Fig. 1
Left adrenal incidentaloma with subtle autonomous cortisol secretion in a 37-year-old woman (white arrow); the right adrenal gland is partially atrophied (black arrow). Over 1 year, the cortisol after 1 mg dexamethasone rose from 1.4 to 2.1 μg/dL, the AM ACTH fell from 14 to 2 pg/mL, and the DHEAS fell from 126 to 41 μg/dL. Improvement in weight and blood pressure was noted after laparoscopic adrenalectomy and several weeks of partial cortisol deficiency
Some investigators explored the idea of assessing the adrenal function by tracking the incorporation of radiolabeled cholesterol derivatives within the gland. Using this principle, scintigraphic uptake exclusively to an adrenal adenoma indicates autonomous cortisol production, while symmetrical incorporation of the tracer supports an ACTH-responsive cortisol synthesis. As an example, Valli and colleagues used [131I]-6β-iodomethyl norcholesterol scintigraphy (IMS) in 31 patients with benign cortical adenomas and found that the sensitivity and specificity of the test in detecting MACE was 58 % and 83 %, respectively, if referenced to a dexamethasone-suppressed cortisol of 5 μg/dL (138 nmol/L) and 100 % and 67 %, respectively, for a dexamethasone-suppressed cortisol of 2.2 μg/dL (60 nmol/L) [36]. Barzon and colleagues obtained similar results with [75Se]-selenio-6α-methyl-19-norcholesterol [49]. These studies, however, are limited by burdensome protocols, scarce availability of the tracers, and high cost. Furthermore, because in contrast to primary aldosteronism , adrenal cortisol excess correlates closely with the size of the adenomas, there is little value of the scintigraphic studies over routine hormonal tests and cross-sectional imaging [50].
Pathogenesis of Dysregulated Adrenal Cortisol Synthesis
The molecular pathogenesis of adrenal Cushing syndrome is covered in detail in another chapter. To summarize, Assie and colleagues identified inactivating mutations of armadillo repeat containing 5 (ARMC5) in 18 of 33 patients with macronodular adrenal hyperplasia and hypercortisolism, 3 of which presented with MACE [51]. ARMC5 mutations have been found in patients with both familial and sporadic macronodular adrenal hyperplasia with a range of hypercortisolism [52–54]. Beuschlein and colleagues found somatic mutations in PRKACA, which encodes the main catalytic subunit of protein kinase A (PKA), in cortisol-producing adenomas associated with overt Cushing syndrome but not in 40 patients with MACE [55]. Germline PRKACA duplications were identified in 14 % of patients with Cushing syndrome due to bilateral adrenal hyperplasia, and three other groups reported PRKACA mutations in 35–69 % of cortisol-producing adrenal adenomas [56–58], primarily mutation L206R, which leads to constitutive PKA activation [57, 58]. Of all the patients with PRKACA mutations across these four studies, only four patients had MACE [57, 58]. Thus, ARMC5 and PRKACA mutations are found in some patients with MACE but more often in those with overt hypercortisolism.
Beyond the genetic and epigenetic aspects contributing to excessive ACTH-independent adrenal cortisol synthesis, several additional factors modulate the effects of excessive glucocorticoids in the target tissues. These include the cortisol binding globulins, tissue-specific glucocorticoid activating and inactivating enzymes, and glucocorticoid receptor (NR3C1) polymorphisms. Recent studies have identified polymorphisms in the 11β-hydroxysteroid dehydrogenase type 1 (11βHSD1) and glucocorticoid receptor genes that are protective against a Cushingoid phenotype, including cognitive impairment [59] and diabetes [60]. 11βHSD1 knockout mice with circulating glucocorticoid excess were protected from the glucose intolerance, hyperinsulinemia, hepatic steatosis, hypertension, myopathy, and dermal atrophy of Cushing syndrome [61]. The intricate interplay between various factors that constitute the basis of a specific phenotype remains elusive and deserves to be dissected further.
