.00 (0.99–1.01) .96       
Change in trend after intervention A 1.06 (0.88–1.28) .55       
Change in trend after intervention B 0.93 (0.76–1.12) .43       
Change in trend after intervention C 1.02 (0.98–1.06) .33       
Non-TW MRSA in previous week 1.04 (0.97–1.12) .28     1.05 (0.98–1.13) .18 
Posterior model P valuea <.01  0.54  0.11  0.09  
TW MRSA         
Intervention A 2.01 (0.77–5.02) .14     2.12 (1.09–4.26) .03 
Intervention B 0.73 (0.20–2.71) .64       
Intervention C 3.23 (1.14–10.23) .04 4.37 (1.76–11.37) .002 1.70 (1.86–16.43) .002 6.69 (2.43–19.05) <.001 
Trend before interventions 1.05 (1.03–1.08) <.001 1.06 (1.04–1.08) <.001 1.07 (1.05–1.09) <.001 1.05 (1.03–1.08) <.001 
Change in trend after intervention A 0.88 (0.78–0.98) .03 0.91 (0.88–0.94) <.001 0.89 (0.86–0.91) <.001 0.89 (0.86–0.92) <.001 
Change in trend after intervention B 1.07 (0.95–1.22) .26       
Change in trend after intervention C 0.93 (0.87–0.99) .03 0.95 (0.91–0.99) .02     
TW MRSA in previous week 1.10 (0.90–1.32) .34       
Posterior model P valuea <.01  0.24  0.14  0.13  

NOTE. Results from 4 models are shown: a full model containing all covariates and 3 models with the highest posterior model probability selected by the Bayesian model averaging procedure (models 1–3). IRRs of covariates included in each model are shown. Intervention A is the education program; intervention B, patient cohorting; and intervention C, anticeptic protocol. The 95% CIs were always between 0 and 0.004 for the outbreak strain of methicillin-resistant Staphylococcus aureus (MRSA) sequence type 239 (TW) models and were always between 0.04 and 0.12 for the intercepts of the non-TW models. CI, confidence interval; IRR, incidence rate ratio.

a

The posterior model P value measures the degree to which the data support the model (the closer these values are to 1, the greater the evidence in favor of the model).

Table 3

Estimated Bayesian Model-Averaged Incidence Rate Ratios (IRRs) for TW and Non-TW Methicillin-Resistant Staphylococcus aureus (MRSA) Acquisitions

 Non-TW MRSA TW MRSA Variable 
 
 

 
Variable IRR (95% CI) Probability ≠ 1a IRR (95% CI) Probability ≠ 1a 
Intervention A 0.91 (0.53–1.58) 15 1.12 (0.63–2.01) 17 
Intervention B 1.12 (0.60–2.09) 18 0.71 (0.18–2.88) 25 
Intervention C 0.30 (0.19–0.47) 100 3.85 (0.80–18.59) 83 
Trend before interventions 1.00 (1.00–1.00) 1.06 (1.03–1.08) 100 
Change in trend after intervention A 1.00 (1.00–1.00) 0.93 (0.85–1.02) 75 
Change in trend after intervention B 1.00 (1.00–1.00) 0.99 (0.92–1.06) 15 
Change in trend after intervention C 1.00 (1.00–1.00) 0.96 (0.88–1.05) 57 
 Non-TW MRSA TW MRSA Variable 
 
 

 
Variable IRR (95% CI) Probability ≠ 1a IRR (95% CI) Probability ≠ 1a 
Intervention A 0.91 (0.53–1.58) 15 1.12 (0.63–2.01) 17 
Intervention B 1.12 (0.60–2.09) 18 0.71 (0.18–2.88) 25 
Intervention C 0.30 (0.19–0.47) 100 3.85 (0.80–18.59) 83 
Trend before interventions 1.00 (1.00–1.00) 1.06 (1.03–1.08) 100 
Change in trend after intervention A 1.00 (1.00–1.00) 0.93 (0.85–1.02) 75 
Change in trend after intervention B 1.00 (1.00–1.00) 0.99 (0.92–1.06) 15 
Change in trend after intervention C 1.00 (1.00–1.00) 0.96 (0.88–1.05) 57 

