Anna M. Buchner1, Prateek Sharma2, and Michael B. Wallace3 1 University of Pennsylvania School of Medicine, Philadelphia, PA, USA 2 Veterans Affairs Medical Center, University of Kansas School of Medicine, Kansas City, KS, USA 3 Mayo Clinic, Jacksonville, FL, USA Over the last decade, a major role of digestive endoscopy has become the early detection and prevention of gastrointestinal (GI) malignancies. This has been achieved through accurate detection of precursor lesions, including the most subtle and minute ones and their subsequent resection. Therefore, current diagnostic endoscopy aims at detection and characterization of all lesions and confirmation of their neoplastic potential. The current advances in endoscopy have provided endoscopists with novel modalities to reach those goals. Digestive endoscopy has evolved from standard white light endoscopy (WLE) to contrast‐enhanced endoscopy. Conventional white light video endoscopy, which has been used initially as a primary diagnostic modality, was found to be associated with a disproportionate miss rate for subtle lesions (e.g., flat adenomas). Numerous studies have demonstrated that even experienced gastroenterologists miss up to 6% of advanced adenomas and 30% of all adenomas [1, 2]. Since subtle dysplastic and early neoplastic lesions may be too small, flat, or depressed to be detected during regular standard WLE, new contrast‐enhanced endoscopic technologies have been introduced to maximize their detection. In experienced hands, these methods truly represent a significant advance in endoscopy practice and may improve the diagnostic yield of significant lesions, decrease the burden of insignificant biopsies, and lead to appropriate on‐ table decisions, such as endoscopic resection of malignant lesion and leaving lesions in situ if they are nonneoplastic. The training in contrast‐enhanced endoscopic technologies has become a critically important issue, and it has to be undertaken in order to maximize the clinical usefulness of these new technologies. Contrast‐enhanced endoscopy includes the following image‐enhanced endoscopy (IEE) technologies: These technologies are used in conjunction with high‐resolution and high‐definition (HD) endoscopes. High‐resolution endoscopes (HRE) with high‐density couple‐charged device (CCD) (600,000–1,000,000 pixels per CCD) produce high‐resolution images with increased spatial resolution for the detection of minute abnormalities in mucosal glandular and vascular structures. In conjunction with a movable lens for magnification endoscopy, the focal distance may be controlled to allow detailed examination of the mucosal surface at close range (<3 mm). This new generation of endoscopes can provide enlargement of the image up to 115× as compared with 30× with standard endoscopes and can be used in conjunction with contrast‐enhancement technologies such as chromoendoscopy/NBI techniques in diagnosing neoplastic and nonneoplastic lesions, detecting flat and subtle lesions, and evaluating the depth of malignant lesions. Some of these techniques can provide in vivo microscopic evaluation of the GI mucosa known as optical biopsy obtained in real time. Furthermore, over the last decade, HD endoscopy has become a new norm replacing older standard definition (SD) endoscopy systems. High‐definition systems have been recognized as an essential tool for lesion detection with 2–4% gain in adenoma detection rate in average risk patients compared with SD [4, 5]. The use of HRE has been combined with chromoendoscopy (dye staining) in an attempt to improve detection of mucosal abnormalities. The key point in chromoendoscopy is applying various dyes to visualize subtle GI lesions and to define surface‐staining patterns, with subsequent targeted biopsies of those lesions. Kudo pit pattern classification for colonic lesions became widely adopted, and it allows to distinguish five types of staining pattern of the mucosa [6]. Type 1 with round pits and type 2 with stellate pits represent nonneoplastic lesions, whereas type 3 tubular pits, type 4 gyrus‐like pits, and type 5 irregular pits correspond to neoplastic lesions (Figure 15.1). Various studies from the early 2000s in the pre‐HD era showed that chromoendoscopy (methylene blue or indigo carmine) targeted biopsies increased the diagnostic yield for dysplastic lesions, including flat lesions, in patients with ulcerative colitis and Crohn’s colitis as opposed to using standard colonoscopy [7–10]. Therefore, chromoendoscopy has become part of the surveillance guidelines in patients with