Introduction

In 2005, Warren and Marshal received the Nobel prize for physiology or medicine for their discovery about Helicobacter pylori (H. pylori) being one of the common human bacterial pathogen that colonizes in the stomach causing gastroduodenal disorders (Niyaz 2005). This discovery changed the peptic ulcer disease paradigm from a chronic disease of unknown causes to an infectious disease that can be cured with the use of antibiotics and acid secretion inhibitors (Basset et al. 2002). Currently, evidence supports that H. pylori is the causative agent of several gastroduodenal disorders including active chronic gastritis, peptic ulcer, and acute erosive gastritis and is a major risk factor for gastric carcinoma and MALT lymphoma (Amieva and EL–Omar 2008; Kuipers 1999; Malfertheiner et al. 2002).

H. pylori infection is a very common and chronic infection around the globe and varies among and within populations, but it is almost universal in developing countries (Dorji et al. 2013; Jang et al. 2010). The risk of infection is multifactorial and is related to the standards of living, hygiene, and sanitation (Ahmed et al. 2007), including the contamination of drinking water and ingestion of fecal-contaminated vegetables as well as lifestyle habits (Kusters et al. 2006; Zhu et al. 2014).

Mizoram, a small state located in Northeastern India, is one of the hotspot regions for H. pylori infection and also the state that has the highest incidence of gastric cancer in India (Ghatak et al. 2016). Although the reasons for the high levels of H. pylori infection in the region are not fully understood, the standard of living in the state is low with more than 20% of the population living under the poverty line (Phukan et al. 2006). In addition, Mizo people have a unique diet and lifestyle factor; they are predominantly non-vegetarians with high intake of animal fat in the form of pork meat and fermented pork fat (saum) that is used to prepare vegetable stew (bai) (Ghatak et al. 2016; Phukan et al. 2006). Also, this population has a peculiar habit of consumption of tobacco smoke–infused aqueous solution called “tuibur.” About 5 ml of tuibur is kept in the mouth for few minutes and then is usually spit out; occasionally, some individuals may also swallow the product. We have previously reported the production and distribution of tuibur (Madathil et al. 2018). The unregulated makeshift industry of tuibur production uses unfiltered water and is usually located in unhygienic conditions. Presence of carcinogenic TSNAs (tobacco-specific nitrosamines), NNN (N’-nitrosonornicotine), and NNK [4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone] in tuibur can be a potential risk factor for gastric carcinoma (Muthukumaran 2016). Furthermore, tuibur even if not swallowed may modify the saliva pH making it more alkaline, thus increasing the risk of H. pylori infection in the mouth which may serve as a reservoir of the bacteria (Lalmuanpuii and Muthukumaran 2016).

H. pylori and tuibur have been associated with risk of gastric cancer (Ghatak et al. 2016). Moreover, although there is evidence for some risk factors for H. pylori infection, the exact risk factors for the bacterial infection remains largely unknown. In addition, no studies have investigated the environmental and lifestyle factors associated with H. pylori in a tribal population living in a hilly mountainous region with unique lifestyle habits. Hence, the present study hypothesizes that tuibur use might be a major risk factor associated with H. pylori infection.

Methods

Study population

A total of 863 patients with gastritis who were scheduled for an endoscopy in Trinity Diagnostic Centre, Aizawl District, Mizoram, Northeast India were recruited during February 2014 to August 2016. All the participants included were Mizo and presented with dyspeptic symptoms or gastrological problems requiring upper GI endoscopy. Patients who had taken antibiotics for gastrological problems for 2 weeks, or with gastrointestinal bleeding, are pregnant, or with a history of gastrectomy were excluded from the study because of the high chances of false-negative result in urease test for H. pylori. We used a convenience sample to increase the probability to capture potential individuals with H. pylori infection. The study was approved by the ethical committee of Civil Hospital, Aizawl (No. B.12018/1/13-CH(A)/IEC/36) as well as the Mizoram University ethical committee.

Data collection

An in-person interview using a validated and standardized questionnaire in both languages, Mizo and English, was used to collect information on array of exposures including socio-demographic characteristics (e.g., age, levels of education, occupation, residential location, marital status, number of children), medical history, health status, family history of H. pylori infection, and lifestyle factors (e.g., tuibur consumption, tobacco and alcohol use). Smoking tobacco and alcohol consumption were collected as self-reported measures (regular, occasional, former, or never consumers). In addition to the frequency of use, combinations of different smokeless tobacco products (e.g., tuibur, betel nut, pan, zarda) and dietary habits (e.g., strictly vegetarian diet, raw or uncooked vegetable consumption, fermented pig fat (saum), salt and pickle consumption) were also collected. Finally, we also recorded information on sanitation and drinking water sources (e.g., unfiltered, filtered [using ceramic, active carbon/charcoal, UV sterilization, and reverse osmosis] and boiled).

