Adrenal incidentaloma

• An adrenal incidentaloma is an adrenal mass discovered by chance on imaging done for a reason that is unrelated to the adrenal glands; the patient usually feels well.
• Most are benign, nonfunctioning adrenocortical adenomas, but a small and important minority are cancers, pheochromocytomas, hormone-producing adenomas, or metastases. ^[3]
• The work-up asks three questions in parallel: is the mass benign or malignant on dedicated adrenal imaging, is it secreting cortisol, catecholamines, or aldosterone, and does size or behaviour justify surgery.
• Imaging starts with an unenhanced CT; a homogeneous mass measuring 10 Hounsfield units or less is benign, and the 2023 European guideline removes any size limit from that statement. ^[2][3]
• Everyone gets a 1 mg overnight dexamethasone suppression test; plasma or urine metanephrines unless the CT already shows no more than 10 Hounsfield units; and an aldosterone-to-renin ratio when there is hypertension or low potassium.
• Surgery is an individualised, multidisciplinary decision, tilted toward an expert high-volume adrenal surgeon when cancer is suspected, the mass exceeds 4 cm, hormone excess is significant, or the lesion grows on follow-up.

By the end of this page you should be able to:

1. Define an adrenal incidentaloma and explain why the 1 cm threshold is used.
2. Use unenhanced CT, washout studies, and MRI to separate benign adenomas from indeterminate or suspicious masses.
3. Plan a complete hormonal screen and interpret the 1 mg dexamethasone suppression test, including mild autonomous cortisol secretion (MACS). ^[2]
4. State when adrenalectomy is recommended and which operative approach fits.
5. Explain how non-operated lesions are followed, and where guidelines genuinely disagree.
6. Outline the special approaches needed for bilateral disease, very young patients, pregnancy, and frail older adults.

The strict definition used by the European Society of Endocrinology (ESE) and the European Network for the Study of Adrenal Tumors (ENSAT) excludes masses found during screening of patients with hereditary syndromes and masses found during staging of an extra-adrenal cancer. ^[1] The word itself was coined in 1982. ^[1] A mass picked up this way is silent by definition, so "incidentaloma" describes how it was found, not what it is; only careful characterisation tells you whether it is a harmless adenoma or something that needs treatment.

Adrenal cancer found incidentally is uncommon. Adrenocortical carcinoma has an estimated incidence of 1 to 2 per million people per year, ^[1] and accounts for about 2% of adrenal incidentalomas. ^[1] Up to 10% are functional. ^[1]

Prevalence depends on how hard you look and in whom. Autopsy series have a median adenoma prevalence of about 3%, while CT series have a median incidental mass prevalence of about 1.9%. ^[1] About 2% of the general population carries one, rising to over 7% in people over 70 years, with very few under 40. ^{[1] [2]}

The glands are involved in similar numbers. The mean diameter at discovery on CT is about 30 mm, ranging from 8 mm to 230 mm. ^[1] Although most lesions are nonfunctioning cortical adenomas, the discovery still matters: it is the only window in which an early-stage cancer, a pheochromocytoma, or an aldosterone-producing adenoma can be caught before it becomes symptomatic.

Most adrenal incidentalomas are benign adrenocortical adenomas. The Korean Endocrine Society 2017 guideline reports that nonfunctional benign adenoma is found in about 80% of adrenal incidentaloma cases; other series report pheochromocytoma in 3% to 6%, adrenocortical carcinoma in 2% to 5%, and metastasis in 1% to 2% (adrenocortical carcinoma is nearer to 2% in unselected series). Surgical series contain higher fractions of cancer and pheochromocytoma, because patients are referred. ^{[1] [3]}

Adrenocortical carcinoma deserves particular respect. It is rare (1 to 2 per million people per year) ^[1] but serious. Forty to sixty percent of these cancers are functional. ^[1] Cushing syndrome is the commonest hormonal presentation (about 45% of functional cancers in Sherlock 2020), about 25% cosecrete glucocorticoids and androgens, and about 10% secrete androgens alone. ^{[1] [3]} Multisecretion from more than one zone strongly suggests cancer rather than benign adenoma.

Other masses include myelolipoma, cyst, ganglioneuroma, haematoma or haemorrhage, macronodular adrenal hyperplasia, and metastasis (common in patients with known cancer). Size, heterogeneity, high attenuation, and irregular margins all push suspicion upward.

Each mass needs a dedicated adrenal imaging protocol; an incidental nodular report added to a scan done for something else is not enough. ^[2] The central imaging question is separating lipid-rich benign adenomas from indeterminate and malignant masses.