NOTE. Incidence rate ratios (IRRs) were derived from model averages taken from the best 9 and 17 models fitting the non-outbreak strain of methicillin-resistant Staphylococcus aureus (MRSA) sequence type 239 (non-TW) and TW MRSA acquisition data, respectively (models with posterior probabilities less than one-twentieth that of the most likely model were excluded). Intervention A was the education program, intervention B was patient cohorting, and intervention C was anticeptic protocol. CI, confidence interval.

a

The columns headed Probability ≠ 1 give estimated probabilities for an effect associated with each parameter. Low values indicate little evidence of an effect, and values close to 100 indicate increasingly strong evidence.

Analysis of the TW data showed strong evidence of an initial increasing trend in acquisitions before any interventions (Table 2 and Table 3). There was strong evidence that the educational campaign and/or cohorting reduced acquisitions of TW by causing a reduction in either trend or level (the cumulative posterior probability of models not showing a reduction in transmission associated with at least 1 of these interventions was <1%), and the 3 models with the greatest support (Table 2) all suggested a decreasing trend after the educational campaign. However, after accounting for model uncertainty, there was only weak evidence in support of either intervention on its own, although a reversal of the increasing trend in TW MRSA associated with the educational campaign was estimated to be more likely than not (Table 3). Interestingly, the full model and the 3 best-fitting models all indicated a sudden increase in transmission of TW MRSA associated with the antiseptic protocol, although the estimated magnitude of the effect varied greatly (Table 2). Accounting for model uncertainty weakened the evidence in support of this effect, but nonetheless the estimated probability remained >80% (Table 3). In summary, regression analysis provided strong evidence that the antiseptic protocol caused an immediate reduction in transmission of non-TW MRSA and moderate evidence of an immediate increase in transmission of TW MRSA.

Carriage of qacA/B and biocide MBCs of TW and non-TW MRSA strains

This apparent diverging effect of the antiseptic protocol on transmission of TW and non-TW MRSA strains raised the possibility that they differed in their susceptibility to either or both chlorhexidine and triclosan. This was assessed first by detecting carriage of qacA/B. Twenty-one temporally distinct acquisition isolates of TW and non-TW strains were assessed for carriage of qacA/B. All TW MRSA strains and only 1 of 21 non-TW isolates carried both qacA/B genes. Chlorhexidine and triclosan MBCs for 5 TW MRSA and 5 qacA/Bnegative non-TW MRSA were assessed (Table 4). The TW MRSA strains had a 3-fold higher chlorhexidine MBC than the non-TW MRSA strains, whereas triclosan MBCs of TW and non-TW MRSA strains were identical.

Table 4

Chlorhexidine and Triclosan Minimum Bactericidal Concentrations (MBCs) for Selected TW and Non-TW Methicillin-Resistant Staphylococcus aureus (MRSA) Acquisition Isolates

Isoalte Triclosan MBC, g per 100 mL Chlorhexidine MBC, g per 100 mL 
TW MRSA 0.0025 0.0078 ± 0.0004 
Non-TW MRSA 0.0025 0.0026 ± 0.0008 
Isoalte Triclosan MBC, g per 100 mL Chlorhexidine MBC, g per 100 mL 
TW MRSA 0.0025 0.0078 ± 0.0004 
Non-TW MRSA 0.0025 0.0026 ± 0.0008 

NOTE. The minimum bactericidal concentrations (MBCs) are reported as mean ± SD of 5 outbreak strains of methicillin-resistant Staphylococcus aureus (MRSA) sequence type 239 (TW) of non-TW MRSA acquisition isolates.

