| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Studies |
Endocrinology Division, Department of Medicine and Department of Physiology, Faculty of Medicine, 14049–900 Ribeirao Preto, Sao Paulo, Brazil
Address all correspondence and requests for reprints to: Ayrton C. Moreira, M.D., Department of Medicine, Faculty of Medicine, 14049–900 Ribeirao Preto, Sao Paulo, Brazil.
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Twenty-four-hour urine collection for the measurement of UFC has been considered a gold standard for the diagnosis of CS (1, 2, 7, 8). This method has a sensitivity of 95% (5, 9) and usually provides a clear-cut distinction between patients with hypercortisolism and normal controls. However, in addition to the disadvantage of urine collection at out-patient and pediatric clinics (2, 5, 9, 10), in 4% of obese patients the cortisol production rate may be increased, leading to erroneous diagnosis of CS (9). The basis for the DEX suppression test lies in the deranged feedback relationship between ACTH and cortisol in CS. The diagnostic accuracy of the DEX suppression test may be compromised by incomplete 24-h urine collection, daily fluctuations in basal plasma cortisol, or failure to take the drug, mainly by out-patients submitted to a standard 2-day test (2, 3). Finally, although the morning plasma cortisol level may be normal in CS, the diurnal pattern of cortisol secretion is lost, as nocturnal levels are usually increased (11, 12). However, circadian rhythm is one of the most difficult tests of the hypothalamo-pituitary-adrenal axis to perform properly. The test requires hospitalization, the sample should not be obtained the first night in the hospital, the patient should not be told a blood sample will be collected, and its accuracy has been disappointing (3, 5).
In this context, measurement of cortisol in saliva may be helpful. Cortisol concentrations in saliva are independent of flow rate and transcortin fluctuations and reflect those in the free fraction of plasma (13, 14). In addition, salivary samples are obtained by noninvasive stress-free procedures, are easier to collect, even at home, and can obviate many problems such as situations where skilled personnel are not available (15). Salivary samples may be collected many times a day and may provide information of great diagnostic significance in the evaluation of CS (16). Recently, Papanicolaou et al. (17) and Raff et al. (18) reported that late night salivary cortisol determination is a sensitive marker (93% and 92%) for the diagnosis of CS in patients with mild hypercortisolism. Because of the continuing need for improved noninvasive means of distinguishing pseudo-Cushing from CS, the aim of this study was to assess salivary cortisol as a tool in screening for CS by evaluating hypercortisolism on the basis of circadian rhythm and the overnight 1-mg DEX suppression test for out-patients.
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
We investigated in our department (between 1994–1997), after written informed consent was obtained, 30 normal nonobese healthy volunteers (13 men and 17 women; mean age ± SD, 28.3 ± 6.6 yr), 18 obese patients including some women with conditions resembling CS features such as essential hypertension and idiopathic hirsutism or menstrual disorders (4 men and 14 women; age, 23.1 ± 10.2 yr), and 33 patients with CS (2 men and 31 women; age, 25.5 ± 11.1 yr). The mean body mass index values were 23.5 ± 2.8, 39.5 ± 8.4, and 29.3 ± 6.4 kg/m2, respectively, in the above groups.
Patients with both CS and obesity were referred to us with suspected clinical features of hypercortisolism. The diagnosis of CS was established by the lack of circadian variation in plasma cortisol levels and the lack of a fall in the plasma cortisol response to low dose (2 mg/day for 2 days) DEX administration. Of the 33 patients with CS, 20 were found to have Cushing’s disease, 10 had cortisol-secreting adrenal tumors, 2 had primary pigmented nodular adrenocortical dysplasia, and 1 had adrenocortical nodular hyperplasia based on standard tests, including determination of plasma ACTH levels by RIA, high dose DEX suppression, and ovine CRH (oCRH) tests (19, 20). Plasma ACTH levels (mean ± SE) were 112 ± 26, 17 ± 3, 24 ± 0.9, and 11 pg/mL, respectively, in the above groups. In patients with pituitary-dependent CS, the percent suppression (mean ± SE) of plasma cortisol levels after the high dose DEX suppression test was 36 ± 8%, and there was an increase in plasma cortisol and ACTH after the oCRH test (20). In 13 patients with adrenal CS, all failed to suppress cortisol plasma levels in the high dose DEX test, and there was no increase in plasma cortisol or ACTH after oCRH treatment. Finally, the etiology of CS was confirmed by image studies, curative surgery, or positive tissue pathology.