Unenhanced CT of the adrenals is the first investigation. ^[1] Attenuation is measured in Hounsfield units (HU). In a homogeneous lesion, a value of 10 HU or less on unenhanced CT is consistent with a benign lipid-rich adenoma. ^[1] The 2023 ESE/ENSAT guideline, summarised by Park and Kim 2023, takes this further: a homogeneous mass with no more than 10 HU is classified as benign regardless of size, with no further imaging and no follow-up. ^[2] This removed the older 2016 ESE size condition (below 4 cm), to avoid years of unnecessary scanning for clearly benign masses. ^[3]

Practical details that matter:

• The region of interest should cover at least two-thirds of the tumour circumference, avoiding necrosis or calcification. ^[1]
• The number applies to homogeneous lesions. In a heterogeneous mass the average can hide a malignant component, so heterogeneity itself is a warning sign rather than a single density reading. ^[1]
• Mean Hounsfield units for pheochromocytoma are 30 to 35, and only 0.5% of pheochromocytomas measure 10 HU or less. ^[1] This is the reason a clearly lipid-rich CT appearance allows metanephrine testing to be omitted: the 2023 guideline does not recommend plasma or urine metanephrines for masses with 10 HU or less. ^[2]

About 30% of benign adenomas are lipid-poor (attenuation above 10 HU) and cannot be distinguished from malignancy on unenhanced CT. ^{[1] [3]}

For lesions above 10 HU, an adrenal washout CT is a good next test (unenhanced, then 60 to 90 seconds after intravenous contrast, then a delayed image at 10 or 15 minutes). Adenomas enhance rapidly but wash out rapidly; cancers enhance rapidly but wash out slowly. Thresholds suggesting a benign adenoma are:

• absolute washout above 60%, calculated as (early-enhanced HU minus delayed HU) divided by (early-enhanced HU minus unenhanced HU), times 100 ^[1]
• relative washout above 40%, calculated as (early-enhanced HU minus delayed HU) divided by early-enhanced HU, times 100 ^[1]

Pheochromocytomas usually wash out slowly but can occasionally mimic benign adenomas. If suspicion remains, do not accept false reassurance; use the multidisciplinary team and further imaging. ^[1]

MRI with chemical shift imaging detects intracellular fat. Recent guidelines reserve MRI as a problem-solving tool, used first when CT is undesirable, for example in pregnancy, in children, or when iodinated contrast is contraindicated. ^[1] Accuracy for lipid-rich adenomas is high (around 95 to 98% on visual and quantitative analysis; chemical-shift MRI is reported anywhere from 81% to 100% in other series). ^[1]

PET with CT is used selectively when malignancy is suspected and other imaging is ambiguous. A low-uptake functional adenoma makes surgery avoidable; high uptake supports surgery, because the negative predictive value is excellent. ^[3] Biopsy is rarely needed and is justified only when the mass is nonfunctional, non-benign on imaging, and the result would change management. ^[3]

Size predicts cancer risk, although it is not absolute:

• under 4 cm: roughly 2% risk of adrenocortical carcinoma
• 4.1 to 6 cm: roughly 6% risk
• over 6 cm: roughly 25% risk

^[1] A large Italian series found 90% of carcinomas over 4 cm at presentation ^[1], and a 4 cm cut-off had 93% sensitivity but only 24% specificity in that series ^[4]; size alone is therefore insufficient. That is why anything indeterminate above 4 cm is best treated surgically.

Where imaging is unequivocally benign, the 2023 ESE approach is no follow-up imaging at all. Where the mass is indeterminate, immediate further imaging with a second modality is preferred over interval follow-up; if an interval scan is chosen, one repeat non-contrast CT or MRI at 6 to 12 months is suggested (2023 ESE), whereas the 2017 Korean guideline starts follow-up at 3 to 6 months and continues for 1 to 2 years. ^{[1] [2] [3]}

Significant growth is defined differently by guideline: the 2023 ESE guideline uses an increase of more than 20% that is also at least 5 mm in maximum diameter, while the Korean 2017 guideline uses an increase of 1 cm or more. ^{[2] [3]} Growth below these thresholds can be re-imaged after 6 to 12 months. ^[2]

Park 2023 and Sherlock 2020 cite a reassuring natural history: in a systematic review of nonfunctioning lesions, mean growth was only 2 mm over a median of 52.8 months; clinically significant enlargement (at least 10 mm) occurred in only 2.5% of patients, and none developed adrenal cancer. ^[1] Slow growth is reassuring on average, but it does not prove benignity in an individual. ^[2]