Anatomical sites of MRSA isolation

To assess whether increased chlorhexidine MBCs of TW strains translate into reduced efficacy in vivo, a comparison was made of TW and non-TW site colonization before and after introduction of the antiseptic protocol. The percentage of patients from whom non-TW strains were isolated at screening, respiratory, skin breach, and wound sites was significantly reduced after introduction of the antiseptic protocol, whereas there was no significant reduction in TW isolation (Table 5).

Table 5

Methicillin-Resistant Staphylococcus aureus (MRSA) Isolation Sites from Patients Discharged from the Intensive Care Unit (ICU) before and Admitted to the ICU after Introduction of the Antiseptic Protocol

Variable Before antiseptic protocol After antiseptic protocol 
TW MRSA    
No. of patients 82 28  
Screen sitesa 28 (34) 11 (39) .31 
Respiratory tractb 52 (63) 15 (53) .18 
Catheter tipsc 41 (50) 12 (43) .26 
Skin breachesd 45 (55) 16 (57) .42 
Non-TW MRSA    
No. of patients 480 274  
Screen sites 359 (75) 140 (52) <.001 
Respiratory tract 261 (54) 97 (36) <.001 
Catheter tips 81 (17) 32 (12) .03 
Skin breaches 145 (30) 59 (22) .005 
Variable Before antiseptic protocol After antiseptic protocol 
TW MRSA    
No. of patients 82 28  
Screen sitesa 28 (34) 11 (39) .31 
Respiratory tractb 52 (63) 15 (53) .18 
Catheter tipsc 41 (50) 12 (43) .26 
Skin breachesd 45 (55) 16 (57) .42 
Non-TW MRSA    
No. of patients 480 274  
Screen sites 359 (75) 140 (52) <.001 
Respiratory tract 261 (54) 97 (36) <.001 
Catheter tips 81 (17) 32 (12) .03 
Skin breaches 145 (30) 59 (22) .005 

NOTE. Data are no. (%) of patients from whom MRSA was isolated at the indicated site at any time during their stay in the ICU. Patients whose ICU stay spanned the day when antiseptics were introduced are excluded. TW indicates outbreak strain of MRSA sequence type 239.

a

Screen sites were anterior nares, axillae, perineum, and throat (rectal swabs are excluded).

b

Respiratory tract samples were sputum, bronchial lavage or washings, and tracheal aspirates.

c

Catheter tips were from all central venous and arterial catheters.

d

Skin breaches were predominantly surgical wounds, including tracheostomy sites, drain sites, and ulcers.

Discussion

This study demonstrates that introduction of a chlorhexidine-based surface antiseptic protocol into an ICU where MRSA prevalence is stable at ∼20% can lead to an immediate and sustained reduction in transmission of susceptible MRSA stains. Although basic infection control interventions were already in place and actively reenforced, this does not mean that a more sustained and effective campaign would not have had an effect comparable to that of the antiseptic protocol. However, the data contribute to a body of evidence indicating that use of antiseptics, either as part of a decolonization strategy for known MRSA-colonized patients or administered to all ICU patients, can be useful in controlling MRSA [8–14]. It is recognized that surface antiseptics have only limited efficacy at eradicating MRSA in multisite-colonized patients [13, 34–37], at least in part prompting systematic reviews to recommend against their use to prevent MRSA transmission [1, 38]; however, the evidence suggests that the suppression of colonization, evidenced by a reduction in colonization sites after introduction of antiseptics rather than attainment of culture negativity, can reduce transmission. All patients received an antiseptic at multiple sites, so determining the relative benefit of individual components would require detailed prospective studies. The contribution of triclosan as a skin cleanser in MRSA-negative patients is also unclear; however, the fact that MBCs of TW and non-TW strains were identical suggests that it did not make a major contribution.

