Bacterial gastroenteritis in children below five years of age: a cross-sectional study focused on etiology and drug resistance of Escherichia coli O157, Salmonella spp., and Shigella spp.

Gastroenteritis is a common infectious disease in children, which results in high mortality and morbidity, especially in resource-poor countries. This study presents the selected main gastroenteritis causing bacteria, including Escherichia coli O157, Salmonella and Shigella species delineating their prevalence and resistance pattern to the clinically used antibiotics. A total of 346 stool specimens from children of clinically suspected gastroenteritis below five years have been obtained, and bacterial pathogens were recovered using selective media, biochemical and serologic tests. We found that 35 (10%) of them were confirmed bacterial gastroenteritis (BGE) with either Escherichia coli O157 (14%), Salmonella spp. (43%), and Shigella spp. (43%). Nevertheless, the prevalence of E. coli O157, Salmonella spp., and Shigella spp. in this study were documented as 1.45%, 4.34%, and 4.34%, respectively. Moreover, 2 (40%) of E. coli O157, 4 (26.67%) of Salmonella spp., and 14 (93%) of the Shigella spp. were found as multidrug-resistant. Nonetheless, Shigella spp. isolates showed 100%, 87%, and 73% resistance to cotrimoxazole, ciprofloxacin, and azithromycin respectively, while E. coli O157 showed 80% resistance to cotrimoxazole and 60% to ciprofloxacin and azithromycin. On the contrary, cotrimoxazole, ciprofloxacin, azithromycin, and tetracycline were resistant in 40%, 47%, 27%, and 53% of Salmonella spp. isolates respectively. These findings generate significant insights on the prevalence and antibiotic resistance of the three major gastroenteritis causing bacteria in the study area. Therefore, it will help physicians and policymakers selecting the right antibiotics in essential cases, particularly, reevaluation of recommending cotrimoxazole, ciprofloxacin, and azithromycin in the management of bacterial gastroenteritis.

Kingdom (Tam et al. 2012), though they mostly play a role in developing countries where childhood mortality resulting from diarrheal diseases is mostly happened (Guerrant et al. 2013). Shigella spp., enterotoxigenic E. coli, Vibrio spp., and Salmonella spp. are the most common causes of bacterial gastroenteritis in developing countries (O'Ryan et al. 2014), even though etiologic agents may vary with geographical region. World Health Organization (WHO) documented that gastroenteritis caused 477,293 (8%) deaths of children under five years in 2016 globally (WHO 2016). Worldwide, around 68% of diarrheal disease has been documented in young children taking 2.5 million lives, and it is recognized as the fifth top reason for child mortality (Hartman et al. 2019). In children under five years, acute gastroenteritis (AGE) is a principal cause of mortality and morbidity, 19% of deaths and 10% of hospital admission (Oppong et al. 2020) in resource-poor countries where proper hygiene, sanitation, potable water is deficient, and contaminations from food products are high (Farfán-García et al. 2020).
In developing countries, around half a million deaths of infants and young children happened due to AGE, where Shigella is the top etiologic agent (Kotloff 2017). In China, Shigella was found as the most frequent bacterial agent causing AGE (Wang et al. 2015), whereas, in India, E. coli has been recognized as the dominant (31%), followed by Shigella (24%) (Shrivastava et al. 2017;Bruzzese et al. 2018). Worldwide, in 2013, the deaths of children below five years due to shigellosis have been documented between 28,000 to 48,000 (Williams and Berkley 2018). Shigella flexneri and Shigella sonnei are the most frequent species causing infections. Shigella dysenteriae, which produces Shiga toxin recognized as a highly virulent strain, could cause epidemics with high mortality in developing countries. A high level of drug resistance, especially to ciprofloxacin and azithromycin, was observed in those outbreaks (Barrett and Fhogartaigh 2017).
According to an estimation, nontyphoidal (NTS) Salmonella are responsible for 93.8 million cases of gastroenteritis and 155,000 deaths globally per annum (Majowicz et al. 2010). A proportion of childhood diarrhea and travelers' diarrhea (TD) in developing countries are occurred by NTS, where Salmonella enterica var. Typhimurium and var. Enteritidis are frequently recovered (Barrett and Fhogartaigh 2017). A study in China revealed that diarrheagenic E. coli (5%) and Salmonella spp. (8%) were responsible for acute diarrhea in children below five years (Tian et al. 2016). Shiga toxin-producing Escherichia coli (STEC) O157 is a public health concern due to having potential for severity though it rarely caused gastroenteritis (Adams et al. 2016). Nevertheless, STEC O157: H7 is broadly distributed in the environment and foods, and humans typically get infected by contaminated food and drink or direct exposure with animals (Ahmed et al. 2017). Moreover, the expression of different virulence factors, extended survival capability in the environment, low infective dose, and challenges to treatment made STEC O157: H7 an enteric pathogen of principal concern around the world (Lupindu 2018).
Globally, antibiotic resistances have been increasing the risk of morbidity, mortality and economic burden, where developing countries are the primary victim due to the over and unnecessary uses of antibiotics, drugs with compromised quality, misuse of antibiotics in poultry and cattle feed, lack of continuous effective surveillance, and poverty (Chowdhury et al. 2015). The emergence of antibiotic-resistant E. coli O157 has been documented in different parts of the world in recent years. Over the past several decades, various strains of Salmonella spp. and Shigella spp. became resistant to the most widely used antibiotics with increasing frequency (Barrett and Fhogartaigh 2017). In bacterial gastroenteritis, investigations on causative agents and antibiotic treatment are generally not required. Recommendation of antibiotics is given to critically ill children having chronic conditions of particular risk factors. Types and regimens of antibiotics should be considered based on patient conditions, suspected bacteria, and local epidemiology (Buzesse et al. 2018). Importantly, antibacterial resistance data also have to be taken into consideration before selecting antibiotics. Thus, to reduce mortality, morbidity, and healthcare expense caused by BGE, the local pattern of bacterial resistance is considerably essential (Buzesse et al. 2018).
To delineate the prevalence and antibiotic resistance pattern of the three major gastroenteritis causing bacteria, E. coli O157, Salmonella spp., and Shigella spp., in children below five years of age, we have conducted a cross-sectional study in two hospitals of Chattogram district in Bangladesh from September 2017 to July 2018.

