Strategies to Prevent MRSA Transmission and Infection in Acute-Care Hospitals

Implement a MRSA active surveillance testing (AST) program for select patient populations as part of a multifaceted strategy to control and prevent MRSA. (Quality of evidence: MODERATE). Recommendations for specific populations may have different evidence ratings.

Active surveillance for MRSA in conjunction with decolonization can be performed in targeted populations prior to surgery to prevent postsurgical MRSA infection. (Quality of evidence: MODERATE)

1. A large meta-analysis demonstrated a reduction in MRSA surgical site infection (SSI) when active surveillance was coupled with targeted nasal decolonization of MRSA carriers prior to undergoing surgery with hardware. Several other studies, including large clinical trials, have demonstrated a similar reduction in both SSI and nosocomial disease when employing S. aureus active surveillance and targeted decolonization of carriers. (See MRSA Decolonization, recommendation 2, in the Additional Approaches section.)

2. Please refer to the “Compendium of Strategies to Prevent Surgical Site Infections in Acute Care Hospitals: 2022 Update” for recommendations regarding active surveillance and decolonization for organisms other than MRSA (eg, all S. aureus).

Active surveillance with contact precautions is inferior to universal decolonization for reduction of MRSA clinical isolates in adult ICUs. (Quality of evidence: HIGH)

1. A 43-hospital cluster-randomized trial in ICUs (REDUCE MRSA Trial) directly compared (1) active surveillance for MRSA coupled with contact precautions, (2) active surveillance for MRSA coupled with contact precautions and targeted decolonization, and (3) stopping active surveillance, continuing contact precautions for known MRSA carriers, and performing universal decolonization for all ICU patients. Universal decolonization with chlorhexidine bathing and nasal mupirocin was superior to the other arms, resulting in a 37% reduction in MRSA clinical isolates (from 3.4 per 1,000 ICU days to 2.1 per 1,000 ICU days) and a 44% reduction in all-cause bloodstream infections (6.1 per 1,000 ICU days to 3.6 per 1,000 ICU days). Universal decolonization should be pursued in lieu of targeted actions informed by active surveillance for the purpose of reducing MRSA.

Hospital-wide active surveillance for MRSA can be used in conjunction with contact precautions to reduce the incidence of MRSA infection. (Quality of evidence: MODERATE)

1. Most hospitals across the United States do not perform active surveillance for all patients. However, between 2007 and the beginning of the COVID-19 pandemic, US Department of Veterans’ Affairs (VA) acute-care hospitals conducted hospital-wide active surveillance. In 2007, VA acute-care hospitals nationwide launched a MRSA control program that included universal nasal active surveillance for MRSA, contact precautions for MRSA carriers, hand hygiene, and increased institutional awareness of infection control. This program resulted in significant reductions in healthcare-associated MRSA infections and MRSA transmission in ICU and non-ICU settings. By 2017, hospital-onset MRSA infections at VA hospitals had declined 66% compared to baseline, and hospital-onset MSSA infections had declined 19%. Questions have arisen regarding what aspects of the VA policy led to the decline, especially relative to active surveillance. Questions have also been raised about the generalizability of findings at VA acute-care hospitals to other hospitals. Hospitals that do not want to conduct whole-hospital active surveillance should consider instituting a more targeted policy based on high-risk patients or high-risk encounters. In addition, hospitals should consider using their baseline risk assessment and additional MRSA monitoring and assessments to help guide their decision making. (See the Risk Assessment and Contact Precautions recommendations in the Essential Approaches section above and Section 5 “Performance Measures” below.) Active surveillance testing has cost implications. Cost and yield considerations should be used to help guide cost-effective policies to attain reductions in MRSA transmission and disease.

Active surveillance can be performed in the setting of a MRSA outbreak or evidence of ongoing transmission of MRSA within a unit as part of a multifaceted strategy to halt transmission. (Quality of evidence: MODERATE).

1. During outbreaks, serial (eg, weekly until outbreak is over) AST can provide important information about the scope of the outbreak, and AST helps identify new cases to enable communication and response (eg, contact precautions, decolonization).

2. See the Decolonization recommendations below for discussion of components of a multimodal strategy.

3. The CDC 2020 NICU guidelines provide information regarding application of this recommendation in the neonatal ICU.

4. See the “Screen HCP for MRSA infection or colonization” recommendation below for additional discussion regarding use of AST for HCP.

