Clinical Course of Adrenal Myelolipoma

A Long-Term Longitudinal Follow-up Study

Oksana Hamidi; Ram Raman; Natalia Lazik; Nicole Iniguez-Ariza; Travis J. McKenzie; Melanie L. Lyden; Geoffrey B. Thompson; Benzon M. Dy; William F. Young Jr; Irina Bancos


Clin Endocrinol. 2020;93(1):11-18. 

In This Article


Clinical Characteristics

Of 4957 consecutive patients with adrenal tumours, 305 (6.2%) were diagnosed with myelolipoma based on radiographic and/or histopathological data (when available). Median age at diagnosis was 63 years (range, 25–87). Overall, 305 patients (n = 168, 55% men) presented with 321 myelolipomas: 141 (43.9%) right-sided and 180 (56.1%) left-sided. A total of 289 (94.8%) patients had unilateral and 16 (5.3%) had bilateral myelolipomas (Table 1). At initial evaluation, median tumour size was 2.3 cm (range, 0.5–18.0) and median unenhanced CT attenuation was −37.8 Hounsfield units (HU) (range, −110, 41) (Figure 1, Panels A and B). Of 2 myelolipomas with unenhanced CT attenuation >10 HU, 1 had acute haemorrhage (26 HU) and 1 had small central calcifications (41 HU). In the latter case, magnetic resonance imaging showed a loss of signal on the out-of-phase sequence within the peripheral aspect of the mass consistent with macroscopic fat.

Figure 1.

A 33-y-old woman presented with intermittent right upper quadrant pain. She also reported constipation, early satiety and nausea. Computed tomography identified a 6.4 × 7.1 × 9.2-cm heterogeneous lesion containing macroscopic fat, arising from the medial limb of the right adrenal gland shown on the axial (Panel A) and coronal images (Panel B). The lesion contained internal diffusely distributed soft tissue components and perceived 'claw sign', consistent with a large adrenal myelolipoma. Biochemical workup was unremarkable. Symptoms subsided and the patient opted for observation

Most myelolipomas were discovered incidentally (n = 264, 86%), whereas some were found on cross-sectional imaging performed for cancer staging (n = 27, 9%), or during workup of mass effect symptoms (n = 14, 5%), including back/flank pain (n = 12), positional dyspnoea (n = 1) and provoked pulmonary embolism due to compression of the inferior vena cava (n = 1). Of 12 (4%) patients who were evaluated for congenital adrenal hyperplasia (CAH), 2 (0.7%) were diagnosed with CAH and 10 had either testing for CAH (n = 9) or thorough history for CAH (n = 1). A total of 19 (6.2%) had concomitant adrenocortical adenoma (contralateral = 12, ipsilateral = 7).

Five (1.7%) patients underwent diagnostic biopsy of the adrenal myelolipoma. There were no reported biopsy-related complications.

Patients were followed for a median of 54 months (range, 0.03–267). Adrenalectomy was performed in 37 (12%) patients with no reported surgical complications. Of 163 (53%) patients with ≥6 months of follow-up imaging performed at a median of 55 months (6–239), overall tumour growth/y was 0 mm (−6, 14). Overall duration of follow-up of patients with ≥6 months of follow-up imaging was 87 months (6–240). At study conclusion, 67 (22%) patients died of unrelated to adrenal mass reason.

Myelolipoma <6 cm vs ≥6 cm

A total of 44 (14.4%) patients had large (≥6 cm) myelolipomas, whereas 11 (3.6%) patients had myelolipomas ≥10 cm. When compared to <6 cm tumours, myelolipomas ≥6 cm were more likely to cause mass effects symptoms (31.8% vs 0%, P < .0001), be bilateral (9/44, 20.5% vs 7/261, 2.7%, P < .0001), have radiographic haemorrhagic changes (13.6% vs 1.1%, P < .0001) and more commonly underwent adrenalectomy (52.3% vs 5.4%, P < .0001). There was no difference in sex, age at diagnosis or evidence of hormonal hyperfunction between patients with <6 cm vs ≥6 cm myelolipomas (Table 1). Three out of 44 (6.8%) patients with ≥6 cm myelolipoma underwent surgery due to acute haemorrhage, whereas none of myelolipomas <6 cm required surgery for this indication.