Clinical assessment for signs of hormone excess (Cushing syndrome, virilisation, gynaecomastia, bruising, proximal weakness, blood-pressure control, palpitations) plus biochemical testing is the other pillar. Pheochromocytoma and cortisol excess are excluded in everyone; aldosterone excess is excluded in patients with hypertension or low potassium. ^[1]

Clinical clues include paroxysmal headache (up to 90% of symptomatic cases), sweating (60 to 70%), palpitations, tremor, and labile or paroxysmal hypertension; 5 to 15% are normotensive at presentation, especially with small or familial tumours. ^[1]

Biochemistry is plasma free or 24-hour urinary fractionated metanephrines, with 90 to 95% sensitivity and a negative predictive value close to 0.99. ^[1] Specificity is 85 to 89% (77% in older patients). ^[1] The 2023 ESE guideline omits this test when unenhanced CT shows 10 HU or less; the Korean 2017 guideline still recommends it for all. ^{[2] [3]} Draw blood after at least 30 minutes supine into pre-chilled heparinised tubes, and avoid interfering drugs and caffeine for 24 hours beforehand. ^[1]

The 1 mg overnight dexamethasone suppression test is the recommended screen. ^{[1] [3]} Take 1 mg in the evening and measure serum cortisol the next morning. Interpretation differs between the older guidelines and the 2023 revision:

The main conceptual change in 2023 is that mild autonomous cortisol secretion (MACS) is defined as a post-dexamethasone cortisol above 50 nmol/L (above 1.8 microg/dL) with no subdivisions by degree. ^[2] Earlier guidelines used a "possible ACS" middle band. The post-dexamethasone value is best handled as a continuous risk variable rather than a yes-or-no. ^[2] Older patients suppress less often, and in a frail patient with limited life expectancy the test may be omitted altogether. ^[2]

Confirming MACS: 2023 ESE recommends demonstrating ACTH-independency (suppressed or low-normal morning ACTH) and repeating the DST, because false-positive DST results are common, and true MACS progresses to overt Cushing syndrome in less than 1% of cases. ^[2] Additional tests to gauge the degree of cortisol secretion (24-hour urinary free cortisol, a salivary cortisol taken near bedtime, and ACTH) may be useful. ^[2]

When classical features are present, the Endocrine Society uses two of three tests: a 24-hour urinary free cortisol collection, a bedtime salivary cortisol, and the 1 mg overnight DST. ^[1] The Korean 2017 guideline additionally describes a 2-day low-dose DST (0.5 mg every 6 hours for 48 hours), with hypercortisolism confirmed when serum cortisol is 1.8 microg/dL (50 nmol/L) or more. ^[3]

MACS is the commonest functional abnormality (prevalence up to 20% in Sherlock 2020; 5% to 25% in the Korean 2017 guideline depending on definition). ^{[1] [3]} It is associated with hypertension, insulin resistance, type 2 diabetes, obesity, dyslipidaemia, and increased mortality. ^[1] Comorbidities attributable to cortisol excess are estimated at 15 to 40%. ^[2] A large international retrospective cohort (2022) found higher mortality with autonomous cortisol secretion, with an age and sex effect. ^[2]

The 2023 guidelines therefore emphasise assessing comorbidities attributable to cortisol excess, especially hypertension and type 2 diabetes, so that they can be treated; randomised trial data on whether surgery improves cardiovascular or mortality outcomes are still lacking. ^[2]

Aldosterone excess is a minority of incidentalomas (normokalemic primary aldosteronism in up to 5.5%). ^[1] The clue is hypertension, with or without hypokalaemia. Only 9 to 37% of patients with primary aldosteronism are hypokalaemic, so a normal potassium does not exclude it. ^[1] Screen patients with hypertension or unexplained hypokalaemia with the plasma aldosterone-to-renin ratio, which has 68 to 94% sensitivity and a negative predictive value approaching 100%. ^[1] Confirmation (saline infusion, fludrocortisone suppression, or salt loading) and subtype characterisation then follow the Endocrine Society guideline. ^{[1] [3]}

Sex hormones and steroid precursors are measured when clinical or imaging features suggest carcinoma or pheochromocytoma: virilisation, feminisation, elevated DHEAS. ^[3] About 60% of adrenocortical carcinomas are functional, and presentation is most often Cushing syndrome (Sherlock 2020: about 45%); multisecretion strongly suggests cancer. ^{[1] [3]}

