Outcomes and opportunities: evaluating an unselected cohort of patients with adrenal incidentaloma at a tertiary hospital in New Zealand

Abstract

Background

The prevalence of adrenal incidentalomas is increasing, along with the incidence of functional or malignant lesions. This study assessed how imaging of adrenal incidentalomas at a tertiary hospital in Auckland, New Zealand, without a dedicated adrenal incidentaloma service, conformed to the American College of Radiology (ACR) recommendations for radiological screening, as well as how using a non‐washout imaging approach affected clinical and service outcomes. Secondary outcomes included determining how biochemical screening conformed to the European Society of Endocrinology (ESE) guidelines and patient‐related factors influencing completion of ACR screening.

Methods

A retrospective audit of adrenal incidentalomas was performed using a local healthcare database

Results

Data from 10 278 consecutive patients with computed tomography scans were evaluated. The overall prevalence of incidentalomas was 1.4%, and 143 lesions were detected. Of these, 65/143 (45.5%) followed the ACR imaging guidelines, and 21/143 (14.7%) followed the ESE guidelines. Older adults were 2.9 times less likely to complete ACR and ESE follow‐up compared to younger adults, while people with lesions >4 cm versus ≤4 cm were more likely to complete ACR follow‐up. Only 11 patients were referred to endocrinology, and this subgroup had a significantly higher rate of ACR adherent evaluation (81.8% vs 12.1%, P = 0.0002). Using a non‐washout method of radiological follow‐up resulted in 80 fewer scans.

Conclusion

Most patients with adrenal incidentalomas did not receive guideline‐recommended imaging or biochemical follow‐up. The establishment of integrated adrenal incidentaloma clinics and the use of non‐washout follow‐up scans may improve both clinical outcomes and patient experience.

Introduction

Adrenal incidentalomas are clinically inapparent masses in the adrenal glands that are detected during radiological investigation for other clinical indications. ¹ , ² , ³ , ⁴ , ⁵ , ⁶ The prevalence of adrenal incidentalomas has dramatically increased from around 1% in the 1980s to 4% in the late 2000s when higher resolution computed tomography (CT) scanners were introduced ¹ ; prevalence also increases with age. ² The majority of these adrenal incidentalomas are non‐functioning benign lesions, but it is essential to exclude malignancy and assess functional status. ³

In terms of radiological evaluation, American College of Radiology (ACR) recommendations are most commonly used and are discussed in diagram form in Figure S1: the recommendations divide adrenal incidentalomas into four distinct groups based on size and density on non‐contrast scanning including, firstly, those with diagnostically benign imaging features (typically less than 10 hu) (no radiologic follow‐up required), then those which are likely benign with a size more than 1–2 cm with no prior imaging that require further follow‐up with an adrenal CT in 12 months and finally those with indeterminate imaging features of 1–4 cm in diameter that may require an adrenal CT with or without washout dependent on density, and those with >4 cm in diameter (resection, biopsy or further imaging with CT positron emission tomography required). ⁴ , ⁵

There is a wide range of acceptable testing protocols for the endocrine evaluation of adrenal incidentalomas. ⁶ , ⁷ The European Society of Endocrinology (ESE) has published recommendations for assessing hormone excess for patients with adrenal incidentalomas, and these are similar to other published guidelines. ⁶ , ⁸

Recommendations include careful clinical assessment, including clinical examination for symptoms and signs of adrenal hormone excess. ⁴ These include a recommendation that all patients with adrenal incidentalomas undergo a 1‐mg overnight dexamethasone suppression test to screen for Cushing's syndrome or mild autonomous cortisol excess (MACS), screening for phaeochromocytomas with plasma metanephrines in all patients with imaging characteristics not in keeping with a benign adenoma (typically HU >10 and/or heterogeneous and/or lipid poor report descriptors) and selective renin‐aldosterone screening for primary hyperaldosteronism of those with hypertension and/or hypokalaemia only.

