- Open Access
Success rates of MRSA decolonization and factors associated with failure
Antimicrobial Resistance & Infection Control volume 11, Article number: 143 (2022)
We evaluated the success rate of MRSA decolonization directly after treatment and after one year in patients who were treated at the outpatient MRSA clinic of a large university medical centre to identify potential contributing factors to treatment success and failure.
Data from November 1, 2013 to August 1, 2020 were used. Only patients who had undergone complete MRSA decolonization were included. Risk factors for MRSA treatment failure were identified using a multivariable logistic regression model.
In total, 127 MRSA carriers were included: 7 had uncomplicated carriage, 91 had complicated carriage, and 29 patients had complicated carriage in combination with an infection. In complicated carriers and complicated carriers with an infection final treatment was successful in 75.0%. Risk factors for initial treatment failure included having one or more comorbidities and not testing the household members. Risk factors for final treatment failure were living in a refugee centre, being of younger age (0–17 years), and having one or more comorbidities.
The results of this study indicate that patients with a refugee status and children treated at the paediatric clinic have a higher risk of MRSA decolonisation treatment failure. For this reason, it might be useful to revise decolonization strategies for these subgroups and to refer these patients to specialized outpatient clinics in order to achieve higher treatment success rates.
Methicillin-resistant Staphylococcus aureus (MRSA) is a human pathogen and an important cause of a wide range of infections. MRSA has become endemic in health care institutions worldwide. Carriage of MRSA is associated with a higher risk of infection than carriage of methicillin-susceptible Staphylococcus aureus. [1, 2]. In the Netherlands, the MRSA prevalence in 2018 was only 1.2% . This low prevalence can be explained by the national search and destroy (S&D) policy in combination with careful and restrictive antibiotic use. The S&D policy focuses on isolation of MRSA carriers and of patients with an increased risk of MRSA carriage, outbreak management, and follow-up of MRSA carriers. The goal of this policy is successful eradication of MRSA carriage [4,5,6].
MRSA eradication treatment serves two purposes: the prevention of infection and the prevention of further transmission. MRSA treatment success rates directly after the decolonization attempt have been reported in one previous study published in 2011, showing a success rate of 56% among complicated MRSA carriers. This study recommended to testing household members of MRSA carriers and treating them simultaneously as the index case if they are positive . This was included in the most recent national Dutch protocol in 2012 . The success rates after implementation of these updated guidelines have not been evaluated yet. The success rate after one year of follow-up, when the absence of MRSA-carriage is considered to be definite, has not been reported before. Evaluating therapy success rates and determinants for therapy failure can contribute to further improvement of MRSA decolonization strategies. Additionally, new potential risk factors may have arisen in the past years. To our knowledge, this is the first study on MRSA decolonization that takes refugee status into account as potential risk factor.
The aim of this study was to evaluate the success rate of MRSA decolonization directly after treatment as well as after one year among patients who were treated at the outpatient MRSA clinic of a large university medical centre in order to identify potential contributing factors for success that could be used to further improve treatment.
Patients and methods
Study design and study population
A retrospective cohort study was conducted among patients who had started and finished MRSA decolonization treatment for MRSA carriage with follow-up cultures taken between November 1, 2013 and August 1, 2020 in the Radboud university medical centre (Radboudumc). Only patients who had undergone complete MRSA decolonization were included. Patients who were treated solely for MRSA infection or load reduction prior to a surgical procedure were not included. Patients were also excluded if their medical records and/or microbiological culture results were unavailable. Those who did not complete the first three follow-up culture rounds were also excluded.
MRSA treatment protocol
The SWAB (Dutch Working Party on Antibiotic Policy) protocol for treating MRSA carriers is used as standard MRSA treatment protocol in Dutch hospitals . Table 1 shows a short summary of this treatment protocol and illustrates the differences and similarities between the treatment of uncomplicated and complicated MRSA carriage.
Definitions and outcomes
Uncomplicated and complicated MRSA were defined according to Dutch MRSA treatment guidelines [7, 8].