There were 3 patients with large (≥6 cm) symptomatic myelolipomas who were followed conservatively. One patient had bilateral myelolipomas (a 10.9-cm right myelolipoma with radiographic haemorrhagic changes and a 2.7-cm left myelolipoma). The large right myelolipoma showed mild decreased in size on follow-up imaging (from 10.9 to 10 cm over 22 months). The decrease in size was attributed to involution of the haemorrhage within the mass. The 2.7-cm left adrenal myelolipoma remained stable in size and appearance. The second patient followed conservatively demonstrated stable size of the 9-cm asymptomatic left adrenal myelolipoma on imaging 2 years after initial diagnosis. The third patient with a 6-cm myelolipoma and abdominal pain was simultaneously diagnosed with prostate cancer. Mass effect symptoms resolved without intervention.

Myelolipoma With Radiographic Haemorrhagic Changes

Radiographic haemorrhagic changes within the myelolipoma were noted in 9 (3%) patients (3 women and 6 men), most noted in large tumours (median tumour size of 7.0 cm; range, 1.8–18.0). Five patients presented with mass effect symptoms (median size, 9.0 cm; range, 4.5–18.0 cm) and 4 were found to have myelolipomas incidentally (median size, 4.6 cm; range, 1.8–7.0). Haemorrhagic changes were more common in tumours ≥6 cm as opposed to <6 cm (n = 6 [13.6%] vs n = 3 [1.1%], P < .0001). Only 1 patient with haemorrhagic myelolipoma was on anticoagulation, while haemorrhage was thought to be unprovoked in 8 patients. Among 5 patients treated with adrenalectomy, indications for surgery included: acute haemorrhage (n = 3), diagnostic surgery (n = 1) and mass effect symptoms (n = 1). Two patients underwent laparoscopic resection, 2 had open surgery and 1 had laparoscopic surgery that was converted to open. None had congenital adrenal hyperplasia or concomitant adrenal adenoma.

Hormonal Evaluation

Of 126 (41.3%) patients evaluated for adrenal hormone excess, the majority (n = 114, 90.5%) demonstrated no evidence of autonomous hormonal hyperfunction. Among those with available data, 96 patients underwent evaluation for pheochromocytoma, and no cases of pheochromocytoma were found. Of 92 patients with biochemical evaluation for autonomous cortisol secretion (excluding one patient with persistent Cushing disease due to ACTH-producing pituitary adenoma), 3 (3.3%) were noted to have autonomous cortisol secretion attributed to concomitant contralateral (n = 2) or ipsilateral adrenocortical adenoma (n = 1) (Table 2).

Of 74 patients that underwent biochemical evaluation for aldosterone excess, primary aldosteronism was noted in 9 (12.2%) patients: due to concomitant ipsilateral (n = 3) or contralateral adrenocortical adenoma (n = 3), or bilateral idiopathic adrenal hyperplasia (n = 3).

Tumour Growth

A total of 163 (53.4%) patients had ≥6 months of follow-up imaging from initial diagnosis and were included in the analysis of tumour growth, with median duration of overall clinical follow-up of 87 months (range, 6–240 months). Ten (6.1%) patients had bilateral myelolipomas, whereas 153 (93.8%) had unilateral tumours (74 right and 99 left) (Table 3).

Of these, 26 (16.0%) myelolipomas increased by ≥1.0 cm over the duration of follow-up. Myelolipomas with ≥1.0 cm growth were larger at initial diagnosis (3.6 vs 2.2 cm, P = .01), more likely to have haemorrhagic changes (11.5% vs 0.7%, P = .007) and more commonly treated surgically (34.6% vs 3.7%, P < .0001), compared to myelolipomas with <1 cm overall growth.

At last follow-up, overall tumour change ranged from −10 mm to 115 mm (median, 0 mm) with growth rate ranging from −6 to 14 mm/y (median, 0 mm/y).

Surgical Resection

A total of 37 (12.1%) patients underwent adrenalectomy: 21 (56.8%) patients underwent laparoscopic adrenalectomy, 15 (40.5%)—open adrenalectomy and 1 (2.7%)—a laparoscopic adrenalectomy that was converted to open resection (pathology showed an 8.9 cm myelolipoma with fat necrosis and organizing haemorrhage) (Table 4). Indications for surgery included: prevention of tumour growth due to large tumour size or tumour growth at follow-up (n = 12), diagnostic surgery (n = 10), mass effect symptoms (n = 5), concomitant ipsilateral adrenal adenoma with hormonal excess (n = 4), acute haemorrhage (n = 3) and concomitant resection during surgery for another indication (n = 3) (Table 1).

Among 12 patients who underwent surgery for prevention of further tumour growth (due to large initial tumour size or tumour growth at follow-up), median initial tumour size was 8.0 cm (range, 1.5–15.0) and median rate of tumour growth was 5.6 mm/y (range, 2.8–14).