Cortisol is secreted by the adrenal gland in a pulsatile manner in accordance with the pulsating secretion of ACTH. The secretion of cortisol displays diurnal variations, with the maximum levels (6.5 to 25.4 µg/dL [range, 180 to 700 nmol/L]) in early morning and minimum level (<3.6 µg/dL [100 nmol/L]) at midnight. Because the diurnal variations characteristic of normal persons are absent in patients with CD, testing by random collection of blood is not recommended [26]. Additionally, it is recommended that diurnal cortisol levels be checked more than twice.

The 24-hour urine free cortisol (UFC) test is a highly sensitive method of testing [2,27]. If the 24-hour UFC level exceeds a normal range, it is defined as a positive finding. The normal ranges are as follows: >90 µg/dL per 24 hours as measured by a radioimmunoassay or >50 µg/dL per 24 hours as measured by denaturing high-performance liquid chromatography/immunochemiluminometric assay. The 24-hour urine creatinine levels and urine volume should also be checked to assess the adequacy of urine collection [28,29].

It is recommended that this test be performed at least 2 to 3 times to minimize diagnostic errors, and further interpretation of the results may also be necessary in some situations. For example, if the glomerular filtration rate of the patient is <60 mL/min, then the amount of free cortisol in the urine may be reduced. Additionally, excessive amounts of cortisol can saturate the circulating cortisol-binding globulin (CBG) moieties and result in detectable increases in free-filtered cortisol in the urine. When conducted on hospitalized patients, the sensitivity and specificity of the 24-hour UFC test are better than those of the 1-mg overnight DMST test [27].

Under physiological conditions, serum cortisol concentrations reach a nadir at midnight or late at night. Salivary cortisol levels display a very high correlation with serum cortisol levels, which makes this measure a convenient method for the measurement of cortisol concentrations without inflicting stress on outpatients. For these tests, saliva should be collected during sleep or between the hours of 23:00 and 24:00, and it is recommended that the levels of salivary cortisol be checked more than twice. If the concentration of salivary cortisol at room temperature exceeds the cutoff value of 0.13 µg/dL (3.6 nmol/L) [30,31], it is defined as a positive finding. Both the sensitivity and specificity of this test are approximately 95% [26]. A "sleeping" midnight serum cortisol level of <1.8 µg/dL (50 nmol/L) effectively excludes a diagnosis of CS, while an "awake" midnight serum cortisol level of >7.5 µg/dL (207 nmol/L) is highly suggestive of CS [2,30]. However, the late night salivary cortisol test is not typically available in Korea.

The overnight DMST is the most widely used screening test for CS and CD [31,32]. In this test, 1 mg dexamethasone is administered orally between the hours of 23:00 and 24:00, and blood is collected between the hours of 08:00 and 09:00 on the following day to measure serum cortisol levels. If the serum cortisol concentration is >1.8 µg/dL (50 nmol/L), it is defined as a positive finding. In the case of obese patients and some patients suffering from depression and/or alcohol withdrawal syndrome, further attention is needed, because cortisol is not being suppressed in these situations.

The desmopressin stimulation test [33,34] depends upon the increase in ACTH release that occurs in 85% of patients with CD following intravenous injection of desmopressin. In this test, patients are intravenously administered 4 µg desmopressin, and blood is collected over a period of up to 2 hours at 30-minute intervals. If the serum ACTH level increases by >50%, the finding is defined as positive. This test is useful for determining ACTH levels following surgery for CD, rather than for the discrimination of CD, because the V3 receptor that binds desmopressin is found in 30% of tumors outside the pituitary gland that secrete ACTH.

The sella magnetic resonance imaging (MRI) is an essential radiological tool for the diagnosis of CD necessary to assess the anatomical structure of the pituitary gland prior to surgery and to confirm whether pituitary adenomas are present. Compared with IPSS or a positron emission tomography scan, a sella MRI scan has the highest level of accuracy for tumor localization, with a positive predictive value of 86% [35]. In studies involving Korean patients with CD, concordant results between MRI scans and intraoperative surgical biopsies were observed in 77.8% of cases [36].

