Key points

Introduction

Colonoscopy is the preferred method for colon cancer screening as recommended by the American College of Gastroenterology. However, the use of conventional colonoscopy is associated with a lower colorectal cancer (CRC) protection for proximal versus distal tumors. The phenomena known as interval cancers are associated with low adenoma detection rates (ADRs), and may be related to nonvisualized flat or nonpolypoid lesions. Multiple adjunctive techniques and technologies have been studied to improve ADRs. These include narrow band imaging, high definition, magnified endoscopy, and chromoendoscopy. In this article, we review chromoendoscopy and its role as an adjunct to colonoscopy. We also discuss the role of chromoendoscopy for the detection of dysplasia in inflammatory bowel diseases (IBDs).

Chromoendoscopy is the application of dye on the colonic surface that allows for a more detailed analysis of mucosal abnormalities. In 1977, Tada and colleagues used a dye-spraying method with indigo carmine (IC) and methylene blue in the lower gastrointestinal tract. They demonstrated that the dye-spraying method was useful to detect small mucosal changes. Colonoscopy with dye spraying or chromoendoscopy has since become a standard diagnostic tool for gastroenterologists.

Introduction

Colonoscopy is the preferred method for colon cancer screening as recommended by the American College of Gastroenterology. However, the use of conventional colonoscopy is associated with a lower colorectal cancer (CRC) protection for proximal versus distal tumors. The phenomena known as interval cancers are associated with low adenoma detection rates (ADRs), and may be related to nonvisualized flat or nonpolypoid lesions. Multiple adjunctive techniques and technologies have been studied to improve ADRs. These include narrow band imaging, high definition, magnified endoscopy, and chromoendoscopy. In this article, we review chromoendoscopy and its role as an adjunct to colonoscopy. We also discuss the role of chromoendoscopy for the detection of dysplasia in inflammatory bowel diseases (IBDs).

Chromoendoscopy is the application of dye on the colonic surface that allows for a more detailed analysis of mucosal abnormalities. In 1977, Tada and colleagues used a dye-spraying method with indigo carmine (IC) and methylene blue in the lower gastrointestinal tract. They demonstrated that the dye-spraying method was useful to detect small mucosal changes. Colonoscopy with dye spraying or chromoendoscopy has since become a standard diagnostic tool for gastroenterologists.

Technique of chromoendoscopy

Chromocolonoscopy involves examining colonic mucosa after it is sprayed with dye. Dye staining may also be achieved by a capsule that is ingested immediately after a polyethylene glycol (PEG) lavage. However, the more widely used method involves a dye-spray catheter. Many spray catheters are commercially available for single use or multiple uses. For most procedures, the spray catheter is introduced through the working channel in the colonoscope, the dye is sprayed onto the mucosal surface of the colon in a continuous spray, and then any extra pooled dye is aspirated. Most techniques involve segmental dye spraying, in which the colon is stained in small sections, typically 10 cm at a time.

Other techniques for dye spraying have been described. In contrast to the traditional high-volume technique, Pohl and colleagues used a low-volume spraying technique. During continuous extubation of the colonoscope, 0.4% IC was applied by an assistant to achieve diffuse coverage of the mucosa. The volume of dye used with this method was smaller and the average time required for the procedure was less than reported in other studies. Other dye-application techniques are discussed in subsequent sections.

An important requirement for chromoendoscopy is good bowel preparation. Although the luminal surface is usually cleaned with water or a mucolytic agent, such as N-acetylcysteine (NAC), large amounts of stool prohibit adequate staining and mucosal visualization.

Chromoendoscopy can be performed using pan-colonic or localized techniques. Pan-colonic chromoendoscopy is usually performed to increase detection of adenomas or dysplasia in IBD. Targeted dye spraying is used to delineate a lesion’s margins or to differentiate neoplastic from non-neoplastic lesions.

There are 3 different kinds of stains: absorptive, contrast, and reactive. The most commonly used stains are methylene blue and IC.

Absorptive Stains

Absorptive stains are taken up by specific epithelial cells and allow for better characterization of the mucosa. Methylene blue, Lugol iodine, Toluidine blue, and Cresyl violet are the most commonly used absorptive stains. The application of absorptive and contrast dyes can be very different and this is illustrated in Fig. 1 . In Fig. 2 , a polyp is shown before and after the application of IC.

