Optical coherence tomography in the gastrointestinal tract

INTRODUCTION — Components borrowed from the telecommunications industry have been applied to medical imaging to improve resolution as never before. Optical coherence tomography (OCT) is an emerging medical imaging technology that relies on the backscattering of light to obtain cross-sectional images of tissue. Many of the early applications of OCT were in ophthalmology, where the transparency of anterior structures of the eye facilitated high-resolution imaging of the retina. There have been multiple studies using OCT in the gastrointestinal tract. These have demonstrated the feasibility of this technology to enhance endoscopic imaging of the superficial layers of the esophagus, stomach, bile ducts, pancreatic duct, and colon. OCT imaging has demonstrated anatomic structures such as crypts and glands that could potentially permit endoscopists to diagnose mucosal abnormalities such as Barrett's esophagus. In 2013, an OCT system designed for imaging the esophagus became commercially available. (See 'Technical advances' below.)

PHYSICS — OCT is similar in principle to ultrasonography but uses light waves rather than acoustical waves. As in B-mode ultrasonography, a quantitative measurement of backscattering is performed at each axial depth, and the measurements are repeated at different transverse positions. In this manner, a linear or radial two-dimensional map of backscattering strength is acquired [1-3].

Measurement of optical backscattering is performed by low coherence interferometry [4,5]. This method uses a low coherence light source such as a superluminescent diode, which typically has a coherence length of approximately 20 micrometers. The incident light is split in two by an optical beam splitter, with one beam directed to the tissue via an optical fiber and the other beam directed to a mirror located at a precisely controlled distance. The backscattered light from the tissue is combined with the reflected light from the mirror. This results in interference only when the path lengths match to within the 20 microns of the coherence length of the light source. A quantitative measurement of optical backscattering at different depths is obtained by measuring the degree of interference at each mirror position as the mirror is moved.

The coherence length of the light source determines the maximal axial resolution that can be obtained. Transverse resolution is determined by the spot size of the focused beam directed at the tissue and the amount that the apparatus is translated at each during the scan; it is typically also approximately 20 microns. OCT is typically performed with near infrared light because tissue is relatively transparent; longer wavelengths penetrate deeper into biological tissues at these frequencies. Scattering of light in tissue limits the depth of scanning to approximately 1 to 2 mm in the gastrointestinal tract, generally restricting OCT imaging to the mucosa and submucosa when performed during endoscopy.

TECHNICAL ISSUES — OCT is typically performed using catheters passed through the accessory channel of standard gastroscopes, colonoscopes, or duodenoscopes (picture 1):

●Radial and linear scanning catheter probes create radial and linear images, respectively

●A water interface, such as is required for ultrasound imaging, is not required

●Tissue contact is not required

●Resolution is within the range of approximately 7 to 20 micrometers

●Scanning depth is limited to 1 to 3 mm

●Scanning at four frames per second provides real time imaging

As with endoscopic ultrasound systems, both radial scanning and linear scanning catheters have been described. Unlike endoscopic ultrasound, OCT can be performed through air so tissue contact or coupling is not required [6]. Scanning depth is limited to 1 to 2 mm because of scattering of light by tissue. Most of the systems described achieve a resolution of approximately 7 to 20 micrometers, which is sufficient for visualizing mucosal glands, crypts, and villi, but not cellular features such as nuclear dysplasia. In contrast, high-frequency ultrasound (20 MHz) resolution is typically 100 to 200 micrometers, which is insufficient to visualize many of the multicellular features seen with OCT. In newer OCT systems, a 512 by 512 pixel image can be acquired in one-quarter of a second; older systems required several seconds to scan an image, which sometimes lead to blurring from patient motion.

TECHNICAL ADVANCES — Research in optical coherence tomography (OCT) is continuing on several fronts. Reports of substantially higher resolution OCT systems relying on femtosecond laser pulses support the potential for additional improvements in imaging capabilities of gastrointestinal epithelium [7]. A catheter-based femtosecond laser system that can be used during endoscopy has been developed [8,9]. In 2013, an OCT system designed for esophageal imaging (NvisionVLE) became commercially available. The device utilizes a through-the-scope balloon catheter that can be positioned in the distal esophagus where it acquires images of a 6-cm circumferential segment in an automated scan.

