The significance of Helicobacter pylori infection in infants and children is still unclear. Several studies have shown a high prevalence of infection in early childhood in developing countries (1-4). However, most studies did not include children under two years of age. Furthermore, in most studies, serology was the only diagnostic method used, although little is known about the development of antibodies to H. pylori in children less than 24 months of age. In addition, impaired immunity due to malnutrition may interfere with the interpretation of serologic results (5), and passive immunity with IgG antibodies transferred from mothers to infants may also constitute a confounding factor. Blecker et al. (6) have shown that all infants with positive serology at birth, born to H. pylori-positive mothers, became H. pylori IgG antibody-negative at the age of 3 months. Therefore, it is difficult to know the exact time of infection and when eradication of the organism occurs after treatment by use of serologic methods only.
The use of the 13C-urea breath test (UBT) to detect H. pylori has been widely accepted in adults and children, and the method has high sensitivity and specificity (7). The method has recently been validated in infants and children (8,9). Although the UBT is easy to perform, an expensive mass spectrometer is needed for analysis of the breath samples, and the method may not be affordable in developing countries.
Because children in lower age groups seem to be asymptomatic, it is not ethical to confirm a positive serology or UBT result by gastroscopy to obtain gastric biopsy specimens for detection of H. pylori. Therefore, there is a need for alternative methods of detecting H. pylori infection in childhood.
Polymerase chain reaction (PCR) is widely used to detect a variety of microorganisms (10) and has been shown to be a rapid, sensitive, and specific method for diagnosing H. pylori in gastric tissue specimens. (10-13). Some investigators have also tried to detect H. pylori in the feces of infected humans by PCR (14-16). Different sample processing methods such as immunomagnetic separation (IMS), have been evaluated as a first step to avoid fecal inhibitors of Taq polymerase in the PCR reaction (17). The IMS process enriches target bacteria, thus making PCR more sensitive. Moreover, it is inexpensive and simple and can be used repeatedly to study the natural history of H. pylori infection in developing countries. The aim of the present study was to compare serology (EIA and immunoblot), UBT, and IMS-PCR in detecting H. pylori infection in infants and young children in a periurban community in Bangladesh.
PATIENTS AND METHODS
The study was performed in Nandipara, seven miles northeast of Dhaka city in Bangladesh. It has a population of 3000 in an area of 2.5 square miles. There are approximately 500 children less than 5 years of age in the community. The average family size is 4.5 members. The people use municipal water for drinking and cooking. However for bathing and washing, they depend on ditch water. The environmental sanitation of the area is very poor, and most inhabitants use hanging latrines that contaminate the ditch water. Most houses are mud-walled with attached dry bamboo roofs. A weekly clinic has been run by International Centre for Diarrhoeal Disease Research (ICDDR) in Bangladesh to treat minor illnesses in the population since 1986.
The study population consisted of 81 asymptomatic children between 4 and 24 months of age who were recruited from the community. Eleven children were excluded from the study because of missing serum samples and two because of missing stool samples. The remaining 68 children (36 boys and 32 girls) had a mean age (±standard deviation) of 11.5 ± 5.7 months and a mean weight of 7.2 ± 1.3 kg. Ten to 15 children at a time were screened for H. pylori at six different occasions during 1995. Sixty-eight of the children were investigated for occurrence of Helicobacter pylori infection using three different diagnostic methods: serum examined for IgG antibodies to H. pylori using an immunoblot (IB) and an enzyme immunoassay analysis (EIA); a UBT performed as a single-point simplified test; and stool samples examined for presence of H. pylori with IMS-PCR. All children were totally or partially breast fed.
One hundred µl of capillary blood was obtained from the children in capillary micropipettes and diluted with 1 ml of normal saline. The serum was separated by centrifugation and frozen at -20°C in Dhaka at ICDDR and transported in dry ice and stored at -70°C in Sweden until analysis was performed. Sera were examined for IgG antibodies to H. pylori with EIA and IB as previously described (19).
The antigens for EIA and IB were prepared from whole organisms using H. pylori strain CCUG 17874 from the culture collection at the university of Gothenburg, Sweden. Preparation of the antigen was made using acid glycine extraction at pH 2.2, according to a standard method (20).
