Characteristics of helicobacter pylori infection in a national referral hospital in Bhutan

Introduction: Helicobacter pylori (H. pylori) is a bacterium causing chronic gastric infection and may cause gastric cancer. It was necessary to see the trend of infection, especially in symptomatic patients. This retrospective descriptive study was aimed to describe the characteristics of H. pylori infection in Bhutanese patients referred for an endoscopy to the National Referral Hospital, Thimphu. Methods: The sample of the study was randomized 380 medical records of the patients who underwent upper gastrointestinal endoscopy and Rapid Urea Test for symptomatic dyspepsia and peptic ulcer. Data was collected using a survey form designed by the researchers. Data analysis was done using descriptive statistics and either Chi-square or Fisher’s exact test. Results: The prevalence of H. pylori infection was very high (76.6%). The mean age of the infection was 42 with a range from 15 to 84 years. The highest prevalence of infection was observed in the age group 20-29 years (82.7%) and lowest in the oldest age group 70-84 years (66.7%). The analysis showed no significant difference in infection amongst age groups, gender, and endoscopic findings to the positive results at 5% significant level except for monthly prevalence (p<0.001). Gastritis was the commonest endoscopy finding (153/380) and gastro-duodenitis had the highest positivity rate (88.9%). Conclusions: The prevalence of infection was relatively high compared with previous studies. Young and middle-aged adults had a high prevalence and this group needs to be given priority for screening and eradication treatment considering limited resources to prevent associated gastric cancer in Bhutan.


INTRODUCTION
Helicobacter pylori (H. pylori) is a bacterium that is widespread in human population and is responsible for most common chronic symptoms of upper gastrointestinal disorders like gastritis, duodenal ulcer, gastric ulcers and gastric adenocarcinoma 1 and it is an important public health problem in Bhutan. The prevalence of H. Pylori infection in Bhutan is the highest among the South Asian countries. The infection was estimated to be approximately 73% in the community-based studies published in 2013 2 and 86% in 2014 3 . The prevalence was around 43-61% in a neighbouring country Nepal 4 and 67% in Bangladesh 5 .
H. pylori infection is a serious health concern as its infection results in a higher risk of acquiring gastric cancer 3 . The incidence rate of gastric cancer in Bhutan is twice as high as in the United States. Among all gastro-intestinal related cancers, approximately 46% were gastric cancer in Bhutanese patients 6 . As the infection was found high in all age groups 3 it is expected that an increasing number of patients would develop chronic gastritis and develop cancer placing higher health burden to hospitals and healthcare cost.
Research studies on H. pylori infection in Bhutan and other countries in the South Asia region had just started recently 2-4,7 . More evidence from studies with hospital-based research will be helpful to support the plan for prevention and eradication of the infection especially when there is a lack of information from the recent studies. Also, factors contributing to the high prevalence of H. pylori infection were associated with poverty, poor hygiene and sanitation, geographical areas, environment, and educational level 3 . There has been rapid development in Bhutan leading to improvements in the socio-economic conditions of its people 8 . However, the trend of its infection, diagnosis, and treatment remains unclear for the last five years. The Rapid Urea Test (RUT) is highly sensitive (95.6%) and specific to detect active H. pylori infection 9 . The facility to diagnose by RUT in gastric biopsy sample is currently available at Jigme Dorji Wangchuck National Referral Hospital (JDWNRH) since January 2018 and Central and Eastern Regional Referral Hospitals at Gelephu and Mongar have just started. The hospital-based study was needed as these patients are always symptomatic and seeking health care services. The severity of the infection, as well as other comorbidities, may be different compared with the findings of the community based-survey.
Therefore, this hospital-based study aimed to characterize the trend of H. pylori infection in terms of demography, seasonal differences and endoscopy findings in Bhutanese people.

METHODS
This is a retrospective descriptive study based on the data of RUT results of all patients who underwent upper gastrointestinal endoscopy (endoscopy) for dyspepsia and peptic ulcer symptoms at JDWNRH from 1 st January 2018 to 31 st December 2018. Considering an estimated 3000 individuals who have undergone endoscopy during the study period and expected prevalence of 73%, using Yamane's Table 10,11 with a precision of determination value (d) at 5%, confidence level 95%, a sample size of 360 was estimated. Upon retrieving the records of endoscopy, only about 1000 individuals had undergone RUT. A systematic random technique was used to sample the records, with a selection of first sample corresponding to random number selected between 1 to 3, and selecting every third individual until 31 st December 2018. Inclusion criteria were the records of all patients who underwent endoscopy with the RUT investigation during the study period. Those patients without confirmed test results and unrecorded endoscopy findings were excluded from the study.
Following systematic random sampling, a total of 380 samples were selected by the end of December 2018 and were analysed.
The data collection proforma was developed by the researchers and was validated by three professional experts who were general specialist, gastroenterologist, and general surgeon. The reliability of the instrument was field-tested for 20 initial records. The data were analyzed using SPSS software version 25 (IBM Corp. Released 2017). Descriptive analysis was performed and summarized as mean, standard deviation, frequencies, and percentage. Chi-square and Fisher's exact test were used to evaluate the association between the age group, gender, monthly variation and endoscopic findings with the RUT test results. A statistical association between RUT and demographic and other explanatory variables were assessed at 5% significance level. This study was conducted with the approval of the Research Ethics Board of Health, Ministry of Health, Thimphu, Bhutan vide approval letter number REBH/Approval/2018/104.

