Multidrug resistance pattern and molecular epidemiology of pathogens among children with diarrhea in Bangladesh, 2019–2021

Antimicrobial and multidrug resistance (MDR) pathogens are becoming one of the major health threats among children. Integrated studies on the molecular epidemiology and prevalence of AMR and MDR diarrheal pathogens are lacking. A total of 404 fecal specimens were collected from children with diarrhea in Bangladesh from January 2019 to December 2021. We used conventional bacteriologic and molecular sequence analysis methods. Phenotypic and genotypic resistance were determined by disk diffusion and molecular sequencing methods. Fisher’s exact tests with 95% confidence intervals (CIs) was performed. Prevalence of bacterial infection was 63% (251 of 404) among children with diarrhea. E. coli (29%) was the most prevalent. E. coli, Shigella spp., V. cholerae, and Salmonella spp., showed the highest frequency of resistance against ceftriaxone (75–85%), and erythromycin (70–75%%). About 10–20% isolates of E. coli, V. cholerae and Shigella spp. showed MDR against cephem, macrolides, and quinolones. Significant association (p value < 0.05) was found between the phenotypic and genotypic resistance. The risk of diarrhea was the highest among the patients co-infected with E. coli and rotavirus [OR 3.6 (95% CI 1.1–5.4) (p = 0.001)] followed by Shigella spp. and rotavirus [OR 3.5 (95% CI 0.5–5.3) (p = 0.001)]. This study will provide an integrated insight of molecular epidemiology and antimicrobial resistance profiling of bacterial pathogens among children with diarrhea in Bangladesh.

Nucleotide sequence and phylogenetic analyses.The phylogenetic analysis revealed that the partial amplicons of qnrB (468 bp) isolated from E. coli were closely related with the previously reported reference qnrB  Mutational analysis of bla TEM .After aligning the sequences against reference sequences of respective resistance genes we searched for the presence of mutations.However, we did not find any significant mutations in the tet, qnrB and mcr-1 genes.We found significant mutations in the bla TEM resistance genes.In most of the study strains the amino acids up-to 158 residues were highly similar with the reference sequences.