Screen HCP for MRSA infection or colonization if they are epidemiologically linked to a cluster of MRSA infections. (Quality of evidence: LOW)

1. HCP can become transiently or persistently colonized with MRSA and can be the source of hospital outbreaks.

   1. Routine screening of HCP for MRSA is not currently recommended in the endemic setting.

   2. Screening of HCP can be an important component of an outbreak investigation if HCP have been epidemiologically linked to a clonal cluster of MRSA cases or if there is evidence of ongoing transmission despite comprehensive implementation of basic MRSA control measures.

   3. See MRSA decolonization below and Section 6: Implementation Strategies for discussion of targeted decolonization therapy regimens that can be used for the treatment of MRSA-colonized HCP.

Use universal decolonization (daily CHG bathing plus 5 days of nasal decolonization) for all patients in adult ICUs to reduce endemic MRSA clinical cultures. (Quality of evidence: HIGH)

1. The previously described REDUCE MRSA Trial demonstrated that universal decolonization of ICU patients with daily CHG bathing and 5 days of twice-daily mupirocin was superior to screening and targeted decolonization as well as to screening and targeted contact precautions for prevention of MRSA-positive clinical isolates and all-cause bloodstream infection.(See Section 4 Additional Approaches for Preventing MRSA, Active Surveillance Testing recommendation.) For determining the applicability of this regimen to hospitals, trial benefit occurred at fairly low endemic levels of >3 MRSA clinical cultures per 1,000 ICU days. This approach has been demonstrated to be cost-effective, including sparing the cost of screening.

2. Climo et al reported that universal CHG alone in adult ICUs reduced bloodstream infections by 28% and reduced the composite of MRSA and vancomycin-resistant enterococcal (VRE) acquisition by 23%. Derde et al (2013) also demonstrated that CHG bathing decreased MRSA acquisition in ICU standardization phases leading up to an RCT that showed no benefit of either conventional or rapid PCR MRSA screening and isolation over high compliance hand hygiene and universal CHG bathing.

3. Even though universal CHG alone does not decolonize carriers, it is effective in reducing transmission of MRSA from carriers to noncarriers in ICUs. Thus, for the purpose of optimally addressing MRSA, the combined effects of mupirocin plus universal CHG are recommended.

4. Finally, although universal decolonization has been found to be superior to screening and targeted decolonization, hospitals may have other reasons for screening patients for MRSA. These may include outbreak response, desire for surveillance data, desire to implement contact precautions for known carriers, and clinical reasons related to restricting empiric anti-MRSA therapy or preoperative vancomycin prophylaxis to known MRSA carriers.

Perform preoperative nares screening with targeted use of CHG and nasal decolonization in MRSA carriers to reduce MRSA SSI in surgical procedures involving implantation of hardware. (Quality of evidence: MODERATE)

1. Note that decolonization can be applied universally as an alternative.

2. Preoperative targeted screening and decolonization of S. aureus carriers is commonly performed for surgical procedures involving the placement of hardware to reduce SSI. Although most studies have evaluated S. aureus outcomes and are not specific to MRSA, a large meta-analysis of RCTs and other studies involving surgeries with hardware similarly found that targeted or universal nasal decolonization reduced S. aureus SSI and that nasal decolonization of MRSA carriers reduced MRSA SSI.

3. S. aureus outcomes were not the target of this guidance document or its search strategy. Nevertheless, we highlight some the S. aureus evidence here because MRSA is a subset of S. aureus. In a large, 20-hospital, interventional cohort study of cardiac, hip, and knee surgeries, targeted nasal decolonization reduced S. aureus SSI. Additionally, in a post-hoc analysis of a single-center RCT of 1,697 patients undergoing arthroplasty or spinal fusion, universal nasal 5% povidone-iodine was superior to universal 2% mupirocin for S. aureus deep SSI. Universal nasal decolonization without nasal screening can be employed for pragmatic reasons to spare the logistics for screening for S. aureus or MRSA. Use of povidone-iodine may also be chosen for pragmatic reasons because it does not require a prescription, including prescription-related transportation needs or insurance copays that may affect patient adherence.

4. Please refer to the “Compendium of Strategies to Prevent Surgical Site Infections in Acute Care Hospitals: 2022 Update” for recommendations regarding decolonization for organisms other than MRSA.

