Approach to the Patient with an Incidental Adrenal Mass

Abstract

Introduction

The frequent use of cross-sectional imaging in routine clinical practice, including computed tomography (CT) and magnetic resonance imaging (MRI), has been both a blessing and a curse. The high-resolution images with functional correlates speeds the evaluation of symptoms and follow-up of disease, but often the studies identify abnormalities for which the ordering physicians were not searching. The adrenal glands, ~15 g endocrine organs situated superior to the kidneys, are a common locus of such incidental findings. Like the thyroid and pituitary gland, the adrenals are prone to developing benign tumors with age, but as internal organs unlike the thyroid, adrenals are not palpable. Like the pituitary, the adrenal is composed of several cell types. The outer cortex produces steroids, while the chromaffin cells of the medulla produce catecholamines, primarily epinephrine. The cortex itself is divided into three zones: The outer zona glomerulosa, which synthesizes the mineralocorticoid aldosterone, primarily under the control of angiotensin II and elevated plasma potassium; the zona fasciculata, which synthesizes cortisol, primarily under the control of corticotropin (ACTH); and the zona reticularis, which synthesizes the androgen precursor dehydroepiandrosterone (DHEA) and its sulfate (DHEAS), also under ACTH stimulation.

Tumors might arise from any of these zones or from neoplastic cells that do not precisely mimic a particular zone of origin. Tumors of the medulla are called pheochromocytomas, and the catecholamine excess can cause hypertension and paroxysms of sweating, palpitations, and acute blood pressure elevations. These tumors are rarely malignant, and catecholamine production is typically proportionate to size. On the other hand, most incidentally discovered pheochromocytomas are small and thus minimally symptomatic¹. Cortical adenomas can produce aldosterone, cortisol, a mixture of these two hormones, and rarely DHEAS. Adrenocortical carcinoma (ACC) represents <0.1% of adrenal tumors but are highly malignant and often secrete multiple steroids and steroid hormone precursors such as 11-deoxycortisol and pregnenolone². Cortical adenomas that primarily produce aldosterone (aldosterone-producing adenoma, APA) often cause hypertension that is poorly responsive to conventional treatments and cause hypokalemia in a minority of cases. Most cortisol-producing adenomas (CPAs) are difficult to ascertain clinically, but even mild hypercortisolemia can also cause hypertension as well as glucose intolerance and bone loss. When severe, CPAs result in overt ACTH-independent Cushing syndrome and warrant treatment. A variation on the CPA is macronodular adrenocortical hyperplasia, a “goiter of the adrenals,” in which both adrenals are enlarged and nodular. Similar to the CPA, these tumors are relatively inefficient cortisol producers and are typically large before ascertained. Lastly, the differential for adrenal masses also includes less common and non-hormone producing pathologies such as myelolipomas, cysts, lymphomas, hematomas, hemangiomas and infiltrative diseases, as well as tumors arising from adjacent tissues that grow against the adrenals, such as ganglioneuromas, renal cell carcinomas, and sarcomas of the retroperitoneum.

To summarize, most incidentally discovered adrenal masses are cortical adenomas that are poorly functional. Most commonly, these tumors make variable amounts of cortisol proportionate to their size, and a central part of the evaluation is to gauge how much cortisol is produced and how significant this ACTH-independent synthesis is above the normal, regulated production to the patient’s physiology. APAs of any size, in contrast, can cause hypertension and should be strongly considered in any patient with hypertension and an adrenal mass. Pheochromocytomas and ACCs are rare and often large, with distinct imaging characteristics compared to cortical adenomas. Other primary adrenal masses and metastases or para-adrenal tumors complete the differential diagnosis. Most patients in whom an adrenal mass is incidentally discovered will have initial laboratory screening tests and one follow-up imaging study in a year, after which no directed testing is necessary. The purpose of the evaluation, therefore, is primarily to identify the minority of patients with tumors of greater significance due to hormone production or biologic activity and to focus attention on these patients in greater detail.

