This study provides the results of a survey on the diagnosis and management of hypercortisolism, among Italian centres affiliated to SIIA and ESH managing patients with hypertension. The survey covered a large proportion of Italian territory, by involving 82 centres in 14 regions with almost 80.000 patients, representing a large population in a real-life setting (the proportion of resistant hypertension was in line with the literature) [21, 22].

The survey highlighted that screening of hypercortisolism still remains an issue across different medical specialties. Importantly, roughly one third of the centres would not screen for hypercortisolism even in the presence of hypertension and specific features of Cushing’s syndrome. This holds true independently of geographical distribution of the centre, prevalent medical specialty or type of the centre (excellence vs non-excellence). An impressive high proportion of underscreening has also been identified in patients with hypertension and adrenal mass (60% of the centres). Notably, young hypertensive patients < 50 years are not undergoing screening for hypercortisolism in the majority of the centres (roughly 70%), with no difference among medical specialists. A large rate of underscreening of hypercortisolism was also identified in patients with resistant hypertension (40% of the centres), with an even higher proportion of underscreening among cardiologists. Even if the data collection was not planned for calculating the prevalence of cases of Cushing’s syndrome and MACS among hypertensive patients, some estimates can be extrapolated. When only first evaluations of patients were considered (since the likelihood of diagnosing hypercortisolism might be higher in this category of patients), the prevalence of new diagnoses is 0.3% (average of 63 cases/21.890 patients/year) for Cushing’s syndrome and 0.5% (average of 98 cases/21.890 patients/year) for MACS. Those numbers are much lower than expected, since the prevalence of Cushing’ s syndrome and MACS among hypertensive patients was up to 2% and 1–8%, respectively, in previous studies [23,24,25]. The proportion of patients receiving a diagnosis of hypercortisolism among the total cohort of patients evaluated by the centres are even lower (0.08% for Cushing’s syndrome and 0.12% for MACS).

According to the current guidelines, patients with specific features of hypercortisolism, like easy bruising, facial plethora, striae rubrae, myopathy or weakness in proximal muscles and unusual features for age, should all be screened for Cushing’s syndrome [1, 2, 16, 26]. The presence of these features, in combination with other signs or symptoms potentially related to hypercortisolism, even if less specific, increase the pre-test probability of Cushing’s syndrome, and should prompt a rapid screening for hypercortisolism. In particular, some specific characteristics associated with hypercortisolism has been recognized among hypertensive patients. In a recent consensus published by the working group on secondary hypertension of the ESH, the screening of Cushing’s syndrome should be performed in young patients (< 40 years) with at least grade 2 hypertension, children with hypertension, subjects with resistant hypertension independently of age, and patients with non-dipping blood pressure profile [16]. Moreover, the presence of hypokalemia should also prompt the screening for hypercortisolism. Furthermore, the analysis of white blood cells might also be helpful in the screening of Cushing’syndrome, according to recent data [27, 28]. Patients with adrenal incidentalomas should be screened all for MACS, independently of the presence of specific features of Cushing’s syndrome. MACS is not associated with catabolic signs of Cushing’s syndrome, by definition. However, hypertension is frequently associated with this disorder, with a pooled relative risk of 1.24 (95% confidence interval 1.16–1.32) in 31 studies, when MACS is defined by cortisol levels after 1 mg-DST > 1.8 µg/dL) [29].

