Endoscopic Ultrasound and EUS-Guided Endotherapy


Overview

Endoscopic ultrasonography (EUS) was first introduced in the 1980s and has since become a cornerstone of endoscopic diagnosis of pancreaticobiliary disease, tissue acquisition via fine-needle aspiration (FNA), and staging of gastrointestinal malignancy. As experience with directed needle access into adjacent organ systems increased, endoscopists were able to expand the utility of EUS to therapeutic interventions. This evolution includes EUS-guided drainage of pancreatic fluid collections, abscesses, cholangiography, biliary drainage, pancreatic duct access, gallbladder drainage, directed cancer therapies, celiac nerve block and neurolysis, cyst ablation, and, most recently, enteroenteral anastomosis. In this chapter, the role of EUS-directed intervention in pancreaticobiliary disease will be reviewed in detail, and the indications, the mechanics, the outcomes, and the adverse events will be examined.

Endoscopic Ultrasonography

EUS combines luminally based endoscopic video capabilities with an ultrasound probe. Echoendoscope designs include catheter-based miniprobes and radial array echoendoscopes that provide a 360-degree sonographic view. For therapeutic purposes, curvilinear (or “linear”) scopes with a 100-degree to 120-degree view allow for sonographically directed advancement of needles, stents, and devices. Both forward and oblique endoscopic viewing linear scopes exist, and device channels range from 2.8 to 4.2 mm in diameter. All current models of EUS scopes have Doppler capability for vascular imaging.

EUS imaging provides detailed and rich diagnostic information in pancreaticobiliary disease. Imaging quality is dependent on the echoendoscope itself, the ultrasound processor, and ultimately the end-user's ability. The entire pancreas, including the pancreatic parenchyma, pancreatic duct, and uncinate processs, can be seen. The bile duct is easily visible and accessible, as is the left lobe of the liver, the gallbladder, and lymph nodes in the porta hepatis, the celiac region, and the gastrohepatic ligament. As such, in the hands of experienced endosonographers, EUS is highly sensitive and useful in the diagnosis of chronic pancreatitis, pancreatic malignancy, choledocholithiasis, cholangiocarcinoma, and pancreatic cysts.

EUS-guided endotherapy is widening in its reach in the pancreaticobiliary system and now includes a number of indications:

  • 1.

    Celiac nerve block and celiac neurolysis

  • 2.

    Drainage of pancreatic fluid collections (pseudocysts and walled-off necrosis [WON]) ( Chapter 56 )

  • 3.

    Pancreatic duct access and drainage

  • 4.

    Biliary drainage (see Chapter 32 )

  • 5.

    Gallbladder drainage

  • 6.

    EUS-guided ablation and cancer therapy

Celiac Nerve Block and Neurolysis

Both pancreatic cancer and chronic pancreatitis can cause debilitating pain. Narcotic medications are frequently necessary for pain control, but as pain becomes increasingly intractable and as dosage increases, medication side effects such as nausea, constipation, and drowsiness may result. Celiac plexus neurolysis (CPN) with ethanol injection for pancreatic cancer patients and celiac plexus block (CPB) with steroids for chronic pancreatitis patients can both be useful adjuncts in the palliation of chronic pain. The celiac plexus surrounds the celiac artery trunk, and several ganglia lie interconnected in the region, transmitting pain signals from the pancreas to the splanchnic nerves and then to the central nervous system. In the case of pancreatic cancer, it is believed that the pain is caused by perineural invasion of the pancreatic nerves. Neurolysis with ethanol, or less commonly with phenol, aims to destroy the ganglia and neural pathways, resulting in neuronal degeneration and fibrosis, but histopathologic evidence suggests only axonal and fascicular damage by chemical injection, allowing for the persistence of neuronal tissue and only short-term benefit. Direct ganglionic injection may increase neuronal damage and thus provides a more effective and durable response. In the case of chronic pancreatitis, CPB with triamcinolone and bupivacaine theoretically treats the perineural inflammation associated with chronic pancreatitis, although bupivacaine alone has been shown to be equivalent to a mixture of the two.

CPN and CPB can be performed percutaneously under radiologic guidance, surgically, or under EUS guidance. Percutaneous posterior approaches are effective, but rarely (<1%) spinal cord trauma or ischemia can occur and result in neurologic complications such as lower limb weakness, paresthesias, and paraplegia. Thus anterior approaches using EUS-guided CPN are more attractive. Regional anatomy and landmarks are generally easily visualized with linear echoendoscopes, and fine-needle injection can be performed precisely and safely.

All patients with pancreatic cancer or chronic pancreatitis and chronic abdominal pain are potential candidates for CPB/CPN. It is generally suggested that the technique be reserved for patients with pain severe enough to require narcotics, and particularly if there is evidence of adverse effects from chronic opioid use.