Approach to the Patient

The assessment of the patient with an incidentally identified adrenal mass can be operationally divided into three parts. First, the review of the imaging study should be at a level of detail that might not be found in the report. Size is a very important factor, not only in raising concern for malignancy or an uncommon tumor but also as a predictor of hormone (specifically cortisol) production. Imaging characteristics are likewise important, as discussed below. One must also visualize the contralateral adrenal gland and determine whether it is normal, atrophic, or also nodular – to narrow the differential diagnosis and tailor the approach. Second, the history and physical exam is very important in guiding the direction and intensity of the evaluation. The presence of hypertension might suggest pheochromocytoma, but primary aldosteronism is roughly 100 times more common and should be considered first. Cushingoid features and symptoms should be vigorously probed. Recent changes in weight and control of blood pressure, lipids, or glucose are tip-offs to tumor function, and medication increases are another variable to assess. Third, standard biochemical testing follows, and higher risk individuals might have additional testing. The screening tests used for adrenal hyperfunction are designed to have high sensitivity, but at the cost of significant false-positive rates. Consequently, the pre-test probability of patient having the disease is critical to the interpretation (positive predictive value) of the screening tests.

Biochemical Evaluation

Why is biochemical evaluation of hormone secretion needed in the assessment of adrenal nodules? Which hormones are assessed? Which adrenal nodules should be screened?

While the vast majority of incidentally identified adrenal nodules are non-functional (produce negligible amounts of hormones), around 10–15% of are hormonally active³, leading to clinically significant consequences. Adrenal nodules can be evaluated for cortisol, aldosterone, and catecholamine secretion, with assessments guided by clinical findings and imaging characteristics, while taking into account relative prevalence and the potential consequences of a missed diagnosis.

Approximately 6% of adrenal nodules secrete excess cortisol³, so given this relatively high prevalence, ACTH-independent cortisol production should be investigated in all nodules^4–6. The degree of glucocorticoid secretion can range from profound, presenting with overt signs of classic Cushing syndrome such as facial plethora, violaceous striae, and proximal muscle weakness, to mild ACTH-independent cortisol secretion. Though physical manifestations of hypercortisolism may be absent, mild ACTH-independent cortisol excess is associated with higher prevalence of conditions such as hypertension, diabetes, obesity, or osteoporosis, as well as higher cardiovascular and all-cause mortality⁷. On imaging, cortisol hypersecretion is more often found in those with large (>2.5 cm) lipid-rich adrenal lesions or in bilateral disease^7,8.

Primary aldosteronism (see Chapter “Primary Hyperaldosteronism: Approach to Diagnosis and Management”) is the most common cause of secondary hypertension with a prevalence ranging from 4% in the hypertensive primary care population to around 10–20% in those with resistant hypertension^9–12. Though hypokalemia is frequently described as a classic manifestation of primary aldosteronism, the majority of patients actually present with normokalemia, so current guidelines recommend screening for this condition in all hypertensive patients presenting with an adrenal nodule^13,14. On imaging, primary aldosteronism can present as a unilateral and small (typically <2 cm in size) adenoma, as well as bilateral hyperplasia, or with completely normal adrenal imaging^10,15.

Pheochromocytomas, adrenal tumors that produce catecholamines, comprise approximately 3% of adrenal nodules³. Classically, patients present with hypertension and paroxysms of headaches, palpitations, and sweating, but up to 20% might be normotensive¹⁶. Given the absence of symptoms in some patients and the potentially devastating cardiovascular consequences of missing this diagnosis, many have advocated for screening all adrenal incidentalomas for excess catecholamine secretion^4,17. However, more recent data has emphasized the importance of imaging characteristics in guiding this decision, finding that lesions with an unenhanced attenuation of <10 Hounsfield units (HU) on CT scan are only very rarely associated with a true pheochromocytoma^18,19. Thus, given the not insignificant risk of false positive results that arise in biochemical testing for this condition, some experts now advocate biochemical pheochromocytoma screening only for adrenal lesions with unenhanced attenuation of >10 HU²⁰. Additionally, pheochromocytoma should always be ruled out prior to adrenal biopsy or any planned surgical intervention given the risk of hypertensive crisis that can occur with manipulation of a catecholamine-producing lesion⁴.

What are the best screening tests to identify abnormal biochemical activity in an adrenal nodule? What is considered a “positive” result?

