UTILITY OF ENDOSCOPIC NARROW-BAND IMAGING FOR IDENTIFYING H. PYLORI-ASSOCIATED GASTRITIS IN CHILDREN

Utilidade da imagem endoscópica de banda estreita para identificar gastrite associada ao H. pylori em crianças

ABSTRACT

Background:  Conventional white light endoscopic (WLE) findings of H. pylori-associated gastritis are often non-specific and may not correlate with histology. Narrow band imaging (NBI), an optical digital technique employed for the visualization of vessels and patterns of gastric mucosa may improve identification. We evaluated the role of NBI in detecting H. pylori-associated gastritis and classifying its severity.

Methods:  Institution-based prospective observational study conducted between May 2021-October 2022. Children presenting with chronic abdominal pain (>1-month duration) were evaluated. Eligible children underwent gastroscopy with NBI and gastric biopsies for rapid urease test and histopathology. NBI gastroscopic findings were classified into five grades as per the classification by Alboudy et al. The association of NBI grade with the presence and severity of H. pylori gastritis on histopathology was analysed.

Results:   Ninety children (mean age 12.65±3.91 years), 52 (57.7%) males with median duration of symptoms of 4.5 (3-12) months underwent gastroscopy. H. pylori was detected on histopathology in 29 (32%) patients. NBI findings suggested a mucosal abnormality in 27/29 (93.1%) children with H. pylori on histopathology. H. pylori positive gastritis was significantly more common among those with higher (≥3) NBI grades as compared to those with lower NBI grades (61% vs10%, P<0.001). No significant association was found between NBI grade and the severity of H. pylori gastritis (P=0.75). NBI exhibited better sensitivity (0.82) compared to WLE (0.55) in identifying H. pylori-associated gastritis. On receiver operating characteristic curve analysis, NBI had higher area under curve (0.79 vs 0.65) as compared to WLE.

Conclusion:  NBI morphological pattern is a useful tool in identifying patients with H. pylori-associated gastritis.

Keywords:
Helicobacter; paediatric; gastritis; chronic abdominal pain

HIGHLIGHTS

• What is already known about this subject:

Endoscopy is routinely performed for the evaluation of H. pylori-associated diseases.

White light endoscopy (WLE) impression alone is often inaccurate and has poor sensitivity to identify H. pylori.

• What are the new findings:

Narrow band imaging (NBI) exhibits better sensitivity compared to WLE in identifying H. pylori-associated gastritis.

Patients with higher grades of NBI (≥3) were 14.2 times more likely to have H. pylori-positive gastritis on histopathology.

RESUMO

Contexto:  Os achados endoscópicos convencionais em luz branca (CLB) da gastrite associada ao H. pylori muitas vezes são inespecíficos e podem não correlacionar-se com a histologia. A imagem de banda estreita (IBE), uma técnica ótica digital empregada para a visualização de vasos e padrões da mucosa gástrica, pode melhorar a identificação. Avaliamos o papel da IBE na detecção da gastrite associada ao H. pylori e na classificação de sua gravidade.

Métodos:  Estudo observacional prospectivo baseado em instituição, conduzido entre maio de 2021 e outubro de 2022. Crianças apresentando dor abdominal crônica (>1 mês de duração) foram avaliadas. Crianças elegíveis foram submetidas a gastroscopia com IBE e biópsias gástricas para teste rápido de urease e histopatologia. Os achados gastroscópicos por IBE foram classificados em 5 graus, conforme a classificação de Alboudy et al. A associação do grau de IBE com a presença e gravidade da gastrite por H. pylori na histopatologia foi analisada.

Resultados:  Noventa crianças (idade média de 12,65±3,91 anos), 52 (57,7%) do sexo masculino, com mediana de duração dos sintomas de 4,5 (3-12) meses, foram submetidas a gastroscopia. H. pylori foi detectado na histopatologia em 29 (32%) pacientes. Os achados por IBE sugeriram anormalidade na mucosa em 27/29 (93,1%) crianças com H. pylori na histopatologia. A gastrite positiva para H. pylori foi significativamente mais comum entre aqueles com graus mais altos (≥3) de IBE em comparação aos com graus mais baixos de NBI (61% vs 10%, P<0,001). Nenhuma associação significativa foi encontrada entre o grau de NBI e a gravidade da gastrite por H. pylori (P=0,75). A IBE apresentou melhor sensibilidade (0,82) em comparação à CLB (0,55) na identificação de gastrite associada ao H. pylori. Na análise da curva característica de operação do receptor, a IBE apresentou maior área sob a curva (0,79 vs 0,65) em comparação à CLB.

