INTRODUCTION
Helicobacter pylori is the primary etiologic agent involved in gastric diseases in humans with worldwide distributions1. In 2005, H. pylori was identified as a microbiologic contaminant of water, and its role in gastric diseases was further assessed2–4. Improvements in sanitary conditions and higher human development indices have reduced the prevalence rate of H. pylori infection in developed countries. However, the prevalence rate in developing countries remains high1.
Populations with lower socioeconomic levels present with higher prevalence rates of H. pylori infection5–7. In developing countries, the prevalence rate is higher than 70%, whereas in developed countries, such as Canada, Australia, and Switzerland, the prevalence rate can be as low as 30%. The prevalence rate of H. pylori infection in Brazil, particularly in its poorest areas, is similar to that in Africa, which is around 70-90%1,8.
Brazil has a population of over 200 million individuals with varying ethnic and geographic backgrounds. The country is heterogeneous with respect to sanitary coverage, and a large proportion of its population still do not have access to high-quality water9,10. Access to high-quality water is not evenly distributed across Brazil and is highly dependent on sanitary facilities and infrastructure.
H. pylori is considered a carcinogen because it can induce the development of lesions such as atrophy, metaplasia, and dysplasia, leading to adenocarcinoma11,12. A meta-analysis showed that the risk of developing gastric cancer increased by two-fold when individuals were infected with H. pylori strains that contained the cagA+ gene13–15. The use of antibiotic therapy for the treatment of H. pylori infection decreased the development of gastric cancer, particularly when the treatment was started prior to the onset of precursor lesions. Controlling environmental factors, such as high salt intake, nitrites, and smoking, has decreased the burden of H. pyloriinfection14. Therefore, decreasing the risk factors of H. pylori infection is one of the measures that can prevent the development of gastric cancer. This study aimed to evaluate the risk factors associated with H. pyloriinfection in patients undergoing endoscopy to validate the cause of dyspeptic symptoms in an urban population in northeast Brazil and to compare the urease test and polymerase chain reaction (PCR) assay results with histopathological findings.
METHODS
Study participants
The study was conducted in Campina Grande, State of Paraiba, Northeast Brazil, a city with a population of 402,912 inhabitants. Both male and female participants and those aged 18 years and older who agreed to participate in the study were recruited from a referral center for gastric diseases. Participants were excluded from the study if they used nonsteroidal antiinflammatory proton pump inhibitors or antibiotics a month prior to examination or if they lived outside the City of Campina Grande. We did not include participants living outside Campina Grande because we could not validate the source of drinking water. Moreover, participants who were pregnant, those who underwent gastrectomy, or those who were unable to undergo endoscopy were excluded. Participants were asked to complete a survey questionnaire with demographic, social, economic, and lifestyle questions. Of the 1,257 participants scheduled to undergo endoscopy between February 2012 and October 2012, 759 underwent endoscopy, of which 200 met the inclusion criteria and were recruited.
Endoscopy assessment
After clinical evaluation, the participants underwent endoscopy, which was performed by a trained endoscopist. Macroscopic data included topography, localization, and type of injury. Systemic anesthesia was conducted with midazolam and fentanyl. During the procedure, gastric biopsies were carried out (two in the antrum, two in the body, and one in the incisura angularis) for histological study, which is in accordance with the recommendations of the Brazilian Society of Digestive Endoscopy.
Determination of Helicobacter pylori infection
H. pylori infection was determined via histological examination, urease test, and PCR. Histological examination is the gold standard test for H. pylori infection. Two fragments were collected from the antrum to determine the presence of urease (Urotest Kit, RenyLab, Barbacena, MG, Brazil), and all test results were read within 24 hours of tissue collection.
