Public Release: 

New technology may mean immediate diagnosis for patients with GI diseases

Other research shows current analysis of colon cancer predictors too subjective

American Gastroenterological Association

SAN DIEGO, CA (May 18, 2008) - A new real-time microscopic technique could change the way gastrointestinal diseases are detected. Research presented today at Digestive Disease Week® 2008 (DDW®) shows that confocal laser endomicroscopy (CLE) effectively and immediately identifies suspicious patterns and precancerous cells and may eliminate the need in many cases for biopsy to diagnose gastrointestinal conditions. This could lead to earlier diagnosis and treatment of conditions including reflux disease, colon cancer and inflammatory bowel disease. DDW is the largest international gathering of physicians and researchers in the fields of gastroenterology, hepatology, endoscopy and gastrointestinal surgery.

Other research presented indicates that the current methods of using aberrant crypt foci, or ACF, to determine the presence of colon cancer may be unreliable due to the subjectivity of the analysis by endoscopists.

"Up until now, determining the condition of a patient required days or even weeks while patients waited for a diagnosis; further, it has been difficult to detect subtle precancerous lesions or even areas within a larger lesion, often leading to time-consuming and labor-intensive procedures as well as uncertainty about missing something important" said Pankaj J. Pasricha, MD, FASGE, professor of medicine, gastroenterology and hepatology, Stanford University School of Medicine. "New techniques such as those discussed today will change the way we diagnose patients, allowing us to treat them more accurately, quickly and appropriately."

Confocal Laser Endomicroscopy is an Effective and Safe Diagnostic Tool in GI-Endoscopy (Abstract #S1162)

Confocal laser endomicroscopy (CLE), a tiny microscope used at the tip of an endoscope to provide highly magnified images, allows for expanded diagnostic possibilities during endoscopy, which can help doctors detect suspicious activity immediately.

"This is a very promising technique for real-time microscopic imaging of the gastrointestinal tract," said Kerry B. Dunbar, MD, fellow at the Johns Hopkins University School of Medicine division of gastroenterology and hepatology. "Previously, it took a few days or a week to find out a diagnosis because we'd have to take photos and then do a mucosal biopsy. But with CLE, we can see what's going on at the time of the endoscopy and then diagnose and treat patients immediately, instead of waiting a week or more."

Researchers performed 2,102 CLE examinations in 1,771 patients using diagnostic imaging tools including intravenous fluorescein (19 percent), topical acriflavine (0.05 percent) and cresyl violet (1 percent). Confocal images were then graded according to confocal pattern classifications for the presence of normal, inflammatory, regenerative and neoplastic changes.

Most importantly, the data revealed 422 cases with reflux disease (including Barrett's esophagus); 84 cases with esophageal dysplasia and cancer; 99 cases of colon cancer; 206 cases of adenoma; 75 cases of gastritis; 22 cases of gastric dysplasia and cancer; 39 cases of celiac disease; and 893 cases of inflammatory bowel disease.

The overall accuracy rate for CLE was 91 percent in the upper GI tract and 93 percent in the lower tract. "This has the potential to help patients more quickly," said Dr. Dunbar, adding that given the rapid progression of cancers, earlier detection and treatment is critical.

Confocal Laser Endomicroscopy in Barrett's Esophagus and Associated Neoplasia: A Prospective Randomized Controlled Crossover Trial (Abstract #673)

A new study shows that doctors are now able to identify the same amount of precancerous cells using images from confocal laser endomicroscopy (CLE), as would have been found using traditional biopsy. The finding is important because it could reduce the number of biopsies taken because doctors using CLE would know whether precancerous cells existed, and therefore would not have to needlessly biopsy tissue that may not be abnormal.

Investigators used CLE to examine patients with Barrett's esophagus (BE), a condition that can often lead to esophageal cancer, in an effort to detect tiny precancerous lesions before they grow. One challenge with BE is that it can require several tissue biopsies, which are expensive, invasive and can cause bleeding. Using CLE, researchers were able to magnify the lining of the esophagus 1,000 times, which enabled investigators to take biopsies only of areas that looked suspicious, instead of performing many unnecessary random biopsies.

Researchers looked at 30 patients undergoing BE surveillance (Group 1) and those with suspected high-grade dysplasia (Group 2) using CLE and standard endoscopy (SE). The results of this ongoing study showed that there was a significant reduction in the mean number of biopsies taken during CLE, compared to SE (0.5 vs. 11.9, respectively). Eighty percent of the patients with BE undergoing routine surveillance did not need mucosal biopsies because no dysplasia was seen during the in vivo microscopic examination.

"This suggests that confocal laser endomicroscopy with targeted biopsy significantly improves diagnosis of Barrett's esophagus, and offers great promise for reducing the number of biopsies needed to look for dysplasia in the esophagus," said Kerry B. Dunbar, MD, fellow at the Johns Hopkins University School of Medicine division of gastroenterology and hepatology. "It may allow us to detect and treat early cancers or high grade dysplasia in BE at the same visit."

