 Kurdistan Journal of Applied Research (KJAR) Print-ISSN: 2411-7684 | Electronic-ISSN: 2411-7706 Special Issue: 2nd International Conference on the Health & Medical Sciences | DOI: 10.24017/science.2018.2.29 Received:1 June 2018 | Accepted: 12 July 2018 Evaluation the Efficacy of ELISA IgG, IgM and IgA Tests for Diagnosis of Helicobacter pylori Hogir Mohammed Shukri Saadi Ali Yahya Saeed Department of Biology Department of Biology College of Science College of Science University of Duhok University of Duhok Duhok, Iraq Duhok, Iraq Dna_proten@yahoo.com ali.saeed@uod.ac 172 Abstract: It has been well recognized throughout the world that Helicobacter pylori is the main cause of gastric ulcer and stomach carcinoma. Laboratory diagnosis of H. pylori infection is made by invasive and non-invasive methods. Invasive methods require endoscopy which is uncomfortable an unacceptable by the most patients. Therefore, non-invasive methods particularly serological tests are easier and comfortable for patients. A total of 86 patients with ages ranging from 18-77 years old (43 males and 43 females) who were referred to the Duhok Hepatology & Gastroenterology center/Azadi Teaching Hospital for endoscopic examination from June to October, 2013 were enrolled in the study. From each patient 5 ml of blood was collected under a septic condition and sera were separated for serology. Data from each participant were recorded in a special questionnaire form after consent agreed upon on ethical and scientific committee of the Azadi hospital. The efficacy of three different ELISA tests (IgG, IgM, IgA) was assessed taking culture as a gold standard method. H. pylori was found in 70.93%, 30.23% and 5.81% by ELISA IgG, IgA and IgM tests respectively. The highest percentage (70.93%) of H. pylori positive cases were found by ELISA IgG and the lowest percentage (5.81%) was recorded in ELISA IgM. The study found a significant relationship between age groups and H. pylori positive cases by ELISA IgG in which positive cases increased with the increase of ages. No statistical correlation was found between the sex, smoking status and residency of studied patients with H. pylori infection. A significant correlation was found between patients without endoscopic lesions and positive cases for H. pylori by ELISA IgG in which 75.67% of patients without endoscopic lesions were positive for anti-H. pylori antibodies compared to 41.66% in patients with endoscopic lesions. The efficacy of IgG, IgM and IgA tests was calculated compared with golden standard tests and ELISA IgG characterized by the highest values of sensitivity (75.75%), negative predictive value (36.00%) and accuracy (86.00%) but with lowest specificity (45.00%) and similar positive predictive value (81.96%) with IgM and IgA. The lowest sensitivity was 5.97% and accuracy 25.56% with ELISA Ig M. From the results of the present study we can conclude that among ELISA tests, anti-H. pylori IgG is a reliable serological test for the diagnosis of H. pylori infection. Keywords: ELISA, Antibody, Evaluation, H. pylori 1. INTRODUCTION Since the discovery of H. pylori by Barry Marshall and Robin Warren from Perth, Western Australia in 1983, a pivotal change has occurred in the overview of both microbiologists and clinicians toward peptic ulcer etiology and treatment strategies. Before this discovery, microbiologists believed that no microorganisms could thrive in such harsh destructive gastric environment and physicians also believed that peptic ulcer is mainly due to the overproduction of gastric acids against which treatment was directed to suppress acid production. Nowadays, it is well recognized that H. pylori is the principal cause of gastric ulcer and the main hazard factor for the development of gastric carcinoma and mucous associated lymphoid tissue lymphoma. Clinicians started to change their treatment strategy from acid suppressing drugs to both anti-acid and antimicrobial drugs regime ]1[. H. pylori is a Gram negative, curved bacilli with 4-6 unipolar sheathed flagella. It grows slowly. It is also fastidious in its growth requirements in that needs special media supplemented with antimicrobial agents and provided with micro aerobic conditions (5% O2, 5-10% CO2, 85% N2) ]2[. H. pylori was first named