Natural orifice transesophageal mediastinoscopy and thoracoscopy: a survival series in swine

Introduction: Transesophageal endoscopic mediastinoscopy (MX) and thoracoscopy (TX) could reduce pain, eliminate intercostal neuralgia, provide better access to the posterior mediastinal compartment and pulmonary hilum, and improve cosmesis. The purpose of this study was to demonstrate the feasibility of transesophageal natural orifice translumenal endoscopic surgery (NOTES) and to determine the complications that might be seen in surviving animals. Methods: Using cap endoscopic mucosal resection and blunt dissection, a 15–20 cm submucosal tunnel was created in the esophagus and an endoscope passed through the tunnel into the mediastinum. One swine underwent MX; three swine underwent both MX and TX. The mediastinal compartment, hilar lymph nodes, pleura, lung, and esophagus were identified. Esophageal closure was obtained via submucosal tunnel flap-valve alone (two swine) or reinforcement with mucosal clips (two swine). The esophagus, mediastinum, and thorax were examined at necropsy. The esophagus was excised and sent for pathological examination. Results: NOTES MX and TX provided excellent visualization of mediastinal and thoracic structures. Pleural biopsy was easily accomplished. All animals survived the procedure, ate well, and showed no ill effects. Swine were sacrificed at either 8 or 12 days postoperatively. At necropsy, mild atelectasis was noted in each animal. One animal (mucosal clip closure) developed a fluid collection in the submucosal tunnel. There was no evidence of mediastinitis or thoracic contamination in any animals. Conclusions: Transesophageal endoscopic mediastinoscopy and thoracoscopy provide excellent visualization of mediastinal and intrathoracic structures. Pleural biopsy can be easily obtained under direct visualization. Structures that are difficult to visualize via traditional cervical mediastinoscopy and thoracoscopy are seen well with this approach. The submucosal tunnel creates a flap-valve that, alone, may be sufficient for preventing esophageal leak. These procedures can be performed safely in swine with short-term survival. Further study with a larger sample size and longer survival is warranted