Resistant HTN represents an important case finding feature for endocrine HTN that is often associated with obesity, sleep apnea, and reduced renal function among other conditions. With the current obesity epidemic, approaching an endocrine assessment of HTN can be difficult [47]. There are many (neuro)endocrine alterations in obese patients with and without sleep apnea, including an increase in leptin, cortisol, insulin, and aldosterone levels and a decrease in growth hormone, prolactin, and testosterone [48]. Among National Health and Nutrition Examination Survey (NHANES) participants with CKD, only 37 % had a BP < 130/80 mmHg, leaving 63 % with a BP higher than 130/80 mmHg. In ALLHAT, 34 % of patients taking at least two antihypertensive drugs continued to have uncontrolled HTN after 5 years of follow-up [49]. For patients with truly resistant HTN, the new recommendations include withdrawing medications that do not lower BP, and consideration of a mineralocorticoid antagonist, amiloride, or doxazosin as alternatives. Furthermore, renal denervation or baroreceptor stimulation should be considered if optimal drug therapy is deemed ineffective. These invasive approaches should be reserved for patients with clinic values > 160 mmHg SBP or > 110 mmHg and with BP elevation confirmed by ambulatory BP monitoring. In treating resistant hypertensive patients, renal nerve ablation is associated with a decrease in plasma noradrenaline but not in renin [50].

Of note, the exact prevalence of resistant HTN in patients with CKD remains unknown, but may reach to 30 % of all patients with uncontrolled HTN in the USA [51, 52]. Large-scale population-based studies such as the US NHANES suggest that 12 % cases of HTN are resistant to pharmacological therapy [53]. Importantly, apparent treatment resistance is often related to poor adherence to antihypertensive therapy, which is reflected by findings that about 40 % of patients with newly diagnosed HTN discontinue their treatment during the first year [52]. In all patients with an albumin excretion rate of > 30 mg/24 h, the use of an ACE inhibitor or ARB is recommended [54]. A recent review of the evidence on the role of dietary salt and potassium intake in cardiovascular health and disease revealed that modest salt restriction while increasing potassium intake serves as a strategy to prevent or control HTN and decrease cardiovascular morbidity and mortality [55].

CKD affects approximately 15 % of the adult population, and intensive BP lowering apparently can protect against kidney failure events in patients with CKD, especially among those with proteinuria [56]. A systematic review and meta-analysis suggest that a 10 mmHg reduction in SBP might result in an overall reduction of 22 % in the risk of kidney failure [56].

One important question in this regard is when to screen for secondary causes of HTN, including endocrine HTN . HTN in young patients and refractory HTN (characterized by poorly controlled BP on > 3 antihypertensive drugs) should alert the physician to screen for secondary causes. The clinician should carefully screen for cardinal signs and symptoms of CS, hyper- or hypothyroidism, acromegaly, or pheochromocytoma . The importance of endocrine-mediated HTN resides in the fact that, in most cases, the cause is clear and can be traced to the actions of a hormone, often produced in excess by a tumor such as an aldosteronoma in a patient with HTN due to PA . More importantly, once the diagnosis is made, a disease-specific targeted antihypertensive therapy can be implemented, and in some cases, surgical intervention may result in complete cure, obviating the need for lifelong antihypertensive treatment.

A detailed medical history and review of systems should be obtained . The onset of HTN and the response to previous antihypertensive treatment should be determined. Consideration should be given to compliance to prescribed antihypertensive regimen. A history of target organ damage (i.e., retinopathy, nephropathy, claudication, heart disease, abdominal or carotid artery disease) and the overall cardiovascular risk status should also be explored in detail .

As in other causes of HTN, the clinician should question the patient about dietary habits (salt intake, etc.); weight fluctuations (recording exact dry weights, etc., in patients with chronic renal disease); and use of over-the-counter drugs and health supplements, including teas and herbal preparations, recreational drugs, and oral contraceptives. Moreover, a detailed family history may provide valuable insights into familial forms of endocrine HTN. The review of systems should include disease-specific questions. Most patients with HTN due to pheochromocytoma are also symptomatic. Symptoms may include headaches, palpitations, anxiety-like attacks, and profuse sweating, similar to symptoms of hyperthyroidism. Pheochromocytoma, CS, and other endocrine conditions may represent secondary causes of diabetes mellitus [57] .