Discussion
According to the world health organization diarrhea is the 2 nd major cause of mortality and the leading cause of malnutrition among children under five years 1-3 .The health burden of diarrheal diseases is severe in developing countries like Bangladesh 3 .The circulation of antimicrobial resistance strains of these bacterial pathogens is intensifying the health burden of diarrheal disease in Bangladesh [13][14][15] .We detected significantly higher prevalence of bacterial infection (63%) than any other previous reports from Bangladesh 1,3,9 .Further, the proportionate incidence of diverse pathogens was alarming.We found E. coli (29%) in higher frequency followed by Shigella spp.(17%) and V. cholerae (13%).The diversity of bacterial pathogens associated with diarrhea in children are similar with previous studies 3,9 .The higher prevalence of Shigella spp.requires further investigation and V. cholerae may indicate the constant presence of cholera endemic in these regions.Findings are in good agreement with previous studies in Bangladesh, China, India, Iran and WHO data from developing countries 2,8,[15][16][17][18][19][20][21][22][23][24][25] .
We found significantly diverse pathogens in children with diarrhea.Mixed infection with rotavirus, norovirus, adenovirus and human bocavirus was simultaneously found in the positive specimens of bacterial infection.Dual infection by rotavirus and E. coli was associated with (15%) of the cases followed by V. cholerae and Shigella spp.(7.7%).Further, we detected that 31% specimens had co-infection with two pathogens and 12% with three pathogens.Few previous studies have also reported the presence of multiple pathogens in the same specimens in Bangladesh.Our findings are in good similarity with the previous findings in Savar and Tangail in Bangladesh 3,9,18 .
Association of coinfection with the clinical symptoms were also analyzed in this study.We found that presence of coinfection was significantly associated with higher odds of diarrhea, vomiting, dehydration and abdominal pain in children under five.The presence of multiple pathogens in a single patient makes it difficult to intervene or treat the conditions earlier.Among the co-infected samples, we could not determine which pathogen infected earlier.However, the presence of coinfection was associated with severe and chronic outcomes in children under five.These findings are supported by the data of previous studies 3,5,8,9,18 .These findings also indicated that the existing hygienic conditions or setting of the study area favored the spread of multiple pathogens associated with diarrhea.
Significantly higher prevalence of antimicrobial resistance (AMR) pathogens from family Enterobacteriaceae was confirmed by the disk diffusion method in this study.About 60% of the isolates of E. coli, V. cholerae, Salmonella spp., and Shigella spp.were resistance against ceftriaxone, erythromycin, norfloxacin, ciprofloxacin and tetracycline.These antibiotics are most commonly used in treatment of diarrheal diseases caused by Enterobacteriaceae and supported by the CLSI guidelines.These findings are supported by previous studies in Savar and Tangail Bangladesh and other developing countries 3,15-21,24-27 .However, we found higher incidence of antimicrobial resistance pathogens than the previous studies in Tangail 18 and Savar 3 , which suggests that the problem of AMR is increasing rapidly and widely in Bangladesh.These findings suggest that use of any of these antibiotics against E coli, V cholerae, Salmonella spp. or Shigella spp.will be less effective option in treatment, which is also supported by previous studies 24,25 .Except imipenem, meropenem and colistin, significant resistance activity was found in all of the detected bacterial pathogens among children.Among the tested antibiotics, the highest frequency (nearly 70%) of sensitivity was found against imipenem and meropenem.Though colistin is not used clinically in treatment of diarrhea caused by Enterobacteriaceae, we found higher prevalence of colistin resistance pathogens.This outcome is supported by previous studies in Bangladesh 3,18,27,28 .