According to the results of this survey, when the screening of hypercortisolism is planned, among the three screening tests for Cushing’s syndrome, 24 h-UFC is the most frequently used, followed by 1 mg DST and LNSC, the latter performed almost exclusively by endocrinologists. 24 h-UFC is being largely used for screening of Cushing’s syndrome, since it reflects the 24 h urinary output of free cortisol, with a sensitivity of 94% and a specificity of 93% [30]. Nevertheless, incomplete urine collection and impaired GFR may lead to false negative results. Therefore, confirmation by a repeated urinary free cortisol assay is suggested by the guidelines, even though intra-individual variations of cortisol excretion is an issue that should be taken into account when interpreting the results [1, 2, 16, 31]. The 1 mg DST has shown a high sensitivity (99%) with a specificity of 91% for diagnosis of Cushing’s syndrome [30]. Given the high sensitivity at the cortisol cut off of 1.8 µg/dL, the 1 mg DST is appropriate for screening purposes, after having excluded potential sources of bias (incomplete or lack of ingestion of dexamethasone by the patient, changes in cortisol binding globulin or albumin, and drugs altering the CYP3A4 activity) [31]. Collection of LNSC is a simple and non-invasive test that can be performed at home by the patients, with sensitivity and specificity for the diagnosis of Cushing’s syndrome > 90% [32]. The reliability of this test may be reduced by night shifts or altered sleep–wake cycles, blood contamination of the sample, and large intra-individual variability, therefore repeated sampling is suggested by the guidelines [1, 2]. Considering the performance and the pitfalls of each test, a larger use of LNSC is desirable as a valid method for improving the low screening rate of Cushing’s syndrome among hypertensive patients.

Although 24 h-UFC and LNSC are appropriate for screening Cushing’s syndrome, they are not valid methods for diagnosing MACS, because those tests may be often normal in this condition [2]. According to the most recent guidelines, MACS should be diagnosed by using 1 mg DST as a single test, by using a revised cut off for cortisol of 1.8 µg/dL, without further grading of severity based on additional hormonal tests [2].

It is noteworthy that an important proportion of centres use inappropriate tool for screening hypercortisolism, like morning ACTH and cortisol (roughly half of the centres) and CT or MRI scan (12% of responders). Baseline ACTH and cortisol should not be considered as screening tests for hypercortisolism, because ACTH may be within the normal range in pituitary Cushing’s syndrome and morning cortisol may not be elevated. Morning ACTH is a useful test for subtyping Cushing’s syndrome, once the diagnosis has been established by appropriate testing [1, 2]. The use of imaging as a screening tool rise the possibility to detect incidentalomas, which are very common and are not necessarily associated with hypercortisolism. In particular, adrenal incidentalomas are discovered in 3–4.2% of patients undergoing abdominal imaging and pituitary incidentalomas in 15–21 per 100.000 in the general population (16–36% of pituitary adenomas) [3, 33].

The results of the survey also highlight an inappropriate referral to proper centres of patients with hypercortisolism. Indeed, half of the non-excellence centres are not aware of a centre with expertise in Cushing’s syndrome (a situation dramatically different from the excellence centres, with 100% of awareness) and less than half of them would refer a patient with hypercortisolism to an endocrinologist. These data point toward an insufficient network of collaboration between non-excellence centres managing patients with hypertension and tertiary endocrinological centres, leading to a referral of patients with hypercortisolism that is insufficient (60% of the patients are not addressed to an expert endocrinological centre) and inappropriate (the remaining patients are referred to endocrinologists who are likely not expert in managing Cushing’s syndrome).

Therefore, the survey highlights an unsatisfactory screening rate for Cushing’s syndrome among hypertensive patients, which may explain in part the reason for the delay in the diagnosis of this condition [19]. The prompt recognition of Cushing’s syndrome is pivotal, to address the patient to a proper treatment to control the hormonal excess and manage hypertension, since the latter has been associated independently with increased mortality and duration and severity of hypercortisolism [5, 34,35,36]. Even though the association between hypertension and MACS is based mostly on retrospective and observational data, two randomized controlled trials have demonstrated that adrenalectomy may be effective in improving blood pressure in hypertensive patients with adrenal incidentalomas and MACS [15, 16]. Ongoing studies will provide more data on the efficacy of surgical (NCT02364089) [37] and medical treatment with steroidogenesis inhibitors (EudraCT: 2019–002008-41) [38] on hypertension in MACS (a full list of trials is available at clinicaltrials.gov).

The main limitation of this study is that the results were based on a self-reported survey, without a through revision of the data provided by each centre. However, this approach is appropriate to understand the current management of hypercortisolism among patients with hypertension on a large scale. Moreover, we did not investigate the referral rate to hypertensive centres from general practitioners. Indeed, the strength of this study are the large number of centres involved, spanning across the country, with a well-balanced representation of excellence and non-excellence centres.