Biliary Tract Disease: Cysts, Leaks, Strictures, and Sphincter Dysfunction

0:00Welcome to Board Pearls. This is episode four of four of the Biliary Tract Disease chapter, in the Pancreatic and Biliary Disease module. This episode is the structural and iatrogenic bile duct: the congenital choledochal cysts and their cancer-driving anatomy, the post-cholecystectomy bile leak and stricture, and sphincter of Oddi dysfunction as it's now classified.

0:20The last episode stayed with the acquired duct, the stone and the sepsis it drives. This one turns to the duct that's structurally abnormal from birth or made abnormal by an operation, where the recurring move is to read the anatomy and let it dictate whether an endoscopic fix can work at all.

0:37Start with the congenital version, where the duct is structurally abnormal and you have to recognize a cancer-driving anatomy. Choledochal cysts are congenital dilations of the bile duct, and the Todani classification places the cyst by anatomy, which determines both the cancer risk and the operation. The first and most common type is fusiform dilation of the extrahepatic duct, frequently associated with a pancreaticobiliary maljunction. The second is a true extrahepatic diverticulum, uncommon, with low malignancy risk. The third is a choledochocele, an intraduodenal cystic dilation of the distal duct that often presents with pancreatitis because it intermittently obstructs the pancreatic orifice. The fourth is combined intrahepatic and extrahepatic cysts, or multiple extrahepatic cysts. And the fifth is Caroli disease, multiple intrahepatic dilations only.

1:29The reason the boards care is the cancer risk: cholangiocarcinoma develops in cyst epithelium at roughly ten to thirty percent over an adult lifetime, highest in the first type and lower in Caroli, with gallbladder and pancreatic cancer also elevated because the same maljunction physiology bathes the gallbladder in refluxed enzymes, and the mechanism is decades of chronic exposure of cyst epithelium to pancreatic enzymes and stagnant bile driving the metaplasia-dysplasia-carcinoma sequence, so longer exposure means higher risk, which is why malignancy rises sharply with age and early operation is preferred even in childhood.

1:29The operation follows the anatomy. The first, second, and fourth types all need complete cyst excision, cholecystectomy, and a Roux-en-Y hepaticojejunostomy, and the first type in particular cannot be cured by a Whipple, because a Whipple resects the distal duct but leaves the proximal cyst epithelium and its cancer risk in place. The diverticulum gets simple diverticulectomy, the combined type with diffuse intrahepatic disease may need segmentectomy or transplant, and the choledochocele is the exception, where the intraduodenal location and very low malignancy risk make endoscopic sphincterotomy or limited excision sufficient.

2:38Pancreaticobiliary maljunction sits behind many of these cysts, a long common channel of eight millimeters or longer where the ducts join above the sphincter, so without the sphincter to separate them pancreatic juice refluxes freely into the biliary tree, and most maljunction cases have associated cyst formation while the rest have maljunction without cysts, and in those the gallbladder bears the brunt and gallbladder cancer develops in about a third, which is why prophylactic cholecystectomy is part of the operation whenever maljunction is found, cyst or no cyst. After cyst excision, current guidance recommends lifelong MRCP every three to five years to surveil retained at-risk epithelium, which can still progress to cholangiocarcinoma decades later.

3:19Caroli disease, the fifth type, presents with recurrent cholangitis and intrahepatic stones because the dilation produces stasis and bacterial seeding, and Caroli syndrome is Caroli disease plus congenital hepatic fibrosis with portal hypertension, both reflecting a ductal-plate malformation that connects to the kidney, which is why they commonly associate with autosomal recessive polycystic kidney disease. Treatment depends on distribution, so unilobar disease can be cured by segmental resection while diffuse disease or Caroli syndrome with advanced portal hypertension needs transplant, with antibiotics managing individual cholangitis episodes and ursodeoxycholic acid sometimes used to reduce stone burden.

