| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Clinic for Endocrinology, Diabetes, and Clinical Nutrition (A.V., J.J.P., H.Z., U.K., B.M.) and Department of Central Laboratories (P.H.), University Hospital Basel, 4031 Basel, Switzerland
Address all correspondence and requests for reprints to: Alexander Viardot, M.D., Diabetes and Obesity Research Program, Garvan Institute of Medical Research, 384 Victoria Street, Sydney-Darlinghurst, New South Wales 2010, Australia. E-mail: alex{at}viardot.com.
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Results: NSC showed a superior reproducibility in healthy volunteers with a low day-to-day variability as reflected by an intraclass correlation coefficient of 0.78. The receiver operating characteristic curve-estimated cutoff of 6.1 nmol/liter (0.22 µg/dl) demonstrated a sensitivity and specificity of 100% (area under the receiver operating characteristic curve, 1.0; 95% confidence interval, 0.94–1.0) in the diagnosis of CS. NSC, 24-h UFC [after adjusting the local laboratory cutoff to 504 nmol/d (183 µg/d)], and the urinary cortisol/creatinine ratio showed a tendency to be superior to 1 mg dexamethasone suppression test in correctly identifying CS. In late pregnancy, the preserved diurnal variation at a higher level of salivary cortisol reduced the specificity of NSC to 75%.
Conclusion: Based on its remarkable reproducibility, easy noninvasive nature, and at least similar diagnostic performance, NSC appears to be a preferable alternative to 24-h UFC as a first-line screening test for CS. The cutoff values of NSC, 24-h UFC, and urinary cortisol/creatinine ratio have to be carefully adjusted using assay and center-specific reference ranges of sufficiently large populations.
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The 24-h urinary free cortisol (UFC) excretion has traditionally been used to screen for CS. However, improper collection technique, insufficient analytical specificity of the introduced immunoassays and, especially, the different and assay-specific cutoff values are drawbacks of this test (15, 16).
The 1 mg overnight dexamethasone suppression test (DST) is widely used, is relatively simple to perform, and has a reported high sensitivity of up to 98% but a less-than-optimal specificity [
70% at cutoff value of 50 nmol/liter (1.8 µg/dl)] (11, 16, 17). Furthermore, in patients receiving drugs inducing cytochrome P450-related enzymes or in patients with renal or hepatic failure (18), DST is prone to give either false-negative or false-positive results, respectively.
ACTH and cortisol are secreted in a circadian rhythm, reaching the highest levels in the morning and the lowest levels at approximately midnight. The loss of this diurnal rhythm distinguishes CS (19) from pseudo-CS and pregnancy, respectively. Accordingly, a single midnight serum cortisol was reported to have a sensitivity and specificity of up to 100% for diagnosing CS (20, 21). However, the procedures involved are cumbersome because standardized, stress-free blood sampling requires hospitalization and the placement of an iv line.
Salivary cortisol reflects the unbound biologically active form of serum cortisol, is therefore not influenced by alterations of protein binding, and represents about 3–6% of the total serum cortisol concentration. Its concentration is not influenced by the salivary flow rate, and the equilibrium is quickly reestablished after changes in serum cortisol levels (22, 23, 24, 25, 26). Advantages of salivary cortisol tests are the easy and noninvasive collection procedure and its stability at room temperature for at least 7 d (27), which is very convenient in an outpatient setting.
Nighttime salivary cortisol (NSC) measurement for the diagnosis of CS has been shown to be a useful test. A sensitivity of 93–100% using a single midnight salivary cortisol measurement has been reported (28, 29, 30, 31, 32).
The reproducibility of this test and its validity in particularly challenging situations, such as late pregnancy, has not been investigated prospectively. In addition, the methods applied to set the cutoff points for test interpretation were quite variable and not standardized. Here we show in a prospective outpatient study the reproducibility of NSC in healthy volunteers (NO) and the excellent sensitivity and specificity of this test for diagnosis or exclusion of hypercortisolism in patients with CS, in patients in which CS was suspected but not confirmed [unconfirmed (UC)], obese individuals (OB), and women in late pregnancy (PR).