Conclusão:  O padrão morfológico da IBE é uma ferramenta útil na identificação de pacientes com gastrite associada ao H. pylori.

Palavras-chave:
Helicobacter; pediatria; gastrite; dor abdominal crônica

INTRODUCTION

Helicobacter pylori (H. pylori) is one of the most prevalent infections worldwide, impacting nearly half the global population¹. In children suffering from chronic abdominal pain, the prevalence of H. pylori infection varies between 10% and 50%^2-5.

Early and precise diagnosis is important for successfully managing and preventing complications from persistent H. pylori infection. Several invasive and non-invasive tests are available to diagnose H. pylori infection.

Endoscopy and biopsy are the gold standards for diagnosing H. pylori infection. A number of factors including the size and location of biopsies, the stain utilised, the use of proton pump inhibitors (PPI), the use of antibiotics and the subjective assessment of the pathologists analysing the samples affect the accuracy of histopathology. The biopsy site is probably the most important factor among these variables.

Narrow-band imaging (NBI) is an endoscopic optical image enhancement technique that enhances the vessels and patterns of the mucosal surface and offers better pit patterns and vascular details. There is some published literature in adult patients to suggest that NBI findings helps in detecting abnormal mucosa harbouring H. pylori and improves the yield of the biopsies. However, there is a paucity of pediatric literature on this subject^1,6-11.

Adult data cannot be extrapolated to paediatric age-group because children often develop a tolerogenic immune response to H. pylori and have limited gastric inflammation in comparison to adults. A lower grade of inflammation will likely translate into a lesser degree of mucosal injury evident on endoscopic examination¹².

We hypothesized that more accurate identification of abnormal mucosa by NBI may enhance the ability of pediatric gastroenterologists to diagnose H. pylori-associated gastritis and improve the diagnostic yield of biopsy samples. Therefore, this study was conducted to study the role of NBI in detecting H. pylori-associated gastritis in children.

METHODS

This institutional-based, prospective observational study was conducted between May 2021 to October 2022. The study included children aged 5 to 18 years with chronic abdominal pain for whom endoscopy was advised as part of their evaluation. Chronic abdominal pain was defined as episodic or daily abdominal pain lasting for at least one month or more¹³. Children with acute abdominal pain and those with previously diagnosed organic or surgical conditions were excluded from the study.

Children having a history of use of proton pump inhibitors (PPI) or antimicrobials were advised to stop the medication and endoscopy performed after 2 and 4 weeks, respectively.

Gastroscopy was performed using an gastroscope (Olympus® GIF HQ 190 and Olympus® GIF H170). First, a conventional WLE mode was used, and findings of were noted. The gastroscope’s NBI mode was then enabled, and a thorough examination of the antrum, incisura, and fundus was performed. Image were recorded digitally, and still, pictures of the observation sites were saved. Conventional Narrow band images were classified into five morphological patterns according to the classification of Alaboudy et al. as follows - type 1: regular arrangement of collecting venules, type 2: cone-shaped gastric pits, type 3: rod-shaped gastric pits with prominent sulci, type 4: ground glass-like morphology, type 5: dark brown patches with bluish margins and irregular border⁶.

Image evaluation was done by a single experienced (>5 years using NBI) adult gastroenterologist who was blinded to the WLE and histopathology findings. Site-directed six biopsy samples were taken from the gastric antrum and body - five for histopathology and one from antrum for rapid urease test (RUT).

Histological demonstration of H. pylori was done using the Warthin-starry stain. A histopathological examination was performed by a single pathologist who was blinded to the endoscopic findings. The severity of gastritis was graded as per the Sydney classification¹⁴.