4:01The next problem is iatrogenic and one of the highest-yield frameworks in the chapter. Bile leaks present within ten days of cholecystectomy, with abdominal pain, bile peritonitis, fever, leukocytosis, elevated enzymes, and bilious drain output, and the Strasberg classification defines the lesion. Type A is a leak from the cystic duct stump or a small accessory subvesical duct, the most common and lowest-pressure leak. Type B is occlusion of an aberrant right hepatic duct without an active leak. Type C is a leak from a sectioned aberrant right hepatic duct that was cut but not ligated. Type D is a lateral injury to the side wall of the main duct with continuity preserved. And type E is complete transection or stricture with no continuity, subclassified by the Bismuth level relative to the hepatic confluence.

4:49The reason Strasberg matters is the management split, and the mechanism of endoscopic treatment is hydraulic: bile normally flows from the duct into the duodenum under a pressure gradient, and the sphincter imposes resistance that holds bile back, so any side hole in the duct preferentially leaks because the sphincter resistance pushes outflow through the lowest-pressure exit, and an endoscopic sphincterotomy plus a transpapillary stent reduces that resistance, so bile flows preferentially through the duodenum away from the leak and it seals in the large majority of type A injuries within four to six weeks. Stenting alone is more effective and safer than sphincterotomy alone, with sphincterotomy added when a retained stone contributes to the gradient, and type C and D injuries with luminal continuity can also be managed endoscopically because the stent can bridge the injury.

5:33Type E is the exception, because complete transection means no continuity, so there's no duct above the injury for a stent to reach, and it requires surgical reconstruction with a Roux-en-Y hepaticojejunostomy, with the Bismuth level predicting surgical complexity, because higher, more proximal injuries demand more proximal reconstruction and carry higher stricture rates and worse outcomes.

5:54So the man with bilious drain output, an intact duct on MRCP, and extravasation from the cystic duct stump is a type A, and he goes to ERCP for a sphincterotomy and stent, and the leak closes. The woman ten days out with jaundice, a bilirubin of fourteen, and an MRCP showing complete absence of filling above the porta hepatis has a type E, with no continuity for a stent to bridge, and she needs hepaticojejunostomy.

6:19Two imaging points: MRCP is the right test before ERCP because it shows continuity non-invasively and identifies the level before you commit to an intervention, and ERCP itself can miss injuries, because a right posterior duct injury can mimic normal opacification and a complete transection can be misread as poor cannulation rather than transection, both of which MRCP catches because it images bile-filled ducts without depending on pressurized contrast from below.

6:48Bile duct injury complicates a few per thousand laparoscopic cholecystectomies, usually from clipping or thermal injury during dissection of Calot's triangle, and anatomic variants raise the risk substantially, the most dangerous being a right posterior sectoral duct inserting low, which can be mistaken for the cystic duct and clipped, and that's the structural argument for intraoperative cholangiography.

7:11Post-cholecystectomy strictures present later, months to years out, in a patient returning with progressive jaundice and a focal short-segment stricture near the confluence, and the same level classification applies, with initial management endoscopic for most, using serial balloon dilation with multiple side-by-side plastic stents over a year, or covered metal stents in some centers, to remodel benign strictures into durable patency, reserving surgical hepaticojejunostomy for refractory strictures and complete transections, and deploying brushings, biopsies, and cholangioscopy when malignancy needs to be excluded.

7:40The last decision in the bile duct is sphincter of Oddi dysfunction, where the modern reframe changed what the right answer looks like. It's obstruction at the sphincter without a stone, from either passive fibrosis or active spasm, invoked in three contexts: recurrent biliary pain after cholecystectomy without structural cause, idiopathic recurrent acute pancreatitis, and biliary pain in a patient with an intact gallbladder and no stones, which is the least studied and most controversial. Before invoking it in a post-cholecystectomy patient, one physiologic point has to be cleared, because the bile duct dilates after cholecystectomy, up to about ten millimeters normally, since the gallbladder used to hold bile during fasting and now the duct does, so that normal post-cholecystectomy caliber must be distinguished from pathologic dilation first.

