🩺 Endoscopic Ultrasound-Guided Biliary Drainage Simulator

When ERCP Fails — The Case for EUS-Guided Biliary Drainage

Endoscopic retrograde cholangiopancreatography (ERCP) is the first-line approach to biliary decompression, but transpapillary cannulation fails in roughly 5–10% of attempts even in the hands of expert endoscopists. EUS-guided biliary drainage (EUS-BD) has emerged over the past decade as the preferred salvage strategy in these failures, and as the primary approach when the papilla is anatomically unreachable altogether.

• 5–10%: ERCP failure rate (even at expert-volume centers)
• ~30–40%: PTBD re-intervention rate (tube dislodgement/occlusion)
• >10×: EUS-BD adoption growth (over the last decade)
• Yes: External catheter avoided (internal drainage, no skin tube)

Why transpapillary cannulation fails

Failed biliary cannulation at ERCP is not a single entity — it reflects several distinct anatomic and pathologic obstacles that converge on the same endpoint: the endoscopist cannot get a wire into the bile duct from the duodenal side.

Common causes include periampullary or duodenal tumor infiltration distorting the papilla, a peri-diverticular papilla that cannot be aligned with the scope axis, impacted stones or strictures that resist standard sphincterotome cannulation, and prior surgery that has physically removed or bypassed the papilla from the reachable lumen. In malignant hilar or distal obstruction, tumor bulk can obliterate the ampullary orifice entirely.

Historically, the fallback after failed ERCP was percutaneous transhepatic biliary drainage (PTBD) — a interventional-radiology-placed external catheter through the liver parenchyma into a dilated intrahepatic duct. PTBD is effective but carries a persistent external tube, a meaningful quality-of-life burden, and re-intervention rates around 30–40% due to tube dislodgement, occlusion, or catheter-site infection.

Surgically altered anatomy as a primary indication

A second, growing indication for EUS-BD is anatomy in which the papilla is endoscopically unreachable regardless of cannulation skill. This includes Roux-en-Y gastric bypass (the papilla lies beyond a long efferent limb inaccessible to a standard duodenoscope), pancreaticoduodenectomy (Whipple resection, which removes the papilla and reconstructs a hepaticojejunostomy), and Billroth II gastrectomy (an afferent-limb approach to the papilla that is technically difficult and carries a higher perforation risk with conventional side-viewing scopes).

In these settings, EUS-BD converts what would otherwise require deep enteroscopy-assisted ERCP (long, technically demanding, with lower success rates) or PTBD into a single-session, internal-drainage procedure performed transgastrically or transduodenally from wherever the echoendoscope can safely position against a dilated duct.

EUS-BD has displaced PTBD as the preferred second-line drainage strategy in expert centers: comparative series show similar technical and clinical success with significantly fewer unplanned re-interventions and better patient-reported quality of life, since there is no external catheter to manage, dress, or accidentally dislodge.

Locating and Sizing the Target Duct — Transgastric and Transduodenal Views

Before any needle is passed, the echoendoscope must find a duct that is both anatomically reachable and large enough to puncture safely. This stage is pure diagnostic ultrasound: identifying dilation, confirming a puncture window free of interposed viscera, and clearing the trajectory of vasculature with color Doppler.

• ≥5 mm: Minimum safe duct size (ideally >6–8 mm)
• Segment II/III: Transgastric target (left intrahepatic ducts)
• Extrahepatic CBD: Transduodenal target (from duodenal bulb)
• Routine: Doppler use (excludes interposed vessels)

Two acoustic windows: stomach and duodenal bulb

A linear-array echoendoscope provides real-time, side-viewing ultrasound alongside an interventional working channel — the same platform used for EUS-FNA — which allows the operator to visualize the duct, plan a needle path, and puncture through the same instrument without exchanging scopes.

From the gastric position (typically high on the lesser curve or cardia), the left intrahepatic ducts — segment II and III branches — are visualized transgastrically. This view is preferred when the duodenum is inaccessible (altered anatomy, duodenal tumor obstruction) and forms the basis of hepaticogastrostomy.

From the duodenal bulb, the extrahepatic common bile duct is visualized transduodenally as it courses adjacent to the portal vein. This view is preferred when the duodenal bulb is reachable and undistorted, and forms the basis of both choledochoduodenostomy and the antegrade rendezvous technique.