Tissue fragments were fixed in 10% formalin and were prepared according to standard histology protocol. Slides were stained with hematoxylin-eosin and were analyzed by a single pathologist. Deoxyribonucleic acid (DNA) was extracted from formalin-fixed paraffin-embedded tissues obtained from gastric biopsy16. Eight slices (10μm) from each sample were collected, and DNA extraction was performed using QIAamp DNA from paraffin-embedded tissues (QIAGEN Inc., Chatsworth, California, the USA) according to the manufacturer’s instructions. The presence of H. pylori DNA was confirmed using PCR amplification of the 16S-ribosomal DNA (rDNA) using specific primers (5′-CTGGAGAGACTAAGCCCTCC-3′ and 5′-GAGGAATACTCATTGCGAAGGCGA-3′]16. PCR reactions were performed with 20μL containing 1× PCR buffer, 0.2 mmol/L of each deoxynucleotide, 0.2μM of each primer, and 1 unit of Taq polymerase. The PCR cycles were 95°C for 2 min; (95°C for 1 min; 58°C for 1 min; and 72°C for 1 min) 40X; 72°C for 10 min; and 10°C. Gel electrophoresis (2% agarose) was used to determine whether PCR products had the correct length of 229bp. Ten samples positive on PCR were sent for sequencing at Macrogen Inc. (Seoul, Korea) to confirm the results of the PCR assay.
Histopathological examination
The diagnoses were made according to the updated Sydney System for chronic gastritis when the presence of lymphocytes in the lamina propria was higher than expected17. The existence of plasma cells was considered an indicator of chronic inflammation. The presence of neutrophils was considered a marker of active gastritis. The loss of glandular tissue was considered a marker of gastric mucosal atrophy17.
Ethical considerations
The study protocol was reviewed and approved by the ethical review board of Universidade Federal do Rio Grande do Norte. The study was conducted in accordance with the ethical standards of the Brazilian national committee of Ethics in inhuman research, which follows the principles of the Declaration of Helsinki. The approval number is CAAE: 0251.0.051.127-10. Informed consent was obtained from all the participants.
Statistical analysis
The comparisons were based on the H. pylori infection status, which was determined via histological examination. Fisher’s exact and Wilcoxon rank-sum tests were used to compare categorical and continuous variables, respectively. The comparison of sensitivity and specificity between diagnostic methods (Urease vs PCR) were determined by McNemar’s chi-square. Logistic regression models were established to further examine demographic, lifestyle, and socioeconomic factors associated with the risk of developing H. pylori infections. All analyses were performed using Stata (StataStatistical Software: Release 11. College Station, TX, United States) with 95% confidence interval (CI), and we set the significance level (α) at 0.05.
RESULTS
Comparison of diagnostic methods for H. pylori infection
The prevalence rate of H. pylori infection was 69.5% (139/200), 60% (120/200), and 58% (116/200) via the urease test, histological examination, and PCR assay, respectively. The urease test was more sensitive than PCR assay [92% (95% CI: 0.85, 0.96) vs. 82% (95% CI: 0.74, 0.89), p=0.012] for diagnosing H. pylori infection, whereas PCR assay had a higher specificity than the urease test [78% (95% CI: 0.67, 0.86) vs. 64% (95% CI: 0.52, 0.74), p=0.034]. Table 1 shows the performances of the two methods with respect to histological examination, which is the gold standard test for H. pylori infection.
Test | Histological examination | ||
---|---|---|---|
negative | positive | Total | |
Urease | |||
negative | 51 | 10 | 61 |
positive | 29 | 110 | 139 |
total | 80 | 120 | 200 |
PCR | |||
negative | 62 | 21 | 83 |
positive | 18 | 98 | 116 |
total | 80 | 119* | 199 |
PCR: polymerase chain reaction. Urease: sensitivity = 110/120 = 92% and specificity = 51/80 = 64%. PCR: sensitivity = 98/119* = 82% and specificity = 62/80 = 78%. *One sample failed in the PCR assay.
Demographic and clinical characteristics of the study population
All participants were from the same geographic region; 72.5% (145/200) of the participants were women. The sociodemographic data of the population are shown in Table 2. No difference was observed in the educational and socioeconomic levels of the H. pylori-positive or H. pylori-negative groups. Current smoking habits [adjusted odds ratio (OR): 5.19; 95% CI: 1.29-20.86; p=0.02] and intake of non-treated water (OR: 2.20; 95% CI: 1.02-4.72; p=0.044) were associated with H. pylori infection (Table 3). Other variables, such as age, hand washing, and cup sharing, were not associated with H. pylori infection (Table 3).
