Association of Endoscopic Features of Gastric Mucosa with Helicobacter pylori Infection in Chinese Patients

The aim of this study is to identify and consolidate reliable endoscopic features associated with H. pylori infection in gastric mucosa, which is one of the major causes of gastric cancer. A total of 256 Chinese patients with symptomatic stomach disturbances were enrolled. Pathological examination was conducted using a light microscope and biopsy specimens stained with hematoxylin-eosin. Endoscopic examination was performed using a high resolution video endoscope. The association between endoscopic features and pathological H. pylori diagnosis was compared, and endoscopic features significantly associated with H. pylori infection were identified. A total of 14 endoscopic features were observed. Six of the 14 endoscopic features, including mucus on the gastric mucosa, diffuse redness, spotty redness of fundic mucosa, enlarged fold, mucosal edema, and RAC (type D and type I), were highly associated with H. pylori infection and were significantly sensitive and specific predictors for H. pylori diagnosis. The type R RAC was not significantly associated with H. pylori diagnosis. Our results indicate that conventional endoscopy features can be used to diagnose H. pylori in Chinese patients and can help determine the risk factor for gastric cancer.


Introduction
Gastric or stomach cancer typically originates from the mucus-producing cells on the inside lining of the stomach. As early symptoms are rare, it is often diagnosed at an advanced stage. Stomach cancer is more common in certain countries, such as China and Japan, than others, such as the United States [1].
The precise cause of gastric cancers remains unclear, but possible risk factors include smoking, high body mass index, genetic factors, and diets rich in salty or smoked foods [2][3][4]. Infection with Helicobacter pylori (H. pylori), a Gramnegative, microaerophilic, and spiral-shaped bacterium, is also known to be a common cause of gastric cancer [5]. The bacterium has evolved various mechanisms to promote its survival in the acidic environment of the stomach and can colonize the gastrointestinal tract [6].
H. pylori infection causes chronic inflammation of the gastric mucosa [7] and induces infiltration of mono-and polynuclear cells into the gastric mucosa. Persistent infection can induce atrophic changes and intestinal metaplasia. H. pylori infection contributes to a wide variety of upper gastrointestinal tract diseases, including gastroduodenal ulcer, gastric adenocarcinoma, gastric mucosal-associated lymphoid tissue lymphoma, and gastric hyperplastic polyps [8]. Successful eradication of H. pylori can improve gastritis and prevent H. pylori associated diseases [9]. Eradication of H. pylori can also prevent or delay development of precancerous lesions and gastric cancer [10].
H. pylori infection can be diagnosed using noninvasive tests such as antibody detection and urea breath test. H. pylori specific antibodies can be detected in whole-blood testing inexpensively, but the test has a relatively high false negative rate. The urea breath test uses 13 C and 14 C is more sensitive but also more expensive [11,12]. H. pylori antigens can be detected in the stool, with similar sensitivity and specificity to antibody testing [13].
Endoscopic atrophy closely correlates with gastric cancer [14,15]. Many risk factors for gastric cancer, such as H. pylori associated gastritis, gastric atrophy, or intestinal metaplasia [5][6][7][8][9][10], can be diagnosed using endoscopic inspection. If the gastric mucosa appears normal, with none of the aforementioned risk factors, lesions are less likely to be present. In these cases, magnified endoscopy may aid diagnosis.
While endoscopic inspection can identify risk factors for gastric cancer, the endoscopic features of gastric cancer, particularly those of H. pylori infection, have not been widely accepted [16][17][18][19]. This endeavor is confounded by the complex nature of many endoscopic features, which can be influenced by geographic locations and the ethnicity of patients. For example, features typical of gastric cancer in Asian patients may not be present in Caucasian patients [18]. Endoscopic features may also differ between Japanese and Chinese patients, despite the high incidence of stomach cancer in both populations.
In this study, we analyzed the association between endoscopic features and pathological diagnosis of H. pylori infection in Chinese patients. Our goal is to further consolidate the endoscopic diagnosis of H. pylori. To the best of our knowledge, this is the first such report concerning the endoscopic diagnosis of H. pylori infection in a Chinese population.

Patient Information.
A total of 256 patients, aged between 19 and 83 years, participated in this study, including 118 male and 138 female patients. The patients were admitted to the Affiliated Hospital of Qingdao University (Qingdao, China) with symptomatic stomach disturbances between October and December, 2015, and referred for endoscopic exam. These patients all resided in Shandong province, a peninsula near the Yellow Sea. All patients provided written informed consent prior to biopsy and endoscopic exam. This study was approved by the Ethics Committee of Qingdao University and was carried out in accordance with the Declaration of Helsinki.

Pathological Examination.
Biopsy specimens were collected from each patient at the following sites: the greater curvature of the antrum; the lesser curvature of the antrum; the lesser curvature of the angulus; the greater curvature of the middle corpus; and the lesser curvature of the middle corpus. Biopsy specimens were stained with hematoxylineosin (HE) and examined using a light microscope for the presence or absence of H. pylori infection [20]. Each patient was considered H. pylori positive if any of the biopsy areas contained H. pylori.