High Resolution Confocal Endomicroscopy Probe System for in vivo Diagnosis of Colorectcal Neoplasia (Late Breaking Abstract)

Investigators have found that the confocal endomicroscopy system (CFM) can immediately distinguish benign from precancerous polyps of the colon with very high accuracy. The procedure allows doctors to both diagnose and treat the condition at the time of the procedure, and it prevents unnecessary biopsies and removal of tissue.

Doctors currently perform an endoscopy to diagnose a number of conditions including acid reflux, diarrhea, precancerous or cancerous conditions, or colon polyps. They examine the tissue with an endoscope to determine whether it appears to be abnormal. However, for specific diagnosis, they also need to take a biopsy and send it for pathological analysis.

Investigators used a catheter probe-based CFM system which can be passed through the accessory channel of almost any endoscope. The CFM probe provides images of cells and the tiny blood vessels around the cells, similar to what pathologists look at under a standard microscope. Investigators performed the procedure on 26 patients in two phases.

First, they removed the polyps and waited for pathological results. Then they looked at the images to determine which cells appear abnormal and normal. They found that using CFM, they could predict the sensitivity and specificity of the polyps with 89 percent accuracy. Furthermore, investigators looked at small polyps that appeared to be benign that would normally have been removed. Using CFM, they were able to accurately detect that they were benign 98 percent of the time.

"This technique will fundamentally change how the two fields interact," said Anna M. Buchner, MD, PHD, of the Mayo Clinic, Jacksonville. "This procedure allows us to become endopathologists, and the result is that we will be able to better manage patients with gastrointestinal disorders."

Dr. Buchner hopes that these results can get the accuracy rate near 100 percent within the next year. She added that CFM will require additional training so that gastroenterologists can learn how to interpret the images and develop computer-aided diagnosis of images.

Investigators are now exploring ways to further increase the accuracy of this procedure, and apply it to other GI conditions that would have required biopsy or tissue removal.

Performance of High Magnification Colonoscopy (HMC) in Detecting Aberrant Crypt Foci (ACF) (Abstract #M1989)

The use of aberrant crypt foci (ACF) to identify predictors of colon cancer is unreliable and unsafe because they do not provide a histological confirmation with any accuracy.

"Attaining more accurate predictors of colorectal cancer is critical, since although it is one of the most common cancers, it is relatively rare in the general population," said Akshay Gupta, MD, fellow at the division of gastroenterology at the University of Michigan Medical School.

ACF are a potential, immediate endpoint - or determinant - of colorectal cancer. While several endoscopic criteria have been used to identify them, their reproducibility, or agreement rate across endoscopists regarding what they indicate, has not been studied. Therefore researchers in this study sought to explore reproducibility as well as which criteria most strongly predict the confirmation of ACF.

Previous studies have tried to look at polyps as a predictor of cancer, but it is difficult since it is rare to see many polyps in one patient. However, ACF are more prevalent and therefore provide a better assessment.

In a previous study, four investigators examined three sets of images for ACF. The investigators then discussed the differences among the images in an effort to develop a consensus protocol for assessment of ACF. For the current study, researchers assessed 113 new images of suspected ACF by two of the participating endoscopists.

Each image was scored for endoscopic criteria used to identify ACF including number of crypts (abnormalities in small intestinal or colonic tissue), diameter and thickness of crypts, staining of the lining of the crypt compared to normal mucosa (the inner lining of the organ and body cavity), and lesion margin. To assess how often and how the raters agreed, one of the investigators evaluated the images a second time three months later.

Researchers found that among the endoscopic criteria used to identify ACF, the rater agreement was good for the number of crypts, moderate for lesion margin and worse for all of the others. "Unfortunately there were no differences in the histology results based on any of the endoscopic criteria," said Dr. Gupta. "Therefore these results raise concern over the use of ACF as a biomarker of colon cancer because there are no uniformly accepted criteria - the data shows too much variability to be useful in detecting precancerous activity."


Digestive Disease Week® 2008 (DDW®) is the largest international gathering of physicians, researchers and academics in the fields of gastroenterology, hepatology, endoscopy and gastrointestinal surgery. Jointly sponsored by the American Association for the Study of Liver Diseases (AASLD), the American Gastroenterological Association (AGA) Institute, the American Society for Gastrointestinal Endoscopy (ASGE) and the Society for Surgery of the Alimentary Tract (SSAT), DDW takes place May 17-22, 2008 in San Diego, Calif. The meeting showcases more than 5,000 abstracts and hundreds of lectures on the latest advances in GI research, medicine and technology.

Disclaimer: AAAS and EurekAlert! are not responsible for the accuracy of news releases posted to EurekAlert! by contributing institutions or for the use of any information through the EurekAlert system.