Enteric etiological surveillance in acute diarrhea stool of United States Military Personnel on deployment in Thailand, 2013–2017

Background Diarrhea remains a major public health problem for both civilian and military populations. This study describes the prevalence of acute diarrheal illness etiological agents, their antibiotic resistance distribution patterns, the resulting impact upon military force health protection, and potential prevention and treatment strategies. Results Forty-eight acute diarrhea stool samples from US military personnel deployed to Thailand from 2013–2017 were screened for enteric pathogens using ELISA, the TaqMan Array Card (TAC), and conventional microbiological methods. These isolates were also evaluated using antimicrobial susceptibility testing (AST) against ampicillin (AMP), azithromycin (AZM), ceftriaxone (CRO), ciprofloxacin (CIP), nalidixic acid (NA), erythromycin (ERY), and trimethoprim-sulfamethoxazole (SXT) using commercial methodology. Susceptibility results were interpreted following the CLSI and NARM guidelines. Questionnaire data obtained from 47/48 volunteers indicated that 89.4% (42/47) reported eating local food and the most common clinical symptoms were nausea and abdominal pain (51%; 24/47). Multiple bacterial species were identified from the 48 stool samples with diarrhea etiological agents being detected in 79% (38/48) of the samples distributed as follows: 43.8% (21/48) Campylobacter jejuni and Campylobacter species, 42% (20/48) diarrheagenic Escherichia coli, and 23% (11/48) Salmonella. Co-infections were detected in 46% (22/48) of the samples. All C. jejuni isolates were resistant to CIP and NA. One C. jejuni isolate exhibited resistance to both AZM and ERY. Lastly, an association between exposure to poultry and subsequent detection of the diarrhea-associated pathogens E. coli and P. shigelloides was significant (p < 0.05). Conclusion The detection of Campylobacter isolates with CIP, AZM and ERY resistance has critical force health protection and public health implications, as these data should guide effective Campylobacteriosis treatment options for deployed military members and travelers to Southeast Asia. Additional research efforts are recommended to determine the association of pathogen co-infections and/or other contributing factors towards diarrheal disease in military and traveler populations. Ongoing surveillance and AST profiling of potential disease-causing bacteria is required for effective disease prevention efforts and treatment strategies.


Background
Exposure to enteric pathogens is one of the major causes of diarrheal infections in both traveler and military populations [1]. Previous studies have reported that military personnel acquired infectious diarrhea during military exercises [2]. The risk of diarrheal infection is regionally dependent, particularly for civilian travelers and military personnel in transition from industrialized countries into developing countries [3,4]. Reported incident rates for bacterial diarrheal disease in military and travelers caused by enterotoxigenic Escherichia coli (ETEC), Campylobacter, and Shigella were between 38 and 45% in previous reports from various countries [5]. Typical treatment for traveler's diarrhea includes the use antibiotics to include ciprofloxacin, azithromycin and rifaximin [6]. However, enteric pathogens and their associated antibiotic resistance patterns evolve over time and vary by region [7,8]; therefore, access to up-to-date data on the global epidemiology of present diarrheal agents and their respective resistances are vital for diminishing the risk of diarrheal infection [6].
There are five Pacific region countries with which the US has a functional security alliance, including Thailand. The Armed Forces Research Institute for Medical Sciences (AFRIMS), based in Thailand, has coordinated studies of deployed US military to Thailand (i.e., the annual US-Thai "Cobra Gold" joint military forces exercise) for several years. Documented studies from previous exercises in Thailand demonstrated that US soldiers suffer consistent diarrhea attack rates during their first few weeks in country [9][10][11]. Despite modern preventive methods, diarrhea remains a primary concern for force health protection and therefore mission success for deployed military personnel in Thailand. Thus, the main objective of this study was to report the prevalence and clinical symptoms of diarrheal etiologic agents and bacterial pathogen antimicrobial susceptibility (AST) patterns affecting deployed US military personnel in Thailand for Cobra Gold exercises conducted in 2013 to 2017. This information will be useful in formulating more effective prevention and treatment strategies for these acute illnesses in deployed US forces.