A H. pylori infection was diagnosed if the child had H. pylori bacteria on histopathology and had a positive RUT. NBI grades were dichotomized into two groups: Higher (grades 3, 4 & 5) and lower (1 & 2). We evaluated the association between H. pylori infection status and NBI gastric mucosal morpho­logic features. Additionally, the association between NBI grades and severity of gastritis was analysed.

Statistical analysis: continuous variables were expressed as median (range/interquartile range) and mean. Categorical variables were summarized as frequency and percentage (%). The chi-square test or ANOVA (>2 groups) were used to study the association among categorical variables. Fisher’s exact test and student’s t-test (Paired t-test, when applicable) were used to study the association in continuous variables. Receiver operating curves (ROC) were drawn to calculate the area under curve. A P-value <0.05 was considered statistically significant in this study.

Statistical analysis was performed using the IBM Statistical Package for the Social Sciences v.26.0 (SPSS, Armonk, NY; IBM Corp).

The study was approved by the hospital’s research ethics board and informed consent was obtained from all participants.

RESULTS

A total of 90 children were enrolled in the study, comprising of 52 males (57%), with a mean age of 12.65±3.91 years. The mean duration of abdominal pain was 4.5 months (IQR 3-12).

Conventional WLE findings were suggestive of H. pylori infection in 31 (34%) of the 90 subjects. The most common finding (n/%) was antral erythema (15/16.6%), followed by mucosal edema (7/7.7%), nodularity (6/6.6%), gastric ulcer (2/2.2%), and gastric erosion (1/1%).

NBI grades 1 to 5 were present in 30 (33%), 19 (21%), 28 (31%), 11 (12%) and 2 (2.2%) of subjects, respectively. (Figure 1) NBI higher grades were present in 39 (45.5%), while lower grades were present in 51 (54.4%) participants.

FIGURE 1
Narrow band imaging (NBI) Grading for H. pylori-associated gastritis - representative images from our cohort.

H. pylori-positive gastritis was seen in 32% (29/90) subjects. Mild, moderate, and severe H. pylori gastritis was present in 20 (69%),7 (24.1%), and 2 (6.9%) respectively.

On correlation of the WLE findings with the histopathological presence of H. pylori-associated gastritis, it was found that 17/29 (58.6%) of these children had findings suggestive of H. pylori on WLE. These were antral erythema (n=5), mucosal edema (n=3), antral nodularity (n=6), gastric ulcer (n=2) and gastric erosion (n=1). The remaining 12 children who were found to have H. pylori on histopathology had normal WLE. Amongst H. pylori negative children, 14 had an abnormal appearing mucosa (antral erythema - 10, mucosal edema -4) on WLE.

NBI findings suggested a mucosal abnormality in 27/29 (93.1%) children with H. pylori on histopathology. A linear association was found between increasing NBI grade and H. pylori gastritis with 6.6% HP gastritis positivity in the lowest NBI grade increasing to 100% in the highest NBI grade (P<0.01). (Table 1) Furthermore, H. pylori-positive gastritis was significantly more common in the higher NBI group as compared to lower NBI group (61% vs 10%, P<0.001).

TABLE 1
Association between narrow band imaging grading and the presence of H. pylori-associated gastritis.

	H. pylori positive		H. pylori
	(n=29)		negative (n=61)
	n	Percentage	n	Percentage
Grade 1	2	6.67	28	93.33
Grade 2	2	10.53	17	89.47
Grade 3	15	53.57	13	46.43
Grade 4	8	72.73	3	27.27
Grade 5	2	100	0	0

The unadjusted odd ratio for the association of NBI groups and H. pylori was 14.2. The patients with higher grades of NBI (≥3) were 14.2 times more likely to have H pylori-positive gastritis on histopathology (P<0.05).

No significant association was found between NBI grade and the severity of H. pylori gastritis. (Table 2) The higher NBI group had a higher proportion of patients with moderate-severe H. pylori gastritis, although the difference was not statistically significant. (8/24 vs 1/5, P=1.00).

TABLE 2
Association of NBI grades and severity of H. pylori gastritis.

Severity of gastritis				Total n
NBI	Mild	Moderate	Severe
Grade 1-2	4	1	0	5
Grade 3-5	16	6	2	24
P value	0.75

• NBI: narrow band imaging.