8:31The historic classification divided patients into three types by objective findings, and the modern reframe splits them cleanly. The former first type, biliary pain plus liver enzyme elevation plus a duct dilated above eight to ten millimeters, is now understood as a true mechanical stenosis, a fibrotic narrowing, and the right answer is empiric biliary sphincterotomy without manometry, because response rates are uniformly high regardless of what manometry shows and the manometry only adds procedural risk. So the patient with several episodes of severe biliary pain after cholecystectomy, transient enzyme spikes during attacks, a twelve-millimeter duct, and no stones is the prototype, and she goes to ERCP for sphincterotomy.

9:15The former second type, now called suspected sphincter dysfunction, is heterogeneous, with pain relief after sphincterotomy more likely when manometry is abnormal but a subset of manometry-negative patients genuinely responding, so decision-analysis supports empiric sphincterotomy by an experienced endoscopist, accepting some added complication risk to treat the manometry-negative responders.

9:37And the former third type, biliary pain alone with normal labs and imaging, is no longer a viable diagnosis, because a randomized trial of these patients comparing sham endoscopy against sphincterotomy found the sham group did at least as well on pain-related disability and manometry didn't predict response, so these patients should not undergo ERCP with or without manometry, and management is non-procedural, identifying and treating functional GI overlap with pain-focused therapies and neuromodulators. The young patient with post-cholecystectomy pain, mildly elevated transaminases, a six-millimeter duct, no opioid use, and a clean workup doesn't need a sphincter procedure, she needs a trial of a low-dose tricyclic and dietary modification.

10:19The unifying mechanism is concrete: the former first type is true mechanical stenosis that ablation cures, the former third type is functional pain that a procedure can't help and only exposes to pancreatitis risk, and the former second type is a heterogeneous middle where empiric sphincterotomy outperforms a manometry-driven strategy. And the pancreatitis risk in suspected sphincter dysfunction is exceptionally high, because instrumenting the sphincter triggers ductal pressure changes, so any sphincter assessment demands aggressive prophylaxis with rectal indomethacin one hundred milligrams, peri-procedural lactated Ringer's, and a prophylactic pancreatic duct stent in high-risk anatomy. Manometry should also be avoided in idiopathic recurrent acute pancreatitis without clear ductal disease, where the modern workup favors endoscopic ultrasound for microlithiasis and MRCP for pancreas divisum before any procedural sphincter assessment, and patients with biliary pain, an intact gallbladder, and no stones aren't candidates for sphincterotomy outside trials.

11:19Pull the structural and iatrogenic duct together. Choledochal cysts are the congenital version of a single logic, where decades of epithelial exposure to refluxed enzymes drive cholangiocarcinoma and the Todani anatomy dictates the operation, with complete excision the rule and the choledochocele the low-risk exception. Bile leak and stricture are the iatrogenic version, where the Strasberg injury level decides whether endoscopic stenting can work at all, because the stent needs continuity to bridge, so type A seals with a stent while type E needs reconstruction. And sphincter of Oddi dysfunction divides into a true stenosis that benefits from sphincterotomy, a functional pain that doesn't and shouldn't be instrumented, and a heterogeneous middle where empiric sphincterotomy beats a manometry-driven workup. Across all three, the move is to match the intervention to the mechanism, not to the symptom.

12:14That closes the biliary tract chapter. The next chapter shifts from the diseases to the procedures that treat them: endoscopy practice and sedation, covering pre-procedure fasting and antithrombotic management, sedation depth and pharmacology including the periprocedural GLP-1 guidance, and the standard adverse-event lexicon.

12:34For the full chapter, the practice vignettes, and the topic-tagged question bank, head to board pearls dot com. You'll find the rest of the series on Apple Podcasts, Spotify, or wherever you listen to podcasts. That brings us to the end of episode four of four of chapter twenty-seven, and I'll see you in the next one.