Autonomous Cortisol Secretion in Adrenal Incidentalomas Linked to Increased Mortality

Adrenal gland cancer, computer illustration.
Investigators sought to measure the association between mortality and levels of autonomous cortisol secretion in patients with adrenal incidentalomas.

Elevated autonomous cortisol secretion was found to be associated with a 3-fold increase in mortality in patients with adrenal incidentalomas, according to research results published in the Annals of Internal Medicine.

To evaluate the association between mortality and autonomous cortisol secretion in patients with adrenal incidentalomas, researchers conducted a population-based, retrospective cohort study ( identifier NCT03919734) involving consecutive patients from Sweden identified between 2005 and 2015. Individual outcome events data were collected from the national registers of the Swedish National Board of Health and Welfare.

Exclusion criteria included the presence of metastatic cancer, adrenal incidentalomas smaller than 1 cm, nonadenoma lesions, pheochromocytomas, primary aldosteronism or clinical Cushing syndrome, oral glucocorticoid treatment, inhaled steroid use or use of medications affecting dexamethasone metabolism, and systemic estrogen therapy.

The primary study outcome was all-cause mortality. The investigators also studied cause-specific mortality (eg, from cardiovascular disease, cancer, infection, or other disease). Secondary outcomes included a composite of major cardiovascular events (MACE) including death from a cardiovascular cause, nonfatal myocardial infarction, nonfatal stroke, hospitalization for heart failure, and revascularization.

The total cohort included 1048 patients (median age, 64.9 years; 58.5% women) who were followed for up to 14 years.

At baseline, 45.1% of participants had cortisol levels of 50 nmol/L or higher following a 1-mg dexamethasone suppression test (cortisolDST); 52.9% had hypertension; 18.7% had diabetes; and 20.6% had a medical history that included at least 1 cardiovascular event. A total of 54 patients underwent adrenalectomy. Patients with cortisolDST levels of 50 nmol/L or higher were more likely to be older and have a history of hypertension, previous cardiovascular disease, heart failure, diabetes, reduced renal function, or adrenalectomy.

Over a median follow-up of 6.38 years (interquartile range [IQR], 4.28-8.73), 16.2% of patients died. Per the investigators, mortality was increased in patients with cortisolDST of 83 to 137 nmol/L (hazard ratio [HR], 2.30; 95% CI, 1.52-3.49) or 138 nmol/L or higher (HR, 3.04; 95% CI, 1.86-4.98).

In reviewing the cohort in its entirety, the researchers found a “fairly linear relation” between mortality and cortisolDST until cortisolDST reached 200 nmol/L. In evaluating patients with cortisolDST of less than 200 nmol/L, a linear increase in risk was identified within this range.

The researchers also sought to predict patient mortality by the size, bilateralism, corticotropin (ACTH), and dehydroepiandrosterone (DHEAS) of adrenal incidentalomas.

When predicting mortality by size, bilateralism, ACTH, and DHEAS of adrenal incidentalomas, the researchers found HRs for unilateral adrenal incidentaloma, bilateralism, ACTH levels less than 2.0 pmol/L, and DHEAS levels less than 1.04 µmol/L were 1.00, 1.02, 1.02, and 1.47, respectively. HRs were adjusted for age, sex, smoking status, estimated glomerular filtration rate (<60 mL/min/1.73 m2), previous cardiovascular disease, and In-transformed cortisolDST.

In 65 patients, the cause of death was registered as cardiovascular disease; 51 patients died of cancer, 6 of infection, and 48 of other diseases. Patients who had a cortisolDST level of 83 nmol/L or higher experienced increased cardiovascular mortality.

Over a median follow-up of 5.17 years (IQR, 3.16-7.47 years), 14.4% of patients had at least 1 MACE. Those with a cortisolDST level of 138 nmol/L or higher had an increased risk of MACE (HR, 2.41), vs those with cortisolDST levels between 50 and 82 nmol/L and 83 to 137 nmol/L, who had a similar risk as patients with cortisolDST 50 nmol/L or less. In patients without pre-existing cardiovascular disease, MACE HRs for patients in each cortisolDST group were 1.10, 0.97, and 2.27, respectively.

Study limitations include the availability of only indirect evidence for causality between autonomous cortisol secretion and mortality, the use of electronic records to collect baseline data, and possible cortisolDST elevation from sources other than autonomous cortisol secretion.

“CortisolDST seems to be a linear risk factor for death and cardiovascular disease in patients with an [adrenal incidentaloma] and [autonomous cortisol secretion], and a plasma cortisol level of 83 nmol/L or higher is associated with a 2- to 3-fold increase in mortality,” the researchers concluded. “We suggest treatment of known cardiovascular risk factors in these patients and incorporation of our results in the decision about which patients to recommend for adrenalectomy.”


Kjellbom A, Lindgren O, Puvaneswaralingam S, Löndahl M, Olsen H. Association between mortality and levels of autonomous cortisol secretion by adrenal incidentalomas: a cohort study. Ann Intern Med. Published online May 25, 2021. doi:10.7326/M20-7946