long‐standing inflammatory bowel disease (IBD) colitis especially when performed by experts and in high‐risk IBD patients with a prior history of dysplasia, family history of colorectal cancer, and history of primary sclerosing cholangitis (PSC) [11–16]. In the United States, chromoendoscopy with HD colonoscopy in patients with long‐term IBD has been the suggested method of surveillance introduced by the SCENIC (Surveillance for Colorectal Endoscopic Neoplasia Detection and Management in Inflammatory Bowel Disease Patients: International Consensus Recommendations) international consensus statement and endorsed by the American Society for Gastrointestinal Endoscopy (ASGE) and American Gastroenterological Association (AGA) [14, 15, 17]. Subsequently the recent ACG clinical guideline on management of Crohn’s disease in adults has also recommended the use of chromoendoscopy in patients at particular high risk for colorectal neoplasia including patients with PSC and personal history of dysplasia [13]. At the present time, chromoendoscopy with HD colonoscopy systems has been primarily utilized in tertiary referral centers, although efforts for its broader implementation have been undertaken [14, 15, 18, 19]. The SCENIC consensus statement provides a summary of the available data for the management of dysplasia and its surveillance in patients with IBD [14, 15]. Additionally, it addresses how to perform surveillance colonoscopy for the detection of dysplasia and how to describe and manage visualized dysplasia. The SCENIC consensus statement recommended standard terminology to describe dysplasia and replaced confusing terms such as dysplasia‐associated lesion mass and adenoma‐like mass. Lesions and their features are now described according to the new SCENIC classification system. They are divided into polypoid or nonpolypoid lesions using the modified Paris classification, the presence of ulcer and clarity with which the dysplasia border can be seen. The Kudo pit pattern classification is recommended to classify nonneoplastic, superficial, and deeply invasive neoplasia [6, 14, 15, 20]. Decisions regarding endoscopic resection or surgical management of the lesion are made based upon these endoscopic features. The most recent systematic reviews and meta‐analyses demonstrated that chromoendoscopy was superior to SD WLE for lesion detection but chromoendoscopy did not lead to increased lesion detection when compared with HD white light colonoscopy [21, 22]. In the United States, chromoendoscopy with targeted biopsies remains the preferred method for dysplasia surveillance and dysplasia characterization and is performed primarily in tertiary academic centers by trained endoscopists in high‐risk patients with IBD, including long‐term IBD and prior history of severe colitis requiring escalation therapy, dysplasia, or PSC. However, random biopsies plus targeted biopsies remain an alternative when chromoendoscopy is not available or optimal (e.g., poor bowel preparation, presence of pseudopolyps, active inflammation). Further studies are still needed to address reimbursement issues and training processes prior to the broad implementation of chromoendoscopy in routine clinical practice in the United States. Both traditional chromoendoscopy with 2% Lugol’s iodine [23] and virtual chromoendoscopy methods such as NBI [24] have been shown to improve detection of squamous cell neoplasia in the esophagus. The methods for Lugol’s application are similar to other chromoscopy techniques but are contraindicated in individuals with iodine allergy. Lugol’s is a negative stain for squamous dysplasia; thus normal epithelium stains dark brown, whereas dysplasia does not stain (Figure 15.2). In NBI, the key pattern is development of increased and distorted intrapapillary capillary loops (IPCLs) (Figure 15.3). Several investigators reported the improved detection of intestinal metaplasia and dysplasia in Barrett’s esophagus (BE) using dye‐based chromoendoscopy methods compared with a random biopsy protocol [25–31]. Sharma et al. described three mucosal patterns visualized with indigo carmine and magnification endoscopy in patients with BE (ridged/villous, circular, irregular/distorted) with the ridged or villous patterns found to be associated with intestinal metaplasia while the irregular or distorted pattern with Barrett’s high‐grade dysplasia (HGD) or superficial adenocarcinoma [32]. The presence of the ridged or villous pattern had high sensitivity, specificity, and positive predictive value (97, 76, and 92%, respectively) for detecting