Cholangiocarcinoma and Biliary Strictures

A biliary stricture describes a narrowed duct, while cholangiocarcinoma names one possible cause. I would keep the anatomical obstruction, the evidence for malignancy, and the need for drainage separate throughout the workup.

Clinical overview

Perihilar and distal cholangiocarcinoma often become apparent through jaundice, pruritus, or cholangitis. An intrahepatic tumor may instead present as a liver mass, with no early obstruction of the main drainage pathway. Painless jaundice raises concern for malignancy but does not identify its origin: pancreatic cancer, ampullary disease, and benign narrowing remain possible.

Primary sclerosing cholangitis (PSC) is an important predisposing condition. Its background strictures also make a new malignant narrowing difficult to recognize. I would ask what has changed from the patient’s previous ductal pattern, rather than interpret every abnormal segment as a newly discovered lesion.

The ACG guideline distinguishes establishing the cause from restoring drainage. I read this as two linked decisions with different stopping rules: falling bilirubin after stenting demonstrates improved drainage, but the malignancy question may remain unresolved.

Anatomy and pathophysiology

Location determines what becomes obstructed

Intrahepatic cholangiocarcinoma arises proximal to the second-order ducts. Perihilar disease lies between the second-order ducts and the cystic duct insertion; distal disease lies below that insertion. The BSG guideline uses these anatomical subtypes because their staging and treatment pathways differ.

A hilar lesion can isolate right and left drainage systems or individual sectors. Consequently, a single extrahepatic duct measurement cannot summarize the obstruction. A distal lesion can obstruct a shared downstream pathway, whereas a peripheral intrahepatic mass may affect only adjacent branches.

The cystic duct relationship also matters for gallbladder interpretation. Distension is a secondary response whose appearance depends on drainage anatomy, previous inflammation, and gallbladder function. I would not use an enlarged gallbladder as a histological test for malignant obstruction.

Shared tissue responses explain overlap

Cholangiocarcinoma can grow as a mass, infiltrate along the duct wall, or grow intraductally. Desmoplasia means abundant fibrous stroma accompanying tumor cells. Contrast retention in this extracellular tissue helps explain delayed enhancement; a narrow infiltrated duct need not contain a conspicuous bulky mass.

Benign strictures can also contain fibrosis and inflammatory cells. Surgical injury can heal by scarring, and ischemic injury can damage the duct wall. Chronic pancreatitis can narrow the intrapancreatic duct. I read the resulting imaging overlap as a limitation of tissue specificity, not simply insufficient image resolution.

Diagnostic workflow and imaging findings

Localize the obstruction before naming its cause

On ultrasound, I would trace visible dilated ducts toward a transition and record which downstream segments remain unseen. Color Doppler helps distinguish adjacent vessels from ducts. Bowel gas can obscure the distal common bile duct, so a visible proximal segment cannot establish distal patency.

The history should include cholecystectomy, transplantation, pancreatitis, PSC, previous drainage, and stent exchanges. These events change both the prior probability of different causes and the appearance being interpreted.

In a hypothetical examination, the intrahepatic ducts are dilated, the distal duct is not enlarged, and a shadowing stone is lodged at the gallbladder neck. Mirizzi syndrome, compression of the adjacent common hepatic duct by an impacted neck or cystic duct stone, becomes a concrete alternative to a primary duct tumor. Demonstrating that anatomical relationship matters more than assigning a generic “obstruction” label.

Give CT and MRI different questions

Multiphasic contrast CT can assess a mass, hepatic arterial and portal venous relationships, lymph nodes, and distant disease. MRI combines tissue characterization with MRCP, whose heavily T2-weighted images map fluid-containing ducts. Irregular shoulders, asymmetric wall thickening, enhancement, and an associated mass increase suspicion, but smooth tapering does not prove benignity.

The ESGE guideline supports MRI/MRCP for evaluating the level and cause of a suspected stricture in patients with jaundice or biochemical cholestasis. MRCP mapping and contrast-enhanced tissue assessment provide related but distinct information.

Where clinically feasible, staging images should precede instrumentation, as recommended by BSG. A stent can decompress the ducts, introduce air, and accompany reactive wall changes. I would preserve the pretreatment images because a later study may no longer show the original extent clearly. Urgent treatment of infection can take priority over an ideal imaging sequence.

Treat negative cytology as a sampling result

ERCP provides duct access for drainage, brush cytology, and intraductal biopsy. A brush principally collects cells accessible at the luminal surface. An infiltrative lesion embedded in fibrous tissue may shed few diagnostic cells, while reactive atypia can complicate interpretation.

