This study reports the management of 15 patients with adrenal myelolipomas at a tertiary adrenal surgery referral centre. Indications for up-front surgery were indeterminate features on radiology (n = 2), attributable symptoms (n = 1) and mass effect on surrounding organs (n = 1). No patient underwent surgery based on the size of the lesion alone. Of the 11 patients undergoing surveillance, only 1 subsequently underwent surgery due to the development of indeterminate radiological features on subsequent imaging with new onset abdominal symptoms. Adrenal myelolipoma was the final histological diagnosis in all cases. In all patients undergoing surveillance, the rate of growth was low at 0.13 cm/year, including 6 of 10 patients with lesions > 4 cm on initial investigation. These findings suggest that surveillance for presumed myelolipomas in the absence of concerning radiological features or symptoms is safe strategy.

In our cohort, adrenal myelolipoma comprised of 2.8% of total cases of adrenal lesions. This incidence is lower than the literature with case series reporting incidence between 4 and 10% [7, 21,22,23]. This may reflect a referral bias that not all cases of adrenal myelolipoma have been reviewed at the adrenal multidisciplinary tumour board. The mean age for diagnosis was between fourth to fifth decade with range of age at initial diagnosis from 27 to 70 years, with female predominance (Table 1). This has been similar to reported cases in different studies [7, 24,25,26]. The size of the adrenal lesions ranged from 2 to 15 cm, comparable to the reported literature [14, 27]. All resected adrenal lesions within this cohort were benign.

There is a gap in understanding the true progression of these tumours which influences the decisions in management of adrenal myelolipoma. The recommendations based on size are encompassed under the guidelines of benign adrenal lesions (< 4 cm and > 4 cm) as per the European Society of Endocrinology clinical practice guidelines (ESE/ENSAT) 2016 [28]. In our prospective series representing one of the largest single-centre experiences, > 65% adrenal myelolipoma tumours were > 4 cm: surgery group mean size was 7.4 cm (4.7), and surveillance group was 4.3 cm (1.5) (Table 1). Although the surgery group mean size was greater compared to surveillance group, almost 40% of patients within surveillance group in comparison with 26% patients in surgery group had adrenal myelolipoma size of > 4 cm (Fig. 2). This shows that despite the mean difference in size between both groups, size was not the only independent factor in the decision process.

Indications for surgery in adrenal myelolipoma are based largely on retrospective studies of case-based reports and expert opinions. Gershuni et al. (2014) in a review of 16 patients with adrenal myelolipomas noted the following indications for adrenalectomy: abdominal pain (44%), large tumour > 8 cm (50%), atypical radiological appearance (31%) and inferior vena cava compression (7%) [14]. Shenoy et al. (2015) in retrospective review of 584 cases recommended the following indications for surgery: symptomatic tumour, size > 4 cm, metabolically active tumour and suspicion of malignancy on imaging study [10]. Weber et al. (2019) reported 12 patients with adrenal myelolipoma who underwent surgery and categorised them into 3 clinical groups: the presence of local symptoms, radiological suspicious sign or the presence of local symptom and radiological suspicious sign [17]. Calissendorff et al. (2021) recommended option of surgery for large adrenal myelolipoma (> 6 cm) or indeterminate lesion and definitive surgery for mass effect or acute haemorrhage within the adrenal myelolipoma [5]. In our study (n = 15), the indications for surgery were for radiologically indeterminate lesions (n = 2), symptomatic and indeterminate lesions (n = 1), symptomatic lesions with a change in size (n = 1) and mass effect on surrounding organs (n = 1). These decision factors are consistent with the reported literature.

Hamidi et al. in their large case series of 321 patients suggested adrenal myelolipomas > 6 cm in size are more symptomatic [7]. Han et al. (1997) reported that nearly 50% of those from observed group either had an increase in size or became symptomatic [8]. In our study, majority of the patients remained asymptomatic despite the growth in size of adrenal myelolipoma. Within our surgery group, three patients with symptoms had adrenal myelolipoma measuring > 6 cm. One of them was a case of giant adrenal myelolipoma measuring 15 cm causing mass effect on intra-abdominal adjoining organs. However, within surveillance group, one patient with a lesion < 4 cm became symptomatic and indeterminate on imaging, while one patient with size > 6 cm remained asymptomatic. There was no statistical significance found on correlation between age of diagnosis and size of the adrenal myelolipoma (Fig. 2). This suggests that symptoms and age of diagnosis are likely independent factors, which need to be taken into consideration along with size to make individualised decisions for surgery versus surveillance.