Interrupted time-series studies are the strongest quasi-experimental designs for assessing longitudinal effects of interventions, and segmented regression analysis is recommended for their analysis [39]. However, when several interventions are introduced close in time, accurately assessing the effect of each is difficult because multicolinearity can lead to unreliable estimates. One possible approach is to eliminate covariates using standard model selection algorithms (such as backward and forward selection). However, these are known to generate estimates that are too precise and P values that are too small, because they ignore uncertainty in model choice [31, 40]. We overcame this problem by using a Bayesian model averaging approach, which accounts for model uncertainty when obtaining parameter estimates by combining results of multiple models weighted by the posterior probability that they are correct. This reduces the chance of making type I errors. Even after accounting for such model uncertainty, the estimated effect of the antiseptic protocol on non-TW acquisition remained highly significant (Table 3).

One concern about the use of antiseptics is the lack of effect on and, therefore, selection of resistant strains, which was consistent with observations in this study. The TW strains carried qacA/B genes, demonstrated 3-fold increased chlorhexidine MBCs in vitro (similar to previously published qacA/B-positive MRSA strains [21]), and were not affected in anatomical site colonization or transmissibility after introduction of the antiseptic protocol. Indeed, all models provided evidence that the antiseptic protocol increased transmission of TW MRSA, although this was less clear when model uncertainty was fully accounted for. The MBC of TW MRSA was, however, still well below the chlorhexidine concentrations used, which might be explained by greater activity and/or expression of qacA/B on body surface localized TW MRSA and, therefore, a higher MBC in vivo than was demonstrated in vitro. Alternative explanations, such as ineffective application or inactivation on skin, seem less likely, because it is hard to envisage how this would translate into such a markedly different effect on TW and non-TW strains when their MBCs were so close. This continued transmission of TW strains cautions against widespread use of chlorhexidine in situations in which prevalence of qacA/B is high. It will be interesting to hear about experiences with chlorhexidine use in countries where qacA/B strains are more highly endemic and whether the carriage of qacA/B can account for some of the decolonization failures observed in randomized studies in which chlorhexidine is used as part of the protocol [12, 13, 15, 36, 37].

The TW MRSA models provided some evidence that hand hygiene or cohorting may have reduced TW MRSA acquisitions; however, this may be an artefact because the exponential increase in TW MRSA is clearly not sustainable in a small ICU. A better assessment of the effect of interventions on transmission of MRSA in small units would make use of individual-level data within a mechanistic transmission model to account for the fact that, as the numbers colonized increase, fewer patients remain at risk. This is an important area for further research.

This article and our previous report [24] on the TW MRSA strain together demonstrate that hospital MRSA strains can differ in their virulence and response to infection control interventions. That such strains exist should be borne in mind when extrapolating effectiveness of infection control interventions from one situation to another. The storage and analysis of MRSA strains identified in intervention studies may help to develop a more targeted approach to introduction of enhanced infection control measures.

Acknowledgments

We thank Dr Jodi Lindsay and Joshua Cockfield for restriction modification typing of the MRSA strains.

Financial support. Guy's and St Thomas' Charity (to R.B. and J.D.E), Department of Health, via the National Institute for Health Research comprehensive Biomedical Research Centre award to Guy's and St Thomas' National Health Service Foundation Trust in partnership with King's College London (to J.D.E.). R.B. receives 50% salary support in the form of an unrestricted educational grant from Novartis UK.

Potential conflicts of interest. All authors: no conflicts.

References

1
Loveday
HP
Pellowe
CM
Jones
SR
Pratt
RJ
A systematic review of the evidence for interventions for the prevention and control of methicillin-resistant Staphylococcus aureus (1996–2004): report to the Joint MRSA Working Party (Subgroup A)
J Hosp Infect
 , 
2006
, vol. 
63
 
Suppl 1
(pg. 
S45
-
70
)
2
Calfee
DP
Salgado
CD
Classen
D
, et al.  . 
Strategies to prevent transmission of methicillin-resistant Staphylococcus aureus in acute care hospitals
Infect Control Hosp Epidemiol
 , 
2008
, vol. 
29
 
Suppl 1
(pg. 
S62
-
80
)
3
Coia
JE
Duckworth
GJ
Edwards
DI
, et al.  . 
Guidelines for the control and prevention of methicillin-resistant Staphylococcus aureus (MRSA) in healthcare facilities
J Hosp Infect
 , 
2006
, vol. 
63
 