Salivary cortisol assessment at 0900, 1700, and 2300 h
Saliva samples (1 mL) were collected in a plastic tube by direct spitting during a 15-min period at 0900, 1700, and 2300 h at the out-patient clinic or at the patient’s home (21). The oral cavity was previously cleaned with tap water. The saliva samples were stored at 4 C until the following morning, when they were brought to the laboratory. After centrifugation at 2000 rpm, the transparent supernatants were stored at -20 C until assayed.
Overnight 1-mg DEX cortisol suppression test
One milligram of DEX was administered orally at 2300 h. The next morning, a post-DEX sample was taken for assay of salivary cortisol between 0800–0900 h.
Assays
Salivary cortisol measurements were made using a previously described RIA method in 25-µL samples of saliva without prior extraction or chromatography. This method previously demonstrated a good correlation (r = 0.95) with plasma free cortisol determined by equilibrium dialysis (22). The assay sensitivity was 62 ng/dL. The intra- and interassay coefficients of variation (CVs) were 5.5% and 11%, respectively. The assay showed parallelism between the standard curves and those obtained with different dilutions of saliva with high cortisol values. All samples obtained from each subject were analyzed in duplicate in the same assay.
Statistical analysis
A normal circadian pattern of salivary cortisol was defined as one in which both afternoon and nighttime salivary values were less than 83.5% of the morning level for each subject. This value was the result of the subtraction of 3 times the mean intraassay CV (16.5%) from the morning value, which was taken as 100% (15). We also used a second analysis of salivary cortisol fluctuations in circadian characterization by the method of Krieger et al. (23). In addition, for group analysis purposes, all individual data were combined. The mean afternoon and mean nighttime salivary cortisol values must be less than 67% of the mean morning value, which is taken as 100%. This percentage was the result of the subtraction of 3 times the interassay CV (33%).
The sensitivity and specificity for salivary cortisol were determined at different times and at various suppression levels. Levels of salivary cortisol above the 90th percentile of the normal control and obese patient groups were defined as cut-off points to calculate the sensitivity and the specificity of the test in distinguishing CS. These cut-off points cannot be considered absolute because they are defined by our dataset and may vary somewhat when applied to a new patient population.
Data are expressed as the mean ± SEM. Results that were below the detection limit of the assay were defined as the value of the detection limit. Friedman’s ANOVA was performed for multiple comparisons. The Wilcoxon-Mann-Whitney test was used when appropriate. Significance was assumed when P < 0.05.
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Figure 1
shows individual salivary
cortisol values obtained at 0900, 1700, and 2300 h and after the
1-mg DEX suppression test for normal controls, obese subjects, and CS
patients. The mean (±SEM) values in each group are
presented in Table 1. ANOVA demonstrated
significant differences among the 0900, 1700, and 2300 h salivary
cortisol levels for the normal control (P = 0.0001),
obese (P = 0.0002), and CS (P = 0.02)
groups. Circadian pattern of cortisol was completely characterized in
the normal control and obese groups, but not in the CS group.
