Multidrug-resistant Staphylococcus aureus Colonization in Healthy Adults Is more Common in Bhutanese Refugees in Nepal than Those Resettled in Ohio

Although studies have shown that human migration is one of the risk factors for the spread of drug-resistant organisms such as methicillin-resistant Staphylococcus aureus (MRSA), surveillance studies examining MRSA among refugee populations in the US are lacking. This study aimed to assess the prevalence and molecular characteristics of S. aureus among Bhutanese refugees living in Nepal and resettled in Northeast Ohio (NEO). One hundred adult Bhutanese refugees from each geographic location were enrolled between August 2015 and January 2016. The participants were interviewed to collect demographic information and potential risk factors for carriage. Nasal and throat swabs were collected for bacterial isolation. All S. aureus isolates were characterized by spa typing and tested for the presence of Panton-Valentine leukocidin (PVL) and mecA genes; selected isolates were tested by multilocus sequence typing (MLST). The overall prevalence of S. aureus was 66.0% and 44.0% in NEO and Nepal, respectively. In Nepal, 5.8% (3/52) of isolates were MRSA and 1.1% (1/88) in NEO. Twenty-one isolates in NEO (23.9%) were multidrug-resistant S. aureus (MDRSA), while 23 (44.2%) in Nepal were MDRSA. In NEO, 41 spa types were detected from 88 S. aureus isolates. In Nepal, 32 spa types were detected from 52 S. aureus isolates. spa types t1818 and t345 were most common in NEO and Nepal, respectively. The overall prevalence of PVL-positive isolates among S. aureus in Nepal and NEO was 25.0% and 10.2%. ST5 was the most common sequence type in both locations. Bhutanese refugees living in Nepal and resettled in NEO had high prevalence of S. aureus and MDRSA. The findings suggest a potential need for CA-MRSA surveillance among the immigrant population in the U S and among people living in Nepal, and a potential need to devise appropriate public health measures to mitigate the risk imposed by community-associated strains of S. aureus and MRSA.


Introduction
Staphylococcus aureus is a common opportunistic bacterium. This bacterium is an important pathogen due to combination of toxin-mediated virulence, invasiveness, and antibiotic resistance [1]. Although it may be part of the normal human microbiota, it can cause wide range of diseases from skin and soft-tissue infections (STIs) to severe invasive disease such as infective endocarditis, osteomyelitis, and toxic shock syndrome [2]. S. aureus is also a major cause 2 BioMed Research International S. aureus accounts for nearly 20% of bloodstream infections in the hospital setting [7]. The treatment of S. aureus infections is often challenging due to the emergence of multidrug-resistant strains [8,9]. Although drug resistance has been seen as a major risk in healthcare settings, a similar increasing trend is observed in community-acquired infections [10].
Traditionally regarded as a nosocomial pathogen [11], MRSA infection outside of hospital has increased in incidence over the last decade [12] and has emerged as a major public health concern worldwide [13]. For example, community-acquired methicillin-resistant S. aureus (CA-MRSA) is the leading cause of identifiable skin and soft-tissue infections (SSTIs) observed in US emergency rooms [14]. An estimated 80,461 invasive MRSA infections occurred nationally in 2011. Of these, 16,560 were community-associated infections [15]. The economic impact of CA-MRSA is tremendous. A recent publication suggested these infections impose an annual burden of $478 million -$2.2 billion on thirdparty payers and $1.4 billion -$13.8 billion on society in the US [16]. Because of lengthy hospital stays, increased costs, and higher mortality, MRSA infections have imposed an increased disease burden in the last decade [12,17].
The disease burden caused by S. aureus is poorly understood in developing countries [18]. Although there have been some studies conducted in the context of Nepal, very little is known about the molecular epidemiology and transmission dynamics of S. aureus in this country. Most of the studies conducted in Nepal have been focused on the hospital environment, examining only the prevalence and phenotypic characterization of S. aureus [19][20][21][22][23][24]. As such, the molecular epidemiology of S. aureus in community settings has not been investigated, and S. aureus colonization parameters among the general population or specific groups such as Bhutanese refugees is unknown.
In the US, little is known regarding the prevalence and epidemiology of MRSA in immigrant populations. Although some infections, including tuberculosis and syphilis, may be detected during the pre-or postarrival screening, MRSA colonization is not examined, as there has been no policy or protocol for such screening by the US health institutions.
In the past several years, the US has granted asylum to Bhutanese immigrants of Nepalese origin that were displaced from Bhutan due to political reasons. As a result, Bhutanese of Nepali origin were one of the largest groups of refugees resettled in the US, accounting for 19% of the total 322,565 refugees admitted into the US between 2008 and 2012 [25]. This study investigated the prevalence and molecular epidemiology of S. aureus carriage in adult Bhutanese refugees living in camps in eastern Nepal and resettled in Northeast Ohio (NEO), United States.

