Prevalence of Helicobacter pylori and Its Associated Factors among Healthy Asymptomatic Residents in the United Arab Emirates

The United Arab Emirates (UAE) has been under continuous populational influences from Asia, Europe, and Africa, making it an ideal site for epidemiological studies on Helicobacter pylori. However, there has been a paucity of well-designed prevalence studies on H. pylori from UAE. The aim of this study was to determine the prevalence of H. pylori and its associated risk factors in the UAE. A prospective cross-sectional study was conducted on healthy asymptomatic residents of UAE. Socio-demographic, lifestyle, and gastrointestinal characteristics of participants were obtained through a questionnaire in parallel within the stool sample collection. A total of 350 participants were included in this study and were tested for H. pylori using the stool antigen test (Premier Platinum HpSAT). Out of the total tested study participants, 41% were found to be H. pylori-infected. Logistic regression analysis has shown a significant association between H. pylori infection and gender, age, ethnicity, profession, domestic overcrowding, source of drinking water, and gastrointestinal characteristics of participants. Based on the results from this study, we suggest that preventive measures against H. pylori infection should be considered worthy by public health authorities.


Introduction
Helicobacter pylori (H. pylori) is a gram-negative microaerophilic bacterium that colonizes the gastric mucosa of more than half of the world's population with high geographic variability [1]. H. pylori infection is generally acquired during childhood and persists life-long in the absence of treatment with antibiotics. Most of the infected individuals remain asymptomatic for a long period. As a result, long-term colonization of H. pylori can damage the gastric mucosa causing various diseases of the upper gastrointestinal tract such as chronic gastritis, peptic ulcer, and gastric malignancies, particularly gastric cancer and gastric mucosa-associated lymphoid tissue (MALT) lymphoma [2,3]. H. pylori has been recognized as a Class I carcinogen by the International Agency for Research on The ethnicities of the participants were distributed as follows: Arab (163, 46.6%), Asian (165, 47.1%), and African (22,6.3%). In term of household income, most of the participants had less than AED 10000 (2700 USD) monthly family income (133, 45.4%). The occupation of participants was distributed as following: Employed as a professional (37,10.6%), employed as an industrial worker (97, 27.7%), employed as a nanny (56, 16%), and student (145, 41.4%). Only 4.4% of the participants were unemployed. All children participants were considered as students since stool samples from this age group were collected from schools and nurseries. More than half of the participants lived with 4 to 6 persons per house (207, 59.1%) compared to participants living with more than 7 persons per house (112, 32%). In term of a number of persons per room, more than half of the participants reported living in close proximity with others, having three persons or above per room (194, 55.4%). In terms of having contact with domestic animals and type of drinking water, most of the participants confirmed no contact with domestic animals (305, 87.1%) and that mineral bottled water was the source of their drinking water (283, 80.9%) ( Table 2).

Gastrointestinal Characteristics
When participants were asked about having experienced gastrointestinal disturbances, the majority have reported no previous or current gastrointestinal disturbances (288, 82.3%). Among the few participants having experienced gastrointestinal disturbances, abdominal pain was the major symptom (29,8.3%), followed by bloating (20, 5.7%). In terms of parent gastric diseases history, the majority did not report any history case of familial gastric diseases (293, 83.6%). Among the few participants having a familial history of gastric diseases, peptic ulcer was reported in the first place (31,8.9%) compared to other gastric diseases including gastric cancer and MALT lymphoma ( Table 2). None of the participants have reported any past, recent, or current intake of Proton Pump Inhibitors (PPI) for H. pylori eradication treatment (data not shown in the table).