Individual analysis showed that in contrast to 97% of the normal
subjects and 73% of the obese subjects, 43% (13 of 30) CS patients
had a persistence of a diurnal variation in salivary cortisol, but the
levels were set abnormally high. The individual daily salivary cortisol
variation estimated by intraassay CV-based analysis was the same for
all subjects, except two, compared to that obtained by the method of
Krieger et al. (23). The salivary cortisol levels of
patients with CS were significantly elevated at 0900, 1700, and
2300 h compared to those of obese patients (P =
0.001, P = 0.002, and P = 0.001,
respectively) or normal subjects (P = 0.0001,
P = 0.0001, and P = 0.0001,
respectively). In contrast, the 0900, 1700, and 2300 h salivary
cortisol levels of obese patients did not differ from those of normal
volunteers. Although mean salivary cortisol levels were higher for CS
patients, there was a marked overlap between groups at 0900 and
1700 h. The 2300 h salivary cortisol levels were undetectable
in 60% of normal controls (18 of 30) and 38% of obese patients (7 of
18). The highest levels detected at this time were 232 and 392 ng/dL
for normal controls and obese patients, respectively. The range of
2300 h values found in Cushing’s syndrome was almost entirely
outside the range of values found in the normal and obese groups. The
overlap was due to 5 patients (15%) in the CS group.
|
|
Salivary cortisol levels in CS were significantly elevated after the 1-mg DEX test compared to those in obese or normal groups (P = 0.0001). In contrast, there was no difference in salivary cortisol levels between obese and normal volunteers. In 28 of 30 control individuals, salivary cortisol was undetectable after the 1-mg DEX test, and the 2 normal subjects presented low detectable values (76 and 92 ng/dL). On the other hand, the post-1-mg DEX salivary cortisol level was detectable in every patient with CS. There was no overlap between normal and CS ranges. However, there was overlap between obesity and CS. Two of 18 obese patients exceeded a value of 232 ng/dL, the lowest found in the CS group.
Sensitivity and specificity
In the present series, a single out-patient 0900, 1700, and
2300 h measurement and overnight 1-mg DEX salivary cortisol levels
above the 90th percentile of the control values had sensitivities and
specificities of 65.6%/87.8%, 81.8%/86.6%, 100%/87.8%, and
100%/100%, respectively (Table 2). When
we used salivary cortisol above the 90th percentile of the obese
patients’ values, the sensitivities and specificities were
78.1%/88.8%, 57.6%/88.8%, 93.3%/93.3%, and 91.4%/94.4%,
respectively, at 0900, 1700, and 2300 h and after 1 mg DEX. The
sensitivity improved (91.4% to 100%) in response to the combination
of 2300 h and overnight 1-mg DEX salivary cortisol tests to
differentiate between obese and CS patients.
|
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The absence of a diurnal rhythm has been considered a hallmark of the diagnosis of CS (11, 12). However, circadian rhythm in patients with Cushing’s disease has actually been demonstrated in some patients (25, 26). In this study, 43% of the CS patients presented diurnal variation of salivary cortisol assessed at 1700 and 2300 h, but the levels were detectable and abnormally high in all cases. Studies of spot or timed UFC between 2200–2300 h (7, 10) or a single sleeping midnight plasma cortisol measurement (27) have demonstrated that at 2300 h or midnight there was no overlap between normal subjects and CS. However, in these two studies, the control group was not ideal and would preferably have been formed from patients, particularly women with conditions resembling CS, such as obesity with essential hypertension and idiopathic hirsutism and polycystic ovarian syndrome. Recently, it was demonstrated that a single serum cortisol value above 7.5 µg/dL at midnight discriminated CS from pseudo-CS patients with 93% sensitivity and 100% specificity (28). We reported salivary cortisol levels obtained at 1700 and 2300 h for 33 patients with CS compared to values in normal and obese groups. Whereas there was no difference in salivary cortisol evaluated at 1700 and 2300 h in the CS group, salivary cortisol was statistically higher at 1700 than at 2300 h in both the normal and obese groups. These data mean that the discrimination between CS and normal or obese groups improves in the late evening. In addition, the range of values found in CS patients at 2300 h was almost entirely outside the range of values found for the normal and obese groups. However, there was an overlap due to five patients with CS. A cut-off point for salivary cortisol above the 90th percentile of the normal subjects at 2300 h (168 ng/dL) produced a sensitivity of 100% and a specificity of 88%. This sensitivity is comparable to those reported in two recent studies using a single sleeping midnight plasma cortisol determination in patients hospitalized for a period of at least 48 h (27, 28). However, in routine practice, saliva can be obtained anywhere and at any time; therefore, this test is easily applicable at out-patient clinics. Our study further showed that the cut-off point of salivary cortisol above the 90th percentile (280 ng/dL) of the obese group at 2300 h produced a sensitivity of 93.3% and a specificity of 93.3%, demonstrating that the ideal control group should be patients with pseudo-CS states. This cut-off point of 280 ng/dL obtained from our data is similar to that recently reported (29), showing that the midnight salivary cortisol cut-off point of 270 ng/dL also detected 13 of 14 children with CS.