Screen for MRSA and provide targeted decolonization with CHG bathing and nasal decolonization to MRSA carriers in surgical units to reduce postoperative MRSA inpatient infections. (Quality of evidence: MODERATE)

1. Note that decolonization can be applied universally as an alternative.

2. In a multinational trial of 33 surgical units in 10 hospitals involving 126,750 admissions, an intervention of enhanced hand hygiene plus universal screening and targeted decolonization of MRSA carriers reduced MRSA clinical cultures by 12% per month. In a secondary analysis of clean surgical patients, universal screening and targeted decolonization of MRSA carriers reduced MRSA clinical cultures by 15% per month and MRSA infections by 17% per month.

3. S. aureus outcomes were not the target of this guidance document nor its search strategy. Nevertheless, we highlight the key S. aureus evidence here because MRSA is a subset of S. aureus. In an RCT of 1,000 mostly surgical patients that evaluated universal inpatient screening for S. aureus and CHG and mupirocin for identified carriers, a significant 58% reduction was achieved in inpatient S. aureus infection among carriers. In addition, decolonization can reduce postsurgical inpatient infections beyond SSI. The Mupirocin and the Risk of Staphylococcus aureus (MARS) Study was a 3,864-person RCT of the addition of mupirocin to preoperative chlorhexidine (CHG) for S. aureus carriers undergoing a variety of surgical procedures (ie, general, gynecologic, neurologic, oncologic, and cardiothoracic surgery) with and without hardware. This mupirocin addition significantly decreased nosocomial S. aureus infections by 51% among S. aureus carriers, although it did not significantly reduce S. aureus SSIs.

Provide CHG bathing plus nasal decolonization to known MRSA carriers outside the ICU with medical devices, specifically central lines, midline catheters, and lumbar drains, to reduce MRSA-positive clinical cultures. (Quality of evidence: MODERATE)

1. The Active Bathing to Eliminate Infection (ABATE Infection) Trial was a 53-hospital cluster-randomized trial involving nearly 340,000 patients comparing routine care to universal decolonization with CHG bathing plus targeted nasal mupirocin for known MRSA carriers. Active screening was not a component of this trial. No overall reduction in the composite outcome of MRSA or VRE carriage, nor all-cause bloodstream infections was detected. However, in a post-hoc analysis, non-ICU patients with medical devices had a significant 37% reduction in MRSA and VRE and a significant 32% reduction in all-cause bloodstream infections. Patients with medical devices (specifically, central lines, midlines, and lumbar drains) were only 10% of inpatients, but they had 37% of MRSA and VRE cultures and 56% of all-cause bloodstream infections.

Consider postdischarge decolonization of MRSA carriers to reduce postdischarge MRSA infections and readmission. (Quality of evidence: HIGH).

1. The Changing Lives by Eradicating Antibiotic Resistance (CLEAR) Trial was an RCT to decrease postdischarge infections in MRSA carriers comparing routine care to postdischarge decolonization (CHG bathing, CHG mouthwash, nasal mupirocin) given for 5 days twice monthly for 6 months. The trial involved 2,121 MRSA carriers. Decolonization significantly reduced MRSA infection (most requiring rehospitalization) by 30% in the 1-year follow-up period, with a number needed to treat of 30. The impact of a shorter duration of decolonization is not known, but the risk of postdischarge infection was higher with closer proximity to discharge.

2. Postdischarge decolonization was first systematically performed by the Dutch Search and Destroy program to decolonize MRSA carriers to prevent infection. These postdischarge efforts require coordination and investment uncommonly adopted by hospitals. Because population-based medicine continues to be a goal for HAI prevention across the continuum of care, assessments of pragmatic implementation and adherence need to be addressed.

Neonatal ICUs should consider targeted or universal decolonization during times of above-average MRSA infection rates or targeted decolonization for patients at high risk of MRSA infection (eg, low birth weight, indwelling devices, or prior to high-risk surgeries). (Quality of evidence: MODERATE)

1. S. aureus is a leading cause of HAI in neonatal intensive care units (NICUs). Neonates in the NICU, especially low-birthweight neonates, are at high risk of invasive S. aureus disease. Because most neonates have never left the hospital, neonates usually develop MRSA colonization or infection as a result of hospital-based transmission. Neonates acquire MRSA from colonized parents, HCP, or the environment. MRSA is the most commonly reported cause of NICU outbreaks, so when neonates in the NICU are identified with a hospital-onset MRSA infection, further assessment is warranted to identify an ongoing cluster of transmission.