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
We evaluated 26 consecutive patients referred to our clinic at the University Hospital Basel (Basel, Switzerland) for evaluation of CS. In 12 patients, the diagnosis of CS was confirmed by either histology or hypocortisolism after surgery. In two patients with presumed ectopic ACTH production without localization, the diagnosis of CS was confirmed by very high ACTH and cortisol levels in the presence of several clinical features of CS and by normalization of previous pathological tests and clinical features under pharmacological treatment. The etiologies of all CS were as follows: five CS with pituitary microadenoma, three adrenal CS with unilateral adenoma, two ectopic ACTH production with unknown localization, and two metastatic neuroendocrine cancers with ACTH secretion. The latter two subjects were inpatients due to progressed disease; all other subjects were evaluated in an ambulatory setting.
To explore normative data for our assay, we recruited 20 NO patients, mainly staff members from the endocrine and medical outpatient clinics.
Whereas the CS and NO groups served as our derivation sample, we evaluated a validation sample, including suspected hypercortisolism. In 14 of the 26 patients were of the UC group. They all had some Cushingoid signs, e.g. visceral obesity and hirsutism, associated with one or more pathological screening tests before admission (24-h UFC and/or DST). The diagnosis of CS was excluded after repeated biochemical testing and absence of progression toward overt CS during at least 6 months. We evaluated further 16 patients with severe obesity, defined as a body mass index greater than 35 kg/m2, referred for evaluation for weight reduction surgery, 20 healthy PR subjects (35th up to 41st gestational week, median of 38th week), and five treated CS (TC), four of them cured or under adequate pharmacological treatment.
For logistical reasons, saliva collection after the 1 mg overnight DST was only obtained in four and urinary creatinine measurement in six CS patients. The number of positive controls for the NSC decreased to eight due to insufficient sample volume in four of the CS patients.
Study design
All subjects collected their urine for 24 h from 0730 to 0730 h, for measurement of UFC, creatinine, and the calculated cortisol/creatinine ratio. Salivary sampling for cortisol quantification occurred on the same day at 0730, 2330, and 0730 h the morning after the administration of 1 mg dexamethasone at midnight before. Simultaneously with the saliva sampling, a serum sampling was performed after the DST for cortisol quantification. To test the intraindividual day-to-day variability, the salivary cortisol measurements were performed by the healthy subjects on three separate days. Pregnant women collected only a 24-h UFC and salivary cortisol but did not undergo an overnight DST. Saliva was collected by chewing a cylindrical cotton swab (Salivette, Sarstedt, Germany) for about 2 min after rinsing the mouth with water. At least 3 h before the collection, the subjects were told not to eat, brush their teeth, or exert physical activity to avoid a possible elevation of the cortisol level. The specimens were refrigerated at home at a temperature of 2–8 C and were brought within 2 d to the laboratory at the Department of Central Laboratories at the University Hospital Basel for additional processing. The received samples were centrifuged at 2600 rpm (in Heraeus centrifuge; Kendro, Osterode, Germany) for 5 min, the cotton swab was removed, and the collected saliva sample was frozen at –20 C until assayed. Written informed consent was obtained from all subjects evaluated. The protocol was approved by the local ethics committee.
Assays
For salivary cortisol measurements, we used the commercial kit CORT-CT-2 RIA (CIS Biointernational, Gif-sur-Yvette, France), which is approved for serum, urine, and with an adapted standard curve for saliva cortisol [analytical sensitivity given by the manufacturer is 0.8 nmol/liter (0.029 µg/dl)]. Cross-reactivity with major steroidal substances is as follows: for cortisol (100%), prednisone and dexamethasone < 0.1% and prednisolone 45%.
The reference range indicated by the manufacturer is 5.3–61.8 nmol/liter (0.19–2.24 µg/dl) at 0800 h and 1.2–12.3 nmol/liter (0.04–0.45 µg/dl) at 2000 h.
Based on our own data, the intraassay coefficient of variation at 6.7 nmol/liter (0.24 µg/dl) was 4.25%, and the interassay coefficient of variation at 11.0 nmol/liter (0.40 µg/dl) was 3%; recovery of cortisol was 106%.