Suppl 1
(pg. 
S1
-
44
)
4
Siegel
JD
Rhinehart
E
Jackson
M
Chiarello
L
2007 Guideline for isolation precautions: preventing transmission of infectious agents in health care settings
Am J Infect Control
 , 
2007
, vol. 
35
 
10 Suppl 2
(pg. 
S65
-
164
)
5
Nijssen
S
Bonten
MJ
Weinstein
RA
Are active microbiological surveillance and subsequent isolation needed to prevent the spread of methicillin-resistant Staphylococcus aureus?
Clin Infect Dis
 , 
2005
, vol. 
40
 (pg. 
405
-
9
)
6
Cepeda
JA
Whitehouse
T
Cooper
B
, et al.  . 
Isolation of patients in single rooms or cohorts to reduce spread of MRSA in intensive-care units: prospective two-centre study
Lancet
 , 
2005
, vol. 
365
 (pg. 
295
-
304
)
7
Girou
E
Pujade
G
Legrand
P
Cizeau
F
Brun-Buisson
C
Selective screening of carriers for control of methicillin-resistant Staphylococcus aureus (MRSA) in high-risk hospital areas with a high level of endemic MRSA
Clin Infect Dis
 , 
1998
, vol. 
27
 (pg. 
543
-
50
)
8
Gould
IM
MacKenzie
FM
MacLennan
G
Pacitti
D
Watson
EJ
Noble
DW
Topical antimicrobials in combination with admission screening and barrier precautions to control endemic methicillin-resistant Staphylococcus aureus in an intensive care unit
Int J Antimicrob Agents
 , 
2007
, vol. 
29
 (pg. 
536
-
43
)
9
Ridenour
G
Lampen
R
Federspiel
J
Kritchevsky
S
Wong
E
Climo
M
Selective use of intranasal mupirocin and chlorhexidine bathing and the incidence of methicillin-resistant Staphylococcus aureus colonization and infection among intensive care unit patients
Infect Control Hosp Epidemiol
 , 
2007
, vol. 
28
 (pg. 
1155
-
61
)
10
Sandri
AM
Dalarosa
MG
Ruschel de Alcantara
L
da Silva Elias
L
Zavascki
AP
Reduction in incidence of nosocomial methicillin-resistant Staphylococcus aureus (MRSA) infection in an intensive care unit: role of treatment with mupirocin ointment and chlorhexidine baths for nasal carriers of MRSA
Infect Control Hosp Epidemiol
 , 
2006
, vol. 
27
 (pg. 
185
-
7
)
11
Milstone
AM
Passaretti
CL
Perl
TM
Chlorhexidine: expanding the armamentarium for infection control and prevention
Clin Infect Dis
 , 
2008
, vol. 
46
 (pg. 
274
-
81
)
12
Vernon
MO
Hayden
MK
Trick
WE
Hayes
RA
Blom
DW
Weinstein
RA
Chlorhexidine gluconate to cleanse patients in a medical intensive care unit: the effectiveness of source control to reduce the bioburden of vancomycin-resistant enterococci
Arch Intern Med
 , 
2006
, vol. 
166
 (pg. 
306
-
12
)
13
Simor
AE
Phillips
E
McGeer
A
, et al.  . 
Randomized controlled trial of chlorhexidine gluconate for washing, intranasal mupirocin, and rifampin and doxycycline versus no treatment for the eradication of methicillin-resistant Staphylococcus aureus colonization
Clin Infect Dis
 , 
2007
, vol. 
44
 (pg. 
178
-
85
)
14
Cookson
BD
Bolton
MC
Platt
JH
Chlorhexidine resistance in methicillin-resistant Staphylococcus aureus or just an elevated MIC? an in vitro and in vivo assessment
Antimicrob Agents Chemother
 , 
1991
, vol. 
35
 (pg. 
1997
-
2002
)
15
Dupeyron
C
Campillo
B
Bordes
M
Faubert
E
Richardet
JP
Mangeney
N
A clinical trial of mupirocin in the eradication of methicillin-resistant Staphylococcus aureus nasal carriage in a digestive disease unit
J Hosp Infect
 , 
2002
, vol. 
52
 (pg. 
281
-
7
)
16
Walker
