Diarrhea caused by enteric infections is a major factor in morbidity and mortality worldwide. The World Health Organization (WHO) has estimated that billion episodes of infectious diarrhea occur each year and are especially prevalent in infants and children younger than 5 years   .
Enterotoxigenic Escherichia coli (E. coli), or ETEC, and Shigella species remain major contributors to acute enteric infections, particularly in the developing countries, where they account for about one billion cases of diarrhea annually. They are responsible for almost one-third of child deaths from diarrhea, as well as many deaths in older age groups    .
The ETEC infection in developing countries is usually frequent in infants younger than 2 years of age, with a decrease after 5 years of age. Repeated ETEC infections among children in these countries are not rare and may be due to both environmental and immunological factors. The diarrhea produced by ETEC is characterized by a rapid onset of watery stool (without blood or inflammatory cells) due to intestinal colonization and production of one or more plasmid-encoded enterotoxins  .
Infection caused by Shigella species (shigellosis) is endemic worldwide and is responsible for an estimated 120 million cases of severe dysentery annually, the majority of which occur in children from developing countries. Shigellosis symptoms include abdominal pain, cramps, fever, vomiting and bloody diarrhea and with mucus in stool   .
Diarrheal illness rarely requires antimicrobial treatment and often treatment is given empirically depending on the severity of the disease and on the risk of complications. To guide the empirical choice of antibiotics, it is crucial to know both which pathogens are most likely to be infecting the patient in a particular geographic area and the most effective antibiotics for treating them  .
In addition, the overuse and misuse of antibiotics in the treatment of diarrhea could lead to an increase of antibiotic resistance. Several studies from developing countries showing worrying trends in multiple resistance among enteric pathogens such as Escherichia coli and Shigella spp.  .
Despite the high prevalence of ETEC and Shigella infections, few studies have been conducted to investigate their prevalence and susceptibility pattern in children our country. Therefore, the aim of this work was to determine the current prevalence and antimicrobial susceptibility patterns of ETEC and Shigella isolates from children in Merida city, Mexico, to provide information for the selection of appropriate empirical treatment of these infections in our region.
Specimen collection: This study was conducted in Merida city and was approved by the Ethical Committee of the Regional Research Centre Dr Hideyo Noguchi of the Autonomous University of Yucatán (UADY) (protocol CEI-CIR-UADY-2012-15). This was a descriptive study conducted over period of one year from January 2013 to January 2014.
Microbiological Analysis: A total of 200 stool samples of children with diarrhea referred to the Clinical Analysis Laboratory of Community Services of the Faculty of Chemistry, UADY, and Friendship’s hospital of Mérida City from Mexico. There were spread on Mac Conkey agar, xylose lysine deoxycholateagar (XLD), and Salmonella-Shigella agar(S-S) agar and were incubated at 37˚C for 18 - 24 h. The presumptive colonies were fully identified using standard biochemical tests, including Triple Sugar Iron (TSI), Sulphur Indole and Motility (SIM), Urea, Simmons citrate, and Lysine Iron Agar (LIA) according to results described on Table 1.
Identification of Enterotoxigenic Escherichia coli by PCR: E. coli clinical isolates were processed for isolation of genomic DNA as previously described  . In brief, overnight liquid cultures were centrifuged, and the pellet was resuspended in TE buffer (10 m MTris-HCL pH 8.0, 5 m MEDTA), boiled for 10 min, and centrifuged again. The supernatant containing a crude DNA extract was used as a DNA template for PCR assays.
Multiplex PCR reactions with specific ST primers (5’-GCTAAACCAGTAGAGCTCTTCAAAA-3’ and 5’-CCCGGTACAAGCAGGATTACAACA-3’) to amplify a 147-bp fragment and specific LT primers(5’-GCACACGGAGCTCCTCAGTC-3’ and 5’-TCCTTCATCCTTTCAATGGCTTT-3’) to amplify a 218-bp fragment were performed as previously reported by Rugeles et al. (2010).
Serological identification of Shigella species: All Shigella clinical isolates were identified by biochemical properties by slide agglutination test, using commercially available antisera (Serobac, BD BBLTM, USA). These antisera were S. dysenteriae poly A, S. flexneripoly B, S. boydii poly C, and S. sonnei poly D.Briefly, strains were subcultured on tryptic soyagar (Difco) and tested for agglutination on glass slides. Slides were divided into two sections with a wax pencil. A drop (20 μl) of 0.85% NaCl solution was placed in one section for use as a negative
Table 1. Biochemical identification of E. coli and Shigella spp.
control, and a drop of the appropriate antiserum was placed in the other section. By using a sterile inoculation loop, a portion of the culture was emulsified with the NaCl solution and a second independent portion of the culture was mixed in the section of the slide containing the corresponding antiserum. The slide was then gently rocked, and relative agglutination was scored after 60 s. Two reference strains, Shigella flexneri ATCC 12022 and Shigella sonnei ATCC 9290, were used as control strains for positive agglutination.