As compared with absorptive stains, pooling with contrast stains cannot be easily managed with water. Although the water will wash away excess stain in the former, it will wash away all of the latter dye.

( A, B ) A polyps is shown before and after the application of indigo carmine.

Following are brief descriptions of commonly used stains:

Methylene blue

This stain is actively absorbed by mucosa of small intestine and colonic epithelium. Methylene blue is poorly absorbed by damaged mucosa or by nonabsorptive surfaces like squamous epithelium, areas of active inflammation, and intraepithelial neoplasia. The level of absorption correlates with the amount of cytoplasm and goblet cells present. Methylene blue chromoendoscopy requires the application of a mucolytic, such as NAC, because mucus prevents active absorption of the dye. Thus, a suggested algorithm for the application of methylene blue includes the following:

• •
  

  The patient’s colon is prepped with a PEG solution

  
  
• •
  

  The targeted mucosa is washed with water

  
  
• •
  

  A mucolytic, such as NAC (10%–20%), is applied

  
  
• •
  

  The mucosa is again washed with water

  
  
• •
  

  The dye is sprayed at a concentration of less than 1%

  
  
• •
  

  Excess dye is washed with water after 2 minutes of dye staining

Patients should be advised that the staining can last up to 24 hours and that they may have discolored stools. In addition, because it is an absorptive dye, they may experience discolored urine. Repici and colleagues conducted an open-label trial of safety and efficacy of a methylene blue capsule in healthy volunteers. They administered 200 mg and 400 mg delayed release tablets of methylene blue in 22 healthy volunteers who had completed a PEG lavage. It was hypothesized that the tablets are better absorbed from colonic mucosa, as it is free from debris and fecal material. Peak levels of the active ingredient were seen in 16 hours. No serious adverse reactions were reported. The most common adverse event was elevated liver enzymes that returned to baseline in 15 days.

Methylene blue has some other attendant risks associated with its use as an absorptive stain. In vitro studies have demonstrated that white light and methylene blue can cause single-strand DNA breaks. Davies and colleagues demonstrated that methylene blue and not IC caused DNA damage. More discussion regarding risks of methylene blue can be found in the IBD section.

Cresyl violet

Cresyl violet is as less commonly used dye that can be used in combination with IC to better delineate colonic lesions. Kudo and colleagues used this agent with magnification colonoscopy to delineate pit patterns, which are discussed later in this article. Cresyl violet 0.2% is applied to the mucosa, which is previously stained with IC. This preferentially stains the pit margins and provides clear definition of each pit pattern.

Chromoendoscopy for polyp morphology

Multiple studies have examined the utility of chromoendoscopy in detection of adenomas during screening and surveillance colonoscopy. Selected trials are shown in Table 1 . Many of these have been single-arm chromoendoscopy studies. These trials demonstrated an increase in flat and small adenomas, many of which were often proximal.

A review on chromocolonoscopy and adenoma detection requires a brief review of polyp morphology. Polyps can be classified as protruding or nonprotruding. One classification is the Japanese Research Society Classification (JRSC), which defines flat lesions as those with a height that is less than one-half the measured diameter. The Paris classification classifies polyps into protruding and nonprotruding based on whether the lesion protrudes into the lumen a distance of at least 2.5 mm or the approximate width of a standard snare catheter or jaws of a closed biopsy forceps. A representation of the Paris Classification is shown in Table 2 . Although many polyps may be flat, there is a distinct clinical significance among those that are elevated, flat, or depressed. The less commonly seen IIc or depressed lesions have the highest likelihood of advanced pathology, such as high-grade dysplasia.

Flat colorectal adenomas were first described by Muto and colleagues in 1985. Mitooka and colleagues demonstrated 37 minute flat lesions in 32 of 1152 patients who had a colonoscopy after swallowing an IC capsule. Rembacken and colleagues examined 1000 patients for flat or depressed lesions. Any mucosal abnormality was sprayed with 0.2% IC: 36% (117) were flat and 0.6% (2) appeared depressed; 54% of the advanced neoplasms were flat or depressed. Although such studies highlight the significance of depressed lesions, it should be noted that most adenomas are IIa or flat elevated and do not contain advanced pathology.

