Helicobacter pylori (H. pylori) infection is a common condition with an estimated half of the world's adult population having been exposed to this organism.[1] H. pylori is an important and a common bacterial pathogen infecting upper gastrointestinal (GI) track and causing various symptoms due to inflammation of the GI track. Prevalence of this infection varies worldwide being as low as completely negative to higher than 80 percentage in developing countries.[2,3] The prevalence rates are higher in developing nations.[3,4] Microbiologically, H. pylori are gram negative, spiral, and flagellated bacilli found under the mucous layer in gastric pits adjacent to the gastric epithelial cells.[5,6] Endoscopic biopsy based tests like rapid urease testing (RUT) and histopathology can be done to identify H. pylori. However, histopathology is considered the gold standard and carry high sensitivity and specificity of more than 90%. But such facilities may not be available in resource poor setting areas of the world; therefore, in low resource communities with high H. pylori prevalence, empirical treatment could be more practical then diagnostic tests.[7,8] Non-invasive tests such as urea breath test, serological immunoglobulin G, A, and M serology, stool antigen test, and saliva antibody test are not easily available in our country.

There are many studies on this topic but there are limited information about the prevalence of H. pylori in Western Nepal. The present study evaluates the prevalence of H. pylori infection among patients undergoing gastrodudenoscopy for various upper GI symptoms.

This observational, cross-sectional, and analytical study was carried out at Endoscopic unit and Department of Pathology of Lumbini Medical College Teaching Hospital (LMCTH). Ethical clearance was obtained from Institutional Review Committee (IRC) of the institute. Secondary data were collected from 1^st of August 2017 to 31^st of October 2017. Data were collected retrospectively from the medical records of all patients who underwent gastrodudenoscopy from 1^st of January 2015 to 30^th of June 2017.

Base line data were obtained and the demographics, indications, and histopathological findings were recorded. Data were entered in Microsoft Excel™. Data analysis was carried out using statistical package for social sciences, version 21 (SPSS™ Inc. Chicago, IL, USA). Descriptive results were presented as mean, standard deviation, frequency, and percentage. Association between categorical independent variables and outcome variable was analyzed using Pearson’s Chi-square test or Fisher exact test. T-test was done to compare mean of two groups. P value of less than 0.05 was deemed statistically significant.

There were 256 patients who fulfilled inclusion/exclusion criteria. Among them, H. pylori infection was present in 63 (24.6%) cases. Thus, the prevalence of H. pylori infection among patients undergoing gastroscopic biopsy was 24.6%. There were 34 (54%, N = 63) males and 29 (46%) females with H. pylori infection. There was no statistically significant association (p = 0.82) between gender and presence of H. pylori infection as shown in Table 1.

The age range of patients undergoing gastrodudenoscopy was from 15 to 88 years (median = 47) while the mean age was 47.07 years (SD =16.5) indicating a symmetric distribution of age of the patients. Mean age of H. pylori infected male was 52.5 years (SD = 14.5) and that of female was 48.3 years (SD = 16.5). This difference of age among gender was not statistically significant (t = 1.07, df = 61, p = 0.29). Thus, the age of male and female patients infected with H. pylori was comparable between groups.

Frequency of H. pylori infected cases according to age is shown in Fig 2. It demonstrates that the infection is common between 40 to 60 years of age.

Relationship between various symptoms and presence of H. pylori infection is presented in Table 2. Heart-burn was the commonest (n = 120, 46.9%) complain followed by dyspepsia (n = 61, 23.8%). Among various symptoms, GI bleed was only found to be significantly associated with H. pylori infection (p = 0.006).

Most common histopathological diagnosis were gastritis (n = 158, 61.7%) followed by gastrodudenitis (n = 50, 19.5%). Frequency of other histopathological diagnosis and their association with H. pylori infection is presented in Table 3. None of the diagnosis were found to be significantly associated with H. pylori infection.