Study Population.
A community-based cross-sectional study was conducted among Bhutanese refugees living in Nepal and NEO in between August 2015 and January 2016. The study sites were Beldangi refugee camp in Jhapa district of Nepal and Akron and Cleveland metropolitan area in NEO.
A convenience sample of 100 Bhutanese refugees aged 18 years and older, upon their consent to participate in the study, was recruited in each site. In both study sites, participants were recruited in community settings. Most of the participants in Nepal were enrolled in their home. A small portion of participants were invited to come to Primary Health Center for interview and sampling for their convenience. However, a meeting room was used for sampling. In NEO, the participants were recruited by visiting Nepalese grocery stores, Nepali restaurants, and the households of the participants and via distribution of flyers in the community.

Survey Instrument and Study
Variables. Data obtained from the questionnaire included sociodemographic information such as family size, education, income, and occupation; medical information such as presence or absence of asthma, heart disease, diabetes, autoimmune disease, and cancer; presence or absence of risk factors such as recent history of hospitalization, hospital visit, history of antibiotic treatment, time spent in jail, participation in team or contact sports, ownership of pets, and history of S. aureus or MRSA infection in the respondent or his/her family members; occupational exposure; and some miscellaneous exposures such as exposure to manure, pork, chicken, turkey meats, live animals, and types of soap used in home. An outcome variable "S. aureus and MRSA colonization positive" was assessed in association with exposure variables such as environmental and occupational exposures.

Data Collection.
A set of pretested semistructured questionnaires was used to collect information from study participants. A face-to-face interview was conducted with each participant by the research team. Questionnaires, first developed in English language were translated into Nepali language, and the Nepali language questionnaire set was used to interview the participants in both study sites. The purpose of the study, its potential harm, and benefits were explained to each participant. Verbal and written consent was obtained from each participant to participate in the study. A copy of Nepali consent form was provided to each participant to take with them. The Institutional Review Board (IRB) of Kent State University and of the Nepal Health Research Council (NHRC) approved the study.

Sample Collection and Bacterial
Culture. Biological specimens were collected from nares and throats using sterile swabs as described previously [26] and transported to the laboratory on blue ice packs. Samples were processed within 24 hours of collection. Samples were inoculated into 5ml (1X concentration) Baird Parker Broth with tellurite enrichment (Sigma products-Sigma-Aldrich, St. Louis, MO) in a 15 ml sterile plastic tubes by twisting the swab vigorously and incubated for 24 hours at 37 ∘ C. One microliter cultured broth was inoculated onto Baird Parker agar (BPA) with EY tellurite enrichment (BD) and was incubated for 48 hours at 37 ∘ C. Isolates were subcultured onto Columbia colistin-nalidixic agar with 5% sheep blood (Columbia CNA; Ramel) and tested via the catalase test, coagulase test, and using a S. aureus  [27]. Bacterial isolates resistant to three or more antibacterial or resistant to oxacillin were designated as multidrug-resistant S. aureus (MDRSA) [28].

Molecular Characterization.
Genomic DNA was extracted using the Wizard Genomic DNA preparation kit (Promega, Madison, WI). The staphylococcus protein A (spa) gene was amplified as described [29,30]. The detection of the mecA and PVL genes was determined by PCR [31,32]. spa types were assigned using Ridom5 StaphType software (version2.2.1; Ridom GmbH, Wü rzburg, Germany). The Based upon Repeat Pattern (BURP) algorithm was used to group spa types into genetic clusters based on their genetic proximity [33]. A selected number of isolates were subjected to multilocus sequence typing (MLST) as described previously [34]. The most common, unique, and/or isolates of interest were chosen for MLST (e.g., those which were present in high numbers, low numbers, or unique in some manner including a novel spa type). Sequence types (STs) were assigned using organism specific MLST database (https://pubmlst.org/saureus/). PHYLOViZ software v2.0 was used to analyze Global optimal eBURST of STs and to draw minimum spanning tree and relatedness of STs as described by Francisco et al. [35,36]. Positive (USA 300) and negative controls (reaction mixture without DNA template) were used in mecA, PVL, and spa PCR.