13C-Urea Breath Test
Patients were examined in the postprandial state after a 2-hour fast. Breath sampling was performed using a mask (Åsmund-Laerdal AS, Staranger, Norway) connected with a one-way valve to an anesthesia gas tube. After approximately 10 breath cycles, samples of expired air were drawn by syringe from the tube and transferred to screw-capped glass tubes. Basal (predose) samples were obtained, and the children were given a test meal consisting of 100 ml of milk formula. Ten minutes later, 40 mg of 13C-urea, dissolved in 10 ml of water, was administered. After 30 minutes, breath samples (postdose) were collected again. All samples were collected in duplicate, and the ratio of 13C:12C determined by gas chromatography-mass spectrometry with a dedicated instrument (BreathMAT, Finnigan MAT, Bremen, Germany). The results were expressed as the relative enrichment of 13C in basal samples and postdose samples and the results reported as the change in relative enrichment. Values of 5δ‰ or more were considered positive (7).
Immunomagnetic Separation-Polymerase Chain Reaction
Stools were collected from children in the study area by health workers and stored at -20°C at ICDDR and transported to Sweden on dry ice. In Sweden, the samples were stored at -70°C until subsequent analyses were performed.
Precoated magnetic beads (Sheep Anti-Rabbit IgG Dynabeads M-280; Dynal AS, Oslo, Norway), were mixed with polyclonal anti-H. pylori rabbit antibodies (19), coupled by a direct technique according to the manufacturer's instructions, and stored at 4°C.
Immunomagnetic separation was performed according to the method of Nilsson et al. (21) Briefly, 0.2 g feces was homogenized with 10 ml of phosphate-buffered saline (PBS; 0.1 M, pH 7.2), containing 0.1% bovine serum albumin (BSA, Fraction V, Sigma Chemical Co, St. Louis, MO, U.S.A.) and 0.1% hexadecyltrimethyl-ammoniumbromide (CTAB; Sigma). Fecal slurries were centrifuged for 1.5 minutes at 200g. Supernatants (1 ml) were centrifuged again at 12,000g for 5 minutes (Microfuge E; Beckman Instruments, Palo Alto, CA, U.S.A) and the pellets resuspended in 1 ml PBS and 0.1% BSA. To the resuspended cell pellets, 30 µl of anti-H. pylori antibody-coated magnetic beads was added and the mixtures incubated with gentle agitation at 4°C for 60 minutes. Magnetic particles were separated from the suspension using a magnetic device (MPC-M, Dynal) and washed three times for 10 minutes with PBS and 0.1% BSA. The beads were then resuspended in 30 µl sterile distilled water, boiled for 10 minutes, briefly chilled on ice, centrifuged (12,000g; 7.5 minutes), and the supernatants transferred to fresh microcentrifuge tubes and frozen until analyzed with PCR.
Polymerase Chain Reaction Conditions
Primers based on the urease A gene of H. pylori, as originally described by Clayton et al., (12) were used for PCR amplification. A conventional PCR-buffer was used for the amplification, consisting of 10 mM Tris-HCl (pH 8.3), 50 mM KCl, 2.5 mM MgCl2, 0.01% BSA, 0.2 mM of each deoxynucleotide triphosphate (Boehringer-Mannheim, Mannheim, Germany), and 0.5 µM of each oligonucleotide primer (SDS; Köping, Sweden). Taq polymerase (1.5 U; Boehringer-Mannheim) was added and the reaction mixture was overlaid with 40 µl of mineral oil (Sigma). Polymerase chain reaction was performed in a reaction volume of 50 µl in a thermal cycler (PTC-100; MJ Research, Watertown, MA, U.S.A.). Amplification consisted of initial denaturation at 94°C for 5 minutes followed by denaturation at 94°C for 1 minute, primer annealing at 47°C for 1 minute, and extension at 72°C for 1 minute. Samples were amplified through 35 consecutive cycles. The final cycle included extension for 10 minutes at 72°C. Amplified products (15 µl) were analyzed in agarose gel electrophoresis with an agarose concentration of 2% (wt/vol). The PCR products were visualized by excitation under ultraviolet light after 30 minutes of staining with 0.5 µg/ml ethidium bromide. At each amplification event, a negative control, consisting of water instead of template DNA, was included as well as an IMS H. pylori-negative fecal sample.