There are several tests that can be used to screen for excess cortisol production by an adrenal nodule, each one investigating a different potential pathophysiologic defect. Typically, the diagnosis of ACTH-independent cortisol secretion is only made if more than one test is abnormal. The preferred initial test is the low-dose dexamethasone suppression test, which identifies inappropriate cortisol secretion in the absence of ACTH stimulation. In this test, the patient is instructed to take 1 mg dexamethasone at 11 PM to midnight and obtain serum cortisol measurement at 8–9 AM the next morning. Various cut-offs have been proposed to identify abnormal cortisol secretion, ranging from 1.8 μg/dL (50 nmol/L) to 5 μg/dL (138 nmol/L) with sensitivities and specificities of this test differing based on the cut-off used. A value of 1.8 μg/dL essentially excludes clinically meaningful hypercortisolemia but is subject to a 20% false-positive rate^6,21 whereas a higher cut-off of 5 μg/dL is highly specific at the expense of missing some clinically important cases of ACTH-independent cortisol secretion^4,21. Thus, as an initial screening test, several guidelines recommend using the more sensitive 1.8 μg/dL cut-off as an indication that ACTH-independent cortisol secretion is possible, and therefore, additional workup is indicated (Figure 1)^5,6. A simultaneous dexamethasone level ensures that the medication has reached the target range for suppression. Other commonly used screening tests, late-night salivary cortisol and 24-hour urine free cortisol, are insensitive for detecting mild forms of hypercortisolemia^4,6. Rather than additional testing of cortisol production per se, indirect testing for attenuation of the hypothalamic-pituitary-adrenal axis are more useful to support the diagnosis of ACTH-independent hypercortisolemia in a patient with an adrenal mass and an abnormal dexamethasone suppression test. These tests include measurement of plasma ACTH and serum dehydroepiandrosterone sulfate (DHEAS), both expected to be near or below the lower end of the normal range but not necessarily suppressed or undetectable. The plasma ACTH must be drawn in the early morning when values are normally the highest, and an ACTH <15 pg/mL (<3.3 pmol/L) is suspicious of adrenal hypercortisolemia^4,22,23. Because DHEAS production from the normal adrenal gland, like cortisol, is ACTH-dependent, chronic dampening of ACTH release due to the heightened negative feedback from a CPA will lower circulating DHEAS concentrations. A very good metric for diagnosing ACTH-independent hypercortisolemia is a significant fall in serial DHEAS measurements over time. Given the multitude of options for cortisol testing, controversy regarding cut-offs and potential pitfalls of each test, any abnormal result should prompt a referral to endocrinology for further evaluation.

Open in a separate window

Figure 1.

Biochemical Screening for Hypercortisolemia

*Based on patient’s clinical features and/or if adrenal lesion >2.5cm

ACTH = adrenocorticotropic hormone; DHEAS = dehydroepiandrosterone sulfate; DST = dexamethasone suppression test; LNSC = late-night salivary cortisol; UFC = 24h urine free cortisol

Screening for primary aldosteronism consists of measuring serum aldosterone and plasma renin (Figure 2). Although many antihypertensives influence the renin-angiotensin-aldosterone axis, medication withdrawal is generally unnecessary. A positive case-detection test includes a suppressed plasma renin activity <1 ng/mL/h or direct renin <10 pg/mL and a serum aldosterone ≥10 ng/dL (277 pmol/L) ^24,25. These tests should be drawn simultaneously in the morning after the patient has been ambulatory for about 2 hours. Rather than using aldosterone-to-renin ratios, we advocate for interpreting the plasma renin and serum aldosterone levels in clinical context, which includes consideration of their antihypertensive regimen and serum potassium level. Resistant hypertension and hypokalemia raises pretest probability of primary aldosteronism; however, hypokalemia attenuates aldosterone production even from tumors and can cause false-negative screens, so hypokalemia should be corrected prior to testing. Elevation of plasma renin due to antihypertensives (such as calcium channel blockers, angiotensin-converting enzyme inhibitors, angiotensin receptor blockers, other vasodilators) or diuretics, particularly mineralocorticoid receptor antagonists (spironolactone and eplerenone), is a second major cause of false-negative screens, but simultaneously elevated aldosterone is a clue to a medication effect, which warrants repeat screening after medication withdrawal. False-positive tests, in contrast, are rare when the aldosterone and renin are interpreted sequentially and not only assessed together as the aldosterone-to-renin ratio. As with other cases of potential hormonal hypersecretion, a positive screening test warrants referral to an endocrinologist for confirmatory testing.

Open in a separate window

Figure 2.