Our study showed an overall prevalence of H. pylori infection of 24.6% among patients undergoing gastroscopy and biopsy. The prevalence was high in the age between 40 to 60 years. Frequency curve of H. pylori infection did not show any trends increasing with the advancing age. This can be due to fewer number of elderly patients.[9] This finding is similar to that of the study laid by Tarkhasvili et al.[10] However, studies by Shokrzadeh et al.[11] and Kaore et al.[12] reported increasing H. pylori infection in age groups of 20-40 years than the older age groups. Recently, several studies have reported declining prevalence of H. pylori infection over the last decade.[9,11]

We did not get a significant difference of H. pylori prevalence according to the gender. There are also some studies which have not found any sex differences.[13,14] In contrast, a study by Kaore et al.[12] found a higher prevalence in males. The reason for observed difference was not known but better hygienic practice may be the reasons for the lower prevalence in female.[15] To prove a strong correlation, it may require further studies.

There were several indications of gastrodudenoscopy. Heart burn was the most common (44.4%) followed by vomiting (25.4%), and dyspepsia (17.5%) among others. However, GI bleeding was only found to be statistically significant (p = 0.006). One of the study showed that a high prevalence of occult non steroid anti-inflammatory drug usages was found in H. pylori negative (PUD) patients which was based on elevated serum thromboxane.[16] In our study, 50% patients had non H. pylori PUD which might be the NSAID usages but we were not able to assess the association between NSAID and non H. pylori PUD because of incomplete data. In the present study, the commonest histopathological finding was gastritis (n = 158, 61.7%). However, the correlation of different histopathological findings with H. pylori was not statistically significant. This is in concordance with the observation of Jemilohun. et al.[17] This may be due to low number of cases being evaluated in our study. H. pylori have been found in 90% of patient with chronic gastritis, 95 % with duodenal ulcer, 70% with gastric ulcer and 50% with gastric cancer.[18] However, our study H. pylori was found in 24.1% of the patients with gastritis, 66.7% with duodenal ulcer, 40% with gastric ulcer, and 20% with carcinoma stomach. Such variation in prevalence is probably because of different dietary habits, life styles, immunological factors, and genetics of different countries.

There were several indications of gastrodudenoscopy. Heart burn was the most common (44.4%) followed by vomiting (25.4%), and dyspepsia (17.5%) among others. However, GI bleeding was only found to be statistically significant (p = 0.006). One of the study showed that a high prevalence of occult non steroid anti-inflammatory drug usages was found in H. pylori negative (PUD) patients which was based on elevated serum thromboxane.[16] In our study, 50% patients had non H. pylori PUD which might be the NSAID usages but we were not able to assess the association between NSAID and non H. pylori PUD because of incomplete data. In the present study, the commonest histopathological finding was gastritis (n = 158, 61.7%). However, the correlation of different histopathological findings with H. pylori was not statistically significant. This is in concordance with the observation of Jemilohun. et al.[17] This may be due to low number of cases being evaluated in our study. H. pylori have been found in 90% of patient with chronic gastritis, 95 % with duodenal ulcer, 70% with gastric ulcer and 50% with gastric cancer.[18] However, our study H. pylori was found in 24.1% of the patients with gastritis, 66.7% with duodenal ulcer, 40% with gastric ulcer, and 20% with carcinoma stomach. Such variation in prevalence is probably because of different dietary habits, life styles, immunological factors, and genetics of different countries.

There are several limitations in our study. Firstly, the retrospective study design is inherently associated with limitations. Secondly, few number of cases cannot be generalized into a larger population. Lastly, endoscopic procedures and histopathological evaluation was carried out by different physicians, surgeons, and pathologists. There was a chance of inter-observer variability.

H. pylori infection was prevalent in 24.6% of the cases undergoing gastrodudenoscopy and biopsy. It was most common between 40 to 60 years of age. Male and female were equally likely to be infected. Heart burn was the most common symptom but gastrointestinal bleeding was the only significantly associated symptom with H. pylori infection. None of the histopathological diagnosis were significantly associated with presence of H. pylori infection.