Barrett’s metaplasia [32] However, some studies using methylene blue spraying have failed to demonstrate a detection benefit for either Barrett’s metaplasia or dysplasia [25, 31]. There has also been a report that raises the issue of DNA damage resulting from methylene blue staining and white light illumination although these have more recently been refuted [33, 34]. Similar conflicting results have been found with studies using acetic acid (a mucolytic agent that alters cellular protein structure) and crystal violet staining [35–37]. Therefore, the initial enthusiastic results have been found later to vary, possibly due to differences in technique, operator experience, and the prevalence of BE within the particular patient population under investigation [38, 39]. Further use of the chromoendoscopy techniques was also described in other upper GI pathologies, including gastric cancer and esophageal cancer [40, 41]. The main value of those techniques, however, appears to be detection of small and flat lesions not seen by conventional colonoscopy and furthermore distinction between neoplastic and nonneoplastic lesions [42–44]. A study by Soetikno confirmed the high prevalence of flat, nonpolypoid colorectal lesions detected with the use of dye‐based chromoendoscopy [44]. The application of dye‐based chromoendoscopy appears to be most beneficial in high‐risk populations including patients with hereditary syndromes such as Lynch syndrome. Various studies have demonstrated that chromoendoscopy improved adenoma detection rates in patients with Lynch syndrome [16, 45]. In addition, pancolonic topical dye spraying has been also evaluated for detection of adenomas and serrated lesions during a routine colonoscopy exam. Large randomized controlled trials confirmed significant increase in adenomas and a nearly significant increase in advanced lesions [46, 47]. A systemic review and meta‐analysis of seven randomized controlled trials comparing colonic lesions detection in patients assigned to dye chromoendoscopy demonstrated higher adenoma detection rates, OR 1.53 (95% CI 1.31–1.79) with dye‐based chromoendoscopy [48]. Despite its potential for colonic diseases and all the recent efforts for the standardization and implementation, chromoendoscopy has not been widely adopted into general clinical practice in the United States given primarily increased procedure time and lack of reimbursement. As of now it remains the preferred method used primarily in high‐risk patients in tertiary academic centers by expert endoscopists [14, 18, 19, 49, 50]. A simpler chromoendoscopy method that could be widely implemented is to mix powdered indigo carmine (2.0 g) with 1 L of sterile normal solution and use this as part of a powered washing device connected to the colonoscope. This method allows rapid, convenient pan‐chromoendoscopy concurrent with standard washing of the colon wall (see Video 15.1). Recently, an oral tablet with delayed‐release methylene blue (MB‐MMX, Cosmo Pharmaceuticals, Lainate, Italy) has been evaluated. The drug is administered the evening prior with standard bowel preparation and stains the colon blue. In a large multicenter randomized controlled trial versus placebo, MB‐MMX increases adenoma detection rate by over 8% with no associated serious adverse events [51]. The novel use of orally administered methylene blue should eliminate many of the technical barriers to implementation of colonic chromoendoscopy. Despite the well‐established benefits of dye‐based chromoendoscopy, the limited application in Western countries has led to efforts to develop so‐called virtual chromoendoscopy with no requirements of the use of topical stains. NBI is a commercially available method of optical chromoendoscopy that improves detection of mucosal abnormalities without the messy, time‐consuming problems associated with vital dye‐staining chromoendoscopy. This is currently the best studied advanced endoscopic imaging technique for the detection of Barrett’s dysplasia and detection and characterization of colorectal polyps. While conventional WLE uses the full visible wavelength range (400–700 nm) to produce a red–green–blue image, NBI in combination with magnification endoscopy illuminates the tissue surface using special filters that narrow the red–green–blue bands and simultaneously increase the relative intensity of the blue band. Various studies have demonstrated the value of this technology in the evaluation of patients with upper GI lesions, including BE dysplasia and gastric lesions [52, 53]. Wolfsen et al. [54] demonstrated