Biochemical Screening for Primary Aldosteronism

AVS = adrenal vein sampling; PRA = plasma renin activity, or direct renin – PRA of 1 ng/mL/h ~ direct renin of 10 pg/mL; MRA, mineralocorticoid receptor antagonist

Patients with pheochromocytomas reliably demonstrate elevated levels of metanephrines in a 24-hour urine specimen or in the plasma (Figure 3). Plasma metanephrines are perhaps simpler to obtain in the primary care setting but are not usually drawn in the ideal way (supine after an indwelling catheter has been in place for 30 minutes) and so are often plagued by false-positive results¹⁷. In particular, the stress of a venipuncture frequently results in activation of the sympathetic nervous system and slight elevations of plasma normetanephrine, up to twice the upper limit of normal. As a result, about 3 out of 4 patients with an incidentaloma who have positive plasma metanephrines do not end up having a true pheochromocytoma²⁶. In addition, the normal range of normetanephrines observed in individuals with essential hypertension are higher than those without the condition. Despite these limitations, elevations of either plasma normetanephrines or metanephrines >3–4 times the upper limit of normal ^4,17 or any elevation of both metanephrines and normetanephrines¹⁷ increases suspicion of a true catecholamine-producing tumor. While anti-hypertensive agents do not need to be held, psychiatric drugs that block catecholamine reuptake (most commonly tricyclic antidepressants and selective serotonin or norepinephrine reuptake inhibitors) and high doses of adrenergic agonists will interfere with these assays and so should be tapered and discontinued at least two weeks prior to testing¹⁷. If pheochromocytoma cannot be excluded from initial screening, an endocrinologist should complete the evaluation with additional testing on an individualized basis according to the degree of clinical suspicion. Options for additional testing that specialists may pursue include repeating the biochemical testing (immediately or a few months later), obtaining a supine plasma sample or a 24-hour urine collection (if not obtained previously), checking for elevated chromogranin A, performing a clonidine suppression test, or obtaining additional imaging¹⁷. To avoid false-positive results in patients with low index of suspicion, 24-hour urine metanephrines can be employed as the initial screen when feasible. In addition to the aforementioned medications as well as transient stress, the main causes of false-positive urine metanephrines are sleep apnea and clonidine withdrawal, which can occur within hours even despite consistent dosing schedules.

Open in a separate window

Figure 3.

Biochemical Screening for Pheochromocytoma*

*Recommend screening all adrenal nodules for pheochromocytoma, although future guidelines might only recommend screening if the nodule density is >10 Hounsfield units

What are the potential pitfalls of these screening tests?

There are potential drawbacks to each of the screening tests for ACTH-independent hypercortisolemia, so these factors need to be taken into account when performing these tests (Table 1). Various medications can alter dexamethasone metabolism through induction or inhibition of cytochrome P450 3A4, and rapid dexamethasone clearance leads to inadequate suppression of ACTH and ultimately a false positive result. Measuring dexamethasone levels can help identify those situations and aid in the correct interpretation of the cortisol results; alternatively, a larger dose of dexamethasone (3 mg) to assure ACTH suppression can be used²⁷. Additionally, the cortisol assay measures total cortisol levels, and about 90% of cortisol is protein-bound and biologically inactive. Alterations in corticosteroid-binding globulin and albumin may lead to artificially higher (i.e. in women taking estrogen-containing contraception pills) or lower levels (i.e. in those with nephrotic syndrome or hypoalbuminemia). Measurement of free cortisol in the saliva avoids this problem but may not be interpretable in a shift worker whose circadian rhythm is altered. Last, urine free cortisol levels can be elevated in cases of high fluid intake (>5 L/day) or depressed if renal impairment is present.

Table 1.

Potential Sources of Error in Biochemical Testing

Test	Causes of False Positives	Causes of False Negatives
Dexamethasone suppression test	Drugs that increase dexamethasone metabolism  Anti-epileptics  Pioglitazone  Rifampin Increased CBG  Estrogen (ie OCPs, pregnancy)	Drugs that impair dexamethasone metabolism  Aprepitant  Cimetidine  Diltiazem  Fluoxetine  Itraconazole  Ritonavir Decreased CBG  Nephrotic syndrome  Critical illness  Malnutrition  Liver disease
Salivary cortisol	Contamination by topical steroids Licorice ingestion Shift workers Tobacco use
Urine free cortisol	>5L/day fluid intake	CrCl <60ml/min Cyclic CS Mild forms of hypercortisolism
Metanephrines	Drugs  Amphetamines  Buspirone  Decongestants  Ethanol  Levodopa  Prochlorperazine  SNRIs  TCAs  Withdrawal from clonidine Acute illness and/or hospitalization Autonomic dysfunction Hemodialysis (plasma)	Reduced renal filtration (urine)
Plasma Renin Activity	Direct renin inhibitors	ACEi/ARB Diuretics Mineralocorticoid receptor antagonists Pregnancy Sodium restriction
Aldosterone	Hyperkalemia	Hypokalemia