However, CD is frequently caused by adenomas that are too small to be detected by classic MRI scans. For example, more than 18% of patients with CD have microadenomas that are not visible on MRI scans [37]. Because the contrast enhancement of pituitary adenomas can be gradual or rapid after contrast administration, the use of dynamic contrast-enhanced MRI scans are recommended rather than classic MRI scans, which are obtaining only T1- and T2-weighted images before and after contrast administration. For cases in which the tumor is located in the far anterior aspect or far posterior aspect of the pituitary gland, the use of dynamic gadolinium-enhanced imaging with simultaneous acquisition of coronal and sagittal planes is helpful for the localization of a pituitary adenoma [38]. If pituitary adenomas are not detected using the aforementioned MRI procedures, half-dose contrast MRI scans or spoiled gradient recalled (SPGR) acquisition associated with the steady-state technique could be helpful. When MRI is executed using a lower injection dose of the contrast medium (0.05 mmol/kg gadolinium, for example), it becomes easier to distinguish microadenoma from normal pituitary tissue. [39]. Additionally, the use of SPGR MRI images with 1-mm sections facilitates detection of microadenomas, compared with spin echo MRI scans [40,41].

In the majority of CD patients, secretion of ACTH is promoted by CRH and desmopressin [3,4,5,6,7,8,9]. The CRH stimulation test can be used to confirm the responses of serum ACTH and cortisol levels after intravenous injection of 100 µg CRH in the morning after fasting; however, a definitive cutoff value for this test to distinguish between CD and ectopic ACTH secretion (EAS) has yet to be established. Moreover, the responses of serum ACTH and cortisol are different between ovine and human CRH [9]. Nevertheless, increases in the serum ACTH and cortisol levels >50% within 30 minutes of CRH administration could suggest the presence of CD [9]. In Korea, the CRH stimulation test is not usually employed due to its high cost; typically, desmopressin is used to stimulate the secretion of ACTH. However, the cutoff values for the desmopressin stimulation tests that can provide an accurate differential diagnosis have yet to be determined.

IPSS is the most important test for the differentiation of CD from EAS in patients diagnosed with CS [42,43]. According to a study that assessed Korean patients with CD using IPSS, 100% of patients tested positively for CD [36]. IPSS is a method that compares ACTH levels by collecting blood from peripheral vessels and both sides of the inferior petrosal sinus (IPS) after promotion of ACTH secretion via either desmopressin or CRH administration; the interpretation of the findings does not differ with regard to the stimulator used [10]. When desmopressin is used as the stimulator, the optimal injection dose is 10 µg [11] and is administered intravenously over a period of 15 seconds [10]. When CRH is used as the stimulator, the optimal intravenous injection dose is 100 µg [44].

Sampling should be executed from the bilateral IPS and peripheral vessels prior to and several times within the 10-minutes period following administration of the ACTH stimulator. CD is suggested by IPS/peripheral (IPS/P) ACTH ratios >2 prior to administration of the stimulator or by ratios >3 following administration of the stimulator. An ACTH ratio from both sides of the petrosal sinus of >1.4 following administration of the stimulator may imply lateralization of the ACTH-secreting tumors in the pituitary gland [45]. However, the primary purpose of IPSS is to distinguish between CD and EAS, because the presence of lateralization during this procedure is concordant with intraoperative biopsies in only 38.9% of Korean patients with CD [36]. Additionally, it is possible to obtain false negative results due to venous anomalies of the petrosal sinus or to the inexperience of the personnel performing the procedure. In such cases, it is possible to confirm whether or not the catheterization was successful by using prolactin values [46]; IPSS results that show an increase in prolactin levels >1.8-fold relative to the peripheral prolactin value indicate successful catheterization. A normalized ratio >1.3 ([dominant peak post-stimulation IPS/P ACTH]/[ipsilateral basal IPS/P prolactin]) implies the presence of CD, while a value <0.8 implies EAS [47].

An 8-mg overnight DMST is very useful for the differential diagnosis of CD from EAS. Relative to the traditional high-dose DMST method, which requires the administration of dexamethasone over a 2-day period, the 8-mg overnight DMST is widely used due to its convenient drug administration and lack of requirement of a 24-hour urine test. Additionally, the overnight DMST has a sensitivity of 95%, which was reported to be superior to that of the traditional high-dose DMST test [12,13]. In the overnight DMST, an 8-mg dose of dexamethasone is administered between 23:00 and 24:00, and the blood cortisol level is measured at 8:00 on the following day. CD is inferred when serum cortisol levels are decreased by >50% from basal levels or if the serum cortisol level is <5 µg/dL [12,13,14,15].