Serum cortisol was quantified by ELISA (Immulite One instrument, Diagnostic Products Corporation, Los Angeles, CA; Bühlmann Laboratories, Allschwil, Switzerland).
UFC was measured by the same instrument (Immulite One), whereby the free urinary cortisol assay was performed on extracted samples (extraction with methylene chloride in a hood; urine/methylene chloride, 1:4), and the recovery was 75%.
This method was validated in previous experiments (data not shown) and is in use in the routine laboratory. The cutoff of 248 nmol/24 h (90 µg/d) used until now was taken from the literature (33), a practice still followed by many medical centers.
Statistical analysis
Data are expressed as median and range. Results below the detection limit of the assay were set to the limit value. For unpaired not normally distributed data, we used Mann-Whitney U test for group comparison and Spearman’s rank test for correlations. P < 0.05 was considered statistically significant.
Variance components and intraclass correlations were estimated using random-effects ANOVA model using the lone-way procedure in Intercooled STATA (version 8; StataCorp, College Station, TX).
Variation can be expressed by dividing overall variation into between- and within-subject variation and describing the percentage of overall variation attributable to each component. If we let 
2b be the between-subject variance and 2w be the within-subject variance, then 2b/(2b + 2w) is the intraclass correlation coefficient (ICC), and 100 x ICC is the percentage of variation explained by between-subject variation. The remaining variation is the within-subject variation, which is the combined biologic and analytic variation (34). The ICC can therefore be used as an index of reproducibility for a test, being ideal close to 1.0 (close to 100% of the observed variation explained by between-individual variation).
For estimation of the cutoff values, we used receiver operating characteristic (ROC) analysis for all diagnostic settings, which were optimized for sensitivity. For these calculations, the PR and TC groups were excluded. Because late pregnancy is obvious and its hormonal changes unique, we did not optimize the cutoffs for this special setting. Because TCs are not a well defined group, subjects being under different and, in one case, insufficient treatments, their exclusion from defining the cutoff was considered as reasonable.
The quality of the tests was expressed as the area under the ROC curve (AUCROC), an index of the probability of correctly identifying and excluding the disease (CS) and being independent from the prevalence of the disease as opposed to the diagnostic accuracy. Given the small number of individuals, the limitations of the ROC analysis and the lack of reaching significant differences between the ROC curves in our setting must be acknowledged.
We used MedCalc software (version 7.4.3.0; MedCalc, Mariakerke, Belgium) for ROC analyses and Statistica (version 6; StatSoft, Tulsa, OK) for other statistical comparisons.
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
, presented as median (range).
|
A random-effects ANOVA estimated the between-subject and within-subject variation in repeated measures of salivary cortisol in healthy volunteers. The calculated ICC of NSC was 0.78 (i.e. 78% variation explained by between-subject variation and 22% by within-subject variation), whereas the ICC of the morning salivary cortisol and the morning to NSC ratio was only 0.47 and 0.45, respectively.
Median and range of the three saliva samples of the 20 healthy subjects are 2.1 (0.3–4.3) nmol/liter [0.07 (0.01–0.16) µg/dl)] at nighttime and 11.6 (2.5–25.4) nmol/liter [0.42 (0.09–0.92) µg/dl] in the morning. Figure 1 shows each NSC measurement of all 20 individuals.
|
As expected, patients with CS, compared with all other groups (NO, UC, OB, PR, and TC) had higher morning (P = 0.0027) and NSC (P < 0.0001) levels, higher 24-h UFC (P < 0.0001) and urinary cortisol/creatinine (P < 0.0001) ratios, and higher salivary (P = 0.0043) and serum cortisol (P < 0.0001) concentration after the DST but lower morning to NSC ratios (P = 0.0001) (Table 1
and Fig. 2, A–C).
|
Compared with NO, and apart from patients with CS, PR showed the highest 24-h UFC (P = 0.0019), urinary cortisol/creatinine ratios (P = 0.0001), and NSC (P < 0.0001), whereas no significant difference was seen in the morning salivary cortisol (P = 0.056) (Table 1).