Southern Blot Hybridization
To confirm that H. pylori DNA was amplified in the PCR analysis, Southern blot hybridization was performed on a set of representative positive and negative samples. Amplified fragments of PCR product were run in electrophoresis and transferred onto a nitrocellulose membrane (Schleicher & Shull, Dassel, Germany) by the alkaline blotting procedure (22). The membrane was heated at 80°C for 2 hours and prehybridized in 5× SSC (20× SSC; 3 M NaCl, 0.3 M sodium citrate), 2% blocking reagent (Boehringer-Mannheim), 0.1% N-lauroylsarcosine and 0.02% sodium dodecyl sulfate (SDS) at 65°C for 4 hours. The amplified urease A (H. pylori urease) fragment of the reference strain H. pylori CCUG 17874 was used as a probe and was prepared by the hexanucleotide priming technique with the digoxygenin-deoxyuridine triphosphate labeling kit (DIG; Boehringer-Mannheim, GmbH, Germany). After hybridization, at 65°C for 12 hours, and washing, twice with 2× SSC and 0.1% SDS and twice with 0.1× SSC, 0.1% SDS, detection of hybridized probes was performed with the nucleic acid detection kit. Labeling and detection of the probe were performed according to the manufacturer's instructions.
Informed consent was obtained from the parents of the children. The study was approved by the local ethics committee at the ICDDR.
Of the 68 children included in the study, 41 (60%) were determined to be H. pylori-positive by IMS-PCR, 39 (57%) by UBT, and 6 (8.8%) by IB. All children were negative according to the results of EIA (Table 1). Of the 6 children positive by IB, 4 were also positive by UBT and PCR, and 2 were positive by UBT only (Fig. 1). Using all methods together, IB, IMS-PCR, and UBT, the H. pylori prevalence in the children was 78% (n = 53). Adding the 11 children in whom serology was not performed did not change the overall results. The concordance (PCR- and UBT-positive or PCR- and UBT-negative) between the two methods was 62% (n = 42).
The selected samples that generated either an intense or a weak positive signal in IMS-PCR, all hybridized to the probe at stringent conditions in the Southern blot analysis. Negative samples from IMS-PCR did not hybridize in Southern blot (Fig. 2).
This is the first study designed to evaluate three noninvasive methods to diagnose H. pylori infection in infants in a developing country. We found that 57% of the studied infants were UBT-positive for H. pylori, which supports the idea of early acquisition of H. pylori in endemic areas (23). Similar results were also found in a study by Bunn et al. (24) in Gambian children using UBT and in a study performed on Nicaraguan children with persistent diarrhea (25).
Although UBT is recognized as a gold standard method (8,9), there may be some methodologic problems in using it in infants, and the gas chromatograph-mass spectrometer used for analysis may be too costly in developing countries such as Bangladesh. However, we found it practical to keep the breath samples in screw-capped glass tubes until analysis was performed in Sweden.
Factors that may influence the UBT results may be other urease-producing bacteria in the mouth or the gastric mucosa or bacterial overgrowth and rapid transit time, especially in highly endemic areas. Although other urease-producing bacteria may interfere with the results, this was not recognized as a methodologic problem because of the high urease-producing capacity of H. pylori. (26). Conversely, rapid transit time may produce false-negative results. Most of the children we screened had loose stools, but none of them had chronic diarrhea.
Other epidemiologic studies based on serology in developing countries in this age group reported approximately 16% (2) and 14% (3)H. pylori-positive children. We found 9% positivity using IB alone. However, we observed a statistically nonsignificant tendency that older children (13-24 months; n = 25) had a higher positivity rate with immunoblot (16%) than did the younger children (4-12 months; 2 of 43 positive [5%]). This may indicate an even lower immune response to H. pylori in this age group. Consequently, it seems that serology is not a good screening method in this population. Therefore, the prevalence of H. pylori infection using serology in developing countries may have been underestimated. The explanation for this low seropositivity may be the result of nonseroconversion or it may be that the methods used have not been validated in this type of population, as shown by Crabtree et al. (27) and Goossens et al. (28) Furthermore, the kits used for serology may not be specific for the H. pylori strains prevalent in developing countries such as Bangladesh. Animal studies have also shown that it may take up to 3 months before development of antibodies against H. pylori occurs (29). Almost all our children were totally or partially breast fed. Breast milk containing IgA antibodies against H. pylori in seropositive mothers may provide some protection against H. pylori infection, as suggested by Bunn et al., (24) and it is therefore important to find other methods for detecting those who carry the infection.
Analysis by IMS-PCR of stool samples from the study group showed 60% positivity. A previous study in an adult population showed good correlation between results of IMS-PCR and serology (21). Magnetic separation provides a two-step purification of H. pylori, facilitating removal of inhibitors of Taq polymerase found in human stool. A representative panel of PCR-amplified samples was analyzed using Southern blot hybridization with digoxygenin-labeled PCR fragments generated by the urease A (HPU) primers. It has been shown that these primers do not amplify DNA from the bacterial species tested so far, including other Helicobacter species, Campylobacter, and a wide range of different urease-producing bacteria (12,21).