Data from New Zealand have shown poor adherence to radiological follow‐up and that this is influenced by radiology report recommendations. ⁹ Furthermore, few studies have evaluated the proportion of patients following a particular scanning or diagnostic testing protocol for follow‐up. ¹⁰ One of the major barriers to adhering to the ACR recommendations is the lack of clarity for non‐specialists regarding follow‐up CT scans and the logistical barrier of ordering CT adrenal scan with washout in a timely manner. ¹¹

Follow‐up scans recommended by multiple external guidelines are not without cost; a 2009 New Zealand review found that exposure to ionising radiation for follow‐up scans led to a higher potential risk of malignancy. ¹² Moreover, false‐positive rates from recommended endocrine investigations in this patient group can reach as high as 50%. These follow‐up scans have been found to take up to 25% of the allocated budget in one UK centre. ¹³ This becomes especially relevant when exploring the studies underlying the ACR model, in which contrast washout follow‐up was validated, and these studies were both in high‐malignancy‐prevalence populations; therefore, they were not truly representative of incidentalomas. ¹⁴ , ¹⁵ This has led to sustained interest in novel approaches and methods for the radiological evaluation of adrenal incidentalomas.

Seow et al. recently developed a simple three‐category model (Fig. S2) that utilises non‐contrast CT follow‐up scans. It divides follow‐up recommendations for adrenal incidentalomas based on size and imaging characteristics: category 1 requires no follow‐up scans, category 2 requires a follow‐up non‐contrast scan in 12 months, and category 3 requires multidisciplinary consultation MDM due to a high risk of malignancy. This model has not yet been evaluated in an international context. ¹⁰

This retrospective audit assessed how the imaging of adrenal incidentalomas at an Auckland tertiary hospital conformed to the published ACR recommendations for radiological screening. Secondary outcomes included the extent to which biochemical screening conformed to ESE guidelines and whether a non‐contrast‐based follow‐up approach (the Seow model) provided clinical and economic benefits.

Methods

Study design and setting

This study used data from the Health New Zealand Te Whatu Ora Waitemata District in Auckland, New Zealand. The audit included all adult inpatients (aged ≥18 years) who underwent abdominal CT scans in 2023.

Case identification and eligibility criteria

Adrenal incidentalomas were identified by searching the Waitemata District Picture Archiving and Communication System (PACS). An adrenal incidentaloma was defined as any adrenal mass ≥1 cm in diameter detected incidentally on full abdomen CT imaging for indications unrelated to known or suspected adrenal disease. This scan had to be the first for an emergency department or inpatient patient encounter. If a patient had multiple scans in 2023, only the first was included in the study.

The exclusion criteria were applied to remove cases that did not meet the definition of adrenal incidentaloma or in which confounding conditions existed. The exclusion criteria for cases were as follows:

• A prior adrenal mass already noted on earlier imaging reports (i.e. the adrenal mass was not a new incidental finding).

• The imaging study was performed to evaluate or follow up a known or suspected adrenal pathology (rather than being a truly incidental discovery)

• A known adrenocortical tumour or hormonal condition at the time of the scan, such as primary hyperaldosteronism, phaeochromocytoma or congenital adrenal hyperplasia

• Known malignancy at the time of the CT scan

• Under 18 years of age (≤17 years)

After applying the inclusion and exclusion criteria, the eligible cases constituted the final cohort of adrenal incidentalomas to be audited.

Ethical approval

As this study involved a retrospective review of anonymised data and did not involve any patient contact or intervention, formal ethical approval was deemed not to be required by the institutional ethics review board of the research and knowledge centre, North Shore Hospital (Auckland, New Zealand).

Data collection and variables

Data were collected from the picture archiving and communication system (PACS) and linked electronic medical records and entered into a secure, de‐identified database. Collected variables included the patient's National Health Index (NHI) number, date and type of CT, key radiological details (e.g. Hounsfield Unit (HU) value, maximum lesion diameter) and whether further imaging or endocrinology referral was recommended. Management and follow‐up data were also collected, including additional imaging, new malignancy diagnosis related to the incidentaloma, endocrinology referral, clinical evaluation for hormone excess and biochemical investigation. Clinical records and laboratory results were reviewed to collect data on biochemical investigations (whether performed, tests used and results).

Assessment of imaging follow‐up

Each case was classified based on whether further imaging was indicated and obtained, resulting in the following categories: no further imaging, ACR guideline consistent; no further imaging, ACR guideline discordant; further imaging, ACR guideline consistent; and further imaging, ACR guideline discordant. Next, the number of scans recommended by the ACR was compared with that recommended by Seow's three‐category model. ¹¹ The proportion of patients requiring a follow‐up scan was a co‐primary endpoint. The proportion of malignant cases detected was a secondary outcome.

