MRSA carriage among healthcare workers in a Vietnamese intensive care unit: a prospective cohort study

Introduction

Colonization is the presence of an organism on or in a host with growth and multiplication to a sufficiently high concentration but does not invade the host’s tissues or cause disease (1). The causal relationship between prior colonization and subsequent infections has been well-established in high-resource clinical settings (1). Staphylococcus aureus is a well-described organism of the normal human flora, frequently colonizing the nose, pharynx, and skin (1). Most S. aureus isolates are sensitive to currently used antibiotics; thus, infections caused by this agent can be effectively treated. However, the emerging methicillin-resistant S. aureus (MRSA) has resulted in significant morbidity and mortality in susceptible patients (2). Notably, patients colonized with MRSA are more likely to develop S. aureus infection compared to methicillin-sensitive S. aureus (MSSA)-colonized or MSSA-non-colonized patients (2). In hospital settings, MRSA colonization among healthcare workers (HCWs) is a huge challenge because they may spread MRSA to their patients as a result of poor infection control practices (3-5). MRSA outbreaks in hospitals are epidemiologically associated with MRSA-colonized or MRSA-infected HCWs, especially those who had exfoliative skin conditions, skin infections, or respiratory tract infections (3,5). According to the current clinical practice guidelines, routine screening of HCWs for MRSA is not recommended. However, it is suggested that screening for MRSA can be beneficial in circumstances including (i) if transmission continues in a ward, despite active control measures; or if epidemiological aspects of an outbreak are unusual; or if there is evidence suggesting persistent MRSA carriage among HCWs; and (ii) if new MRSA carriers have been found among patients in a ward, and thus, HCWs with skin lesions should be identified and screened (5).

Currently, little is known about the magnitude of MRSA among HCWs, especially in low-resource settings like Vietnam, likely due to the lack of a routine screening program that is costly. To the best of our knowledge, available data focus on the magnitude of MRSA in those who are not intensive care unit (ICU) staff such as healthy adults, medical conference attendees, and ICU patients (6-8). Despite this, in 2006, an outbreak of severe community-acquired MRSA infections following routine immunization was reported in Ho Chi Minh City (HCMC), Vietnam (9). The outbreak investigation found that HCWs’ insufficient hand hygiene during routine injection led to the transmission of MRSA between children. A recent study conducted in the adult ICU at the Hospital for Tropical Diseases (HTD), HCMC, Vietnam, reported that 16.2% of patients acquired MRSA colonization during their ICU stay (10). This study also found that MRSA accounted for more than half (66.7%) of all S. aureus infections and suggested the role of HCWs in transmitting MRSA, leading to hospital-acquired infections (10). To strengthen MRSA prevention and control practices in Vietnam and other comparable settings, this study was conducted in the adult ICU of the HTD, which is among the largest local hospitals for infectious diseases in Vietnam, to examine the antimicrobial susceptibility profile of S. aureus isolates and the patterns of MRSA carriage among HCWs.

Materials and methods

Results

Antimicrobial susceptibility of S. aureus isolates

A total of 128 S. aureus isolates were cultured and included 123 (96.1%) from nasal swabs and 5 (3.9%) from hand swabs. MRSA accounted for 71.1% (91/128) of all S. aureus isolates (Fig. 1), with almost all (97.8%, 89/91) cultured from nasal samples. Almost three-quarters of 128 isolates were resistant to erythromycin (71.9%, 92/128) and clindamycin (70.3%, 90/128), while one-third (34.4%, 44/128) were resistant to ciprofloxacin. Vancomycin-resistant S. aureus was not detected. All S. aureus strains were fully sensitive to sulfamethoxazole-trimethoprim and rifampicin.