There is a need for new noninvasive methods to detect H. pylori infection, especially in children in developing countries. Polymerase chain reaction is an easy method that can be used for this purpose; however, the analysis detects H. pylori DNA but does not distinguish between live or dead bacteria. A rapid gastrointestinal transit makes this method more suitable in children. There are reports of cultured H. pylori in biopsy specimens obtained in the duodenum (30) and of H. pylori cultured from stool of infected children (31). IMS-PCR may be a valuable method, but further studies must be performed in which IMS-PCR results on fecal specimens are compared with those in culture of gastric biopsy samples.
There was some discrepancy between the PCR and UBT results. Colonization with intermittent shedding of bacteria in the stool may explain PCR negativity and UBT positivity in some children. Whereas UBT reflects active infection with live bacteria, PCR may also detect dead bacteria and coccoid forms of H. pylori that do not produce urease.
Using several noninvasive methods at the same time to detect H. pylori in endemic areas may produce different results as shown in our study. This may be of importance when interpreting epidemiologic studies in which endoscopy is not performed.
We conclude that serology and an inferior epidemiologic tool for screening for H. pylori infections in young infants and children in developing countries. The IMS-PCR analysis of H. pylori in stool and the UBT are promising noninvasive diagnostic tests for children. Furthermore, because they may reflect different disease stages of H. pylori infection, they may be useful for epidemiologic and therapeutic studies.
Acknowledgment: The authors thank Drs. Rasheduzzaman Shah and Aminul Islam and Makshud Ali Khan, Asma Khatun, Arif Hussein, and Afia Akhter for their help in the study.
Supported in part by Swedish International Development Cooperation Agency (Sida), the Swedish Medical Research Council (Grant 16X-04723), the Jerring Foundation, Enskale, Sweden, Swedish Society of Medicine, Stockholm, Sweden, and funds from the Karolinska Institute and the International Centre for Diarrhoeal Disease Research, Bangladesh.
1. Sarker SA, Rahman MM, Mahalanabis D, et al. Prevalence
of Helicobacter pylori
infection in infants
and family contacts in a poor Bangladesh community. Dig Dis Sci 1995;40:2669-72.
2. Oliveira AMR, Queiroz DMM, Rocha GA, Mendes EN. Seroprevalence of Helicobacter pylori
infection in children of low socioeconomic level in Belo Horizonte, Brazil. Am J Gastroenterol
3. Sullivan PB, Thomas JE, Wight DGD, et al. Helicobacter pylori
in Gambian children with chronic diarrhoea and malnutrition. Arch Dis Child
4. Klein PD, Graham DY, Gaillour A, Opekun AR, Smith EOB. Water source as risk factor for Helicobacter pylori
infection in Peruvian children. Lancet
5. Chandra PK, Kumari S. Nutrition and immunity: An overview. J Nutr
6. Blecker U, Lanciers S, Keppens E, Vandenplas Y. Evolution of Helicobacter pylori
positivity in infants
born from positive mothers. J Pediatr Gastroenterol Nutr
7. Logan RPH, Dill S, Bauer FE, et al. The European 13
C-urea breath test
for the detection of Helicobacter pylori. Eur J Gastroenterol Hepatol
8. Vandenplas Y, Blecker U, Devreker T, et al. Contribution of the 13
C-urea breath test
to the detection of Helicobacter pylori
gastritis in children. Pediatrics
9. Rowland M, Lambert I, Gormally S, et al. 13
C-urea breath test
for the diagnosis of Helicobacter pylori
infection in children. J Pediatr
10. Randhawa PS. Application of PCR for the diagnosis of infectious diseases affecting the gastrointestinal tract and liver. In: Erlich GD, Greenberg SJ, eds. PCR-based diagnostics
. Boston: Blackwell Scientific, 1994:231-52.
