Diagnosis of Helicobacter pylori infection using immunochromatography among patients attending Tamboul Hospital in Gezira State, Sudan: a cross-sectional study [version 1; peer review: 1 not approved]

Background: Helicobacter pylori causes a major health problem worldwide; more than half of the world’s population are infected with this pathogen. The diagnosis of the infection was initially made through invasive methods, but now non-invasive methods have been developed to make diagnosis easier. This study aimed to screen the presence of H.pylori antibodies and antigen among symptomatic and asymptomatic patients at Tamboul City in Gezira State. Methods: A cross-sectional study was conducted in Tamboul city, Gezira State, Sudan between March 2016 and December 2019 to compare between antigen and antibody tests results used for diagnosis of H. pylori infection among symptomatic and asymptomatic Sudanese patients. Stool and blood samples were collected and analyzed for presence of antigen and antibodies to H. pylori using immunochromatography (ICT) cards. Results: Serum and stool samples were collected from 100 patients; 50 were symptomatic and 50 were asymptomatic. In symptomatic patients, 18/50 (36%) were men (32; 64%, women) with mean age of 16.7±24.6 years. In this group, 35/50 (70%) showed positive results for stool antigen, while 30/50 (60%) were positive for serum antibodies. In asymptomatic patients, 19/50 (38%) were men (31; 62%, women) with mean age of 16.7±20.4 years. In this group, 18/50 (36%) were positive for stool antigen and 25/50 (50%) for serum antibodies. There was a significant association between antigen results and patient group (P=0.001), but there was an insignificant association between Open Peer Review


Introduction
Helicobacter pylori is a Gram-negative, microaerophilic, spiral-shaped bacteria that inhabits the mucous layer of the antral gastric (Basher et al., 2011). Humans are living things infected (Abdallah et al., 2014). H. pylori is typically symptomless and has no clinical manifestations. Signs and symptoms associated with the disease are mainly due to gastric or peptic ulcer disease or duodenal inflammation. Other symptoms, such as nausea, vomiting, and stomach pain, can be related to other gastrointestinal diseases (Gold et al., 2014). The prevalence of H. pylori is still high in most countries, with approximately onethird of adults infected in northern Europe and North America, while the prevalence in southern and eastern Europe, South America and Asia is higher than 50% (Eusebi et al., 2014). Up to 93.6% of adults in developing countries are infected with H. pylori (Olokoba et al., 2013) and the prevalence of infection in Sudan was estimated to be between 65.8% and 80% (Abdallah et al., 2014).
The attachment of H. pylori to the gastric mucosa is mediated by the Lewis (Le6) antigen. The availability of these receptors may be reduced in blood groups A and B compared to blood group O. H antigen is also an important gastroduodenal mucosal cell receptor for H. pylori attachment (Alcout et al., 2000).
Laboratory testing for H. pylori infection is very important part of the diagnosis process for gastric and duodenal inflammatory disease. A number of different diagnostic test methods, both invasive and non-invasive, are available. Invasive methods, such as antibody (Ab) detection is cheap and easy to perform, but may give false positive results (Quartero et al., 2000). The stool antigen (Ag) test has recently become more acceptable as it is non-invasive, convenient for patients and can be performed easily even in small laboratories. Other non-invasive tests may also include detection of H. pylori antigens in saliva (Logan & Walker, 2001;Malfertheiner et al., 2007).
This study aimed to screen the presence of H.pylori antibodies and antigen among symptomatic and asymptomatic patients at Tamboul City in Gezira State.

Study design
This descriptive, cross-sectional, hospital-based study was conducted at Tamboul Hospital in Gezira State, Sudan, between March 2016 and December 2019.

Ethical considerations
Ethical approval was obtained from the Scientific Research Committee of the College of Medical Laboratory Sciences, Sudan University of Science and Technology . Written informed consent was initially obtained from participants for participation in the study and publication of data; however, the researchers noticed that participants preferred to agree verbally. After requesting permission from the Scientific Research Committee to change the consent process, verbal informed consent was obtained from participants for participation in the study and publication of data.

Patients
Patients were selected by non-probability, convenience sampling and were recruited by physicians.
Patients with and other without symptoms of H. pylori infection with different age and gender were included in this research.
Exclusion criteria: symptomatic patients undergoing treatment of H. pylori infection and patients with other diseases.
A structured questionnaire containing demographic data and medical history was used to collect information from participants.