Patients with CS often complain of weight gain, insomnia, depression, easy bruising, and fatigue. Acne and hirsutism (in women) can also be observed. The challenge these days is to recognize patients with evolving CS among the many obese and often poorly controlled diabetic individuals. Primary hyperaldosteronism is manifested by mild to severe HTN [58–60]. Hypokalemia can be present but it is not a universal finding, and there is normokalemic PA . Polyuria, myopathy, myoglobinuric acute renal failure, and cardiac dysrhythmias may occur in cases of severe hypokalemia [61]. Hypokalemia is associated with kidney function impairment in patients with PA [62, 63], because chronic hypokalemia could induce chronic renal failure through characteristic tubulointerstitial damage consisting of vacuolization of epithelial tubular cells and interstitial fibrosis [64]. In the study of Takanobu and colleagues [65], however, preoperative serum potassium levels did not differ significantly among untreated PA patients. This result suggests that excretion of potassium could have already declined in patients with progressive renal damage, such as epidermal growth factor receptor (eGFR) < 60 mL/min/1.73 m² .

Many adrenal tumors are today being discovered incidentally during imaging procedures. The challenge for health-care providers is to identify those lesions that may be potentially harmful by either excessive hormone secretion, compressive effects on other structures adjacent to the adrenal glands, or spreading metastases . When caring for a patient with impaired renal function and an incidentally discovered adrenal tumor, screening for excessive hormone secretion from the tumor may be challenging, as many of our contemporary hormonal cutoff values, including stimulation and suppression tests, are based on patient populations with normal renal function. (Fig. 16.2) In both patient groups, those with and those without normal renal function, the transition from normal to subclinical and to finally full-blown clinical hormonal oversecretion often is difficult to recognize. When evaluating a patient with impaired renal function and an adrenal incidentaloma, one should attend to unexplained hypokalemia, typical cushingoid features, signs of hyperandrogenism, spells including headaches, palpitations, sweats, severe or refractory HTN, and familial syndromes that are associated with adrenal tumors (e.g., von Hippel–Lindau (VHL) syndrome, multiple endocrine neoplasias) .

In patients with a positive screening test, subsequent confirmation by various testing modalities is necessary. The most used investigations are listed in Table 16.1 .

PA has first been described in a Polish patient with HTN and hypokalemia who experienced normotension and normokalemia after removal of an adrenocortical adenoma [66]. In patients with HTN and adrenal incidentaloma, the median prevalence of PA is estimated to be 2 %. In patients with resistant HTN, the estimated prevalence of PA is between 17 and 23 % [67]. After applying the aldosterone-to-renin ratio (ARR) not only to hypokalemic but also normokalemic hypertensives, the diagnosis of PA increased up to 15-fold [68].

The ARR, commonly used to screen for PA, has also been viewed as an index for salt sensitivity and recently has been linked to the development of CKD in a longitudinal, observational study of 698 Japanese individuals. The mean follow-up in that study was 9 years, and those people with a higher ARR more frequently developed CKD [69]. In the German Diabetes and Dialysis Study, including 1255 diabetic hemodialysis patients, the combined presence of high plasma aldosterone > 200 pg/ml and high serum cortisol > 21 mcg/dl increased the risk of sudden cardiac death as well as all-cause mortality compared to patients with an aldosterone level less than 15 pg/ml and a cortisol value less than 16.8 mcg/dl [70].

Although PA has long been viewed as a relatively benign form of HTN [71, 72], recent studies suggest that long-term exposure to high aldosterone levels might lead to cardiovascular and renal structural damage that seems to occur independent of the BP level [73, 74]. Furthermore, significant histological damage to the kidney was noted in PA patients [63, 75].

