DIAGNOSTIC ABILITY OF BLUE LIGHT IMAGING FOR PREDICTING DEEP SUBMUCOSAL INVASION IN COLORECTAL LESIONS

Aims:

Blue Light Imaging (BLI) was recently introduced as a novel imaging technology allowing for enhanced visualization of the mucosal surface and vascular pattern morphology. Data regarding the applicability of BLI for prediction of deep submucosal invasion of colorectal lesions is missing.

Main study objective was to assess the potential of BLI for prediction of deep submucosal invasion of colorectal lesions.

Methods:

Consecutive patients undergoing screening or surveillance colonoscopy were prospectively evaluated using a high-definition endoscope with BLI capability. Circumscript lesions were examined with BLI before taking biopsy specimens or performing endoscopic resection. BLI images were graded according to surface and vascular pattern morphology and correlated with conventional histopathology in a prospective and blinded fashion.

Results:

120 cases were included. BLI yielded high-quality images in all cases. Based on pit pattern and vascular alterations BLI could predict the presence of deep submucosal invasion with high sensitivity (95%), specificity (91%) and accuracy (93%). Positive and negative predictive values of BLI for in vivo diagnosis of deep submucosal invasion were 88% and 95%, respectively.

Conclusions:

BLI is a novel diagnostic tool allowing for real-time prediction of deep submucosal invasion of colorectal lesions with high accuracy. This becomes of crucial importance in clinical practice and could lead to an optimized and rapid diagnosis of neoplastic changes during ongoing endoscopy and an individualized management approach.