It is recognized that gastroscopy can miss intestinal metaplasia, dysplasia and early gastric cancer. This could conceivably be due to the fact that these lesions may only present as subtle mucosal changes on conventional white light endoscopy (WLE) and thus be easily missed. In narrow band imaging (NBI) a rotating interference narrow band filter is interposed after the xenon light source such that when the NBI mode is switched on, discrete blue and green wavelengths are used and this improves mucosal surface contrast and facilitates visualization of mucosal details. A new NBI system is available that allows brighter illumination. We hypothesize that bright -NBI is superior to WLE in detecting focal gastric lesions such as gastric intestinal metaplasia, dysplasia and early gastric cancer in subjects undergoing gastroscopy.

Background It is recognized that gastroscopy can miss intestinal metaplasia, dysplasia and early gastric cancer (EGC)[1-3]. This could conceivably be due to the fact that EGC, as well as premalignant changes, may only present as subtle mucosal changes on conventional white light endoscopy (WLE) and thus be easily missed.

Chromoendoscopy has been used routinely in some centers during upper gastrointestinal (UGI) endoscopy in an attempt to improve mucosal contrast and detect subtle mucosal lesions[4]. However, this is not the norm in most centers due to the inconvenience of dye spray. Narrow band imaging (NBI) facilitates the examination of the mucosal surface without a need for chromoendoscopy. It is based on the principle that the depth of penetration of light is wavelength dependent, with shorter wavelengths resulting in more superficial penetration i.e. blue light penetrates most superficially (mucosal imaging) while red light penetrates the deepest (submucosal imaging). In NBI a rotating interference narrow band filter is interposed after the xenon light source such that when the NBI mode is switched on, discrete blue and green wavelengths are used and this improves mucosal surface contrast and facilitates visualization of mucosal details[5]. Magnifying endoscopy (ME) is useful in assessing the gastric mucosal pit pattern and microvessel of gastric mucosal lesions, thus providing the possibility of predicting the histological nature of the mucosal lesion. Studies from Japan, a country with a high population risk for gastric cancer, which were based on preselected high risk patients or patients with known precancerous lesions or EGC, showed that NBI and NBI-ME could diagnose precancerous lesions such as intestinal metaplasia (IM)[6] as well as EGC[7-11].

However, there is a lack of published data on the clinical utility of NBI in the detection of gastric intestinal metaplasia, dysplasia and EGC in the unselected general population. A recent study from Singapore looked at the utility of NBI and NBI-ME in patients aged 35 to 70 years undergoing diagnostic UGI endoscopy and found that the detection rate of focal lesions was increased from 43.7% to 58.8% with NBI, compared to WLE. Among the additional lesions detected, NBI-ME identified 97.1% as IM, 1.4% as EGC and 1.4% as benign. The EGC was a synchronous cancer missed by WLE but detected by NBI[12].

The current NBI system is limited by the dark endoscopic view. This has the potential limitation of making the detection of subtle lesions more difficult and lesions could be missed. A new prototype NBI system has been designed that has a brighter appearance. This prototype bright NBI coupled with high definition resolution is likely to overcome this drawback of original NBI.

Hypothesis We hypothesize that bright -NBI is superior to WLE in detecting focal gastric lesions such as gastric intestinal metaplasia, dysplasia and EGC in subjects undergoing diagnostic upper GI endoscopy.

Materials and methods

1. Setting and trial design:

   A randomized controlled study.

2. Subjects:

   Inclusion criteria: 1) Subjects aged > 50 years undergoing diagnostic or screening upper GI endoscopy; 2) ability to provide a written consent to trial participation. Exclusion criteria: 1) presence of active gastrointestinal bleeding; 2) presence of coagulopathy precluding biopsies; 3) absence of informed consent.

3. Informed Consent:

   The protocol will be approved by local Institutional Review Board. Trial subjects will sign informed consent prior to UGI endoscopy.

4. Interventions:

   Subjects are randomized to undergo UGI endoscopy in either WLE or bright light NBI modes, using a high definition gastroscope (Olympus 190 or 290 series). An endoscopy assistant will then open the next numbered sealed envelop at the instruction of the endoscopist. For subjects randomized to WLE, the entire examination will be performed using white light. For subjects randomized to NBI, the NBI mode will be switched on before insertion into the stomach. Once the gastroscope is inserted into the duodenum, the endoscopist can switch back to WLE to examine the duodenal mucosa. Before withdrawal from the duodenum back into the stomach, the NBI mode will be switched on and gastric mucosal examination carried out in the NBI mode. After the gastric examination findings using NBI have been recorded, the endoscopist is allowed to re-examine the gastric mucosa using white light, but any additional findings will not be recorded in the study data sheet. During withdrawal into the esophagus, the endoscopist can switch back to WLE to examine the esophageal mucosa.

   Recorded parameters include 1) presence or absence of focal lesions and the suspected diagnosis such as a) IM; b) cancer; c) early cancer; d) gastric ulcer or erosion; e) other focal lesions. Switching from WLE to NBI and vice versa is allowed at the discretion of endoscopist after detection of a lesion. Endoscopists may elect to remove these lesions using WLE. Magnification is allowed in either group. IM is diagnosed on white light endoscopy based on by whitish color change with plaques, patches, or homogeneous discoloration on the gastric mucosa[13].High definition white light endoscopy may possibly visualize the villous pattern. IM is diagnosed on NBI based on the criteria such as light blue crest and villous morphology[14, 15]. Early cancer may present on white light endoscopy with varied non-specific features such as a subtle polypoid protrusion, a superficial plaque, ulcer, depression, or even a mucosal discoloration. However distortions of the mucosal pit pattern and microvascular pattern on NBI have been found to be characteristic[6].

