ERCP CBD Stone Extraction: Balloon vs Basket vs Lithotripsy (2026)

After an adequate sphincterotomy, start ordinary CBD stones (about 10 mm or smaller) with a balloon. Difficult or large stones get limited sphincterotomy plus endoscopic papillary large-balloon dilation sized to the distal duct. If that fails, crush, then cholangioscopy. Unstable cholangitis: drain first and come back.

Experienced teaching points

Clinical Pearls

After an adequate sphincterotomy, start with a balloon for stones about 10 mm or smaller. Baskets miss small stones.
Limited sphincterotomy plus EPLBD is first-line for difficult or large stones. Size the balloon to the distal duct.
Do not trap a large stone in a non-crushing basket. Impacted-basket rescue is an out-of-scope lithotriptor.
If EPLBD fails, use mechanical lithotripsy, then cholangioscopy-guided EHL or laser.
Unstable cholangitis: drain first. Leave a plastic stent and come back rather than persisting.

Clinical Bottom Line

Situation	2026 practical answer
Ordinary stone after an adequate sphincterotomy (about <=10 mm)	Start with a balloon. Ishiwatari: single-catheter clearance 92% balloon vs 80% basket. Small stones slip through basket wires. Switch devices if needed. Finish with an occlusion cholangiogram.
Difficult or large stone (often >15 mm, barrel-shaped, or a tapered distal duct)	Limited sphincterotomy plus endoscopic papillary large-balloon dilation, sized to the distal CBD (usually 12 to 20 mm). Karsenti: index clearance 96% vs 74%, mechanical lithotripsy 4% vs 36%.
EPLBD failed or unsafe	Through-the-scope mechanical lithotripsy. Do not grab a large stone with a non-crushing basket. Impacted-basket rescue is an out-of-scope metal sheath, not harder pulling.
Impacted, intrahepatic, or lithotripsy-refractory	Cholangioscopy-guided electrohydraulic or laser lithotripsy. Buxbaum: one-session clearance 93% vs 67% conventional. Bang: after failed balloon/basket, 94% vs 73% large-balloon sphincteroplasty.
Sick, coagulopathic, incomplete clearance, or the clock is gone	Drain with a plastic stent and come back. Tokyo: severe cholangitis gets drainage first, not heroic same-session clearance.
Small-balloon EPBD without sphincterotomy	Niche: coagulopathy or altered anatomy, small stones. Not Western default. Disario: PEP 15.4% vs 0.8% after EST, trial stopped early, 2 deaths.

What do I do tonight?

If the papilla is intact, run the PEP bundle in parallel: rectal NSAID, wire-guided cannulation, pancreatic stent if the PD is repeatedly accessed. That protocol lives on the PEP prevention article.
If the stone is about 10 mm or smaller after an adequate sphincterotomy: balloon first, basket in the room, occlusion cholangiogram at the end.
If the stone is large, barrel-shaped, or the distal duct is tapered: limited sphincterotomy plus EPLBD sized to the distal CBD (usually 12 to 20 mm). Stop if a waist will not yield.
If EPLBD fails or is unsafe: TTS mechanical lithotripsy with a crush-compatible basket.
If the stone is impacted, intrahepatic, or lithotripsy-refractory: cholangioscopy-guided EHL or laser.
If the patient is septic, coagulopathic, or the clock is gone: plastic stent, stop, come back.

What do the societies say?

Is this a size chart or an extraction algorithm?

The old mapping (balloon under 10 mm, basket 10 to 15 mm, lithotriptor above 15 mm) is too rigid. Tool choice follows the orifice, the distal duct, and whether the patient can tolerate a long first session. ESGE 2019 still offers extraction to every fit patient with a CBD stone [1]. Leaving stones in situ on GallRiks was associated with later pancreatitis, cholangitis, or obstruction in 25% versus 13% after removal.

Run the PEP bundle in parallel on an intact papilla. This page is how you get the stone out once you are in.

Balloon or basket after sphincterotomy?

After sphincterotomy, both devices are acceptable. Two RCTs disagree on the endpoint, not on the anatomy:

Ishiwatari (stones <=10 mm, CBD <=15 mm) [6]: complete clearance with the assigned catheter 92.3% balloon vs 80.0% basket. One-session clearance after allowing the other device was 97% in both arms.
Ozawa BasketBall (stones <11 mm) [7]: complete removal within 10 minutes 81.3% basket vs 83.9% balloon, not different.

Read that as: start with a balloon, especially for small stones that baskets cannot grasp. Have a basket in the room. Saito's 904-patient observational series found balloons finished more <=10 mm cases without a device switch, with more bleeding (3.3% vs 0.7%). That is a caution, not an RCT.

When is limited sphincterotomy plus EPLBD first-line?

Limited EST plus EPLBD is ESGE first-line for difficult stones (strong, high-quality) [1]. It is not the same operation as 8 mm EPBD without a cut.

Karsenti (stones >=13 mm, n=150) [10]: complete EST plus large balloon vs complete EST conventional. Index clearance 96.1% vs 74.0%. Mechanical lithotripsy 3.9% vs 35.6%. Rescue balloon dilation cleared 15 of 19 conventional failures. Teoh [9]: limited EST plus balloon cut lithotripsy from 46% to 29%, and from 91% to 58% when the stone was >=15 mm. Stefanidis [11]: EST plus large balloon vs EST plus mechanical lithotripsy for 12 to 20 mm stones, fewer complications (4.4% vs 20%), driven by cholangitis after lithotripsy.

Heo 2007 is the seminal RCT but mixed stone sizes, so lithotripsy rates look similar. Do not quote Heo 2007 as proof that EPLBD fails to reduce lithotripsy. Facciorusso's network meta still ranks EST plus large balloon over EST alone for index clearance and less lithotripsy.

