Adrenal Adenoma vs Met (washout)

You scan a belly for a kidney stone and there it is — a little lump sitting on top of the kidney like a hat. An adrenal nodule. Most of the time it means absolutely nothing, but now and then it's the one finding that changes a cancer patient's whole staging. The job is telling those two apart without poking the patient with a needle. Good news: the adrenal gland practically hands you the answer if you know which questions to ask.

The cast of characters

The adrenal adenoma is the overwhelming favorite. These show up on routine abdominal CTs as an incidentaloma — a thing you weren't looking for that decided to come to the party anyway (see incidentaloma frameworks). Most are non-functioning, meaning they sit there quietly and don't pump out extra hormones. The plot twist that makes them easy to catch: a lot of them are packed with intracellular fat.

The villain we're ruling out is the metastasis. Adrenals are a popular landing spot for spread from lung, breast, melanoma, kidney, and others — partly because they have a rich blood supply, like a busy highway off-ramp. A met has no reason to be full of fat, and it tends to be greedy with contrast.

Findings by modality

CT, unenhanced

Here's the elegant part. Fat is less dense than water and soft tissue, so it reads low on CT. If an adrenal nodule is crammed with intracellular fat, its average Hounsfield unit value — the radiologist's density scale, with water set at zero — gets pulled down.

The rule: on an unenhanced CT, a homogeneous mass measuring ≤10 HU is a lipid-rich adenoma. The ACR white paper treats that as diagnostic regardless of size, and you can essentially call it benign and walk away. One measurement, one circle, done. It feels almost too easy, and most of the time it is. Two housekeeping rules make the number trustworthy: draw the region of interest over the bulk of the lesion (not the edge, not a necrotic pocket), and measure on a proper density-friendly acquisition — the ACR suggests 120 kVp and 3 mm collimation with axial and coronal reconstructions.

The unenhanced series also catches the other "done at a glance" lesions: a mass with macroscopic fat is a myelolipoma, and a mass that simply doesn't enhance — a change of less than 10 HU between the pre- and post-contrast images — is a cyst or a hematoma. All three need no further workup.

Figure · CT

CT, adrenal-protocol (washout)

Some adenomas are lipid-poor — they skipped the fat and measure above 10 HU, sitting right in the gray zone with the mets. Annoying. This is where the dedicated adrenal-protocol CT earns its keep, and it's the ACR's examination of choice for an indeterminate mass 1 to <4 cm that measures >10 HU.

The protocol has three acts. An unenhanced acquisition is reviewed in real time (at reduced dose, with limited z-axis coverage — you only need the adrenals); if the mass already shows a benign feature, the patient goes home. If not, contrast is power-injected and a dynamic phase at 60–90 seconds is acquired, followed by a 15-minute delayed acquisition.

The concept is about timing, not just brightness. Both adenomas and mets light up on the dynamic phase. The difference is what happens next: an adenoma releases contrast quickly — it washes out — while a metastasis clings to it and stays bright. Think of a paper towel versus a sponge that's already saturated: pour water on both, and the paper towel sheds it fast while the sponge just sits there heavy and wet.

MRI, chemical shift

MRI has its own fat detector, and it's a clever one. Chemical-shift imaging acquires in-phase and opposed-phase T1 images; when fat and water share a voxel, their signals cancel on the opposed-phase image. A lipid-rich adenoma therefore loses signal on opposed-phase compared with in-phase images, judged against the spleen (which has no fat to lose and makes a fair reference — the liver is a poor one because it may be fatty itself). The ACR considers that drop almost always diagnostic of a lipid-rich adenoma regardless of size, which makes MRI a good way to characterize an adenoma when CT is indeterminate. A metastasis, having no intracellular fat, stays the same on both.

Figure · MRI

PET/CT

In a patient with a known cancer, FDG PET/CT is the problem-solver when CT and MRI still can't decide, and the ACR white paper specifically floats biopsy or PET/CT when a mass shows central necrosis, which raises the likelihood of metastasis. Adenomas are typically quiet on FDG; mets are typically not. There is no robust pooled accuracy figure for adrenal PET to quote here, so treat it as a strong adjunct rather than a number.

The numbers

Everything on this page distills to a handful of values. Here they are in one place.

Let me show you the arithmetic once so the formulas stop looking scary. Say a mass measures 30 HU unenhanced, 90 HU on the dynamic phase and 45 HU at 15 minutes (numbers I've made up for the exercise). Absolute washout is (90 − 45) ÷ (90 − 30) × 100 = 75%, comfortably above 60%: adenoma. If you only had the two post-contrast scans, relative washout would be (90 − 45) ÷ 90 × 100 = 50%, above 40%: same answer.