The overnight 1-mg DEX suppression test (30) is simple and reliable to segregate Cushingoid disorders from CS on an out-patient basis (1, 30). However, attempts have been made to improve diagnostic accuracy or simplicity (31). In the present study, we tested whether using salivary instead of plasma cortisol determination may simplify the 1-mg DEX suppression test. In fact, using as a criterion a cut-off point for salivary cortisol above the 90th percentile of the normal control (62 ng/dL) and obese (392 ng/dL) groups, the test had sensitivities and specificities of 100%/100% and 91.4%/94.4%, respectively. Based on our data, the overnight 1-mg DEX suppression test had a similar sensitivity and specificity as the 2300 h salivary cortisol determination.
We also evaluated the performance of the combination of both 2300 h collection and the overnight 1-mg DEX suppression salivary cortisol test to establish an easier way to differentiate CS from obese patients. Our data indicated that when the 2300 h collection and the overnight 1-mg DEX suppression salivary cortisol tests were combined, the sensitivity (100%) was higher than that of either test performed individually. Raff et al. (18) reported similar improvement in sensitivity combining an elevated late night salivary cortisol and an elevated 24-h UFC. The improvement in the sensitivity of the combined test may be a result of reducing laboratory error through an additional measurement. Flack et al. (8) estimated that at cut-off points for maximal sensitivity at 100%, specificity can vary by 4% only because of the daily variation in baseline cortisol secretion. In addition, our combined evaluation is based on two different abnormalities of the hypothalamo-pituitary-adrenal axis in CS: the altered circadian rhythm and the resistance to corticosteroid feed-back.
Our data indicate that salivary cortisol determinations in a sample collected at bedtime (2300 h) and in the overnight 1-mg DEX suppression test are sensitive markers for the diagnosis of CS. In addition, the combination of the two tests improves the sensitivity to differentiate between obese and CS patients and may be useful for out-patient evaluation. Therefore, we recommend the association of both tests, because the collection of saliva rather than blood is a practical, simple, and noninvasive out-patient alternative to screen for hypercortisolism.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Received July 31, 1998.
Revised November 2, 1998.
Accepted December 8, 1998.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
-hydroxyprogesterone in management of congenital adrenal
hyperplasia. J Pediatr Endocrinol. 4:255–262.