Assessment of biochemical workup

Another secondary outcome was how endocrine testing conformed to the ESE guidelines. ⁶ Biochemical workup was classified as complete evaluation (all recommended biochemical tests performed), partial evaluation (some, but not all, indicated tests performed, or non‐guideline‐recommended tests used) or no evaluation

Assessment of impact of the Seow model

The clinical impact of the Seow model ¹¹ was determined by comparing the proportion of participants requiring follow‐up scans using this approach with ACR guidelines (the null hypothesis was that there would be no difference). The economic impact of the two approaches was determined by multiplying the difference in the number of scans by the average cost of CT adrenal scans in Auckland public hospitals, stated to be NZ$ 513.66 per investigation. ¹⁶

Statistical analysis

Data were analysed using both descriptive and comparative statistics. Firstly, the incidence of cases in each category of imaging management and each category of biochemical workup was determined. Secondly, concordance with the ACR recommendations (primary outcome) was expressed as a percentage of the total incidentaloma cases. Finally, for secondary outcomes, differences between groups that were and were not concordant with the guidelines were analysed using the chi‐squared test, or Fisher's exact test when expected cell counts were small.

Multivariate methods were used to explore the associations between patient factors and the likelihood of guideline‐concordant management (for both imaging and endocrine outcomes). Continuous variables, such as patient age and adenoma size, were converted into categories based on predefined cut‐offs (<65 vs ≥65 years and ≤4 vs >4 cm ¹⁷ , ¹⁸ ) to facilitate multiple linear regression.

Statistical significance was set at P < 0.05. Patients with missing data were excluded from the relevant analyses. All statistical analyses were performed using GraphPad Prism software (version 10; GraphPad, Boston, MA, USA).

Results

Study population

A total of 10 278 scans were ordered for inpatients at North Shore Hospital during the study period, of which 10 031 were excluded (Fig. 1). Of the 247 adrenal incidentalomas originally identified, 104 did not meet the size criteria for an incidental adrenal tumour (diameter <1 cm) or were described as having a bulky appearance. Of the 143 incidental lesions defined as incidental adrenal tumours, 80 (55.9%) were newly discovered, and the remainder had been previously detected. The study population was elderly (mean age, 69.1 years), and hypertension (45.5%) and diabetes (23.9%) were the most common comorbidities (Table 1).

Figure 1.

Distribution of adrenal incidentalomas.

Table 1.

Baseline demographics and clinical characteristics of the study population

Characteristics	Patients (n = 143)
Age, years	69.1 (67.8, 72)
Male sex, n (%)	74 (51.7)
Smoking status, n (%)
Current smoker	16 (11.2)
Ex‐smoker	54 (31.2)
Non‐smoker	73 (51.0)
Diabetes, n (%)	36 (23.9)
Hypertension, n (%)	67 (45.5)
Known malignancies/metastases, n (%)	13 (9.1)
Size of adrenal incidental nodules, cm	2.6 (2.2, 3.0) cm
Hounsfield Unit >20, n (%)	21 (14.7)
Known functional lesions, n (%)	3 (1.2)

Note: Values are presented as the mean and 95% confidence interval or as the percentage of patients (%).

Abbreviations: ACR, American College of Radiology; ESE, European Society of Endocrinology.

Prevalence and incidence

The overall prevalence of adrenal incidentalomas was 1.4%, and the incidence was 0.78%.

Radiological and overall biochemical follow‐up and ACR/ESE guidelines

The percentage of patients receiving care in accordance with the ACR imaging guidelines was 45.5% (65/143). Of these, 78.5% (51/65) completed their recommended follow‐up scan (Fig. 2). Of the 78 patients who did not complete the ACR follow‐up, 62.8% (49/78) completed a follow‐up scan for another indication.

Figure 2.

Proportion of American College of Radiology (ACR)‐defined follow‐up groups that completed follow‐up computed tomography (CT) scans. ACR+, ACR positive; ACR−, ACR negative.

The percentage of all incidental lesions managed according to the ESE screening guidelines was 17.5%, including 8.4% with partial ESE screening and 74.1% with no ESE screening.

ACR follow‐up in comparison with Seow model

Overall, across the entire cohort of 143 patients, only 20 patients required an additional scan when following Seow's model, and this approach would have resulted in 80 fewer scans than the ACR follow‐up recommendations.

From an economic perspective, performing 80 fewer follow‐up scans would save approximately NZ$ 41 092, with an additional NZ$ 2400 saving on the cost of contrast.

When the participants who received ACR‐compliant follow‐up (n = 65) were re‐evaluated using Seow's three‐category criteria, 14 (21.5%) were categorised in category 2 (requiring further scanning), and 51 (78.5%) were in category 1 (not requiring further scanning).