Campylobacter like organism, later named Campylobacter pylori. Based on the DNA sequence, H. pylori was placed under new genus of Helicobacter ]3[. The distribution of H. pylori infection varies among developed and developing nation in which 70-90% of the people living in developing nations are colonized by H. pylori compared to 50% in those living in developed countries ]4[. Colonization with H. pylori varies even among populations of the same country according to age, socioeconomic status, hygienic practice, geographical area and overcrowding conditions ]5[. Laboratory identification of H. pylori infection is made by several different techniques including both invasive and non-invasive method. Invasive methods are based on gastric biopsies taken during endoscopy like culture, histopathology and rapid urease test (RUT). Invasive methods are costly and inconvenient for most patients 173 they might also give false negative results, particularly in patchy colonized and treated patients. Furthermore, these methods are not suitable as a screening test for large scale of population studies. Therefore, non-invasive methods particularly serological tests remain the alternative choice for screening the sero-prevalence of H. pylori in large populations during epidemiological studies ]6[. Serological tests like ELISA may be superior to invasive methods in cases with gastric atrophy in which the number of microorganisms is so small to be undetectable by invasive methods. Serological tests also reveal antibodies generated by intracellular survival H. pylori after treatment ]7[. 2. LITERATURE REVIEW Among serological tests, Enzyme-linked immunosorbent assay (ELISA) is the most commonly used for detecting specific antibodies in the serum, as it is easy to perform, in-expensive, widely available, and suitable also for large- scale screening ]1[. H. pylori ELISA kit (ACON Lab, USA) was used by [17] for detecting anti-H. pylori IgG antibodies in the serum samples collected from patients with gastritis. They found that 81% of 58 patients were positive. H. pylori IgG and IgA ELISA kits (Germany) was used by [14] and found that 41(58.5%) and 48 (68.5%) from 70 studied patients were positive by IgG and IgA, respectively. The efficacy of ELISA IgG for the detection anti- H. pylori antibodies was assessed by [27] and found that 119 (39.1%) out of 304 tested serum samples were positive with a sensitivity of 99% and specificity of 82%.Another study by ]24[ found that 487 (69.0%) out of 706 tested serum samples were positive for anti-H. pylori IgG by ELISA technique (Orgenics® Immunocomb II), while ]12[ found 85 (73.91%) from 115 patients were positive for anti-H. pylori IgG antibody titers by ELISA IgG kit. They found that the test had high sensitivity (100%) but with low specificity (55.3%) in comparison to urease test, culture, histopathology and leukostix test. A study conducted by [13] in Iran using ELISA IgG, IgM and IgA kits (IBL International GMBH) for detecting seroprevalence of H. pylori in 339 patients (114 male and 225 female) who suffered from gastric symptoms from northwest of Iran. They found that the overall seropositivity rates were as follows: anti H. pylori IgG 73%; anti H. pylori IgM 43% and anti H. pylori IgA 25%. In a study carried out by ]16[ and used ELISA IgG, IgM and IgA (Monobind Inc., USA) for serological distribution of anti- H. pylori Ab in population of Sulaimani governorate/Iraq. A total of 335 adult and children volunteers from Chamchamal and Sulaimani cities in Kurdistan region/ Iraq were tested and found that 20.4% had positive IgM, 32.3% positive IgG and 58.2% positive IgA. ELISA IgG, IgM and IgA kits (NovaLisa, NovaTec, Germany) used by [15] to screen the seroprevalence of H. pylori among asymptomatic healthy Omani blood donated individuals. Among 133 healthy asymptomatic individuals, IgG H. pylori antibody was found in 83 (62.4%), IgM H. pylori antibody in 21 (15.1%) and IgA H. pylori antibody in 11 (8.7%). In another study by [11] using ELISA kits for IgG, IgA and IgM and found that 68 from 102 patients were positive .Out of these, IgG type was positive in 49 (72%), IgA in 56 (82.3%), and IgM in 25 (36.8%). 