   Lesion size is measured by a comparison to known diameter of open forceps and grouped into size categories. The decision to perform hot/ cold biopsies or snaring is left to the endoscopist. Each lesion is retrieved separately for pathologic examination.

   Recognizing that the incidence rate of cancer may be low from the study populations[16], the decision was made to use as a surrogate intestinal metaplasia, a known premalignant lesion, as the main outcome variable.

5. Statistics:

   1. Methods of randomization:

      Randomization sequence will be generated using a computer program in blocks of 20 of equal assignment pertaining to individual endoscopists who participate in the trial.

   2. Primary outcome variables:

      Detection rate of IM

   3. Secondary outcome variables:

      1. Detection rate of gastric cancer
      2. Aggregate detection rate of focal gastric lesions The diagnostic performance of the imaging modality will form ancillary analyses. Histology will be the gold standard.

   4. Sample size calculation:

   In the UGI endoscopy study that compared NBI with WLE in the general population, it was found that most focal lesions were due to IM and that after age 50 years, the prevalence rate of IM was 36%[12]. In this study, WLE detected IM in 4.4% of cases, whereas NBI detected IM in 16.2% of cases. We arbitrarily assumed that HD WLE increase the detection rate of IM to 10%, while keeping the IM det4ection rate by NBI to 16.2%. We determine our sample size to detect a 6.2% difference with a power of 80% and a two sided type 1 error of 0.05. A total of 464 patients will be required.

6. Timetable of work:

The study will continue until the intended subject size of 464 subjects is reached. .

References

1. Hosokawa O, Tsuda S, Kidani E, et al. Diagnosis of gastric cancer up to three years after negative upper gastrointestinal endoscopy. Endoscopy 1998; 30:669-74.
2. Yalamarthi S, Witherspoon P, McCole D, Auld CD. Missed diagnoses in patients with upper gastrointestinal cancers. Endoscopy 2004; 36:874-9.
3. Raftopoulos SC, Segarajasingam DS, Burke V, et al. A cohort study of missed and new cancers after esophagogastroduodenoscopy. Am J Gastroenterol 2010; 105:1292-97.
4. Kiesslich R, Neurath MF. Magnifying chromoendoscopy for the detection of premalignant gastrointestinal lesions. Best Pract Res Clin Gastroenterol. 2006; 20: 59-78.
5. Kuznetsov K, Lambert R, Rey JF. Narrow band imaging: potential and limitations. Endoscopy 2006; 38:76-81.
6. Yao K, Anagnostopoulos GK, Ragunath K. Magnifying endoscopy for diagnosing and delineating early gastric cancer. Endoscopy 2009; 41:462-7.
7. Kato M, Kaise M, Yonezawa J, et al. Trimodal imaging endoscopy may improve diagnostic accuracy of early gastric neoplasia: a feasibility study. Gastrointest Endosc 2009; 70:899-906.
8. Kaise M, Kato M, Urashima M, et al. Magnifying endoscopy combined with narrow band imaging for differential diagnosis of superficial depressed gastric lesions. Endoscopy 2009; 41:310-5.
9. Yokoyama A, Inoue H, Minami H. Novel narrow band imaging magnifying endoscopic classification for early gastric cancer. Dig Liver Dis 2010; 42:704-8.
10. Nakamura M, Shibata T, Tahara T, et al. The usefulness of magnifying endoscopy with narrow band imaging to distinguish carcinoma in flat elevated lesions in the stomach diagnosed as adenoma by using biopsy samples. Gastrointest Endosc 2010; 71:1070-5.
11. Ezoe Y, Muto M, Horimatsu T, et al. Magnifying narrow band imaging versus magnifying white light imaging for the differential diagnosis of gastric small depressive lesions: a prospective study. Gastrointest Endosc 2010; 71:477-84.
12. Ang TL, Fock KM, Teo EK, Tan J, Poh CH, Ong J, Ang D. The diagnostic utility of narrow band imaging magnifying endoscopy in clinical practice in a population with intermediate gastric cancer risk. Eur J Gastroenterol Hepatol. 2012;24(4):362-7Lin BR, Shun CT, Wang TH, Lin JT. Endoscopic diagnosis of intestinal metaplasia of stomach: accuracy judged by histology. Hepatogastroenterology 1999;46:162-166.
13. Uedo N, Ishihara R, Iishi H, Yamamoto S, Yamamoto S, Yamada T, Imanaka K, Takeuchi Y, Higashino K, Ishiguro S, Tatsuta M. A new method of diagnosing gastric intestinal metaplasia: narrow-band imaging with magnifying endoscopy. Endoscopy. 2006 Aug;38(8):819-24.
14. Pimentel-Nunes P, Dinis-Ribeiro M, Soares JB, Marcos-Pinto R, Santos C, Rolanda C, Bastos RP, Areia M, Afonso L, Bergman J, Sharma P, Gotoda T, Henrique R, Moreira-Dias L. A multicenter validation of an endoscopic classification with narrow band imaging for gastric precancerous and cancerous lesions. Endoscopy 2012;44(3):236-46.
15. Fock KM, Ang TL. Epidemiology of Helicobacter pylori infection and gastric cancer in Asia. J Gastroenterol Hepatol 2010;25:479-486.

Inclusion Criteria:

1. Subjects aged > 50 years undergoing diagnostic or screening upper GI endoscopy
2. ability to provide a written consent to trial participation.

Exclusion Criteria:

1. presence of active gastrointestinal bleeding
2. presence of coagulopathy precluding biopsies
3. absence of informed consent.