The percentage of patients recommended to undergo follow‐up scanning in the ACR‐compliant subgroup was significantly lower in the Seow group than in the ACR‐compliant subgroup (21.5% vs 78.5%, P < 0.0001, Fisher's exact test).

When those participants not completing ACR follow‐up (n = 78) were re‐evaluated using the Seow model, only six (7.7%) were in category 2 (requiring further scanning) and 72 (92.3%) in category 1 (not requiring further scanning). Again, the percentage of participants recommended to have follow‐up scans in this ACR non‐compliant subgroup was significantly lower in the Seow group than in the ACR non‐compliant subgroup (7.7% vs 62.8%, P < 0.001, Fisher's exact test).

Non‐adrenal malignancies and co‐existing adrenal lesions

In terms of adrenal lesions in patients associated with subsequently diagnosed malignancy (n = 13), pancreatic cancer (n = 3) and melanoma (n = 3) and carcinoma of unknown primary (n = 2) being the most common diagnoses, with gastric, uterine, lung, head and neck, and endometrial cancers being the other diagnostic types. Only five lesions could be characterised as definitive metastasis to the adrenal glands on a variety of follow‐up imaging modalities. The remaining eight lesions were characterised as co‐existing adrenal adenomas on follow‐up imaging.

Of the 13 individuals with known malignancies with co‐existing adrenal lesions, only two were in the ACR‐compliant subgroup, both of which had follow‐up scans. Both of these lesions would be characterised as Seow category 1 follow‐up (no follow‐up scan recommended); in this group of individuals with known malignancies and a co‐existing adrenal lesion diagnosis, there was no significant difference in the rate of follow‐up scans when the Seow group were compared to the ACR‐compliant subgroup (p = 0.3333, fisher's exact test).

For the remaining 11 non‐ACR‐compliant subgroup, 10 (91%) had follow‐up scans, but only two (18.2%) would be classified as needing a follow‐up scan using Seow's model, with a lower rate of follow‐up scans in the Seow Group compared to the non‐ACR‐compliant subgroup (P = 0.002, fisher's exact test).

Biochemical screening follow‐up by indication

All patients were recommended to complete a 1‐mg dexamethasone suppression screening test; however, this occurred in only 39 patients (27.3%).

Plasma metanephrines were measured in 84 (58.7%) patients; however, ESE guidelines recommend measurement in only 42 patients. Of these 42 patients, 25 (59.5%) completed plasma metanephrine screening. Of the 67 patients with hypertension and/or hypokalaemia for whom renin/aldosterone testing is recommended, 26 (38.8%) underwent testing.

Only 11 (7.6%) patients in the whole study cohort were referred to endocrinology for subspecialty review. Patients referred to endocrinology were significantly more likely to complete all endocrinology tests than those who were not (81.8% vs 12.1% respectively; P < 0.0001).

Functional lesions

Three hormonally active lesions were identified, two of which were aldosterone‐producing adenomas, and one was an adenoma that had excess aldosterone secretion from both glands (based on adrenal vein sampling). No new cases of phaeochromocytomas and/or adrenal Cushing's disease were documented during the study.

Variables associated with the completion of ACR/ESE follow‐up

Patients aged >65 years versus ≤65 years, those with an adenoma size of >4 cm versus ≤4 cm, and those with high‐ versus low‐density lesions were significantly more likely to complete ACR follow‐up (Table 2). Adherence to the ESE screening guidelines was significantly more likely in patients aged >65 years versus ≤65 years and those with versus without hypertension and diabetes (Table 3).

Table 2.

Factors associated with American College of Radiology follow‐up

Variable	Odds ratio	95% confidence interval	P‐value
Age > 65 versus ≤65 years	−2.9	−2.4, −12.5	0.0040
Size of incidental nodules >4 versus ≤4 cm	2.9	1.2, 6.9	0.0051
Current Smoking and previous smoking versus non‐smoking	1.0	−16.5, 5.4	0.3132
Diabetes (yes vs no)	1.2	−12.5, 3.2	0.2481
Hypertension (yes vs no)	1.7	−35.5, 3.1	0.0998
Hounsfield Unit >20 versus ≤20	2.0	2, 33.3	0.0465
Ethnicity (non‐European vs European)	1.7	−8.5, 50	0.0899

Note: Bold indicates p value < 0.05.

Table 3.

