Endoscopic horizon stabilization in natural orifice translumenal endoscopic surgery: a randomized controlled trial

Spatial orientation in natural orifice translumenal endoscopic surgery (NOTES) has been identified as a potential barrier to clinical application. We aim to evaluate a triaxial inertial sensor and software that automatically corrects any movements on the roll axis of the flexible endoscope, allowing for stabilization of the image horizon during NOTES operations in a randomized controlled trial.

A total of 18 participants (11 surgeons/7 gastroenterologists) performed a transgastric task in the ELITE simulator, which included navigation to the appendix and gallbladder, diathermy of the appendix base and gallbladder fossa, and clipping of the cystic duct using a single-channel gastroscope. Each participant performed the task twice with randomization to horizon stabilization occurring at the second attempt. The primary end point was change in overall performance (time taken and errors made) between the first and second attempt, and secondary end points were absolute performances in the second attempt and subjective evaluation.

Without horizon stabilization, there was a median improvement of 42.4% in time taken and 38% in number of errors made from the first to the second attempt; however, with the software turned on, there was a statistically significant deterioration of 4.9% (P = .038) in time taken and an increase in errors made of 183% (P = ns).

Although the software corrects the view to that preferred during surgery, the endoscopic control mechanism as well as the exit point of the instrument are altered in this process, leading to a deterioration of overall performance. Potential solutions include deploying intermittent horizon stabilization or using a robotic interface to achieve fully aligned perceptual-motor control.