Su1252 ENDOSCOPIC ULTRASOUND TRENDS IN UNITED STATES: PRACTICE CHANGES OVER A DECADE

Abstract

The endoscopic ultrasound (EUS) fine needle aspiration (FNA) technique has been refined improving tissue acquisition and allowing targeted analgesic injection. During the last years, new research and guidelines advocate for conservative management of low risk pancreatic cysts. Simultaneously, an increasing number of reports have shown the limitations of celiac plexus block and neurolysis to treat chronic abdominal pain. We aims to describe changes in EUS practice over a decade. The Clinical Outcomes Research Initiative (CORI v.3) database was reviewed for EUS procedures between 2000 and 2010. EUS procedures were included if they had complete demographic data and details on FNA performance, celiac plexus block or neurolysis. Linear regression models compared temporary trends in FNA completion, celiac plexus block or neurolysis. Subgroup analysis evaluated FNA practice by different procedure indications (i.e. pancreatic mass, pancreatic cyst, esophageal lesion, and rectal lesion). EUS complications and medications are not consistently recorded in CORI, and were not included in our study design. Initial database review revealed 20,230 EUS procedures meeting our inclusion criteria. Mean age was 61.08 ±14.19 years, 11,936 (59%) patients were male, 17,027 (84%) non-Hispanic white, 1,476 (7%) non-Hispanic black and 1,197 (6%) Hispanic. The majority of procedures were completed as outpatients 14,782 (90%) and 8,202 (41%) had a fellow or trainee involved. Procedure indications included evaluation of a pancreatic mass (1,951 [10%]), pancreatic cyst (1,815 [9%]), esophageal mass (1,506 [7%]), and rectal or anal mass (1,087 [5%]). 275 patients (1%) underwent a celiac plexus block or neurolysis. Only 9,834 (49%) procedures provided details of the device used, with the majority done with a radial probe (77%). Overall FNA utilization increased over the study period (regression coefficient 0.10 [95%CI 0.02 to 0.18], P=0.013). Subgroup analysis revealed a decreasing trend in pancreatic cyst aspiration (-0.55 [-0.80 to -0.30], p <0.001). No significant trend was identified on use of FNA for pancreatic masses (0.20 [95%CI -0.17 to 0.56]), esophageal lesions (-0.05 [95%CI -0.40 to 0.30]) or rectal lesions (-0.74 [95%CI -1.93 to 0.45]) (p=0.29, 0.779, and 0.225 respectively) (Figure 1). An increase in celiac plexus block or neurolysis was seen during the time period evaluated (regression coefficient 0.75 [0.42 to 1.07], p<0.001). Despite an overall increase in FNA utilization in EUS procedures, there was a decreasing trend in pancreatic cyst aspiration. We hypothesize this is secondary to better management and radiological modalities for surveillance of pancreatic low risk lesions. Celiac plexus injection rate remained the same overtime. Efforts to limit these procedures to a select patient population should continue.