Chromoendoscopy for polyp detection

Several randomized controlled studies examining pancolonic chromocolonoscopy were performed in high-risk patients who were symptomatic or had a family or personal history of colorectal neoplasia. Brooker and colleagues observed no increase in overall adenoma detection in one of the earliest randomized trials. However, the chromocolonoscopy group had more diminutive adenomas proximal to sigmoid colon, more patients with 3 or more adenomas, and more non-neoplastic lesions detected. The extubation time with chromocolonoscopy was also statistically significantly longer as compared with the control group.

Lapalus and colleagues randomized 292 patients with a personal history of colorectal adenomas and/or family history of colorectal cancer. Patients received a conventional colonoscopy followed by randomization to repeat conventional colonoscopy or dye spraying with structural enhancement. There was no difference in the number of patients with at least one adenoma detected or in the number of detected adenomas. However, this study also observed an increase in detection of proximal adenomas and increased examination duration in the dye group.

In a trial by Hurlstone and colleagues, the investigators controlled for extubation time and observed a statistically significant increase of detected adenomas in the chromocolonoscopy group. There were more diminutive adenomas smaller than 4 mm, diminutive and flat polyps in right colon, and an increased number of patients with multiple adenomas. They recommended use of chromocolonoscopy for index colonoscopies to risk stratify patients. Le Rhun and colleagues observed in their randomized trial that chromocolonoscopy detected more polyps but it required longer procedure time. Stoffel and colleagues conducted their study with high-risk patients who had at least one previous colon polyp and/or colorectal cancer. A major critique of chromoendoscopy trials is that the increased adenoma detection rates may result from an increased observation time. In the Stoffel and colleagues trial, patients were randomized to chromoendoscopy versus conventional endoscopy with intense inspection to counter the time effect of dye spraying. The study concluded that chromoendoscopy detected additional adenomas in more subjects than intensive inspection (44% vs 17%) and identified significantly more missed adenomas per subject (0.7 vs 0.2, P <.01). The adenomas detected with chromoendoscopy were significantly smaller and more often right-sided. Chromoendoscopy took longer than conventional with intense inspection. One important observation was the increased number of normal biopsies in the patients who had chromocolonoscopy.

Based on the previously mentioned studies, Kahi and colleagues randomized 660 average-risk screening patients to chromocolonoscopy with high-definition white light or high-definition white light only. They observed that chromocolonoscopy detected marginally more small and flat adenomas than white light alone. In addition, they observed a difference between study sites, implying that the benefit of chromoendoscopy may vary between endoscopists. Although this study controlled for mucosal examination time, the overall procedure time was greater in the chromocolonoscopy arm.

Using a novel dye-spraying technique as described previously, Pohl and colleagues enrolled 1008 patients in the largest randomized controlled trial of pancolonic chromocolonoscopy. The proportion of patients with at least one adenoma was significantly higher in the chromocolonoscopy group. The pancolonic chromocolonoscopy group had an increased overall ADR with an increased detection of flat and serrated adenomas. Because serrated lesions may be implicated in interval cancers, this benefit of chromocolonoscopy may be important clinically. There was a trend toward an increased rate for detection of advanced adenomas. Unlike the Kahi and colleagues’ trial, this group recommended chromocolonoscopy for average-risk individuals.

Other studies have examined the impact of chromocolonoscopy in targeted segments of the colon. Although some examine the distal colon, others have examined the proximal colon. These studies confirm the findings of pancolonic chromoendoscopy studies with regard to the small size of detected lesions and increased procedure time for the chromoendoscopy group.

Chromoendoscopy and hereditary nonpolyposis colorectal cancer

There are 2 published studies evaluating the use of chromoendoscopy in patients with hereditary nonpolyposis colorectal cancer (HNPCC). Lecomte and colleagues enrolled 36 consecutive patients with HNPCC. Patients had a complete high-resolution colonoscopy and, after a second intubation of cecum, the second step was chromoendoscopy of proximal colon on withdrawal. Chromoendoscopy increased detection of flat adenomas compared with conventional. It also detected more adenomatous lesions in the proximal colon than regular colonoscopy.

Huneburg and colleagues studied a total of 109 patients with HNPCC. Forty-seven patients had standard colonoscopy followed by chromocolonoscopy, and 62 patients had narrow band imaging (NBI) followed by chromocolonoscopy. In comparison with white light and NBI, chromocolonoscopy detected statistically significantly more lesions per patient. In particular, these were hyperplastic polyps and flat adenomas.