Statistical Analysis.
Frequency distribution and proportions were calculated for the categorical variables while mean (± standard deviation), median, and range were calculated for continuous variables. Chi-square test or Fisher's exact test and logistic regression were used to assess the association between outcome variables and the potential risk factors. The crude odds ratio (OR) was calculated with 95% confidence intervals. Statistical significance was assessed at = 0.05 level. All statistical analyses were conducted using SAS statistical software (Version 9.3, SAS Institute Inc., Cary, NC).

Demographic Information.
A total of 400 samples (200 nasal and 200 throat) were collected from 200 participants (100 from each study site). The mean age of the participants was 30.7 (SD 10.9) years in NEO and 36.8 (SD 13.3) years in Nepal. In NEO, 58% and 42 % of the participants were male and female, respectively, while the percentage of male and female participants in Nepal was 29% and 71%, respectively. In Nepal, the majority of the participants was from the Kirat religion (79%), married (86%), and had a formal education (60%). In NEO, the majority of the participants had formal education (86%) (  The overall prevalence of MRSA (based on the presence of the mecA gene) was 3.0% (3/100; 95% CI 0 -6.3%) in Nepal and 1.0% (1/100; 95% CI 0 -2.9%) in NEO. No significant difference was observed in MRSA prevalence between Nepal and NEO (p = 0.62, OR 0.32; 95% CI 0.03%-3.19%). Tables 2 and 3 show participants' demographics as well as medical, occupational, and environmental risk factors previously associated with the presence or absence of S. aureus. In NEO, a higher prevalence of S. aureus was found among males, participants who handled meat (pork and poultry) in the last six months, and participants who had skin infections in the last six months. In Nepal, a higher prevalence of S. aureus was found among females, participants who had children attending day care, and found who were diagnosed with skin infections and autoimmune diseases (Tables 2 and  3).