The multidrug resistance (MDR) profiling of the isolated pathogens was alarming.Significantly higher prevalence of MDR was detected (about 15%) among the isolates of E. coli, V. cholerae and Shigella spp.against several antibiotics including ceftriaxone, erythromycin, norfloxacin and chloramphenicol.This finding is supported by previous studies in Bangladesh, China and Iran 3,14,17,18,[25][26][27][28] .However, the report of MDR from pathogenic bacteria isolated from children with diarrhea is less available in Bangladesh.In addition, higher prevalence of MDR against these important and commonly used antibiotics is staggering.The prevalence of MDR strains of pathogenic bacteria might be a significant cause of severe and prolonged health outcomes among children with diarrhea in Bangladesh.The rapid and uncontrolled increase in use of antibiotics as human treatment and animal feed and treatment options has significantly contributed to the rise and spread of MDR, which is supported by the previous studies 28 .
Integration of molecular characteristics and epidemiological data can contribute significantly in tracing the origin and transmission of AMR and MDR strains of pathogenic E. coli, V. cholerae, Salmonella spp.and Shigella spp.Which will ultimately aid in tracing the sources of infection and contribute in reduction of health burden among children.Antibiotic resistance genes against different group of antibiotics were found among the isolates of E. coli, V. cholerae, Salmonella spp.and Shigella spp.Resistance genes against ciprofloxacin (qnrB), cotrimoxazole (sxt) and colistin (mcr-1) were found in lower frequency among the isolates of E. coli (⁓10%), V. cholerae (⁓13%), Salmonella spp.(⁓14%) and Shigella spp.(⁓15%).Antibiotic resistance gene bla TEM was found in highest frequency among the isolates of Salmonella spp.(45%) followed by E. coli (36%), and V. cholerae (36%).Further, tetA gene was detected among 18-27% isolates of these enteropathogenes.These findings are in good agreement with previous studies in Bangladesh 3,18 .However, the diversity and prevalence of resistance genes were significantly higher in this study than previous reports.Further, we found that the presence of resistance genes in pathogenic E. coli, V cholerae, Salmonella spp., Shigella spp.were significantly associated with the phenotypic resistance properties against ceftriaxone, cotrimoxazole, ciprofloxacin, tetracycline and colistin.The findings of these study are supported by the previous studies in Bangladesh and China 3,14,[17][18][19][20][26][27][28] . This i one of the first study to report the association between phenotypic and genotypic resistance of bacterial pathogens in Bangladesh.These findings are alarming for the pediatric health in Bangladesh as these bacterial pathogens contribute for a significant number of cases.Both AMR and MDR pathogens with the presence and association of resistance genes with treatment failure among the children with diarrhea require extensive studies in future.
The prevalence of resistance E coli, V cholerae, Salmonella spp.and Shigella spp.were higher in this study than previous studies in Bangladesh, India, Nigeria, Pakistan and Iran 13,16,18,[23][24][25] .This phenotypic resistance and presence of resistance genes indicate probable continuous present of AMR pathogens among children in Bangladesh.It is established that misuse and over use of antibiotics have contributed to the origin and spread of resistance properties among pathogens.Importance should be given to appropriate diagnosis of pathogens before suggesting and use of antibiotics in children with diarrhea.Strict and widespread public health surveillance, education and policy are required to prevent further misuse and overuse of antibiotics in Bangladesh.
Findings from this study will add knowledge in the pathogenic diversity of children with diarrhea.Further, integrated approach to determine the prevalence of antimicrobial resistance and multidrug resistance isolates added significant information in the field.This study will contribute in determining and understanding the actual scenario and causes of antimicrobial resistance problem and their molecular elements, which will aid in accurate diagnosis and treatment of diarrheal patients in Bangladesh.