NBI exhibited better sensitivity (0.82) compared to WLE (0.55) in diagnosing H. pylori gastritis, while specificity was equivalent (0.75) for both methods. On Receiver operating characteristic curve analysis, NBI had a higher area under curve (AUC) 0.79 vs 0.65 as compared to WLE (Figure 2).

FIGURE 2
Receiver operating characteristic curve comparing white light endoscopy (WLE) and narrow band imaging (NBI) in detecting H. pylori-associated gastritis in children.

DISCUSSION

In this prospective study we found that the histopathological prevalence of H. pylori in our cohort of children with chronic abdominal pain was 32%. NBI findings, especially higher grades aided in identifying H. pylori-associated gastritis and fared better than WLE.

WLE findings that have been described in literature in patients with H. pylori gastritis include diffuse redness, enlarged gastric folds (mucosal edema), antral nodularity, gastric erosions and ulcerations. Amongst these findings antral nodularity has been described as the most sensitive and specific¹⁵. In our study we found that while antral erythema and mucosal edema were non-specific findings and were also found in nearly a quarter of children who were H. pylori negative, the presence of nodular mucosa in the antrum, gastric erosions or ulcerations heralded the presence of H. pylori with reasonable accuracy. However, these findings were present in only 31% (9/29) patients and 40% (12/29) had a normal appearing mucosa on WLE. This highlights the need for better techniques to improve the H. pylori diagnostic yield of an endoscopy.

We found that NBI helped in identifying H. pylori- associated gastritis with better accuracy than WLE. This study’s findings align with existing literature supporting NBI’s potential as a sensitive and specific method for detecting H. pylori infection. In a previous pediatric study carried out by Ozgur et al., 165 children were evaluated of which 56 (33.9%) were found to have H. pylori¹. The sensitivity, specificity and AUC of NBI for H. pylori detection was 92.8%, 62.4% and 0.77 respectively. However, in this study the authors did not compare WLE and NBI findings.

Among the NBI patterns, we found that the finding of a type 1 pattern or regular arrangement of collecting venules (RAC) helps in “ruling out” H. pylori infection, with 94% (28/30) of children with this finding being H. pylori negative. On the other hand, only 6% of H. pylori positive children showed RAC. This finding i.e. RAC is probably more important in children as compared to adults as RAC becomes less visible with increasing age and in an adult study, RAC was present in 78.4% of H. pylori negative individuals¹⁶.

Children with a higher NBI grade had a higher proportion of children with moderate-severe gastritis on histopathology. However, we did not find a statistically significant association between the NBI grade and the severity of gastritis. This is in contrast to Tongtawee et al. study of 200 adults in which the authors found a good statistical correlation between the gastric mucosal morphologic pattern and the severity of gastritis⁷. We speculate that a small sample size and a smaller proportion of children with severe inflammation (NBI grade 5 occurring in only 2.2% of patients versus 30% in their study) could have skewed our results (type 2 error). In the future, a study with a larger sample size may help in addressing this question with more certainty.

The strength of our study is that it is one of the few studies that address the role of NBI in H. pylori gastritis in children. We followed standard recommendations for obtaining biopsies and diagnosing H. pylori. Inter-observer variations of endoscopy and histopathological findings were avoided using a single gastroenterologist and histopathologist.

However, our study has some limitations too. In a single-centre study, the external generalizability of the results needs to be studied. Given the limited sample size, additional multi-centre studies may be needed to draw more definitive conclusions.

To conclude, NBI is an effective tool for diagnosing H. pylori infection in the pediatric population. We speculate that this technique can significantly enhance the detection of abnormal mucosa and direct targeted biopsies for H. pylori-associated gastritis during a gastroscopy.