Open in a separate window

ACEi = angiotensin-converting enzyme inhibitor; ARB = angiotensin II receptor blocker; CBG = cortisol-binding globulin; CrCl = creatinine clearance; CS = Cushing’s syndrome; OCP = oral contraceptives; TCA = Tricyclic antidepressants; Sources ^6,25,63

Although the testing for primary aldosteronism can be influenced by many classes of anti-hypertensive agents, initial screening can usually be attempted without making changes to a patient’s medical regimen. The mechanism of most antihypertensives is either volume depletion or vasodilation, which raises plasma renin and subsequently aldosterone. In contrast, beta-blockers lower renin, but also lower aldosterone. Consequently, screening results are always interpretable as long as the plasma renin is suppressed²⁸. This principle is true even for patients taking mineralocorticoids receptor antagonists such as spironolactone and eplerenone²⁹. However, if plasma renin is not suppressed, potentially interfering medications should be held for 2–4 weeks, hypokalemia corrected and salt intake liberalized³⁰. Alternative anti-hypertensives that do not impact aldosterone and renin levels include alpha-antagonists, calcium channel blockers, and hydralazine, so these agents can be used to control blood pressures while other agents are withheld¹⁴. (see Chapter “Primary Hyperaldosteronism: Approach to Diagnosis and Management” for more information on the evaluation of hyperaldosteronism.)

As alluded to previously, while anti-hypertensive agents can and should be continued, certain psychoactive medications and stimulants will impact the measurement of plasma and urine metanephrines, so these should be tapered and discontinued a few weeks prior to testing (Table 1).

Once a patient screens positive for hormone secretion, what are the next steps in diagnosis and management?

While surgery is the standard of care treatment of overt Cushing syndrome, management of mild ACTH-independent cortisol secretion, often referred to as “subclinical Cushing’s” is less straightforward. Several recent studies found increased morbidity and mortality in patients with mild ACTH-independent cortisol secretion, presumably due to hypertension, glucose intolerance and dyslipidemia. While several small studies have shown improvements in glycemia, blood pressure, and dyslipidemia after surgical resection of a CPA, clear evidence of a mortality benefit from treatment has not been demonstrated^31–37. Thus, given the relatively low-quality evidence available thus far, current guidelines do not unequivocally favor surgical management for all patients with mild ACTH-independent hypercortisolemia. Instead, it is best to take an individualized approach to management with more serious consideration of surgery in those likely to derive the most benefit such as younger patients (age <40) or those with multiple cortisol-mediated comorbidities, especially if any are progressive or uncontrolled⁵. If surgery is pursued, glucocorticoid replacement might be necessary perioperatively given the risk of contralateral adrenal suppression and transient post-operative adrenal insufficiency³⁸. Patients with postoperative adrenal insufficiency are difficult to identify from preoperative laboratory testing but might be ascertained from a cosyntropin stimulation test on the first postoperative day³⁹. If there is clinical equipoise regarding the diagnosis, continued biochemical monitoring can be pursued annually for up to five years, because while hypercortisolism can manifest over time, it does not typically progress to overt Cushing syndrome^4,40,41. Medical management of hypertension with spironolactone is generally successful, and a variety of glucose lowering agents can be used to control diabetes. However, the other consequences of hypercortisolemia, including the catabolic features, require additional targeted therapies that carry either high cost or significant risks⁴².