Animal studies support the role of aldosterone in the progression of renal vascular disease [76–79]. In desoxycorticosterone acetate (DOCA)-salt rats, micropuncture studies have demonstrated glomerular hyperfiltration, reflecting increased glomerular blood flow consequent upon vasodilation of both afferent and efferent arterioles, and ascribed to volume expansion [80]. This renal vasodilation may be in large part via a direct mineralocorticoid effect on the vessel wall [81] and is in sharp contrast to the vasoconstrictor response to increased renal perfusion pressure in essential HTN. Studies in dogs indicate that an increase in renal artery pressure, which raises glomerular filtration rate and fractional sodium excretion, is essential in allowing the kidneys to escape from the chronic sodium-retaining action of aldosterone and to achieve sodium balance and a stable level of arterial pressure without severe volume expansion and ascites [82]. Preclinical studies in animal models clearly show that inappropriate aldosterone levels for sodium status can produce extensive renal damage [83].

Although preclinical studies indicate that aldosterone per se might cause important renal damage [76, 78], the clinical evidence supporting a direct role of this hormone as a potential contributor to renal dysfunction is limited.

Cross-sectional studies on renal function in PA have shown a high degree of variability in the prevalence of clinically relevant renal damage [84–86]. In fact, the majority of initial reports indicated that PA is less likely to cause overt renal damage [71, 72, 87, 88].

Two recent studies, a short-term [89] and a long-term follow-up [90], demonstrated that the renal dysfunction of PA is closely related with the hemodynamic adaptation of the kidney to the effect of aldosterone excess. In particular, the long-term follow-up study has shown that the condition is characterized by partially reversible renal dysfunction, in that glomerular hyperfiltration is corrected and urinary albumin excretion significantly reduced after either surgical or medical treatment of PA [89]. In agreement with the findings of studies that were conducted in more experimental settings [80–82], longitudinal studies consistently demonstrate that the hallmark of renal dysfunction in PA is reversible glomerular hyperfiltration that contributes to increase urinary albumin losses.

In a short-term study of adrenal adenoma patients after adrenalectomy, the evidence favors a major hemodynamic driver for the increased albuminuria of PA [89]. Patients with proteinuria have lower preoperative serum potassium levels than those without proteinuria [91].

This contrasts with the partial response in terms of BP and albuminuria seen in patients with higher pretreatment plasma renin levels, a possible marker for more severe renal structural damage [89].

Patients with an aldosteronoma are heterogeneous concerning mechanisms for impaired eGFR: those whose eGFR improved after adrenalectomy had lower preoperative plasma renin than those with decreased eGFR after the operation. In those patients with decreased eGFR after removal of the aldosteronoma, preoperative plasma renin was higher and not completely suppressed by the elevated aldosterone . This suggests the coexistence of a high aldosterone state due to the adenoma and an additional nonsuppressible renin state from other causes such as essential HTN. This nonsuppressible renin state would result in additional kidney damage that cannot be reversed by adrenalectomy [91, 92].

Correct interpretation of renal function in patients can be difficult using conventional eGFR before surgery, because in patients at an early stage, subtle kidney impairment might be masked by the glomerular hyperfiltration peculiar to PA preoperatively [93]. Recent studies have shown that many PA patients show a significant decline in eGFR within 6 months of follow-up without any further decrease or increase later on, because the glomerular hyperfiltration state has disappeared after treatment [62].

The occurrence of hypoaldosteronism after unilateral adrenalectomy can be a potential risk factor for postoperative development of CKD, because low aldosterone with improving HTN might decrease growth factor receptor (GFR) and lead to renal impairment [94].

Excessive production of aldosterone in PA patients has been noted to lead to a higher frequency of cardiovascular events, as compared with patients suffering from essential HTN [95]. Furthermore, CKD itself has been suggested to show a significant association with risks of death, cardiovascular events, and hospitalization [96] .