Study design
A prospective acute diarrhea study was conducted in February of each calendar year, 2013-2017, at the following field sites: Lopburi, Phitsanulok, Chonburi (Samaesarn/Utapao), and Chanthaburi (Baan Chan Khem).
Diarrheal cases were defined as three or more loose stool in the previous 24 h, starting no more than 72 h before presentation, with concurrent clinical symptoms such as nausea, vomiting, and abdominal or bowel pain. After obtaining informed consent, US military service members who presented with these criteria and symptoms self-reported on an administered questionnaire the following: stool frequency and description, poultry exposure, local food consumption, and any additional clinical symptoms. Stool grading (formed, soft, loose, or watery) was assessed by US military medical staff. The stool grading "loose" is described the stools that appear softer than normal whereas "watery" is specified to the stools appearance that no solid pieces and all liquid. The study was approved yearly by the Walter Reed Army Institute of Research institutional review board, Silver Spring, Maryland, USA.

Antimicrobial susceptibility testing
Isolated enteric pathogens, except Campylobacter and Arcobacter, were evaluated by AST following standard Kirby-Bauer method to the following antibiotics, ampicillin (AMP), azithromycin (AZM), ceftriaxone (CRO), ciprofloxacin (CIP), nalidixic acid (NA) and co-trimoxazole (SXT), using commercially prepared discs according to the manufacturer's instructions (Becton, Dickinson and Company, USA). Susceptibility results were interpreted following the Clinical and Laboratory Standards Institute (CLSI) guidelines [13]. Campylobacter and Arcobacter isolates were evaluated using E-tests (Biomérieux, NC, USA). The minimal inhibitory concentration (MIC) was Keywords: Enteric etiology, Acute diarrhea, United States Military Personnel, Thailand defined as the lowest concentration of an antimicrobial agent that completely inhibited visible growth and was read at the point where the elliptical zone of inhibition intersected the MIC scale on the strip. Due to the limitation of CLSI guidelines for Campylobacter, the National Antimicrobial Resistance Monitoring System (NARM) 2013 criteria for AZM, CIP, erythromycin (ERY) and NA were followed for these isolates [14].

Statistical methods
Statistical analysis was conducted using IBM SPSS Statistics version 24.0. Chi squared tests were sued to determine if the association between patient questionnaire data and subsequent pathogen identification was significant.
Co-infections (defined as more than one etiologic agent) were detected in 46% (22/48) of the study samples with one sample from 2016 containing seven enteric pathogens: Aeromonas veronii bv sorbria, Arcobacter butzleri, C. jejuni, EPEC, Plesiomonas shigelloides, V. cholera, and V. parahaemolyticus. Stool samples containing the diarrheagenic E. coli and P. shigelloides were found to be most commonly associated with those US service members who were exposed to poultry (p = 0.02). One surprising observation was the absence of any typical etiologic agents for 2/10 samples that were classified as bloody diarrhea.
All pathogenic bacterial isolates obtained from 2015 to 2017 were further sub-cultured to perform AST, with resulting antibiotic resistance profiles contained in Table 3. 100% of the Salmonella isolates were resistant to AMP and 44.4% resistant to SXT. 52.9% (9/17) of the diarrheagenic E. coli isolates were resistant to AMP and 100% (4/4) of the EPEC isolates resistant to SXT. One C. jejuni and one A. butzleri isolate were resistant to AZM. The AZM-resistant C. jejuni was also resistant to ERY. 100% of the C. jejuni isolates were resistant to CIP and NA. All of the Plesiomonas and Aeromonas isolates were susceptible to all tested antibiotics.

