Helicobacter pylori infection is probably one of the most common chronic bacterial infections worldwide. The infection is acquired predominantly in childhood, and in most patients, its course is without complications. Nevertheless, in a small percentage of infected individuals, peptic ulcer disease (1), gastric carcinoma (2), or mucosa-associated lymphoid tissue lymphoma (3) develop. Once acquired, the infection persists for years and elicits mucosal and serum immune responses in most infected persons (4).
Differences in virulence among bacterial strains may explain the diverse possible outcomes of H. pylori infection (5). Among the putative virulence factors, the cag A gene, located in the cag pathogenicity island, has been shown to encode for a highly immunogenic outer-membrane protein of molecular mass 120 to 140 kd. It has been suggested that H. pylori strains that contain the cag pathogenicity islands are more virulent and are associated with complications such as peptic ulcer and distal gastric carcinoma (6). The cag pathogenicity island contains genes that encode proteins with similarities to components of secretion systems and seems to induce an increased inflammation in the gastric mucosa through release of interleukin 8 by epithelial cells (7).
Although enzyme-linked immunosorbent assays (ELISAs) have been considered sensitive and specific to establish CagA status in adults (8,9), we are unaware of studies validating commercial ELISAs for CagA in childhood. This would be important because there is evidence that ELISA has poor accuracy for the diagnosis of H. pylori infection in young children (10). Therefore, the aim of this study was to evaluate the accuracy of a commercial ELISA assay to detect anti-CagA antibodies in children compared with the gold standard detection of cag A by polymerase chain reaction (PCR) from biopsy-cultured bacterial strain or gastric tissue DNA.
PATIENTS AND METHODS
This project was approved by the Ethics Committee of Hospital das Clínicas, Universidade Federal de Minas Gerais, Minas Gerais, Brazil, and informed consent was obtained from children (whenever possible) and their parents.
Portions of the sera tested in the present study were from children included in a previous study for validation of a commercial ELISA for the diagnosis of H. pylori infection (10). We studied sera from 115 children (50 boys, 65 girls; mean age, 9.2 ± 3.4 years; range, 2–16 years) who underwent upper gastrointestinal endoscopy for evaluation of symptoms referable to the upper gastrointestinal tract such as recurrent abdominal pain, vomiting, or hematemesis. Exclusion criteria were age younger than 2 years; antimicrobial drugs during the 6 months before endoscopy; current H 2 -receptor antagonists or nonsteroidal antiinflammatory drugs; portal hypertension, coagulation disorders, or anatomic obstacles preventing endoscopy. Sixty-two children were H. pylori positive (21 with duodenal ulcer) and 53 were H. pylori negative.
At endoscopy, biopsy specimens were obtained from the antral and oxyntic gastric mucosa for culture, urease test, and histologic study (11). Fragments of the gastric mucosa were also obtained and frozen at −80°C for direct PCR if necessary. Children were considered to be H. pylori positive if the culture was positive or if both urease and histologic analysis results were positive. Results were considered inconclusive if either urease or histologic analysis results were positive.
All H. pylori strains isolated were evaluated for the presence of ure A and cag A. Tissue PCR for ure A and cag A was performed in two H. pylori-positive patients with negative culture. Two cag A-positive strains (ATCC 49503 and NCTC 11637) and one cag A-negative strain (Tx30A) were used as positive and negative controls, respectively, for cag A detection. The same H. pylori strains and two Proteus mirabilis and Escherichia coli human isolates were used as positive and negative controls in the reaction for ure A detection. A reagent-negative control reaction, in which DNA samples were replaced with distilled water, was performed with each batch of amplification to exclude contamination.
For cag A detection, genomic DNA from the bacterial strains or from the gastric tissues underwent PCR amplification using two sets of oligonucleotide primers previously described by Kelly et al. (12) and Peek et al. (13). The amplified PCR products were resolved in 1% agarose gels containing Tris/borate/edetic acid by using 100 bp as a molecular weight marker. The strains were considered to be cag A positive when at least one of the reactions was positive.
For detection of ure A, genomic DNA from the bacterial samples and gastric tissues was amplified by use of a set of synthetic oligonucleotide primers, as described by Clayton et al. (14).