Duct diameter thresholds and Doppler vessel mapping

Duct caliber is the single most important safety parameter at this stage. A target duct should generally measure at least 5 mm, with 6–8 mm or larger considered ideal: a wider duct gives the needle a larger, more forgiving target, reduces the chance of a through-and-through puncture of the far wall, and provides more working room for guidewire manipulation and subsequent tract dilation.

Below roughly 5 mm, technical difficulty and complication risk rise sharply — the needle tip is harder to keep confidently within the ductal lumen, aspiration of bile is less reliable, and the margin for error during tract dilation and stent deployment narrows considerably. Color and power Doppler interrogation of the planned needle trajectory is performed routinely to exclude interposed gastric, splenic, or portal venous branches, since the needle path is otherwise blind to vasculature on grayscale imaging alone.

19-Gauge Puncture, Cholangiogram, and Guidewire Advancement

With a suitable duct identified, the procedure moves from diagnostic to interventional: a 19-gauge needle crosses the gut wall into the duct under simultaneous EUS and fluoroscopic guidance, bile aspiration and contrast injection confirm correct placement, and a guidewire is coiled within the biliary tree to secure access for everything that follows.

• 19G: Needle gauge (standard FNA-type needle)
• 0.025–0.035": Guidewire caliber (stiff, hydrophilic-tip options)
• 2: Confirmation steps (bile aspiration + cholangiogram)
• 2: Guidance modalities (real-time EUS + fluoroscopy)

The puncture sequence

Under continuous ultrasound visualization, the 19G needle is advanced through the gastric or duodenal wall directly into the target duct in a single controlled pass, keeping the needle tip visible on the screen at all times to avoid inadvertent injury to adjacent structures. Once the needle is believed to be intraductal, bile is aspirated through the needle — free flow of bile is the first and most reliable confirmation of correct ductal entry.

Contrast is then injected through the needle under fluoroscopy to obtain a cholangiogram, outlining the intrahepatic or extrahepatic biliary tree, defining the level and length of any stricture, and confirming that the needle tip truly communicates with the ductal system rather than a false passage.

Guidewire advancement and route selection begins here

A 0.025–0.035 inch guidewire is passed through the needle lumen into the duct and manipulated — under fluoroscopic control — either to coil freely within the intrahepatic branches (securing access for a hepaticogastrostomy) or to be advanced antegrade, down the bile duct, across the native papilla or stricture, and out into the duodenal lumen (the first step of the rendezvous technique).

This single guidewire maneuver is the fork in the road for the entire procedure: successful antegrade passage across the papilla opens the option of a rendezvous approach that preserves normal transpapillary anatomy, while failure to cross — due to a tight stricture, tortuous duct, or altered anatomy — commits the case to a transluminal stent route (hepaticogastrostomy or choledochoduodenostomy) instead.

Losing guidewire access after the initial 19G puncture is one of the most feared technical mishaps in EUS-BD — with the tract only just created and no stent yet in place, wire loss can necessitate re-puncture and meaningfully raises the risk of bile leak. Maintaining wire position across every subsequent exchange is a central technical discipline of the procedure.

Choosing the Drainage Route — Hepaticogastrostomy, Choledochoduodenostomy, or Rendezvous

With guidewire access secured, the operator chooses among three technical routes, dilates the fistula tract, and deploys a stent to create durable bile flow. The choice of route is driven chiefly by which lumen is accessible and whether the wire successfully crossed the papilla — not by operator preference alone.

• 8×8 / 10×10 mm: LAMS diameters used (electrocautery-enhanced)
• Duodenum inaccessible: EUS-HGS use case (or altered anatomy)
• Duodenum accessible: EUS-CDS use case (extrahepatic CBD target)
• Preserves anatomy: EUS-RV advantage (normal transpapillary flow)

Three routes, three anatomic logics

Hepaticogastrostomy (EUS-HGS) places a stent directly from an intrahepatic duct (typically segment II/III) into the stomach, creating a new bilioenteric fistula. It is the route of choice when the duodenum cannot be reached at all — malignant duodenal obstruction, surgically altered anatomy such as Roux-en-Y bypass, or a prior Whipple reconstruction — because the stomach remains reliably accessible from a standard peroral approach.

