Published online May 14, 2021. doi: 10.3748/wjg.v27.i18.2238
Peer-review started: January 27, 2021
First decision: March 29, 2021
Revised: March 31, 2021
Accepted: April 23, 2021
Article in press: April 23, 2021
Published online: May 14, 2021
Processing time: 103 Days and 4.2 Hours
In Correa’s model of gastric carcinogenesis, Helicobacter pylori (H. pylori) infection, gastric atrophy and intestinal metaplasia (IM) are linked to gastric cancer development. The low level of serum pepsinogens (PG) was known to be highly associated with extensive atrophic gastritis.
High-resolution and magnifying narrow-band imaging (M-NBI) facilitate the detailed examination of gastrointestinal mucosa. M-NBI endoscopy can be used to diagnose H. pylori infection and classify gastritis by histological severity. Moreover, recent improvements in the resolution (> 1 million pixels) of gastrointestinal endoscopy have enhanced image quality, facilitating characterization of the gastric mucosal pattern. Close observation of the gastric corpus mucosa by standard endoscopy without magnification enables prediction of H. pylori gastritis and precancerous lesions.
To date, there was no comparative data regarding the usefulness of standard and M-NBI endoscopy for H. pylori infection and gastric precancerous conditions. We compared the diagnostic performance of standard and M-NBI endoscopy for H. pylori gastritis and precancerous conditions.
In total, 254 patients who underwent gastroscopy were prospectively enrolled. Standard endoscopy findings of the gastric mucosal surface were classified into mosaic-like appearance (type A), diffuse homogenous redness (type B), and irregular redness with groove (type C). Gastric mucosal patterns visualized by M-NBI endoscopy were classified as regular round pits with polygonal sulci (type Z-1), more dilated and linear pits without sulci (type Z-2), and loss of gastric pits with coiled vessels (type Z-3). We evaluated the utility of the two endoscopic classifications for the diagnosis of H. pylori gastritis, gastric atrophy, IM, and a serum PG I/II ratio of ≤ 3.
The diagnostic accuracy of standard and M-NBI endoscopy for H. pylori gastritis was 93.3% and 96.1%, respectively. Regarding gastric precancerous conditions, the diagnostic accuracy of standard and M-NBI endoscopy was 72.0% vs 72.6% for moderate to severe atrophy, and 61.7% vs 61.1% for IM in the corpus, respectively. Compared to type A and Z1, types B+C and Z-2+Z-3 were significantly associated with moderate to severe atrophy [odds ratio (OR) = 5.56, P = 0.001; OR = 8.67, P = 0.007] and a serum PG I/II ratio of ≤ 3 (OR = 4.48, P = 0.004; OR = 5.69, P = 0.029).
Close observation of the gastric corpus mucosa by standard and M-NBI endoscopy enables diagnosis of H. pylori infection and gastric precancerous conditions. Furthermore, our results suggest an association of endoscopic mucosal patterns with moderate to severe atrophy and a serum PG I/II ratio of ≤ 3.
By gastric mucosal observation in detail, optical diagnosis of H. pylori–related gastritis may be achieved in real time. In the future, a multicenter trial is required to confirm the reliability of our results.