In uncomplicated MRSA carriage:
The MRSA is exclusively localized in the nose,
There is no active infection and there are no skin lesions,
The MRSA is sensitive in vitro to the antibiotics to be prescribed,
And there is no foreign material that forms a connection between the internal and the external patient environment.
For complicated MRSA carriage at least one of the following criteria are met:
Located in throat, perineum, or skin lesions, independent of nasal carriage
There are active skin lesions
There is foreign material that forms a connection between the internal and the external environment
MRSA is in vitro resistant to mupirocin
Previous treatments according to the recommendations for uncomplicated carriage have failed.
Patients who had an active infection in addition to their MRSA carriage were categorized as ‘Infection and complicated carriage’. When looking at the treatment success rates, this group was merged with the complicated carriers.
Treatment outcome was assessed at two timepoints and for each patient we determined ‘initial treatment success’ and ‘final treatment success’. Treatment outcome was either ‘successful’ or ‘failed’.
Initial treatment success was defined as three consecutive negative culture sets obtained at least 48 h after completion of the first treatment and distributed over a period of at least seven days. When a patient also had negative MRSA cultures after two months and one year, they were defined as having final treatment success. Patients in whom decolonization treatment was not attempted after one or multiple failed treatments and remained MRSA carriers were defined as having treatment failure. Patients who had ‘initial treatment success’ but who later had positive cultures again were considered ‘recurrent MRSA carriers’. Patients who missed the two-month and/or the one-year cultures were included in the study but were marked as ‘lost to follow-up’ for determination of final treatment outcome, as it remains unknown if there was a recurrence of MRSA in these patients. Therefore, the patients who did not complete the full follow-up of one year were excluded in the analysis of final treatment outcome. Figure 1 shows an overview of the patient treatment outcomes of MRSA decolonization treatment.
Data were collected from the electronic patient system (EPIC). Information on MLST type was collected from the Medical Microbiology Laboratory database and the National MRSA Surveillance database Type-Ned from the Dutch Institution of Public Health and the Environment. A list of possible risk factors and determinants for treatment success were retrieved from literature and expert opinion. Data were entered in a secure research data environment in a database built in Castor.
The source of infection was divided into five categories: unknown, hospital-acquired, livestock-associated, community-acquired, and living in a refugee centre. An unknown source was specified as a patient without any known risk factor. Hospital-acquired MRSA was defined as any hospital admission in the Netherlands in the past year where the patient came into contact with an MRSA positive person and patients who were admitted or treated in a hospital outside the Netherlands. MRSA from patients who were involved with livestock were categorized as livestock-associated MRSA. MRSA from patients who had contact with confirmed MRSA positive persons outside the hospital were categorized as community-acquired MRSA. Refugee status was categorized based on living in a refugee centre or otherwise. In cases where patients could be placed in more than one category, a decision was made based on the most likely origin. Household members of an index patient were defined as a person staying in the same house during the day and night and having at least one shared facility .
For the data analysis, Statistical Package for the Social Sciences (SPSS) v25 and RStudio v1.2.5033 were used. Patients were divided based on their initial and final treatment outcomes and initial and final MRSA decolonization treatment success rates were calculated. Univariable logistic regression analysis was done to identify determinants for initial and final treatment failure for categorical variables. The Fisher’s exact test was used for dichotomous variables. P values were not calculated for variables with sample sizes smaller than five. Age was divided into three age groups: 0–17 years, 18 to 64 years, and 65 years and older. Patients under 18 years were treated at the paediatric MRSA outpatient clinic. Household size was categorized in three groups: 1–2, 3–4, and 5 or more household members, and was categorized as ‘unknown’ if this information was unavailable.
Multivariable logistic regression analysis was then performed to assess the associations between determinants of interest and both initial treatment outcome (i) and final treatment outcome (ii). All variables with a p value of < 0.1 in the univariable analysis were included in the multivariable analysis. Before entering these variables in the multivariable analysis, all potential variables were checked for multicollinearity by using the Variance Inflation Factor (VIF). Variables with a VIF larger than 4.0 were removed from the final model. Significance was defined as a p value of < 0.05. The enter method was chosen for the multivariable analysis. The results were presented by using adjusted odds ratios (OR) with their respective 95% confidence intervals (CI).