3. MATERIALS AND METHODS Serological tests Each serum sample was analyzed by ELISA IgG, IgM and IgA kits (Monobind Inc., USA) following the procedure supplied by kits. Statistical analysis Data was statistically analyzed using (statistical analysis system) SAS software (Version, 2010) by Log linear model. The first Model: Yij = µ+Ai+Eij used for analyzing the effect of age, smoking, gender and resident on Serological tests (IgG, IgM and IgA). Yij = the studied test µ =over all mean Ai = the independent factor Eij = random error The second Model: Yijk=µ+Ai+Bi+(A٭B)ij+Eijk used for analyzing the effect of Culture on Serological test(IgG, IgM and IgA). Yijk = the studied serological test. µ = over all mean Ai = the effect of first standard Bi = the effect of second standard (A٭B)ij = the interaction of both standard Eijk= random error (SAS, LTD, 2010) 4. RESULTS Three different ELISA kits (IgG, IgM and IgA) were used for the detection of H. pylori infection in 86 patients using a culture method as a gold stander method. H. pylori was detected in 37.2%, 70.93%, 5.81% and 30.23% by ELISA IgG, ELISA IgM and ELISA IgA respectively compared to 37.2% by cultural method as shown in Table 1. The maximum percentage of H. pylori was recorded with ELISA IgG (70.93%) followed by ELISA IgA (30.23%), while the lowest percentage of H. pylori was recorded with ELISA IgM test (5.81%). Table 1. Serological tests and cultural methods used for detection of H. pylori. Tests Positive for H. pylori Negative for H. Pylori number % number % IgG 61 70.93 25 29.06 IgM 5 5.81 81 94.18 IgA 26 30.23 60 69.76 Culture 32 37.2 54 62.79 A high percentage of H. pylori infection was found with increased ages with ELISA IgG in which a high percentage 88.23% was found in individuals with the age group 41-50 years old compared to 50% in the age group less than 20 years old. ELISA IgA recorded high percentages of positive cases 70% in young age less than 20 years old compared to low percentages with increasing 174 ages, while in ELISA IgM, the picture was irregular in which only 11.53% and 12.5% of cases were positive for H. pylori in age 20-30 and 31-40, respectively, as shown in Figure 1. An important correlation was found between ages and H. pylori positive cases with ELISA IgG in which positive cases increased with the increase of age but no significant correlation was detected between age groups ELISA IgM and IgA tests for H. pylori infection. Figure1. Percentages of H. pylori infection in age groups using ELISA techniques. H. pylori was detected in 34 (74.06%), 3 (6.97%) and 15 (34.88%) in female patients by ELISA IgG, IgM and IgA respectively. Among male patients, H. pylori was detected in 27 (62.79%), 2 (4.65%) and 11 (25.58%) by ELISA IgG, IgM and IgA, respectively, as shown in Figure 2. No significant correlation was found between sex and ELISA (IgG, IgM and IgA) methods for screening H. pylori. Figure 2. Percentages of H. pylori infection in sex groups by ELISA IgG, IgM and IgA tests. H. pylori was detected in 13 (65%), 1(5%) and 2 (10%) smoker patients by ELISA IgG, IgM and IgA, respectively, while 14 (21.21%, 0 (0%) and 8 (12.12%) were positive for H. pylori in non-smoker patients by ELISA IgG, IgM and IgA respectively, as shown in figure 3. No significant correlation was found between smoking and ELISA techniques. Figure 3. Percentages of H. pylori infection in smokers and non-smokers patients by ELISA IgG, IgM and IgA tests. H. pylori was found in 33 (60%), 2(4%) and 14 (28%) in patients lived in the urban areas by ELISA IgG, IgM and IgA, respectively, while 28 (77.77%), 3(8.33%) and 12 (33.33%) were positive for anti- H. pylori antibodies in patients who lived in rural areas by ELISA IgG, IgM and IgA respectively, as shown in Figure 4. No significant correlation was found between residence and ELISA tests. Figure 4. Percentages of H. pylori infection among resident groups using ELISA techniques. The sensitivity, specificity, positive predictive value, negative predictive value and accuracy of ELISA IgG, IgM and IgA were calculated using culture as golden standard methods. Among 86 patients, 61 were positive and 25 were negative by ELISA IgG test in which the true positive, false positive, true negative and false negative results were 50, 11, 9 and 16 cases, respectively. Only 5 cases were positive and 81 cases were negative by ELISA IgM test in which the true positive, false positive, true negative and false negative results were 4, 1, 18 and 63 cases, respectively. By ELISA IgA test 26 cases were positive while 60 cases were negative in which true positive, false positive, true negative and false negative results were 22, 4, 16 and 44 cases, respectively, as shown in Table 2. 