Study Populations
This prospective controlled, open-label, multicenter feasibility study was conducted at 9 hospitals in Beijing, China. This study was conducted according to the Standards for the Reporting of Diagnostic Accuracy Studies (STARD) initiative [(7)] and the Declaration of Helsinki. Out-patients aged 40 years or older at each center between March 2014 and June 2017 were considered for enrollment in this study. We excluded patients who underwent gastrectomy, had got the diagnosis of GC before, refused to take part in the study, and couldn’t accept the examination.
Informed consent was obtained from each participant, and the study was approved by the institutional review board of each participating hospital (2014/8/25-2014/9/30).
Moreover, the study were registered in Chinese Clinical Trial Register (ChiCTR:2000028733).
Study Design
Firstly, we surveyed the current status of conventional WLG procedures as the base-line data at each center. All clinical information and procedure details were collected, including preparation before examination, the time during operation(from insert into pharynx to withdrawal from pharynx), and photos taken during examination. Patients and endoscopists filled in the satisfaction questionnaire for each endoscopy examination. The endoscopically detected gastric lesions were recorded, as well as their pathological results including inflammation, intestinal metaplasia, intraepithelial neoplasia, and early and advanced staged GC.
Based on the endoscopy procedure experience and SSS in Japan, we set a standardized operation procedure for WLG after the discussion by more than endoscopy experts from all of the 9 centers. The standardized procedure can map the entire stomach without blind spots. The procedure includes 3 steps: 1) Preparation before the endoscopy. Patients were asked for fasting and water deprivation for more than 6 hours. Fifteen minutes before the procedure, patients were asked to drink a mixture of mucolytic and defoaming agents, which was 50ml water mixed with 8000U chymotrypsin or 10000U pronase, 1 g sodium bicarbonate, and 5 ml simethicone.
They also took the lidocaine mucilage for local pharyngeal anaesthesia before examination.
2) Rinsing out the mucus and froth from gastric mucosa during the endoscopic examination. 3) Standard operation procedure. Endoscopists were asked to take at least 34 + 1 pictures with careful observation during the entire examination. All the pictures included: 2 pictures for laryngeal part of pharynx when entry and exit. 5 + 1 pictures for esophagus, each for introitus oesophagi, upper segment (22 cm from incisor), middle segment (28 cm from incisor), lower segment (34 cm from incisor), and cardia (including dental line). +1 meant an esophagogastric junction picture when dental line migrating upward. 22 pictures for stomach (Fig.
1), including 4 quadrants of gastric antrum, lower body, middle-upper body with the antegrade view and fundus-cardia with retroflex view, and 3 quadrants of middle-upper body, incisura with retroflex. 5 for duodenum, including 4 quadrants of duodenal bulb and one picture of descending duodenum with papilla. These were the minimum required standard. If lesions were found in the examination, additional pictures could be taken.
Then we applied the standard procedure for white light endoscopy to out-patients at each center. All clinical and operation information were recorded. The endoscopically detected gastric lesions and the pathological results were recorded as mentioned above. Patients and endoscopists filled in the satisfaction questionnaires after each endoscopy examination. The satisfaction was graded into 5 scales, which were “very satisfied, satisfied, general, not satisfied, very dissatisfied” respectively and endowed 5,4,3,2,1 points. We made the comparative analysis between the base-line data and standard procedures.
Endoscopy Examination
All examinations were performed by endoscopic specialists who had been trained for more than two years on endoscopy in 9 institutes. All patients were offered local pharyngeal anesthesia with lidocaine mucilage before examination, without intravenous sedation drugs. We used the same types of endoscopy (GIF-XQ260, GIF-H260, Olympus Medical Systems, Tokyo, Japan) and high-resolution liquid-crystal monitors. We used a fixed structure enhancement setting and color tone for the video processor.
Pathologic Evaluation
Specimens for histological analysis were placed in 10% formalin solution and processed in the routine manner. The biopsy specimens were evaluated using H&E staining. The pathologist was blinded to the clinical and endoscopic findings. Histologic evaluation and diagnoses were performed at each center by two experienced pathologists according to WHO criteria for tumors [9] and the updated Sydney system for gastritis [10]. We used a central system of consultation with a main expert pathologist (Wei-xun Zhou). If an indeterminate lesion was encountered, it was scheduled to be reviewed by this consulting pathologist in making a final diagnosis.
Statistical Analysis
Statistical analysis was performed using SPSS version 17 software (SPSS, Chicago, IL). The continuous variables were expressed as mean ± standard deviation for normal distribution data or medians (25%, 75% quantile) for non-normal distribution data. Continuous data were compared using the independent sample t test or the Mann-Whitney U test. Categoric data were compared using the Pearson χ2 test (continuity corrected χ2 when minimum expected count was < 5; Fisher’s exact test was used when minimum expected count was < 1). All P values were two-tailed, and P value < 0.05 was considered statistically significant.