The scale shows density in Hounsfield units. R indicates right.
Customize your JAMA Network experience by selecting one or more topics from the list below.
Pasternak JD, Seib CD, Seiser N, et al. Differences Between Bilateral Adrenal Incidentalomas and Unilateral Lesions. JAMA Surg. 2015;150(10):974–978. doi:10.1001/jamasurg.2015.1683
Adrenal incidentalomas are found in 1% to 5% of abdominal cross-sectional imaging studies. Although the workup and management of unilateral lesions are well established, limited information exists for bilateral incidentalomas.
To compare the natural history of patients having bilateral incidentalomas with those having unilateral incidentalomas.
Design, Setting, and Participants
Retrospective analysis of a prospective database of consecutive patients referred to an academic multidisciplinary adrenal conference. The setting was a tertiary care university hospital among a cohort of 500 patients with adrenal lesions between July 1, 2009, and July 1, 2014.
Main Outcomes and Measures
Prevalence, age, imaging characteristics, biochemical workup, any intervention, and final diagnosis.
Twenty-three patients with bilateral incidentalomas and 112 patients with unilateral incidentalomas were identified. The mean age at diagnosis of bilateral lesions was 58.7 years. The mean lesion size was 2.4 cm on the right side and 2.8 cm on the left side. Bilateral incidentalomas were associated with a significantly higher prevalence of subclinical Cushing syndrome (21.7% [5 of 23] vs 6.2% [7 of 112]) (P = .009) and a significantly lower prevalence of pheochromocytoma (4.3% [1 of 23] vs 19.6% [22 of 112]) (P = .003) compared with unilateral lesions, while rates of hyperaldosteronism were similar in both groups (4.3% [1 of 23] vs 5.4% [6 of 112]) (P > .99). Only one patient with bilateral incidentalomas underwent unilateral resection. The mean follow-up was 4 years (range, 1.2-13.0 years). There were no occult adrenocortical carcinomas.
Conclusions and Relevance
Bilateral incidentalomas are more likely to be associated with subclinical Cushing syndrome and less likely to be pheochromocytomas. Although patients with bilateral incidentalomas undergo a workup similar to that in patients with unilateral lesions, differences in their natural history warrant a greater index of suspicion for subclinical Cushing syndrome.
Incidentally discovered adrenal lesions, or incidentalomas, continue to be a clinical conundrum. The prevalence of these lesions has been reported to be as high as 1% to 5% of abdominal cross-sectional imaging studies performed for reasons unrelated to the adrenal gland.1-4 Once an incidentaloma is discovered, its functional status and malignant potential should be evaluated. Multiple organizations have published guidelines outlining the evaluation of adrenal incidentalomas.4,5 Surgery is generally recommended for larger or growing lesions, as well as those with hormonal hypersecretion. Although the workup of bilateral and unilateral incidentalomas is similar, the results of small studies6-8 suggested differences in the 2 groups. We sought to determine the characteristics and natural history of patients having bilateral adrenal incidentalomas to provide greater insight into their workup and management.
We conducted a retrospective analysis of a prospectively collected database of consecutive patients referred to a tertiary care multidisciplinary adrenal conference at the University of California, San Francisco (UCSF) Medical Center. Patients evaluated within a 5-year period between July 1, 2009, and July 1, 2014, were reviewed. All patients with a referral to clinical endocrinology or endocrine surgery were automatically reviewed by a multidisciplinary adrenal conference consisting of endocrine surgeons, endocrinologists, and radiologists. Patient medical records were evaluated for demographics, adrenal lesion characteristics, functionality, intervention (if any), and outcome. Patients with adrenal incidentalomas were those who underwent imaging unrelated to adrenal disease or active surveillance for cancer. Those patients being screened for adrenal manifestations of genetic syndromes were not included in the analysis. The UCSF Committee on Human Research, the university’s institutional review board, approved the study protocol. The protocol was for medical record review and required no additional written or verbal consent beyond the general consent already signed by all patients treated at the UCSF Medical Center.
All patients underwent screening for pheochromocytoma and primary aldosteronism according to published guidelines.4,5 To screen for pheochromocytoma, all patients had plasma free metanephrines measured. Collection of 24-hour urinary fractionated metanephrines was used as an adjunct for patients with diagnostic uncertainty. Screening for hypersecretion of aldosterone included the ratio of plasma aldosterone to plasma renin and (in some patients) 24-hour urinary aldosterone measurement. These patients underwent further biochemical workup, including plasma potassium and renal function testing. Because these patients had bilateral lesions, positive testing for primary aldosteronism necessitated adrenal venous sampling.