Patient selection
Children under five years of age, both male and female, exhibited acute gastroenteritis symptoms and were admitted to Chattogram Maa-O-Shishu General Hospital's diarrheal ward and Chattogram Upazila Health Complex Patiya, Chattogram were selected for this study employing a simple random sampling method. The ward's responsible physician has diagnosed the gastroenteritis in patients according to the definition of World Health Organization (WHO) as "who have diarrhea more than three times per day, with abnormal stool specimen such as watery stool, loose stool, mucus and blood in stool" (WHO 2017).

Ethical consideration
We have taken written permission from the ethical committee of the Department of Microbiology at the University of Chittagong to conduct the study, and written consent has been taken from the patient's attended guardian with informing aims and objectives, merits, and demerits of participation in the study.

Specimen collection
Fresh stool specimens were collected in sterile plastic vials and placed in an insulated box at 4 °C and transported to the laboratory at the Microbiology department, of the University of Chittagong, or Industrial Microbiology Division, Bangladesh Council of Scientific and Industrial Research (BCSIR), Chattogram.

Isolation and identification of bacteria
The stool specimen was enriched in Luria-Bertani broth (Himedia, India) for 24 h at 37 °C, and broth culture was then inoculated onto Salmonella-Shigella agar (Himedia, India) and Hektoen enteric agar (Himedia, India) plates. The plates were incubated for 24 h at 37 °C. Individually recognizable colorless colonies onto Salmonella-Shigella agar plates were primarily selected as Salmonella or Shigella species. Motile isolates that were indole negative, citrate positive, produced red alkaline slant and yellow acidic butt with black colored H 2 S generation in triple sugar iron (TSI) agar were confirmed as Salmonella spp. Conversely, isolates that were non-motile, indole positive, citrate negative, and produced red alkaline slant and yellow acidic butt without H 2 S generation were confirmed as Shigella spp. (Assefa and Girma 2019). On the contrary, green or blue-green, moist, and raised colonies on Hektoen agar plates were presumptively identified as Shigella spp., while colonies with green or blue-green color with or without black center were recognized as Salmonella spp. (Assefa and Girma 2019). Furthermore, for the isolation of E. coli O157, the stool specimen was inoculated in modified tryptone soya broth containing novobiocin (Lab M, UK) for 24 h at 37 °C for the selective enrichment. Then, broth culture was streaked on cefixime-tellurite sorbitol MacConkey agar (CT-SMAC) (Lab M, UK) plates and incubated at 37 °C for 24 h. Colorless bacterial colonies on CT-SMAC agar were primarily selected as E. coli 0157 and then confirmed with slide agglutination with the E. coli O157 Latex test kit (Oxoid, UK).

