Assessing adrenal lesions

When we do an adrenal protocol CT, the aim is to first do a non-contrast volume to identify the lesion(s) and measure their density, and use this information to decide whether contrast is required. This is a brief teaching resource to show you how to identify the adrenal glands on CT, how to find adrenal nodules, what to measure, why we do the measurement, and how to measure it. Once you have viewed this page, you have the option of doing the test cases.

Please note that the cases are best viewed using Chrome or Safari, not Internet Explorer, and the measuring of the density of the adrenal nodules works best on a desktop or laptop.

Normal adrenal glands

The adrenal glands are paired endocrine organs that lie just superior to the kidneys. They have a V or Y shape, and the limbs are called the anterior, medial and lateral limbs. The normal thickness of a limb is ~2-3 mm. To find the adrenals, scroll up from the top of the kidneys, and they are located just lateral to the crura of the diaphragm which lie on either side of the aorta. It is good practice to also look at them on the coronal images.

You can scroll through the images in the Radiopaedia case below to see what the normal adrenal glands look like.

Adrenal nodules

Nodules of the adrenal glands are commonly found on CT scans. Estimates vary from 3% to 7%, but our own data showed that we found them in 2% of CTs (i.e. a lot!).

There are two main things to consider with these lesions. The first is whether they are functional, meaning that they are secreting hormones. This is identified by blood and urine tests performed by the endocrinology department.

The second question is whether they are benign ("adenoma"), or small malignant lesions, and this is where we come in.

If the mean density of the lesion is less than 10 Hounsfield Units (HU), it means it has lipid content and is almost certainly an adenoma, and is known as a "lipid rich adenoma". For these, no further imaging is required. The problem with most abdominal CT is that we do them with intravenous contrast, and then we can't accurately assess for lipid content. So when a nodule is identified on a contrast CT, it is often labelled an "indeterminate", until we do an adrenal protocol CT, starting with a non-contrast volume.

On the non-contrast volume, if the mean density of the lesion is greater than 10 HU, we proceed to giving intravenous contrast and then measure the density and using these measurements to calculate the washout. This can provide evidence of a lesion being an adenoma, despite not being lipid-rich. You can read more about that here if you wish.

Finally, there will be some patients coming for an adrenal protocol CT not because they had a lesion incidentally detected on a CT for another reason, but because the referring doctors are suspicious they may have a lesion because of hormone issues. In those patients, you may not see any adrenal abnormality on the non-contrast scan, so there is no need to then give contrast.

How and what to measure

When you measure the density of the lesions, you will be doing so on the CT workstation, using the region of interest (ROI) tool. For the purposes of this teaching and assessment, I am using Pacsbin. This video shows you how to navigate Pacsbin, find the ROI tool, and I talk through two separate cases and show how I measure the mean density.


  1. Look at the report of the previous CT; it will tell you how many adrenal nodules there are, which side, and how large

  2. Do a non-contrast phase and find the adrenal glands just off the tops of the kidneys

  3. Find the nodule (s) and draw a ROI that is just inside the edge of the lesion - having a decent sized ROI is important for accuracy

  4. When all the numbers pop up, the one that you are looking for is the MEAN - ignore all of the others

  5. Do three separate ROIs and do a rough calculation of the mean of these

  6. It may be easier to do the measuring on the coronal images

  7. If the mean is less than 10 HU, you can stop the scan - the lesion is a lipid rich adenoma

  8. If the mean is more than 10 HU, proceed to giving contrast as per the protocol

  9. If it is a screening scan (i.e. not already known to have an adrenal lesion, but clinicians are looking for one as has hormone excess) then there may not be a lesion to see and in this case you can stop the scan

Try the demo cases yourself

Now you can have a look at these same two cases on Pacsbin and try to measure the ROI yourself. The measurement tools only work when you do this on a desktop or laptop. Once you are happy with this, try the test cases.