Technique

The celiac plexus is located around the trunk of the celiac artery. Ganglia are variably visible sonographically, and detection rates vary greatly, from 73% to 89%. ( Fig. 33.1 ). If a central approach is taken, the needle is advanced just superior and anterior to the celiac artery take-off. If a bilateral approach is taken, the scope is rotated clockwise and counterclockwise to either side of the celiac axis until the artery is no longer visible, and the needle is advanced into the space adjacent to the trunk. Different needles can be used, generally 22-gauge or 19-gauge FNA needles, or specially designed fenestrated 20-gauge needles (Cook Medical, Winston-Salem, NC). If ganglia are targeted, sonographers should look for hypoechoic oblong structures measuring between 0.5 and 4.5 cm with central hyperechoic foci or bands. Mixtures vary, but 5 to 10 mL of 2.5% bupivacaine mixed with 15 to 20 mL of absolute ethanol is injected centrally or divided and injected bilaterally or into visualized ganglia. In CPB, ethanol is included as part of the injectant, but most sonographers use a mixture of 5 mL of diluted triamcinolone (40 to 80 mg) and 10 to 20 mL of bupivacaine, injected centrally in the celiac region above the trunk.

FIG 33.1, Celiac axis viewed with linear echoendoscope, with visible celiac ganglion (white arrow).

Efficacy

Overall efficacy of CPN for pain reduction in inoperable pancreatic cancer is estimated at 80% in meta-analysis. In a meta-analysis, EUS-CPN has been shown to reduce pain compared with opioids at 4 and 8 weeks and to reduce opioid consumption. Central injection has been shown to be as effective as bilateral injection. If ganglia are identified, direct ganglia injection has been shown to be more effective than standard CPN, with an almost 30% absolute improvement in partial and complete response rates.

Alcohol-based CPN for chronic pancreatitis has been shown to provide relief in 59.4% of patients. CPB for chronic pancreatitis has been shown in meta-analysis to provide significant relief of pain in 51.4% of patients and is more effective compared with percutaneous CPB. No difference has been found in randomized controlled trials of central versus bilateral injection or in a randomized controlled trial of bupivacaine alone versus bupivacaine and triamcinolone. Duration of response is estimated at 10 weeks (range 1 to 54 weeks) after the initial procedure, and repeated CPB appears safe and effective if there is response to the initial CPB.

Adverse Events

The most common adverse events from CPB/CPN are transient hypotension (1%), diarrhea (4% to 15%), and transient increase in pain (9%). Infections are also reported, including retroperitoneal abscess, and some sonographers recommend administration of prophylactic antibiotics. Despite improved safety compared with percutaneous approaches, anterior spinal cord infarction with paralysis has been reported. There are also rare reports of lethal complications such as necrosis and perforation of the stomach and aorta.

Drainage of Pancreatic Fluid Collections

EUS-guided drainage of pancreatic fluid collections (EGDPFC) is now a well-established procedure. The nomenclature for pancreatic fluid collections has evolved and it is important to recognize the differences between subtypes, as successful drainage and avoidance of adverse events (AEs) is partly dependent on recognition of subtypes and how to best manage them (see Chapter 56 ). Acute fluid collections form shortly after the onset of pancreatitis, within the first 4 weeks. They are homogeneous and without debris or encapsulation. Most will not become infected and will resorb without intervention. Pseudocysts are often the result of a maturing acute fluid collection. They do not have necrotic material and are normally round, with a thick wall, and usually do not become infected ( Fig. 33.2, A ). If large, they may be symptomatic and require drainage. If a significant amount of necrosis is present, resulting computed tomography (CT) findings are defined as acute necrotic collections, with nonenhancing areas of variable attenuation. Over time, generally 4 weeks after onset, these mature and develop a wall, referred to as WON, which can cause unremitting pain, gastric or duodenal obstruction, or infection ( Fig. 33.2, B ).

FIG 33.2, Pancreatic fluid collections vary in etiology, appearance, composition, and behavior. A, An example of a pseudocyst within the uncinate process. Note round, thick-walled appearance. B, Typical walled-off necrosis 4 weeks after onset of severe acute pancreatitis. Note heterogeneous enhancement with small pocket of air and start of rind formation around the fluid collection.

Treatment of pancreatic fluid collections should be reserved for organized collections (pseudocysts and WON) that are causing symptoms or complications. Acute fluid collections are rarely infected and should generally not require drainage, and attempting drainage of unorganized fluid or necrosis will often lead to worse outcomes. Recent efforts have been targeted at endoscopic management of WON, and EUS has become a cornerstone of therapy, as it has been shown to be clearly superior to conventional “blind” endoscopic drainage, with lower cost and shorter hospital stay compared with surgical cystgastrostomy. Several variations of care now exist, there are no clear standard approaches, none can be generalized, and the approach taken is often more dependent on local practice and expertise. Transgastric or transduodenal stenting can be performed with plastic stents, covered self-expanding metal stents (SEMS), or lumen-apposing stents (LAMS). Some have advocated for the use of nasocystic drains for repeated flushing, whereas others have had remarkable success with dual-modality drainage, in which a percutaneous drain is placed in addition to transgastric stents and gradually upsized to allow for drainage of liquefying necrotic tissue. More recently, endoscopic necrosectomy has been performed in centers worldwide, to debride the necrotic cavity after access through dilated transluminal tracts or through preexisting metal stents ( Fig. 33.3 ).

FIG 33.3, Necrosectomy in a 46-year-old man with walled-off necrosis. A 15-mm Axios lumen-apposing stent was placed transgastrically into the necroma (A) and the necroma was entered (B) . The majority of necrosectomy was performed using a snare (C and D) .

You're Reading a Preview

Become a Clinical Tree membership for Full access and enjoy Unlimited articles

Become membership

If you are a member. Log in here