CT gastrointenstinal bleed (protocol)

A CT gastro-intestinal bleed protocol utilizes a multiphasic technique to detect active gastrointestinal bleeding (as well as other potential non-bleeding bowel disease ¹. 

Note: This article is a general guideline for evaluating CT gastrointestinal bleeds. Protocol factors are variable as they are dependent on the CT scanner, software, patient demographics and the radiologist.

A typical CT assessment for a gastrointestinal bleed would look as follows:

Indications

Typical indications for a suspected bleed include ^1,2,3:

• rectal bleeding

• hematemesis/hemoptysis

• melena (black, tarry stools)

• abdominal pain

• trauma

• vomiting darker blood

• chest pain

• weight loss

Clinical assessment in conjunction with the patient's history and risk factors can facilitate locating the GI bleed (i.e. upper or lower GI) ¹.

Purpose

The primary purpose of this examination is to evaluate for active gastrointestinal bleeding, however other differentials could also be ruled out through the multiphase technique. The multiphase protocol best demonstrates abdominal structures in three phases which can be used to evaluate an overt or occult bleed and the source of the bleed ^1,2. Whilst the protocols are site dependent, protocols are dependent on the type of bleed and patient presentation. The aim of the examination is to maximize sensitivity and specificity whilst using the lowest radiation dose possible.

• overt bleeding or trauma presentation

  • non-contrast

  • arterial phase

    This phase best demonstrates vascular anatomy, vascular abnormalities such as arteriovenous malformations (AVMs) and masses such as carcinoid tumors. Luminal extravasation in best visualized in this phase which can identify the source of a bleed ²

  • venous phase

•  occult bleed or suspected small bleed ¹

  • non-contrast

    This phase highlights any hyperdense material that pre-exists within the bowel lumen and helps distinguish any blushing indicating an active hemorrhage ³

  • late arterial phase

  • portal venous phase

    This phase visualizes the bowel wall due to the structure's perfusion rate and intraluminal extravasations as it has a greater sensitivity when the images are compared with a non-contrast or arterial phase ¹

• single phase protocols - used independently ¹

  • enteric phase

    This phase best demonstrates the bowel wall as the scan is performed at the optimal time for contrast perfusion through the bowel walls which is around fifty seconds post injection ¹.

    • this protocol would be similar to a CT enterography protocol

  • portal venous phase

  • split bolus

Technique

• patient position

  • supine with their arms above their head

• scout

  • diaphragm to iliofemoral arteries, inclusive of renal arteries and the bifurcation point 

• scan extent

  • diaphragm to iliofemoral arteries, inclusive of renal arteries and the bifurcation point

• scan direction

  • craniocaudal

• contrast injection considerations

• monitoring slice (region of interest) - varies depending on scanner capabilities and department protocols

  • descending aorta at the level of the carina or renal arteries

• threshold

  • dependent on scanner but usually 150 HU ³

• injection 

  • 75 mL of non-ionic iodinated contrast with a 40 -50 mL saline chaser at 3.5 - 5mL/s

• multiphasic approach

  • non-contrast (optional)

  • arterial

    • contrast volume: 100 - 125ml at 3.5 - 4 mL/s with a saline chaser of 50ml ³. Approximately 1ml of contrast per kg

    • bolus track: descending aorta

    • scan delay: 15 - 30 seconds from the time of injection

  • portal venous

    • contrast volume: 70 -90ml at 2 – 4ml/s with a saline chaser of 50ml. Approximately 1ml of contrast per kg.

    • bolus track: descending aorta

    • scan delay: 60 - 75 seconds from the time of the injection

• scan delay

  • dependent on the phase used

• ​respiration phase

  • single breath-hold: inspiration

• multiplanar reconstructions

  • axial images: axial to the body axis

  • coronal images: coronal to the body axis

  • sagittal images: sagittal to the body axis

  • slice thickness: soft tissue ≤3 mm

• maximum intensity projection

  • axial images: axial to the body axis

  • coronal images: coronal to the body axis

  • sagittal images: sagittal to the body axis

  • slice thickness: soft tissue ≤3 mm

  • 3D reconstruction using soft tissue thins

Practical Points

• dose optimization

  • correlating patient presentation and clinical history is essential in determining the potential location of a bleed and using the appropriate protocol ³

  • correct patient positioning will facilitate dose modulation and multiplanar reconstructions

  • dual energy CT technique can be useful as non-contrast reconstructions can be virtually developed reducing the radiation dose delivered to the patient ¹

• depending on the patient's age, presentation and radiologist reporting, single phase protocols are often utilized when there is no active rectal bleeding