Recurrent abdominal pain (RAP) is a common gastrointestinal disorder of school-aged children. Approximately 90% of affected children do not have any underlying organic disease (1). However, some investigators report a higher prevalence of dyspeptic symptoms and peptic ulceration in adult patients who had RAP in childhood (2). Moreover, duodenal ulcer (DU) is strongly associated with chronic and recurrent symptoms: approximately 90% of children with primary duodenal ulceration have abdominal pain (3).
There is compelling evidence that the Gram-negative spiral bacterium Helicobacter pylori is an important cause of chronic gastritis and peptic ulceration in childhood (4). However, the association between H. pylori infection and gastrointestinal symptoms in the absence of ulcer is still controversial (5,6). Numerous epidemiologic studies have indicated that H. pylori is not a cause of recurrent abdominal pain in children (7–9). For instance, a critical review by Macarthur et al. (9) failed to show a relation between H. pylori and RAP in children. However, other studies report a high frequency of H. pylori infection in children with RAP (2,10,11). For example, immunoglobulin G (IgG) antibodies against H. pylori were detected in 43% of Danish children who had RAP (12). If H pylori is indeed associated with RAP, determination of IgG antibodies directed against H. pylori in dyspeptic children may represent a useful noninvasive screening procedure to replace more invasive, uncomfortable, and expensive endoscopy to diagnose H. pylori–associated gastroduodenal disease, in particular, peptic ulcers (13,14).
Some studies in children (15) and adults (16) have found no difference in anti–H. pylori IgG concentrations between subjects with DU and those with H. pylori–associated gastritis without ulcers. However, a more recent study by Oliveira et al. (17) reported that serum anti–H. pylori IgG levels were significantly higher in children with DU than in those without ulceration. In 1995, using the same urease-based detecting antigen, we found that chronic childhood DU was accompanied by higher anti–H. pylori IgG titers than those in children without ulcer (14). The reason for these contradictory results remains unexplained.
PATIENTS AND METHODS
Between January and December 1998, 129 pediatric outpatients (mean age, 12.1 ± 2.3 years; range, 10–15 years), seen consecutively with upper abdominal or gastrointestinal symptoms requiring endoscopic evaluation were included in the study. Forty-one of the patients were boys, and 88 were girls. Among the patients, 52 were Russian and 67 were native Asian (Bashkirs). The subjects who had been treated with any medication (bismuth salts, H 2 -blockers, antacids, antibiotics) during the previous month were not included in the study. Fully informed parental consent was obtained for all the procedures in each case. Anatomic abnormalities were excluded by abdominal ultrasonography. Long-term esophageal pH testing was performed on 33 selected patients. The presence of chronic inflammatory bowel disease, lactose malabsorption, and celiac disease was assessed using standard screening diagnostic procedures: urinalysis, complete blood count, erythrocyte sedimentation rate, and evaluation for parasitic infestation were performed in all patients.
A detailed history and physical examination were obtained from each patient. Information about symptoms was elicited using a standardized questionnaire that was completed by the children and their parents before diagnostic endoscopy. A history of RAP was sought in each case (18). The severity of symptoms was graded from 0 to 6. Symptoms assessed included abdominal pain, bitter taste, nausea, vomiting, heartburn, and hematemesis. A score of 1 point was given for each factor. Gormally et al. (6) outlined a similar symptom score.
Endoscopy and Biopsy
Severe chronic abdominal pain, possibly indicating organic disease, was the main indication for endoscopy in all patients. Esophagogastroduodenoscopy was performed using an Olympus GIF XP 20 fiberoptic endoscope (Olympus Optical Co., Ltd., Tokyo, Japan) after overnight fast without prior sedation. Standard endoscopic forceps (FB-19K; Olympus Optical Co., Ltd.) were used to obtain mucosal biopsy specimens. Endoscope and biopsy forceps both were cleaned and disinfected in 2% glutaraldehyde and washed before every endoscopy. During endoscopy, four biopsy specimens were taken from the antral mucosa, within 5 cm of the pylorus. In all patients, biopsy specimens were taken from abnormal-appearing mucosa. Histology, rapid urease test, and polymerase chain reaction (PCR) were performed in each subject (19).