The ASGE guideline suggests adding fluoroscopic biopsy to brush cytology. It also supports selected use of cholangioscopy-guided sampling for nondistal strictures and EUS-guided sampling for distal lesions or suspected nodal spread. These choices address where diagnostic tissue is accessible.

I would retain “insufficient,” “atypical,” “suspicious,” and “malignant” as separate sampling conclusions. Collapsing the first three into benign labels destroys information about why the procedure failed to settle the question. Persistent radiological concern requires reconciliation with sample adequacy and the subsequent course.

Choose the needle route with treatment in view

Direct transperitoneal sampling of a potentially transplant-eligible perihilar primary requires particular caution. Heimbach and colleagues reported an association between transperitoneal fine-needle aspiration and dissemination in a transplant-evaluation cohort. This was observational evidence, not a randomized estimate of biopsy harm.

I would therefore distinguish sampling the primary hilar lesion from sampling an appropriate suspicious node or distant lesion. The target, route, and proposed curative pathway belong in multidisciplinary planning before needle placement. “Tissue needed” does not specify that every accessible target is equally appropriate.

Differential diagnosis and management context

PSC can produce multiple strictures; IgG4-related sclerosing cholangitis can resemble malignant narrowing, particularly with autoimmune pancreatitis. A postoperative anastomotic stricture has a different anatomical history from a newly appearing hilar abnormality. External compression by pancreatic cancer is malignant obstruction without being cholangiocarcinoma.

CA 19-9 can rise during benign obstruction or cholangitis, and its expression depends partly on Lewis antigen biology. Serum IgG4 is also insufficient alone because elevation can occur in other diseases. ESGE places these tests alongside imaging and tissue assessment. I would record whether CA 19-9 was obtained before or after drainage rather than compare values without that context.

Drainage planning at the hilum must identify which functioning liver territories will actually drain. For surgical candidates, assessment includes ductal extent, vascular involvement, distant spread, and the anticipated liver remnant. Selected transplant pathways and oncological treatment require specialist evaluation. BSG emphasizes this multidisciplinary approach.

I would avoid treating “underwent resection” as synonymous with “imaging definitively established cancer.” Surgery can be undertaken despite unresolved preoperative histology when the overall evidence and treatment options justify it.

Implications for medical AI

Construct a timeline rather than a binary folder

For a gallbladder ultrasound dataset linked to biliary investigations, I would record the index scan, drainage procedures, each sampling result, and the final adjudicated diagnosis. A later malignant resection specimen can establish disease while an earlier brush remains correctly described as nondiagnostic.

I would also distinguish primary gallbladder cancer, cholangiocarcinoma, pancreatic malignancy, and benign obstruction. Pooling these as “malignant gallbladder images” could train a model to detect downstream consequences shared by several diseases.

Test intervention history as an alternative explanation

A concrete shortcut arises if malignant cases contribute post-ERCP scans with echogenic stents or pneumobilia, while benign controls contribute untreated scans. The model can exploit the workup pathway even when the tumor itself is outside the saved view.

My first audit would compare the frozen classifier on the complete cohort and on pretreatment examinations, retaining patient-level separation. I would then inspect performance within stent-status groups and review false positives in benign patients who also underwent drainage. This would test an observed association; it would not isolate every effect of treatment or referral.

Use anatomical disagreement to make the audit difficult

I could assemble gallbladder examinations with Mirizzi syndrome, benign distal strictures, and confirmed biliary malignancy, matched where feasible for duct dilatation and gallbladder distension. Readers would separately annotate direct lesions, secondary obstruction signs, and assessability.

My question would be whether a malignancy score distinguishes etiologies once the obvious obstruction pattern is shared. Another feasible comparison is between frames containing the gallbladder lesion and frames showing only dilated ducts from the same examination. If scores remain high in the latter, I would investigate contextual reliance before claiming recognition of malignant mural morphology.

References

• Elmunzer et al., ACG Clinical Guideline: Diagnosis and Management of Biliary Strictures, American Journal of Gastroenterology 2023.
• Rushbrook et al., British Society of Gastroenterology guidelines for the diagnosis and management of cholangiocarcinoma, Gut 2024.
• Fujii-Lau et al., American Society for Gastrointestinal Endoscopy guideline on the role of endoscopy in the diagnosis of malignancy in biliary strictures of undetermined etiology: summary and recommendations, Gastrointestinal Endoscopy 2023.
• Facciorusso et al., Diagnostic work-up of bile duct strictures: European Society of Gastrointestinal Endoscopy (ESGE) Guideline, Endoscopy 2025.
• Heimbach et al., Trans-peritoneal fine needle aspiration biopsy of hilar cholangiocarcinoma is associated with disease dissemination, HPB 2011.