that in patients evaluated for BE with dysplasia, NBI detected significantly more patients with dysplasia and higher grades of dysplasia with fewer biopsy samples compared with standard resolution endoscopy (Figure 15.4 and Video 15.2) . The meta‐analysis of 14 studies with 854 patients demonstrated that NBI offered an increased diagnostic yield for the detection of dysplasia or cancer of 34% compared with WLE with random biopsies (risk difference (RD) = 0.34; 95% CI 0.14–0.56, p < 0.0001) [55].The subgroup analysis of traditional dye‐based chromoendoscopy and NBI showed no difference in dysplasia detection [56]. Subsequent meta‐analysis has confirmed the improved efficacy of NBI in detecting BE‐associated intestinal metaplasia and HGD as compared with traditional WLE [57]. Furthermore, based on the meta‐analysis and systemic review conducted by ASGE confirming high sensitivity, specificity, NPV (94, 94, 97%) of NBI in BE evaluation, NBI has been endorsed to guide targeted biopsies for detection of dysplasia during surveillance of patients with nondysplastic BE instead of traditional random biopsies for routine surveillance [58]. In routine endoscopic practice, the main advantage of NBI is to highlight mucosal abnormalities and thus direct targeted biopsies and, once dysplasia is confirmed, further endoscopic therapy with mucosal resection. NBI features of specialized intestinal metaplasia with a flat villous‐type mucosal pattern and branching vessels as opposed to HGD with disrupted mucosa and irregular blood vessels have been proposed in three main classification systems from Kansas, Amsterdam, and Nottingham [59–61]. An NBI classification system to detect and characterize dysplasia has been introduced and subsequently validated by the Barrett’s International NBI Group (BING) [61]. A systematic review by Curvers et al. [55] summarized the promising performance and clinical utility of NBI in upper GI endoscopy with a focus on the primary detection of premalignant lesions and the differentiation between neoplastic and nonneoplastic lesions. Despite initial promising studies [62–66], subsequent studies revealed conflicting results with no improvement of adenoma detection rates with the use of this novel technique as compared with traditional dye‐based chromoendoscopy [67–69]. Adler et al. [68] demonstrated the increased adenoma detection rates only in the initial phase of the study, but not in the second phase of the study, suggesting that exposure to NBI (and perhaps the awareness that small, flat adenomas are common) leads to increased adenoma detection with both white light and enhanced methods. The study by Rex et al. [67] further supports this concept since they observed that both HD white light and NBI led to nearly doubling of the adenoma detection rate compared with their own well‐documented historical controls. Results of another prospective randomized back‐to‐back trial comparing NBI to conventional colonoscopy for adenoma detection demonstrates that the miss rate for polyps, and for adenomas, is lower with HD‐NBI than for standard colonoscopy [70]. Subsequent meta‐analyses comparing NBI with standard or HD‐WLE for the detection of adenomas (ADR) failed to demonstrate significant differences between these modalities in improvement of adenoma detection rates [71–74]. Furthermore, it was noted that NBI may be better than standard WLE and equal to HD‐WLE for polyp and adenoma detection [71, 73]. Since then, the initial studies using the second‐generation technologies with brighter illuminations such as the Olympus 190 series Narrow Band Imaging and the Fujifilm Blue Laser Imaging had showed increase in ADR [75–77]. NBI has now undergone two generational improvements (160–180 series and then 180–190 series). While the studies above were unable to demonstrate significant improvements in ADR with each generation, the combined improvements over two generations do appear to increase ADR [78]. The use of NBI for the detection of dysplasia in long‐standing ulcerative colitis has not led to improvement of neoplasia detection, and NBI has not been recommended to be used during colorectal neoplasia surveillance in IBD patients [13, 14, 69, 79, 80]. On the other hand, East et al. [81] demonstrated that NBI could play a role in adenoma detection in high‐risk group patients. The investigators found that a second additional examination with NBI use doubled the total number of adenomas detected in 62 patients with hereditary nonpolyposis colorectal cancer [81]. In a prior systematic review, summarized data on the