Two biopsy specimens were fixed in 10% neutral-buffered formalin and then embedded in paraffin. Sections (5 μm) were cut with a microtome, stained with hematoxylin and eosin, and evaluated under light microscopy. Mucosal inflammation was identified under high dry magnification (×150). Esophagitis was diagnosed by use of the standard histologic procedure (presence of intraepithelial leukocytes or basal cell hyperplasia). As outlined in Table 1, the score of antral gastritis severity was assessed according to the grading system described by Mitchell et al. (20). H. pylori colonization of gastric mucosa was studied using the modified Giemsa technique, as described previously (14). Histopathologic evaluation was performed by a pathologist who was not aware of clinical and endoscopic status.
Rapid Urease Testing
One of the antral biopsy specimens was placed into a vial containing 0.2 mL “De-Nol test” broth (19). For each investigation, one vial with medium only (that is, without gastric tissue) served as a negative control. The vials were incubated at room temperature for 4 hours; a color change from tan to pink indicated positive test results.
Polymerase Chain Reaction
Polymerase chain amplification, based on specific urease A gene internal sequences, was also used for H. pylori detection on the fourth gastric biopsy sample. The PCR was performed with the primers described by Clayton et al. (21). Results were considered positive when the product of amplification was identified.
After an overnight fast, 5 mL whole blood was obtained from each subject before endoscopy. Serum was frozen at −20°C until used for analysis. Serum samples were coded and evaluated blindly within 1 week.
H. pylori–specific IgG antibodies were assayed by indirect enzyme-linked immunoassay using a highly purified urease-containing bacterial antigen (anti–H. pylori EIA; Roche, Zürich, Switzerland). Results are expressed as optical density measurements (14). The optical density cutoff point was established according to the recommendations of the manufacturer. All assays were performed in duplicate with appropriate positive and negative internal controls. Children were considered to be infected with H. pylori when the serum antibody titer was more than 0.12 optical density units (ODU). The assay was concurrently validated in 100 children of similar age whose H. pylori status was known from the results of gastric biopsy.
H. pylori Identification
A combination of three positive tests (Giemsa stain, urease test, and PCR) was used as the “gold standard” for detection of H. pylori. In this study, the immunoassay was not considered to be an essential part of the diagnostic standard.
Results are expressed as either mean ± standard deviation or as the median for the score of symptoms and for anti–H. pylori IgG titers. Differences between groups were compared by Student t test and chi-square (χ 2 ) test, as appropriate. Correlation between anti–H. pylori antibody level, inflammatory infiltrate of mucosa, symptom score, and duration of symptoms were obtained using the Spearman rank test.
Considering the combination of the three positive biopsy-related tests as the diagnostic standard for the detection of H. pylori infection, 104 (80.6%) of 129 children were H. pylori positive. All the children shown by histology to be positive for H. pylori infection were also positive by PCR and urease testing.
The results of endoscopic features and histologic findings are shown in Table 2. Histologic examination of biopsy specimens obtained from infected patients showed evidence of chronic gastritis and presence of the infection (Table 2). All 31 children with endoscopic evidence of DU were positive for H. pylori.
Histologic evidence of peptic esophagitis was shown in nine patients. Each of the children with esophagitis also had pathologic acid reflux determined by pH recordings. These children were indistinguishable within the study group. No histologic changes of eosinophilic gastroenteritis, celiac disease, and chronic inflammatory bowel disease were found in our series. Duodenal aspirate for the identification of Giardia and ankylostomiasis was always negative. Stool cultures and results of microscopic examinations for ova and parasites were always negative. Abdominal ultrasonography did not identify any organic disease condition of kidney, gallbladder, or pancreas. Urinalysis results and complete blood count were always normal. Lactose malabsorption was excluded on clinical grounds.
Clinical data are summarized in Table 3. There was no association of H. pylori infection with gender or ethnicity. The mean age of the patients with H. pylori infection was not older than the mean age of patients without H. pylori (P > 0.05). The score of symptoms and the duration of complaints also were similar in both groups of the patients. Children with H. pylori infection had a higher frequency of family history of peptic ulcer compared with those who were H. pylori negative (47% vs. 4%, P < 0.05).