Discussion

The prevalence of MRSA carriage among our ICU HCWs was 29.1% (16/55, 95% CI: 18.8-42.1%). Specifically, 24.5% of HCWs were MRSA persistent nasal carriers. In Vietnam, previous studies found that the prevalence of MRSA colonization among healthy adults and ICU patients was 4.2% (28/662, 95% CI: 2.9-6.1%) (8) and 8.6% (72/838, 95% CI: 6.9-10.7%) (6), respectively, which is lower than the rate found in HCWs. Our findings are also in line with previous studies in Europe and the United States, which documented a higher rate of MRSA nasal colonization in HCWs compared with non-healthcare professionals (17,18). This can be explained by the increased exposure of HCWs to patients, hospital environment, and potential MRSA-contaminated objects including medical devices compared to non-healthcare individuals. Indeed, a multicenter study conducted in Nepal, which is a comparable country, found that MRSA can be isolated from the commonly used medical devices in ICU settings such as stethoscopes, pulse oximeters, ventilators, and defibrillators (19). The biofilm-forming property enables S. aureus to survive longer on the surfaces of these instruments, which are potential sources of nosocomial infections (19). A global review found that the estimated average MRSA carriage rate in HCWs was 5% (3). In detail, the rate of MRSA carriage among HCWs in Europe was 3.4% (95% CI: 3.1%-3.7%), the United States 4.2% (95% CI: 3.8%-4.7%), Africa 15.5% (95% CI: 13%-18.4%), the Middle East 6.1% (95% CI: 5.2%-7.2%), Australia and New Zealand 9.7% (95% CI: 8.5%-11.1%), and Asia excluding Vietnam 9.8% (95% CI: 8.4%-11.4%) (3). The prevalence of MRSA carriage in our HCWs was substantially higher than these reports and data from MRSA-endemic settings, where the prevalence in HCWs was 8.1% (95% CI: 7.4%-8.9%) (3). Globally, information on the burden of MRSA among ICU HCWs is scarce. Only 4.7% (95% CI: 4%-5.4%) of ICU staff from other regions were found to be MRSA carriers (3), which was lower compared with our data. Our findings further confirm the burden of MRSA in ICU settings in Vietnam, supported by high rates of MRSA colonization and infection in Vietnamese ICU patients (6,10). In light of this, Vietnam should be listed as a country with hyperendemic MRSA. Our study also indicates that MRSA carriage in HCWs in Vietnam is an urgent health problem that needs to be addressed, though this warrants further large-scale studies.

Nurses are associated with a higher risk of MRSA colonization. A meta-analysis showed that the risk of MRSA colonization among nurses was 2.6 (95% CI: 1.8-3.7) times higher than other healthcare staff including doctors and nursing assistants (17). This is probably due to the more frequent and close contact of nurses with patients compared with other healthcare staff. However, we noticed a higher proportion of nursing assistants who were colonized with MRSA compared with doctors and nurses, although this difference was not statistically significant. Suboptimal infection control practices have been indicated as a risk factor for MRSA carriage in HCWs (20). Despite the availability of local infection control guidelines, recent studies found that infection control compliance among HCWs is suboptimal in Vietnam (20,21). Especially, nursing assistants are found to have a lower infection control knowledge compared with other healthcare staff (22). This may explain the high proportion of MRSA carriage among our nursing assistants. To address this, infection control education programs should be tailored to meet the nursing assistants’ level of knowledge, and an audit program to measure infection control practice, especially hand hygiene, should be reinforced. Indeed, tailored infection control programs have been proven to be effective in comparable developing countries (23).

We also found that HCWs with gastritis or sinusitis had a higher prevalence of MRSA carriage compared with those without these comorbidities, although this association was not statistically significant. It is documented that HCWs with sinusitis are at an increased risk of transmitting MRSA in hospital settings and have been implicated in several MRSA outbreaks (3,24). Presently, the MRSA universal screening policy is controversial because of the lack of robust evidence for the effectiveness of such a costly measure. Recommendations for this infection control policy are suggested to be made by healthcare professionals based on their specific contexts (25). We believe that in low-resource settings with a high burden of MRSA like Vietnam, an infection prevention and control program needs to be designed to actively screen for MRSA among HCWs with these comorbidities for prompt interventions.