REFERENCES

• ¹ Özgür T, Özkan TB, Erdemir G, Özakın C, Yerci Ö. The diagnostic value of endoscopic narrow band imaging in helicobacter pylori gastritis in children. Turk J Gastroenterol. 2015;26:112-6.
• ² Spee LA, Madderom MB, Pijpers M, van Leeuwen Y, Berger MY. Association between helicobacter pylori and gastrointestinal symptoms in children. Pediatrics. 2010;125:e651-69.
• ³ Das BK, Kakkar S, Dixit VK, Kumar M, Nath G, Mishra OP. Helicobacter pylori infection and recurrent abdominal pain in children. J Trop Pediatr. 2003;49:250-2.
• ⁴ Mohsen AAH, Manji AJ, Hashim JM, Obeid RM. The significance of Helicobacter pylori Infection as a cause of recurrent abdominal pain in children. Curr Pediatr Res. 2018;22:82-7
• ⁵ Singh M, Prasad KN, Yachha SK, Saxena A, Krishnani N. Helicobacter pylori infection in children: prevalence, diagnosis and treatment outcome. Trans R Soc Trop Med Hyg. 2006r;100:227-33.
• ⁶ Alaboudy AA, Elbahrawy A, Matsumoto S, Yoshizawa A. Conventional narrow-band imaging has good correlation with histopathological severity of Helicobacter pylori gastritis. Dig Dis Sci. 2011;56:1127-30.
• ⁷ Tongtawee T, Kaewpitoon S, Kaewpitoon N, Dechsukhum C, Loyd RA, Matrakool L. Correlation between Gastric Mucosal Morphologic Patterns and Histopathological Severity of Helicobacter pylori Associated Gastritis Using Conventional Narrow Band Imaging Gastroscopy. Biomed Res Int. 2015;2015:808505.
• ⁸ Tongtawee T, Dechsukhum C, Leeanansaksiri W, Kaewpitoon S, Kaewpitoon N, Loyd RA, et al. Improved Detection of Helicobacter pylori Infection and Premalignant Gastric Mucosa Using “Site Specific Biopsy”: a Randomized Control Clinical Trial. Asian Pac J Cancer Prev. 2015;16:8487-90.
• ⁹ Pimentel-Nunes P, Libânio D, Lage J, Abrantes D, Coimbra M, Esposito G, et al. A multicenter prospective study of the real-time use of narrow-band imaging in the diagnosis of premalignant gastric conditions and lesions. Endoscopy. 2016;48:723-30.
• ¹⁰ Liaquat Memon H, Taha Yaseen R, Ali Khalid M, Lail G, Shahzad S, Ul Haque MM, et al. Diagnostic Accuracy of Narrow-Band Imaging in Predicting Helicobacter pylori Gastritis in Patients With Dyspepsia. Cureus. 2024;16:e54756.
• ¹¹ Paulose A, Kurupasseril AJA, Kunnathuparambil SJ, Kunnathuparambil SG, Panakkal RP. Efficacy of endoscopy and NBI in diagnosing Hp in chronic gastritis patients presenting with significant dyspepsia: A prospective observational study. Int J Med Pub Health 2024;14:424-30.
• ¹² Reyes VE. Helicobacter pylori Immune Response in Children Versus Adults. Med Res Arch. 2022;10:3370.
• ¹³ Di Lorenzo C, Colletti RB, Lehmann HP, Boyle JT, Gerson WT, Hyams JS, et al. Chronic Abdominal Pain In Children: a Technical Report of the American Academy of Pediatrics and the North American Society for Pediatric Gastroenterology, Hepatology and Nutrition. J Pediatr Gastroenterol Nutr. 2005;40:249-61.
• ¹⁴ Dixon MF, Genta RM, Yardley JH, Correa P. Classification and grading of gastritis. The updated Sydney System. International Workshop on the Histopathology of Gastritis, Houston 1994. Am J Surg Pathol. 1996;20:1161-81.
• ¹⁵ Chatrangsun B, Vilaichone RK. Endoscopic Diagnosis for H. pylori Infection: White Light Imaging (WLI) vs. Image-Enhanced Endoscopy (IEE). Asian Pac J Cancer Prev. 2021;22:3031-8.
• ¹⁶ Glover B, Teare J, Patel N. Assessment of Helicobacter pylori status by examination of gastric mucosal patterns: diagnostic accuracy of white-light endoscopy and narrow-band imaging. BMJ Open Gastroenterol. 2021;8:e000608.

• Disclosure of funding:
  none
• Declaration of use of artificial intelligence:
  none

Edited by

• Associate Editor:
  Ricardo Guilherme Viebig.

Publication Dates

• Publication in this collection
  20 Dec 2024
• Date of issue
  2024

History

• Received
  07 July 2024
• Accepted
  23 Sept 2024