Indeterminate adrenal lesions pose a challenging dilemma in management. ESE/ENSAT recommends three options for indeterminate lesions in adrenal incidentalomas (AI): immediate additional imaging with different modality, interval imaging with 6–12 months or surgery [28]. Calissendorrff et al. recommend considering interval imaging or surgery depending on the clinical, biochemical and differential diagnosis parameters when considering options in indeterminate lesions [5]. Melck et al. from their case series reported 7 of the 16 patients in the observed group for indeterminate lesions crossed over to surgery at mean of 13.1 months [29]. In our study to consider an adrenal lesion indeterminate, the patient underwent combination of different imaging modalities: computed tomography (CT), magnetic resonance imaging (MRI), positron emission tomography (PET-CT) and iodine-131-metaiodobenzylguanidine (MIBG) scan. Out of them, three patients had indeterminate lesion in imaging workup for adrenal myelolipoma. After adrenal multidisciplinary tumour board discussion, all three underwent adrenalectomy, based on rationale that indeterminate lesions can masquerade and harbour otherwise ambiguous adrenal pathologies [30, 31]. Adrenal myelolipoma was confirmed on pathology specimen in all three cases. Our study supports surgical intervention in indeterminate lesions as suggested as one of the options by the ESE/ESNAT guidelines [28].

Within the surveillance group with radiologically confirmed adrenal myelolipoma, further decision of surveillance versus no surveillance remains a matter of discussion. Adrenal myelolipomas grow in an indolent manner and remain asymptomatic in majority of cases. Within the surveillance group, our study showed that tumours are slow growing at growth rate of 0.13 cm/year (size range 0.2 to 1.9 cm/year’s range 3 to 18), and in majority of patients, this growth did not trigger move to surgery on adrenal multidisciplinary tumour board discussion. Campbell et al. (2017) in their case series consisting of 69 patients showed that during mean interval surveillance of 3.9 years, 16 tumours showed average growth rate of 0.16 cm/year (0.08–0.71 cm/year). On further analysis of our surveillance group, adrenal myelolipomas with initial size of > 4 cm showed mean growth rate (SD) cm per year by 0.20 (0.12) cm, and those of < 4 cm showed mean growth rate (SD) cm per year by 0.11 (0.16) cm (p = 0.421). This would suggest that adrenal myelolipoma measuring > 4 cm tends to grow at faster rate compared to those < 4 cm; however, the underlying factors driving this unpredictable growth pattern may have confounded to the statistical insignificant result. Campbell et al. (2017) showed association with both younger age and longer duration of time interval between CT scan (mean age: growth group 50 years and no-growth group 62 years) [9]. This would suggest that these tumours if observed over a longer period will grow gradually; however, in majority, cases will still remain within insignificant size to cause any symptoms. In our study, depending on their period of surveillance (1–18 years), patients underwent between 1 and 5 computed tomography interval scans. As the majority of these patients did not need surgical intervention, the potential benefit of repeated interval imaging should be balanced against the radiation exposure from computed tomography imaging.

In rare instances, these tumours can show abrupt growth pattern. Within our study, one patient on radiology interval imaging showed change in size from 5.5 to 12 cm after 11 years of follow-up. After 4 years of interval imaging, the tumour size reduced to 7.3 cm, and patient remained asymptomatic. Weber et al. (2019) reported change in one patient followed for over 30 years showing increase of 0.15 cm/year from 45- to 92-mm size, while three patients showed > 5-mm growth within 6 months of follow-up [17]. This demonstrates that in exceptional cases, these tumours can exhibit unpredictable growth pattern; however, the majority show slow linear growth. The underlying growth triggers causing sudden changes in size are poorly understood, and over a long duration of follow-up, very few cases will show significant abrupt change in size.

The current study was retrospective as it would be difficult to obtain a prospective study given its rare and clinical course of adrenal myelolipoma. Due to its low prevalence, the true statistical interpretation is limited to show causality or correlation to various factors triggering the growth pattern. There is also gap in understanding the true interval and duration of radiological imaging within the surveillance group.