Clinical Consequences of MACE
Cardiometabolic Profile and Related Outcomes in MACE
Research conducted over a decade ago found that surrogates of cardiovascular risk, including blood pressure, fasting glucose, homeostatic assessment model-insulin resistance (HOMA-IR) index, lipoproteins and triglycerides, fibrinogen, waist-to-hip ratio, and mean carotid artery intima-media thickness, were significantly worse in patients with MACE than in age-, sex-, and body mass index (BMI)-matched controls [62]. More recent studies have linked mild hypercortisolism with cardiometabolic morbidity and mortality. In a first large cross-sectional study, Di Dalmazi and colleagues stratified patients with AI and hypercortisolism in an intermediate group, with a cortisol after dexamethasone between 1.8 and 5 μg/dL, or >5 μg/dL, respectively. They found that the prevalence of type 2 diabetes mellitus and coronary heart disease increased in parallel with progressively higher degrees of hypercortisolism, as compared with patients with nonfunctioning adrenal adenomas [13]. The same group longitudinally followed a cohort of 198 patients with AI (mean follow-up, 7.5±3.2 years), and they found that the incidence of cardiovascular events and related mortality was higher in patients with subclinical hypercortisolism (cortisol >1.8 μg/dL) [9]. Worsening hypercortisolism during follow-up was independently associated with cardiovascular events and mortality.
Another Italian multicenter study retrospectively analyzed the outcomes of 206 patients with AI followed for a median of 6 years. Of these, 11.6 % patients were classified to have subclinical hypercortisolism, based on a cortisol after dexamethasone >5 μg/dL, or at least two other indicators of altered HPA axis (low ACTH, increased urinary free cortisol, and cortisol >3 μg/dL after dexamethasone) [15]. Subclinical hypercortisolism was associated with a higher incidence of cardiovascular events and worsening of at least two metabolic parameters (weight, glycemic, lipid, and blood pressure control), independent of age. Debono and colleagues retrospectively studied a similar size cohort of patients with AI followed for 4.2±2.3 years in the UK [12]. During the time interval studied, 18/206 patients died, and of these, 17 patients had a cortisol >1.8 μg/dL after dexamethasone. Mortality was higher in patients with a cortisol after dexamethasone >5 μg/dL vs. 1.8–5 μg/dL, and half of the deaths were attributed to cardiovascular causes. The mean time to death was 3.2 years, and the age of death was lower than the life expectancy at birth for the general population in the same area. Taken together, these studies strongly suggest that chronic hypercortisolism is a direct contributor to cardiovascular events and related mortality even when subtle, and that the impact directly increases with the degree of hypercortisolism.
Metabolic Bone Disease
The deleterious effects on bone metabolism of overt glucocorticoid excess, both endogenous and exogenous, have been widely documented [63]. Evidence that mild hypercortisolism leads to osteoporosis and fragility fractures emerged predominantly from Italian cohorts [13, 14, 34, 64–67]. In a cross-sectional study of 219 patients evaluated for osteoporosis without any known secondary causes, subclinical hypercortisolism was present in 5 % of patients and in 10 % of the subset who also had vertebral fractures [64]. Similarly, a 2-year longitudinal study of 103 consecutive patients with AI documented a higher incidence of vertebral fractures in patients with MACE [14]. Patients with MACE experienced worsening of their spinal deformity index, independent of age, gender, BMI, bone mineral density, baseline spinal deformity index (SDI ), and menopause duration. In another cohort including 287 patients with AI, both bone mineral density and bone quality, as measured by the SDI, were significantly worse in patients with MACE [34]. The trabecular bone score, another index of bone quality, was found to be worse amongst patients with AI who had MACE, and this parameter was proposed to be a useful predictor of fractures [68].