Endoscopy and biopsy sampling

Endoscopic samples collected from the antrum portion of the participant’s stomach were used to detect H. pylori infection based on the rapid urease test kit. Endoscopy was carried out using Fujinon VP-4450 HD Series on patients after an overnight fast. Two biopsy specimens per patient (replicates) from the mucosa of the gastric antrum were obtained by endoscopy and were placed in a small screw-capped bottle containing 0.2-ml sterile normal saline to maintain humidity. From the biopsy specimens collected, one was used for the rapid urease test and the other for PCR amplification of urease C gene.

Rapid urease test for H. pylori status

The endoscopic sample was screened by a commercially available rapid urease kit (RUT DRY Test kit, Gastro Cure System, WB, India) (Brown et al. 2002). The endoscopic samples were directly placed into the well of the kit. The urease enzyme produced by H. pylori rapidly hydrolyzes urea in the well, producing ammonia. The rise in the pH of the medium by ammonium ions was detected with a pH-based color indicator. In the case of H. pylori infection in the samples, there will be an immediate change in color from yellow to pink (H. pylori-positive). Samples that remained yellow, that is, without color change were inferred as an absence for H. pylori infection (H. pylori-negative). This test was read after 2 h for positive and negative results. Cross validation was performed by PCR amplification of urease C gene using forward primer 5′ -AAGCTTTTAGGGGTGTTAGGGGTTT- 3′ and reverse primer 5′ -AAGCTTACTTTCTAACACTAACGC- 3′ with thermal condition of 95 °C for 7 min followed by 35 cycle of 95 °C for 1 min, 54 °C for 40 s, 72 °C for 1 min, and final extension of 72 °C for 5 min. PCR amplicons were visualized on a 1.5% agarose gel (Lu et al. 1999).

Statistical analysis

The characteristics of the population were explored using descriptive statistics. The association between tuibur and H. pylori infection was investigated using unconditional logistic regression. We performed a hierarchical modeling strategy to evaluate the change in the strength of this association when adjusted for potential confounder. Four sets of confounders were considered. (i) Model 1: tobacco and alcohol factors (smoked tobacco, smokeless tobacco, and alcohol consumption); (ii) model 2: demographic factors (age, sex, marital status); (iii) model 3: socioeconomic position (no. of persons in household, drinking water sources, and sanitation); and (iv) model 4: dietary factors (raw food, salt, saum, pickles). Odds ratios and 95% confidence intervals were estimated. All statistical analyses were performed using R statistical package ver3.3.0 (The R Foundation for Statistical Computing).

Results

A total of 943 patients was invited to participate in the study, and 863 accepted the invitation to participate in the study (response rate of 91.5%). However, the percentage of men refusing to participate in the study was higher (75% of the total refusal). More than half of the samples (475 (55%)) were tested positive for H. pylori infection by both rapid urease test (RUT) as well as urease C gene PCR-based methods. The age of the participants ranged from 8 to 92 years, with comparable distribution between H. pylori-positive and -negative groups. The H. pylori-positive group had a slightly higher proportion of male (58.5%), married people (83.4%), those with below matriculation level of education (52%) compared with the H. pylori-negative group (55.4%, 79.9%, 47.2%, respectively). Most of the participants had access to a public water distribution system (81.1% and 83.8% H. pylori-positive and -negative, respectively). However, a lower proportion of participants in the H. pylori-positive group drank filtered water. Dietary factors showed a similar distribution between H. pylori-positive and -negative groups (Table 1).

Table 1 Distribution of sociodemographic and lifestyle factors among H. pylori-positive and -negative participants
Full size table

The majority of the participants in the H. pylori-positive group were tobacco chewers (74.1%) and smoked forms of tobacco (57.3%) compared with 63.9% and 35.6% among the H. pylori-negative group, respectively. There was approximately three times higher proportion of tuibur users in the H. pylori-positive (15.2%) compared with the H. pylori-negative group (5.9%).