Venous blood samples were drawn from each child at the time of endoscopy. The serum was separated, divided into aliquots, and stored at −20°C before testing. Sera were assayed for anti-CagA antibodies by a commercial Helicobacter p120, CagA ELISA kit (Viva Diagnostika, Hürth, Germany). Helicobacter p120 cagA is an assay used for the detection of immunoglobulin G (IgG) specific for p120 antigen. The antigen used in the kit is a highly purified protein of 120 kd, and the assay was performed according to the recommendations of the manufacturer. Briefly, patient serum samples were diluted and allowed to react for 1 hour at 37°C in wells coated with the antigen. After a washing step, the peroxidase-conjugated goat antihuman IgG antibody was added and the plate was incubated as previously described. After removal of the unbound conjugate by a washing step, the plate was incubated with a substrate solution containing tetramethylbenzidine and hydrogen peroxide for 30 minutes at room temperature. The enzymatic reaction was stopped by the addition of acid, and absorbance at 450 nm was determined within 15 minutes. Positive and negative serum controls were included in each assay. The concentrations of IgG antibody in the serum samples were determined by using a calibrator with a known value. As suggested by the manufacturer, values less than 5 units were considered negative and those of more than 7.5 units were considered positive. Values between 5 and 7.5 units were considered equivocal and were retested.
The performance of the test was evaluated by determining the sensitivity, specificity, and positive and negative predictive values with 95% confidence intervals (CIs). To compare IgG concentrations between groups, the children infected by cag A-positive strains with duodenal ulcer were age matched (± 1 year) with children without duodenal ulcer infected by cag A-positive strains. Statistical analysis was performed by using the two-tailed chi-square or Fisher exact test for nonparametric values and the Student t test for comparison of the mean age and IgG concentration between groups. Odds ratios and 95% CIs were estimated by logistic regression controlling for age, gender, and disease. The Spearman correlation was used to analyze the association between age and IgG levels. The level of significance was set at P < 0.05.
Helicobacter-infected children with duodenal ulcer were older (mean age, 11.5 ± 1.8 years;P = 0.02) and frequently were male (15 boys, 6 girls;P = 0.01) compared with infected children without duodenal ulcer (mean age, 9.7 ± 2.8 years; 13 boys, 28 girls).
The cag A was observed in biopsy-cultured specimens (strains, 41; tissue, 2) of 43 of 62 patients (69.3%), including 20 of 21 H. pylori-positive children (95.2%) with duodenal ulcer and 23 of 41 H. pylori-positive children (56.1%) without duodenal ulcer (Table 1). A positive and independent association was also observed between cag A and age (Table 1).
Three children without duodenal ulcer and infected by H. pylori-cag A negative strains had indeterminate results on the ELISA assay and were excluded from accuracy calculations. The data from the remaining 112 children (Table 2) were used for the final analysis. Anti-CagA antibodies were detected in the sera of 41 children (95.3%) infected by cag A-positive strains, 3 children (18.7%) infected by cag A-negative strains, and 6 H. pylori-negative children (11.3%). The sensitivity, specificity, and positive and negative predictive values of ELISA to detect anti-CagA antibodies were 95.3% (95% CI, 82.9%–99.2%), 87.0% (95% CI, 76.2%–93.5%), 82.0% (95% CI, 68.1%–91.0%), and 96.8% (95% CI, 87.8%–99.4%), respectively. The sensitivity of the test was similar in the groups of H. pylori-positive children with (20/20; 100%) and without (21/23; 91.3%) duodenal ulcer (P = 0.49).
When children were stratified by age (group 1[2–6 years], 26 cag A negative, 0 cag A positive; group 2 [7– 11 years], 21 cag A positive and 30 cag A negative; and group 3 [12–16 years], 22 cag A positive and 13 cag A negative), the sensitivity of the test was 90.4% and 100% (P = 1.0), and the specificity was 86.6% and 92.3% (P = 0.37) in groups 2 and 3, respectively. The specificity of the test was 84.6% in group 1, and the sensitivity cannot be calculated in this group because none of the children were colonized by a cag A-positive strain.