Helicobacter pylori | |||
---|---|---|---|
Characteristics of the participants | negative (n=80) | positive (n=120) | p value* |
Sex, n (%) | |||
male | 19 (23.8) | 36 (30.0) | 0.419 |
female | 61 (76.2) | 84 (70.0) | |
Age group (years), n (%) | |||
18-39 | 22 (27.5) | 59 (49.2) | 0.004 |
40-59 | 38 (47.5) | 46 (38.3) | |
≥60 | 20 (25.0) | 15 (12.5) | |
Median age (IQR), years | 47 (38–59.5) | 40 (32.5-52.0) | 0.005 |
Median BMI (IQR), kg/m2 | 26.0 (23.1-30.1) | 26.4 (23.5-30.2) | 0.741 |
Education, n (%) | |||
low | 40 (50.0) | 44 (36.7) | |
medium | 23 (28.8) | 43 (35.8) | 0.179 |
high | 17 (21.2) | 33 (27.5) | |
Income, n (%) | |||
low | 9 (11.2) | 13 (10.8) | |
medium | 61 (76.3) | 96 (80.0) | 0.738 |
high | 10 (12.5) | 11 (9.2) | |
Household crowding, n (%) | |||
<1 person/room | 60 (75.0) | 78 (65.0) | 0.09 |
≥1 person/room | 20 (25.0) | 42 (35.0) | |
Smoking status, n (%) | |||
never | 50 (62.5) | 74 (61.7) | |
current | 3 (3.8) | 18 (15.0) | 0.018 |
former | 27 (33.7) | 28 (23.3) | |
Alcohol drinking, n (%) | |||
never | 43 (53.8) | 62 (51.7) | |
current | 15 (18.7) | 42 (35.0) | 0.009 |
former | 22 (27.5) | 16 (13.3) | |
Coffee drinking, n (%) | |||
yes | 70 (87.5) | 109 (90.8) | 0.299 |
no | 10 (12.5) | 11 (9.2) | |
Water supply, n (%) | |||
mineral water | 31 (38.8) | 22 (18.3) | |
municipal | 41 (51.2) | 78 (65.0) | 0.005 |
others | 8 (10.0) | 20 (16.7) | |
Hand washing, n (%) | |||
yes | 66 (82.5) | 84 (70.0) | 0.032 |
no | 14 (17.5) | 36 (30.0) | |
Cup sharing, n (%) | |||
yes | 51 (63.8) | 84 (70.0) | 0.220 |
no | 29 (36.2) | 36 (30.0) |
IQR: interquartile range; BMI: body mass index. *Fisher’s exact test was used for categorical variables, and Wilcoxon rank-sum test was utilized for.quantitative variables.
Helicobacter pylori infection and its relationship with gastric diseases
Gastritis and duodenal ulcer were common findings in both individuals with and without H. pylori infection. Belching was the only symptom (p=0.05) associated with H. pylori infection (Table 4). There was no association between H. pylori infection and a family history of either peptic ulcer (p=0.457) or gastric cancer (p=0.447) (Table 4). Mild gastritis was more frequently observed in individuals who were H. pylori negative (p<0.001) than in those with moderate or severe gastritis, which was commonly associated with H. pylori-positive individuals (p<0.001) (Table 5). No difference was observed between the two groups in terms of gastroduodenal ulcer (p=0.474). There was no association between alcohol intake (p>0.05) and the grade of inflammation. Cigarette smoking (p=0.020) and consumption of municipal water (p=0.044) were associated with mild and moderate gastritis.