0 20 40 60 80 100 ≤ 20 20-30 31-40 41-50 51-60 ≥ 60 P e rc e n ta g e ( % ) Age groups IgG IgM IgA 0 10 20 30 40 50 60 70 smoker non smoker P e rc e n ta g e % Smoking IgG IgM IgA 175 Table 2. True positive, false positive, true negative and false negative results of ELISA IgG, IgM and IgA tests. 5. DISCUSSION Among serological tests, Enzyme-linked immunosorbent assay (ELISA) is the most commonly used for detecting specific antibodies in the serum, as it is easy to perform, in-expensive, widely available, and suitable also for large- scale screening [1]. The results of ELISA IgG were in agreement with those recorded by other researchers [8, 9, 10, 11, 12, 13] who found 76%, 69%, 73.68%, 73.9%, and 73% respectively. While the results were in disagreement with the results of ]14,15,16,17] who found 58.5%, 62.4%, 32.3% and 81%. The causes of these dissimilar results may be attributed to the kind of captured antigens used for detection of IgG, geographical distribution of the H. pylori infection and stage of infection. The ELISA IgA results were similar to those found by ]13[, who found that 25% of patients were positive for H. pylori infection, while dissimilar to that found by ]14,11,15,16], who found 68.5%, 82.3%, 8.7% and 58.2% respectively. This dissimilarity in the results can be attributed to the same factors mentioned in ELISA IgG. The results of ELISA IgM were dissimilar to those found by most studies, such as ]1,11,13,15,16] who found 44% ,43%, 36.8%, 20.4%, 15.1%, respectively. This dissimilarity of the results may be due to the low number of samples covered by the study or the patients were in the chronic stages of the infection. Positive cases of H. pylori increased with the increase in age by ELISA IgG in which a high percentage (88.23%) of positive cases was in the age group 41-50 years old patients compared to 50% in the age group less than 20 years old (Figure 1). A significant relation was found between H. pylori positive cases and the increase of ages. The picture was different with the results of ELISA IgA in which a high percentage (70%) of positive cases was found in young patients under 20 years old compared to 40% in older patients more than 60 years old, but did not show significant correlation. The results were in disagreement with those of ]13[, who found that H. pylori seropositivity for IgA increased with the increase in age. The pattern of positive cases by ELISA IgM was not significant and irregular in which no positive cases were found in the age group less than 20 years old, 41-50, 51-60 and above 60 years old patients. The findings of the study were similar to those found by ]18,16,19, 20, 13, 21[ who found a significant effect of age on the levels of H. pylori positive cases in which high percentages of positive cases for anti-H. pylori IgG antibodies were increased with the increase in ages, these results in disagreement with those of ]12[ who found no correlation between H. pylori seropositivity and age. This may be due to the chances of acquisition of infection which tend to increase with age. The results of ELISA IgM were inconsistent and not significant this may be due to several factors, such as low number of patients tested because of the cost constraint, the nature of the antigen used in the kit or the patients were in the chronic stage of infection and IgM antibodies appeared in the early stage of infection then decline. No significant correlation was detected between sex and (IgG, IgM and IgA ELISA`s Tests) for finding of positive cases of H. pylori. The results were similar to those found by ]1 [ for IgM, ]23, 22, 24[ for IgG and ]16[ for IgG, IgM and IgA, while in disagreement with the results of ]15[ who found that males were more infected than females by ELISA IgG, IgM and IgA. No significant correlation was found between the smoking factor and seroprevalence of H. pylori. The results were dissimilar to those found by ]18,21,10[, who found a significant effect