RESULTS
The results showed that among the 380 records examined, 237 (62.4%) were female and 143 cases (37.6%) were male, and age ranged from 15 to 84 years (mean=42.03, SD=16.3). Of those, 291 patients (76.6%) tested positive to H. pylori infection. The infection rate was high in the young adults in the age range of 20-29 years (82.7%) followed by the middle age with 80.0% (age range of 40-49 years). The lowest infection rate was found in the oldest age group range of 80-89 years (66.7%) as depicted in Table 1.
The analysis of age within the RUT positive group found differences in the pattern of distribution. The infection rate was  Table 1. The youngest age group of 10-19 years also had a lower infection rate (5.2%, 15/291).
The H. Pylori infection was found higher among females (184/237, 77.6%) than males (107/143, 74.8%) but not significantly different. Gastritis was the commonest endoscopic findings (153/380) but gastro-duodenitis had the highest rates of infection (88.9%). A high proportion of H. pylori infection was also found in the group of individuals having normal mucosa as shown in Table 1.
The results in Table 1 and Figure 1 show that higher frequencies of positive RUT results were found from June to November months (10.7 -14.1%) and the highest peak was in August (14.1%). The association between the age group, gender, monthly variation and endoscopic findings with the RUT test results showed no significant association of H. pylori infection among age groups (p=0.793), between gender (p=0.534) or categories of endoscopic findings (p=0.368), except monthly variation (p=0.000).

DISCUSSION
The prevalence of H. pylori infection in the current study was 76.6% showing an alarming sign for the risk of developing stomach cancer. This prevalence is much higher compared with the findings of previous studies by Vilaichone and team which reported the prevalence of 54.6% using RUT 2 . The difference of

Figure 1. Monthly distribution in the frequency of H. Pylori test (RUT) result
over 20% prevalence from the recent study may be attributed to the variation between community-based study and the hospitalbased study as more proportions of symptomatic patients are screened in hospital-based study. With a similar mountainous populace in Nepal, the prevalence of H. pylori infection reported was only 61.3% 4 . Together, the present findings confirm the prevalence of H. pylori infection using the RUT test in this study is higher than that of the previous studies in Bhutan, Bangladesh, and Nepal 2-5,12 . It is worth discussing that the different methods of investigating H. pylori infection may result in a different range of prevalence rates even when studied in the same setting and group of symptomatic patients seeking treatment. Using a similar setting for investigating H. pylori infection, the hospital-based study of Dorji et al 3 reported 86% prevalence rate which had a higher prevalence rate compared to this study. It may be due to the difference in the investigation method used which was based on the serological test. As known, serology shows the lowest specificity and accuracy in comparison with other tests and the antibody titers can remain high for months after the elimination of the infection, attributing to higher prevalence rate in the previous study.
To identify the high-risk group, this study investigated the factors of age, gender, endoscopy findings and periods being tested. The results of these factors related to the prevalence of H. pylori infectionare discussed further. About 83% (242) positive cases were under 60 years and the high prevalence rates of infection were the groups of early and middle-aged adults. This phenomenon is congruent with previous studies in South Asia and other countries that the prevalence of H. pylori infection decreased with increasing age 12 and showed a lower prevalence in the older age group 4,5 . Although this study did not find any significant association between age and H. pylori infection (p>0.05), it is apparent that the younger age groups consisted of a high-risk group that may develop gastric cancer in later years of their lives. This shows that any program for the elimination of H. pylori infection should target even young adult groups of 20-29 years.
The role of gender as a risk factor for H. pylori infection is still being debated. Meta-analysis studies recently reported male predominated H. pylori infection in adults [13][14][15][16] . Again, it is noticeable that the result of this study found a higher prevalence of infection in females (184/291, 63.2%) and a higher infection rate within the female gender (184/237,77.6%). This study is congruent with the previous study conducted in Bangladesh where higher prevalence was detected in females 5 . In contrary, the previous community-based study of Shiota and his team 12 , showed a high percentage of positive test results in the male Bhutanese group. However, a statistically significant difference in the prevalence of the infection rate between the gender group was not found (p>0.05).
All types of endoscopy findings were associated with very high H. Pylori infection and prevalence was similar (p>0.05). The hospital currently has only a simple white light endoscope where images cannot be magnified and it is difficult to even for an experienced endoscopist to visualize the gastric findings in detail. With the ongoing training of endoscopists in the hospital, the observational findings in the gastric mucosa are expected to be improved and uniform reporting is expected in future.
Climatic conditions may be related to detecting a high prevalence of H. pylori infection in the summer season as shown in Figure 1. H. pylori bacteria normally grow within a temperature range of 30°C to 37 °C, with optimum growth at 37 °C and survives in complex foodstuffs such as milk, vegetables, and ready-to-eat foods 1 . The prevalence in this study varied with the months compared to a study in Venezuela 17 which found that the frequency of H. pylori infection was significantly high in the rainy season but contradictory to another study in Iran 18 which found the higher frequency in winter months than in spring and hot summer months. Therefore, more studies are needed to see any seasonal variation in detecting H. pylori infections.
Other findings found were lack of information on geographic distribution, histopathology sampling and treatment and this could be attributed to the existing documentation practices of the hospital.

CONCLUSIONS
This study demonstrates again that the prevalence of H. pylori infection in Bhutan is very high and increasing, especially in young and middle-aged adults. This indicates that the majority of the Bhutanese population are at high risk of developing gastric cancer later in their life.

LIMITATIONS AND RECOMMENDATIONS
Since this study was based on retrospective hospital data, the other risk factors for H.pylori infection could not be assessed. As an example, the data on geographical location, treatment, treatment outcomes and follow up endoscopy were not accessible due to a lack of information in the record. This can be improved with prospective study designs in the future.The impact of this study on clinical practice is a strong recommendation to provide a noninvasive diagnostic test for younger groups and reevaluation for the eradication of H. pyloriinfection after treatment.