Sample size
Sample size was determined according to the formula below. Due to high cost, the sample size was kept at 100 sera and stool samples; 50 symptomatic patients and 50 asymptomatic patients.
Data collection Serum sample. Five ml blood sample were collected from each patient in a plain blood container. Serum was then separated by centrifugation (3000 rpm for five minutes). H. pylori antibodies (IgG and IgM) were detected in the serum using ALL TEST (Hangzho Biotics,China;. 2 drops of sample (about 100µl) were added into the sample well (S) of the test device and then 1 drop buffer was added to this. As the test began to work, a purple color moving across the window to the center of the device.
The presence of two-color bands as test band (T) and control band (C) within the results window, no matter which band appeared first, indicated a positive result. It should be noted that some invalid results are because the directions have been followed incorrectly or the test may have deteriorated beyond the expiration date (Ansorg et al., 1991).

Stool sample.
Patients were asked to provide a stool sample at the hospital. Fresh stool samples were collected into spooncover and outer-labeled stool container for antigen detection. Using a wood stick, a small portion of the stool sample was transferred into buffer, incubated for 2 minutes and then two to three drops of the mixture were poured in the hole of the ICT of H. pylori stool antigen detection (ALLTEST; Hangzho Biotics, China; catalog number 30226). The color migrated from the well containing the tested sample in the ICT device (Ansorg et al., 1991). The presence of two bands (test band (T) and control band (C)) within the result window, no matter which band appeared first, indicated a positive result.

Statistical analysis
Statistical Package for Social Science program (SPSS) version 20 was used for analysis. Frequencies were presented as tables and figures. Chi-square test was used to compare between variables and level of significance was set as P ≥ 0.05.

Results
A total of 100 stool and 100 sera specimens were collected from 50 asymptomatic and 50 symptomatic patients, with different gender, age and blood groups. Age was classified as following: <15 years, 15-50 years, and >50 years old.
In asymptomatic patients, the mean age was 16.7±24.6 years. The majority of the patients in this group were women (32; 64%), In symptomatic patients, the mean age was 16.7± 20.4 years, and the majority of patients were women (31; 62%).
For both groups of patients, those who had previous treatment for H. pylori had low frequency of infection compared to those with no previous treatment. For both groups, 60% had family members who had also been diagnosed with H. pylori infection (Table 1).
There was a higher number of symptomatic patients with a positive antigen result (35/50; 70%) compared with asymptomatic patients (18/50; 36%). There was a significant association between antigen result and patient group (P=0.001; Table 2).
Similarly, there was a higher number of symptomatic patients with a positive antibody result (30/50; 60%) compared to asymptomatic patients (25/50; 40%). There was no significant association between antibody results and patient group (P=0.317; Table 3).
Symptomatic patients in the 15-50 years age group had the highest number of positive results for antigen and antibodies (antigen, 27 (54%); antibody, 24 (48%)). In asymptomatic patients for this age group, a positive antigen and antibody result was seen in 17 (34%) and 23 (46%) patients, respectively. However, there was no association between age group and positive result for antigen and antibodies to H. pylori in either group of patients.
Patients with O positive blood group showed higher positive results than other blood groups for both antigen and antibody in both asymptomatic and symptomatic groups (asymptomatic O positive, Ab 14 (28%), Ag 17 (34%); symptomatic O positive, Ab 10 (20%), Ag 7 (14%)). The association between  blood group and antigen/antibody presence was not significant in either group of patients.
Patients in both groups with family members who also had H. pylori infection showed higher positive antigen results than those without family members who had the infection (symptomatic patients with family infection, 21 (42%), without family infection 14 (48%); asymptomatic patients with family infection, 12 (21%), without family infection 6 (12%). There was no significant association between antigen results and family member infection.
Asymptomatic patients who had no previous treatment for H. pylori showed higher antigen positivity (12/50; 24%) than those with previous treatment. This was similar in symptomatic patients (no previous treatment Ag, 25/50; 50%). There was no significant association between previous treatment and antigen result in asymptomatic and symptomatic patients.