Antimicrobial susceptibility testing: ETEC and Shigella strains were analyzed for their antimicrobial susceptibility by Kirby-Bauer disc diffusion method according to the Clinical and Laboratory Standards Institute (CLSI) guidelines. The antibiotics tested were: ampicillin (AMP, 10 μg), amikacin (AK, 30 μg), carbenicillin (CB, 100 μg), cephalothin (CF, 30 μg), cefotaxime (CTX, 30 μg), ciprofloxacin (CIP, 5 μg), chloramphenicol (CL, 30 μg), gentamicin (GE, 10 μg), netilmicin (NET, 30 μg), nitrofurantoin (NF, 300 μg), norfloxacin (NX, 10 μg), and trimethoprim/sulfamethoxazole (SXT, 30 μg). E. coli ATCC 25922 was used for quality control purposes according to the CLSI standards. After 24 hrs of incubation the diameter of zone of inhibition was measured in millimeter (mm) and the sensitivity pattern of each isolate was recorded accord to the inhibition zone size scale provided by the CLSI standards. For the data analysis, isolates categorized as intermediate were considered as resistant isolates.
3. Results and Discussion
During the 12-month study period a total of 200 stool specimens were examined from children with diarrhea. The subjects comprised 89 males (44.5%) and 111 females (55.5%), with a median age of 2.3 years (SD ± 1.3) and ranging from 0 - 5 years old.
Escherichia coli strains were detected in 124 out of 200 (62.0%) samples analyzed. Of these E. coli strains, 18 (14.5%) were positive for ETEC by PCR (Figure 1). The genes lt and st were successfully amplified in all samples, indicating that all strains correspond to ETEC.
Figure 1. Identification of ETEC by PCR based on the lt and st genes. Ethidium bromide stained agarose gel showing PCR products of 147 and 218 bp for st and lt genes, respectively; after amplification of genomic DNA extracted from E. coli isolates from diarrheic stools in children. Lane 1: 50 bp molecular weight marker; lane 2: positive control; lane 3: negative control; lanes 4 - 6: positive strains for ETEC.
In our study, the prevalence rate of ETEC infection was of 9.0% among children with diarrhea under 5 years of age. These results agree with the rates determined in Turkey and Iran   . However, this rate is lower than the reports from Tunisia and Egypt   .
The highest prevalence of ETEC infection (55.5%) was found between the 2 and 3-year-old-children, followed by the children under one year of age and between the 1 and 2-year-old-children, with a prevalence of 27.7% and 16.7%, respectively (Table 2). These findings reflect that ETEC was an important pathogenic bacteria in the population of children studied.
Some reports indicate that ETEC infection is the first cause of diarrhea illness experienced by infants from endemic areas of developing countries during their first year of life   . In a study conducted in our country in 2009, Estrada-García et al.  determined a 37.5% prevalence rate for ETEC in children (<2 years) with acute diarrhea. In that study for 0 - 1-year-old group; ETEC was detected only in 17% and for our study the prevalence was 28.7% for children in the same age. Nevertheless our study demonstrated that the infection for ETEC was more prevalent in children < 2.5 years of age. Our findings are supported by previous studies of incidence of ETEC infections in developing countries, which revealed that ETEC was detected at a constant rate from cases of acute diarrhea during the first two years of life, with a de-crease of infections after three years  .
In our study, both toxins coding genes was amplified in all ETEC strains isolated from children with diarrhea. LT expressing ETEC strains are considered to be less pathogenic than those expressing ST  .
Among the 200 stool samples subjected to analysis, 12 (6.0%) were positive for Shigella species infection in children under five years old. The highest prevalence of Shigella infection (9 of 12, 75.0%) was found in children under one year of age, and especially those aged 7 - 11 months (Table 2). Shigella flexneri was identified in 8 out of 12 Shigella isolates (66.7%), corresponding the most prevalent specie detected in these children, with a median age of 1.3 years (SD ± 0.8), followed by Shigella boydii (4/12; 33.3%), with a median age of 1.2 years (SD ± 0.1). Neither Shigella sonnei nor Shigella dysenteriae were detected in any of the children studied.