How good is the test

The 10 HU cutoff is a deliberate compromise, and the literature analysis behind it shows why. Pooling ten CT studies covering 495 adrenal lesions (272 benign, 223 malignant), the sensitivity for calling a lesion benign ran from 47% at a 2 HU threshold up to 88% at 20 HU, while specificity fell from 100% at 2 HU to 84% at 20 HU. Slide the threshold up and you catch more adenomas but start letting mets through; slide it down and you never mislabel a met but miss a lot of adenomas. 10 HU is the conventional middle — very high specificity, at the cost of leaving the lipid-poor adenomas undecided.

That's exactly the gap washout closes. In the study that established the combined approach, using unenhanced attenuation ≤10 HU or washout ≥60% correctly categorized nearly all of 166 adrenal masses (127 adenomas and 39 non-adenomas). Combined unenhanced-plus-delayed CT is, in other words, the standard because it works.

For chemical-shift MRI, the concept is well established (signal loss on opposed-phase images is nearly always diagnostic of lipid-rich adenoma) but there is no robust numeric sensitivity or specificity to quote, so I won't invent one. For PET/CT, likewise: no robust pooled figure to quote.

The ACR pathway at a glance

There's no numbered grading system for adrenal masses the way Bosniak grades cysts, but the ACR Incidental Findings Committee's 2017 white paper gives a pathway, and it's worth carrying in your head in abbreviated form.

Table abbreviated from the ACR white paper (Mayo-Smith et al., 2017); the full algorithm, including the pathway for patients with a known cancer, lives in the paper itself.

The differential

Reporting

The referrer wants a verdict, not a tour. Give them:

• Size (largest dimension, and note whether it is 1 to <4 cm or ≥4 cm, because that decides the pathway) and side.
• Unenhanced attenuation in HU, and whether the mass is homogeneous. If it is ≤10 HU, say "consistent with lipid-rich adenoma; no further imaging recommended."
• If a washout study was done, the three HU values and the calculated APW or RPW with the threshold it was measured against.
• Any diagnostic benign feature (macroscopic fat, no enhancement) named explicitly.
• Comparison with prior imaging and the stability interval.
• The recommendation: none, adrenal-protocol CT or chemical-shift MRI, biochemical evaluation, or specialist referral for a ≥4 cm mass — and in a cancer patient, that the lesion is or is not compatible with metastasis.

The bottom line

Most adrenal nodules are sleepy adenomas. Reach for unenhanced density first — ≤10 HU and you're done. If it's lipid-poor and stuck in the gray zone, let time break the tie with a washout study: ≥60% absolute or ≥40% relative means adenoma, and anything that hangs onto its contrast stays on the suspect list. Wrap every measurement in the patient's story, and that little hat on the kidney usually turns out to be exactly as boring as it looks.

References

• Mayo-Smith WW, Song JH, Boland GL, et al. Management of Incidental Adrenal Masses: A White Paper of the ACR Incidental Findings Committee. J Am Coll Radiol 2017;14(8):1038–1044. Used for: the ≤10 HU unenhanced threshold, the ≥60% absolute and ≥40% relative washout cutoffs and their formulas, the <10 HU non-enhancing definition, macroscopic fat (myelolipoma) and the benign calcified mass as diagnostic benign features, the adrenal-protocol timing (unenhanced, dynamic at 60–90 s, 15-min delay), the 120 kVp and 3 mm technique notes, the 1 to <4 cm and ≥4 cm size rules, the "no growth-rate cutoff" statement, the central-necrosis and PET/CT guidance, and the chemical-shift MRI criterion in "Findings by modality," "The numbers," "The ACR pathway at a glance," "The differential," and "Reporting."
• Caoili EM, Korobkin M, Francis IR, et al. Adrenal masses: characterization with combined unenhanced and delayed enhanced CT. Radiology 2002;222(3):629–633. Used for: the ≤10 HU and ≥60% washout criteria and the 166-mass (127 adenomas, 39 non-adenomas) result in "How good is the test."
• Boland GW, Lee MJ, Gazelle GS, Halpern EF, McNicholas MM, Mueller PR. Characterization of adrenal masses using unenhanced CT: an analysis of the CT literature. AJR Am J Roentgenol 1998;171(1):201–204. Used for: the sensitivity (47% at 2 HU to 88% at 20 HU) and specificity (100% at 2 HU to 84% at 20 HU) ranges and the 495-lesion (272 benign, 223 malignant) pooled analysis in "How good is the test," and the 10 HU compromise.
• Israel GM, Korobkin M, Wang C, Hecht EN, Krinsky GA. Comparison of unenhanced CT and chemical shift MRI in evaluating lipid-rich adrenal adenomas. AJR Am J Roentgenol 2004;183(1):215–219. Used for: the chemical-shift (in-phase versus opposed-phase signal loss) concept in "Findings by modality"; no numeric index is taken from it.