EUS-guided celiac ganglia neurolysis: a clinical and human cadaver study (with video)

Background and Aims There is little evidence that structures targeted during EUS-guided celiac ganglia neurolysis (EUS-CGN) are celiac ganglia and that selective ethanol injection into ganglia is feasible. We aimed to visualize celiac ganglia, confirm that these structures are ganglia, and visualize ethanol spread after EUS-CGN and EUS-guided celiac plexus neurolysis (EUS-CPN). Methods First, celiac ganglia were sought during 97 consecutive EUS procedures. Second, ganglia were identified in a prosected human cadaver by placing a linear echoendoscope next to the celiac trunk and removing the underlying tissue for histology. Finally, various EUS-CGN and EUS-CPN techniques were performed in human cadavers; EUS-CGN was performed with 1 mL ethanol in 1 ganglion, 1 mL per ganglion (both low volume), and 4 mL per ganglion (high volume). EUS-CPN was performed with a central (20 mL) and a bilateral (2*10 mL) approach. Transverse sections (75 μm) were obtained and photographed to allow visualization of the spread of ethanol. Results A total of 204 ganglia were detected in 83 patients. Mean (± standard deviation) size of the long axis was 8.1 mm (± 7.4 mm). Histology of the removed region in the cadaver showed only nerve cell bodies. After low-volume EUS-CGN in cadavers, ethanol spread well beyond the targeted ganglion. After high-volume EUS-CGN in cadavers, a larger ethanol spread was seen, which also reached unidentified ganglia; the spread was comparable to the spread after EUS-CPN. Conclusions Specific EUS-CGN is not feasible because ethanol spreads well beyond the targeted ganglion. Unidentified celiac ganglia are better reached with high-volume EUS-CGN, and this would likely result in a more thorough neurolysis. High-volume EUS-CGN should be preferred to low-volume EUS-CGN.