Prevalence of H. pylori Infection in Asymptomatic Healthy UAE Residents
Analysis of the collected stool samples of the 350 participants has shown that 145 participants were H. pylori-infected and therefore the prevalence of H. pylori infection was considered 41% including both children and adults. Bivariate analysis has shown that H. pylori infection exhibited significant differences across several sociodemographic, lifestyle, and gastrointestinal characteristics ( Table 2). Participants from Ajman showed the highest prevalence (35,64.8%) to H. pylori compared to Abu Dhabi, Sharjah, and Dubai. Female participants were more prone to H. pylori infection (69, 53%) compared to male participants (76, 35%) with a statistically significant difference (p = 0.001). In term of age groups, a statistically significant difference was found between H. pylori infection and the age group of participants. The prevalence of H. pylori was found to increase significantly with the age of participants (p = 0.032). Adults participants aged between 16 and 30 years old showed the highest prevalence for H. pylori (37, 52.9%), followed by participants 30 years old and older (63, 46%). Since people of several ethnicities reside in UAE, it was of great interest to investigate which ethnicity is more prone to H. pylori infection. A significant statistical difference was found between H. pylori infection and the ethnicity of the participants (p < 0.0001) ( Table 2). African residents in UAE presented the highest prevalence to H. pylori (18,81.8%) compared to Asian (77, 46.7%) and Arab participants (50, 30.7%) ( Figure 1). Among the Asian group, participants from India were found more infected by H. pylori (39%) followed by participants from Pakistan (23%) and the Philippines (21%). Among African participants, the prevalence of Ethiopian participants presented the highest positivity to H. pylori (71%). In Arab participants, Emiratis were the most prone to H. pylori infection (55%) compared to other Arab nationalities. All other Arab participants presented a similar prevalence to H. pylori. Taking together all H. pylori-infected participants, the prevalence was the highest among the Ethiopian nationality (71%) compared to the other nationalities (data not shown).
In term of the number of persons per house, no statistically significant difference was found between H. pylori infection and these factors (data not shown). However, the number of persons per room was found to play an important role in H. pylori infection with a statistically significant difference (p = 0.0001). Half of the participants living in crowded conditions (≥3 members per room) were infected by H. pylori (97, 50%) ( Table 2).
In term of the drinking water source, even though most of the participants confirmed drinking mineral bottled water (283, 80.9%), a statistically significant difference was found between H. pylori and tap water as a source of drinking water (p = 0.001). More than half of participants drinking tap water were found positive to H. pylori (34,58.6%). In term of gastrointestinal characteristics, even though 82.3% of the involved participants were healthy and asymptomatic without any gastrointestinal disturbances, few participants have mentioned suffering from gastrointestinal disturbances. Interestingly, a statistically significant difference was found between the H. pylori infection and the gastrointestinal disturbances (p = 0.001). Participants suffering from abdominal pain were found highly prevalent to H. pylori (22, 75.9%) followed by bloating (8, 40%). When family history of gastrointestinal diseases was investigated, it was found that participants having one of their parents with previous or current peptic ulcer diseases were highly prevalent to H. pylori (28,90.3%) with a statistically significant difference (p < 0.0001). There was no statistically significant difference found between nutritional resources and H. pylori infection ( Figure 3). difference (p =0.0001). Half of the participants living in crowded conditions (≥ 3 members per room) were infected by H. pylori (97, 50%) ( Table 2).
In term of the drinking water source, even though most of the participants confirmed drinking mineral bottled water (283, 80.9%), a statistically significant difference was found between H. pylori and tap water as a source of drinking water (p = 0.001). More than half of participants drinking tap water were found positive to H. pylori (34,58.6%). In term of gastrointestinal characteristics, even though 82.3 % of the involved participants were healthy and asymptomatic without any gastrointestinal disturbances, few participants have mentioned suffering from gastrointestinal disturbances. Interestingly, a statistically significant difference was found between the H. pylori infection and the gastrointestinal disturbances (p = 0.001). Participants suffering from abdominal pain were found highly prevalent to H. pylori (22,75.9%) followed by bloating (8, 40%). When family history of gastrointestinal diseases was investigated, it was found that participants having one of their parents with previous or current peptic ulcer diseases were highly prevalent to H. pylori (28,90.3%) with a statistically significant difference (p < 0.0001). There was no statistically significant difference found between nutritional resources and H. pylori infection (Figure 3).

Association Between H. pylori Infection and Socio-Demographic, Lifestyle and Gastrointestinal Characteristics
Binary logistic regression analysis revealed that female participants had a significantly higher likelihood of being diagnosed as H. pylori-infected compared to a male participants, Odds ratio (OR) 2.52, 95% CI, 1.16-5.50, p = 0.020. Furthermore, Asian and Africans were more likely to have H. pylori infection, OR 7.98, 95% CI, 2.32-27.48, p = 0.001 and OR 29.75, 95% CI, 4.41-200.94, p < 0.0001, respectively, compared to Arabs. In addition, participants who shared a bedroom with 3 or more persons were more likely to have H. pylori infection, OR 3.76, 95% CI, 1.41-10, p = 0.008 compared to those who shared a bedroom with one or two persons (Table 3).