Discussion
H. pylori infection is a co-factor in the development of three important upper gastrointestinal diseases: duodenal or gastric ulcers, gastric cancer and gastric mucosa-associated lymphoid tissue lymphoma (McColl, 2010).
In the present study, H. pylori antibodies were present in 25/50 (50%) asymptomatic patients and 30/50 (60%) symptomatic patients. These results were similar to a study carried by Naji et al. (2014) in Yemen, which found antibodies in 72%. In contrast, our results are higher than those found by Douraghi et al. (2013) in Iran, who found antibodies in 4.6% of patients, and lower than those found by Alfadil (2016) in Sudan who found antibodies in 48% of patients.
In our asymptomatic patients, H. pylori antigen was detected in 18/50 (36%) patients, and 35/50 (70%) symptomatic patients. These results are similar to a study carried out in Nigeria by Omosor et al. (2018), where stool antigen was found in 56% of patients. Our results for antigen were higher than those reported by Douraghi et al. (2013) in Iran (12.4%) and lower than Naji et al. (2014) in Yemen (49%). The variation may be due to difference in socioeconomic status of the patients.
Our study results did not find a significant association between age groups and H. pylori infection (as diagnosed with antigen/ antibody testing). This is similar to study done in Iran by Alvali et al. (2010), but it differed from Kabir (2007) in Sweden who reported that the percentage of infection increases with age. In our study, the age group 15-50 years showed the highest rate of infection in both symptomatic and asymptomatic patients for antigen and antibodies. That may be because the vast majority of individuals acquire H. pylori infection during childhood, and the infection associated with poor hygiene and overcrowding.
We also found that women were more affected than men both in the symptomatic group (62% vs 38%.) and asymptomatic group (64% vs 36%). This was a similar result to that found in Turkey by Kanbay et al. (2005) (60.6% vs 42.9%), but a dissimilar results to those found in Iran by Metanant et al. (2010) (32.8% vs 36.4%). The differences between studies could be due to random selection of patients regardless of their gender in our study.
In the current study, patients with blood group O positive had the highest frequency of positive H. pylori infection, both for symptomatic (48%) and asymptomatic (38%) patients. These results were the same as another study done in Gezira State, Sudan by Mohammed et al. (2015), who found that 58.1% of blood group O positive patients were positive for H. pylori infection. This is in contrast to a study done in Turkey by Kanbay et al. (2005) who was reported that patients with blood group A positive had a higher frequency of H. pylori infection than other blood groups at 38.2%. This dissimilarity may be due differences in ethnic background, and the fact that blood group O positive is the most frequent blood group in Sudan (Hassan, 2010).
In this study, 12/50 (24%) of asymptomatic patients who had family members with H. pylori infection were positive for antigen and/or antibody. These results are similar to Gravina et al. (2016) in Italy, who found H. pylori infection in 28.4% household contacts. However, this study was higher than Mabeku et al. (2018) in Cameroon (80.39%). In symptomatic patients, the frequency of antigen and antibodies with family member infection were 21(42%) and 20(40%) correspondingly, that were similar to result of Gravina et al. (2016) in Italy who found H. pylori infection in 43.2% household contacts. This finding was higher than Manfredi et al. (2016) in Italy (81.7%), and variation may be due to personal hygiene.
In the current study, the frequency of antigen in asymptomatic patients who had previous treatment for H. pylori were 12% (6/50) while in symptomatic patients this was 20% (10/50). This agreed with a study carried out by Gisbert et al. (2005), but was higher than the results reported by Nanivadekar et al. (1990) in India (2.4%), and lower than the results found by Bapat et al. (2000) in India (60%). This discrepancy may be due to resistance of standard triple therapy. In this study, the association between previous treatment and H. pylori reinfection was not statistically significant.

Conclusions
The frequency of H. pylori antigen was higher than antibodies in symptomatic patients, while the frequency of H. pylori antibodies was higher than antigen in asymptomatic patients.

Milagrosa Montes
Department of Microbiology, Donostia University Hospital-Biodonostia Institute, San Sebastián, Spain The authors describe two ICT methods to diagnose Helicobacter infection in two population groups: symptomatic and asymptomatic. I see two major errors. The first one, to use the detection of antibodies (Ab) to diagnose Helicobacter infection. Serology should only be used for epidemiological studies as it does not differentiate between current and past infections. The second failure that I observe is that speaking of non-invasive diagnostic methods to detect Helicobacter infection, they do not name the urea breath test (UBT), which, to date, is the non-invasive test with the highest sensitivity and specificity, and the one recommended by all Helicobacter infection diagnostic guidelines. My suggestion to the authors would be that they write another paper with the stool antigen data, saying that it is a good technique to detect Helicobacter infection when UBT is not available, it is logical that positivity is higher in symptomatic subjects (70 % vs 36%), and to use the serology data to give the seroprevalence in Gezira State; 55 patients have Ab (30 in the symptomatic group and 25 in the asymptomatic group) a figure lower than the 65-80% reported for Sudan by Abdallah et al.
They should further describe in detail the test used to detect Helicobacter Ag in feces, whether it uses monoclonal or polyclonal Ab. It would be good to give the data of each test in previously treated patients, it would give an idea of the treatment failures. The heading of Table 1, "Demographic variables for patients with Helicobacter pylori in patients from Gezira State, Sudan" is not correct, since the 100 patients of the study are included, those with and those without infection, it should say "Demographic variables of the 100 patients from Gezira State (Sudan) included in the study".

Is the work clearly and accurately presented and does it cite the current literature? Yes
Is the study design appropriate and is the work technically sound?