Discussion
Diarrhea remains a leading cause of acute morbidity and chronic health effects, negatively impacting the health and functionality of both traveler and military populations. US military service members often deploy into developing regions in which enteric pathogens associated with diarrheal disease are prevalent. Campylobacter was the most frequent pathogen identified in this study, which correlates to the high prevalence in travelers with acute diarrhea in previous studies in Thailand in travelers and US military service members participating in previous Cobra Gold exercises [5,24,25]. Campylobacter isolates from this study were also entirely resistant to quinolones (NA) and fluoroquinolone (CIP) antibiotics, which is of additional concern based upon recent evidence indicating that quinolone-and fluoroquinolone-resistant Campylobacter infections are associated with the development of post-infectious long term sequelae to include Guillen Barre Syndrome [26,27]. A. butzleri, a member of the Campylobacteraceae family, was isolated from one stool samples. Arcobacter species are not typically associated with diarrheal disease, however, previous studies showed an 8% prevalence of traveler's diarrhea associated with A. butzleri in Mexico, Guatemala, and India [28]. Study of tourist restaurants in Thailand suggested that Arcobacter was a food-borne pathogen and its isolates were frequently resistant to AZM which is the common therapeutic recommendation for the treatment of diarrhea in Asia [29]. The AZM-resistant C. jejuni was also resistant to ERY, the recommended antimicrobial treatment in invasive cases or to eliminate carrier states. Erythromycin resistance has been reported in Thailand previously [30]. The second most common etiologic diarrhea agent identified in this study was E. coli. ETEC is the leading cause of childhood diarrhea and the most frequent cause of diarrhea in travelers to developing countries [31]. ETEC contribution to diarrheal disease is dependent upon the region of interest and seasonality [32][33][34]. In this study, EPEC was detected more commonly in cases than ETEC. A previous study noted that EPEC was dependent upon co-infection with other pathogenic bacteria to include Aeromonas and Salmonella in travelers who developed travelers' diarrhea [35]. However in our study, EPEC was detected in only one co-infected sample. Non-typhoidal Salmonella (NTS) was the third most common pathogen detected and previous epidemiological studies demonstrated that infection with drugresistant NTS enterica serotypes was associated with excess morbidity [36]. Based on the antibiotic profiles in this study highlighted that AZM should remain firstline treatment for travelers' diarrhea to Thailand [37]. Norovirus genogroups II and I were detected in several of the cases, but are usually associated with outbreaks of diarrhea. Nevertheless, previous studies have shown that norovirus is becoming commonly detected in both children and adults returning from tropical settings though most laboratories do not commonly test for norovirus in a hospital, clinical setting [38].
Plesiomonas and Aeromonas are not normally associated with travelers' diarrhea though this study indicated that these pathogens were detected in samples with other enteric pathogens. These co-infection results, associated with clinical diarrhea in military patients, support evidence from previous studies that Aeromonas contribute towards the development of diarrhea [39]. Meng et al. reported that synergy or antagonism among pathogens likely affected the degree of diarrheal disease severity more than a single infection in children [40], and that the presence of multiple infections dramatically challenged the ability to properly identify the actual etiological agents of diarrhea disease.
There were several limitations to the study. A relatively small number of diarrheal stool samples were collected with no matched control sample which makes stating that the identified pathogen (s) were truly the cause of the diarrhea. Another limitation is the lack of antibiotic profiles for the bacterial pathogens detected in samples from 2013 to 2014 as the main diagnostic methodology used in these years were ELISAs and TAC. Inclusion of conventional microbiological methods allowed for the determination of antibiotic susceptibility profiles. Due to diagnostic limitations, some pathogens remain undetectable by these methods because they require challenging or unknown unfavorable growth conditions. A previous study indicates that C. consisus and C. ureolyticus are emergent-bacterial diarrheal pathogens [41]. However, these organisms are obligate anaerobes that require a H 2 -enriched atmosphere for optimum growth [42]. Methods to identify these pathogens were not used in this study.

Conclusions
Ongoing diarrheal etiologic agent surveillance studies with antibiotic susceptibility testing should continue in large scale US military exercises these studies relay