We identified 127 patients who started and finished MRSA decolonization therapy in the study period. Table 2 shows the patient characteristics of the study population and Table 4 shows the patient characteristics of adults and children separately. Of all the patients, 30.7% had one or more of the following comorbidities: chronic lung disease, skin disease, renal disease that required a solid organ transplant (SOT) or dialysis, or the need of indwelling devices. Indwelling devices were specified as all foreign material that forms a connection between the internal and external environment. There were 21 patients (16.5%) living in a refugee centre, including 15 children. Exposure to livestock was present in 16.5% of the patients. Other MRSA sources were hospital-acquired (23.6%), community-acquired (26.0%), and unknown (17.3%).
Determinants of initial treatment failure
The initial treatment outcome was successful in 70.9% (90/127) of all patients. In complicated carriers the success rate was 70.8% (85/120). In uncomplicated carriers this was 71.4% (5/7). In the univariable analysis (Table 2), the following variables were independently associated with initial treatment failure: age 0–17 years old, having one or more registered comorbidities, living in a refugee centre, certain MLST types, not testing household members, and a household size of 5 or more members. These variables were used for the multivariable analysis, with an exception of MLST due to high multicollinearity with the MRSA source, as certain sequences are predominantly found in specific sources. Age 0–17 (ORa 5.9, 95% CI [1.8–19.0]); comorbidities (ORa 3.4, 95% CI [1.2–9.2]) and not testing household members (ORa 7.4, 95% CI [1.4–40.4]) were all associated with failure of the first decolonization treatment.
Determinants for final treatment failure
Of the total group of 127 patients, 20 patients were excluded in this analysis due to loss to follow-up at two months (n = 14) or at one year (n = 6). The final treatment outcome was successful in 73.8% (79/107) of all patients. In complicated carriers, treatment was successful in 75.0% (75/100) and in uncomplicated carriers this was 57.1% (4/7). The univariable analysis in Table 3 shows the same determinants for final treatment failure as for initial treatment failure. Additionally, having travelled abroad, contact with livestock, perineal carriage and having side-effects of the treatment were independently associated with final treatment failure. Household size, contact with livestock and MLST were left out of the multivariable analysis due to high multicollinearity with MRSA source, as certain sequences are predominantly found in specific sources. The multivariable analysis showed that the following determinants were associated with treatment failure: living in a refugee centre (ORa 39.8, 95% CI [2.7–582.9]), age 0–17 (ORa 5.0, 95% CI [1.1–22.5]), and having one or more comorbidities (ORa 4.7, 95% CI [1.3–17.9]). We found that 92.9% of patients with a refugee status had final treatment failure, compared to 16.1% of non-refugees. In only one refugee, treatment was ultimately successful. Seven patients with a refugee status with initial treatment success were lost to follow-up and could not be included in this analysis.
In total, 35 patients were 0–17 years old (27.6%) and were treated at the paediatric clinic. Patient characteristics of this subgroup can be found in Tables 4, 5, and 6. Whereas 27/92 (29.3%) of the adults had relevant comorbidities, 12/35 (34.3%) of the children had comorbidities. All four children who were adopted had a cleft lip or palate. MRSA from these children were categorized as hospital-acquired MRSA as the children were all partially treated for their cleft lip or palate in their country of origin. In 42.9% of the patients, the source of MRSA was identified as caused by living in a refugee centre. Of the total number of refugees, 71.4% were children. The initial treatment success rate was 42.9%. The highest treatment success rates were seen in children with perineal carriage (70% initial treatment success), compared to nasal carriage (50%), and throat carriage (37%).
The final treatment outcome was successful in 43.3% (13/30) of the patients. In the complicated carriers, the final treatment outcome rate was successful in 42.3% (11/26). The lowest final treatment success rates were seen in children living in a refugee centre (1/11; 9.1%) compared to other sources of infection (56–100%) success rates) and children with carriage in ‘other sites’ than nose, throat, and perineum. These cultures from ‘other sites’ were obtained from a wound, sputum or eyes.