As shown in Table 4, there were no significant differences between ulcer-positive and ulcer-free subgroups within the H. pylori–positive population with respect to age and gender. However, H. pylori infection was more common in Asian children than in Russian children with peptic ulceration. Therefore, there was stated a positive association between peptic ulcer disease and ethnicity.
The mean duration of symptoms and symptom score were not associated with H. pylori–positive peptic ulceration (P > 0.05), but family history of peptic ulcer was significantly associated with endoscopically demonstrated DU. The mean duration of symptoms was similar in children with DU and H. pylori–positive children without peptic ulceration compared with the H. pylori–negative children (t = 0.405 and t = 0.348, respectively, P > 0.05). The mean symptom score among the children with ulcers, H. pylori–infected children without ulcers, and symptom score among H. pylori–negative children also did not differ (t = 1.586 and t = 0.935, P > 0.05).
As shown in Table 4, peptic ulceration was associated with nighttime abdominal pain, fasting abdominal pain, postprandial pain, pain associated with meals, bitter taste, and heartburn compared with the children infected with H. pylori without ulcer (P < 0.05). The periumbilical location of pain also had a significant association with ulcer-positive status (P < 0.05). Apparently, these symptoms are the only symptoms suggestive of organic disease, indicating a need for endoscopic investigation. Epigastric or diffuse locations of pain did not distinguish between the subgroups (P > 0.05). The prevalence of hematemesis, vomiting, nausea, belching, and halitosis was comparable in the children with ulcers and those without ulcers (P > 0.05).
Serum IgG mean optical density value was 1.46 ± 0.39 ODU for the H. pylori–positive children and 0.021 ± 0.01 ODU for the H. pylori–negative patients (P < 0.01). Among the patients with H. pylori infection, serum anti–H. pylori IgG titers were increased in those with peptic ulcer compared with subjects with gastritis alone (2.12 ± 0.57 ODU vs. 1.06 ± 0.24 ODU, P < 0.001). The sensitivity and specificity of serum H. pylori IgG level were both 100% (P < 0.01). The positive predictive value for positive IgG was 100%, and the negative predictive value was 100%. Furthermore, serum IgG antibody titers correlated with chronologic age (r = 0.374, P < 0.05), the degree of gastritis (r = 0.629, P < 0.001), and bacterial load in the gastric mucosa (r = 0.539, P < 0.001).
Table 5 shows the results of the serum IgG assay in H. pylori–positive patients of different age groups. Increased optical density values of anti–H. pylori IgG were evident in the older children with H. pylori infection. The serum IgG titers were not significantly different among endoscopic ulcer-positive children of different ages (P > 0.05).
The gastritis score correlated with the symptom score in the infected and noninfected patients (r = 0.24, P < 0.05 and r = 0.548, P < 0.05, respectively). Nevertheless there was no correlation between symptom scores and IgG titers (r = 0.07, P > 0.05) in the H. pylori–positive group. The duration of complaints in H. pylori–positive patients was not correlated with the level of anti–Helicobacter IgG (r = −0.06, P > 0.05), gastritis score (r = −0.017, P > 0.05), or symptom score (r = 0.189, P > 0.05).
In the patients with H. pylori–negative status there also was no observed correlation between duration of complaints and symptom score (r = 0.086, P > 0.05), and between duration of complaints and gastritis score (r = 0.173, P > 0.05). We did not find any significant correlation between symptom score and gastritis score (r = 0.193, P > 0.05) and IgG titers (r = 0.017, P > 0.05) in the children with peptic ulcer disease. Also, among these patients, no correlation was found between duration of complaints and serum IgG levels (r = −0.121, P > 0.05) or gastritis score (r = 0.08, P > 0.05). In the H. pylori–positive ulcer-free subgroup the gastritis score was correlated with symptom score (r = 0.268, P < 0.05). However, symptom score was not observed to be significantly correlated with the specific IgG levels (r = 0.08, P > 0.05) or duration of complaints (r = 0.186, P > 0.05). The duration of symptoms in ulcer-negative patients infected with H. pylori did not correlate with gastritis score (r = −0.003, P > 0.05) or with anti–Helicobacter IgG antibody levels (r = 0.009, P > 0.05) or symptom score (r = 0.188, P > 0.05).