Factors associated with following the European Society of Endocrinology screening guidelines

Variable	Odds ratio	95% confidence interval (profile likelihood)	P‐value
Age > 65 versus ≤65 years	−3.0	−2.6, −12.6	0.0030
Size of incidental nodules >4 versus ≤4 cm	1.1	−2.4, 7.1	0.2772
Smoking versus ex/non smoking	0.2	−8.4, 10.6	0.8163
Diabetes (yes vs no)	2.1	0.01, 2.86	0.0380
Hypertension (yes vs no)	2.0	0.02, 3.23	0.0475
Hounsfield Unit >20 versus ≤20	1.7	−30, 2.74	0.1049
Ethnicity (Non‐European vs European)	1.2	−50, 11	0.2189

Note: Bold indicates p value < 0.05.

Discussion

This retrospective audit of 143 adrenal incidentalomas identified at a major tertiary hospital in Auckland, New Zealand, in 2023 showed significant disparities between clinical practice and established international guidelines. Our study was conducted at a centre without a specialised adrenal incidentaloma clinic and is the largest single number of reference cases in the New Zealand context.

In comparison with our centre, prior studies have reported a range of prevalence, incidence and follow‐up rates. ¹⁹ , ²⁰ , ²¹ , ²² A United States study revealed a 10‐fold increase in incidence from 4.4 to 47.8 per 100 000 person years between 1995 to 2017, but also found much lower rates of functional tumours, including those with steroid excess. ²⁰ Further UK studies have noted a slightly lower prevalence of incidentalomas compared to our centre at 0.7%, but also that up to 80% of lesions had received no formal endocrine follow‐up. ²¹

In our sample, the adherence rate to the ACR recommendations for imaging follow‐up was only 45.5%. This is similar to the 48% adherence rate reported in an American retrospective study. ²² Adherence to the ESE guidelines for hormonal evaluation was even lower, with a complete biochemical workup performed in only 17.5% of patients. These results highlight a critical gap in the management of adrenal incidentalomas within our region and suggest that a substantial number of patients may be at risk of missed malignancy or functional endocrinological pathology diagnoses.

We observed that older patients were significantly less likely to undergo both ACR‐recommended follow‐up imaging and ESE concordant biochemical testing than younger patients. This may suggest that clinicians are weighing the benefits of further investigations against the overall health status of elderly patients with frailty and comorbidities. One factor associated with better adherence to ACR‐recommended follow‐up was a high unenhanced CT attenuation (>20 HU), which is a well‐established radiological feature associated with potential malignancy. ¹⁴ , ¹⁵ , ¹⁷ , ¹⁸ This finding is reassuring because it indicates that radiologists are appropriately identifying this red‐flag feature and that clinicians are acting on it.

The malignancy rate in our study was under 10%, and hormonally active tumours were identified in only approximately 1% of the cases. A British study conducted in a district hospital found similar rates of malignancy and hormone‐producing tumours: approximately 8.6% of patients had a malignant lesion (mostly metastatic disease), and only 5% had overt hormonal syndromes (approximately 1% pheochromocytomas and 4% Cushing's). ¹⁹ The majority of incidentalomas in both this series and the literature are benign, non‐secretory adenomas. Although extensive workup has identified only a small fraction of malignancies or functioning pathologies, these cases can result in disorders that can cause significant morbidity and mortality (e.g. Cushing's syndrome and pheochromocytoma).

Adherence to the recommended biochemical workup for hormonal evaluation was poor, with only 17.5% of patients undergoing a complete biochemical workup as recommended by the ESE and another 8.4% receiving a partial evaluation. Approximately three‐quarters (74.1%) of the patients had no documented hormonal testing to assess conditions such as subclinical Cushing's syndrome, pheochromocytoma, or primary aldosteronism. The ESE guidelines recommend that all patients with an incidental adrenal receive at least a cortisol suppression test, with additional testing for catecholamine and aldosterone excess when appropriate. ⁶ In our audit, the lack of biochemical screening meant that many cases of potential hormone‐producing tumours may have been missed. For patients referred to endocrinology, 81.8% had full recommended hormonal testing, compared with only 12.1% of those not referred, indicating that endocrinology team involvement dramatically improved guideline adherence. When care was managed solely by primary referring teams, biochemical workup is often not performed. ¹⁸ , ¹⁹ , ²⁰ Other centres have similarly reported suboptimal radiological and biochemical follow‐up of adrenal lesions, with low rates of specialist referral and follow‐up. ²⁰ , ²¹ , ²²