To screen for hypercortisolism, patients underwent an overnight 1-mg dexamethasone suppression test, 24-hour urinary free cortisol test, or morning cortisol and corticotropin testing depending on the preferred modality of the referring endocrinologist. A diagnosis of asymptomatic hypercortisolism or subclinical Cushing syndrome (SCS) was made if at least 1 of the following 3 criteria were met: (1) cortisol level after dexamethasone suppression test exceeding 1.8 µg/dL (to convert cortisol level to nanomoles per liter, multiply by 27.588), (2) 24-hour urinary free cortisol exceeding 50 µg, or (3) multiple low corticotropin levels, suggesting pituitary suppression. In addition, one patient had a midnight salivary cortisol measurement (upper limit of normal, 0.09 µg/dL). The differential diagnosis for patients with bilateral nonsecreting lesions includes benign adenomas or adrenocortical carcinomas (more likely unilateral),9,10 and those found to have hypersecretion underwent further localization using a combination of imaging and adrenal venous sampling if surgery was planned. If there was any concern about the accuracy of diagnosis or regarding malignancy, the patient had repeat imaging and biochemical testing at 3-month to 12-month follow-up.
Statistical analysis was performed using statistical software (STATA, version 9; StataCorp LP). Comparisons of proportions were conducted using Fisher exact test.
Five hundred new patient referrals were presented at the multidisciplinary adrenal conference during the study period. Of this total, 135 patients had adrenal incidentalomas, 23 of whom had bilateral lesions. In this bilateral cohort, the mean age at diagnosis was 58.7 years, and 56.5% (13 of 23) of patients were women. The mean follow-up was 4 years (range, 1.2-13.0 years). Table 1 summarizes the characteristics of the patient cohort.
The analysis of imaging characteristics in the patients with bilateral incidentalomas revealed that right-sided and left-sided lesions were similar in size (mean, 2.4 and 2.8 cm, respectively) (Table 2). In addition, imaging characteristics on computed tomography demonstrated similar median densities on both sides, with 14.8 Hounsfield units (HU) on the right and 12.1 HU on the left. Radiologists at the adrenal conference described as adenomas those adrenal tumors on noncontrast computed tomography with less than 10 HU or with absolute washout exceeding 60%.4 Other tumors outside of these parameters, but without obvious invasion or adenopathy, were considered indeterminate. The description of most of these lesions was consistent with adenomas (73.9% [17 of 23]), while 9 lesions were indeterminate by density or by washout characteristics (26.1% [6 of 23] right and 13.0% [3 of 23] left).
Of 23 patients with bilateral incidentalomas, 30.4% (7 patients) had evidence of hormonal hypersecretion, most of whom had SCS. None had overt, clinically apparent Cushing syndrome. When patients with SCS were analyzed, no patients had diabetes mellitus, 2 patients had hypertension, and 2 patients were obese (one of whom was morbidly obese). In addition, this subgroup was diagnosed using a dexamethasone suppression test in 4 of 5 patients and 24-hour urinary free cortisol in 4 of 5 patients. These patients with SCS were followed up for a mean of 4.6 years (range, 1.5-10.0 years). Examination of the rest of the bilateral incidentaloma group for functionality demonstrated that one patient had primary aldosteronism with bilateral secretion as determined by adrenal venous sampling. The dimensions of the adrenal lesions in this patient were 2.4 cm on the right and 2.8 cm on the left. One patient had a left pheochromocytoma.
Compared with the group of patients having unilateral incidentalomas, the group having bilateral incidentalomas was statistically significantly more likely to have SCS (P = .009) and less likely to have a pheochromocytoma (P = .003) (Table 1). Of the entire adrenal incidentaloma cohort, only 2 patients had adrenocortical carcinoma (1.5%), both of whom had unilateral lesions (5.9 and 7.0 cm, respectively, on the left).
Only 1 patient with bilateral incidentalomas had unilateral resection. This patient had a left pheochromocytoma, and the laterality was determined by imaging characteristics, specifically, differences in HU densities (Figure). The patient underwent initial imaging after being seen with unrelated hematuria. Biochemical workup of this patient showed elevated plasma free metanephrines and 24-hour urinary fractionated metanephrines, with normal 24-hour urinary free cortisol and a normal ratio of plasma aldosterone to plasma renin. The patient with primary aldosteronism had data from adrenal venous testing that suggested bilateral disease, so adrenalectomy was not performed. Those with SCS were serially followed up for any symptoms or worsening biochemical findings. For various reasons, none of the patients with SCS underwent adrenalectomy, including patient preference, minimal or absent symptoms or sequelae of overt Cushing syndrome, or loss to follow-up.
We report the results of a retrospective review of 135 patients with incidentalomas among 500 patients with adrenal disease evaluated at a tertiary care multidisciplinary adrenal conference. Of 135 patients with adrenal incidentalomas, 17.0% (23 patients) had bilateral lesions. This finding is in agreement with other large population series, which found up to 23% bilateral prevalence.11 While there are many studies and guidelines for the evaluation of patients with adrenal incidentalomas, we focused specifically on those with bilateral lesions.