Association Between H. pylori Infection and Socio-Demographic, Lifestyle and Gastrointestinal Characteristics
Binary logistic regression analysis revealed that female participants had a significantly higher likelihood of being diagnosed as H. pylori-infected compared to a male participants, Odds ratio (OR) 2.52, 95% CI, 1.16-5.50, p = 0.020. Furthermore, Asian and Africans were more likely to have H. pylori infection, OR 7.98, 95% CI, 2.32-27.48, p = 0.001 and OR 29.75, 95% CI, 4.41-200.94, p < 0.0001, respectively, compared to Arabs. In addition, participants who shared a bedroom with 3 or more persons were more likely to have H. pylori infection, OR 3.76, 95% CI, 1.41-10, p = 0.008 compared to those who shared a bedroom with one or two persons (Table 3).
In term of gastrointestinal characteristics, it was found that participants with H. pylori infection were more likely to have abdominal pain (OR 6.87, 95% CI, 2.39-19.79, p = 0.001) compared with those of reference level with no gastrointestinal problems. Moreover, it was found that participants with a family history of peptic ulcer were more likely to develop H. pylori infection (OR 22.07, 95% CI, 6.01-81.11, p < 0.0001) compared to those of reference level with no family history of gastrointestinal problems (Table 3).

Discussion
The current study has assessed for the first time the prevalence of H. pylori in asymptomatic healthy children and adult residents of UAE and the risk factors associated with the infection. Previous studies in the UAE have shown variable prevalence results depending on the purpose of the study. For example, the prevalence of H. pylori was found 90% in dyspeptic patients [35,36]. However, in another study evaluating the rate of complications of sleeve gastrectomies, it was found that 44% of symptomatic and asymptomatic patients were H. pylori-infected [37]. In another study performed to find an association between diabetes mellitus 2 and H. pylori infection, a prevalence of 76.7% was found in diabetic subjects compared to 64.8% in non-diabetic subjects [38]. Only two studies conducted in 2002 and 2006 by Bener et al. have investigated the prevalence of H. pylori in asymptomatic UAE patients [33,34]. The first prospective study was conducted on 151 subjects (76 farmers and 75 non-farmers). The prevalence of H. pylori was 74.2% when using the IgG H. pylori antibody and 51% when using the IgA H. pylori antibody [33]. Overall, there was no difference in H. pylori prevalence between the UAE asymptomatic farmers and non-farmers. In another study performed on asymptomatic subjects from low socioeconomic background, the H. pylori prevalence was found 78.4% in industrial workers and 64.3% in referent workers with a statistical significance between the exposed industrial and non-exposed control groups [34]. In summary, the few studies conducted on H. pylori in UAE in the past 26 years presented several limitations and were, therefore, nonrepresentative of the general population in the UAE. The studies were either restricted to dyspeptic subjects or to a special population such as farmers and non-farmers or low socio-economic workers in UAE. None of the performed studies have taken into consideration the asymptomatic and general multinational population residents in UAE including children and adults. Moreover, the serological test used to detect H. pylori presented discordant values between the used antibodies IgG and IgA and do not necessarily reveal an active ongoing infection [33]. Therefore, the current prevalence findings are incomparable with the previous studies [33,34].
In our study, the prevalence of H. pylori was significantly high. Almost half of the study participants were found to be infected by H. pylori. However, the prevalence might be considered low compared to other Arab neighboring countries. In Jordan and Iraq, the prevalence was found to be 77.5% and 78% respectively [39,40]. In Kuwait and Egypt, H. pylori was present in 84% and 86% of individuals, respectively [41,42]. In North African countries, the estimated prevalence in Libya, Morocco, and Tunisia were 76%, 75.5%, and 64%, respectively [43][44][45]. In other countries of the Persian Gulf region, the prevalence of H. pylori was found 79% in Bahrain [46]. Our estimate of H. pylori infection is comparable to the prevalence of the neighboring country Oman (49.1%) but lower than Saudi Arabia (65.9%) [1]. Compared to other countries in the MENA region including Egypt, Libya, Saudi Arabia, Turkey, the prevalence ranged from 50% to 94% in both the children and adult population [1]. In Asia, India and Bangladesh have shown the highest prevalence of H. pylori, especially in adults (90% and 88%, respectively). In Africa, Ethiopia has shown the highest prevalence of H. pylori. More than 95% of Ethiopian adults were found positive for H. pylori [32]. These previous findings may explain why Indian and Ethiopian residents in UAE were more prone to H. pylori infection than other nationalities. Interestingly, it was found that nannies presented a higher prevalence of H. pylori infection (64%) compared to other professions. Since most of the recruited nannies in this study arrived very recently to UAE, it was hypothesized that most of the infected nannies harbored the H. pylori infection from their country of origin. Further studies are needed to explore these findings. Among the recruited nannies, Ethiopian nannies presented the highest prevalence of H. pylori (50%) compared to Filipino (44%) and Indonesian nannies (6%). In parallel, the prevalence of H. pylori in Emiratis was the highest among the Arab Asian participants. Therefore, prevention of gastric cancer and gastro-duodenal diseases should target these highly prevalent parts of the population of UAE.
