March 29, 2019
By Dr Simon Lo, Cedars-Sinai Medical Center, Los Angeles, California
The scope. Device-assisted ERCP procedures are frequently carried out in patients whose biliary orifices are far away from the stomach. This may include those who have undergone bariatric Roux-en-Y gastric bypass (RYGB), biliary diversion surgeries, or a variety of gastrojejunostomies including Whipple surgery. While pediatric or adult colonoscopes may fit into this general category, we typically refer here to single-balloon, double-balloon and spiral overtube enteroscopes as the assisting devices. Most of today’s enteroscopes are 200 cm long and have a 2.8 mm-internal diameter instrument channel. A newer version of the double-balloon enteroscope has a 3.2 mm channel capable of inserting an 8.5-French stent. There is a shorter 152-cm double -balloon enteroscope that may be used to perform pancreaticobiliary procedures. Finally, a through-the-scope balloon may be used to assist enteral advancement of a colonoscope. Of course, this balloon has to be removed from the instrument channel before any pancreaticobiliary interventions are embarked upon.
Scope advancement technique. Single-balloon enteroscopy is probably the simplest to perform. It has a limited range of traveling distance, but that may be sufficient for most purposes except for patients with long-limb RYGB surgery or significant intra-abdominal adhesions. There is no modification of the common technique for passage of double -balloon enteroscopes. It is a tedious procedure, but the reward is a near certainty that it is long enough to reach the biliary orifice. Spiral enteroscopy is rarely practiced these days. Any endoscope supported by the spiral overtube can be used to perform ERCP similarly to the standard single- or double -balloon enteroscope. As opposed to the back-and-forth, pulling and pushing motions used in balloon enteroscopes, spiral enteroscopy relies on a clockwise rotation to drive the scope forward. Getting around the duodenojejunal junction can be quite challenging, but the reward is a stable positioning that is essential for a successful ERCP procedure.
Anatomical landmarks. Each surgically altered anatomy has its unique construction and findings that help guide scope passage. Entering the wrong intestinal limb invariably leads to frustration and sometimes physical exhaustion. We often ask ourselves whether it is time to withdraw the scope and explore the other lumen when the major papilla or hepaticojejunostomy is not readily visualized. Is it wise to look for another intestinal lumen without wasting more time, or is it better to be patient and try to go further down the limb? There are some general rules of thumb that may guide us down the right path (see Table 1).
Table 1 Anatomical landmarks and tips for enteroscope advancement
|Surgery||Traveling wisely up the afferent limb||How to find the biliary orifice|
|Billroth II gastrectomy||Afferent limb is frequently the acute lumen and it crosses the spine (midline) very quickly Efferent limb seems to reach the left lower quadrant readily||The minor papilla is often easier to visualize. If it is rotated to the 4-o’clock position, the major papilla should be seen at the 6–7-o’clock position by scope withdrawal of roughly 2 cm|
|RYGB||The Roux limb opens as the right lumen of the gastrojejunostomy. The afferent jejunojejunostomy is the forward-facing lumen immediately across the anastomosis||The minor papilla is often easier to visualize. If it is rotated to the 4-o’clock position, the major papilla should be seen at the 6–7-o’clock position by scope withdrawal of roughly 2 cm|
|Whipple||The afferent limb is most often found at the 10–12-o’clock position. Same relationship is seen whether it is a conventional or pylorus-preserving Whipple||The biliary orifice is often located at the 8–10-o’clock position and it corresponds with the top point of the afferent loop on fluoroscopy. The pancreatic orifice is likely found at the 4-6-o’clock position, within a 1–2-cm flat patch of jejunal mucosa. It is almost always several cm further upstream from the biliary orifice|
|Roux-en Y hepaticojejunostomy||The jejunojejunostomy opens at a few cm distal to the ligament of Treitz. The afferent limb is recognized as the forward-facing lumen immediately across the anastomosis||Expect to see one to three orifices, depending how high the surgical cut is made in the proximal bile duct. It is almost always located on the left side of the jejunum, but the actual position is quite variable|
|Loop gastrojejunostomy||No distinctive feature related to the afferent limb. It is a random selection of one of the two gastrojejunostomy orifices||The major papilla should be readily reached with an enteroscope. The distance from the gastrojejunostomy is typically short to modest in length|
|Duodenal switch||This is a rather laborious procedure. An alternative method should be used to study the pancreaticobiliary structure|
Cannulation of the intact papilla in surgically altered anatomy. Despite the challenges of passing an enteroscope and finding the pancreaticobiliary orifices, the real difficulty of performing device-assisted ERCP actually begins with bile duct cannulation. Discussion of pancreatic ductal manipulations and treatment is beyond the scope of this article and will not be covered here. The native papilla is best approached with a cap fitted at the tip of the endoscope. It serves to stabilize and even “capture” the papilla for steady probing. Gentle suctioning through the cap may expose the papillary orifice for easier intubation. The papilla should be placed at the 6–8-o’clock position, if possible. Even though it may seem easier sometimes to visualize the papilla at the 12-o’clock position, approaching it from the 5–7-o’clock position where the instrument exits the scope makes it difficult to advance the instrument into the bile duct. Once the bile duct is accessed, the priority is to perform a brief study and, if appropriate, to perform a sphincterotomy. It is rather difficult to get back into the bile duct if the instrument has fallen out of it without a wide-open sphincter. Sphincterotomy should be done with a guidewire in place. Ideally the cutting wire should be perpendicular to the hood of the papilla. Unfortunately the cutting wire in the only commercially available wire-guided sphincterotome does not align well for effective cutting. Therefore, it is best to perform the cutting by modifying the bow-string sphincterotome into a needle knife. With a guidewire maintaining ductal access, the over-the-wire needle knife can be used to cut down towards the guidewire or cut outward (inside-out method) by first extending the needle alongside the guidewire inside the common channel. We prefer to cut downward to avoid the tendency to pull the guidewire out of the bile duct during inside-out cutting.