Antibiotic susceptibility test
Antibiotic susceptibility test (AST) was performed by standard Kirby-Bauer disc diffusion method according to Clinical and Laboratory Standards Institute guidelines (CLSI 2018). The antibiotic discs of azithromycin (15 µg), chloramphenicol (30 µg), cotrimoxazole (25 µg), tetracycline (30 µg), doxycycline (30 µg), and ciprofloxacin (5 µg) (Oxoid, UK) were used in this study. The zone of inhibition for each antibiotic was measured in millimeters, and on the basis of the inhibitory zones, isolates were classified as sensitive, intermediate, and resistant (CLSI, 2018). Isolates resistant to one or more drugs in three or more classes of antibiotics are recognized as multidrug-resistant (MDR) (Magiorakos et al. 2012;Rahman et al. 2021).
In parts of Europe, Japan, and North America, most outbreaks of gastroenteritis are because of enterohemorrhagic E. coli serotype O157: H7, though other serotypes are significantly remained health concerns in the rest of the developed countries. E. coli O157 and Shiga toxinproducing E. coli (STEC) were not commonly found in Bangladesh (Islam et al. 2007). Our study also found BGE positive cases of E. coli O157 were relatively lower (14%, 5/35) than the other two bacteria. The prevalence of E. coli O157 in this study was 1.45%, lower than the other two bacteria. In contrast, a prevalence of over 7% of STEC O157: H7 was reported in the diarrheal patient of Morogoro, Tanzania in 2006 (Raji et al. 2008), whereas 3.4% of isolates were STEC E. coli O157: H7 found in Tunisia (Al-Gallas et al. 2006). Similar studies in Nigeria reported a prevalence of 6% (Olorunshola et al. 2000) in Lagos and 5.4% (Chigor et al. 2010) in Zaria. The pathogen was responsible for diarrhea in children at a proportion of 1.9% in Mozambique (Mandomando et al. 2007;Lupindu 2018).
The emergence of multidrug-resistant (MDR) Shigella spp. is a progressive concern worldwide (Zamanlou et al. 2018). As shigelloses are highly infectious, it is necessary to know the disease's occurrence and antimicrobial susceptibility of the strains to ensure proper clinical care and patient management (Singh et al. 2011). Mild symptoms are self-limited, but antibiotics are prescribed in extreme dysentery cases to shorten the duration of diarrhea (Williams and Berkley 2018). In our study, Shigella spp. showed resistance to cotrimoxazole (100%), ciprofloxacin (87%), azithromycin (73%), chloramphenicol (20%), tetracycline (40%), and doxycycline (20%). Ciprofloxacin-resistant S. dyseteriae 1 first found in Bangladesh in 2003, but S. boydii and S. flexneri were documented in 2007 and 2008, respectively (Talukder et al. 2006;Khatun et al. 2011). Other studies from Bangladesh also revealed that ciprofloxacin-resistant S. sonnei was about 10% in 2007, but in 2011 it was increased by sevenfold (Ud-Din et al. 2013). Zamanlou et al. (2018) reported that 4.2% of Shigella spp. showed resistance to ciprofloxacin in Tabriz, Iran. In contrast, 56.2% of S. flexneri were also resistant to the antibiotic has been documented in India (Aggarwal et al. 2016). A study conducted in Bangladesh between 2001 to 2011 documented that Shigella spp. were resistant to trimethoprim-sulfamethoxazole (89.5%), nalidixic acid (86.5%), mecillinam (10.5%), and ampicillin (9.5%), respectively (Ud-Din et al. 2013). The other study from Bangladesh also stated that 248 (69%) Shigella strains showed resistance to cotrimoxazole (Talukder et al. 2003). The study conducted by Abbasi et al. (2019) revealed that cotrimoxazole (100%), ampicillin (84.2%), cefixime (68.4%), ceftriaxone (63.1%), and ciprofloxacin (10.5%) were affected by Shigella resistance. World Health Organization (WHO) proposed