Table 2. Enterotoxigenic E. coli and Shigella isolates from children with diarrhea in Merida, Mexico.
NI: Not identified.
Shigella species are predominantly isolated from the stools of children with bloody diarrhea causing 50% or more of all episodes in endemic regions of the developing countries. In this study, Shigella infection prevalence was 6.0% (12 of 200). This prevalence was similar to other studies performed across the world with children in the same age group, particularly African and Asian countries      . However, it is lower than the rate reported in a study conducted in our region by Zaidi et al.  where determined a 12.1% prevalence rate for Shigella infection in children (0 - 10 years) with diarrhea in 2011. In that study, 54.0% of children belong to an age group of 1 - 4 years. If the rate for 1 - 4-year-old group had been calculated, the Shigella prevalence rate would probably have been around the same percentage as in our study, since ten of the twelve isolates correspond to children with a median age of 1.5 years (SD ± 0.5). Evaluation of the results of the study by Zaidi et al. and our study shows that the prevalence rate of Shigella in our region is predominately in children less than 5 years.
The geographical repartition and the pathogenicity of the four Shigella species are different by country  . In this study, we found S. flexneri was the most common species, followed by S. boydii. These results agree with previous reports that indicate that S. flexneri is the most commonly isolated species in the developing world and the most frequent cause of bacterial dysentery  .
There were no significant sex-related differences in the incidence of both ETEC and Shigella infections in the children studied. The antimicrobial susceptibility testing results of all 18 ETEC and 12 Shigella strains isolated from the stool samples of children of 0 - 5 years of age are shown in Table 3.
Several of the ETEC strains were resistant to many antimicrobial drugs. The highest level of resistance was detected for ampicillin, carbenicillin, cephalotin, and nitrofurantoin in which all (100%) ETEC isolates were found to be resistant. The resistance rates of ETEC isolates to the other antibiotics were 66.6% for cefotaxime, 61.1% for chloramphenicol, followed by ciprofloxacin and gentamicin
Table 3. Antimicrobial drug resistance of ETEC and Shigella species isolated from children with diarrhea in Merida, Mexico.
AMP, ampicillin; AK, amikacin; CB, carbenicillin; CF, cephalothin; CTX, cefotaxime; CIP, ciprofloxacin; CL, chloramphenicol; GE, gentamicin; NET, netilmicin; NF, nitrofurantoin; NX, norfloxacin; STX, trimethoprim/sulfamethoxazole. (a)-Includes S. flexneri and S. boydii strains.
with a 50.0% of resistance each, and 44.4% ETEC resistance for amikacin. ETEC isolates showed 33.3% resistance to netilmicin and trimethroprim-sulfamethoxazole. The lowest level of resistance was detected for norfloxacin (27.7%).
Resistance patterns of Shigella spp. against the applied antimicrobials is given in Table 2. All of the 12 strains of Shigella (100%) were resistant to ampicillin, carbenicillin, and cephalothin, followed by nitrofurantoin (83.3%). Amikacin was the only antimicrobial to which all Shigella isolates were susceptible.
When we analyzed the results considering resistance patterns in different species, we identified that S. flexneri isolates displayed 100% resistance to ampicillin, carbenicillin, and cephalotin, and 75.5% for nitrofurantoin; while resistance for the other antibiotics were not detected. On the other hand, all S. boydii strains showed resistance to almost all antimicrobials tested except ciprofloxacin, gentamicin and netilmicin, to which 50% showed resistance.
All ETEC and Shigella isolates showed resistance against at least three antibiotics. These isolates were defined as multidrug resistant (MDR) strains. Six MDR patterns were dominated by ETEC isolates (Table 4). Of the 18 ETEC isolates, 8 (44.4%) showed multidrug resistance to nine antibiotics with two different MDR patterns, followed by resistance to seven antibiotics (33.3%), a MDR patterns to six antibiotics with a prevalence of 11.1%, and MDR patterns for eight and four antibiotics, with a prevalence of 5.6% each.
All the isolates of S. flexneri showed two MDR patterns. The most prevalent MDR patterns in S. flexneri strains were AMP-CB-CF-NF with a prevalence of
Table 4. Multidrug resistance patterns of ETEC and Shigella isolates from children with diarrhea in Merida, Mexico.
AMP, ampicillin; AK, amikacin; CB, carbenicillin; CF, cephalothin; CTX, cefotaxime; CIP, ciprofloxacin; CL, chloramphenicol; GE, gentamicin; NET, netilmicin; NF, nitrofurantoin; NX, norfloxacin; STX, trimethoprim/sulfamethoxazole. (n)-number of antibiotics in the MDR pattern.