of smoking on the prevalence of H. pylori. This discrepancy in the results could be contributed to the number of analyzed samples, Methodology, socio-economic status and geographical location. In this study, no statistical correlation was found between residence and H. pylori seropositivity by serology IgG, IgM and IgA tests. The results were comparable to the results of ]22[ and ]19[ who found no significant impact of residence on the seropositivity of H. pylori by ELISA techniques. The sensitivity, specificity, positive predictive value, negative predictive value and accuracy of ELISA IgG were 75.75%, 45.00%, 81.96%, 36.00% and 68.60% respectively, for ELISA IgM were 5.97%, 94.73%, 80.00%, 22.22% and 25.58%, respectively and for ELISA IgA were 33.33%, 80.00%, 84.61%, 26.60% and 44.18% respectively, as shown in Table 2. In the present study, ELISA IgG was characterized by high sensitivity, high negative predictive value and high accuracy in comparison to both IgM and IgA. Lowest sensitivity, lowest positive predictive value, lowest negative predictive value and lowest accuracy were found with IgM. The results were analogous to those reported by ]14,25], but generally the results of this study were lower than those found by ]26,27]. The lower results of the present study may be due to inadequate number of analyzed serum samples because of the cost constraints and the antigenicity of the antigens used in the kits. 4. CONCLUSION 1. ELISA IgG was superior to ELISA IgM and IgA for the detection of anti-H. pylori antibodies. 2. ELISA IgG was characterized by high sensitivity, negative predictive value and accuracy but with lower specificity. 3. ELISA IgM was characterized by the highest specificity but with the lowest sensitivity. 4. ELISA IgA was characterized by moderate sensitivity, specificity, positive predictive value, negative predictive value and accuracy. 5. No significant effect of smoking and residence was found on H. pylori infection. 6. A significant correlation was found between age and H. pylori infection by ELISA IgG in which H. Pylori infection increased with the increase in age. Test +ve True +ve False +ve -ve Tru e – ve False -ve IgG 61 50 11 25 9 16 IgM 5 4 1 81 18 63 IgA 26 22 4 60 16 44 176 REFERENCE [1] R. M. Abu-Mughesieb, "Risk Factors Associated with Helicobacter pylori Infection in Gaza, Palestine,” M. Sc. Thesis The Islamic University-Gaza, Deanship of Graduate Studies, Biological Sciences Master Program, Medical Technology, Facutly of Science," 2007 [2] M. Arshad, M. Akram, U. Shahab, A. Afzal, U. Khan, H. Abdul, and A. M. Mohiuddin, "Helicobacter pylori: An introduction," International Journal of Applied Biology and Pharmaceutical Technology, vol.1, pp. 1337-1351, 2010 [3] V. Gabriel, "Microbiology and Infectious disease,"1997 3rd ed. williams and wilkins U. S. A. [4] A. S. Barik, "Helicobacter pylori Infection in Developing Countries: The Burden for How Long?," Saudi Journal of Gastroenterology,vol. 15(3), pp. 201–207, 2009. [5] M. Amini, A. Karbasi, and H. Khedmat, "Evaluation of eating habits in dyspeptic patients with or without Helicobacter pylori infection," Journal of Tropical Gastroenterology, vol. 30(3) , pp.142-144, 2009. [6] C.P. Dooley, H. Cohen , P. L. Fitzgibbons, M. Bauer, M.D. Appleman, G. I. Perez-Perez, and M. J. Blaser, "Prevalence of Helicobacter pylori infection and histologic gastritis in asymptomatic persons," N England Journal Medical ,vol. 1(23), pp. 1562-1566, 1989. [7] N. Parimala, and M. Ishaq , "Efficacy of sonicated and acid- extractable antigens in the serodignosis of H. pylori infection in peptic ulcer patients," India Journal Medical Microbial, vol. 23 (2), pp. 117-119, 2005. [8] A. M. S. Ibrahim, and M. K. Turab, "The effect of propolis on growth inhibition of Helicobacter pylori isolates from peptic ulcer patient," Kufa Journal For Veterinary Medical Sciences, vol.2(1), pp.44-58, 2011 [9] R. Johaneesssen, K. Bergh, C. Jianu, and P. M. Klevel, "Polymerase chain reaction versus culture diagnosis of Helicobacter pylori infection," Gastroenterology Insights, vol. 5 (1), pp1-6,2013. [10] N. M. Kaore, V. N. Nagdeo, and V. R. Thombare, "Comparative evaluation of the diagnostic tests for Helicobacter pylori and dietary influence for its Acquisition in Dyspeptic patients: A