fluoroquinolone (ciprofloxacin) as first-line therapy for all patients with bloody diarrhea without age restriction, and ceftriaxone is recommended as second-line therapy or alternative antibiotics in children and adults; because of resistance to trimethoprim-sulfamethoxazole, sulfonamides, ampicillin, and tetracycline (WHO 2019). In developing countries, MDR was rising in Shigella spp. mainly to azithromycin, fluoroquinolones, and thirdgeneration cephalosporins . The raised resistance level is mostly because of the horizontal transfer via plasmids, transposons, and integrons (Barrantes and Achí 2016). Though cotrimoxazole is broadly used in the empirical management of diarrheal diseases, and its widespread application has contributed to the emergence of resistance of Shigella spp. (Pourakbari et al. 2010). Previously, a high level of resistance to cotrimoxazole has been reported in Iran (92.2% to 94%) (Pourakbari et al. 2010;Nikfar et al. 2017) and Turkey (95%) (Kacmaz et al. 2014). Our findings were also consistent with their results that 100% of all Shigella spp. found resistant to the antibiotic. Mahmoudi et al. (2017) reported that the frequency of trimethoprim/sulfamethoxazole resistance among Shigella spp. was high (92%), which agrees with reports from Ethiopia (Asrat 2008). These results were similar to Mengistu et al. (2014), Savadkoohi and Ahmadpour-Kacho (Barari and Ahmadpour 2007), and Jomezadeh et al. (2014), which documented trimethoprim/sulfamethoxazole resistance rates of 76.5%, 73.8%, and 80.5%, respectively. These findings may indicate inappropriate antibiotic use in the recent decade (Mahmoudi et al. 2017). Though rates of multidrug resistance (MDR) of Salmonella to traditional first-line agents in both Asia and Africa is increasing (Gordon et al. 2008), Leung et al. (2013) found a marked decline in MDR and decreasing resistance to the ampicillin, chloramphenicol, and cotrimoxazole but increasing resistance to ciprofloxacin in Bangladesh. In our study, Salmonella spp. were found resistant to tetracycline (53%), ciprofloxacin (47%), cotrimoxazole (40%), azithromycin (27%), chloramphenicol (7%), and doxycycline (7%). Several reports from around the world also showed the occurrence of a high level of ciprofloxacin-resistant isolates. The studies conducted by Mahmoudi et al. (2017), Mamuye et al. (2015), Mengistu et al. (2014), and Beyene et al. (2011) found the resistance of Salmonella isolates to trimethoprim/ sulfamethoxazole as 23%, 60%, 37.5%, and 31.5%, respectively.
Bacterial gastroenteritis is self-limiting; therefore, antibiotic therapy is only needed as an intervention to control BGE in infants, older people, and immunosuppressed patients, especially when bacteremia is suspected. Empiric therapy usually starts with oral cotrimoxazole or metronidazole, but parenteral treatment with ciprofloxacin or ceftriaxone could be preferred (Bruzzese et al. 2018).

Conclusions
This cross-sectional study delineated the prevalence of bacterial gastroenteritis in children below five years of age caused by the three significant pathogens Escherichia coli O157, Salmonella spp., and Shigella spp. in the Chattogram district of Bangladesh. The study findings revealed that the prevalence of those bacteria causing gastroenteritis is alarming. Preventive measures are highly recommended to reduce the disease burden and healthcare expense, eventually improving life quality. The findings of antibiotic resistance patterns will help physicians and policymakers select the right antibiotics where necessary, especially making a rational decision while choosing cotrimoxazole, ciprofloxacin, and azithromycin to manage bacterial gastroenteritis cases.