Anti-CagA antibodies were detected more frequently in the sera of duodenal ulcer (20 of 21; 95.2%) children than in H. pylori-positive children without duodenal ulcer (21 of 38; 55.2%;P = 0.003). Anti-CagA antibodies were also detected more frequently in the older children (P = 0.001), even when the duodenal ulcer children were excluded from the analysis (P = 0.01). In addition, there was a positive correlation between age and anti-CagA IgG concentration (P = 0.01). A difference in the IgG levels (P = 0.05) was also observed when the children with duodenal ulcer infected by cag A-positive strains were age matched with children infected by cag A-positive strains but without duodenal ulcer.
Among the putative virulence factors produced by H. pylori, the cag pathogenicity island that includes cag A has been associated with complications of H. pylori infection. A more severe gastritis and the development of peptic ulcer disease associated with cag A-positive status have also been observed in children (8,15,16). The cag A positivity status is determined by positive PCR results or the presence of serum anti-CagA antibodies. However, PCR is a labor-intensive and expensive technique that cannot be recommended for routine clinical practice and epidemiologic studies. Conversely, serologic tests, especially ELISA, are noninvasive, inexpensive, and easy to perform. Although these methods have been widely used to establish CagA status, they were not validated for the pediatric population.
The results of the current study demonstrated that in contrast to other ELISA assays with low sensitivity for H. pylori infection in children (10), the CagA ELISA assay is a highly sensitive test to detect CagA status in young children. This may be because CagA is highly immunogenic, and the immune response to this antigen in children seems to be strong. In fact, it has been demonstrated that children infected by cag A-positive strains have higher levels of IgG anti-H. pylori antibodies (15). The accuracy of the CagA ELISA assay in our study was similar to that we observed in Brazilian adults (9), but greater than that described by Yamaoka (16), who reported sensitivity of 85% and specificity of 68% in Japanese adults. This discrepancy may be explained by different antigens and different gold standards, the higher background prevalence of cag A strains in Japan, and the high frequency of gastric cancer in the Japanese study (25%), which decreases the sensitivity of serologic tests for detection of H. pylori infection (17).
In the present study, we also observed that anti-CagA antibodies were detected more frequently in the older children, even when the duodenal ulcer children were excluded from the analysis. These findings have been confirmed by others (8,19,20). As suggested by Queiroz et al. (18) with regard to children, susceptibility to colonization by cag A-positive strains seems to be linked to age and may be related to different expressions of gastric mucosa adherence molecule, which may be modified by age.
False-positive results were observed for 6 of 53 H. pylori-negative children (11.3%). This may be because these patients may have taken antimicrobial drugs for other purposes, with a resulting decrease in the bacterial load, or the bacterium may be eliminated spontaneously, a fact more frequently observed in young children. Sörberg et al. (21) observed that IgG antibodies to CagA antigen were detected up to 32 months after eradication of the microorganism with antimicrobial drugs.
False-positive results were also observed in 3 of 16 children (18.7%) infected by cag A-negative strains that may be explained by the detection of crossreacting antibodies or by previous infection with a cag A-positive strains, or the patients may be infected simultaneously by cag A-positive and -negative strains. Other authors (18,22) reported a higher rate of false-positive results for anti-CagA ELISA for adults. Cover et al. (22) detected anti-CagA antibodies in 27% patients from whom H. pylori strains lacking cag A were isolated. They hypothesized that these patients could be infected by more than one H. pylori strain. These authors determined the cag A status evaluating single colonies. We avoided this problem by determining the cag A status in a pool of colonies. Thus, infection with multiple H. pylori strains was not a factor in the false-positive results in our study.
In conclusion, ELISA showed high accuracy for the determination of CagA status in children, and it can be useful for diagnosis and epidemiologic studies.
1. Mégraud F, Lamouliatte H. Helicobacter pylori
and duodenal ulcer. Dig Dis Sci 1992; 37: 769–72.
2. Parsonnet J, Friedman GD, Vandersteen DP, et al. Helicobacter pylori
infection and the risk of gastric carcinoma. N Engl J Med 1991; 325: 1127–31.
3. Blecker U, Mckeithan TW, Hart J, et al. Resolution of Helicobacter pylori
-associated gastric lymphoproliferative disease in a child. Gastroenterology 1995; 109: 973–7.