   1. MRSA colonization is an important risk factor for subsequent infection in this population. Quasi-experimental studies have shown that decolonization can reduce MRSA infections during endemic and outbreak settings.

   2. Targeted and universal decolonization approaches have both been successfully used to reduce MRSA in this population. Decolonization reduces MRSA colonization, acquisition and infection in neonates.

   3. Decolonization also reduces MSSA colonization and infections in this population.

   4. Mupirocin and chlorhexidine are the most commonly used decolonization agents in NICUs. In a recent RCT, 66 infants were assigned to intranasal mupirocin, and no product-related moderate, serious, or severe adverse events occurred. Chlorhexidine has been safely used in neonates, but due to potential for skin irritation and systemic absorption, it should be used with caution in premature infants. The US Food and Drug Administration notes that chlorhexidine should be “used with care in premature infants or infants under 2 months of age.” Chlorhexidine is used widely in NICUs and its use increased from 59% in 2009 to 86% in 2015 in a survey of US NICUs. Chlorhexidine-associated adverse events are infrequent, but many NICUs limit chlorhexidine use, especially in preterm infants within the first month of life.

   5. In addition to HCP and the environment, parents can be an important reservoir for S. aureus and can expose their neonates in the NICU. The TREAT PARENTS trial showed that decolonizing parents with intranasal mupirocin and topical chlorhexidine gluconate baths reduced transmission of MRSA and MSSA to neonates in the NICU.

Burn units should consider targeted or universal decolonization during times of above-average MRSA infection rates. (Quality of evidence: Moderate)

1. Higher quality evidence is needed to support a recommendation for routine decolonization of burn patients (unresolved issue).

2. Burn patients are at high risk of MRSA acquisition and infection.

3. Quasi-experimental studies have shown that decolonization can reduce MRSA infections. Decolonization strategies have included universal intranasal mupirocin with chlorhexidine antisepsis, universal decolonization using mupirocin and daily hypochlorous acid solution, and octenidine antisepsis for intact skin and nasal mucosa.

4. Given inconsistent results on the safety of antiseptics to interfere with wound healing, the role of topical antisepsis in this population for MRSA prevention must carefully balance the risk of toxicity and benefit of preventing MRSA infections. Therefore, the decision to implement targeted or universal decolonization in burn patients should be guided by a local risk assessment of MRSA incidence.

Consider targeted or universal decolonization of hemodialysis patients. (Quality of evidence: MODERATE) Higher-quality evidence is needed to support a recommendation for routine decolonization of dialysis patients.

1. MRSA bloodstream infections complicate care of hemodialysis patients. MRSA colonization predisposes individuals to subsequent MRSA infections, and hemodialysis patients have one of the highest risks of MRSA invasive disease, with a risk of 45 per 1,000 patients, which as 100-fold higher than that of the average population.

2. A systematic review and meta-analysis found that intranasal mupirocin with chlorhexidine body washes can eradicate MRSA carriage in hemodialysis patients. Data are not available demonstrating effectiveness of decolonization on reducing MRSA infections. However, a separate systematic review and meta-analysis found an 82% reduction in the risk of S. aureus bacteremia, comparing those who did and did not receive mupirocin.

Decolonization should be strongly considered as part of a multimodal approach to control MRSA outbreaks. (Quality of evidence: MODERATE)

1. Although no clinical trials have tested strategies to control MRSA outbreaks, many quasi-experimental studies have demonstrated successful outbreak control that includes MRSA decolonization as part of a multimodal approach to reduce MRSA transmission and infection.

2. In outbreak situations, decolonization can protect colonized individuals from infection and reduce colonization pressure that may promote transmission.

   1. Intranasal therapy reduces infection risk for individual patients.

   2. Topical skin decontamination reduces bioburden and helps reduce organism transmission.

3. Decolonization can be implemented universally or in combination with AST

   1. In an outbreak setting, active surveillance cultures can help measure the extent of organism spread in the unit and provide organisms for strain typing.

   2. In addition to identifying patients as a reservoir for propagating outbreaks, successful outbreak control may involve screening HCP to detect reservoirs, especially in high-risk units like the neonatal ICU and burn units. HCP have been implicated as reservoirs for MRSA transmission during adult hospital unit and NICU outbreaks and during times of ongoing clonal transmission.After implementation and failure of other basic MRSA prevention and control measures (eg, hand hygiene, contact precautions, enhanced environmental cleaning, screening and decolonizing neonates), screening and decolonizing HCWs has helped successfully control MRSA outbreaks in adult units and NICUs.