Up until now, there has been limited research on the risk factors of failure of MRSA decolonization treatment. The previous Dutch study was published in 2011 and had an initial treatment success rate of 56% among complicated carriers . Success rates have not been evaluated in nearly 10 years while there have been important changes in guidelines, demographics, and MRSA epidemiology. Compared to this previous research , the treatment success rate of the Radboudumc  is higher. This study showed a final treatment success rate of MRSA decolonization therapy of 75% for complicated carriers. There was a limited number of uncomplicated carriers, as they are normally treated by their GP and not referred to an MRSA outpatient clinic . In our study, all uncomplicated carriers were household members of index patients with a complicated carriage and were therefore treated simultaneously at the clinic. When looking at final treatment failure, our results showed that patients aged 0–17 years old, patients with relevant comorbidities, and patients living in a refugee centre had a significantly higher risk of treatment failure. The same risk factors were identified for initial treatment failure, with an addition of patients whose household members were not tested for MRSA.
The previous Dutch study showed that having chronic pulmonary disease, activities of daily living (ADL) dependency and the presence of devices, were strong predictors of treatment failure . Another important finding was that treatment failure is associated with not testing and treating household members . Similar results were found in a study conducted, in which it was observed that devices, relevant comorbidities and ADL dependency are independently associated with MRSA treatment failure . In our study, the majority of household members were tested for MRSA. Since previous research only used three consecutive negative culture sets to determine treatment failure, the results from our initial treatment failure were used for comparison. In the multivariable analysis of our data, the probability of treatment failure is higher when one or more household members were not tested, thus confirming the results of previous research. This was not seen in the multivariable analysis of final treatment failure. This could possibly be explained by testing household members in latter decolonization attempts that were initially not tested. Our study also confirmed having relevant comorbidities were associated with treatment failure. Therefore, our results support previous research and the updated Dutch national protocol.
Other factors that have been described as risk factors for MRSA eradication treatment failure are recent antibiotic use, presence of a skin wound, previous hospitalization, and older age . Although these variables have been tested in our study, we did not find an association with failed MRSA decolonization. This could be partially explained by the older age of the study population compared to our study, as children were not included in this study . In our study, age 0–17 years was associated with treatment failure (ORa 4.1, 95% CI [1.2–14.5]), while older age was not. This can be explained by the high percentage and severity of the underlying comorbidity in the children included in our study. Treatment compliance could also be of concern. Additionally, part of the children were refugees, which was also shown to be an important risk factor for treatment failure. Because of this, these children were not representative of the general population. To our knowledge, this is the first study that has also looked specifically at MRSA decolonization treatment in children. Our findings do not imply that decolonization treatment in children should be discouraged.
As stated previously, research has shown that ADL dependency may be associated with treatment failure [2, 10]. Another study also suggested that tonsillectomy might be able to improve treatment success in persistent MRSA carriers . Unfortunately, due to the lack of systematic registration of these variables in medical records, this could not be analysed in our study. Additional information on antibiotic resistance and side effects of the treatment could provide further insight as well. Therefore, a prospective study with preferably a larger sample size is still favourable to fully assess the impact of all risk factors of interest.