Conclusion
We detected high prevalence of E coli, V cholerae, Salmonella spp.and Shigella spp.among children with diarrhea in Bangladesh.Co-infection of viruses was also found at higher frequency among the children.Further, we detected higher incidence of antimicrobial and multidrug resistance isolates among these pathogenic bacteria in children with diarrhea.Findings from this study also suggested that presence of co-infection increased the severity and complications of the disease among children.Findings from this study will contribute in policy making to understand and reduce the health burden associated with antimicrobial and multidrug resistance enteropathogens.

Methods
Ethical approval.The ethical clearance was obtained from the Biosafety, Biosecurity & Ethical Committee at the Jahangirnagar University.The informed consent was taken from the participants before taking the survey.

Method guidelines.
The authors confirm that all the methods were performed in accordance with the relevant guidelines and regulations.All methods were performed in accordance with the our previously published articles 3,5,7,8 .For bacterial culture and antibiotic sensitivity test, guidelines from ATCC (https:// www.atcc.org/) and CLSI (https:// clsi.org/) were followed, respectively.

Study population and fecal specimens.
A total of 404 fecal specimens were collected from patients with diarrheal diseases from 3 clinics of two different localities (Savar and Tangail) in Bangladesh through January 2019 to December 2021.Samples were collected from Enam Medical College (135 of 404) and Dip Clinics (125 of 404) in Savar and Kumuduni Women's Medical College (144 of 404) in Tangail.Convenience sampling method was used to enroll the participants.Epidemiological and demographic data were collected from children aged below 18 years with acute gastroenteritis.The participants of this study were divided into seven age groups including the range in months, 1-3, 4-11, 12-23, 24-35, 36-47, 48-60 and > 60 (Table 1).Clinical symptoms data of children were analyzed.Informed consent was obtained from the parents of patients.Inclusion criteria to take samples were patients age below 18 years, have symptoms of diarrhea according to WHO definition, reported in the clinic's outpatients, inpatients or emergency department.Exclusion criteria were patients aged above 18 years, did not report of their diarrheal cases in these clinics during the study period, missing demographic data, and having other health complications.One fecal specimen was collected from every participant.After collection, the specimens were stored at − 20 °C.Transportation of the fecal specimens were conducted by maintaining − 20 °C temperature in ice box.All the specimens were stored at − 20 °C after being transported under maintaining proper conditions 3 .All methods were performed in accordance with the our previously published articles 3,5,7,8 .
Antibiotic susceptibility test of enteric bacterial pathogens.Antibiotic susceptibility test was performed by following the Kirby-Bauer Disk Diffusion Test and according to the CLSI guidelines 3,29 .Antibiotics of nine different groups were used, following the CLSI manual (Table 5).The susceptibility of the pathogens to each antibiotic was determined by measuring the zone diameter, which was further classified as resistance, intermediate, or susceptible according to the CLSI guidelines.E. coli ATCC 25922 strain was used concurrently as control to determine the validity of the test protocol.
Bacterial genome extraction and polymerase chain reaction.The bacterial genome was extracted by using the previously described boiled DNA method 3,30 .Further, genome of virus was isolated and purified (Supplementary material).The isolated bacteria were identified by using polymerase chain reaction (PCR).The www.nature.com/scientificreports/following sequences of 16S rRNA primers (F-AGT TTG ATC CTG GCT CAG and R-ACC TTG TTA CGA CTT) were used for PCR reaction 3 .The PCR reaction mixture contained 12.5 µl of 2X master mix (GoTaq Green Master Mix, Promega, USA), 1 µl of forward (F) primer, 1 µl of reverse (R) primer, 6.5 µl of nuclease-free water, and 4 µl of template DNA (Eppendorf, Germany).The final volume of reaction mixture was 25 µl.The PCR reaction was conducted in a thermal cycler (2720 Thermal Cycler, Applied Biosystems, USA).The PCR reaction was performed at 94 °C for 5 min, followed by 35 cycles at 94 °C for 30 s, 53 °C for 30 s, and 72 °C for 60 s 3 .The details of the PCR conditions for the detection of antimicrobial resistance genes are provided in the supplementary material.

Statistical analysis.
We used percentage to present categorical variables and mean/ median for continuous variables.Inferential statistics (p value) was applied for the analysis.We calculated odds ratio (OR) of categorical variables by using two-tailed Chi-square or Fisher's exact tests with 95% confidence intervals (CIs).For twotailed tests p value < 0.05 were considered as significant.We performed data analysis by using Statistical Product and Service Solutions (SPSS v24.0) software (IBM, US).

Figure 2 .
Figure 2. Phylogenetic relationship of antimicrobial resistance genes isolated from bacterial pathogens.(a) Partial sequences of qnrB gene from E. coli were used for constructing this tree, (b) Partial sequences of mcr-1 gene from E. coli were used for constructing this tree, (c) Partial sequences of bla TEM gene from E. coli were used for constructing this tree, (d) Partial sequences of tetA gene from V. cholerae were used for constructing this tree.The trees were constructed by using maximum likelihood model with a bootstrap value of 1000 by using MEGAX.The scale bars in the trees indicated nucleotide substitutions per site for each tree separately.Partial sequence of 16S rRNA from Bacillus spp was used as outgroup in tree A and partial sequence of carbapenem resistance gene from E. coli was used as out group in tree (b), (c) and (d). https://doi.org/10.1038/s41598-023-41174-6

Table 1 .
Risk of infection by different diarrheal pathogens among children according to their demographic conditions.

Table 2 .
Risk analysis for disease outcome among children infected with mono-pathogens and co-pathogens of diarrhea.

Table 4 .
Association of phenotypic and genotypic resistance of bacterial pathogens isolated from pediatric patients with diarrhea.Phen Phenotypic, Gen Genotypic.

Table 5 .
Performance standards for E. coli, V. cholerae, Salmonella spp.and Shigella spp.isolated from children with diarrhea to antimicrobial susceptibility testing performed in this study.S Sensitive, I Intermediate, R Resistance.

Table 6 .
List of primers used in this study.bp base pairs, T a Annealing temperature.