Once a patient screens positive for primary aldosteronism, endocrinologists will often confirm the diagnosis by demonstrating lack of normal aldosterone suppression after salt loading (see Chapter Primary Hyperaldosteronism: Approach to Diagnosis and Management). Most commonly this is done by collecting a 24-hour urine sample after three days of oral salt loading, but some centers administer 2 L saline intravenously over 4 hours and measure serum aldosterone instead. Spontaneous hypokalemia coupled with a completely suppressed plasma renin and an aldosterone level >20 ng/dL is diagnostic of this condition and so does not require additional confirmatory testing¹⁴. The most common forms of primary aldosteronism are a unilateral APA, which is surgically curable, or bilateral hyperaldosteronism, which is medically managed^43,44. Thus, if a patient is an appropriate and willing surgical candidate, adrenal vein sampling is recommended to determine whether this is unilateral or bilateral disease, as adrenal imaging is unreliable in making this distinction and determining lateralization⁴⁵. Furthermore, an imaged adenoma in a patient with primary aldosteronism might not be the source of aldosterone, even when adrenal vein sampling lateralizes to the adrenal with the tumor⁴⁶. Given the technical challenges associated with this procedure, this should be done in a tertiary care center by an experienced interventional radiologist⁴⁷. If the patient is not a surgical candidate, all forms of primary aldosteronism can be managed medically with spironolactone titrated every 4–8 weeks until blood pressure and serum potassium normalize and the plasma renin is no longer suppressed²⁵. Eplerenone can be used instead if adverse effects such as menstrual irregularity or gynecomastia develop with spironolactone.

If an adrenal mass is consistent with pheochromocytoma based on biochemical and imaging findings, surgical resection is indicated. In order to avoid intraoperative hemodynamic instability and crisis, patients need to be prepared for surgery a few weeks before with a combination of alpha-adrenergic blockers such as phenoxybenzamine or doxazosin, followed by beta-blockade until blood pressures and heart rates normalize. To offset the orthostasis often induced by medical therapy, patients are encouraged to liberalize their sodium (>5 g/d) and fluid intake (>2.5 L/d) as well²⁰.

Imaging Evaluation

What are typical imaging modalities used to evaluate adrenal nodules? What imaging characteristics are concerning?

CT and MRI are the most common cross-sectional imaging modalities that identify incidental adrenal lesions and can also further characterize adrenal masses. A number of features on CT and MRI have been identified that help distinguish between benign and malignant lesions (Table 2). In general, benign tumors are smaller, homogenous, have lower HU on pre-contrast CT images and loss of signal intensity between in- and out-of-phase images on MRI.

Table 2.

Imaging Characteristics of Benign versus Suspicious or Indeterminate Adrenal Tumors

Imaging Feature	Benign Lesions	Suspicious or Indeterminate Lesions
Size	<4cm	>4cm
	Size stability over time	Size increase over time
CT	Homogeneous	Heterogeneous
	Regular margins	Irregular margins
	Presence of macroscopic fat	Presence of necrosis or hemorrhage
	Hounsfield units ≤10	Hounsfield units >10
	Relative washout ≥40%	Relative washout <40%
	Absolute washout ≥60%	Absolute washout <60%
MRI	Homogeneous, regular margins	Heterogeneous, irregular margins
	Loss of intensity between in-phase and out-of-phase imaging	No loss of intensity between in-phase and out-of- phase imaging
PET	Low ratio of adrenal SUVmax to liver SUVmax or SUVmean	High ratio of adrenal SUVmax to liver SUVmax or SUVmean

Open in a separate window

Absolute washout is calculated as: (HU_{post-contrast} - HU_washout) / (HU_{post contrast} - HU_pre-contrast).

Relative washout is calculated as: (HU_{post-contrast} - HU_washout) / (HU_{post-contrast})

Nodule size is important, with larger or growing lesions more concerning for malignancy⁴⁸ and more likely to show hyperfunction. Specifically, lesions larger than 4 cm in patients without prior history of cancer are associated with higher probability of adrenocortical carcinoma⁴⁹. Several society guidelines recommend surgical resection for masses greater than 4 cm, with or without other concerning radiologic features^4,5,50. Some exceptions to this size limit include myelolipomas, which are benign tumors identified by the presence of macroscopic fat in the tumor, and tumors that have not enlarged over many months of careful surveillance – an approach often chosen for patients at high risk for surgical complications. Meanwhile, lesions smaller than 4 cm with benign features can often be monitored with follow-up imaging, unless significant hormone production is found. Size stability over 6–12 months makes malignancy unlikely.

The preferred imaging scan to characterize an otherwise identified adrenal tumor is a triphasic CT scan without and with contrast, plus a washout scan. The CT images without contrast assess the density of tissues in the body compared to water, which is allocated the tissue attenuation value of 0 measured in HU. HU are only meaningful in homogenous lesions, as heterogeneous lesions by definition have various attenuations in different regions of the tumor. Of note, non-contrast CT images are required for interpretation of HU; HU of adrenal lesions incidentally found on contrast-enhanced scans alone cannot be interpreted. Most benign adrenal adenomas are lipid-rich, have homogenous appearance, regular borders, and low Hounsfield units on CT. The conventional cutoff of ≤10 HU is 71% sensitive and 98% specific for benign lipid-rich adenomas⁵¹. In fact, some guidelines state that tumors less than 4 cm in size, homogenous, and have HU ≤10, are benign, and no further imaging is required⁵.