A remarkable difference between our study and the previous Dutch study is that the Netherlands has been receiving more refugees since 2015 [13, 14]. Refugees are often from countries with a higher prevalence of resistant micro-organisms [15,16,17]. To our knowledge, this is the first study on MRSA decolonization that takes refugee status into account as potential risk factor. Refugee status was associated with treatment failure. Many refugees were lost to follow-up as MRSA recurrence was often seen in their household members, which led to the termination of collecting more cultures, or because of transfer to another refugee centre. Our local public health and infectious diseases specialists observed that successful decolonization may be harder to achieve in people living in a refugee centre, which was confirmed by this study. Refugees are often not treated for MRSA due to the continuous exposure to other people with MRSA. They often only receive treatment if there is a high medical urgency such as a planned surgery or a medical condition. Additionally, many refugees have large households and several households usually share facilities in a refugee centre. This makes it harder and less feasible to test all household members, which is a risk factor for treatment failure. People living in a refugee centre often face less optimal hygienic living conditions. According to guidelines, it is recommended that patients change their clothing, bedding and towels regularly when being treated for MRSA . This might not be feasible for people living in a refugee centre as there are often restrictions when using washing machines. With the current situation in Ukraine, many EU countries, including the Netherlands, will be welcoming Ukrainian refugees. Data have shown that high rates of antimicrobial resistance, including MRSA, are reported in Ukraine [18, 19]. Therefore, these findings may also have an important implication for treating Ukrainian refugees.
A practical challenge is that many refugees in the Netherlands have financial restrictions. Although antibiotics are covered by medical insurance, the recommended antiseptics are not. These costs are often a barrier that cannot be overcome and because of which treatment is eventually postponed until these patients have better living and financial conditions. This highlights the importance of involving the refugee centres and regional public health services to improve the conditions of refugees to achieve higher success rates. Finally, antibiotic policy in the Netherlands often differs from other countries. There is a likelihood that refugees might be colonized with MRSA types that are harder to treat.
This was an observational retrospective cohort study, that had some limitations. Due to being a single-centre study the sample size was relatively small. Furthermore, people who visit the MRSA outpatient clinic or the paediatric clinic often already have several risk factors or comorbidities, which can influence treatment outcome negatively. This may have led to a higher prevalence of certain risk factors. The overall treatment success rates that were found in our study may therefore be lower compared to MRSA treatment success rates in primary care or regional hospitals. An additional uncontrolled factor is that not everyone who visits the MRSA outpatient clinic is treated. In patients with a continuous exposure to MRSA no decolonization attempt was started as it was considered highly likely that treatment would fail. In this group, only in patients with high medical urgency, load reduction or decolonization was attempted. High urgency also indicates that the patient may have underlying problems.
It was not possible to assess all patient data of HCWs who were treated for MRSA due to privacy reasons. As a result, not all relevant HCWs could be included in this study. This could have contributed to a more biased selection. For patients who were referred to the MRSA outpatient clinic, the MRSA genotyping sequence was not always available. Likewise, cultures of re-colonized patients were often not re-typed to assess whether the patient was colonized with a new strain. We assumed that these patients had a recurrence rather than a reinfection, as MRSA prevalence in the Netherlands is only 1.2% . Information regarding MLST types was however limited and could not be used for the multivariable regression analysis.
In conclusion, current treatment success of MRSA carriage based on the SWAB guideline is good. Since previous research has shown that transmission occurs in about half of the cases from an index person to household members, current success rates are likely due to the renewed policy where household members are tested and treated at the same time . Household members should still be tested and treated at the same time as the index patient. One of the more significant findings to emerge from this study is that patients with a refugee status and children with important comorbidity have a higher probability of treatment failure. For this reason, it might be useful to revise decolonization strategies for these groups and to refer these patients to specialized outpatient clinics. For people living in refugee centres, it seems to be important to also involve other parties, such as the refugee centres and regional public health services, to address these difficulties.
Availability of data and materials
The datasets generated and analysed during the current study are not publicly available due to privacy reasons but are available from the corresponding author on reasonable request.
Activities of daily living
Livestock associated MRSA
Methicilin-resistant Staphylococcus aureus
Search and destroy
Solid organ transplant
Stichting werkgroep antibioticabeleid
Variance Inflation Factor
Werkgroep Infectie Preventie
Pantosti A, Venditti M. What is MRSA? Eur Respir J. 2009. https://doi.org/10.1183/09031936.00007709.
Ammerlaan H, et al. Eradication of carriage with methicillin-resistant Staphylococcus aureus: determinants of treatment failure. J Antimicrob Chemother. 2011;66:2418–24.