Use gowns and gloves when providing care to or entering the room of all adult ICU patients, regardless of MRSA colonization status. (Quality of evidence: MODERATE)

1. A cluster-randomized trial conducted in 20 adult medical and surgical ICUs compared the effect of universal glove and gown use for all patient contact and when entering any patient room with standard practice (ie, the use of gowns and gloves only for patients known to be infected or colonized with antimicrobial-resistant organisms) on the rate of acquisition of antimicrobial-resistant gram-positive organisms and healthcare-associated infections. Although the investigators found no difference in the primary outcome of acquisition of either MRSA or VRE, there was a significantly greater relative reduction in the prespecified secondary outcome of MRSA acquisition in intervention units compared to control units (40.2% vs 15%; P = .046).

   1. On intervention units, contamination of HCW clothing was 70% lower during the intervention period than during standard practice in the postintervention period (7.1% vs 23%; OR, 0.3; 95% CI, 0.2–0.6). In addition to the use of gowns and gloves, a lower frequency of HCP visits (4.28 vs 5.24 per hour; P = .02) and higher hand-hygiene compliance (78.3% vs 62.9% upon exit; P = .02) in the intervention arm compared to the control arm may have played a role in the observed difference in MRSA acquisition between the 2 groups. In subsequent mathematical modeling, the decrease in MRSA acquisition was found to be primarily due to the gown-and-glove use intervention, with additional but smaller effects from improved hand hygiene and lower HCP–patient contact rates.

   2. In a subsequent secondary analysis of data from this trial, the intervention was associated with a nonsignificant decrease in acquisition of antibiotic-resistant gram-negative bacteria (rate ratio, 0.90; 95% CI, 0.71–1.12). This finding suggests that universal gown-and-glove use when providing care in adult ICUs may provide benefits in addition to the potential to reduce MRSA transmission.

Universal MRSA decolonization

1. Additional study is needed to determine the incremental benefit of the addition of mupirocin to daily chlorhexidine bathing in the adult ICU because the REDUCE MRSA study used both mupirocin and CHG for their decolonization arm.

2. Additional study is needed to evaluate the role of routine universal decolonization of NICU patients.

Mupirocin and chlorhexidine resistance: The risk for development of resistance to mupirocin and/or chlorhexidine as they become more widely used is currently unknown, although some centers have reported increased rates of resistance.

1. Chlorhexidine: Although some published data have demonstrated reduced susceptibility in vitro to chlorhexidine among staphylococci by at least 2 mechanisms of resistance, the definitions used in these studies often use an MIC threshold far below standard CHG applications (eg, often an MIC of 8 µg/mL is used to define “resistance,” even though 2% CHG applies 20,000 µg/mL to the skin). Clinical trials have evaluated, but have not identified, the emergence of resistance to CHG.

2. Mupirocin resistance has been studied extensively; however, the ability of hospital laboratories to provide mupirocin resistance data is limited.

   1. Mupirocin resistance is phenotypically categorized into 2 levels based on the minimum inhibitory concentration (MIC). Low-level resistance (MICs of 8–256 mg/mL), and high-level resistance (MICs > 512 mg/mL). The molecular mechanism of low-level mupirocin resistance involves point mutations and is mediated by plasmid encoded genes in high-level mupirocin.

   2. A recent meta-analysis described a global increase in the prevalence high-level mupirocin resistance among clinical S. aureus isolates over time. Because mupirocin remains the most effective antibiotic for MSSA and MRSA decolonization, a reduction in its effectiveness presents a risk.

   3. Emergence of mupirocin resistance following increased use has not been reported consistently. The use of universal ICU decolonization with mupirocin in the REDUCE MRSA Trial was not associated with emergence of mupirocin resistance when evaluating thousands of MRSA isolates from the trial. Additional studies of mupirocin resistance have been hampered by a lack of availability of routine susceptibility testing in most hospital laboratories. Large-scale studies on decolonization failure associated with increased mupirocin use are needed to provide an understanding of the risk.

MRSA-colonized HCP: The optimal use of AST to identify asymptomatic carriage of MRSA among HCP and the optimal management (eg, decolonization therapy, follow-up monitoring) of MRSA-colonized HCP have not been definitively determined.