Upon administration of contrast, the HU of the lesion are measured at two time points: about 5 minutes, corresponding to peak enhancement, and 15 minutes during washout. These values are used to calculate the relative and absolute contrast enhancement washout percentages. Benign tumors have faster washout compared to malignant tumors. A relative washout of ≥40% and an absolute washout ≥60% are associated with benign tumors^52,53. The converse, however, is not always true; benign pheochromocytomas and some cortical adenomas can demonstrate delayed contrast washout.

The MRI technique of chemical shift imaging has also been used, which relies on the differences in resonance frequencies of proton nuclei for water and lipid molecules in a magnetic field, due to their distinct chemical environments. When the signal intensity is compared with the in-phase images, lipid-rich adenomas have loss of signal intensity on the out-of-phase images, whereas malignant tumors or pheochromocytomas do not^54,55. Due to lack of standardization in MRI techniques and measurements, this imaging modality is generally less preferred but avoids ionizing radiation.

When should I monitor adrenal nodules with repeat imaging versus refer to a specialist?

The European Society of Endocrinology clinical practice guidelines recommend that for nonfunctioning adrenal nodules that are <4 cm in size, homogenous, and have a density of ≤10 HU on non-contrast CT, no further evaluation or follow-up imaging is needed⁵. Nevertheless, most endocrinologists conservatively perform a single follow-up (generally non-contrasted) scan for most tumors, particularly those between 2 and 4 cm, in 6 to 12 months after the initial study based on size. As discussed earlier, the probability of hypercortisolemia increases with size, and repeat screening should be considered for tumors >2 cm unless unequivocally normal on initial testing. For example, a 50-year old patient with a 3 cm cortical adenoma has a dexamethasone suppressed cortisol of 1.5 μg/dL, ACTH of 18 pg/mL, and DHEAS of 80 μg/dL (low-normal). Given the marginally normal results in a patient with a fairly large mass, repeat testing for hypercortisolemia is warranted in 6–12 months, independent of the decision to repeat imaging. If imaging and biochemical stability is observed after 6–12 months, no further evaluation is generally warranted, although subsequent rescreening is prudent if the clinical status changes. Lesions with concerning features, such as those >4 cm, heterogeneous appearance, or >10 HU, can be referred to an endocrinologist or endocrine surgeon independent of hormone production.

What other imaging should a primary care provider order?

We recommend deferring the decision for additional imaging to specialists. Under most circumstances, CT or MRI will both provide specialists with adequate information to determine the most appropriate management. Approximately 12% of the time, adrenal tumors may be indeterminate on CT or MRI⁵⁶. For large or hormone-producing masses, specialists might recommend further evaluation with flurodeoxyglucose positron-emission tomography ([¹⁸F]-FDG-PET/CT) or meta-iodobenzguanine ([¹²³I]-MIBG) scans. [¹⁸F]-FDG-PET/CT can be useful to rule out malignancy, particularly in patients with known or suspected cancer^57–60. [¹²³I]-MIBG scan uses an iodine-labeled norepinephrine analog that is transported via the catecholamine reuptake system, which is specifically accumulated in pheochromocytomas but not cortical tumors^61,62.

How does imaging impact clinical management?

Several features on imaging can impact clinical management. Small, lipid-rich adenomas that are nonfunctional on hormonal testing generally require no intervention. Lesion that are larger, heterogeneous, or have higher Hounsfield units may warrant further diagnostic imaging, biopsy or surgery. For lesions that warrant surgery, surgeons may prefer open over laparoscopic surgery for large tumors with concern for malignancy or pheochromocytoma, because spillage of tumor cells can spread disease and thwart a curative operation^4,5.

Summary

Most incidentally discovered adrenal masses are benign and poorly functional adrenal tumors. Nevertheless, these tumors do require further investigation. Physicians should remain alert for concerning imaging findings and features of hormone excess. Screening biochemical studies in conjunction with history, physical exam, and review of the images are used to triage patients to specialists, determine if follow-up imaging is necessary, and recommend therapies for co-morbidities.

Acknowledgments

Footnotes

References