4. Mégraud F. Advantages and disadvantages of current diagnostic tests for the detection of H. pylori
. Scand J Gastroenterol 1996; 31: 57–62.
5. Covacci A, Falkow S, Berg DE, et al. Did the inheritance of a pathogenicity island modify the virulence of Helicobacter pylori
? Trends Microbiol 1997; 5: 205–8.
6. Covacci A, Censini S, Bugnoli M, et al. Molecular characterization of the 128Kda immunodominant antigen of Helicobacter pylori
associated with cytotoxicity and duodenal ulcer. Proc Natl Acad Sci U S A 1993; 90: 5791–5.
7. Tummuru MKR, Sharma AS, Blaser MJ. Helicobacter pylori pic
B, a homologe of the Bordetella pertussis
toxin secretion protein, is required for induction of IL-8 in gastric epithelial cells. Mol Microbiol 1995; 18: 867–76.
8. Fusconi M, Vaira D, Menegatti M, et al. Anti-CagA reactivity in H. pylori
-negative subjects. A comparison of three different methods. Dig Dis Sci 1999; 44: 1691–5.
9. Rocha GA, Oliveira AMR, Queiroz DMM, et al. Low sensitivity of serological tests for the detection of anti-CagA antibodies in patients with gastric carcinoma. Gut 1999; 45: A126.
10. Oliveira AMR, Rocha GA, Queiroz DMM, et al. Evaluation of enzyme-linked immunosorbent assay for the diagnosis of Helicobacter pylori
infection in children from different age groups with and without duodenal ulcer. J Pediatr Gastroenterol Nutr 1999; 28: 157–61.
11. Rocha GA, Oliveira AMR, Queiroz DMM, et al. Immunoblot analysis of humoral immune response to Helicobacter pylori
in children with and without duodenal ulcer. J Clin Microbiol 2000; 38: 1777–81.
12. Kelly SM, Pitcher MC, Farmery SM, et al. Isolation of Helicobacter pylori
from feces of patients with dyspepsia in the United Kingdom. Gastroenterology 1994; 107: 1671–4.
13. Peek Jr, RM Miller GG, Tham KT, et al. Detection of Helicobacter pylori
gene expression in human gastric mucosa. J Clin Microbiol 1995; 33: 28–32.
14. Clayton CL, Kleanthous H, Coates PJ, et al. Sensitive detection of H. pylori
by using polymerase chain reaction. J Clin Microbiol 1992; 30: 192–200.
15. Luzza F, Contaldo A, Imeneo M, et al. Testing for serum IgG antibodies to Helicobacter pylori
cytotoxin-associated protein detects children with higher grades of gastric inflammation. J Pediatr Gastroenterol Nutr 1999; 29: 302–7.
16. Yamaoka Y, Kodama T, Graham DY, et al. Comparison of four serological tests to determine the CagA or VacA status of H. pylori
strains. J Clin Microbiol 1998; 36: 3433–4.
17. Queiroz DMM, Mendes EN, Rocha GA, et al. Serological and direct diagnosis of H. pylori
in gastric carcinoma: a case-control study. J Med Microbiol 1999; 48: 501–6.
18. Queiroz DMM, Mendes EN, Carvalho AST, et al. Factors associated with Helicobacter pylori
infection by a cagA-positive strain in children. J Infect Dis 2000; 181: 626–30.
19. Plebani M, Guariso G, Fogar P, et al. Effect of cag
A status on the sensitivity of enzyme immunoassay in diagnosing Helicobacter pylori
-infected children. Helicobacter 1999; 4: 226–32.
20. Malfertheiner P, Gerards C, Leodolter A, et al. Pitfalls in Helicobacter pylori
diagnosis. In: Hunt RH, Tytgat GNJ, eds. Helicobacter pylori : Basic Mechanisms to Clinical Cure.
Amsterdam: Kluwer Academic Publishers, 1998;274–92.
21. Sörberg M, Engstrand L, Ström M, et al. The diagnostic value of enzyme immunoassay and immunoblot in monitoring eradication of Helicobacter pylori
. Scand J Infect Dis 1997; 29: 147–51.
22. Cover TL, Glupczynski Y, Lage AP, et al. Serologic detection of infection with cag
A + H. pylori
strains. J Clin Microbiol 1995; 33: 1496–1500.