The addition of color Doppler to OCT has been used to identify subsurface blood vessels in experimental and clinical settings [10,11]. One group used a color Doppler OCT device in an in vivo rat model and concluded that the device could be useful clinically in monitoring hemostatic intervention [10].

A second group used Doppler OCT to study a variety of conditions in the gastrointestinal tract including cancer, esophageal varices, and gastric antral vascular ectasia [11]. The combination of high-resolution imaging and Doppler provided striking images of the microcirculation in various disease states.

Spectroscopic OCT imaging is an extension of ultrahigh-resolution OCT. It uses the broad spectral bandwidth of the optical source to obtain information from the spectral content of the backscattered light. In vitro studies in Barrett's esophagus showed that spectral OCT improved the contrast of OCT images [12]. It may also be possible to quantify localized tissue hemoglobin oxygenation by spectroscopic OCT [13]. These advancements in OCT imaging show promise for future clinical applications.

OCT of the esophagus — Several studies have described features of normal and abnormal esophageal mucosa on OCT during endoscopy. In the normal esophagus, several distinct layers are clearly visualized: a relatively homogeneous epithelium, a high-backscattering lamina propria, a low-scattering muscularis mucosa, a high-scattering submucosa, and a low-scattering and thick muscularis propria (picture 2) [12,14-16]. By contrast, the uniformly layered structure is disrupted in Barrett’s esophagus and multiple crypt and gland-like structures are seen as pockets of low backscattering (image 1) [9,17]. Several cases of esophageal adenocarcinoma have been described where large, low-scattering pockets (corresponding to mucin) and a disorganized appearance were noted [18-20]. Finally, OCT has been used to stage esophageal squamous cell carcinoma [21].

Barrett's esophagus — In the first prospective evaluation of OCT in the esophagus during endoscopy, three criteria were used to diagnose specialized intestinal metaplasia (Barrett's esophagus) [22]:

●Lack of normal esophageal or gastric morphology

●Inhomogeneous tissue contrast

●Presence of submucosal glands

The presence of at least two of the three criteria was found to be 97 percent sensitive and 92 percent specific using histopathology as the standard. However, it is difficult to directly compare these results to previous studies estimating the accuracy of visual identification of intestinal metaplasia on endoscopy because the OCT study set consisted mainly of normal esophagus, Barrett's esophagus, and normal stomach. In practice, the main difficulty in diagnosing intestinal metaplasia visually during endoscopy arises in distinguishing Barrett's from inflammation in the distal esophagus and cardia, and in distinguishing intestinal metaplasia from the less worrisome gastric metaplasia. All of the false positives in this study were due to tissue from the gastric cardia with or without inflammation, and it is therefore likely that the specificity would have been substantially lower if the study would have included a more clinically representative set of samples.

Several studies have evaluated the accuracy of OCT in detection of dysplasia in Barrett's esophagus [23-28]. Because the resolution of standard OCT is insufficient to assess cellular features of dysplasia, an assessment of dysplasia is generally based upon larger scale features such as epithelial scattering intensity and the presence of irregularly shaped or enlarged mucosal glandular structures.

In one study, 177 OCT images were reviewed by a blinded investigator and assessed for high-grade dysplasia/intramucosal carcinoma by evaluating the degree of surface maturation and gland organization [25]. Sensitivity and specificity were 83 and 75 percent, respectively, when these features were combined to form a dysplasia index.

A second group used OCT to assess Barrett's esophagus for the presence of any level of dysplasia (low- or high-grade or cancer) [26]. The endoscopist performing the procedure evaluated the OCT images for dysplasia, relying on features such as focal areas of decreased light scattering and focal loss of mucosal structure and organization. Sensitivity and specificity were 68 and 82 percent, respectively. This group also described a computer-aided diagnosis algorithm for detection of dysplasia that had a sensitivity and specificity of 82 and 74 percent, respectively [27].

Two additional studies evaluated the performance characteristics of an OCT system utilizing endoscopic mucosal resection (EMR) specimens from patients with Barrett’s esophagus. OCT had a sensitivity of 83 to 86 percent and a specificity of 71 to 88 percent for differentiating neoplastic Barrett’s esophagus (ie, high-grade dysplasia or intramucosal carcinoma) from non-neoplastic Barrett’s esophagus [23,24].

A reusable, tethered OCT capsule for evaluating the esophagus has been described. In a study including 17 patients with Barrett's esophagus, diagnostic images generated by the OCT capsule showed strong correlation with endoscopic visualization for determining the maximal extent of Barrett's mucosa [29].

Finally, OCT has been used to detect buried Barrett's epithelium following radiofrequency ablation [30]. (See "Barrett's esophagus: Treatment with radiofrequency ablation", section on 'Buried Barrett's esophagus'.)

Squamous cell carcinoma — OCT has been used in the preoperative staging of esophageal squamous cell carcinoma. In a study of 62 patients, OCT was used to categorize the depth of invasion into one of three categories: epithelium or lamina propria, muscularis mucosa, or submucosa [21]. The OCT image criteria were established using findings from the first 16 patients. The criteria were subsequently evaluated in the other 46 patients, comparing the results of OCT with the results of the histologic examination of the resected specimens. The accuracy of OCT for determining involvement of the epithelium/lamina propria, muscularis mucosa, or submucosa was 95, 85, and 91 percent, respectively.

A second study included 123 patients with 131 superficial esophageal squamous cell carcinomas. OCT accurately characterized 118 (90 percent) of the tumors [31]. Five tumors (4 percent) were overstaged, and eight tumors (6 percent) were understaged. Of the tumors that were limited to the epithelium/lamina propria, OCT correctly staged 88 of 93 (95 percent), and of the tumors that involved the muscularis mucosa/submucosa, OCT correctly staged 30 of 38 (79 percent). The accuracy of OCT in staging tumors limited to the epithelium/lamina propria was higher than that of high-frequency probe-based endoscopic ultrasound (95 versus 81 percent, p<0.01).

OCT of the stomach — There have been no large studies of OCT in the stomach. OCT images of the stomach are generally characterized by low tissue contrast [18]. It can be difficult to differentiate the superficial mucosal layers from the muscularis mucosa [14,15]. The relatively thick superficial glandular epithelium also causes poor visualization of the deeper layers, making it difficult to evaluate the muscularis propria [20].

OCT of the colon and small bowel — The full thickness of the colonic wall can often be seen on OCT: the superficial mucosa with visualization of crypts, the highly scattering submucosa, the weakly scattering muscularis propria, and the thin, highly scattering serosal stripe [14,32]. Images of small bowel clearly demonstrate mucosal villi and can be used to identify villous atrophy in celiac disease [33]. Adenomas in the colon have been described with expanded glands in the superficial mucosa, mucosal cysts, and an uneven surface. Adenocarcinoma has also been described with complete loss of mucosal architecture [19]. In ulcerative colitis, superficial mucosal ulcers and other mucosal characteristics can be seen on OCT that correlate well with histologic findings [34,35].

One study suggested that hyperplastic and adenomatous polyps had a distinct appearance on OCT [36]. Hyperplastic polyps generally had an organized crypt pattern and overall scattering intensity that was relatively similar to normal tissue, while adenomatous polyps demonstrated an absence of an organized crypt pattern and a decrease in overall scattering intensity.

Another study of colectomy specimens found that OCT could help predict transmural inflammation, making it potentially useful for distinguishing Crohn disease from other forms of colitis [37]. The authors subsequently used the ex-vivo criteria to evaluate 40 patients with known Crohn disease and 30 patients with known ulcerative colitis [38]. A disrupted layered structure was apparent in 36 patients with Crohn disease (90 percent) compared with only 17 percent of patients with ulcerative colitis. Sensitivity and specificity of the disrupted layered structure for distinguishing Crohn disease from ulcerative colitis were 90 and 83 percent, respectively. The authors concluded that OCT may be useful for distinguishing Crohn disease from ulcerative colitis. Whether the test characteristics found in this study will hold true (especially when evaluating patients with indeterminate colitis) remains to be determined.

OCT of the bile and pancreatic ducts — The role of OCT in the bile ducts is still being determined. At least three groups have described results of in-vivo OCT of the bile duct [39-41]. The OCT probe is inserted into the bile duct via the accessory channel of a duodenoscope. The flexibility and narrow caliber of the OCT probe (2.6 to 2.8 mm in diameter) permit relatively easy cannulation. Nevertheless, cannulation with the probe was attempted after sphincterotomy.

●The first study found that the connective tissue layer and underlying retroperitoneal tissue seen on OCT were similar to those seen histologically [39]. However, the authors found that the resolution of the images was limited.

●In the second study, five patients (three of whom had malignant biliary strictures), underwent OCT during ERCP [40]. Of the three patients with malignant disease, one had metastatic colorectal cancer and the other two had cholangiocarcinoma. Regions of disorganized microstructure were believed to represent metastatic cancer based upon the similarity of images to images acquired in patients with esophageal adenocarcinoma. OCT images of cholangiocarcinoma showed villiform papillary structures not seen in patients with benign disease.

●In the third study, intraductal OCT was used in 37 patients with biliary strictures to detect malignancy [41]. OCT criteria for malignancy included unrecognizable layer architecture and the presence of large, non-reflective areas compatible with tumor vessels. Thirty-five of the patients were able to undergo OCT (the remaining two could not due to tight strictures). Of those 35 patients, 19 had malignant strictures, and 16 had benign strictures. When both OCT criteria were met, the sensitivity for OCT was 53 percent, with a specificity of 100 percent. If at least one criterion was met, the sensitivity was 79 percent, with a specificity of 69 percent. When the results of OCT were combined with the results from biopsies and/or brushings, the sensitivity increased to 84 percent, with a specificity of 69 percent (the sensitivity and specificity for biopsies/brushings alone were 67 and 100 percent, respectively).

There is limited experience with the performance of OCT in the pancreatic duct. An ex-vivo study described OCT findings in 10 freshly resected surgical specimens using a 1.2 mm diameter OCT probe [42]. A significant advantage of the 1.2 mm diameter probe is that it fits within the typical transparent cannulas used during ERCP, which should make probe insertion into the pancreatic duct technically simpler. The normal pancreatic duct was visualized as a three-layer structure: a thin superficial hyporeflective band corresponding to the epithelium, a thick hyperreflective connective-fibromuscular layer, and a deep hyporeflective layer corresponding to acinar tissue near the duct. By contrast, adenocarcinoma was characterized by loss of the three-layer morphology and multiple minute, non-reflective areas. OCT images from inflamed or dysplastic noncancerous areas often appeared similar to normal, suggesting that OCT may not be sensitive for precancerous changes in the duct.

Laser therapy is becoming widespread in medicine and surgery. Control and dosimetry rely primarily on visual feedback of the thermal effect. The potential to use OCT to direct surgical ablation was studied in a rat model [43]. One group used OCT as a targeting method in a number of organ sites. An argon laser was directed against the tissue. OCT imaging was then successfully used to assess the thermal tissue damage, suggesting that OCT may have a future role in image-guided surgical procedures. An ophthalmological study suggested that there may be an application for OCT in monitoring photodynamic therapy [28]. Photodynamic therapy is under study for treating esophageal carcinoma, and there may be a future role for OCT in monitoring treatment.

SUMMARY AND RECOMMENDATIONS

●Optical coherence tomography (OCT) can be used to obtain high-resolution images of the epithelium during endoscopic procedures throughout the gastrointestinal tract. Resolution is approximately 20 micrometers, which is sufficient to resolve architectural details such as glands and crypts, but not nuclear dysplasia. (See 'Technical issues' above.)

●Early studies have mostly described features seen in normal and disease states. Additional studies suggest that OCT may be sensitive and specific in diagnosing Barrett esophagus and in staging superficial squamous cell cancer of the esophagus. (See 'OCT of the esophagus' above.)

●By demonstrating areas with worrisome architectural features, OCT may eventually assist endoscopists in selecting appropriate biopsy sites when performing cancer surveillance for patients with Barrett esophagus or ulcerative colitis. OCT can also distinguish adenomatous from hyperplastic polyps. (See 'OCT of the colon and small bowel' above.)