There are limited reliable data available on the incidence and prevalence of DU in children (22). In developed countries, DU is not a common condition in children (23). For example, in Italy the detection rate of ulcer disease was 7.8 in an average of 180 pediatric gastrointestinal endoscopies peformed each year (24). In Canada, an approximate incidence of peptic ulcer was 1 case per 2,500 hospital admissions (3). Our study evaluated the frequency of peptic ulcer disease and H. pylori infection in school-aged children with chronic abdominal complaints living in the Ural area of Russia. A high frequency of peptic ulcers among these children probably reflects the high incidence of peptic ulcer disease in Russia (19,22,25). In our area, the overall incidence of childhood peptic ulcer disease was 82.0 per 100,000 children in 1999. The finding of 24% of school-aged children with RAP having duodenal ulcers shows a high prevalence rate of DU. Recently, Ashorn et al. (26) concluded that DUs might be found in approximately 4% of children with RAP syndrome. The same figure (5%) was shown by Mitchell et al. (20) in Australian children with gastrointestinal symptoms. In Italy, approximately 12% of children with nonspecific upper abdominal symptoms had DU (27). However, two latter studies included children with the age range from 1 to 18 years. In the age groups as young as 6 months to 1 year old, RAP is an uncommon diagnosis (9). Obviously, any preschool pediatric population cannot be considered as a comparison group for subjects of school age with RAP. This may be a reason for discordant results compared with our selected population, which includes school-aged children only. At the same time, our data are mainly in accordance with the findings of another study in older pediatric patients (28), in which 18% of the children who underwent endoscopy because of RAP, were noted to have DU.
A strong association between H. pylori and DU, which was shown in our series, is similar to that reported in older pediatric patients with RAP in Western countries (4,5,15,23,26,27).
H. pylori was identified in 80% of the school-aged patients residing in the Ural area of Russia who underwent endoscopy for RAP. H. pylori infection was not related to age, gender, and ethnicity. The clinical symptom score and the mean duration of complaints did not differentiate children with H. pylori colonization from those who were not infected. H. pylori infection in these children was associated with a family history of peptic ulcer disease (P < 0.05).
It has been postulated that H. pylori infection associated with duodenal ulcer is an important cause of symptoms in childhood (5,6,29). The symptoms, which we found indicative of the presence of peptic ulcer, were nighttime pain with nocturnal awakening, fasting pain relieved by food, postprandial pain, pain associated with meals, heartburn, and bitter taste (P < 0.05). A previous history of peptic ulcer in first-degree relatives was a more frequent finding in the ulcer-positive subgroup, which agrees with the reference data (29). The severity of symptoms in patients with peptic ulcer was not associated closely with the severity of antral gastritis accompanying ulcer disease. Also, we were not able to provide any evidence of an association between duration of complains and gastritis severity in H. pylori–infected children, which was reported by Mikelsaar et al. (30).
An important clinical pattern emerging from this study was that peptic ulcer disease was more frequent in native Asian patients. This is the first report showing a possible predisposition of an ethnic group of Asian persons (Bashkirs) to childhood peptic ulcer.
Recently, Oliveira et al. (17) and Oderda et al. (31) reported that anti–H. pylori IgG antibodies are significantly increased in pediatric patients with DU compared with those without ulceration. Our previous data (14) and current findings support the conclusion that the H. pylori–infected children with endoscopically proven DU have significantly higher serum IgG titers than do children without an ulcer. This information is similar to that of Andersen et al. (32), who suggested that an increased level of IgG to H. pylori measured by enzyme-linked immunoassay is a reliable predictor of peptic ulceration. These findings contrast with those of Drumm et al. (15) concerning the absence of a difference in specific anti–H. pylori IgG levels between pediatric patients with and without an ulcer. This discrepancy probably is the result of using different detecting antigen preparations. As a working hypothesis, we propose that H. pylori–associated peptic ulcer in childhood may be immunologically mediated (14). These data are not surprising in the view of the fact that “the more severe inflammation, the greater the chance of ulceration”(33). The increase in IgG production could lead to mucosal damage similar to an Arthus reaction (34). It has been postulated that H. pylori evolved lifestyles in which inflammation is a necessary condition for its survival (35). The more robust immune response to H. pylori in children with DU also may be related to the severity of gastric inflammation and bacterial load on the gastric mucosa (14,17). Furthermore, there was a correlation between the age of the infected patients and antibody titer, confirming the experience of Oliveira et al. (17) and De Giacomo et al. (13) and our own preliminary findings (14).