• Repeat hormonal testing is a point of genuine disagreement: the Korean 2017 guideline recommends annual hormone testing for 4 to 5 years (as does AACE 2009), while the 2023 ESE guideline recommends against repeated hormonal work-up in patients with a nonfunctioning tumour unless new clinical signs appear or comorbidities worsen. ^{[2] [3]}
• For patients with MACS, annual re-assessment of cortisol-attributable comorbidities is recommended, and progression to overt Cushing syndrome is expected in less than 1%. ^[2]
• In a patient with an extra-adrenal cancer and an adrenal mass, pheochromocytoma must still be excluded biochemically before any biopsy or surgery. ^[3]

There is no single size that commands surgery in every case. Decisions are individualised by a multidisciplinary team (endocrinologist, radiologist, pathologist, experienced adrenal surgeon), weighing likelihood of malignancy, degree of hormone excess, age, general health, and patient preference. ^[2] Surgery for suspected or proven adrenocortical carcinoma should take place in a referral centre with a surgeon performing more than 15 adrenalectomies per year; the 2023 ESE guideline defines a high-volume surgeon as at least 12 per year, with more than 20 desirable. ^{[1] [2] [3]}

Indications for adrenalectomy, summarised from the 2023 ESE and 2017 KES guidelines:

1. Functional tumours causing clinically significant hormone excess (Cushing syndrome, pheochromocytoma, aldosterone-producing adenoma, sex-steroid-secreting tumours). ^{[1] [3]}
2. Hormonally inactive masses larger than 4 cm, because of malignancy risk. ^{[1] [3]}
3. Imaging features suggesting malignancy: irregular margin, heterogeneity, non-uniform enhancement, adjacent invasion, non-benign washout. ^{[1] [3]}
4. Growth beyond the defined threshold during follow-up. ^{[2] [3]}
5. Young patients (under 40), children, adolescents, and pregnant women with an indeterminate mass, because malignancy and functional lesions are more likely. ^{[2] [3]}
6. Compressive myelolipomas, cysts, or metastases, when clinically indicated.

Laparoscopic adrenalectomy is the gold standard for most small and medium-sized benign lesions and for functional tumours, with shorter stay, less pain, and lower morbidity than open surgery. ^[1] The 2023 ESE guideline recommends a minimally invasive approach for benign adrenal tumours requiring surgery because of hormone excess, and also for tumours up to 6 cm with radiological signs of malignancy but without local invasion. ^[2] The Korean guideline agrees that laparoscopic removal is preferred for most pheochromocytomas, and recommends open resection for infiltrative pheochromocytomas over 6 cm. ^[3]

Open resection is recommended for larger tumours (higher capsule-rupture risk) and whenever there is local invasion or neighbouring organ involvement. Conversion from laparoscopic to open should be done early, before the capsule is disrupted. ^[1] Solitary adrenal metastases are often confined within the gland and can be removed laparoscopically if completely resectable. ^[1]

• Cushing syndrome and MACS: peri-operative stress-dose glucocorticoid cover is required, because the contralateral adrenal is suppressed. Wean to maintenance oral glucocorticoid and use synacthen testing to decide when replacement can stop. Patients need sick-day rules and education before discharge. ^{[1] [2] [3]}
• Primary aldosteronism: stop or reduce the mineralocorticoid-receptor antagonist after surgery and monitor potassium and blood pressure. ^[1]
• Pheochromocytoma: alpha-blockade for 7 to 14 days, volume expansion, a liberal sodium diet, and beta-blockers where tachyarrhythmia is a concern; teams should be ready for profound hypotension after the adrenal vein is divided. ^{[1] [3]}

Bilateral incidentalomas affect about 10 to 15% of cases. ^{[1] [3]} They go through the same clinical, radiological, and hormone evaluation as unilateral lesions. The 2023 ESE guideline uses a four-option schema: (1) bilateral (macronodular) hyperplasia, (2) bilateral adenomas, (3) two morphologically similar but non-adenoma masses, (4) two morphologically different masses. ^[2] Bilateral macronodular hyperplasia or bilateral adenomas with MACS deserve individualised options, and bilateral adrenalectomy is not recommended without overt Cushing syndrome. ^[2] In bilateral disease, remember to test for adrenal insufficiency (metastases, infiltrative disease, haemorrhage) and for congenital adrenal hyperplasia (early-morning 17-hydroxyprogesterone) where appropriate. ^[3]