ES
Vasquez
JE
Dula
R
Bullock
H
Sarubbi
FA
Mupirocinresistant, methicillin-resistant Staphylococcus aureus: does mupirocin remain effective?
Infect Control Hosp Epidemiol
 , 
2003
, vol. 
24
 (pg. 
342
-
6
)
17
Sidhu
MS
Heir
E
Leegaard
T
Wiger
K
Holck
A
Frequency of disinfectant resistance genes and genetic linkage with beta-lactamase transposon Tn552 among clinical staphylococci
Antimicrob Agents Chemother
 , 
2002
, vol. 
46
 (pg. 
2797
-
803
)
18
Mayer
S
Boos
M
Beyer
A
Fluit
AC
Schmitz
FJ
Distribution of the antiseptic resistance genes qacA, qacB and qacC in 497 methicillin-resistant and -susceptible European isolates of Staphylococcus aureus
J Antimicrob Chemother
 , 
2001
, vol. 
47
 (pg. 
896
-
7
)
19
Mitchell
BA
Paulsen
IT
Brown
MH
Skurray
RA
Bioenergetics of the staphylococcal multidrug export protein QacA: identification of distinct binding sites for monovalent and divalent cations
J Biol Chem
 , 
1999
, vol. 
274
 (pg. 
3541
-
8
)
20
Smith
K
Gemmell
CG
Hunter
IS
The association between biocide tolerance and the presence or absence of qac genes among hospital-acquired and community-acquired MRSA isolates
J Antimicrob Chemother
 , 
2008
, vol. 
61
 (pg. 
78
-
84
)
21
Vali
L
Davies
SE
Lai
LL
Dave
J
Amyes
SG
Frequency of biocide resistance genes, antibiotic resistance and the effect of chlorhexidine exposure on clinical methicillin-resistant Staphylococcus aureus isolates
J Antimicrob Chemother
 , 
2008
, vol. 
61
 (pg. 
524
-
32
)
22
Miyazaki
NH
Abreu
AO
Marin
VA
Rezende
CA
Moraes
MT
Villas Boas
MH
The presence of qacA/B gene in Brazilian methicillin-resistant Staphylococcus aureus
Mem Inst Oswaldo Cruz
 , 
2007
, vol. 
102
 (pg. 
539
-
40
)
23
Wang
JT
Sheng
WH
Wang
JL
, et al.  . 
Longitudinal analysis of chlorhexidine susceptibilities of nosocomial methicillin-resistant Staphylococcus aureus isolates at a teaching hospital in Taiwan
J Antimicrob Chemother
 , 
2008
, vol. 
62
 (pg. 
514
-
7
)
24
Edgeworth
JD
Yadegarfar
G
Pathak
S
, et al.  . 
An outbreak in an intensive care unit of a strain of methicillin-resistant Staphylococcus aureus sequence type 239 associated with an increased rate of vascular access device-related bacteremia
Clin Infect Dis
 , 
2007
, vol. 
44
 (pg. 
493
-
501
)
25
Batra
R
Eziefula
AC
Wyncoll
D
Edgeworth
J
Throat and rectal swabs may have an important role in MRSA screening of critically ill patients
Intensive Care Med
 , 
2008
, vol. 
34
 (pg. 
1703
-
6
)
26
Hudson
IR
The efficacy of intranasal mupirocin in the prevention of staphylococcal infections: a review of recent experience
J Hosp Infect
 , 
1994
, vol. 
27
 (pg. 
81
-
98
)
27
Ammerlaan
HS
Kluytmans
JA
Wertheim
HF
Nouwen
JL
Bonten
MJ
Eradication of methicillin-resistant Staphylococcus aureus carriage: a systematic review
Clin Infect Dis
 , 
2009
, vol. 
48
 (pg. 
922
-
30
)
28
Barnett
AG
Batra
R
Graves
N
Edgeworth
JD
Robotham
J
Cooper
BS
Estimating the effect of methicillin-resistant Staphylococcus aureus infection on length of stay using a longitudinal model
Am J Epidemiol
 , 
2009
, vol. 
170
 (pg. 
1186
-
1194
)
29
Cockfield
JD
Pathak
S
Edgeworth
JD
Lindsay
JA
Rapid determination of hospital-acquired methicillin-resistant Staphylococcus aureus lineages
J Med Microbiol
 , 
2007
, vol. 
56
 