Unadjusted | Adjusted* | |||||
---|---|---|---|---|---|---|
Variable | OR | 95% CI | p value | OR | 95% CI | p value |
Sex | ||||||
male | 1.00 | – | – | 1.00 | – | – |
female | 0.73 | (0.38-1.39) | 0.333 | 0.68 | (0.31-1.56) | 0.350 |
Age | 0.98 | (0.96-0.99) | 0.011 | 0.98 | (0.96-1.01) | 0.232 |
BMI | 1.00 | (0.95-1.06) | 0.918 | 1.03 | (0.97-1.11) | 0.317 |
Education | ||||||
low | 1.00 | – | – | 1.00 | – | – |
medium | 0.96 | (0.44-2.09) | 0.924 | 0.84 | (0.34-2.06) | 0.703 |
high | 0.57 | (0.27-1.17) | 0.125 | 0.66 | (0.25-1.73) | 0.398 |
Income | ||||||
low | 1.00 | – | – | 1.00 | – | – |
medium | 1.43 | (0.57-3.57) | 0.443 | 0.95 | (0.30-2.99) | 0.934 |
high | 1.31 | (0.39-4.39) | 0.658 | 0.66 | (0.15-2.98) | 0.590 |
Household crowding | ||||||
<1 person/room | 1.00 | – | – | 1.00 | – | – |
≥1 person/room | 1.62 | (0.86-3.03) | 0.136 | 1.22 | (0.60-2.48) | 0.579 |
Smoking status | ||||||
never | 1.00 | – | – | 1.00 | – | – |
current | 4.05 | (1.13-14.5) | 0.031 | 5.19 | (1.29-20.86) | 0.020 |
former | 0.70 | (0.37-1.33) | 0.275 | 1.05 | (0.49-2.24) | 0.904 |
Alcohool intake | ||||||
never | 1.00 | – | – | 1.00 | – | – |
current | 1.94 | (0.96-3.93) | 0.066 | 1.47 | (0.65-3.33) | 0.361 |
former | 0.50 | (0.24-1.07) | 0.075 | 0.45 | (0.18-1.10) | 0.079 |
Coffee drinking | ||||||
no | 1.00 | – | – | 1.00 | – | – |
yes | 1.42 | (0.57-3.51) | 0.453 | 1.30 | (0.45-3.81) | 0.630 |
Water supply | ||||||
mineral water | 1.00 | – | – | 1.00 | – | – |
municipal | 2.68 | (1.38-5.21) | 0.004 | 2.20 | (1.02-4.72) | 0.044 |
others | 3.52 | (1.32-9.44) | 0.012 | 3.87 | (1.30-11.47) | 0.015 |
Hand washing | ||||||
yes | 1.00 | – | – | 1.00 | – | – |
no | 2.02 | (1.01-4.05) | 0.048 | 1.80 | (0.81-3.99) | 0.150 |
Cup sharing | ||||||
no | 1.00 | |||||
yes | 1.33 | (0.73-2.42) | 0.356 | 1.17 | (0.57-2.40) | 0.664 |
OR: odds ratio; 95% CI: confidence interval 95%.*The multiple logistic regression model included all variables mentioned in Table 3 as covariates.
Helicobacter pylori | |||
---|---|---|---|
Clinical findings | negative (n=80) | positive (n=120) | p value* |
Symptoms, n (%) | |||
nausea | 20 (25.0) | 34 (28.3) | 0.362 |
vomiting | 11 (13.8) | 13 (10.8) | 0.341 |
heartburn | 49 (61.3) | 83 (69.2) | 0.157 |
epigastric pain | 57 (71.3) | 92 (76.7) | 0.243 |
loss of appetite | 12 (15.0) | 16 (13.4) | 0.447 |
early satiety | 6 (7.5) | 9 (7.5) | 0.602 |
belching | 11 (13.8) | 29 (24.2) | 0.050 |
regurgitation | 5 (6.3) | 12 (10.0) | 0.254 |
hiccups | 7 (8.8) | 11 (11.2) | 0.565 |
fullness | 36 (45.0) | 52 (43.4) | 0.465 |
globus | 10 (12.5) | 15 (12.5) | 0.582 |
Family history, n (%) | |||
peptic ulcer | 20 (25.0) | 28 (23.3) | 0.457 |
gastric cancer | 12 (15.0) | 16 (13.3) | 0.447 |
*Each symptom was compared in a 2×2 table using one-sided Fisher’s exact test.
Helicobacter pylori* | |||
---|---|---|---|
Microscopic findings, n (%) | negative (n=80) | positive (n=120) | p value** |
intestinal metaplasia | 12 (15.0) | 10 (8.3) | 0.107 |
gastric atrophy | 9 (11.3) | 24 (20.0) | 0.073 |
mild gastritis | 38 (47.5) | 18 (15.0) | <0.001 |
moderate gastritis | 24 (30.0) | 67 (55.8) | <0.001 |
severe gastritis | 2 (2.5) | 33 (27.5) | <0.001 |
*Participants who had multiple symptoms. **Each finding was compared in a 2×2 table using Fisher’s exact test.