11. Bickley J, Owen RJ, Fraser AG, Pounder RE. Evaluation of the polymerase chain reaction
for detecting the urease C gene of Helicobacter pylori
in gastric biopsy samples and dental plaque. J Med Microbiol
12. Clayton CL, Kleanthous H, Coates PJ, Morgan DD, Tabaqchali S. Sensitive detection of Helicobacter pylori
by using polymerase chain reaction
. J Clin Microbiol
13. Hammar M, Tyszkiewicz T, Wadström T, O'Toole PW. Rapid detection of Helicobacter pylori
gastric biopsy material by polymerase chain reaction
. J Clin Microbiol
14. Mapstone NP, Lynch DAF, Lewis FA, et al. PCR identification of Helicobacter pylori
in faeces from gastritis patients. Lancet
15. Enroth H, Engstrand L. Immunomagnetic separation
and PCR for detection of Helicobacter pylori
in water and stool specimens. J Clin Microbiol
16. Van Zwet AA, Thijs JC, Kooistra-Smid AMD, Schirm J, Snijder JAM. Use of PCR with faeces for detection of Helicobacter pylori
infections in patients. J Clin Microbiol
17. Monteiro L, Bonnemaison D, Vekris A, et al. Complex polysaccharides as PCR inhibitors in faeces: Helicobacter pylori
model. J Clin Microbiol
18. Lotterer E, Ramaker J, Lüdtke FE, Tegeler R, Geletneky JV, Bauer FE. The simplified 13
C-urea breath test
: One point analysis for detection of Helicobacter pylori
infection. Z Gastroenterology
19. Nilsson I, Ljungh Å, Aleljung P, Wadström T. Immunoblot assay for serodiagnosis of Helicobacter pylori
infections. J Clin Microbiol
20. Lelwala Guruge J, Schalén C, Nilsson I, et al. Detection of antibodies of Helicobacter pylori
cell surface antigens. Scand J Infect Dis
21. Nilsson HO, Aleljung P, Nilsson I, Tyskiewicz T, Wadström T. Immunomagnetic bead enrichment and PCR for detection of Helicobacter pylori
in human stools. J Microbiol Method
22. Sambrook J, Fritsch EF, Maniatis T. Molecular cloning: A laboratory manual
. 2nd ed. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press, 1989.
23. Neale KR, Logan PH. The epidemiology and transmission of Helicobacter pylori
infection in children. Ailment Pharmacol Ther
24. Bunn JEG, Thomas JE, Harding M, Arthur H, Oostenveld E, Weaver LT. Maternal milk anti-H. pylori
IgA modulates pattern of H. pylori
infection in infancy. J Pediatr Gastroenterol Nutr
25. Kehrt R, Becker M, Brösicke H, Krüger N, Helge H. Prevalence
of Helicobacter pylori
infection in Nicaraguan children with persistent diarrhea, diagnosed by the 13
C-urea breath test
. J Pediatr Gastroenterol Nutr
26. Mobley HLT, Foxall PA. H. Pylori
urease. In: Hunt RH, Tygat GNJ, eds. Helicobacter pylori. Basic mechanisms to clinical care
. Dorchecht, The Netherlands: Kluwer Academic Publishers. 1991:41-58.
27. Crabtree JE, Mahony MJ, Taylor JD, Heatley RV, Littlewood JM, Tompkins DS. Immune responses to Helicobacter pylori
in children with recurrent abdominal pain. J Clin Pathol
28. Goossens H, Glupczynski Y, Burette A, Van den Borre C, Butzler JP. Evaluation of a commercially available second-generation immunoglobulin G enzyme immunoassay for detection of Helicobacter pylori
infection. J Clin Microbiol
29. Wang X, Sturegård E, Rupar R, et al. Infection BALB/c A mice spiral and coccoidal forms of Helicobacter pylori. J Med Microbiol
30. Bontems P, de Olivera Andrade D, Corvaglia L, Heimann P, Scaillon M, Cadranel S. Pattern of Helicobacter pylori
infection in infants
and young children in industrialised countries. J Pediatr Gastroenterol Nutr
31. Thomas JE, Gibson GR, Darboe MK, Dale A, Weaver LT. Isolation of Helicobacter pylori
from human faeces. Lancet
Clinical Quiz Section
Short cases with accompanying photographs of diagnostic biopsies, endoscopic findings, or physical characteristics are welcomed for the newly established Clinical Quiz section. Please send a short history, photo with negative, and answers to quiz questions to:
ESPGHAN Clinical Quiz Editor Dr. Riccardo Troncone; Department of Pediatrics; University Federico II; Via S. Pansini, 5; 80131 - Naples; ITALY; Tel: 39-81-7463394; Fax: 39-81-5469811
NASPGN Clinical Quiz Editor Dr. Joseph F. Fitzgerald; Department of Pediatrics; James Whitcomb Riley Hospital for Children; 702 Barnhill Drive; Indianapolis, IN 46202-5225; Tel: 317-274-3774; Fax: 317-274-8521
Keywords:© 1999 Lippincott Williams & Wilkins, Inc.
Helicobacter pylori; Immunomagnetic separation; Infants; Prevalence; Polymerase chain reaction; Serology; Urea breath test