Frequency of Visualization of Celiac Ganglia by Endoscopic Ultrasound

Abstract

Purpose: Endoscopic ultrasound (EUS) visualization of celiac ganglia has been recently described. It is unknown how often celiac ganglia can be identified during EUS. The aims of this study were to identify the frequency of visualization of celiac ganglia in unselected patients undergoing upper GI EUS and to identify clinical factors that predict visualization of presumed celiac ganglia. Methods: Clinical, demographic, EUS and cytological data were prospectively collected from 200 patients undergoing upper GI EUS. Data collected included: age, alcohol/cigarette use, radial versus linear echoendoscope, presence or absence of presumed ganglia, number of ganglia identified, location, long axis length, shape, echogenicity, margins, presence of central hyperechoic strands and/or hypoechoic connecting fibers and presence of celiac adenopathy. Univariate and logistic regression analyses were performed to determine independent clinical predictors of ganglia visualization. Results: 200 patients, 97 (48.5%) male, mean age 59 ± 15 years, underwent EUS primarily to stage GI cancer (25%) or to evaluate pancreatic pathology (37%). Presumed celiac ganglia were visualized in 162 (81%) patients overall including 38/48 (79%) and 101/118 (86%) undergoing either radial or linear echoendoscope examinations (p= 0.31). There was a significant difference in the mean number of ganglia visualized per patient between radial and linear echoendoscopes, (1.56 ± 0.79 vs. 2.1 ± 1.1, p= 0.001). Female sex and no previous GI surgical intervention were the only significant factors associated with the visualization of celiac ganglia in logistic regression analysis, p= 0.027, p= 0.019 respectively. Identified ganglia were predominately oval in shape, with irregular margins, isoechoic with the adrenal gland, positioned between the celiac artery and left adrenal gland, with hypoechoic connecting fibers. Ten patients had FNA of presumed ganglia to rule out malignant lymphadenopathy. A median number of 2 passes revealed nerve cell bodies in all. Sampled ganglia had similar size and echo features as presumed ganglia. Nine of eleven patients with presumed celiac lymph nodes underwent EUS FNA, which revealed lymphocytes in all and malignancy in 2, without neurons. Conclusions: Presumed celiac ganglia were identified in 81% of patients undergoing upper gastrointestinal EUS exams. Female gender and absence of prior GI surgery predicted the visualization of ganglia.