Advanced imaging: the impressive success story of gastrointestinal endoscopy
26 February 2019 (online)
As we celebrate 50 years of Endoscopy, it is almost 70 years since the gastrocamera was first introduced. In 1950 it became possible to visualize the inside of the stomach by a stiff and long instrument with a photocamera on the tip. This gastrocamera was developed by Uji in collaboration with engineers from Olympus Optical Company of Tokyo, Japan. Imaging was not performed in real time, but pictures were taken blindly in a structured manner and then developed afterwards. This method meant that certain parts of the stomach were very difficult to visualize. In the subsequent years, the gastrocamera was gradually developed further, and in 1972, a new prototype was presented by Chang, which included two bending sections and other new features such as a close-up warning system and a fixed-volume air-pumping system . In the meantime, a fiberoptic endoscope was developed, enabling real-time visualization of the gastrointestinal (GI) tract. However, quality was still quite poor and only one or two endoscopists could view at the same time. An extra side-viewer was needed for the second examiner, but this reduced the brightness of the view as this was divided into two. Even with this rather simple fiberoptic instrument, magnification became possible .
In the 1980 s, video endoscopy became possible and mainstream. Endoscopic images were displayed on a separate screen. Video recording became possible and thus teaching was substantially improved. Today, the quality of the imaging has further improved and high definition is possible. In addition, extra imaging techniques such as the use of dye and specific light techniques have become standard of care. These techniques all aim to improve the detection and characterization of abnormal mucosal patterns in the GI tract. The success story of cancer prevention with GI endoscopy ran in parallel with these technical innovations.
In this editorial, we look back to the start of detailed mucosal inspection during endoscopy. We would like to highlight a landmark paper in this area that was published back in 1974. It was written by Kohli et al., and titled: “Minute endoscopical findings of duodenal mucosa using the dye scattering method” . With the aim of improving the detection of subtle lesions, the authors described the scattering of indigo carmine or methylene blue solution directly onto the duodenal surface through a teflon tube that was inserted through the fiberscope. They applied a large quantity to wash away the duodenal juice, and the superfluous solution was subsequently removed by changing the patient’s position or using suction. In a study using this technique in 173 patients, the authors recognized ulcers and ulcer scars but also minute changes such as Brunner’s gland hyperplasia and micro-erosions in the duodenum. This is one of the first publications on the method of chromoendoscopy, which is still used today. Its use for lesion detection is advocated for dysplasia surveillance in patients with long-standing inflammatory bowel disease, and these techniques are widely used for lesion characterization, especially in Japan.
However, as the use of extra dye spray takes experience, and extra time and effort, its application in daily practice remains limited. Around 2000, endoscopic engineers developed digital chromoendoscopy, where the effect of chromoendoscopy is simulated by a change in light or application of light filters, or the images are adapted in a post-processing manner. This is an easy applicable technique as it demands just a switch of a button on the working shaft of the endoscope, and today most new endoscopes are equipped with such a technique. Although the use of digital chromoendoscopy has not demonstrated increased lesion detection, it has become standard of care for lesion characterization in the upper and lower GI tract. Different classifications for optical diagnosis have emerged in the East and in the West, often used in combination with magnification or near-focus systems. Accurate lesion assessment is important for decision making in treatment management, especially in assessing whether a suspicious lesion is invasive (i. e. cancer) or not. As the arsenal of endoscopic resection techniques is rapidly evolving and treatment of most early GI lesions without deep submucosal invasion seems safe, this is an area of increasing interest. The most frequently used application of optical diagnosis, however, is for diminutive colonic polyps. Here, optical diagnosis is applied in the so-called “resect and discard” and “diagnose and leave in” strategies. Safe introduction of these strategies could result in significant cost reductions, increased efficiency, and reduced burden of colonoscopy for both doctors and patients. However, as interobserver variability in accuracy of optical diagnosis between endoscopists is quite large, specific training and monitoring are necessary to achieve and secure competency .
Would it be useful to assess mucosal GI lesions in even more detail during endoscopy, as the pathologist would do after resection? This technique was developed and introduced about a decade later – microscopic examination, such as endocytoscopy and confocal laser microscopy . Images are impressive, and malignant cells can be readily identified. However, the true clinical value of the system is still under debate.
With recent breakthroughs in artificial intelligence, computer-aided diagnosis for GI endoscopy is rapidly evolving. It could be used for automated lesion detection as well as achieving accurate optical diagnosis, with the potential to reduce or even abolish interobserver variability. The first data on automated on-site optical diagnosis of lesions in the upper and lower GI tract are promising, and indeed if performance meets expectations in daily practice, this technique might be a new game changer in the ever-changing and exciting field of endoscopy.
In this short editorial we are happy to reflect on the breakthroughs of imaging in GI endoscopy. Thanks to all innovators in this field, GI endoscopy has become a true success story in patient care and this interesting field of medicine will surely rapidly evolve further.
- 1 Chang FM, Chi PH, Ashizawa S. Gastrophotography of the upper part of the stomach. Endoscopy 1972; 4: 125-133
- 2 Sasaki N, Iida Y, Okazaki Y. et al. Magnifying endoscopic observation of the gastric mucosa, particularly in patients with atrophic gastritis. Endoscopy 1978; 10: 269-274
- 3 Kohli Y, Nakajima M, Ida K. et al. Minute endoscopical findings of duodenal mucosa using the dye scattering method. Endoscopy 1974; 6: 1-6
- 4 Vleugels JLA, Dijkgraaf MGW, Hazewinkel Y. et al. Effects of training and feedback on accuracy of predicting rectosigmoid neoplastic lesions and selection of surveillance intervals by endoscopists performing optical diagnosis of diminutive polyps. Gastroenterology 2018; 154: 1682-1693
- 5 Neumann H, Kiesslich R, Wallace MB. et al. Confocal laser endomicroscopy: technical advances and clinical applications. Gastroenterology 2010; 139: 388-392.e1-2