Approaching a biliary–enteric anastomosis. Hepaticojejunostomy is much easier to cannulate than the intact papilla in an altered anatomy, as the orifice is often seen in straight views. However, the right hepatic duct is typically more difficult to approach because the afferent limb commonly comes up to anastomose with the bile duct from the right side of the patient, thus favoring instrumentation of the left ductal system. Fluoroscopy may help to direct the tip of the endoscope to engage the right system that is otherwise very hard to approach.
Biliary stenting. Placement of plastic stents is often a tedious process through these skinny long scopes. Even for 7-French stents, a stent constrainer (or positioning sleeve) must be used to prevent the stent’s flap(s) from jamming up the small enteroscope channel. Since a single stent of that caliber is typically inadequate for maintaining ductal patency for longer than a month, we often insert 2 to 4 stents in parallel. This requires successive insertions that take a long time to perform. To overcome this issue, we frequently load up 3 or 4 stents, back-to-back, over a guidewire. After a successful placement, we would re-insert the guidewire into the duct for the second stent and repeat the same process for the third stent. Metal stents can be placed through these scopes as well. In this situation, the single- or double-balloon overtube should be advanced as close to the biliary orifice as possible. Through the scope, a long guidewire (600 cm in length) is inserted into the bile duct. The scope is then removed while the guidewire and overtube are carefully kept in place. A biliary metal stenting device can then be advanced over the guidewire into the bile duct, where the stent is released under fluoroscopic guidance. We have even placed side-by-side metal stents successfully in this manner. Since it is often a struggle to pass a scope through a surgically altered upper GI tract, it is best to minimize the need to repeat an ERCP as much as possible. One way to do so is to make use of a nasobiliary (NB) drain. While the commercially available NB drain sets may be too short to allow a standard exchange, it is quite easy to do so by using a long ERCP catheter to push the NB drain beyond the tip of the enteroscope when it is outside of the patient.
The dos and don’ts when performing device-assisted ERCP:
Do pay attention to bile fluid in the intestinal lumen. The first sight of bile is often within 10–30 cm proximal to the jejunojejunostomy anastomosis. Slow down and look for the anastomosis when you see bilious fluid so that you can take time to correctly select the afferent jejunal orifice to enter. It can be a frustrating experience if you unknowingly zoom past the anastomosis and end up spending the next 30 minutes in the efferent limb.
Do bring the overtube as close to the biliary opening as possible. The tip of the scope needs to be stabilized by the overtube balloon to help with cannulation and further manipulations
Do place tattoo marks to help identify the afferent lumen in future procedures, even if you don’t think that a repeat study is needed at the time. It is particularly important to mark the anastomosis and its afferent limb entrance
Do use fluoroscopy to help find the biliary orifice, especially when there is no endoscopic clue within the intestinal lumen. Bile duct air is the best guide. But when it is not seen, look for the most cephalad intestinal air and direct the scope to that location to search for the orifice
Do use fluoroscopy to confirm the location of the endoscope in the afferent limb of a Billroth II anatomy. It should show a straight scope that crosses the spine within only a few centimeters beyond the stomach.
Do insert a long guidewire into the presumed “wrong” lumen of a Roux-Y construction during withdrawal towards the jejunonjejunostomy anastomosis. It is often difficult to be certain when the scope is coming out of one of the several lumens until it is too late. Having the wire in the lumen allows us to be more confident when searching for the alternative anastomotic opening without going back into the same limb over and over again.
Do apply 0.5 ml of vegetable oil into an enteroscope channel before insertion of most ERCP accessories. The importance of lubricating the scope lumen cannot be overstated. Don’t use an excessive amount of oil, as spilled oil may muddy up the viewing lens of the endoscope.
Don’t use luminal bile to determine whether the scope is within the afferent limb. This is the biggest misconception of most endoscopists. Collections of bile fluid may just as easily be found in the afferent as the efferent limb.
Don’t assume the acute lumen at a jejunojejunostomy anastomosis is the entrance to the afferent limb. In fact, it is almost always the opposite.
Don’t sweep the bile duct by forceful pulling of the stone balloon or withdrawal of the enteroscope. There is the risk of the scope or balloon falling back by a substantial distance and not getting back to the original position. It is better to pull the balloon down to the distal duct and hold it in place, while rotating or torquing the scope to the side. This will allow controlled balloon sweeps without losing bile duct access or the scope‘s falling out of position.
Dr. Simon Lo is a gastroenterologist in Los Angeles, California and is affiliated to the Cedars-Sinai Medical Center. He received his medical degree from NYU School of Medicine and has been in practice for more than 20 years.