This article has been cited by other articles:
 |
|
 |
|
  G. Carneiro, S. M. Togeiro, L. F. Hayashi, F. F. Ribeiro-Filho, A. B. Ribeiro, S. Tufik, and M. T. Zanella Effect of continuous positive airway pressure therapy on hypothalamic-pituitary-adrenal axis function and 24-h blood pressure profile in obese men with obstructive sleep apnea syndrome Am J Physiol Endocrinol Metab, August 1, 2008; 295(2): E380 - E384. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. K. Nieman, B. M. K. Biller, J. W. Findling, J. Newell-Price, M. O. Savage, P. M. Stewart, and V. M. Montori The Diagnosis of Cushing's Syndrome: An Endocrine Society Clinical Practice Guideline J. Clin. Endocrinol. Metab., May 1, 2008; 93(5): 1526 - 1540. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Kidambi, H. Raff, and J. W Findling Limitations of nocturnal salivary cortisol and urine free cortisol in the diagnosis of mild Cushing's syndrome Eur. J. Endocrinol., December 1, 2007; 157(6): 725 - 731. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. H. Golden, S. Malhotra, G. S. Wand, F. L. Brancati, D. Ford, and K. Horton Adrenal Gland Volume and Dexamethasone-Suppressed Cortisol Correlate with Total Daily Salivary Cortisol in African-American Women J. Clin. Endocrinol. Metab., April 1, 2007; 92(4): 1358 - 1363. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. W. Findling and H. Raff Cushing's Syndrome: Important Issues in Diagnosis and Management J. Clin. Endocrinol. Metab., October 1, 2006; 91(10): 3746 - 3753. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. Reimondo, B. Allasino, S. Bovio, P. Paccotti, A. Angeli, and M. Terzolo Evaluation of the effectiveness of midnight serum cortisol in the diagnostic procedures for Cushing's syndrome Eur. J. Endocrinol., December 1, 2005; 153(6): 803 - 809. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. S. Chriguer, L. L. K. Elias, I. M. da Silva Jr., J. G. H. Vieira, A. C. Moreira, and M. de Castro Glucocorticoid Sensitivity in Young Healthy Individuals: in Vitro and in Vivo Studies J. Clin. Endocrinol. Metab., November 1, 2005; 90(11): 5978 - 5984. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Viardot, P. Huber, J. J. Puder, H. Zulewski, U. Keller, and B. Muller Reproducibility of Nighttime Salivary Cortisol and Its Use in the Diagnosis of Hypercortisolism Compared with Urinary Free Cortisol and Overnight Dexamethasone Suppression Test J. Clin. Endocrinol. Metab., October 1, 2005; 90(10): 5730 - 5736. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Yaneva, H. Mosnier-Pudar, M.-A. Dugue, S. Grabar, Y. Fulla, and X. Bertagna Midnight Salivary Cortisol for the Initial Diagnosis of Cushing's Syndrome of Various Causes J. Clin. Endocrinol. Metab., July 1, 2004; 89(7): 3345 - 3351. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Putignano, P. Toja, A. Dubini, F. P. Giraldi, S. M. Corsello, and F. Cavagnini Midnight Salivary Cortisol Versus Urinary Free and Midnight Serum Cortisol as Screening Tests for Cushing's Syndrome J. Clin. Endocrinol. Metab., September 1, 2003; 88(9): 4153 - 4157. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Raff and J. W. Findling A Physiologic Approach to Diagnosis of the Cushing Syndrome Ann Intern Med, June 17, 2003; 138(12): 980 - 991. [Full Text] [PDF]
|
|
|
|
 |
|
 S. L. King and K. M. Hegadoren Stress Hormones: How Do They Measure Up? Biol Res Nurs, October 1, 2002; 4(2): 92 - 103. [Abstract] [PDF]
|
|
|
|
|
|
D. A. Papanicolaou, N. Mullen, I. Kyrou, and L. K. Nieman Nighttime Salivary Cortisol: A Useful Test for the Diagnosis of Cushing's Syndrome J. Clin. Endocrinol. Metab., October 1, 2002; 87(10): 4515 - 4521. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Moro, P. Putignano, M. Losa, C. Invitti, C. Maraschini, and F. Cavagnini The Desmopressin Test in the Differential Diagnosis between Cushing's Disease and Pseudo-Cushing States J. Clin. Endocrinol. Metab., October 1, 2000; 85(10): 3569 - 3574. [Abstract] [Full Text]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
), 18 patients with obesity (
), and
33 patients with CS (
). The mean is represented by the
horizontal bar.