Antibiotic-Resistant Strains of in 50 Antibiotic Treatment-Naïve Children in Northeast Poland Diagnosed by Gastric or Duodenal Biopsy Between February 2019 and May 2022

Background

Helicobacter pylori (H. pylori) infection is very common, with prevalence rates of more than 80% in developing countries and less than 40% in developed countries [1–6], and it is one of the most frequently diagnosed bacterial infections of the gastrointestinal (GI) tract worldwide [1–4]. It was found that H. pylori infection was significantly associated with lower economic status, having a mother/siblings infected with H. pylori, lack of access to a sewage system, and drinking unboiled or non-treated water [6]. Fortunately, improvements in sanitation and hygiene around the world have reduced the incidence of this infection [7]. H. pylori is often transmitted in childhood by saliva through person-to-person interactions and persists in the absence of treatment [1,5,8]. Recently, it has been suggested that certain factors, such as smoking and drinking alcohol, may affect H. pylori infection and the effectiveness of therapy [9–13].

This bacterium causes many diseases, including chronic gastritis, peptic ulcer disease (PUD), gastric mucosa associated lymphoid tissue lymphoma (MALT), and gastric adenocarcinoma [7,14–17]. In children, the most common forms of H. pylori infection are PUD and chronic gastritis, although the latter is usually asymptomatic. In contrast, MALT and gastric adenocarcinoma only develop occasionally in the pediatric population [7,14–17].

According to the latest guidelines of the European Society for Pediatric Gastroenterology Hepatology and Nutrition (ESPGHAN) North American Society for Pediatric Gastroenterology, Hepatology and Nutrition/NASPGHAN, the diagnosis of H. pylori infection in children is based on invasive gastric biopsy-based methods, including positive culture or histopathology (H. pylori-positive gastritis), with at least 1 other positive biopsy-based test (rapid urease test [RUT] or molecular-based assays) [17]. For the test result to be reliable, one should wait at least 2 weeks after discontinuing a proton pump inhibitor (PPI) and 4 weeks after discontinuing antibiotics before testing for H. pylori. If H. pylori infection is identified in a child with gastric or duodenal ulcers, eradication should be initiated. On the other hand, if H. pylori is an incidental finding during endoscopy, without clinical symptoms, treatment of H. pylori can be considered after careful discussion with the child’s parent/guardian about the risks and benefits of such treatment. Choice of antibiotics for H. pylori therapy should be based on the results of antimicrobial susceptibility testing or national/regional antibiotic resistance data [5,17]. It is well known that primary antibiotic resistance depends on geographic regions and should be taken into account to increase the eradication rate [17,18]. Antimicrobial resistance of H. pylori is mainly caused by several mutations [19]. For example, amoxicillin resistance is associated with mutations that cause reduced binding of the agent to the penicillin binding protein (PBP) 1A, PBP2, or PBP3. The development of resistance to metronidazole is complex and mainly related to reduced activation of the prodrug by mutations in the recA and lexA genes and mutations leading to impaired RecA DNA repair. The main mechanism of clarithromycin resistance in H. pylori is based on A2142C, A2142G, and A2143G mutations in the V domain of 23S rRNA, as well as several minor mutations outside the domain. Bacterial resistance can be associated with the presence of several mutations simultaneously, as well as with impaired efflux pumps activity and biofilm production [19]. In everyday practice, the detection of H. pylori resistance to antimicrobials is mainly based on several methods, such as culture-based techniques, PCR-based assays, in situ probe hybridization-based assay, whole-genome-sequencing-based assays, and protein-based molecular assay [19,20].

There are no data on the susceptibility of H. pylori to antibiotics in children with a first episode of H. pylori infection from northeast Poland, so we decided to conduct such a study using the culture-based technique of biopsy specimens taken during esophagogastroduodenoscopy (EGD).

Material and Methods

ETHICS STATEMENT:

The study was approved by the local Bioethics Committee (R-I-002/599/2018). A closed-ended questionnaire that each patient completed included symptoms, the previous year’s medication history, and family history of H. pylori infection and smoking habits. Parents/guardians of all participants provided written informed consent for participation in the study and for the endoscopic examination.