Pathophysiology of MACE
Pathogenic Mechanisms of MACE on Cardiovascular and Glucose Metabolism
While solid evidence exists to explain the deleterious effects of overt cortisol excess, the pathogenic mechanisms derived from mild chronic hypercortisolism remain speculative. The most commonly entertained hypothesis is that even subtle cortisol excess over time has cumulative effects, leading to clinical consequences similar to overt Cushing syndrome, but at a smaller scale. This hypothesis is supported by evidence of an incremental effect of cortisol on cardiovascular events and mortality [9, 12, 13]. In addition, an increased vascular mortality rate has been observed in patients with primary adrenal insufficiency on various glucocorticoid replacement regimens and has been attributed to chronic overtreatment [69]. Metabolic components associated with an increased cardiovascular risk, such as high blood pressure, fasting glucose and insulin, cholesterol, fibrinogen and waist to hip ration, are common to mild and overt hypercortisolism [62]. Cortisol-induced visceral adiposity might explain both the increased insulin resistance and cardiovascular risk in these patients. A retrospective study of 125 patients with AI conducted by Debono and team found that patients with a cortisol >1.8 μg/dL following dexamethasone had significantly more visceral fat than those with nonsecretory adenomas [70]. Furthermore, there was no difference in visceral fat between patients with subclinical and overt hypercortisolism, although only nine women in the latter group were included.
Beyond patients with MACE, data have emerged to support that even apparently nonfunctioning adrenal adenomas are associated with increased cardiovascular risk. In 2009, Yener and colleagues proposed that the increased carotid intima-media thickness was a consequence of insulin resistance associated with subtle cortisol autonomy [71]. The same group later suggested that impaired arterial flow-mediated dilatation and elevated IL-18 might underlie the endothelial alterations in patients with adrenal adenomas and early cortisol autonomy [72]. To eliminate the confounding effects of comorbidities associated with increased cardiovascular risk frequently present in hypercortisolism, Androulakis and colleagues studied a group of 60 normotensive and normoglycemic patients with apparently nonfunctioning AI and 32 healthy controls [73]. Besides absence of clinically overt Cushing syndrome, patients were enrolled if they had normal early morning basal serum ACTH and cortisol levels, preservation of ACTH and cortisol circadian rhythm, and normal 24-h urinary free cortisol excretion. Subsequently, a group of 26 patients was classified as cortisol-secreting, based on low-dose 2-day DST greater than 1.09 μg/dL, cutoff derived from the mean +2 SD values of the control group. The authors found that carotid intima-media thickness measurements were higher and that flow-mediated dilatation was lower in the cortisol-secreting group compared with both nonfunctioning and control groups. In addition, they found that intima-media thickness correlated with cortisol, urinary free cortisol, and cortisol after dexamethasone. The authors concluded that this disproportionate cortisol secretion might potentially lead to microvasculature damage [73].
Another hypothesis to explain some of the cardiovascular profile and outcomes in patients with MACE is that before the cortisol excess becomes apparent, other alterations in steroidogenesis and HPA axis might occur. In support of this hypothesis stand two lines of evidence: (1) cardiovascular risk factors appear to be increased in patients with so-called “nonfunctionin g” adrenal adenomas [71, 73–77], terminology that only excludes cortisol, aldosterone, and catecholamines excess; (2) cortisol secretion typically becomes apparent in large adenomas [15], suggesting that intrinsic enzymatic alterations in the steroid biosynthesis within the tumor might lead to an atypical steroid profile prior to the development of clinical manifestations. Using LC-MS/MS, Di Dalmazi and colleagues have recently measured ten steroids, both at baseline and after cosyntropin stimulation, in patients with adrenal adenomas (66 nonfunctional and 28 subclinical hypercortisolism) and in 188 age- and sex-matched controls [48]. Basal and cosyntropin-stimulated DHEA, androstenedione and, in women, basal testosterone concentrations were lower in patients with MACE than in those with non-secreting adenomas and controls. Increased cortisol and reduced DHEA levels were independently associated with increased waist circumference. Cortisol, but not androstenedione, was independently associated with increased number of cardiovascular risk factors in patients with MACE. Patients with MACE also demonstrated increased production of 21-deoxycortisol and the mineralocorticoid 11-deoxycorticosterone after cosyntropin stimulation. In addition, the ratio between 17α-hydroxyprogesterone and androstenedione was higher in the MACE than in nonfunctioning adenomas group, suggesting alterations in P450c17 and P450c21 activities. A second hypothesis postulated by the authors was that the cortisol excess secreted from an adenoma suppresses ACTH, and this in turn leads to decreased adrenal androgen synthesis from the remaining adrenal tissue. However, although a positive correlation with ACTH was noted for both DHEA and androstenedione, DHEA was also reduced in patients with nonfunctioning adenomas, despite normal ACTH levels. Further studies to assess the common and unusual steroids synthesized both in vivo and in vitro are needed as an initial step; subsequently, it would be important to establish the function of steroids other than cortisol and their links with clinical outcomes.