pt 5
(pg. 
614
-
9
)
30
Wikler
MA LD
Cockerill
FR
Sheehan
DJ
Craig
WA
Tenover
FC
Dudley
MN
Methods for dilution antimicrobial susceptibility tests for bacteria that grow aerobically: approved standard-seventh edition
 , 
2006
Wayne, PA
CLSI
(pg. 
M7
-
A7
(formerly NCCLS)
31
Hoeting
JA
Madigan
D
Raftery
AE
Volinsky
CT
Bayesian model averaging: a tutorial
Stat Sci
 , 
1999
, vol. 
14
 (pg. 
382
-
401
)
32
Team
RDC
R: A language and environment for statistical computing
 , 
2007
Vienna, Austria
R Foundation for Statistical Computing
33
Grundmann
H
Hori
S
Winter
B
Tami
A
Austin
DJ
Risk factors for the transmission of methicillin-resistant Staphylococcus aureus in an adult intensive care unit: fitting a model to the data
J Infect Dis
 , 
2002
, vol. 
185
 (pg. 
481
-
8
)
34
Harbarth
S
Dharan
S
Liassine
N
Herrault
P
Auckenthaler
R
Pittet
D
Randomized, placebo-controlled, double-blind trial to evaluate the efficacy of mupirocin for eradicating carriage of methicillin-resistant Staphylococcus aureus
Antimicrob Agents Chemother
 , 
1999
, vol. 
43
 (pg. 
1412
-
6
)
35
van Rijen
M
Bonten
M
Wenzel
R
Kluytmans
J
Mupirocin ointment for preventing Staphylococcus aureus infections in nasal carriers
Cochrane Database Syst Rev
 , 
2008
, vol. 
4
 pg. 
CD006216
 
36
Watanakunakorn
C
Axelson
C
Bota
B
Stahl
C
Mupirocin ointment with and without chlorhexidine baths in the eradication of Staphylococcus aureus nasal carriage in nursing home residents
Am J Infect Control
 , 
1995
, vol. 
23
 (pg. 
306
-
9
)
37
Wendt
C
Schinke
S
Wurttemberger
M
Oberdorfer
K
Bock-Hensley
O
von Baum
H
Value of whole-body washing with chlorhexidine for the eradication of methicillin-resistant Staphylococcus aureus: a randomized, placebo-controlled, double-blind clinical trial
Infect Control Hosp Epidemiol
 , 
2007
, vol. 
28
 (pg. 
1036
-
43
)
38
Siegel
JD
Rhinehart
E
Jackson
M
Chiarello
L
Management of multidrug-resistant organisms in health care settings, 2006
Am J Infect Control
 , 
2007
, vol. 
35
 
10 Suppl 2
(pg. 
S165
-
93
)
39
Ramsay
C
Brown
E
Hartman
G
Davey
P
Room for improvement: a systematic review of the quality of evaluations of interventions to improve hospital antibiotic prescribing
J Antimicrob Chemother
 , 
2003
, vol. 
52
 (pg. 
764
-
71
)
40
Robins
JM
Greenland
S
The role of model selection in causal inference from nonexperimental data
Am J Epidemiol
 , 
1986
, vol. 
123
 (pg. 
392
-
402
)
(See the editorial commentary by Jarvis, on pages 218–20.)

Comments

0 Comments