Recently, a medical position article reported a consensus conference on H. pylori infection in children and stated that, in the pediatric age group, there is no specific clinical picture of H. pylori infection, indicating a need to use noninvasive tests to screen children with dyspeptic symptoms for the infection (36). The new-generation enzyme-linked immunoassay test presented in our study showed an excellent correlation between serologic evidence of H. pylori infection and demonstration of the infection by other diagnostic methods. Other serologic tests (for example, indirect immunofluorescence test) that correlated well with H. pylori presence did not differentiate between the pediatric patients with and without DU (37). Many endoscopic procedures potentially could be replaced with the present enzyme-linked immunoassay serologic test as an effective screening procedure for the diagnosis of peptic ulceration in school-aged children with dyspeptic symptoms. Therefore, the detection of increased serum H. pylori–specific IgG titers may be a useful screening test in pediatric practice.
The authors thank Professor Phil Sherman, M.D., F.R.C.P.C., University of Toronto, and Giuseppina Oderda, M.D., Universitá del Piemonte Orientale, for the critical reading of this manuscript and for their helpful suggestions, and Dr. Emil Mingzhev, Children's Republic Hospital Ufa, Russia, for language and technical assistance.
1. Apley J. The Child With Abdominal Pains
. Oxford: Blackwell; 1959.
2. Andersen LP, Wewer AV, Christiansen KM, et al. The humoral immune response to Helicobacter pylori
infection in children
with recurrent abdominal pain. APMIS 1994: 102; 457–64.
3. Drumm B, Rhoads JM, Stringer DA, et al. Peptic ulcer disease in children
: etiology, clinical findings, and clinical course. Pediatrics 1988; 82: 410–4.
4. Prieto G, Polanco I, Larrauri J, et al. Helicobacter pylori
infection in children
: clinical, endoscopic and histologic correlations. J Pediatr Gastroenterol Nutr 1992; 14: 420–5.
5. Gormally S, Drumm B. Helicobacter pylori
and gastrointestinal symptoms. Arch Dis Child 1994; 70: 165–6.
6. Gormally SM, Prakash N, Durnin MT, et al. Association of symptoms with Helicobacter pylori
infection in children
. J Pediatr 1995; 126: 753–6.
7. Hardikar W, Feekery C, Smith A, et al. Helicobacter pylori
and recurrent abdominal pain in childhood. J Pediatr Gastroenterol Nutr 1996; 22: 148–52.
8. Bode G, Rothenbacher D, Brenner H, et al. Helicobacter pylori
and abdominal symptoms: a population based study among preschool children
in southern Germany. Pediatrics 1998; 101: 634–7.
9. Macarthur C, Saunders N, Feldman W. Helicobacter pylori
, gastroduodenal disease, and recurrent abdominal pain in children
. JAMA 1995; 273: 729–34.
10. Chong SK, Lou Q, Asnicar MA, et al. Helicobacter pylori
infection in recurrent abdominal pain in childhood: comparison of diagnostic tests and therapy. Pediatrics 1995; 96: 211–5.
11. Heldenberg D, Wagner Y, Heldenberg E, et al. The role of Helicobacter pylori
with recurrent abdominal pain. Am J Gastroenterol 1995; 90: 906–9.
12. Wewer AV, Andersen LP, Christiansen KM, et al. Is Helicobacter pylori
a cause of recurrent abdominal pain in children
>3 years [abstract]? Microb Ecol Health Dis 1991; 4: S110.
13. De Giacomo C, Lisato L, Negrini R, et al. Serum immune response to Helicobacter pylori
: epidemiologic and clinical applications. J Pediatr 1991; 119: 205–10.
14. Studenikin MY, Nijevitch AA, Mutalov AG. Consideration of Helicobacter pylori
infection in childhood: immune response, endoscopic and morphological findings. Acta Paediatr Jpn 1995; 37: 551–6.
15. Drumm B, Perez-Perez GI, Blaser MJ, et al. Intrafamilial clustering of Helicobacter pylori
infection. N Engl J Med 1990; 322: 359–63.