75.0%, and the AMP-CB-CF pattern was the second most prevalent multiresistance pattern with a prevalence of 25.0%. Likewise, all S. boydii isolates were shown to be MDR strains. Two S. boydii isolates showed a MDR pattern to 11 out of 12 antibiotics tested and the other two isolates showed a resistance for 8 antibiotics (Table 4). The aminoglycosides were the most active antibiotics against both Shigella species.
The increase in antibiotic resistance among ETEC and Shigella isolates are becoming a serious problem, particularly in developing countries. The incidence varies with the area of isolation of these strains. In our study, a total of 18 ETEC and 12 Shigella strains isolated from children were tested against 12 different antibiotics. We report a high prevalence of antimicrobial resistance among ETEC and Shigella isolates. All ETEC and Shigella isolates were 100% resistant to nitrofurantoin and to two penicillins: ampicillin and carbenicillin. Previous studies reported a prevalence of ampicillin resistance more than 75% for ETEC isolates from children in developing countries, including other regions of Mexico, although resistance levels does not raise the obtained in the present study    . On other hand, penicillins are not recommended for Shigellaempiric treatment because of the high resistance in many regions of the world   .
Cephalosporins and fluoroquinolones constitute another two major groups. We used two drugs representing the cephalosporins: cephalotin and cefotaxime. All ETEC isolates showed a total resistance to both antibiotics. Shigella isolates were 100% resistant to cephalotin. Interestingly, S. boydii, but not S. flexneri, showed a total resistance to cefotaxime. With the fluoroquinolones: ciprofloxacin and norfloxacin, S. boydii also showed the higher resistance pattern in comparison with S. flexneri. These data indicates that the appropriate antibiotic treatment of shigellosis should depend on identifying resistance patterns to species-level. However, norfloxacin was the most effective drug against ETEC, as 72.3% isolates were sensitive to it.
Aminoglycosides are an important group of drugs highly effective against Gram-negative bacteria. We used amikacin, gentamicin and netilmicin as representatives of this group. Sensitivity pattern was identified in more than 50% of all ETEC isolates for the three aminoglycosides. Netilmicin was more effective showing resistance 33.3% by ETEC strains. Regarding Shigella isolates, no resistance was detected to amikacin. These results are in accordance with some earlier reports of resistance pattern for both enteric pathogens   . Nevertheless, these drugs have generally not been the first choice of treatment for diarrheic illness and the use in clinical practice is limited, and therefore, no higher prevalence of resistant bacteria are expected.
The growing problem of multidrug resistant enteric pathogens is especially common in developing countries from Latin America, Africa and Asia    . All our isolates showed resistance to at least three drugs in different combinations, therefore all the isolates could be classified as MDR. Also, we found a high frequency of MDR to drugs belonging to structurally different antimicrobial groups. Until six different MDR patterns with a minimum of four drugs (AMP, CB, CF, and NF) were present in all ETEC strains. It is unclear why ETEC isolates from this area manifests such high rates of resistance; could be because of differences in antimicrobial usage which is not effectively controlled.
Antibiotics always are used to treat Shigella infection in children, because it may help to reduce the clinical course and complications of illness, and also to reduce the duration of the symptoms. However, several studies have been reported multidrug resistance to the antibiotics most commonly used to treat shigellosis    . In the present study we detected two MDR pattern for S. flexneriisolates, however all isolates were susceptible to 8 of 12 antibiotics tested, including the fluoroquinolones which are recommended as the drugs of choice for shigellosis by WHO. In the case of S. boydii, although the number of isolates were low, can give us a perspective of the multiresistance present in our region. All S. boydii isolates showed two different MDR pattern with a minimum of eight of the 12 antibiotics tested. This finding was in line with some earlier reports conducted in developing countries   .
This study reflects a significant prevalence of ETEC and Shigella strains as responsible of diarrhea episodes in children from southeast Mexico. The majority of the isolated strains showed high rates of resistance to antibiotics which may lead to problems in the cases requiring antibiotic treatment. We conclude that MDR is very frequent in ETEC and Shigella species in our region, therefore the prescription and use of antimicrobial agents to treat diarrhea infections should be considered several factors including specific identification of bacteria and their antimicrobial resistance profile.
This work was supported by the Universidad Autónoma de Yucatán through the Programa Integral de Fortalecimiento Institucional (PIFI).