The field of gastrointestinal (GI) endoscopy has witnessed rapid progress since the
introduction of the first fiberoptic endoscope more than 50 years ago.[1] While the initial focus of diagnostic luminal endoscopy was on making the device
more operator- and patient-friendly, and obtaining clearer and better quality white
light images, the last two decades has been marked by significant progress in technology
to appreciate more minute details of the mucosal surface (image-enhanced endoscopy
[IEE]).[2] This has opened up a new avenue of diagnostic possibilities in real-time.[3] Some of these technologies add contrast to image by manipulation of wavelength of
light and do not require dye spray (digital chromoendoscopy). Examples of digital
chromoendoscopy technology include narrow-band imaging (NBI, Olympus), i-scan (Pentax),
and blue light imaging (BLI, Fujifilm Corporation).[2] Since they provide a detailed, magnified view of the mucosal surface, the characteristics
of surface (pits, villi, etc.) and small vessels can be well-appreciated. However,
it requires training and practice to become familiar with the surface and vascular
patterns seen in normal mucosa and in the presence of lesions. Identifying features
associated with various pathological conditions and developing simple classifications
is one of the cornerstones in promoting uniform, effective, and wider use of IEE technology.
The quantum of research on IEE has been progressively increasing, indicating its growing
popularity and widening clinical application. This issue of the journal contains two
original research articles and one case report on clinical applications of IEE.[4]
[5]
[6] They highlight the potential of IEE in identifying lesions in different parts of
the GI tract.
Helicobacter pylori (H pylori) infection of the stomach results in chronic inflammation, and in a proportion of
patients, this may progress to atrophic gastritis, intestinal metaplasia and eventually
gastric cancer in a minority. These pathological sequelae of infection alter the surface
and vascular characteristics of the gastric mucosal surface which may be recognized
in real-time during IEE.[7]
[8] The normal gastric body mucosa is characterized by round pits, honeycomb shaped
subepithelial capillary network, and regular arrangement of collecting venules (CV).
The most consistent finding observed in H pylori infected stomach is the loss of CV.[7]
[9] In addition, the microsurface (pits), which are round in the body of stomach, may
enlarge in size or become elongated/tubular.[4]
[7] Taken together, these findings have shown a good accuracy in predicting H pylori infection.[10] The study by Balekudru et al5 investigates the utility of IEE (i-scan) in detecting
features associated with H pylori infection . In this prospective study, 68 patients with functional dyspepsia underwent
magnification white light endoscopy (M-WLE), followed by i-scan. The diagnosis of
H pylori infection was based on previously described surface and vascular patterns associated
with this infection.[7] Gastric biopsies were obtained from the greater curve of the body of stomach for
detecting H pylori infection (reference standard). i-scan had a sensitivity and specificity of greater
than 90% in detecting H pylori infection. Magnification white light endoscopy alone did not perform well in detecting
H pylori infection, although the criteria for diagnosing infection on M-WLE were not elaborated.
The lack of biopsy from gastric antrum is likely to underestimate the frequency of H pylori infection in this study, and the authors have acknowledged this limitation. Overall,
this study adds to the growing body of evidence on the utility of IEE for real-time
diagnosis of mucosal pathology. Do the findings from this and other similar studies
mean IEE can be used as a diagnostic tool for H pylori infection? Perhaps the answer currently is “no.” More data are required which compares
IEE with existing direct testing modalities for H pylori infection. Whether the mucosal pattern reverts to normal after successful eradication
of bacteria also needs to be studied further in order to differentiate current from
past infection.[11]
Another article by Desai et al[6] explores the performance of JNET classification in predicting histology of colonic
polyps IEE has shown excellent performance in differentiating non-neoplastic from
neoplastic polyps in the colon.[12] However, neoplastic polyps include adenoma with low- and high-grade dysplasia, superficial
cancer, and deep invasive cancer. Further characterization of neoplastic polyps into
the above pathological categories is essential to make decisions on management. A
widely used classification for this purpose is the NICE classification.[13] The type 2 category in this classification pools together adenomas with various
grades of dysplasia and superficial cancer together. The JNET classification improves
upon this by creating subgroups 2A (adenoma with low-grade intramucosal neoplasia)
and 2B (adenoma with high-grade intramucosal neoplasia or superficial cancer).[14] The authors applied JNET classification to images of 90 colonic polyps obtained
using NBI in 80 patients. The final diagnosis was based on histopathology. The hyperplastic
(non-neoplastic) polyps were accurately identified in > 95% cases. Among the neoplastic
polyps, positive predictive value of type 2B classification in predicting adenoma
with high-grade intramucosal neoplasia or superficial cancer was only 38%. Their observations
convey an important message; classifications developed by experts need to be validated
in clinical practice in different countries to confirm their utility. The authors
must be commended for this important work. In fact, a study of around 3000 colorectal
lesions form Japan showed a PPV of only 50.9% for type 2B lesions. Further work is
required in improving discrimination of neoplastic colorectal lesions.[15] In addition, sessile serrated adenomas, which have malignant potential but may appear
as non-neoplastic polyp on IEE pose another challenge in clinical application of NICE
or JNET classification.[16]
The third paper reports an interesting and rare case of Olmesartan-associated collagenous
gastroduodenitis.[11] The patient was a 54-year-old lady who presented with diarrhea, vomiting and weight
loss. NBI of duodenum showed tubular structures on surface of nodules with micromucosal
pattern. The patient recovered after stopping Olmesartan and repeat NBI showed normal
duodenal villi. IEE may be a useful tool in evaluation of patient with chronic small
bowel diarrhea and malabsorption, as it can show mucosal abnormalities in real-time
and help taking targeted biopsies.[17] However, it is not a substitute for mucosal biopsy as all malabsorptive conditions
may not show mucosal abnormality on imaging.
H pylori infection and colorectal polyps are common clinical conditions in the general population.[18] The clinician performing IEE should be familiar with the features associated with
these lesions.[19] Since H pylori infection is a precursor for gastric preneoplastic and neoplastic lesions, its presence
may warrant a more detailed and careful examination of the gastric mucosa. The challenge
currently is in identifying the target group of patients where IEE could be useful.[20] Similarly, colorectal polyps can be managed by endoscopic mucosal resection, endoscopic
submucosal dissection or referred for surgery, and IEE may help us in characterizing
the lesion to decide on the optimum treatment. The cost of the procedure, prolonged
procedure time and specialized training for interpretation of the enhanced images
are some of the potential barriers for clinical application of IEE. Recent guidelines
have started to define the role of IEE in clinical practice.[20] This, coupled with the ongoing research on artificial intelligence to interpret
endoscopic images, is likely to make image enhancement technology more popular and
user-friendly in the future.