Patients with aldosteronoma and left ventricular hypertrophy had lower eGFR compared to those without left ventricular hypertrophy [91]. The dysfunction of both heart and kidneys may be more closely related to other mechanisms, such as generalized endothelial dysfunction and increased oxidative stress [97, 98]. Renal damage might be underestimated in PA patients preoperatively. In the study of Takanobu, all patients with manifested CKD showed a preoperative eGFR of 60–89 mL/min/1.73 m². Patients with preoperative eGFR of 90 mL/min/1.73 m² did not progress to an eGFR of 60 mL/min/1.73 m² because of higher baseline eGFR.

Utsumi et al. [99] found that the previous presence of dyslipidemia is an independent predictor for the postoperative development of CKD. Additional factors, such as older age, lower diastolic BP, and lower estimated GFR, also influence the development of CKD. In particular, aging is closely associated with declining renal function [100, 101]. Furthermore, older age might affect lower diastolic BP in patients with manifested CKD, as a result of the fall in diastolic BP after approximately 50 years of age [102] .

The ARR is useful to screen for PA in patients in whom there is an expected high prevalence of PA. Such patient groups are those with resistant HTN, HTN and adrenal incidentaloma, HTN grade 2 or 3, HTN and spontaneous or diuretic-induced hypokalemia, HTN and a family history or early-onset HTN or cerebrovascular accident younger than age 40, and all hypertensive first-degree relatives of patients with PA [67]. As there is no general consensus on the ARR cutoff, sensitivity and specificity vary widely. The ARR has good within-patient reproducibility and an accuracy of 80 % for identifying patients with an aldosterone-secreting adenoma [68].

Antihypertensive drugs are the most confounding factor affecting the measurement of aldosterone and renin. Especially, MR antagonists, such as spironolactone, eplerenone, and canrenone, should be discontinued 4–6 weeks prior to screening for PA and prior to adrenal vein sampling (AVS), because these agents can lead to an increase in renin secretion and subsequently aldosterone secretion from the unaffected side (if only one adrenal gland was oversecreting aldosterone). Only half of the patients (PA or essential HTN) can be studied drug-free or on medication with minimal influence on the ARR, with approximately 20 % of patients switching antihypertensive drugs experiencing an increase in BP and several patients suffering serious adverse events requiring hospitalization [103].

In 27 uremic patients on a chronic dialysis program, there was no correlation between BP and renin. After dialysis, renin activity rose significantly [104]. In such patients on chronic hemodialysis, the renin–angiotensin system apparently still functions regarding circulatory homeostasis with challenges by volume loss or loading, as demonstrated with 1.5–2 l of intravenous saline infusion that resulted in an increase in plasma volume by 0.4 l and BP by 10–15 mmHg, but in a decrease in plasma renin activity (PRA) by 40 % [105]. In hyperkalemic patients with chronic renal failure and mild azotemia, PRA and aldosterone levels usually are lower compared to patients with normokalemia [106]. In some patients with PA and end-stage renal disease (ESRD), aldosterone excess exists for many years, if hypokalemia is “masked” by normokalemia in the setting of chronic renal failure which may become unmasked after hemodialysis when hypokalemia due to PA can develop [107]. When interpreting the ARR, one should consider the specific patient population, as elderly and/or black patients often have low PRA values which can result in high ARR while plasma aldosterone levels are rather low. Also, black individuals are more sensitive to aldosterone regarding BP elevation than white people [108]. Most individuals with a plasma aldosterone concentration less than 9 ng/dl have normal suppression after administration of fludrocortisone [109]. When assessing the ARR, direct active renin is rarely measured and standardized/compared to the traditional plasma renin activity [110, 111]. Before measuring the ARR, certain factors should be considered, as mentioned in the Endocrine Society Practice guidelines [67]. Such factors include to correct hypokalemia, to collect blood midmorning, avoiding stasis and hemolysis, maintaining the sample at room temperature and not on ice, separating plasma from cells within 30 min of collection, considering age, the levels of potassium and creatinine, the intake of estrogen-containing medications, as estrogens can increase angiotensinogen synthesis by the liver, and considering other medications that can affect the ARR. Among drugs that have minimal effects on plasma aldosterone levels are the ones listed in Fig. 16.3.