DISCUSSION
This is the first study conducted in the State of Paraíba, Northeast Brazil that assessed the effect of sociodemographic factors on the development of H. pylori infection. This pathogen affects half of the world’s population, and it exceeds 80% in developing countries1. The prevalence rate of H. pylori infection was higher in the population who used non-treated water. These results are consistent with those of studies conducted in other countries in South America and in the other regions of Brazil1. Although Brazil has the eighth largest economy, it still has epidemiological characteristics similar to those of developing countries, where the prevalence rate of H. pylori infection is higher than 80%. Zaterka et al. assessed 1,008 blood donors in São Paulo, and their results showed that the prevalence of H. pylori infection was also high, regardless of sex18.
H. pylori infection was most commonly observed in adults with an average age of 40 years. The prevalence of the infection among elderly individuals in Brazil is usually high, which is greater than 68% of the cases. Our results are in accordance with those of other studies. Gasbarrini et al. assessed health professionals in Rome, Italy and found that individuals who are older and those with fathers with fewer years of education were more likely to be H. pylori positive19. Another study conducted in Kuwait revealed a higher prevalence in women younger than 30 years20.
Income level, educational level, and household crowding were similar in our study population. The association between an increased risk of H. pylori infection and coffee consumption was not statistically significant. H. pyloriinfection was more common among smokers and drinkers, regardless of whether they were light or heavy consumers. Smoking has been considered a risk factor of H. pylori infection18,21,22. The association between H. pylori infection and smoking has been attributed to socioeconomic conditions. Furthermore, smoking is a major risk factor for peptic ulcer disease because it stimulates gastric acid secretion and compromises the mucosal barrier, resulting in gastric atrophy and potentially leading to the development of cancer23. The association between two or more risk factors and H. pylori infection in the same patient also increases the risk of ulcer development24. Chen et al. showed that smoking was associated with an increased risk of duodenal ulcer in patients with H. pylori infection25. Infection stimulates the production of pepsinogen and decreases mucous production, leading to more pathology. H. pylori infection was common among drinkers. However, no association was found after controlling for confounders, as observed in a previous study26.
In our study, individuals who consumed non-treated water and did not wash their hands before meals were more likely to acquire the infection. Sharing of cups among family members was not associated with the development of the infection. H. pylori is considered to be one of the microbiological contaminants of water. The viability of the bacteria is dependent on temperature conditions, which are ideal in the tropics, and this strengthens the assumption that contaminated water can be a natural source of infection4,27. The presence of H. pylori in non-treated water samples has been reported in other studies10. Campina Grande ranked 29th in water quality in Brazil, and it still does not have complete sewage coverage. However, because of frequent water shortage, individuals are more likely to store water at home, and the risk of contamination becomes higher. Hand washing has contributed to a reduction in infections, particularly in the gastrointestinal tract18,28.
Belching was associated with H. pylori infection, although it has a borderline p value. Dyspepsia, which is defined as pain or discomfort in the upper abdomen, is the main clinical manifestation in most patients. Usually, this symptom causes participants to seek medical help, and these individuals are then required to undergo endoscopy11. Moreover, heartburn and epigastric pain were symptoms commonly associated with the persistence of H. pylori infection in human immunodeficiency virus-positive individuals29. Inflammation of the gastric mucosa increased with exposure to the bacteria. We found more H. pylori-positive patients with chronic gastritis, and this result is in accordance with the findings showing that chronic infection increases the severity of gastritis30.
In conclusion, the prevalence rate of H. pylori was high in patients dyspeptic disorders and was associated with the severity of gastritis, and this increased the awareness of the need to improve the city’s non-treated water as a primary intervention to decrease H. pylori infection. PCR is a sensitive and specific diagnostic method that can be implemented in pathology laboratories and can be used as a confirmatory assay for H. pylori infection. In addition, it can help identify the genotype strains of H. pylori that contain the cagA+ gene associated with neoplastic development13.