performance and clinical utility of NBI during colonoscopy did not show a significant improvement in adenoma detection with the use of NBI, but it confirmed the value of NBI for differentiation neoplastic from nonneoplastic colon polyps when used by experts [82] (see Video 15.3). Therefore, in spite of existing controversy on the role of NBI in adenoma detection, the NBI system has been showed to have a relatively high sensitivity of 90–95% and specificity of 80–85% for differentiation neoplastic from nonneoplastic lesions using microvascular features [83]. This level of accuracy is comparable with that of expert chromoendoscopist based on the available studies [49, 84, 85]; thus, potentially the time‐consuming chromoendoscopy may, in time, be replaced by NBI. At the current time, the ability to differentiate between adenomas and hyperplastic polyps appears to be the important clinical utility of NBI in the colon (Figure 15.5). Various studies including meta‐analyses have assessed the accuracy of optical diagnosis for small polyps below 10 mm using NBI system [58, 86]. The NBI International Colorectal Endoscopic (NICE) classification has been introduced and validated in subsequent studies [86, 87]. The meta‐analysis of studies using NBI for optical diagnosis of colon polyp histology confirmed sensitivity and NPV > 90% when optical diagnosis was made with high confidence [88]. An additional value of NBI has been recognized in characterization of sessile serrated polyps in which endoscopic diagnosis is especially challenging due to their subtle endoscopic appearances. In the study by IJspeert et al., the “Workgroup serrAted polypS and Polyposis” (WASP) classification has been proposed and subsequently validated [89]. Specific serrated lesions’ features such as clouded surface, indistinct borders, irregular shape, and dark spots inside the crypt were combined with the known NICE classification. A lesion assessed initially as NICE 1 or NICE 2 with two serrated features was classified as sessile serrated polyp. The accuracy of optical diagnosis of SSP vs. non‐SSP lesion when made with high confidence was 91% and for neoplastic vs. hyperplastic was 89%, while NPV for diminutive neoplastic lesions was 91% [89]. The NBI classifications such as NICE and WASP appear to be useful tools for endoscopists in facilitating the diagnosis of encountered lesions and management decisions such as endoscopic mucosal resection or dissection vs. targeting biopsy vs. referral for surgery when submucosal involvement is suspected (NICE type 3) [90]. While NBI system depends on optical filters within the light source, the electronic system known as flexible spectral imaging color enhancement (FICE), introduced in 2005 by Fujinon, is based on a computed spectral estimation technology that processes the reflected photons to reconstruct virtual images with a choice of different wavelengths. Unlike NBI, it has a processing system instead of optical filter for adjusting wavelength. It has been used to enhance visualization of any GI mucosal lesions including esophageal and gastric lesions. Recently the system has been replaced by the systems that utilizes either light‐emitting diodes (LED) (Eluxeo system) or tandem laser array (Lasereo): white light laser (450 ± 10 nm) as a vivid light source for observation and narrowband blue light/laser imaging (BLI) (410 ± 10 nm) to enhance contrast of capillary and mucosal surface pattern. Due to regulatory issues, the Eluxeo system is marketed in the United States and other Western countries, and the Lasereo is primarily available in Asia. Furthermore, a hybrid mode between these two is called BLI‐bright. When this mode undergoes additional pre‐processing to separate red spectra, it is called as linked‐color imaging. On the other hand, iScan has been also developed by Pentax and is a post‐processing video enhancement modality. All these virtual chromoendoscopy systems lead to enhancement of the tissue microvasculature as a result of the differential optical absorption of light by hemoglobin in the mucosa. These abnormal areas can be defined by the magnification [62,91–94]. These electronic systems of virtual chromoendoscopy were used in esophageal neoplasia demonstrating improvement in detection of early neoplasia [94]. Pohl et al. [62] in their prospective trial compared computed virtual chromoendoscopy system with other modalities such as standard colonoscopy and conventional chromoendoscopy with indigo carmine in low‐ and high‐magnification modes for determination of colonic lesion histology. On