In terms of gender, our findings indicated a statistical difference in the prevalence of H. pylori (p = 0.001) with women more infected than men (53% vs. 35%). These findings are in disagreement with a recent worldwide H. pylori prevalence study which showed no differences in H. pylori infection between male and female subjects [47]. The issue of gender disparity in H. pylori infection is an intriguing topic and further research is needed to understand the mechanisms by which sex may influence the acquisition and persistence of infection. In the current study, female participants were found more likely to be effected by H. pylori infection, possibly because the majority were working as nannies and belonged to low socioeconomic status families. Moreover, they came from highly endemic countries, like Ethiopia. In term of age, our findings showed that the prevalence of H. pylori infection increased with age significantly (p = 0.032) and the obtained results were in agreement with other performed studies [48,49]. In term of living conditions, participants living in crowded conditions (≥3 members per room) were found to be more prone to the infection by promoting the intrafamilial transmission of H. pylori. These findings were in agreement with other study [17,50]. In term of gastrointestinal characteristics, it was found that participants having abdominal pain and family history of peptic ulcer were more prone to H. pylori infection. Usually, most peptic ulcer cases are caused by H. pylori and since H. pylori transmission is intrafamilial, this can explain the high prevalence of H. pylori among these participants.
Apart from the findings of this study, selecting the most suitable test for H. pylori detection in this epidemiological study was one of our study challenges. All the diagnostic tests for H. pylori tests have different purposes and limitations [20,51]. The choice of Premier Platinum HpSA test in our study depended to a large extent on the noninvasiveness requirement. One of the major advantages of the HpSA test is that it reveals active ongoing H. pylori infection and does not require a phlebotomist compared to other serological tests. Serological tests have been used widely in epidemiological studies. However, not all the currently available serological tests have the same accuracy [20,21]. The invasive gold standard method including the endoscopic examination followed by H. pylori culture from the gastric biopsy remains the excellent and more reliable test used for H. pylori detection. Unfortunately, this method requires medical facilities, equipment on-site, and culture conditions which have rendered the application of this method in epidemiological studies expensive and difficult to perform, especially on children. Moreover, for ethical consideration, it is not possible to use an invasive test for an epidemiological study when noninvasive test exists. Detection of H. pylori by molecular techniques such as PCR give the additional possibility to test the antimicrobial susceptibility of H. pylori and give important indications for future treatment. However, the PCR method relies in the first place on the availability of the gastric biopsy. Taking together all the limitations of the above-listed tests, it was found that Premier Platinum HpSA was the most appropriate epidemiological tool with high sensitivity and specificity, especially in children [27,28,51]. Several studies using stool H. pylori antigen tests have shown that the sensitivity and specificity of stool antigen testing are comparable to histology, culture, or urea breath test [24,27]. However, the main limitations of the HpSA are the invalidity of the test on watery diarrheal stools and on patients who have taken antimicrobial medications, proton pump inhibitors, and bismuth preparations. It was demonstrated that previous exposure to antibiotic therapy for H. pylori may result in the conversion of bacillary form to a coccoid form of H. pylori which is a morphological manifestation of bacterial cell death without an infective capacity [24], thus HpSA might detect antigens from the degradation of the two morphological forms of H. pylori [52]. To overcome such limitations, all watery collected stool samples and volunteer participants under recent antimicrobial medications, bismuth, and potent proton pump inhibitors were excluded from the study.
The study has provided a comprehensive overview of the prevalence of H. pylori in Sharjah, Ajman, Abu Dhabi, and Dubai taking into consideration the multinational UAE residents and the different factors associated with this infection. These data can be used to support national initiatives to prevent and eradicate H. pylori, with the goal of reducing the complications of the infection. However, the authors are aware that the current study provided only an overview of the prevalence of H. pylori in the UAE. It is considered a pilot study and the obtained results certainly cannot be generalized toward the whole UAE population due to its limitations such as convenience sampling methods, and low sample size and nature of cross-sectional studies which cannot establish causality. More efforts are in process to conduct further large-scale and multicenter epidemiological studies among the seven emirates in UAE.