Pathogenic Mechanisms of MACE on Bone Metabolism
Cortisol alters bone metabolism by decreasing bone formation and increasing bone resorption [78, 79]. The magnitude at which cortisol excess starts to affect bone metabolism and the relationship between time and degree of cortisol excess remain unclear. Tauchmanova and colleagues assessed the bone density and vertebral fractures in 71 consecutive women with either overt (n = 36) or subclinical (n = 35) hypercortisolism and corresponding controls [65]. Interestingly, bone mineral density and prevalence of any vertebral fractures did not differ between women with overt and subclinical hypercortisolism, defined by a cortisol after dexamethasone >3 μg/dL. Di Dalmazi et al. found that osteoporosis was independently associated with subclinical hypercortisolism, as defined by a cortisol after dexamethasone >5 μg/dL [13], while an intermediate group, with a cortisol after dexamethasone of 1.8–5 μg/dL, was no different than the non-AI group.
The interrelation between sex steroids and cortisol on bone metabolism has been explored in both men and women [65, 67]. In women with MACE, eugonadism was partially protective, but this effect was lost in patients with overt Cushing syndrome [65]. MACE was associated with low bone mineral density and high prevalence of vertebral fractures in eugonadal men [67]; however, a direct comparison with hypogonadal men has not been done. A more intriguing aspect is the role of DHEAS in bone health. Beyond cortisol itself, Tauchmanova et al. found that the cortisol/DHEAS ratio was a predictor of fractures in all patients [65], but to what degree this association is reflective of the cortisol excess alone remains unclear. Other authors have suggested a benefit of DHEAS on bone density. Studies investigating the association between DHEAS and bone mineral density in postmenopausal women have found conflicting results [80–84]. In placebo-controlled studies of DHEA administration in elderly men and women, DHEA was found to have a positive effect on bone mineral density only in women in one study [85] and in both sexes in another [86].
Management of MACE
So far, the evidence to guide appropriate treatment of subclinical hypercortisolism has been modest. The few studies that have looked at management of MACE have used different diagnostic criteria and have enrolled small numbers of patients. In addition, most studies are retrospective and prone to selection bias. As an example, surgery could have been more frequently offered to patients with subtle comorbidities typically associated to hypercortisolism. One prospective study randomized 45 patients with MACE to either laparoscopic surgery (n = 23) or observation (n = 24) and followed them for a mean of 7.7 years [87]. In the surgical group, diabetes mellitus, hypertension, hyperlipidemia, and obesity normalized or improved in 62.5 %, 67 %, 37.5 %, and 50 %, respectively. In contrast, some worsening of diabetes, hypertension, and hyperlipidemia was noted in conservatively managed patients. Similar outcomes were reported by smaller retrospective studies [88–93]. Surgery has also been proposed for patients with MACE and bilateral adrenal nodules, by selectively removing the gland with the largest nodules. In a retrospective study of 33 patients with bilateral AI and MACE followed for up to 4.5 years, Perogamvros and colleagues found that markers of HPA axis dysregulation were significantly improved in the 14 patients who underwent unilateral adrenalectomy [94]. In addition, comorbidities associated with hypercortisolism, such as hypertension, impaired glucose tolerance or diabetes mellitus, dyslipidemia, and osteoporosis, improved in the surgical group, while no changes were noted in the observational group [94]. A recent systematic review of outcomes of adrenalectomy for MACE concluded that, compared with conservative management, surgery cured or improved blood pressure, glucometabolic control, and obesity in 72, 46, and 39 % of patients, respectively [95]. The main limitations to this analysis were the heterogeneity of diagnosis and outcomes followed and the retrospective nature of all but one of the studies included. Furthermore, the interventions in the nonsurgical grou ps were often poorly defined, and no studies have evaluated the outcomes of MACE in patients with intensive comorbidity-specific medical therapy.