It is recommended to withdraw these medications at least 4 weeks before screening: spironolactone, eplerenone, amiloride, triamterene, potassium-wasting diuretics, licorice-containing products. This clearly may represent a challenge in patients with chronic renal failure and HTN.

After screening for PA with the ARR, patients should typically undergo a confirmatory test for which various procedures exist, including oral sodium loading, saline infusion, fludrocortisone suppression, and a captopril challenge. Depending on the degree of chronic renal failure and HTN, such tests may not be safe to be performed in an individual patient. This becomes especially an issue when considering the lack of data in such patient groups. In patients without PA and normal or only slightly impaired renal function, plasma aldosterone usually suppresses to less than 6 ng/dl after 2 l of intravenous saline infusion. Assessing urinary aldosterone levels is also an option in patients with normal renal function who are suspected to have PA [67]. After oral sodium loading, the urinary aldosterone should be less than 10 mcg/day.

Imaging can be performed by computed tomography (CT) or magnetic resonance imaging (MRI). A typical aldosteronoma is shown in Fig. 16.4.

AVS should be performed in patients with PA who are considered for adrenalectomy, unless they are younger than 40 years with marked PA and have a clear unilateral adrenal adenoma, have an unacceptable high risk of adrenal surgery, are suspected to have adrenal cancer, or are proven to have familial hyperaldosteronism type 1 (glucocorticoid (GC) remediable) or type 3 (often due to germline mutations in the KCNJ5 potassium channel gene and treated with bilateral adrenalectomy). MR antagonists or amiloride should be discontinued 4–6 weeks before AVS, and AVS should best be performed in the morning, if cosyntropin stimulation is not used [112]. The selectivity index cutoff value (difference between affected and unaffected sides) should be greater than 2.0 under unstimulated conditions and greater than 3.0 during cosyntropin stimulation. In patients with PA and CKD stage III, stage IV, and stage V, AVS could be performed safely with acute kidney injury occurring in two patients [113] (see Fig. 16.5).

Deciphering the human genome has helped identify several disease-causing genes, including some being involved in PA, i.e., in the K-channel gene KCNJ5, encoding Kir3.4, a member of the inwardly rectifying K⁺ channel family [114] . The presence of these mutations in the KCNJ5 gene alter the K⁺ conductance/selectivity of this channel and consequently increase the Na⁺ conductivity (influx) with a further impact on voltage-gated calcium channels, leading to cellular proliferation in the adrenal cortex [115]. The recurrent somatic mutations (G151R, L168R) in the adrenal potassium channel KCNJ5 have been related to benign aldosterone-producing adenomas (APAs) with initial estimates, reporting almost half of APAs being associated with these mutations [115–118]. Subjects with the KCNJ5 G151E mutation have no features of APAs and hyperplasia, a different clinical course (not progressive), and excellent control of BP with spironolactone .

We now know of familial aldosteronism type 1, type 2, and type 3, although the precise molecular pathogenesis of aldosteronomas still needs to be elucidated [119–122]. FH−1 is caused by adrenocorticotropic hormone (ACTH)-dependent aldosterone secretion, GC remediable, and therefore treated with dexamethasone and/or MR antagonists, whereas FH−2 is non-GC remediable and indistinguishable from sporadic PA. FH−3 is resistant to pharmacotherapy and therefore treated with bilateral adrenalectomy. Given the heterogeneity of tumors not only between but also within one individual, it may be more prudent to analyze family members with a known gene defect and identical tumors in a whole genome- and exome-wide sequencing project for targeted genes known to be involved in cell growth regulation. Some of these patients with familial aldosteronism (overall less than 5 % of patients with PA) may require bilateral adrenalectomy to control their HTN which can occur at young age (sometimes in childhood). Exome sequencing of aldosterone-secreting adenomas has revealed somatic hotspot mutations in the ATP1A1 (encoding an Na(⁺)/K(⁺)ATPase alpha subunit) and ATP2B3 (encoding a Ca(2⁺)ATPase) genes [123, 124] in a small subset of tumors. Similarly, somatic mutations in CACNA1D, encoding a voltage-gated calcium channel, potentially causing increased calcium influx with subsequent aldosterone production and cell proliferation in adrenal glomerulosa cells, have been identified in less than 10 % of aldosteronomas [125] .