Cushing’s syndrome excluded or not established (UC)
UC patients had, compared with NO, similar 24-h UFC, morning and NSC values, but higher serum and saliva cortisol concentrations after the 1 mg overnight DST (P = 0.027 and P = 0.019, respectively). Nevertheless, all measurements were significantly different when compared with CS (all P values <0.042) (Table 1).
Obesity (OB)
Compared with NO, OB showed similar morning and NSC, similar 24-h UFC, despite lower urinary cortisol/creatinine ratios (P = 0.009), and lower serum cortisol levels after DST (P = 0.002). As for the UC group and the NO group, every test was significantly different when compared with CS (all P values <0.015) (Table 1).
Treated Cushing’s syndrome (TC)
Three of the five included patients with TC were in complete remission after surgical treatment and did not have any treatment at the time of evaluation. Two patients were on pharmacological treatment, one with metyrapone alone and one with metyrapone and ketoconazole. The latter patient was known to still have insufficient treatment according to the clinical symptoms and pathological DST. Despite a normal UFC at the time of evaluation, the NSC was high, with 8.7 nmol/liter (0.32 µg/dl), and could therefore detect the underlying disease. The other four patients with sufficient treatment or in remission did not show different test results compared with NO, OB, or UC.
Nightime salivary cortisol (NSC)
The quality of the different tests in the diagnosis of CS was estimated by ROC analysis and expressed as AUCROC (Table 2). Estimating the cutoff value of NSC by ROC, optimized for sensitivity, we found a test-specific cutoff of 6.1 nmol/liter (0.22 µg/dl) to be most appropriate to distinguish between CS and UC, with a sensitivity and specificity of 100% (Fig. 2A). This cutoff was equal to the highest NSC value of the UC group. Using this cutoff in the PR group (which was excluded for cutoff determination), the physiological higher NSC levels reduced the resulting specificity to 75%.
|
In all groups, the morning salivary cortisol showed a high variance, especially in the CS patients, with a clear overlap between CS patients (positive control group) and the NO, OB, and UC groups (negative control group).
This problem was also reflected by the highly variable salivary morning to NSC ratio, leading to a slightly, although not significantly, reduced test performance (AUCROC of 0.945) compared with the NSC alone. Combining NSC with the morning to NSC ratio did not improve the general diagnostic performance in our population, in contrast to a previous report (29).
24-h UFC and urinary cortisol/creatinine ratio
The healthy volunteers showed a 24-h UFC of 322 (163–485) nmol/24 h [117 (59–176) µg/d], which is higher than the usually quoted literature reference range. Adjusting our cutoff for 24-h UFC to 504 nmol/24 h (183 µg/d), according to our reference data from the NO and OB groups, this test demonstrated an excellent sensitivity and specificity (both 100%), without any overlap between the CS and UC groups. A greater gap and therefore an even better differentiation between the CS and UC groups was seen in the urinary cortisol/creatinine ratio, with the best cutoff at 45 nmol/mmol (14.1 µg/g) (Fig. 2B).
The 1 mg overnight DST
The 1 mg overnight DST showed the known relatively low sensitivity (83.3%) at the usual cutoff of 140 nmol/liter (5.0 µg/dl) and the relatively low specificity (79.6%) at the lower proposed cutoff of 50 nmol/liter (1.8 µg/dl). Optimizing the sensitivity by ROC, a cutoff of 86 nmol/liter (3.1 µg/dl) achieved the highest sensitivity and specificity (100%/91.8%) in our population (Fig. 2C).
For the salivary cortisol after 1 mg overnight DST, we estimated the cutoff to be at 1.5 nmol/liter (0.05 µg/dl); however, we did not perform a detailed ROC analysis because of the small number of positive controls (n = 4). There was a significant correlation between salivary and serum cortisol after DST (r = 0.529; P < 0.05). The smaller AUCROC of the serum DST (0.986) and the salivary DST (0.932) may suggest a reduced diagnostic performance of these tests compared with NSC or 24-h UFC, although these differences were not statistically significant.