ECDC E C f D P a C. Data from the ECDC Surveillance Atlas - Antimicrobial resistance. Available from: https://www.ecdc.europa.eu/en/antimicrobial-resistance/surveillance-and-disease-data/data-ecdc.
van Rijen MM, Kluytmans JA. Adjustment of the MRSA Search and Destroy policy for outpatients in the Netherlands: a prospective cohort study with repeated prevalence measurements. Antimicrob Resist Infect Control. 2014;3(1):3–3.
Wertheim HFL, et al. Low prevalence of methicillin-resistant Staphylococcus aureus (MRSA) at hospital admission in the Netherlands: the value of search and destroy and restrictive antibiotic use. J Hosp Infect. 2004;56(4):321–5.
Vos MC, et al. 5 years of experience implementing a methicillin-resistant Staphylococcus aureus search and destroy policy at the largest university medical center in the Netherlands. Infect Control Hosp Epidemiol. 2009;30(10):977–84.
SWAB D W P o A P. Optimalization of the antibiotic policy in the Netherlands XI Revision SWAB Guideline for the Treatment of MRSA Carriage; 2012.
Wertheim HF, et al. Optimisation of the antibiotic policy in the Netherlands. XII. The SWAB guideline for antimicrobial eradication of MRSA in carriers. Ned Tijdschr Geneeskd. 2008;152(49):2667–71.
Mollema FPN, et al. Successful treatment for carriage of methicillin-resistant Staphylococcus aureus and importance of follow-up. Antimicrob Agents Chemother. 2010;54(9):4020–5.
von Dach E, et al. Determinants of successful mrsa decolonization among patients included in a clinical trial of polyhexanide. Antimicrob Resist Infect Control. 2015;4:P192.
Lee AMM, Francois P, Renzi G, Schrenzel J, Vernaz N, Pittet D, Harbarth S. Impact of combined low-level mupirocin and genotypic chlorhexidine resistance on persistent methicillin-resistant staphylococcus aureus carriage after decolonization therapy: a case-control study. Clin Infect Dis Off Publ Infect Dis Soc Am; 2011. 52.
Labordus-van Helvoirt REM, van Rijen MML, van Wijngaarden P. Tonsillectomy for persistent MRSA carriage in the throat—description of three cases. Int J Infect Dis. 2018;67:98–101.
UNHCR UNHCFR. Refugees, internally displaced persons, asylum-seekers and other persons of concern to UNHCR in Europe; 2019. p. https://data2.unhcr.org/en/dataviz/102.
van Heelsum A. Aspirations and frustrations: experiences of recent refugees in the Netherlands. Ethn Racial Stud. 2017;40(13):2137–50.
Ravensbergen SJ, et al. High prevalence of MRSA and ESBL among asylum seekers in the Netherlands. PLoS ONE. 2017;12(4):e0176481–e0176481.
RIVM, R.v.V.e.M. Draaiboek infectieziekten in opvangcentra voor asielzoekers; 2019.
Ravensbergen SJ, et al. Proportion of asylum seekers carrying multi-drug resistant microorganisms is persistently increased after arrival in the Netherlands. Antimicrob Resist Infect Control. 2019;8:6–6.
ECDC. Operational public health considerations for the prevention and control of infectious diseases in the context of Russia’s aggression towards Ukraine; 2022
World Health Organizatio. Antimicrobial resistance surveillance in Europe 2022–2020 data; 2022.
Mollema FP, et al. Transmission of methicillin-resistant Staphylococcus aureus to household contacts. J Clin Microbiol. 2010;48(1):202–7.
This study was supported by internal funding.
Ethics approval and consent to participate
Ethics approval and informed consent were not required for this study. MRSA swabs were obtained as part of routine diagnostic testing and the data were anonymized. The results of the study did not impact the patients’ care.
Consent for publication
The authors declare that they have no competing interests.
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data.
About this article
Cite this article
Yiek, WK., Tromp, M., Strik-Albers, R. et al. Success rates of MRSA decolonization and factors associated with failure. Antimicrob Resist Infect Control 11, 143 (2022). https://doi.org/10.1186/s13756-022-01177-w
- Treatment success