rural hospital based study in central Indi,". Journal of clinical and diagnostic Research, vol. 6 (4), pp. 636-641, 2012. [11] S. Shukla, M. Pujani, A. Agarwal, and A. Rohtagi, "Correlation of Serology with Morphological Changes in Gastric Biopsy in Helicobacter Pylori Infection and Evaluation of Immunohistochemistry for H. Pylori Identification," The Saudi Journal of Gastroenterology, vol.18(6), pp. 369-374, 2012 [12] N. E. Mohsun, R. H. Al-Hadithi, and S. Saadallah, " The Role of ELISA test in the diagnosis of helicobacter pylori infection," Faculty Medicine Baghdad, vol.53(2), pp.311-313, 2011. [13] S. Montazer-Saheb, S. Farajnia, N. Saeedi, R . Yousefzadeh, A. Rafat, and L. Rahbarnia, "Seroprevalence of Helicobacter pylori infection in patients suffering from gastric symptoms in the Northwest of Iran," African Journal of Microbiology Research, vol. 5 (22) , pp. 3616-3619, 2011 [14] M. Sharma, P. Mehta, and P. Vohra, "Comparative Evaluation of Different Diagnostic Techniques Available for Diagnosis of Helicobacter Pylori," International Journal of Scientific and Research Publications,vol.2, pp.1-5, 2012 [15] M. S. Al-Balushi , J. Z. Al-Busaidi , M. S. Al-Daihani , O. Sh. Mohammed, and S. H. Sidgi, " Sero-prevalence of Helicobacter pylori infection among asymptomatic healthy Omani blood donors," Asian Pacific Journal of Tropical Disease,vol.3(2), pp.146-149, 2013 [16] A. Al-Windi, H. H. Ali, and S. Narmin, "Seroprevalence of anti- Helicobacter pylori antibodies in population of Sulaimani governorate / Kurdistan Region / Iraq," Journal of Zankoy Sulaimani, vol.15(3), pp.175-185, 2013 [17] I. Alsaimary, M. Al-Saadon, A. Jassim, and S. Hammadi, "Clinical Findings and Prevalence of Helicobacter Pylori in Patients with Gastritis B in Al-Basrah Governorate," Oman Medical Journal, vol.24(3), pp. 208-211, 2009 [18] H. I. Baqir, M. A. Assam, A. S. Al-Bana, and H. M. Al-Aubaidi, "Sere-prevalence of Helicobacter pylori infection in unselected adult population in Iraq," International Journal of Global Education. 1(3), pp.22-29, 2002. [19] M. Feleke, K. Afework, M. Getahun, A. Solomon, A. Berhanu, N. Takeshi, and O. Fusao, "Seroprevalence of Helicobacter pylori in dyspeptic patients and its relationship with HIV infection, ABO blood groups and life style in a university hospital, Northwest Ethiopia" World Journal of Gastroenterology,vol.12(12), pp.1957-1961, 2006. [20] A.H.M. Alizadeh, S. Ansari, M. Ranjbar, H. M. Shalmani, I. Habibi , M. Firouzi, and M. R. Zali, "Seroprevalence of Helicobacter pylori in Nahavand: a population-based study," Eastern Mediterranean Health Journal, vol.15(1), pp.129-135, 2009. [21] M. A. Khan, and H.O. Ghazi, "Helicobacter pylori infection in asymptomatic subjects in Makkah, Saudi Arabi," Journal Pakistan medical Association, vol. 57 (3), pp.114-117, 2007. [22] S. M. Alavi, S.M. H. Adel, and A. Rajabzadeh, "Seroprevalence study of Helicobacter pylori infection among visitors of cardiac patients in Razi hospital in Ahvaz, Iran," Jundishapur Journal of Microbiology, vol. 3(1), pp.28-31, 2010. [23] F. Megarud, "Comparison of non-invasive tests to detect Helicobacter pylori infection in children and adolescents: Results of Multicenter European study," Journal Pediatr, vol 146(2), pp.198-203,2005. [24] A. Alim, M. Ataş, T. Güne, S. Özkan, and N. Dündar, " Comparison of antigen and antibody detection tests used for diagnosing the Helicobacter pylori infection in symptomatic patients," Basic and Clinical Sciences, vol. 1(4). pp. 61-70, 2010. [25] Sh. Kazemi, H. Tavakkoli, M. R. Habizadeh, and M. H. Emami, "Diagnostic values of Helicobacter pylori diagnostic tests: stool antigen test, urea breath test, rapid urease test, serology and histology," Journal of Research in Medical Science, vol.16(9), pp.1097-1104, 2011. [26] G.A. Rocha, A.M.R. Oliveira, D.M.M. Queiroz, E.N. Mendes, S.B. Moura, C.A. Oliveira, and T.C. A. Ferrari, "Serodiagnosis of Helicobacter pylori infection by Cobas Core ELISA in adults from Minas Gerais, Brazil," Brazilian Journal of Medical and Biological Research, vol.31, pp.1263-1268, 2011. [27] S. Redéen, F. Petersson, E. Törnkrantz,; H. Levander, E . Mårdh, and K. Borch, " Reliability of Diagnostic Tests for Helicobacter pylori Infection," Gastroenterology Research and Practice, pp.1-6, 2011.