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
In our study, NSC had an excellent reproducibility, demonstrated by a low within-subject variation of only 22% of the total variability. In contrast, the morning salivary cortisol showed broad range and variability in all groups, especially in patients with CS. Repeated measures in healthy volunteers showed high within-subject variation of morning salivary cortisol, responsible for 53.4% of the total variability. This is probably due to increased responsiveness of the hypothalamic-pituitary-adrenal (HPA) axis to extrinsic factors in the morning hours and the known pulsatile secretion pattern of ACTH in many patients with CS.
Regarding the diagnostic performance of NSC as a screening tool for CS, our results confirm the findings of previous reports from the literature (28, 29, 30, 31, 32), demonstrating an excellent discriminatory potential in our studied population.
It should be mentioned that the saliva collection procedure can be problematic in special situations. Four of the 12 CS patients failed to provide a sufficient sample volume due to dehydration (n = 1), somnolence/unconsciousness (n = 1), or insufficient cooperation (n = 2). These were all indirect consequences of severe CS and, therefore, compromised patients. To minimize these failures, an optimal instruction of the patient and sequential sampling must be implemented.
The cutoff value for NSC has to be established for each type of assay and population used. Different statistical methods were used in the literature for its determination. Some groups had set the cutoff at the highest value or the 99th percentile of the (negative) control group, which is susceptible to outliers or selection bias of the tested population. We estimated our cutoff using ROC analysis, optimized for sensitivity, at 6.1 nmol/liter (0.22 µg/dl), which corresponded also to the highest value of the UC group, whereas our lowest value of the CS group was 7.0 nmol/liter (0.25 µg/dl).
A recent report (32) proposed a cutoff of 5.52 nmol/liter (0.20 µg/dl), using the same assay as in the present study, to obtain the sensitivity of 100% at a specificity of 96%. Recapitulating these and our data, the appropriate cutoff can be roughly estimated between 5.5 and 6.9 nmol/liter (0.20–0.25 µg/dl). However, importantly, as for all other endocrine tests, normative reference values must be established locally to obtain an optimal diagnostic accuracy.
The limitation of ROC analysis and cutoff estimation using small group sizes has to be acknowledged. With small subject numbers, the cutoff will depend on a few individuals only. Because the cutoff can only be roughly estimated in our study population, located between the upper limit of reference range and the lowest value of the positive controls, it must be chosen as low as possible to achieve best sensitivity when used for screening. Comparison of the test-specific AUCROC does not demonstrate statistical significant differences in the present population due to its limited size. However, the clearly, although not significantly, higher sensitivity and specificity suggest that NSC, UFC, and urinary cortisol/creatinine ratio may have a superior test performance compared with DST.
Pregnancy is accompanied by a physiological hypercortisolism due to the estrogen-mediated rise in corticosteroid-binding globulin and therefore total cortisol. Beside the secretion of cortisol, cortisone, ACTH, and corticotropin-releasing factor by the fetoplacental unit, there seems to be a competition of progesterone and cortisol at the level of the glucocorticoid receptor that results in a reduced cortisol feedback and therefore resetting of the HPA axis (corticostat). This explains the observed rise in free serum or salivary cortisol, maintaining a circadian rhythm at a higher level (6, 7, 8, 9, 10). Because of the unique status of pregnancy, and the fact that a screening test must be optimized for sensitivity, these subjects were excluded for the calculation of the cutoff to prevent an unnecessary rise with the risk of a loss in sensitivity.
The pregnant women in our group showed significantly higher 24-h UFC levels and a higher NSC, which reduces the specificity of the NSC to 75%. Thus, late pregnancy, which is clinically evident, must be taken into account when using this test during pregnancy.