Results from the logistic regression analysis showed that tuibur users increased the odds of H. pylori-positive endoscopy by 3.32 (OR = 3.32; 95% CI = 1.95–5.83) times compared with never users. This association was not attenuated after adjusting for several known risk factors of H. pylori infection. In addition, chewed tobacco users had 1.49 (OR = 1.49; 95% CI = 1.06–2.09) higher odds of having H. pylori-positive endoscopy result (Table 2).

Table 2 Association between tobacco and alcohol habits and H. pylori infection
Full size table

Discussion

Our results suggest an association between the use of tuibur and H. pylori infection among Mizo individuals with gastritis. Although one previous study has reported an association between tuibur consumption and the risk of gastric cancer (Ghatak et al. 2016), to our knowledge, this is the first study investigating the association between tuibur use and H. pylori infection. Moreover, this association was not attenuated when we included important risk factors (e.g., indicators of socioeconomic position) in the models (Table 2).

Although we have not tested the tuibur samples for the presence of H. pylori bacterium, our previous work reported the unhygienic conditions in which tuibur is produced (Madathil et al. 2018). A higher incidence of gastric cancer was reported for the consumers of smoked tobacco as well as tuibur in Northeast India (Misra et al. 2014), and in the present study, we found a strong association between tuibur and H. pylori presence. Furthermore, the average alkaline pH of tuibur is above 9 and may have an impact on the oral cavity and the stomach which may facilitate the viability of H. pylori bacterium (Lalruatfela et al. 2017). The prolonged exposure to smokeless tobacco extract affects the drug-metabolizing enzymes of the gastrointestinal tract, which may be an important factor that determines the susceptibility to different organisms leading to carcinogenesis (IARC Working Group 2007). Tuibur also contains carcinogenic agents like TSNAs, NNN, NNK, and others which is reported as health hazards as well as causes cancer and other diseases. These molecules may have a correlation with H. pylori proliferation by changing pH in the oral cavity as well as in the stomach and together influencing the disease outcome (Muthukumaran 2016; Lalmuanpuii and Muthukumaran 2016).

The association between tobacco and alcohol use and H. pylori infection is inconclusive (Brown et al. 2002; Misra et al. 2014). Studies from European countries have reported positive or negative (Sharma et al. 2016; Ahmed et al. 2006) or inconclusive relationship between tobacco, alcohol use, and H. pylori infection (Sharma et al. 2016; Ahmed et al. 2006; Abebaw et al. 2014; Brown 2000). In our study, tobacco smokers and alcohol consumption were associated with increased odds of H. pylori infection (Table 2). A negative association was observed between the past users of tobacco and H. pylori infection. In fact, changes in alcohol and smoking habits were reported by 43.4% and 55.2%, respectively, of the participants with dyspeptic symptoms.

The use of betel quid (a chewable tobacco product) was associated with increased odds for H. pylori infection in our study. The practice of washing raw leaves of betel vine and areca nut (used to prepare the betel quid) in unhygienic water may explain our results (Abebaw et al. 2014). Interestingly, the results were not grossly attenuated even after adjusting for dietary, water sources, and socioeconomic variables.

Our study has several limitations. First, the study sample only included participants of Mizo origin. However, a high incidence of gastric cancer among this population suited our aim ideally. Incidence rates of gastric ulcer, gastritis, and gastric carcinoma show variations in geographical location and lifestyle habits among Northeast Indian states (Jang et al. 2010; Aziz et al. 2015; Darnindro et al. 2015). Second, for feasibility reasons, we recruited our participants from patients reporting with gastritis at a gastroenterology clinic in Aizawl. Although this strategy allowed us to focus on a group with a high probability of H. pylori infection, it might have biased our results. For example, if tuibur use can induce gastritis regardless of H. pylori infection, our study may have succumbed to collider stratification bias (Cole et al. 2010). However, a direct link between tuibur use and gastritis, not-mediated through H. pylori infection, has not been established in the literature.

Unfortunately, due to the cross-sectional nature of our study, reverse causality bias cannot be ruled out. However, it is rather unlikely that anyone would start using tuibur due to gastritis induced by H. pylori infection. Moreover, our study provides initial evidence for an association and replication of our findings, though prospective and mechanistic studies are warranted. Isolation of H. pylori bacterium from samples of tuibur collected from point of sales and potentially from the oral cavity of regular tuibur users is the next essential step in the scientific inquiry.

In conclusion, we report a positive cross-sectional association between tuibur use and H. pylori infection. Future studies should consider the possibility of the oral cavity of regular tuibur user, being a reservoir for repeated H. pylori infections.