What is the approach to duodenal embolization for hemostasis in cases of severe duodenal bleeding where the bleeding site cannot be detected?

Duodenal Embolization for Hemostasis Without Detecting the Bleeding Site

Prophylactic embolization of the gastroduodenal artery territory should be performed when severe duodenal bleeding is documented by endoscopy or CTA but no active extravasation is visible on angiography, with clinical success rates of 72-83% and comparable outcomes to cases where bleeding sites are identified. 1

When to Proceed with Empiric Embolization

Proceed with prophylactic embolization when:

• Endoscopy has documented severe duodenal ulcer bleeding but failed to achieve hemostasis 1
• High-risk features are present: hemodynamic instability, ulcer >2cm, or ongoing transfusion requirements 1
• Bleeding is documented by endoscopy or CTA even without visible extravasation on angiography 1

The evidence strongly supports this approach, with empiric TAE achieving 83% clinical success and 100% technical success, with no significant difference in outcomes compared to cases where bleeding sites are identified 2. A separate study demonstrated 90% success in stopping bleeding when the long stretch of the GDA including pancreaticoduodenal branches was embolized 3.

Technical Execution Protocol

Perform selective catheterization of both celiac axis and superior mesenteric artery with high-volume contrast and prolonged imaging to address the rich collateral blood supply. 1

Embolization technique:

• Target the entire gastroduodenal artery territory including anterior and posterior superior pancreaticoduodenal arteries 1, 3
• Use microcoils as the preferred embolic agent (though glue may reduce rebleeding rates) 1
• Perform superselective rather than proximal embolization to minimize ischemia risk 1
• Embolize the long stretch of the GDA around the suspected bleeding site, not just one side 3

The technical approach is critical—one study reported failure when only the common hepatic artery side of the GDA bleeding site was embolized, but 90% success when the long stretch including ASPD and PSPD was embolized 3.

Safety Profile and Complications

The risk of bowel ischemia is 1-10% and is minimized by using microcoils rather than particles, avoiding glue when possible in duodenal territory, and performing superselective embolization. 1

• Overall post-procedural complications are significantly lower than surgery 1
• No severe complications were reported in the empiric TAE group, with comparable complication rates to identifiable TAE 2
• Method-related complications are not significantly different between TAE and surgery, though TAE has a higher rebleeding rate 4

Comparative Outcomes

TAE appears safer than surgery, particularly in high-risk patients:

• 30-day mortality: 19-20% for TAE versus 20-22% for surgery 5, 4
• In high-risk patients (APACHE II score ≥16.5), mortality was 23.1% with TAE versus 50.0% with surgery 4
• TAE reduces transfusion requirements from median 14 units before to 2 units after the procedure 6

Optimizing Angiographic Targeting

Use endoscopic findings to guide embolization:

• Endoscopic clip placement at the ulcer site helps direct angiographic targeting 1
• Location of ulcer on endoscopy determines which pancreaticoduodenal branches to prioritize 1
• High-risk stigmata on endoscopy support proceeding with the empiric approach 1

Critical Caveat

Hemodynamically unstable patients despite resuscitation should proceed directly to surgery rather than TAE. 1 This represents the key exception where empiric embolization should not be attempted, as these patients require immediate surgical control.