STUDY DESIGN:

This prospective study included 82 children from northeast Poland, aged 3–18 years old, who had undergone EGD due to GI symptoms between February 2019 and May 2022. Our department is the main pediatric endoscopic center in northeast Poland. Among the reported symptoms that were an indication for EGD were: epigastric pain, nausea, vomiting, anemia of undetermined cause, and the detection of H. pylori antigen in the stool, if performed. Patients who received prior H. pylori eradication treatment were not included in the study.

SAMPLING AND ANALYSIS PROCEDURE:

In patients eligible for EGD, 2 gastric mucosa biopsies for H. pylori culture were taken only if macroscopic changes according to the Kyoto classification (such as nodular gastritis, enlarged folds, and diffuse redness) were visualized [21]. EGD photos of selected patients are shown in Figure 1. The biopsies for culture were placed in cold sterile saline in an Eppendorf tube placed on ice, and were prepared for culture within 1 h. Biopsy was homogenized and plated on Helicobacter Agar modified culture medium from Becton Dickinson (BD, Germany). The bacteria were cultured at 37°C for 3–7 days under microaerophilic conditions. They were then identified based on the colony morphology pattern from the culture (small and translucent colonies with a raised bulge) and positive tests for urease, catalase, and oxidase. In addition, Gram staining of the culture material was performed to confirm the presence of spiral/curved gram-negative bacilli. Susceptibility of cultured strains to amoxicillin (AMX), metronidazole (MTZ), and clarithromycin (CL) was assessed by the E-test method (Liofilchem, Italy). Viable H. pylori colonies were suspended in Mueller-Hinton broth with 5% horse blood +20 mg/l β-NAD (BioMerieux, France) and culture suspensions with a McFarland turbidity of 3 were applied to the inoculation plates. The E-test strip of each antibiotic was placed on a plate and incubated for 3 days at 37°C under microaerophilic conditions. The minimum inhibitory concentration (MIC) was identified by the point of intersection of the inhibition zone with the E-test strip. Strains were considered resistant when the MIC value was >0.12 μg/mL for amoxicillin, >0.5 μg/mL for clarithromycin, and >8 μg/mL for metronidazole. The breakpoints are based on epidemiological cutoff values, which distinguish wild-type isolates from those with reduced susceptibility by using the European Committee on Antimicrobial Susceptibility Testing criteria (EUCAST 2020). To control the quality of drug susceptibility, the reference strain H. pylori ATCC 43504 was used.

STATISTICAL ANALYSIS:

The quantitative data are expressed as median, maximum, and minimum values, while qualitative variables are expressed as absolute frequency and percentage. Differences between groups for frequency of gender, symptoms, and endoscopic features were analyzed by the chi-squared test, and for age by the Mann-Whitney U test. If the group was smaller than 10, then no statistical calculations were made. A P<0.05 value was considered statistically significant.

Results

PATIENTS’ CHARACTERISTICS:

H. pylori strains were isolated from cultured gastric biopsy samples taken during EGD from 50 (61%) patients who had never been treated for H. pylori infection. The data on demographics, symptoms, endoscopic finding, history of medications used during the previous year, family history of H. pylori infection, and smoking habits in this group are shown in Table 1.

In the study group, there were almost equal number of boys and girls, and the median age was 11 years. However, one-third of patients (34%) were younger than 10 years (Table 2). The most frequent symptom was abdominal pain, reported by 76% of patients, followed by nausea (26%) and vomiting (18%). No children with anemia were observed among H. pylori-positive patients. In endoscopic assessment, the nodularity gastritis was the most frequent finding. Positive family history of H. pylori infection was noted in 38% of cases.

Less than half of H. pylori strains isolated from treatment-naïve children (n=24/50; 48.0%) were susceptible to all 3 antibiotics (Figure 2).

ANTIBIOTIC RESISTANCE:

The prevalence of resistance of H. pylori strains isolated from previously untreated children to individual antibacterial drugs is shown in Figure 3. Primary resistance to clarithromycin (CL) was detected in 38% (n=19/50) of the H. pylori strains (MIC range was 0.016–256.00 μg/ml) (Figure 3). Primary metronidazole (MTZ) resistance was found in 30% (n=15/50) of positive cultures (MIC range was 0.016–256.00 μg/ml). Primary resistance to amoxicillin (AMX) was confirmed in 26% (n=13/50) of the H. pylori strains (MIC range was 0.016–256.00 μg/ml). All cultured strains were positive in oxidase, catalase, and urease tests.

The prevalence of resistance to 1, 2, or 3 antibiotics among H. pylori strains isolated from previously untreated children is shown in Figure 4. Resistance to a single antibiotic was found in 14 patients (28%), with the highest rate for CL, followed by MTZ and AMX (14% vs 8% vs 6%, respectively) (Figure 4). Dual drug resistance was observed in 3 children (6%), among whom 2 (4%) had resistance to CL and MTZ, and 1 (2%) to CL and AMX. Triple drug resistance was detected in 9 patients (18%).

COMPARISON OF CHILDREN SUSCEPTIBLE TO ALL 3 ANTIBIOTICS WITH THOSE RESISTANT TO AT LEAST 1 ANTIBIOTIC:

Having realized that the study group was small, we decided to investigate if there were differences in demographics, symptoms, endoscopic finding, history of medications used during the previous year, family history of H. pylori infection, and smoking habits between children susceptible to all 3 antibiotics with those resistant to at least 1 antibiotic (Table 3). There were no significant differences in any of the evaluated parameters.

COMPARISON OF ANTIBIOTIC RESISTANCE ACCORDING TO PATIENT AGE:

We then decided to see if there were differences depending on the age of the patient with a first-diagnosed H. pylori infection in terms of H. pylori resistance to particular antibiotics, personal history of antibiotic use, family history of H. pylori infection, and family smoking habit (Table 2). Considering the age of the patients, we found that the group of children younger than 10 years was more likely to have received antibiotic therapy in the last year. We also observed a higher antibiotic resistance of H. pylori strains from older children than from children younger than 10, but this result was not statistically significant. However, we realize that the number of patients was small, so the observed differences should be interpreted with caution.

COMPARISON OF ANTIBIOTIC RESISTANCE IN THE PRE- AND POST-PANDEMIC PERIOD OF COVID-19:

Given the onset of the COVID-19 pandemic during the course of our study and the associated changes in medical care across the country, we decided to see if this affected the number of diagnosed H. pylori strains resistant to particular antibiotics, personal history of antibiotic use, family history of H. pylori infection, and family smoking habits. It is noteworthy that 46 patients were included in the study in the first year before the pandemic, while during the pandemic, only 36 EGDs were performed for another 2.5 years. Comparing results taken before or during the COVID-19 pandemic, there were no significant differences in prevalence of H. pylori resistance to antibiotics, nor between the other parameters assessed, with the exception of household smoking, which increased during the pandemic (Table 4).

Discussion

Antibiotics have become an important factor of modern healthcare [22]. However, the unnecessary use of antibiotics, especially in common viral infections is connected with development of bacterial resistance [23–26]. The increase in H. pylori resistance may be related to overuse of antibiotics, patient non-compliance with treatment of bacterial infections, virulence of the strain, migration, and overuse of azithromycin during the COVID-19 pandemic [19].

In our study, the H. pylori resistance was most common to clarithromycin (n=19; 38%), followed by metronidazole (n=15; 30%) and the least frequent to amoxicillin (n=13; 26%). Treatment of H. pylori infection in children is based on antibiotic therapy including AMX, MTZ, or CL combined with PPIs [17]. Recent treatment guidelines for H. pylori in children differ from those for adults, which are associated with a lower incidence of complications of H. pylori infection in children. According to Koletzko et al, approximately 10–15% of patients develop serious complications in adult life, such as gastric ulcer disease and gastric cancer [27]. It is therefore advisable to test for H. pylori and treat only when justified, as over-treatment with antibiotics induces drug resistance of this bacterium.

Our study is the first report of antibiotic susceptibility of H. pylori in the treatment-naïve children from northeast Poland. The most unexpected finding was the low percentage (48%) of H. pylori strains sensitive to all 3 antibiotics (CL, MTZ, and AMX). In our group of patients, the highest rate of resistance was observed for CL (38%). Depending on the study region, resistance of H. pylori strains to CL ranged from 12.3% to 46.9% of the population [28,29]. It is worth noting that, compared to most studies conducted in children, we observed a higher rate of primary resistance to CL of H. pylori strains. For instance, in Spain, resistance to CL was 34.7%, in Austria 34%, in Germany 12.3%, in Portugal 23.3%, in northern Italy 26%, in Slovenia 23.4%, and in Turkey 30.1% [18,28–34]. In contrast, CL resistance rates in studies from southern Poland and Germany were higher than in our study, reported as 46.9% and 45%, respectively [29,35]. Increasing resistance of H. pylori to CL in never-treated children with H. pylori infection may be related to the frequent use of this antibiotic in infections, especially of the respiratory tract [5,31,36]. Borka et al suggested that due to the increasing trend of CL-resistant strains in all European countries, this antibiotic should only be used when susceptibility has been confirmed [7].

For another antibiotic, MTZ, the primary resistance of H. pylori strains in our study was 30%. The resistance rate of MTZ varies depending on the population studied, and in European studies it ranged from 14.9% to 59% [29–35], but in studies conducted in China, H. pylori resistance to MTZ was higher and ranged from 75.2% to 81.77% [37,38], perhaps due to the frequent use of MTZ for intestinal, including parasitic, infections, which are more prevalent in Asia than in Europe [39].

We found that among the tested antibiotics, resistance to AMX was the lowest and was found in 26% of the isolated bacteria. However, compared to other observations, this percentage was alarmingly high. In southern Poland, Turkey, and Israel, H. pylori resistance to amoxicillin was not detected [29,34,40]. Amoxicillin is an antibiotic commonly used for infections of the upper and lower respiratory tract or urinary tract, which are the most common infectious diseases in children. Hence, its widespread use may have contributed to the increased resistance of H. pylori to AMX. According to data from the Polish Ministry of Health, the consumption of antibiotics in open medicine amounted to 27 defined daily doses (DDD) per 1000 inhabitants, higher than the average value recorded in the European Union in recent years.

Furthermore, we observed double (for MTZ and CL as well as AMX and CL) and triple resistance in 6% and 18% of H. pylori strains, respectively. Based on available studies, dual resistance of H. pylori to CL and MTZ, as well as CL and levofloxacin, have been observed in Spanish children, but there are no data on its frequency [41]. Similar observations have been reported in China, which may be related to the clinical effectiveness of current eradication therapy [38]. Based on available studies from European pediatric centers, no triple antibiotic resistance of H. pylori has been reported to date, while data from China show a lower rate of this phenomenon than in our study [42]. Such a high rate of multidrug-resistant H. pylori strains noted in our region is alarming and should prompt careful choice of antibiotics in therapy. It is possible that these strains were acquired from infected adult family members, as the route of H. pylori transmission is from person to person [9]. More research is needed to compare H. pylori resistance to antibiotics among members of the same household.

We did not find significant demographic or clinical differences between the group of patients infected by resistant strains and the group infected by susceptible strains. Similar results were noted by Ogata et al [43].

The strength of our study is that we performed the first assessment of H. pylori susceptibility to antibiotics in infected but treatment-naïve children from our region. Our department is the main pediatric endoscopic center in northeast Poland, so most children with H. pylori infection were diagnosed in our department. However, we could not rule out that some of patients from our region were diagnosed in other parts of Poland. Moreover, this was a prospective study. EGDs were performed by the same endoscopists, and the microbiological assessment was performed by the same microbiologists.

The limitation of the study is the small number of subjects. This could be partially explained by the COVID-19 pandemic, which resulted in a reduction in the number of EGDs performed. In addition, we could not compare our results with the adult population from our region because there are not such data. Moreover, conventional methods for antibiotic susceptibility testing include culture, which is considered the reference method for diagnosing infection. However, it has a high risk of failure due to the particular growth requirements of H. pylori, which needs a microaerophilic atmosphere and prolonged incubation for growth, leading to time-consuming limitations for susceptibility testing. Therefore, molecular methods to assess antibiotic resistance of H. pylori are easier to perform and do not take as much time, but unfortunately these methods were not available at our center.

Conclusions

In this study, carried out for the first time in northeast Poland to assess the antibiotic susceptibility of H. pylori, a high percentage of strains resistant to at least 1 antibiotic was found. The results obtained may help in the rational selection of antibiotics for the eradication of H. pylori in children from our region, but also indicate the growing problem of resistance of this bacterium to the drugs used. Further research into H. pylori drug resistance among family members may contribute to the development of more effective treatment regimens.