Criteria for the diagnosis of dysplasia by endoscopic optical coherence tomography

Background. Endoscopic optical coherence tomography provides images of the GI mucosa and submucosa in microscopic detail. It is unknown whether endoscopic optical coherence tomography can reliably detect dysplasia. Colon polyps were used as a model to determine whether dysplasia in GI tissue has characteristic optical coherence tomography imaging features. Methods: Endoscopic optical coherence tomography images of colon polyps and normal colon tissue were obtained at colonoscopy. In real time, endoscopists compared tissue organization and light scattering for polyps and normal mucosa with endoscopic optical coherence tomography. Imaged polyps were removed and evaluated histopathologically. Organization and light scattering, as assessed by endoscopic optical coherence tomography at colonoscopy, were compared for adenomas versus hyperplastic polyps. A computer program also quantified and compared the degree of light scattering for hyperplastic polyps and adenomas. Results: A total of 44 polyps were imaged in 24 patients (30 adenomas, 14 hyperplastic polyps). Endoscopic optical coherence tomography images of adenomas had significantly less structure (p = 0.0005) and scattered light to a lesser degree than hyperplastic polyps (p = 0.0007). Hyperplastic polyps were significantly closer in organization (p = 0.0003) and light scattering (p = 0.0006) to normal mucosa as compared with adenomas. By digital image analysis, the light-scattering property of hyperplastic polyps was closer to normal mucosa compared with adenomas (14.86 vs. 45.81; p = 0.0001). Conclusions: Real-time endoscopic optical coherence tomography imaging differentiated adenomas, hyperplastic polyps, and normal colon tissue. By using the colon adenoma as a model, the endoscopic optical coherence tomography characteristics of dysplasia are loss of tissue organization and reduced light scattering.