Key points are not available for this paper at this time.
Primary aldosteronism (PA) is emerging as the most common form of secondary hypertension, with a prevalence of ~5–10% among hypertensive patients.1,2 The prevalence of aldosterone-producing adenomas (APAs) among patients with PA is ~35–50% in those centers where adrenal venous sampling is expertly performed for the differentiation of APA from idiopathic hyperaldosteronism. Therefore, it is expected that more and more patients with APAs will be detected and surgically cured in the near future. Hence, optimal perioperative management of patients with Conn's syndrome is highly desirable. Postoperative selective hypoaldosteronism with hyperkalemia is a rare, but well-known complication, and monitoring of serum potassium levels after surgery is currently recommended.2 On the other hand, postoperative hypocortisolism is not currently regarded as a noteworthy complication. Indeed, the coexistence of PA and autonomous hypercortisolism from adrenal adenomas producing both aldosterone and cortisol has been reported anecdotally in western countries. A 62-year-old man was referred to our unit on December 2006 for resistant hypertension. He had been discovered with hypertension and hypokalemia at 33 years of age. In the past 10 years, he had been treated with spironolactone 100 mg od, amlodipine 10 mg od, and ramipril 5 mg od. At presentation, physical examination was normal. In particular, no signs of Cushing's syndrome were observed. The patient's blood pressure was 178/102 mm Hg, and body mass index was 23.3 kg/m2. Cardiac ultrasound examination revealed left ventricular hypertrophy (143 g/m2). Laboratory investigations performed 6 weeks after withdrawal of spironolactone showed a combination of PA and adrenocorticotropic hormone (ACTH)-independent hypercortisolism. Indeed, plasma aldosterone was markedly elevated and plasma renin activity relatively suppressed both at baseline and after captopril administration, resulting in a high aldosterone-to-plasma renin activity ratio. Intravenous saline infusion did not suppress plasma aldosterone level. Urinary-free cortisol excretion was threefold the upper limit of the normal range, plasma ACTH at 8:00 AM was suppressed, while plasma cortisol was within the normal range. However, overnight low-dose dexamethasone failed to suppress plasma cortisol level (Table 1). Preoperative and postoperative laboratory data Preoperative and postoperative laboratory data Computed tomography scan of the abdomen revealed a hypodense mass in the right adrenal, 4.5 × 4.5 × 4.0 cm in diameter, and a 5.6 × 5.1 × 4.8 cm mass at the lower pole of the right kidney (Figure 1). Adrenal venous sampling was not performed, in view of the large size of the right adrenal mass, possibly consistent with a malignant lesion, which indicated right adrenalectomy anyway. Moreover, both the long-lasting (~30 years) hypokalemic hypertension and the very high values of both aldosterone and aldosterone-to-renin ratio were unlikely to result from an occult aldosterone-producing microadenoma in the controlateral, radiologically normal adrenal. Right nephrectomy with right adrenalectomy was performed by laparoscopic surgery. Hydrocortisone 100 mg was intravenously administered in the first 2 days after operation, and then discontinued. The weight of the adrenal tumor was 41 g, and its cut surface was golden yellow. The tumor was mainly composed of large clear cells, with scattered areas of compact cells. The histological diagnosis was adrenocortical adenoma (Figure 2). The tumor at the lower pole of the right kidney, 5.6 cm in greatest diameter, was confined within the renal capsule, without evidence of perirenal invasion. Pathological diagnosis was clear cell renal cell carcinoma, Fuhrman nuclear grade 3 (Figure 2). Ten days after surgery, the patient complained of weakness, anorexia, nausea, abdominal pain, postural dizziness. His blood pressure was 120/90 mm Hg without any antihypertensive drug. Laboratory examinations performed 3 weeks after surgery showed hyponatremia, a slight decrease in renal glomerular filtration rate, and complete disappearance of the biochemical findings of primary hyperaldosteronism. Concurrently, levels of plasma cortisol and ACTH were substantially lower and higher, respectively, than those found preoperatively (Table 1). Intravenous dexamethasone 4 mg/day was administered for 5 days, with complete reversal of the symptoms of adrenal insufficiency and normalization of serum sodium concentration. From day 16 to 28 postoperatively, the patient received oral dexamethasone 0.5 mg od, and from day 29 to day 35 oral dexamethasone 0.25 mg od, followed by discontinuation of the replacement therapy. Ten days later, the patient again complained symptoms of adrenal insufficiency. His blood pressure was 105/70 mm Hg. Laboratory investigations showed extremely low urinary-free cortisol excretion, and a blunted cortisol response to ACTH stimulation (Table 1). Replacement treatment with dexamethasone 0.25 mg od was resumed with complete reversal of the symptoms. At the last clinical examination, 9 months postoperatively, the patient was found to be normotensive while on amlodipine 5 mg od and dexamethasone 0.25 mg od. Plasma renin activity was measured by radioimmunoassay for generated angiotensin I with a commercial kit (Radim, Pomezia, Italy), after a 3-h incubation at pH 5.5 at 37 °C. The intraassay and interassay coefficients of variation were <6 and <9%, respectively; the reference range was 0.20–0.80 ng/ml/h supine and 1.50–5.70 ng/ml/h upright. Aldosterone was measured by radioimmunoassay (CisBio international, Gif-Sur-Yvette, Cedex, France). Both intraassay and interassay coefficients of variation were <8%; the reference range was 2.0–12.0 ng/dl supine and 4.0–32.0 ng/dl