Ethical Statement
The study was reviewed and approved by the Research and Ethics Committee of the University of Sharjah [REC-17-04-17-01] (see supplementary "Ethical Approval"). All demographic data related to the volunteer participants were recorded and handled with privacy. All volunteer participants signed their consent to participate in the current study.

Study Design
The cross-sectional study was carried out between September 2017 and April 2018. Four emirates of UAE were included in this prevalence study. A total of 350 stool samples were consecutively collected from healthy asymptomatic children and adult residents in the UAE. Prior to sample collection, the principal investigator made visits to schools, nurseries, housemaid offices, and industrial companies to explain the purpose of the study and to seek consent. To avoid sampling from patients with gastrointestinal disturbances, hospitals and medical centers were excluded. Almost half (48%) of the participants in the study were from Sharjah followed by 33.7% from Abu Dhabi, 15.4% from Ajman and 3% from Dubai. A part-time research assistant was recruited to collect the stool samples and to assist participants in filling out the questionnaire. Resistance to provide a stool sample was one of the challenges encountered. For this reason, the anonymous contribution system was applied during the stool sample collection. To achieve this, two empty boxes were distributed in all collection sites. One box contained the empty stool container and the questionnaire, and the second box was used as a drop box for the stool sample. The part-time research assistant was in charge to check these boxes daily and to aseptically collect and freeze the stool samples at −20 • C. After participants signed the consent form, an enveloped questionnaire was distributed to them along with the stool container. For participating children, parents were requested to sign the consent form and to fill in the questionnaire. The questionnaire and consent forms were written in both English and Arabic. Industrial workers and Ethiopian nannies were assisted by research assistants speaking the same language. For stool sample collection from nurseries and schools, helpers were asked to cooperate with the assigned research assistant. Participants with equivocal questionnaire were taken out of the study. Inclusion criteria were UAE residents aged 1 year and above. Participants subjected to antimicrobials, proton pump inhibitors, and bismuth preparations within the last two weeks from the study period were excluded. The questionnaire consisted of three main sections. The first section consisted of demographic and socioeconomic data which included gender, age, nationality, profession, marital status, education level, family income, number of siblings, number of family members per room, and number of rooms per house. The second section consisted of the lifestyle characteristics of participants. The third section included the gastrointestinal disturbances and family history of H. pylori infection. Participants from several nationalities were involved in the study. They were classified into Arab Asians, Non-Arab Asians, and Africans. Arab Asians were considered from the following countries: UAE, Jordan, Iraq, Syria, Lebanon, and Palestine. Non-Arab Asians were considered from India, Pakistan, Afghanistan, the Philippines and Indonesia. Africans were considered from Sudan, Egypt, and Ethiopia. Body mass index (BMI) was classified into four categories: underweight (≤18.5), normal weight (18.5-24.9), overweight (25.0-29.9), and obese (≥30.3).

H. pylori Stool Antigen Test (Premier Platinum HpSA)
The Premier Platinum HpSA is a sandwich enzyme immunoassay which employs a plurality of monoclonal anti-H. pylori antibodies absorbed to microwells. The collected stool samples were received in an airtight transport container, then stored either at 4 • C for up to three days before the test or frozen immediately upon received at −20 • C until tested. Diluted stool samples, positive and negative controls were added to each antibody-coated microwell. Tests were performed in duplicate according to the manufacturer's instructions. Absorbance equal to or above 0.1 was considered a positive result. As the test has not been validated for use on watery diarrheal stools, all diarrheal stool collected were discarded. Volunteers participants subjected to antimicrobials, proton pump inhibitors, and bismuth preparations within the last two weeks were excluded from the study.

Statistical Analysis
Data were analyzed using the IBM SPSS Statistic 25.0 for Windows. The Chi-Square was used to test group differences. Significant predictors of H. pylori positivity and socio-demographic, lifestyle characteristics, and gastrointestinal symptoms were assessed using multiple logistic regressions. The significance level was determined at p-value < 0.05.

Conclusions
For the first time, the prevalence of H. pylori and its associated risk factors has been evaluated among healthy asymptomatic children and adults residents in the UAE. Almost half of the tested study participants were found to be infected with H. pylori, including children and adults. Moreover, univariate and multivariate logistic regression analysis showed a significant association between H. pylori infection and gender, age, ethnicity, profession, domestic overcrowding, source of drinking water, and gastrointestinal characteristics of participants. Prevention is worthy by taking into consideration the socio-demographic factors, lifestyle, and gastrointestinal characteristics found to be associated with H. pylori infection. Understanding the prevalence pattern of H. pylori in UAE will help in prioritizing public health efforts to better manage the complications of H. pylori infection.