Antibiotic Susceptibility
Testing. All isolates were tested for antibiotic susceptibility. All isolates in Nepal and NEO were resistant to penicillin. Oxacillin resistance was observed for 1.1% (1/89) and 5.8% (3/52) of the isolates in NEO and Nepal, respectively. Resistance to other antibiotics was higher for Nepal than Ohio. Additionally, MDRSA rates were higher in Nepal than Ohio (44.2% and 23.6%, respectively) ( Figure 1).
lower than in a prior study conducted in a community setting in Nepal [43]. A previous study examining the prevalence of MRSA in a school children in Pokhara City of Nepal via nasal swab documented 17.4% (32/184) overall MRSA prevalence [43]. Such differences in S. aureus and MRSA prevalence may be due to geographical, population, and methodological differences. For example, we defined MRSA based on the presence of mecA gene, our sampling took place in eastern Nepal, and participants had to be at least 18 years old to be eligible for our study. Work by Rijal et al. (2008) took place in western Nepal, enrolled only minors (younger than 15 years old), and used a phenotypic method (Kirby-Bauer) for MRSA designation. Although the prevalence of MRSA in NEO is consistent with a previous report that suggested 0.84% of the noninstitutionalized US population were colonized with MRSA, the prevalence of S. aureus is significantly higher in this study (31.6% versus 66.0%) compared to that report [5]. S. aureus prevalence in our study was not associated with demographic factors such as age, gender, marital status, occupation, education, and income in both locations, nor with reported previous antibiotic exposure. The odds of S.   ST  68T5  Throat  --P, TS  t002  ST5  26T5  Throat  --P, C  t11917  ST72  71T5  Throat  --P  t15579  ST2885  74N Nose aureus colonization among participants in NEO was 2.5 times higher than colonization in Nepal (OR = 2.58, 95% CI: 1.45%-4.58%, p <0.001). However, variability in participants' demographic characteristics as well as length of residency in the US may have contributed to these differences. For example, the demographic composition of participants from Nepal was different from those in NEO in regard to gender, religion, marital status, and education. No significant difference was observed in S. aureus colonization among participants who lived less than 5 years in NEO. The prevalence of S. aureus in the nose and throat was higher in NEO than in Nepal (p=<0.05). Although the anterior nares are considered as the most common site of S. aureus colonization [6], the result of this study showed that the colonization of S. aureus was higher in throat compared to the nose in both locations of the study. This result is in consistent with a previous study done in Sweden that found throat as more common site of colonization [44].
Of the 18 drugs tested, resistance was observed to 9 and 12 different antibiotics in NEO and Nepal, respectively ( Figure 1). Compared to NEO, a significantly higher proportion of isolates (23.6% versus 44.2%) were MDRSA in Nepal. The higher proportions of isolates were resistant to different classes of antibiotics in Nepal as compared to NEO. S. aureus were frequently resistant to drugs that are used commonly in Nepal. For instance, ciprofloxacin is one of the most commonly prescribed antibiotics in Nepal [45], and trimethoprim/sulfamethoxazole is one of the most preferred antimicrobial in Nepal due to its wide spectrum and low cost [46]. One of the studies done in a tertiary care hospital in western Nepal demonstrated that approximately 56% and 27% of the MRSA isolates were resistant to trimethoprim/sulfamethoxazole and ciprofloxacin, respectively [24]. In Nepal, 42% and 27% of the isolates were resistant to trimethoprim/sulfamethoxazole and ciprofloxacin, respectively. Antibiotics overuse is a common drug use problem in Nepal [47]. For example, a study conducted in eastern Nepal documented that 84% of the prescriptions contained antibiotics comprising 42.8% of the total number of drugs [48]. In NEO, 22.5% and 18.0% of the isolates were resistant to trimethoprim/sulfamethoxazole and ciprofloxacin, respectively.
The prevalence of PVL-positive isolates was significantly higher among refugees living in Nepal compared to NEO (p=0.02). This is consistent with prior publications demonstrating a high prevalence PVL-positive isolates in clinical specimens in Nepal [49,50].
spa type t345 (ST 3206) was the most common spa type in Nepal. spa type t345 was previously isolated from patients infected with S. aureus in Banglore, India [51]. spa type t1818 (ST96) was most common in NEO. spa type t1818 has been previously isolated from clinical specimens in France [52], clinical samples from intensive care units of a tertiary hospital in Korea [53], and clinical samples from intensive care unit patients in the Netherlands [54]. Spa type t345 was found in both locations. However, spa type t1818 was found only in NEO. Only one isolate of spa types t002 (ST5) and t008 (ST8) each was found in Nepal. In NEO, 7 (8.0%; 7/88) and 4 (4.5%; 4/88) isolates were t002 and t008, respectively. spa type t002 (ST5/USA100) and t008 (ST8/USA300) are common hospital-associated and community-associated S. aureus stain, respectively. ST5 was the most common sequence type in both locations.
There are number of limitations to this study. This study employed a convenience sampling method to collect a limited number of human samples from populations in narrow geographic locations. As such, the generalizability of these findings is limited. Since the study design is cross-sectional, it does not allow for causal inference between exposures and outcomes. Although most of the medical and occupational risk factors were associated with S. aureus colonization, the association was not statistically significant. Hence, the study was not able to establish the demographic as well as medical and occupational exposures potentially implicated with S. aureus colonization. Since the study participants in Nepal and NEO are different and resettled refugees had varying duration of time regarding their residency in NEO, there might be other confounding factors affecting S. aureus colonization. The lack of longitudinal follow-up limits the understanding of the dynamics of persistent colonization of S. aureus and MRSA in the study population as well as the effects of environmental and occupational exposure over time.
The strengths of the study include the acquisition of samples from two locations that enrolled participants from the same ethnocultural heritage. The inclusion of survey questions in addition to human samples provided information concerning exposure variables. To our knowledge, this is the first study conducted to assess the prevalence and molecular characterization of S. aureus in immigrant population in the United States and in a community setting in Nepal. Follow-up studies could examine changes in colonization in a longitudinal manner as immigrants are resettled in Ohio or other areas in the US.

Conclusions
In conclusion, the findings of this study indicate that Bhutanese refugees living in Nepal aand resettled in NEO had high prevalence of S. aureus and MDRSA carriage. S. aureus colonization in throat was significantly higher than nasal colonization in both locations. The prevalence of MSSA was higher among resettled Bhutanese refugee living in NEO as compared to those living in Nepal. Information about S. aureus epidemiology in developing countries is important, as almost 1 billion people cross international boundaries each year. Population mobility imposes threats to the distribution of antimicrobial drug-resistant organisms [10]. MRSA has been isolated from immigrant people in Europe [55,56]. For example, a total of 56 persons who had recently arrived in Sweden between 2002 and 2003 as immigrants had cases of MRSA, giving an overall risk of 15.9 cases/100,000 immigrants [56]. In Spain, of the 19 patients infected with CA-MRSA, 15 were immigrants from South America [55]. The findings of this study emphasize the potential need for surveillance among immigrant population in the US and among people living in Nepal and a potential need to devise public health interventions to mitigate the risk imposed by S. aureus infections.

Data Availability
Data are available upon request to the authors.

Conflicts of Interest
The authors declare that they have no conflicts of interest.