upright. Cortisol and ACTH were measured by LIAISON chemiluminescence immunoassay (DiaSorin, Saluggia, Italy). The intraassay and interassay coefficients of variation were <5 and < 6% for cortisol, and <5% and <9% for ACTH. The reference range of plasma cortisol was 45– 250 ng/ml in the morning and 15– 70 ng/ml in the evening, 10– 85 μg/24 h for urinary-free cortisol, and 5–50 pg/ml for ACTH. PA due to APA and Cushing's syndrome due to cortisol-producing adrenal adenoma (CPA) are usually regarded as distinct conditions, on the basis of the specificity of both the clinical findings and the steroidogenic profile in APAs and CPAs. However, production and concentration of cortisol in APAs are similar to those found in the normal adrenal cortex.3 Even the amount and activity of 17α-hydroxylase and 11β-hydroxylase, two enzymes involved in cortisol synthesis, are not different from those found in the normal cortex.4 Moreover, APAs express mRNAs for both CYP17 and CYP11B1 (ref. 6). Notwithstanding the steroidogenic activities of APAs, biochemical screening for Cushing's syndrome is not recommended as a routine component of the diagnostic evaluation of patients with Conn's syndrome. On the other hand, at least 33 other cases of concurrent autonomous secretion of both aldosterone and cortisol from aldosterone cortisol–producing adenomas (ACPAs) have been reported so far.6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22 Higher expression of mRNA for 17α-hydroxylase has been found in ACPAs compared to APAs.22 Furthermore, immunohistochemical analysis of ACPAs has shown that aldosterone synthase on the one hand, and both enzymes specifically involved in cortisol biosynthesis (i.e., 17α-hydroxylase and 11β-hydroxylase) on the other hand, are localized in two different cell types; this finding suggests the existence of two separate cell populations with different steroidogenic pathways in ACPAs. Particularly, aldosterone synthase is expressed in the prevailing clear cells, while both 17α-hydroxylase and 11β-hydroxylase are expressed in the scattered compact cells.16,17 Accordingly, it is conceivable that in our patient's adrenal adenoma, aldosterone and cortisol were synthesized by the prevailing clear cells and the compact eosinophilic cells, respectively (Figure 2a). ACPAs are on the average larger than APAs, with a mean diameter of 2.7 cm22 compared to that of 1.7 cm for APAs.2 Moreover, no microadenoma (with a diameter < 1 cm) with mixed secretion has been reported so far. Because APAs have the steroidogenic potential to produce cortisol, we hypothesize that APAs could evolve into ACPAs, when the population of cortisol-producing cells exceeds a certain threshold. Overt Cushing's syndrome was present only in 6 of the 33 cases of ACPAs reported so far,7,8,10,11,14 while subclinical Cushing's syndrome was present in the remaining cases. Furthermore, 26 of the 33 cases with mixed syndrome have been reported by Japanese authors, which could reflect either a higher incidence or a higher awareness of the possible coexistence of hypercortisolism and Conn's syndrome in Japan than in other Countries. Thus, the true prevalence of this mixed syndrome is likely to be underestimated, because subclinical Cushing's syndrome is not routinely excluded in patients with PA. Overlooking subclinical Cushing's syndrome in patients with Conn's syndrome may result in unrecognized adrenal insufficiency after surgery. Indeed, acute adrenal insufficiency after unilateral adrenalectomy for PA due to unilateral ACPA was observed both in the present case and in other cases previously reported.6,9,12,16 Therefore, preoperative recognition of subclinical hypercortisolism may prevent postoperative adrenal insufficiency through appropriate glucocorticoid replacement therapy. Furthermore, suppression of cortisol production in the adrenal controlateral to ACPA may impair the diagnostic accuracy of adrenal venous sampling, which is currently regarded as the gold standard in differentiating unilateral from bilateral subtypes of PA. Indeed, cortisol concentration in the controlateral adrenal vein may be much lower than that in the vein draining the ACPA.13,17 Thus, recognition of coexistent hypercortisolism may help prevent misinterpretation of the hormonal concentrations in the adrenal venous blood samples. In the present case, ACPA was combined with an ipsilateral renal cell carcinoma. The coexistence of functional adrenal adenoma and renal cell carcinoma could be co-incidental, because only three other cases of aldosterone-producing adenomas associated with homo- or controlateral renal cancer have been reported so far.23 However, in some autopsy series, the prevalence of nonfunctioning adrenal adenomas in patients with renal cancer was two- to fourfold higher than in patients who died of diseases other than cancer.24 On the other hand, abdominal computed tomography scan in patients with renal cell carcinoma did not reveal an increased incidence of adrenal masses compared to the general population.25 Therefore, an increased frequency of adrenal adenomas in patients with renal cell carcinoma has not been firmly established. In conclusion, the real prevalence of subclinic hypercortisolism in patients with PA due to adrenal adenomas is probably underestimated, because an assessment of the pituitary–adrenal axis is not commonly performed in patients with PA. Because unilateral and autonomous production of cortisol may result in controlateral adrenal suppression, patients with ACPAs are at risk of developing postoperative adrenal failure. Therefore, screening for autonomous hypercortisolism should be recommended in patients with Conn's syndrome, particularly in cases of adrenal tumors greater than 1 cm. The authors declared no conflict of interest.
Rossi et al. (Thu,) studied this question.