The comparison between the ACR recommendations and Seow's three‐category model suggested that Seow's model could simplify radiology follow‐up recommendations and reduce the number of scans for benign lesions. However, both Seow's model and the ACR recommendations share similar limitations regarding malignant lesions. Applying Seow's model, 81.8% of the lesions in our study would not require further scans, 18.2% would need a scan at 6‐ to 12‐month follow‐up, and none needed discussion at a multidisciplinary meeting because all large lesions were not >20 HU. ¹¹ This reduction in avoidable imaging also limits exposure to contrast and radiation, thereby lowering the risk of complications such as contrast‐induced nephropathy while preserving diagnostic safety. If the calculated savings were extrapolated across similar‐sized health regions, the possible national savings to the health system could be even more substantial. Further analysis of both imaging guidelines revealed limitations: a small number of metastatic lesions would not have been recommended for follow‐up under the Seow model. ¹¹ In contrast, ACR‐based follow‐up would have resulted in only a minimal increase in imaging burden with the potential to characterise some of these higher risk cases. ¹¹ These findings highlight the need for the Seow model to be further validated in diverse populations with heterogenous malignancy risk.

We recommend the development of clear local guidelines or institutional protocols for adrenal incidentaloma follow up. Such guidelines should outline when to obtain follow‐up imaging (based on lesion size and imaging characteristics, as per ACR recommendations) and which biochemical tests to perform in all patients (as recommended by the ESE). The lack of unified local guidelines, both in New Zealand and worldwide, contributes to inconsistent practices. ⁹ , ²³ ACR guidelines can be complex and may require new types of radiological scans to be ordered, which has been identified as a key barrier to follow‐up. ²³ A defined protocol could empower radiologists and clinicians to take consistent action when an adrenal incidentaloma is reported. Integration of care is also crucial. Our results suggest that involving endocrinology specialists leads to a more complete biochemical workup of patients. Radiology reports should include prompts for endocrine referral and recommended tests because both national and international studies have reported that the content of the radiology report can significantly influence adherence to guidelines. ⁹ , ²³

Using audit data to create feedback loops (i.e. regularly reviewing local performance against benchmarks) can foster accountability and sustained improvement. In the longer term, these measures could reduce unnecessary imaging, improve the detection of functional or malignant lesions and enhance patient safety and resource utilisation across New Zealand healthcare settings. The use of a specific adrenal incidentaloma clinic has been associated with increased adherence to radiological and biochemical screening, with rates well over 50% both nationally and internationally. ²⁴ , ²⁵ , ²⁶ In one study, 69% of patients were discharged from these clinics. ²⁷ In New Zealand, the abnormal dexamethasone suppression test is the most common hormonal abnormality, with many returning to normal values on repeat testing. The use of artificial intelligence tools also shows potential, with some models having a high positive predictive value for detecting benign adrenal lesions. ²⁸ Further research and evaluation of these approaches may help improve care for patients with adrenal incidentalomas.

The strengths of this study include stringent inclusion and exclusion criteria to capture adrenal incidentalomas and the use of a systematic PACS‐based case identification process. By benchmarking practice against established international guidelines, we provide an objective measure of concordance and highlight areas where clinical practice diverges from the recommended standards. Our statistical analysis also identified several factors associated with the completion of ACR or ESE follow‐up.

Nevertheless, the limitations of our study must also be acknowledged. This was a retrospective, single‐centre audit conducted over a 1‐year period, which may restrict the generalisability of our results to other healthcare settings in New Zealand or internationally. As with retrospective studies, we were limited by the accuracy and completeness of existing documentation and by missing data. Furthermore, the lack of a primary care interface may have introduced selection bias because follow‐up scan referrals from primary care or private providers would be missed by our approach to data collection. In contrast, because both public and private providers use the same regional laboratory provider, our laboratory data are likely to be more complete. The sample size limited the power of the subgroup analysis, particularly with respect to malignancy detection and full biochemical workup.

Conclusion

This retrospective audit identified significant gaps between clinical practice and international guidelines regarding the management of adrenal incidentalomas in Auckland, New Zealand. Several factors predicting guideline‐recommended follow‐up were identified. In addition, we found that the Seow model required fewer follow‐up scans and less contrast than the ACR recommendations, leading to potential cost savings. Therefore, the use of the Seow model should be evaluated in larger and more diverse cohorts. Our work underscores the need for improved guideline adherence to ensure comprehensive assessment and management of adrenal incidentalomas.

Supporting information

Figure S1. Diagrammatic representation of ACR model.⁵

Figure S2. Diagrammatic representation of Seow model.¹¹

Data S1. STROBE checklist completed.

Data availability statement

The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.