Balloon tamponade of GI bleeding

Case

You have been resuscitating the 62 year old man with a massive GI bleed from the previous post. You started your massive transfusion protocol and have him safely intubated. However, he continues to bleed, and you need to transfer him out of your small community hospital…

My approach

Although relatively rare, balloon tamponade is required to control upper GI bleeding when there is going to be a delay to definitive therapy, whether by endoscopy, surgery, or interventional radiology. There are a couple of different devices available, and each hospital is likely to stock only one. You don’t want to have to figure out the nuances of your specific device with a critically ill patient in front of you. This post will cover a general approach that is applicable to all balloon tamponade devices, but I would suggest taking out the device available in your department, reading the attached instructions, and playing with it, so you are sure how your specific device works.

What you need:

• The balloon
• Lube
• 60ml Syringe
• NG tube
• Kelly clamps
• 3 way stopcock (2 or 3)
• Working wall suction, and ‘christmas tree’ adaptor
• Manometer
• Traction (1 L bag of saline and a roll of cling dressing will suffice)

The procedure:

1. The patient really must be intubated first. The pain of the procedure alone is enough that the patient should be intubated and adequately sedated. Additionally, these patients are sick, may need to be transferred, are at high risk of aspiration, and airway management after the device is placed will be significantly more difficult.
2. Position the patient with the head of the bed up at 45 degrees.
3. Check the balloon (or both balloons if your tube has two) for leaks. Sometimes this seems like a waste of time with a critical patient in front of you, but you will waste even more time if you insert a faulty balloon.
4. Prepare an OG tube so that you will be able to suction from above the gastric balloon after it is inflated. (This will tell you if there is ongoing esophageal bleeding.) Some balloons have an esophageal suction port, so you can skip this step. Place the OG and the Blakemore tubes next to each other, so that the tip of the OG lies just proximal to the gastric balloon. Mark both tubes so you can realign them later. (This mark should be at or beyond the 50cm measurement on the balloon, as the rest will be inside the patient’s body).
5. Lubricate the balloon and tube.
6. Insert the balloon via that mouth just like you would an orogastric tube. Using a laryngoscope for visualization and McGill forceps to guide the tube might be helpful. Insert to 50cm. You can verify its location clinically like you would an NG tube (injecting air and listening over the stomach and lungs).
7. Partially inflate the balloon with 50ml of air. Confirm the balloon is below the diaphragm on chest X-ray. This is probably also possible using ultrasound, which would be a lot quicker (see this post by Josh Farkas on PulmCrit). It is very important not to inflate the balloon in the esophagus. It is a large balloon and will decimate the esophagus. This is almost always fatal.
8. After proper positioning below the diaphgram has been confirmed, fully inflate the gastric balloon. (It will take a lot of air. Make sure you know your own device. The Blakemore tube takes 250ml, the Minnesota 500ml, and the Linton 700-800ml.)
9. Once fully inflated, clamp the inflation port. (Pearl: if using bare metal clamps, wrap the tips in tape, cling, or gauze to prevent them from cutting into the rubber.)
10. Pull the balloon back against the fundus. Depending on how long it will take to get the patient to definitive therapy, it might be best to just maintain the pressure manually. Otherwise, note the measurement at the teeth, and maintain the traction by tying a bag of saline to the Blakemore tube (be creative, you can use any kind of knot you want), and hang the MacGyvered contraption over an IV. (Or I guess you could use a commercially available traction device if you have one.)
11. Attach the gastric port to suction.
12. Insert the OG tube to the distance previously marked and suction. If there is significant ongoing bleeding, this localizes the bleeding to the esophagus, and you will need to inflate the esophageal balloon.
13. Due to a high risk of perforation and the variable volume required, the esophageal balloon should be filled using a manometer. Attach the manometer to the esophageal balloon port using a 3 way stopcock and inflate to a target pressure of 25mmHg, then clamp the inflation port.
14. Don’t forget to continue with your general GI bleed resuscitation, massive transfusion strategy, and get the patient to definitive therapy (endoscopy, interventional radiology, or surgery).

The final set-up

Possible contraindications:

This is a temporizing procedure for dying patients, so in my mind the only real contraindications would be use in a stable patient or a non-intubated patient. It also shouldn’t be used as a definite therapy. There are a number of listed contraindications, but as a last resort, your hands may be tied. Possible contraindications to be aware of include:

• Esophageal stricture
• Recent surgery involving the gastroesophageal junction
• Hiatus hernia

Other FOAMed Resources

There are some excellent videos that will demonstrate this a lot better than words. (Sorry, I guess I should have put these right at the beginning of the post)

Minnesota Tube placement from SMACC and the Hennepin County Medical Center:

Minnesota Tube from Social Media and Critical Care on Vimeo.

This series from EM:RAP HD and Jess Mason includes an overview, and specific instructions for the Blakemore, Minnesota, and Linton tubes:

Other resources

Blakemore Tube Placement for Massive Upper GI Hemorrhage on EMCrit

PulmCrit Wee: Ultrasound-guided blakemore tube placement

References

Attar BM. Chapter 63. Balloon tamponade of Gastrointestinal bleeding. In: Reichman EF (ed). Emergency Medicine Procedures, 2e. Toronto: McGraw-Hill; 2013.

Winters ME and Panacek EA. Chapter 41. Balloon tamponade of gastroesophageal varicies. In: Roberts JR, ed. Roberts and Hedges’ clinical procedures in emergency medicine, 6e. Philadelphia,PA: Elsevier; 2014.