16. Booth L, Holdstock G, MacBride H, et al. Clinical importance of Campylobacter pyloridis
and associated serum IgG and IgA antibody responses in patients undergoing upper gastrointestinal endoscopy. J Clin Pathol 1986; 39: 215–9.
17. 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 group with and without duodenal ulcer
. J Pediatr Gastroenterol Nutr 1999; 28: 157–61.
18. Mc Callion WA, Bailie AG, Ardill JES, et al. Helicobacter pylori,
hypergastrinaemia and recurrent abdominal pain in children
. J Pediatr Surg 1995; 30: 427–9.
19. Nijevitch AA, Sataev VU, Alyangin VG, et al. “De-Nol test” for rapid diagnosis of Helicobacter pylori
infection. Acta Paediatr 1999; 88: 234–5.
20. Mitchell HM, Bohane TD, Tobias V, et al. Helicobacter pylori
infection in children
: potential clues to pathogenesis. J Pediatr Gastroenterol Nutr 1993; 16: 120–5.
21. Clayton C, Kleanthous K, Tabaqchali S. Detection and identification of Helicobacter pylori
by the polymerase chain reaction. J Clin Pathol 1991; 44: 515–6.
22. Kneepkens CMF. Peptic ulcer disease in childhood. Acta Endosc 1994; 24: 169–81.
23. Bourke B, Sherman P, Drumm B. Peptic ulcer disease: what is the role for Helicobacter pylori
? Semin Gastrointest Dis 1994; 5: 24–31.
24. Oderda G, Ansaldi N. Peptic ulcer in childhood. Lancet
25. Nijevitch AA, Khamidullina SV, Khamidullina FM. Childhood duodenal ulcer
associated with Helicobacter pylori
and ABO blood groups. Am J Gastroenterol 1999; 94: 1424–5.
26. Ashorn M, Maki M, Ruuska T, et al. Upper gastrointestinal endoscopy in recurrent abdominal pain of childhood. J Pediatr Gastroenterol Nutr 1993; 16: 273–7.
27. Oderda G, Vaira D, Holton J, et al. Serum pepsinogen I and IgG antibody to Campylobacter pylori
in non-specific abdominal pain in childhood. Gut 1989; 30: 912–6.
28. Conti-Nibali S, Sferlazzas C, Fera M-T, et al. Helicobacter pylori
infection: a simplified diagnostic approach. Am J Gastroenterol 1990; 85: 1573–5.
29. Stringer MD, Veysi VT, Puntis JWL, et al. Gastroduodenal ulcers in the Helicobacter pylori
era. Acta Paediatr 2000; 89: 1181–5.
30. Mikelsaar M, Mändar R, Maaroos H-I, et al. Helicobacter pylori
gastritis in children
: a clinical, bacteriological and morphologic study. Microecol Ther 1996; 22: 44–53.
31. Oderda G, Vaira D, Holton J, et al. Helicobacter pylori
with peptic ulcer and their families. Dig Dis Sci 1991; 36: 572–6.
32. Andersen LP, Rosenstock SJ, Bonnevie O, et al. Seroprevalence of immunoglobulin G, M and A antibodies to Helicobacter pylori
in an unselected Danish population. Am J Epidemiol 1996; 143: 1157–64.
33. Lee A. Future research in peptic ulcer disease. Scand J Gastroenterol 1994; 29 (suppl 205): 51–8.
34. Ernst P, Jin Y, Reyes V, et al. The role of the local immune response in the pathogenesis of peptic ulcer formation. Scand J Gastroenterol 1994; 29 (suppl 205): 22–8.
35. Blaser MJ. Helicobacters are indigenous to the human stomach: duodenal ulceration is due to changes in gastric microecology in the modern era. Gut 1998; 43: 721–7.
36. Drumm B, Koletzko S, Oderda G, European Paediatric Task Force on Helicobacter pylori
. Helicobacter pylori
infection in children
. Report of a Consensus Conference held in Budapest, September 1998. J Pediatr Gastroenterol Nutr
37. Rocha GA, Queiroz DMM, Mendes EN, et al. Serodiagnosis of Helicobacter pylori
infection in children
by an indirect immunofluorescence test. J Pediatr Gastroenterol Nutr 1993; 16: 247–1.