Choledochoduodenostomy (EUS-CDS) places a stent from the extrahepatic common bile duct into the duodenal bulb. It is favored when the duodenal bulb is accessible and undistorted, generally offers a more direct, shorter tract than HGS, and is often technically faster.

Rendezvous (EUS-RV) does not place a transluminal stent at all: the guidewire advanced antegrade across the papilla in Stage 3 is grasped by a duodenoscope, which then completes a conventional transpapillary ERCP stent placement. Because it restores normal physiologic bile flow through the native papilla rather than creating a new enteric fistula, EUS-RV is the preferred option whenever the wire successfully crosses — but it is only feasible in a minority of cases.

Tract dilation and lumen-apposing metal stent deployment

Before a stent can be placed across the newly created fistula, the puncture tract must be dilated — using either a balloon dilator or an electrocautery-tipped dilating catheter that cuts and coagulates simultaneously as it advances over the wire, reducing the number of instrument exchanges and the risk of losing access mid-procedure.

The dominant device for HGS and CDS today is the lumen-apposing metal stent (LAMS): a short, saddle- or dumbbell-shaped, fully covered metal stent with wide flanges on each end that cinch the gut wall directly against the duct wall, apposing the two lumens and sealing the tract against bile leakage as it deploys. Electrocautery-enhanced delivery systems allow the stent to be deployed through the same device used for tract access, in a single "hot" step. Common sizes are 8×8 mm and 10×10 mm (flange diameter × saddle length), chosen to match duct depth and the thickness of the intervening wall. In rendezvous cases, no LAMS is needed — a standard plastic or metal biliary stent is placed transpapillary as in routine ERCP.

Technical and Clinical Success, and the Complication Profile of EUS-BD

In expert hands, EUS-BD achieves technical success — defined as successful stent or wire deployment establishing bile flow — in roughly 90–95% of attempts, with clinical success (normalization of bilirubin and resolution of cholangitis or jaundice) in about 85–90%. These figures now rival or exceed those historically reported for PTBD, while avoiding an external drainage catheter altogether.

• 90–95%: Technical success (expert-center series)
• 85–90%: Clinical success (bilirubin normalization)
• 15–20%: Overall complications (any adverse event)
• ~5–10%: Serious adverse events (requiring intervention)

The complication spectrum

The most feared early complication is bile leak, which can progress to localized peritonitis or frank bile peritonitis if the fistula tract is inadequately sealed — this is precisely the failure mode that LAMS flange design is intended to minimize compared with older plastic-stent transluminal techniques. Pneumoperitoneum from insufflation and tract manipulation is common on imaging and usually self-limited, but can occasionally require decompression.

Stent migration — inward into the duct or outward into the gut lumen — is a recognized late complication of both HGS and CDS, sometimes necessitating stent exchange or revision. Bleeding can occur at the puncture or dilation site, particularly if Doppler mapping was incomplete or a small vessel was traversed. Cholangitis can develop if drainage is incomplete or the stent occludes with debris, sediment, or tumor ingrowth over time.

Modifiable risk factors: duct size and anatomy

Two variables tracked throughout this simulation — duct diameter and the presence of surgically altered anatomy — are the dominant modifiable determinants of procedural risk and route selection in real practice. Ducts below the ~5 mm safety threshold are associated with materially higher rates of failed puncture, wire misdirection, and tract complications, because there is less margin for the needle, dilator, and stent flanges to seat correctly within the ductal lumen. Altered anatomy adds difficulty chiefly by removing the duodenal option, forcing a hepaticogastrostomy route and precluding the anatomy-preserving rendezvous technique, and by sometimes distorting the expected position of the left intrahepatic ducts relative to the gastric wall.

Multiple comparative series and meta-analyses report that, relative to PTBD, EUS-BD achieves similar overall efficacy but with significantly fewer unplanned re-interventions and better quality-of-life scores — driven almost entirely by the absence of an external drainage catheter. This durability-without-external-tube profile is the central clinical argument that has moved EUS-BD from a rescue technique to a first-line alternative at high-volume centers.