the basis of this study, virtual chromoendoscopy was able to identify morphological details that efficiently predict adenomatous histology and was superior to standard colonoscopy and equivalent to conventional chromoendoscopy. Investigators have also been able to show that computed virtual chromoendoscopy is a helpful adjunct for surveillance of BE and it appears to be as accurate as traditional chromoendoscopy in the detection of HGIN/early cancer [62]. Suzuki et al. in their study applied a visibility score to WLE, BLI‐bright, and LCI for assessment of flat colorectal lesions and found that LCI increased visibility of colorectal flat lesions, and it was followed by BLI‐bright and then WLE [95]. When only sessile serrated adenoma/polyp lesions were analyzed, LCI remained significantly higher than the other [95]. The system BLI‐bright has been also demonstrated to detect more adenomas per procedure when compared with WLE in real time [96]. Hoffman et al. showed also that HD endoscopy combined with iScan system is significantly superior in detecting colorectal neoplasia as compared with standard video colonoscopy, and it allows prediction of histology of the identified lesions and surface enhancement [97, 98]. The application of iScan system has been also demonstrated in upper GI lesions such as minimal changes of esophagitis [99]. While all these techniques of virtual chromoendoscopy with NBI, FICE, BLI, LCI, and iScan are very promising, further studies are needed to evaluate their final application, efficiency, and cost‐effectiveness in both detection and characterization of GI lesions, including various forms of GI malignancy from subtle dysplastic lesions to early cancer. Fluorescence imaging endoscopy is primarily based on AFI techniques that differentiate tissue types based on their differences in fluorescence emission. Once the tissue is exposed to short wavelength light, endogenous fluorophores are excited, leading to emission of fluorescent light of a longer wavelength (AFI). Because of changes of endogenous fluorophores in normal, dysplastic, and neoplastic mucosa, the altered autofluorescence is reflected by a pseudocolored image of normal mucosa (green color) and dysplasia/neoplasia with varying tones of red/purple color. AFI has been incorporated into multimodality imaging endoscopes that combine HRE and virtual chromoendoscopy with NBI system. In an initial prospective randomized study, AFI significantly increased the diagnostic yield of surveillance endoscopy to detect intraepithelial neoplasia in high‐risk patients with ulcerative colitis [100]. However, AFI has not been demonstrated to enhance the diagnostic yield of screening colonoscopy in a tandem colonoscopy trial by the same group [101]. The recent study by Vleugels et al. study confirmed that AFI did not meet criteria for proceeding to a large non‐inferiority trial, and thus existing AFI technology should not be further investigated as an alternative dysplasia surveillance method [102]. Curvers et al. [103] published the results of four expert endoscopy centers using trimodal imaging for the evaluation of 84 patients with Barrett’s dysplasia. The use of AFI increased the number of patients found to have HGD from 53 to 90%. On the other hand, the use of NBI reduced the false‐positive rate of AFI from 81 to 26%. Thus far, the experience with systems combining AFI, HRE, and NBI is based on the expert endoscopists evaluating the use for detection dysplasia and carcinoma. CLE allows high‐resolution imaging of cellular and subcellular tissue when optical slices of mucosal surface created by detecting reflected light and tissue autofluorescence enhances through the administration of fluorescent contrast agent. It has been used along with chromoendoscopy techniques in evaluation of various GI lesions, including colorectal lesions enabling visualization of the GI tract at a cellular level [104–111
15
Contrast‐Enhanced Endoscopy: Chromo and Optical Contrast Techniques
Introduction
Overview of contrast (image)‐enhancement techniques: chromoendoscopy and other optical techniques
Chromoendoscopy
Chromoendoscopy in inflammatory bowel disease
Chromoendoscopy in esophageal neoplasia
Chromoendoscopy for colorectal polyps and nonpolypoid neoplasia
Narrowband imaging
NBI in Barrett’s esophagus
NBI for colorectal polyp detection
NBI in inflammatory bowel disease
NBI for classification of colorectal neoplasia
FICE, BLI, BLI‐Bright, LCI, and iScan
Fluorescence imaging
Confocal laser endomicroscopy
Stay updated, free articles. Join our Telegram channel