The transsphenoidal approach (TSA) is the primary treatment recommended for patients with CD, because it is associated with a complete remission rate of 65 to 90% for microadenomas, although the rate of remission for macroadenomas is lower than 65% [16]. While transsphenoidal microsurgery is the most frequently employed surgical method for CD patients, endoscopic surgery has also been implemented fairly widely in recent years. A head-to-head comparison of the prognoses for these two surgical methods has yet to be conducted, but no major differences in their surgical outcomes have been reported [17]. A total or partial hypophysectomy can be performed if the microadenomas are not easily visible, but this procedure has a relatively low remission rate (70%) as well as high frequencies of hypopituitarism and surgical complications [18].

The recurrence rate of microadenomas after TSA is 5 to 10% after 5 years and 10 to 20% after 10 years. Patients younger than 25 years were reported to experience recurrence more frequently [18,19]. Compared with microadenomas, macroadenomas have a higher recurrence rate (12 to 45%) and an earlier recurrence point (49 months vs. 16 months, respectively) [20]. However, there are a number of favorable prognostic factors related to surgery, including the fact that microadenomas are detectable on an MRI, the non-invasion of the dura mater or cavernous sinus, the confirmation of ACTH-secreting adenomas in pathological tissue, low serum cortisol levels after surgery, and a prolonged period of hypocortisolism [16].

A majority of CD patients require supplementary glucocorticoid therapy (optimal dose; 12 to 15 mg/m²) following surgery to correct hypocortisolism. Although the dosage can be increased if the hypocortisolism is severe, it should be reduced during the first month following surgery to a supplementary dose if possible. This supplementary glucocorticoid administration can be terminated if the morning serum cortisol level or stimulated serum cortisol is >18 µg/dL [16].

Radiation-based therapies include the use of fractionated radiotherapy and stereotactic radiotherapies, such as Gamma Knife surgery (GKS), CyberKnife, and proton-beam therapy.

The standard assessments for remission after surgery are as follows [16]: (1) if the morning serum cortisol levels the 1st week after surgery are <2 µg/dL, then CD is considered to be in remission; if the levels of serum cortisol are >5 µg/dL for up to 6 weeks after surgery, then additional tests should be performed; and if the levels of serum cortisol are 2 to 5 µg/dL, then close monitoring is necessary due to the high probability of remission; (2) UFC levels should be measured if the results of the serum cortisol analyses are ambiguous, and if the 24-hour UFC level is <20 µg, then CD is in a state of remission; if the 24-hour UFC level is in the range of 20 to 100 µg (normal range), the result is ambiguous; and if the 24-hour UFC level is above the normal range, CD is persistent; and (3) because serum cortisol levels sometimes decrease gradually, it is important to verify whether cortisol levels are at a minimum prior to initiation of any additional treatment.

In the event of recurrent or residual CD after surgery, secondary treatments such as repeat TSA, medical therapy, radiotherapy, or bilateral adrenalectomy can be implemented [16]. Although repeat surgery is associated with a high risk of hypopituitarism, repeat TSA has success rates of 50 to 70% and these rates improve if the microadenomas are visible. It is more advantageous to implement surgery for remnant adenomas as early as possible if the patient is in an active disease stage, but the surgical procedure could be delayed by approximately 4 to 6 weeks, since some patients improve gradually after surgery. Although a bilateral adrenalectomy induces the most rapid improvements in hypercortisolism among surgical methods, it can also induce permanent hypocortisolism and necessitate steroid therapy for the rest of a patient's life. Additionally, serum ACTH assessments and MRI tracking must accompany the bilateral adrenalectomy due to the risk of Nelson's syndrome. Therefore, this may be considered for patients who have recurrent CD, for whom surgical, medical, and radiation-based therapies have failed, or for those who are intolerant of the side effects of these treatment modalities.

Various complications, such as risks within the cardiovascular system, hypercoagulability, bone density decline, psychiatric problems, and cognitive impairments, are only partially ameliorated in patients following CD treatment [21]. The cardiovascular risks are greater in a CD patient than in a normal person even 5 years after a complete remission, and the extent of recovery from these complications is inversely proportional to the duration of exposure to hypercortisolism. Furthermore, the patient's quality of life continues to deteriorate even after remission, and this trend is worse for patients with hypopituitarism.