The 24-h UFC measured by RIA, a commonly used and established method, showed unexpected high values in the NO group, reflecting a known methodological problem (15). Compared with combined competitive protein binding with detailed chromatography (HPLC + RIA), today’s reference method, the daily excretion of free cortisol ranges from 28–117 nmol/24 h (10–42 µg/d) (35). The high values obtained by RIA are due to an unspecific antibody with cross-reactivity to other cortisol metabolites not yet identified and are therefore rather a marker for urinary cortisol metabolites than for free cortisol. In the literature, the reference ranges for 24-h UFC show a wide variation, reflecting the need for assay-specific normative data and cutoff values for the diagnosis of CS. Analyzing results from the urinary cortisol assay used at our institution, the NO group demonstrated relatively high UFC concentrations, implying a marked higher cutoff of 504 nmol/24 h (183 µg/d) compared with the literature [250–300 nmol/24 h or 90–109 µg/d (33, 36)]. After adapting our cutoff level, the 24-h UFC regained an excellent power to discriminate NO and UC from CS with a perfect sensitivity and specificity. The cortisol/creatinine ratio showed an even greater gap between CS and controls, suggesting a possibly better diagnostic performance than UFC. This might be explained by the reduced muscle mass with consequent lower urinary creatinine excretion in CS. Additionally, the urinary creatinine excretion can be used as a control for complete urine collection over 24 h and can correct for other factors (e.g. lean body mass). Consequently, reference values of most urinary hormones are expressed as creatinine ratio. A correction of an incomplete 24-h collection, however, should not be considered because of the different circadian variation of cortisol excretion compared with creatinine.
The 1 mg overnight DST showed the known relatively low sensitivity and specificity according to the used cutoff due to a considerable overlap between CS and negative controls. This was reflected in the reduced AUCROC, although a statistically significant difference between the tests could not be demonstrated due to our relatively small sample size. Despite pharmacological suppression, the high variation of the morning cortisol, due to the pulsatile secretion pattern, could be responsible for the overlap between CS and negative controls. Considering the known problems of dexamethasone metabolism, the value of the DST as a first-line screening test seems questionable, leading to substantial unnecessary additional testing.
Compared with the serum DST, the salivary DST showed a similarly reduced specificity and AUCROC. Until additional evaluation, there is no reason to use salivary cortisol after DST as a screening tool.
Interestingly, in the CS, UC, OB, and NO groups, the NSC and salivary DST showed an excellent correlation (r = 0.706; P < 0.05), indicating a similar responsiveness of the HPA axis to endogenous vs. pharmacological suppression (Fig. 3). NSC was significantly correlated with 24-h UFC, urinary cortisol/creatinine ratio, and serum DST (r = 0.537, 0.569, 0.596, respectively; P < 0.05).
|
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Abbreviations: AUC, Area under the curve; CS, Cushing’s syndrome; DST, dexamethasone suppression test; HPA, hypothalamic-pituitary-adrenal; ICC, intraclass correlation coefficient; NO, healthy volunteers group; NSC, nighttime salivary cortisol; OB, obese patients group; PR, women in late pregnancy group; ROC, receiver operating characteristic; TC, treated Cushing’s syndrome group; UC, unconfirmed; UFC, urinary free cortisol.
Received November 18, 2004.
Accepted July 5, 2005.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  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]
|
|
|
|
|
|
M. B. Elamin, M. H. Murad, R. Mullan, D. Erickson, K. Harris, S. Nadeem, R. Ennis, P. J. Erwin, and V. M. Montori Accuracy of Diagnostic Tests for Cushing's Syndrome: A Systematic Review and Metaanalyses J. Clin. Endocrinol. Metab., May 1, 2008; 93(5): 1553 - 1562. [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]
|
|
|
|
|
|
F. Pecori Giraldi, A. G. Ambrogio, M. De Martin, L. M. Fatti, M. Scacchi, and F. Cavagnini Specificity of First-Line Tests for the Diagnosis of Cushing's Syndrome: Assessment in a Large Series J. Clin. Endocrinol. Metab., November 1, 2007; 92(11): 4123 - 4129. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. K. Baid, N. Sinaii, M. Wade, D. Rubino, and L. K. Nieman Radioimmunoassay and Tandem Mass Spectrometry Measurement of Bedtime Salivary Cortisol Levels: A Comparison of Assays to Establish Hypercortisolism J. Clin. Endocrinol. Metab., August 1, 2007; 92(8): 3102 - 3107. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. M Maguire, G. R Ambler, B. Moore, K. Waite, M. McLean, and C. T Cowell The clinical utility of alternative, less invasive sampling techniques in the assessment of oral hydrocortisone therapy in children and adolescents with hypopituitarism Eur. J. Endocrinol., April 1, 2007; 156(4): 471 - 476. [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]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |
