The antibiotic resistance of Helicobacter pylori to five antibiotics and influencing factors in an area of China with a high risk of gastric cancer

Background H. pylori exhibits antibiotic resistance with regional differences. In this paper, we explored antibiotic resistance of H. pylori to five antibiotics in an area with a high risk of gastric cancer. Results H. pylori resistance rates to metronidazole, levofloxacin, clarithromycin, amoxicillin, and tetracycline were 78.0, 56.0, 31.0, 9.0, and 15.0%, respectively. Double, triple, quadruple, and quintuple resistance rates were 23, 20, 6, and 4%, respectively. The clarithromycin and multidrug resistance rates were significantly higher in males than females (clarithromycin: 44.4% vs 15.2%, respectively, P = 0.002; multidrug: 75.5% vs 37.2%, respectively; P < 0.001). During the three periods of 1998–1999, 2002–2004 and 2016–2017, the resistance rates to levofloxacin and amoxicillin were increasing (OR: 2.089, 95%CI: 1.142–3.821, P = 0.017; and OR: 5.035, 95%CI: 1.327–19.105, P = 0.018, respectively). The antibiotic resistance rates were unassociated with the host disease state. Metronidazole resistance was lower in the vacAs1m1/m2 group than the vacAs1m1m2 group (65% vs 85.7%, respectively; P = 0.026). As for levofloxacin resistance, it was higher with cagA + than cagA− (60.9% vs 23.1%, respectively; P = 0.020) but lower with slyD+ than slyD− (41.4% vs 68.5%, respectively; P = 0.009). Clarithromycin had a lower resistance rate with iceA++ than iceA−+ (19.7% vs 52.4%, respectively; P = 0.017). For amoxicillin, the iceA++ group had a lower resistance rate than the iceA−− group (1.6% vs 27.8%, respectively; P = 0.009). Conclusions The total resistance rates of H. pylori to metronidazole, levofloxacin, clarithromycin, amoxicillin, and tetracycline were high in Zhuanghe. The resistanc rates to levofloxacin and amoxicillin increased over time. Clarithromycin resistance was associated with male and iceA. The resistance of metronidazole was related to vacA. Levofloxacin resistance was concerned with cagA and slyD and amoxicillin resistance was concerned with iceA. While, the antibiotic resistance of H. pylori had nothing to do with the status of gastric disease.


Background
A number of large randomized controlled trials have shown that the eradication of Helicobacter pylori (H.pylori) can reduce the occurrence of chronic atrophic gastritis and intestinal metaplasia [1] and decrease the risk of gastric cancer [2]. Therefore, the successful eradication of H. pylori is of great significance for the prevention and treatment of gastric diseases. The antibiotics that are commonly used for H. pylori eradication include clarithromycin, metronidazole, amoxicillin, and levofloxacin. In the past few decades, the resistance rate has continued to increase.
The antibiotic resistance of H. pylori exhibits regional differences. In Germany [3] and Brazil [4], the antibiotic resistance rate to clarithromycin was 28.7 and 23.2%, respectively, and it was only 7.3% in Ireland [5] and 0% in Malaysia [6]. The resistance rate to levofloxacin was 32.7% in Buenos Aires [7] and 18.2% in the Bogota-Colombia [8], as well as 5.8% in the Hebei Province of China [9]. As reported in the literature, the resistance rate to amoxicillin and tetracycline is generally low [7,10,11], but in Ireland and Vietnam, it was as high as 38.1% [5] and 23.8% [12], respectively.
There are many factors influencing the antibiotic resistance of H. pylori, including age and gender, as well as the host disease status and virulence factors carried by the pathogens. However, there is no consistent report on whether these factors are related to the antibiotic resistance of H. pylori. It is well known that the virulence factors secreted by H. pylori are closely related to different clinical outcomes in gastric diseases. Cytotoxinassociated gene A (cagA) is the most common virulence factor of H. pylori. Some studies have shown that the clarithromycin resistance rate is higher in patients with cagA + than those with cagA − [13], but others have demonstrated that the eradication rate in patients with cagA + was significantly higher than those with cagA − [14]. These studies showed that the influencing factors of H. pylori antibiotic resistance are different. Therefore, it is necessary to carry out further investigations in different areas.
Zhuanghe, Liaoning Province, a high-risk area for gastric cancer, is in northern China. The H. pylori infection rate for the local residents was more than 60%, and the risk of gastric cancer and precancerous diseases was increased for H. pylori-infected individuals [15]. In this study, we investigated the antibiotic resistance of H. pylori to metronidazole, levofloxacin, clarithromycin, amoxicillin, and tetracycline, aimed to provide guidance for the selection of medication, to improve the eradication effect, and to reduce the occurrence of gastric disease.

Patients
A total of 100 strains were collected from the primary cultures of 454 gastric mucosal tissues obtained during an endoscopic biopsy that were all from the Zhuanghe Gastric Diseases Screening Program, started in 1997 [15]. The gastric mucosal tissues were cultured for H.pylori during the 3 periods of 1998-1999, 2002-2004, 2016-2017 were 102, 242 and 110 cases. The number of primary cultured strains was 23, 50, and 27, respectively, so the total number of specimens in the 3 years was 100 strains. The patients were 54 males and 46 females, ranging from 24 to 87 years old (mean age: 50.74 ± 10.942 years). Pathological diagnosis was based on the updated Sydney Classification System, including 21 cases of superficial gastritis (GS), 35 cases of atrophic gastritis (GA), 32 cases of gastric erosion ulcer (GEU), 7 cases of atypical dysplasia (GD), and 5 cases of gastric cancer (GC). Patients were excluded if they had stomach surgery and if they were administered eradication treatment. The study was approved by the Ethics Committee of Chinese Medical University and all participants signed an informed consent.

Antibiotic susceptibility testing
The susceptibility of H. pylori to metronidazole, levofloxacin, clarithromycin, amoxicillin, and tetracycline was tested using an Epsilometer test (E-test, bioMerieux, France). The methods of H.pylori culture and antibiotic susceptibility testing are all the same in different periods. The H. pylori that was in a logarithmic growth period was adjusted to a McFarland standard of 2.0 and smeared evenly with a sterile cotton swab (bioMerieux, France) on Mueller-Hinton agar (Oxoid, Basingstoke, UK), supplemented with 7% defibrinated sheep blood. The concentration range of the E-test strips was 0.016-256 μg/ml for metronidazole, amoxicillin, and tetracycline (bioMerieux, France); 0.016-256 μg/ml for clarithromycin; and 0.016-32 μg/ml for levofloxacin (Kangtai, China). Then, the plates were incubated at 37°C for 48-72 h in a microaerophilic atmosphere with 10% O 2 , 5% CO 2 , and 85% N 2 . Thereafter, the minimum inhibitory concentration (MIC) for each antibiotic was determined after 48-72 h by two people independently. The MIC breakpoints were identified using the recommended Clinical and Laboratory Standards Institute (http://www.clsi.org/) of America. Strains were considered resistant with an MIC of ≥1 mg/ ml for clarithromycin [16], levofloxacin [17], amoxicillin and tetracycline [18] and ≥ 8 mg/ml for metronidazole [4]. The H. pylori reference strain, ATCC 26695, was used for quality control.

DNA extraction
Whole genomic DNA of H. pylori was extracted using a phenol-chloroform method [15]. In brief, 300 μl of TE buffer (10 Mm Tris-HCl [pH 8.3], 0.1 mmol/l EDTA), 100 μl of 10% SDS, and 10 μl of PK enzyme (20 mg/ml) were added to the bacterial precipitate and digested overnight in a water bath at 55°C. The extracted DNA was dissolved in 50-100 μl of TE buffer and stored at − 20°C.

Statistical analysis
All statistical analysis was carried out using SPSS V.18.0 software for Windows (SPSS Inc., Chicago, IL, USA). A binary logistic regression model was used to calculate the odds ratios (OR) and 95% confidence interval (CI). A P value of < 0.05 was defined as significant.
In addition, among the 100 strains of H. pylori, 39 (39%) had single-drug resistance, 23 (23%) had double-drug resistance, and 20 (20%) had triple-drug resistance; meanwhile, 6 (6%) were resistant to four drugs, and 4 (4%) were resistant to all five antibiotics (Table 2). Only 8 (8%) strains were found to be sensitive to all five antibiotics. As for the multidrugresistant strains, most were double or triple resistant. The highest proportion of a double-drug resistant pattern was metronidazole plus levofloxacin, accounting for 73.9%. The highest proportion of a triple-drug resistance pattern was the combination of metronidazole, levofloxacin, and clarithromycin, accounting for 55%. Similarly, after stratifying by gender and age, the results showed that the multidrug resistance rate was higher for males than females (75.5% vs 37.2%, respectively; OR: 5.203, 95%CI: 2.120-12.771, P < 0.001, Table 1). However, there was no significant difference between the single and multidrug resistance rates based on age.

The antibiotic resistance of H. pylori and host disease state
Concerning the relationship between H. pylori antibiotic resistance and host disease status, we found that after adjusting for age and gender, there was no significant difference in the antibiotic resistance for the GS, GA, and GEU groups or the GEU and non-gastric erosion ulcer (NGEU) groups ( Table 1).

Discussions
Antibiotic resistance is the main reason for failure to eradicate an H. pylori infection. In this study, 100 H. pylori strains from Zhuanghe, an area with a high incidence of GC, were included to investigate the antibiotic resistance, to explore the trend over time, and to analyze the factors affecting antibiotic resistance. This study revealed the antibiotic resistance of H. pylori to five antibiotics in an area with a high incidence of GC and provided guidance for the treatment of H. pylori.
Our study demonstrated that the total resistance rates of H. pylori to metronidazole, levofloxacin, and clarithromycin were 78, 56, and 31%, respectively, with high MIC levels; however, the resistance rates to amoxicillin and tetracycline were low, and the multidrug resistance rate was as high as 53%. As compared with low-risk areas for GC in China [27][28][29], neighboring countries [30][31][32][33], and other non-neighboring countries [6,11,34,35], the antibiotic resistance was generally higher in the Zhuanghe area. The high antibiotic resistance increased the risk of H. pylori transmission among the population and caused a widespread and persistent high infection rate of H. pylori in the population; moreover, it also could increase the continued colonization of the stomach in hosts and increase the inflammation of and damage to the gastric mucosa, which may lead to an increased risk of GC.
According to previous studies, H. pylori antibiotic resistance has significantly changed over time [8,33,[36][37][38]. For example, in Korea, the resistance rate to clarithromycin increased from 11 to 60% from 2009 to 2014 [39,40]. In Singapore, the resistance rate to metronidazole increased from 24.8 to 48.2% from 2000 to 2014 [38]. Our results showed that the resistance rates to clarithromycin and metronidazole maintained a stable yet high level and that tetracycline was at a steady yet low level in 1998-2017. The resistance rates to levofloxacin and amoxicillin, as well as multidrug resistance, have demonstrated an increasing trend. Our study showed that the two antibiotics, clarithromycin and metronidazole, were not suitable for eradication therapy in this area. However, tetracycline can be used to eradicate H. pylori but has a shortcoming because of adverse reactions, so the time and dose should be strictly controlled. Developing alternative drugs to help with eradication is also a future goal.
In this study, we found that the resistance rate to clarithromycin for males was significantly higher than   females. Boyanova [41] also had similar conclusions in females and males. In addition, we found males are more prone to multidrug resistance than females. The possible causes of this phenomenon was that macrolide antibiotics were commonly used for the treatment of respiratory diseases and respiratory infections related to smoking and other factors mostly occurred in men that might lead to antibiotic resistance for males with a clarithromycin-based eradication treatment for H. pylori. Previous studies of our group showed that the incidence and mortality of males with GC in the Zhuanghe area were higher than those of females by 2.8 times and 2.3 times, respectively [42]. Thus, the high antibiotic resistance of H. pylori for males might also be one risk factor for GC in addition to environmental exposure factors such as diet, smoking, drinking, and occupation.
In addition, there have been different reports on whether the gastric disease status is related to the antibiotic resistance of H. pylori. A study involving 2751 patients in France showed that the successful eradication rate of H. pylori in patients with a duodenal ulcer was higher than that of patients with a non-ulcer-related dyspepsia [43]. However, some medical literature indicated that these diseases are not related to antibiotic resistance [30,41,[44][45][46][47]. In the present study, we found that the H. pylori resistance rate was not associated with gastric disease. However, while analyzing the reason, on the one hand, it may be geographical differences. On the other hand, it may be because of the small number of cases included in this study. Some virulence factors carried by H. pylori may also affect the clinical outcomes of eradication [48][49][50]. As compared with single-and multi-strain infections, we found that the resistance rate of metronidazole to vacAm1 or vacAm2 was lower than that of a mixed infection with vacAm1m2, to iceA1 or iceA2 alone higher than that of mixed infection with iceA1A2. The resistance rate to amoxicillin in iceA −− was higher than that of a mixed infection. Additionally, the resistance rate to levofloxacin was higher in the cagA + group than the cagA − one. The resistance rate to levofloxacin was higher in the slyD − group than the slyD + one. CagA + or mixed vacAs1m1m2 infection was associated with antibiotic resistance. Patients infected with such strains had more difficulty achieving eradication of H. pylori, and continuous colonization was more likely to induce precancerous lesions and GC. Therefore, decreasing the incidence of GC could be achieved so long as the emergence of antibiotic resistance is avoided during the administration of effective eradication treatment. In addition, a mixed iceA or slyD + strain would be easy to eradicate according to the present study results. However, in some reports both iceA1 and iceA2 infections were unrelated to the eradication outcomes [51,52]. SlyD was a new virulence factor associated with gastric disease [53]. To our knowledge, until now, there was no report on the relationship between slyD and antibiotic resistance. Therefore, the relationship between these two virulence factors and bacterial resistance needs to be further explored.
This study had the following limitations: 1. Since there exists dominant genotype of H.pylori virulence factors eg. cagA positive, vacA s1 m1, and iceA1 positive in this area [15], therefore, the relationship between virulence factors and antibiotic resistance needs to be further investigated. 2. The obtained conclusions in the present study were based in a number of only 100 strains, the factors affecting the antibiotic resistance still needs to further verification in a large samples. 3. Other factors such as host CYP2C19, MDR, and IL-1β gene polymorphism can also influence H. pylori eradication. However, there is a lack of relevant information in our study, so we could not evaluate it.

Conclusions
In conclusion, our study evaluated the antibiotic resistance and influencing factors of H. pylori in Zhuanghe. The total resistance rates to metronidazole, levofloxacin, clarithromycin, amoxicillin, and tetracycline were high. The resistance rates to levofloxacin and amoxicillin increased over time. Clarithromycin resistance was Resistance to all ABs 4 Abbreviations: AB antibiotic associated with male and iceA. The resistance of metronidazole was related to vacA. Levofloxacin resistance was concerned with cagA and slyD and amoxicillin resistance was concerned with iceA. While, the antibiotic resistance of H. pylori had nothing to do with the status of gastric disease. Per this study, clarithromycin-based triple therapy is no longer applicable in this area; metronidazole and levofloxacin should be used with high vigilance; and amoxicillin and tetracycline can be used as candidates for antibiotic treatment. Virulence factors including cagA, vacA, iceA, and slyD can be used to predict antibiotic resistance. Our study is beneficial to the selection of the H. pylori eradication program that is conducive to reducing the infection rate, thus helping in the prevention and treatment of GC and related gastric diseases.