In our series, we demonstrated a higher prevalence of SCS among patients with bilateral incidentalomas. Other small series found similar results.6,12,13 To determine whether one side may be dominant in cortisol secretion, Young et al14 used adrenal venous testing, with resultant surgical intervention that led to some biochemical or clinical improvement. Iodomethyl-19-norcholesterol scintigraphy is not available for use in the United States.15,16
In this series, only one patient with a pheochromocytoma underwent unilateral resection. The side of resection was determined based on computed tomography criteria17 (Figure).
None of 5 patients with SCS underwent adrenalectomy. For those patients who met size criteria for resection (>3 to 5 cm, depending on patient age and comorbidities and tumor characteristics on imaging studies), patient preference for observation was taken into consideration. In the patients with SCS and hypertension or obesity, the degree of hypercortisolism was determined to be mild and unlikely to be responsible for these manifestations. The morbidity of bilateral adrenalectomy is as high as 18%.18
There is evidence to suggest that SCS may be undertreated.19 However, studies have largely been retrospective and heterogeneous in the definition of SCS and with regard to desired clinical outcome goals.20 One prospective randomized study21 showed improved outcome in patients with SCS who underwent laparoscopic resection of unilateral adrenal incidentalomas compared with those who had nonsurgical follow-up. However, patients with bilateral lesions are more complex because the laterality of cortisol hypersecretion must be determined before surgical intervention. The physician must weigh the risks of adrenalectomy by considering inadequate long-term randomized data showing improved outcome after surgery in this group of patients.19,22
There are several limitations to this study. First, it was dependent on a retrospective analysis, with its well-known drawbacks. Second, patients included in this study were referred to a tertiary care university medical center, which may be associated with a selection bias in that this cohort does not represent the overall population of patients with bilateral incidentalomas. However, this bias is somewhat mitigated because both medical and surgical endocrinologists were the sources of referral. In other words, the adrenal conference was not limited to just those referred for surgery. Third, patients in this series had a short follow-up, with a mean of 4 years. Longer follow-up may demonstrate that some patients with SCS will indeed require adrenalectomy because of worsening hypercortisolism or tumor growth. Larger series and longer follow-up will be needed to evaluate the long-term natural history of this population. Fourth, not all of our patients underwent the same testing for hypercortisolism, which led to variability in the diagnostic criteria for SCS. Other groups may have different criteria. Nevertheless, the overall result is consistent with other studies that suggest a higher prevalence of SCS among patients with bilateral incidentalomas. We recommend a standardized workup protocol for SCS in patients with incidentalomas (whether unilateral or bilateral) so as to have more uniform biochemical definition of SCS. This evaluation should include 24-hour urinary free cortisol, an overnight 1-mg dexamethasone suppression test, and morning cortisol and corticotropin testing. Some patients with bilateral incidentalomas and SCS may be candidates for surgery because of size and imaging characteristics of the tumors or the results of adrenal venous testing.
Bilateral adrenal incidentalomas are common, occurring in 23 of 135 patients (17.0%) with adrenal incidentalomas in this series. Recommended workup and surgical indications for patients with bilateral incidentalomas are similar to those for patients with unilateral lesions. In this series, 5 of 23 patients (21.7%) with bilateral incidentalomas had SCS.
Accepted for Publication: April 12, 2015.
Corresponding Author: Quan-Yang Duh, MD, Section of Endocrine Surgery, Department of Surgery, University of California, San Francisco, 1600 Divisadero St, Third Floor, San Francisco, CA 94115 (firstname.lastname@example.org).
Published Online: July 22, 2015. doi:10.1001/jamasurg.2015.1683.
Author Contributions: Drs Pasternak and Duh had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.
Study concept and design: Pasternak, Tyrell, Shen, Duh.
Acquisition, analysis, or interpretation of data: Pasternak, Seib, Seiser, Liu, Cisco, Gosnell, Suh, Duh.
Drafting of the manuscript: Pasternak, Tyrell, Duh.
Critical revision of the manuscript for important intellectual content: Pasternak, Seib, Seiser, Liu, Cisco, Gosnell, Shen, Suh, Duh.
Statistical analysis: Seib, Cisco, Suh, Duh.
Administrative, technical, or material support: Pasternak, Seib, Cisco, Gosnell, Shen.
Study supervision: Tyrell, Gosnell, Shen, Suh, Duh.
Conflict of Interest Disclosures: None reported.
Previous Presentation: This study was presented at the 86th Annual Meeting of the Pacific Coast Surgical Association; February 22, 2015; Monterey, California.
Additional Contributions: Avital Harari, MD (University of California, Los Angeles), Raymon Grogan, MD (University of Chicago, Chicago, Illinois), Elliot Mitmaker, MD (McGill University, Montreal, Quebec, Canada), Michael Tsinberg, MD (St Jude Medical Center, Fullerton, California), Michael Campbell, MD (University of California, Davis, Sacramento), David Greenblatt, MD (Sharp Rees–Stealy Medical Group, San Diego, California), and Anthony Boganey, MD, MBA (Naval Medical Center, San Diego, California), contributed to the University of California, San Francisco, multidisciplinary adrenal conference in the past and assisted in the prospective collection of patient data.