Endoscopic Examination.
Endoscopic exams were performed using a high resolution video endoscope (GIF-H290, Olympus Co., Japan). More than 50 endoscopic images were recorded at fixed sites in the esophagus, stomach, and duodenum. The following distinctive endoscopic features, based on the Sydney System [20], in images of the antrum, angulus, lesser and greater curvature of the lower body, greater curvature of the upper body, and cardia of the stomach were recorded: (1) Mucus on the gastric mucosa: determined by the visibility of mucus on the surface of gastric mucosa; it can be observed mainly in great curvature of the gastric body; it also can be seen in the nonatrophic mucosa of gastric body (1) fundic gland mucosa is soft, thick, and distended (2) pyloric gland mucosa is soft and convexoconcave Each case was determined by at least 2 well-trained endoscopists. Each patient was considered H. pylori positive if at least one of the significant endoscopic features (Table 1) was observed.

Statistical Analysis.
Diagnostic accuracy was investigated using MedCalc software (https://www.medcalc.org/). The 2 -test was used to evaluate endoscopic and histological parameters. The endoscopic diagnosis of H. pylori infection was compared to the histological diagnosis using the following statistical characteristics: specificity, sensitivity, positive predictive value (PPV), negative predictive value (NPV), the area under the curve of receiver operating characteristics (ROC/AUC), and the 95% confidence intervals (CI). < 0.05 was considered to indicate statistical significance. Fourteen of the endoscopic features defined, based on the Sydney System [6,7,21], were observed in our study, including (1) mucus on the gastric mucosa; (2) diffuse redness;
In this patient population, the most reliable endoscopic features of H. pylori diagnosis were spotty redness, enlarged fold, RAC, and mucosal edema (marked with an " * " in Table 2). They were all highly specific and sensitive indicators  (Table 2). Diffuse redness and mucus on the gastric mucosa were also highly associated with H. pylori infection, with 95% specificity and >50% sensitivity. All four types of erosion (flat, raised, hemorrhagic, and bleeding spot) were highly specific (>92%) but insensitive predictors of infection, possibly due Gastroenterology Research and Practice 5

Discussion
Endoscopic features have been reported to be useful tools for the diagnosis of H. pylori infection of gastric mucosa. However, the predictive capacity of these features may vary between ethnicities. In this study, we identified 14 gastric mucosa endoscopic features in 256 Chinese patients with strong symptomatic stomach disturbances. Consistent with a previous study [22], we found the endoscopic features spotty redness and enlarged fold to be highly associated with H. pylori infection. In addition, we found mucus on the gastric mucosa, diffuse redness, mucosal edema, and RAC to also be associated with H. pylori infection, suggesting that these features may be unique indicators of H. pylori infection in Chinese patients. Of the three RAC types, we found that a much higher rate of H. pylori infection was associated with type I and type D. This finding is consistent with a previous study in which magnifying endoscopy was used in a Chinese population [23], suggesting that the type I and type D RAC could be used as reliable endoscopic features for H. pylori diagnosis. It is also important to note that the type R RAC was not found to be associated with H. pylori infection, a finding that may reduce application of additional and unnecessary medical examinations.
It has been suggested that erosion is an indicator of absence of H. pylori infection [21]. In our study, about 55% of patients in which erosion was detected were H. pylori negative ( Table 2), indicating that erosion is not an accurate indicator of the presence or absence of H. pylori infection. Patch redness has also been suggested as an indicator of H. pylori eradication [24]. In this study we observed patchy redness in only 4 patients, and although all 4 patients were H. pylori negative, we feel that this number of cases is insufficient to draw a reliable conclusion.
H. pylori infection has been previously associated with increased incidence of both gastric ulcer and duodenal ulcer. Nonsteroidal anti-inflammatory drugs were reported to reduce the incidence of duodenal ulcer but not gastric ulcer [25][26][27]. In this study, gastric ulcers were only observed in 12 patients, and duodenal ulcers were only observed in 19 patients, so although our results are consistent with previous findings, again, this number of cases is insufficient to draw a reliable conclusion. This limitation also applies to several other endoscopic features, including patchy redness, hyperplastic polyp, red streak, xanthoma, fundic gland polyp, and hyperplastic polyp. In the future, studies enrolling a larger sample size, from multiple hospital facilities, may provide more informative results.
In summary, we identified 14 endoscopic features in 256 Chinese patients with symptomatic stomach disturbances. Our results suggest that at least 6 of the 14 endoscopic features, including mucus on the gastric mucosa, diffuse redness, spotty redness of fundic mucosa, enlarged fold, mucosal edema, and RAC (type D and type I), are highly associated with H. pylori infection and could be used as reliable indicators for H. pylori diagnosis. In addition, we found that type R RAC is likely not associated with H. pylori infection. This finding may help to reduce unnecessary medical examinations. Additional studies with larger sample sizes and more diverse populations may help to characterize the relationship between endoscopic features such as diffuse redness and erosion and H. pylori diagnosis. We believe that our results contribute to the development of rapid H. pylori infection diagnosis, one of the risk factors for gastric cancer.