We here will not review the molecular pathogenesis of ACTH-secreting pituitary tumors, ectopic ACTH-secreting neuroendocrine tumors, or adrenal tumors. However, a few statements based on more recent findings should be made. In general, the molecular pathogenesis of ACTH-independent macronodular adrenal hyperplasia and sporadic cortisol-secreting adenomas has recently been elucidated. Exome sequencing or tumor–normal pairs revealed gain of function mutations in the CTNNB1 (β-catenin) or GNAS (Galphas) genes or somatic mutations in the PRKACA (protein kinase A catalytic subunit) gene with protein kinase activity inducing cortisol production and cell proliferation of affected adrenal cells [216]. Germline duplications of the PRKACA gene result in bilateral adrenal hyperplasia, whereas somatic mutations in this gene cause unilateral cortisol-producing adrenal adenomas [217]. A subset of ACTH-independent macronodular adrenal hyperplasia contains inactivating mutations in armadillo repeat containing 5 (ARMC5), a putative tumor-suppressor gene [218]. Regarding ACTH in the pituitary, most of such tumors occur sporadically with altered gene expression, somatic mutations in various genes, epigenetic changes, and abnormal microRNAs. Their pathogenesis remains largely unknown [219]. Whole exome sequencing of pituitary adenomas reveals different oncogenic mutations in each individual tumor, making it probable that there is no common oncogenic denominator but an abnormal stem cell that leads to abnormal localized proliferation. Nevertheless, there are familial pituitary tumor syndromes, including multiple endocrine neoplasia type 1, Carney complex, familial isolated pituitary adenomas, and others. The pathogenesis of ectopic ACTH-secreting neuroendocrine tumors is largely unknown, although some familial syndromes in which an ACTH-secreting neuroendocrine tumor is part of helped elucidate the biology of these neoplasms. For instance, neuroendocrine tumors of the pancreas in patients with VHL syndrome usually are nonfunctional but can secrete ACTH [220]. VHL syndrome is caused by germline mutations in the VHL gene and subsequent molecular events [221]. Neuroendocrine tumors of the thyroid usually are medullary thyroid cancer and often are caused by germline mutations in the rearranged during transfection (RET) proto-oncogene [222]. Many other organs can harbor ACTH-secreting neuroendocrine tumors, and for each organ, for instance, the prostate, pancreas, adrenal medulla/pheochromocytoma, one would have to consider the respective pathogenesis [223–225].

In patients with a cortisol-producing adrenal tumor, laparoscopic adrenalectomy should be performed to normalize cortisol production. If bilateral adrenalectomy is necessary, proper adrenal hormone replacement should be commenced. For instance, most individuals can be managed by taking oral hydrocortisone, 10 mg in the morning and 5–10 mg late afternoon, in addition to fludrocortisone 50–100 mcg/day [226]. Adrenal function should be assessed in patients with ESRD in order to unmask adrenal insufficient states [227].

In patients with an ACTH-secreting pituitary tumor, transsphenoidal or transcranial surgical removal should be performed, if possible. Postsurgically, transient or permanent adrenal insufficiency can occur and should be handled, as aforementioned. Long-term follow-up is necessary even in patients with initial and long-term remission [228]. In patients with nonpituitary ACTH-dependent CS, ideally the primary neuroendocrine tumor should be found and removed, although surgical resection of metastases might result in short-term remission of CS [60]. For patients who are not or poor surgical candidates, medical therapy of hypercortisolism should be initiated. Depending on country-specific regulations and availability of drugs, one can choose among the following arsenal with each one to be revisited for mild or more impaired renal and/or liver function with respective dose adjustments: metyrapone, ketoconazole, mifepristone RU486, cabergoline, pasireotide, octreotide, etomidate, and mitotane [195, 229–232].