Among MRSA hand and nasal carriers, the distinction between persistent and intermittent carriage is important because persistent carriage is associated with a significantly higher bacterial load than intermittent carriage, resulting in an increased risk of transmitting MRSA to others (3,26). For MRSA hand carriage in our study, persistent carriage was not recorded, but 3.6% (95% CI: 1%-12.3%) of participants were found to be intermittent carriers. This prevalence was not different compared with the MRSA hand carriage rate in HCWs from other regions, including North America (8.3%, 95% CI: 3.5%-14.5%), Asia (4%, 95% CI: 2.1%-6.3%), and Europe (2.5%, 95% CI: 1%-4.5%) (27). Staphylococcal hand carriage in HCWs is usually transient, which means it is detectable after a working shift and gone before the next shift (3). The lack of persistent hand carriage in our study is a reassuring result. Moreover, intermittent carriage is often self-limiting and requires no treatment in healthy people (28). However, the contaminated hands of HCWs who are persistent or intermittent carriers are the main MRSA transmission route in hospitals, which can be prevented by effective hand hygiene (3,12). The overall prevalence of MRSA nasal carriage in our study was 30.1% (95% CI: 19.5%-43.5%), which was not different from the rate reported from Gaza Strip (25.5%, 95% CI: 20%-32%) (29). However, our rate was higher than that reported from Ethiopia (5.8%, 95% CI: 3.5%-9.5%) (30) and Nigeria (8%, 95% CI: 4.6%-13.5%) (31). Furthermore, the prevalence of persistent carriage of 24.5% was 4.5 times higher than the rate of intermittent carriage (5.6%) in our study. Unlike MRSA hand carriage, persistent nasal carriage cannot be managed by hand hygiene (32), but nasal mupirocin has been demonstrated to be efficacious in decolonizing MRSA in HCWs (3). Nasal mupirocin also provides a cost-effective adjunct to other infection control measures, including the screening and isolation strategies in controlling MRSA (33,34). However, nasal decolonization using mupirocin has not yet been implemented in Vietnam. A localized infection control guideline with detailed instructions on how to prevent and control MRSA nasal carriage with a focus on nasal decolonization will provide long-term benefits to both HCWs and patients.

In our study, S. aureus isolates were highly resistant to penicillin (96.1%), erythromycin (71.9%), and clindamycin (70.3%). We also found a moderate resistance rate with ciprofloxacin (34.4%). All S. aureus strains were fully sensitive to sulfamethoxazole-trimethoprim and rifampicin. Our results are consistent with a study reported from Ethiopia, in which 93.1% (27/29) of S. aureus isolates colonizing in HCWs showed resistance to penicillin followed by erythromycin (62.1%) and ciprofloxacin (37.9%) (30). Although a lower resistance rate was documented with clindamycin (17.2%) in this report, a higher resistance rate of 51.7% was recorded for co-trimoxazole compared to our findings (30). The resistance rates with erythromycin (29.1%), clindamycin (11.2%), and ciprofloxacin (9.6%) of the 62 S. aureus isolates colonizing in HCWs in Gaza Strip were also lower than ours (29). Surprisingly, 14.5% of S. aureus isolates were found to be resistant to vancomycin in the Gaza Strip study, while this strain was not detected in our study (29). Higher resistance to the aforementioned antibiotics in our study could be due to excessive use, misuse, and irrational prescriptions of these medications in both hospitals and community in Vietnam (35,36). Therefore, there is an urgent need for robust antimicrobial stewardship programs in combination with adherence to infection control measures to tackle the growing threat of antibiotic resistance in S. aureus. The results of this antimicrobial stewardship would help tailor the MRSA infection prevention and control program to meet the local needs.

There are some limitations in our study. Our sample size was small, and thus may not be able to statistically detect the differences between study groups. However, our sample represented almost all staff (92%, 55/60) in the ICU of a leading tertiary hospital in Vietnam. In addition, it has been found that MRSA counts (colony-forming units/mL) may decrease over time among subjects exposed to a source of MRSA (37). Hence, MRSA counts can provide more insights into the clearance of MRSA among our participants. However, information on HCWs’ MRSA counts was not available in our study. We also did not use genotyping methods to identify the resistance gene mecA, which is commonly used to examine MRSA due to limited financial resources. However, our study was set up to prospectively screen for MRSA carriage among HCWs for 8 consecutive weeks, and an antimicrobial susceptibility profile was performed for all cultured S. aureus isolates. The multiple testing times for antibiotic susceptibility of S. aureus in our study helped increase the possibility of detection of MRSA carriage and reduce the inconsistency of phenotyping resistance compared to the genotyping method. Given that there is no similar study in Vietnam, our study is the first attempt to examine the burden of MRSA among local HCWs. Therefore, we may have missed some possible risk factors of MRSA carriage in HCWs in Vietnam.

In conclusion, our data suggest that southern Vietnam may be an emerging MRSA hotspot, where the study was conducted . HCWs, especially nursing assistants, with comorbidities tended to be MRSA carriers. Infection control education programs should be tailored to meet the different knowledge levels of all HCWs. In addition to strengthening hand hygiene practice and antimicrobial stewardship, infection control guidelines need to be designed to actively screen for MRSA among HCWs with comorbidities for prompt interventions.

Acknowledgments

The authors would like to acknowledge the kind support of all staff of the adult intensive care unit of the Hospital for Tropical Diseases for their voluntary participation in the study.

Disclosures

Conflict of interest: The authors declare no conflict of interest.

Financial support: This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.