Medical treatment for MACE has not been much assessed. One open-label pilot study observed a reduction in insulin AUC in 4/6 patients with MACE treated with the glucocorticoid receptor antagonist mifepristone for 4 weeks [96]. Other strategies to decrease cortisol synthesis have been tried in adrenal tumors with aberrant receptor expression, such as glucose-dependent insulinotropic peptide, catecholamine, serotonin, vasopressin, angiotensin II, leptin, and luteinizing hormone/chorionic gonadotropin receptors [97–102]. Examples of such successful therapies include somatostatin analogs [99, 100], propranolol [101], and leuprolide acetate [102]. Inhibitors of cortisol synthetic enzymes, such as ketoconazole, metyrapone, or LCI699 (osilodrostat), have not yet been formally studied in MACE. While medical treatment for overt Cushing syndrome is reserved for inoperable cases, emerging medical therapies with a favorable safety profile might offer a safe and effective alternative to surgery in MACE, especially if low doses could successfully inhibit the hormone synthesis in these already inefficient adenomas.
Conclusion
It is clear that cortisol excess spans a spectrum of severities and, not surprisingly, establishing rigid lines to define clinically important disease becomes an unrealistic task. Solid evidence has emerged that even mild hypercortisolism has important clinical consequences, including deleterious effects on cardiovascular risk, glucose, lipid and bone metabolism, and even survival. However, numerous aspects remain to be clarified in order to best guide clinical practice for MACE. Beyond the mere association of MACE and unfavorable outcomes, the responsible mechanisms remain speculative. Do other steroid precursors produced by apparently nonfunctioning and/or cortisol-producing adrenal adenomas have direct clinical impact, either by activating nuclear hormone receptors or by different mechanisms? Is it reasonable to conclude that surgery should be offered to all patients with MACE? If not, how should we best follow these patients and when should we recommend treatment? A future research agenda aiming to answer some of these questions should include prospective studies of large cohorts, to characterize detailed steroid profiles and autonomy in these patients and to assess clinical outcomes in three distinctive arms: surgical treatment, steroid synthesis or action blockade, and intensive comorbidity-specific interventions. Until then, defining clinically important hypercortisolism and appropriate management remain rather arbitrary, and decisions must be individualized empirically.
References
1.
2.
3.
Mantero F, Terzolo M, Arnaldi G, Osella G, Masini AM, Ali A, et al. A survey on adrenal incidentaloma in Italy. Study Group on Adrenal Tumors of the Italian Society of Endocrinology. J Clin Endocrinol Metab. 2000;85(2):637–44.PubMed
4.
Kim J, Bae KH, Choi YK, Jeong JY, Park KG, Kim JG, et al. Clinical characteristics for 348 patients with adrenal incidentaloma. Endocrinol Metab (Seoul). 2013;28(1):20–5.
5.
6.
7.
8.
Barzon L, Scaroni C, Sonino N, Fallo F, Paoletta A, Boscaro M. Risk factors and long-term follow-up of adrenal incidentalomas. J Clin Endocrinol Metab. 1999;84(2):520–6.PubMed
9.
Di Dalmazi G, Vicennati V, Garelli S, Casadio E, Rinaldi E, Giampalma E, et al. Cardiovascular events and mortality in patients with adrenal incidentalomas that are either non-secreting or associated with intermediate phenotype or subclinical Cushing’s syndrome: a 15-year retrospective study. Lancet Diabetes Endocrinol. 2014;2(5):396–405.CrossRefPubMed
10.
11.
Di Dalmazi G, Pasquali R, Beuschlein F, Reincke M. Subclinical hypercortisolism: a state, a syndrome, or a disease? Eur J Endocrinol. 2015;173(4):M61–71.CrossRefPubMed
Related posts:
and Treatment of Aggressive Corticotroph Pituitary Tumors
Regulation of Body Composition and Metabolism
Regulation of Neurocognitive and Neuropsychiatric Function
Metabolism as a Regulator of the Tissue-Specific Glucocorticoid Action
Therapies in Cushing’s Syndrome
and Nonneoplastic/Physiological Hypercortisolism: Cushing Versus Pseudo-Cushing Syndromes
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