Nonfunctioning lesions with clear benign imaging can be discharged from adrenal follow-up, under the 2023 ESE approach, although removal of the size criterion is recent and earlier guidelines used a 4 cm cut-off. ^{[1] [2]}

For indeterminate lesions not operated, one repeat CT or MRI is suggested at 6 to 12 months (2023 ESE); the 2017 Korean guideline uses 3 to 6 months, then follow-up to 1 to 2 years. ^{[1] [2] [3]}

Serial hormonal testing remains contested: annual testing for 4 to 5 years (KES 2017, and AACE 2009) versus none unless new features appear (2023 ESE). ^{[2] [3]} Where testing was performed in a Korean cohort with initially nonfunctioning tumours, 47% acquired an asymptomatic hormone abnormality over 5 years. ^[3] The best follow-up interval is individualised by the multidisciplinary team. ^[1]

• Under-40s and pregnancy: urgent assessment and a low threshold for surgery, since cancer and functional lesions are likelier. MRI replaces CT for additional imaging in this group. ^{[2] [3]}
• Frail or elderly: the 2023 guideline allows omitting the DST in frail patients with limited life expectancy (in whom MACS has diminishing impact); work-up should balance benefit against harm. ^[2]
• Adrenal masses in children, adolescents and all patients under 40 are worked up urgently; when disease is bilateral, congenital adrenal hyperplasia is excluded with an early-morning 17-hydroxyprogesterone. ^{[2] [3]}
• Low-resource settings: an unenhanced adrenal CT with Hounsfield-unit measurement is the single cheapest decisive test; washout calculations need no advanced software. Biochemistry available locally (DST, aldosterone-to-renin ratio, basic catecholamines) covers essentials. Expert adrenal surgery is the main gap when carcinoma is suspected, so referral should be early.

• An adrenal incidentaloma is a chance finding of at least 1 cm. Most are benign, nonfunctioning adrenocortical adenomas. ^[1]
• Each lesion needs dedicated adrenal imaging plus a hormonal work-up; unenhanced CT is first, and values of 10 Hounsfield units or less define a benign lipid-rich adenoma.
• The 2023 ESE guideline classifies homogeneous lipid-rich masses (10 Hounsfield units or less) as benign at any size, with no follow-up.
• Indeterminate masses are best triaged by multidisciplinary review; immediate additional imaging is preferred over surveillance scanning.
• Everyone gets a 1 mg overnight dexamethasone suppression test; post-dexamethasone cortisol of 1.8 microg/dL or less means normal suppression. Mild autonomous cortisol secretion is above that level.
• The post-dexamethasone cortisol is interpreted as a continuum in 2023 (MACS), not three separate categories; 2016 ESE and 2017 KES used "possible ACS" for 1.9 to 5.0 and "ACS" above 5.0.
• Plasma or urine metanephrines are essential unless the CT already shows 10 Hounsfield units or less; the aldosterone-to-renin ratio is added in hypertension or hypokalaemia.
• Size is a risk factor, not a diagnosis: 2% cancer risk under 4 cm, 6% between 4.1 and 6 cm, 25% over 6 cm. ^[1]
• Growth signalling malignancy is more than 20% plus 5 mm (2023 ESE) or 1 cm or more (2017 KES). ^[3]
• Surgery follows indication rather than size alone: significant hormone excess, suspicion of cancer, over 4 cm, documented growth, and young or pregnant patients. Asymptomatic benign nonfunctioning masses usually do not require operation. ^[3]
• Laparoscopic removal is standard for benign and functional lesions and for selected lesions up to 6 cm; large or invasive tumours need open surgery.
• Adrenocortical suppression after surgery for cortisol-secreting lesions demands stress-dose glucocorticoid cover, weaning, and synacthen testing before stopping.
• Guidelines disagree on follow-up: 2023 ESE stops imaging and repeated hormones for clearly benign lesions, whereas older guidelines planned serial scans and annual hormones for years.
• MACS rarely becomes overt Cushing syndrome; some patients improve blood pressure and glucose after surgery, but evidence of mortality benefit is limited.
• A low Hounsfield-unit value permits reassurance; heterogeneity should raise suspicion no matter what the measured density is.

Plain-language related topics: adrenocortical carcinoma, pheochromocytoma, primary hyperaldosteronism, Cushing syndrome, congenital adrenal hyperplasia, adrenal metastases, adrenal crisis, hypothalamic-pituitary-adrenal axis suppression, adrenal imaging.