Tu1545 A Prospective Feasibility Study on EUS Guided Broad Plexus in Combination of Celiac Ganglion Neurolysis in Pancreatic Cancer Pain

Abstract

case. Staging based on clinical information and EGD alone was compared to the staging provided by EGD with EUS. A kappa statistic was used to assess interobserver agreement. Results: Two EUS-trained endoscopists who did not perform the original case correctly provided a T-stage in 11/16 and 12/16 cases with EGD alone compared that provided by EGD with EUS. The thoracic surgeon and gastroenterologist not trained in EUS correctly provided a T-stage grade in 9/16 and 8/16 cases, respectively. When we adjusted for the need for neoadjuvant therapy based on T stage only (T1 or T2 versus T3 or T4), gastroenterologists trained in EUS correctly staged the cancer in 13/16 (2 over-staged, 1 under-staged) and 12/16 (3 over-staged, 1 under-staged) cases, respectively. The thoracic surgeon and noninterventional gastroenterologist correctly staged the cancer in 9/16 (2 over-staged, 5 under-staged) and 12/16 (4 under-staged) respectively. The kappa statistic was 0.42 for T1-2 versus T3-4 for all observers and 0.76 for T1-2 versus T3-4 in EUS-trained endoscopists. Conclusions: In physicians with training in EUS, EGD alone allows accurate T-staging of esophageal malignancies in approximately 75% of cases when compared to EGD with EUS. Physicians trained in EUS tend to over-stage lesions on EGD; thus, possibly leading to recommendations of unnecessary neoadjuvant therapy. The kappa statistics demonstrated substantial interobserver agreement among endoscopists trained in EUS and moderate interobserver agreement for all evaluators. In summary, EGD alone cannot replace EGD with EUS for staging of esophageal cancer.