MARVELOUS (Japan) showed EPLBD without EST can clear large stones in expert centers. ESGE still wants limited EST plus EPLBD as the Western first-line. Do not skip the cut because an East Asian trial made it look optional.

How do I size the balloon without perforating?

Park 2013 (n=946) [12]: adverse events 10%, including 9 perforations and 4 deaths. Distal CBD stricture: perforation OR 17. Full-length EST predicted bleeding. Fatal cases kept inflating despite a persistent fluoroscopic waist. Stop if the waist will not yield. Do not dilate through a distal CBD stricture. Do not choose a balloon larger than the distal duct.

When is mechanical lithotripsy the next step?

ESGE: use TTS mechanical lithotripsy when EPLBD fails or is inappropriate [1]. Clearance across series is about 76% to 91%. First-session clearance is often only 50% to 73%. Failure tracks with impaction, stones above 30 mm, and a stone-to-CBD ratio above 1. Out-of-scope metal-sheath lithotripsy is for the impacted Dormia basket. If you think you might need to crush, start with a lithotripsy-compatible basket. Do not grab a large stone with a non-crushing basket.

When do I switch to cholangioscopy?

Buxbaum RCT (n=60, stones >10 mm) [13]: one-session clearance 93% SOC-laser vs 67% conventional (balloon, basket, ML, papillary dilation).
Bang RCT (n=66 after failed balloon/basket) [14]: SOC-laser 93.9% vs large-balloon sphincteroplasty 72.7%. Success associated with a stone-to-duct ratio <=1 and a non-tapered distal duct.
Angsuwatcharakon (n=32 after failed EST and/or EPLBD) [15]: first-session laser 100% vs ML 63%. Small, tertiary.
McCarty meta [16]: fragmentation about 91%, complete single-session about 77%, mean 1.32 sessions. Heterogeneity is real.

ESGE: pick EHL or laser by what you have. Antibiotics are reasonable for cholangioscopy. Do not use cholangioscopy as a free upgrade for every 12 mm stone that would clear with a properly sized EPLBD.

When should I stent and return?

If the patient is unstable, coagulopathic, or the stone will not come, leave a plastic stent and stop. Tokyo 2018: severe cholangitis needs drainage after resuscitation, often within hours if there is shock [4] [5]. Moderate disease: drain within 48 to 72 hours. Difficult, large, or multiple stones in cholangitis: two sessions after drainage. Same-session clearance is for mild or moderate disease without coagulopathy, not for septic shock. Do not persist to heroic same-session clearance in septic cholangitis.

Stents shrink some stones, but on-demand exchange is a cholangitis plan. Di Giorgio: cholangitis 36% on-demand vs 8% with 3-month exchanges. If the stent stays, schedule exchange.

Cholecystectomy within 2 weeks after successful ERCP for choledocholithiasis is still the ESGE line when the gallbladder is in [1].

What about Billroth II and Roux-en-Y?

Billroth II and Roux-en-Y belong in centers that do this weekly. ESGE: EPBD is often safer than a reverse sphincterotomy. This page will not pretend to teach enteroscopy-ERCP.

Do not use routine small-balloon EPBD without EST as the Western default. Disario stopped early: PEP 15.4% vs 0.8% after EST, 2 deaths [8]. Niche only: coagulopathy, altered anatomy, small stones.

Do not cite EPISOD for choledocholithiasis. That trial is SOD type III.

Pitfalls

Pitfall	Better move
Routine small-balloon EPBD without EST in mixed-risk Western practice	Disario stopped early. PEP 15.4% vs 0.8%. Niche only.
EPLBD balloon larger than the distal duct, or dilation through a stricture	Park 2013: distal stricture OR 17 for perforation. Stop if the waist persists.
Full-length EST immediately before EPLBD	Limited EST plus balloon. Full EST predicted bleeding.
Non-crushing basket on a large stone	Impacted basket. Lithotripsy-compatible basket or EPLBD first.
Leaving a plastic stent until the patient is ill	Three-month exchanges if clearance is deferred.
Heroic same-session clearance in septic cholangitis	Drain first. Tokyo two-session plan for difficult stones.
Citing EPISOD	That trial is SOD type III, not choledocholithiasis.

Selected references

Questions doctors ask

Should I start with a balloon or a basket?

Start with a balloon after an adequate sphincterotomy, especially for stones about 10 mm or smaller. Ishiwatari: single-catheter clearance 92% balloon vs 80% basket. Keep a basket in the room and switch if needed.

When do I use large-balloon dilation?

Limited sphincterotomy plus EPLBD is first-line for difficult or large stones, sized to the distal CBD (usually 12 to 20 mm). Karsenti: index clearance 96% vs 74%, with much less mechanical lithotripsy.

Is small-balloon EPBD without sphincterotomy safer in Western practice?

No. Disario stopped early: PEP 15.4% vs 0.8% after EST, 2 deaths. Niche only for coagulopathy or altered anatomy with small stones. It is not the Western default.

When should I leave a stent and come back?

When the patient is septic, coagulopathic, or the stone will not come. Tokyo 2018: severe cholangitis gets drainage first. Difficult stones in cholangitis are a two-session plan.

Does cholangioscopy replace EPLBD for every large stone?

No. Use cholangioscopy-guided EHL or laser when the stone is impacted, intrahepatic, or lithotripsy-refractory. It is not a free upgrade for every 12 mm stone that would clear with a properly sized EPLBD.

Topics

For your teaching file

Save this article as a PDF

Drop your email and we'll open a print-ready version you can save as a PDF — and you'll start getting our weekly GI endoscopy newsletter.

← All articles