| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Article |
Division of Endocrinology, Diabetology and Clinical Nutrition, Department of Internal Medicine, University Hospitals, CH-4031 Basel, Switzerland
Address all correspondence and requests for reprints to: Kaspar Berneis, M.D., University Hospital Bruderholz, Department of Internal Medicine, 4101 Bottmingen, Switzerland. E-mail: kaspar{at}berneis.ch.
Abstract
The metyrapone test is used to test the hypothalamic-pituitary-adrenocortical axis. The present study aims to assess the diagnostic accuracy of combined stimulation of ACTH and compound-S (CMP-S). In addition, we analyzed the safety and practicability of this test as an outpatient procedure.
A total of 327 metyrapone tests were analyzed retrospectively in 185 patients (mean age, 50.3 ± 15.2 yr). One hundred thirteen patients had one test, and 72 patients had between 2 and 6 tests over 1–3 yr. Most patients suffered from pituitary adenomas (60 macroadenomas, 63 microadenomas) or other pituitary lesions (n = 29). Metyrapone (2 g) was given at 2400 h as an outpatient procedure. Blood samples for analysis of ACTH, CMP-S, and cortisol were taken at 0730 h.
Stimulation of adrenal CMP-S and cortisol by pituitary ACTH demonstrated a dose-response curve with the shape of half a geometric parabola. CMP-S reached a plateau when ACTH rose above 175 ng/liter [r = 0.661, P < 0.0001 for ACTH <175 ng/liter; r = 0.083, P = not significant (NS) for ACTH >175 ng/liter], cortisol flattened at ACTH levels above 230 ng/liter (r = 0.633; P < 0.0001 for ACTH < 230 ng/liter; P = NS for ACTH >230 ng/liter). Alternatively, the sum of CMP-S plus cortisol also flattened when ACTH rose above 230 ng/liter (r = 0.696; P < 0.0001 for ACTH <230; P = NS for ACTH > 230 ng/liter).
Receiver operating curve analysis defining a cut-off for ACTH at 150 ng/liter demonstrated a sensitivity of 47% and 67% at a cut-off level for CMP-S at 200 or 260 nmol/liter, respectively. The respective specificity was 82% and 68% for CMP-S. This compared with a sensitivity of 71% and specificity of 69% if the sum of CMP-S plus cortisol of 450 nmol/liter were used as cut-off.
The response curve between CMP-S and ACTH implies a maximally stimulated adrenal cortex at circulating ACTH levels above 175 ng/liter. Single measurement of CMP-S using the cut-off at 200 nmol/liter, as suggested in the literature, yields a poor sensitivity of only 47% compared with ACTH. Despite the relatively high cross-reactivity of CMP-S in the cortisol assay, the sum of CMP-S and cortisol levels with a cut-off value of 450 nmol/liter yields a better diagnostic accuracy compared with CMP-S alone.
THE METYRAPONE TEST was developed specifically to assess the pituitary ACTH reserve (1). Metyrapone inhibits the adrenocortical enzyme 11 ß-hydroxylase, which converts 11-deoxycortisol [compound-S (CMP-S)], to cortisol in the final step of adrenal steroidogenesis, leading to a decrease of circulating cortisol. CMP-S does not have any glucocorticoid activity and thus does not inhibit ACTH production. The decline in plasma cortisol stimulates ACTH production, thereby increasing adrenal steroidogenesis proximal to the enzyme blockade and causing CMP-S to accumulate. Therefore, an advantage of the metyrapone test is that it tests the integral hypothalamic-pituitary-adrenocortical axis.
Dynamic tests using stimulation by ACTH, insulin-hypoglycemia, or metyrapone are the procedures most widely used to assess the integrity of the hypothalamus-pituitary-adrenal (HPA) axis. It has been suggested that the ACTH stimulation test may be the best indicator of recovery of the entire HPA axis after therapy with high-dose glucocorticoids, because adrenal function recovers after pituitary function (2). However, there may be partial degrees of ACTH deficiency with sufficient hormone levels to provide a trophic stimulus to the adrenal gland and prevent its atrophy (3). Thus, adrenal response to ACTH may be preserved despite diminished pituitary ACTH reserve (4). In addition, ACTH testing with the commonly used supraphysiological doses (250 µg) might provoke a response, whereas physiological amounts of ACTH in response to insulin-hypoglycemia or metyrapone do not. Recently, some authors have suggested that the low-dose (1 µg) ACTH test may be equal to metyrapone testing (5, 6). However, Soule et al. (7) demonstrated that 50% of patients with proven ACTH deficiency to metyrapone testing showed a normal response to low-dose ACTH, suggesting that this test lacks sensitivity in assessing secondary adrenal insufficiency.
Insulin-hypoglycemia and metyrapone both assess pituitary ACTH reserve by different mechanisms. Previous studies have found good correlation between these two methods (4, 8). The insulin-hypoglycemia test, however, is uncomfortable for the patients, requires close medical supervision, and should not be performed in certain medical conditions. In contrast, the overnight single dose metyrapone test is simple, safe, and requires only one blood sample for measurements of ACTH, cortisol, and CMP-S. The present study evaluates retrospectively more than 300 overnight metyrapone and thereby assesses the diagnostic accuracy of ACTH, CMP-S, and cortisol, its correlation, and receiver/response operating characteristic (ROC) curves. Furthermore, its practicability and safety are examined. On the basis of the superior diagnostic performance, we propose a new evaluation criteria of the metyrapone test in which direct measurement of ACTH is not available, i.e. the sum of cortisol and CMP-S. Results of stimulation during chronic glucocorticoid therapy should be interpreted with caution, because abnormalities may be due to hypothalamic-pituitary disease or to glucocorticoid therapy itself.
Patients and Methods
Patients (Table 1
)
A total of 327 overnight metyrapone tests [169 in females (52%), 158 in males (48%)], which were performed in 1978–1996 in 185 patients [95 females (51%), 90 males (49%)] with a mean age of 50.3 ± 15.2 yr, were analyzed. Repeated tests were performed in 72 patients (2–6 tests), and single tests in 113 patients. Multiple tests represented routine follow-up tests, usually in intervals from 1–3 yr. Exclusion criteria were Cushing’s disease and primary adrenal insufficiency. Table 1 lists the main endocrine diagnosis in all patients. A total of 143 tests were performed in patients on substitution treatment with glucocorticoids, and 174 tests were performed in patients without glucocorticoids. The substitution doses were as follows: 2.5–10 mg hydrocortisone (n = 24), 11–20 mg hydrocortisone (n = 71), and more than 20 mg hydrocortisone (n = 48). The dose of steroids was unknown in 10 tests, which were therefore not included for comparisons.
|
Patients ingested a single dose of 2.0 g metyrapone (eight capsules of 250 mg Metopirone, Novartis, Basel, Switzerland) at 2400 h at home with a snack to prevent gastric irritation. Patients were instructed not to take their usual dosage of hydrocortisone the following morning to avoid interference between hydrocortisone measurement and cortisol measurement. Blood samples were taken at 0730 h for ACTH, CMP-S, and cortisol as an outpatient procedure. All patients received 50 mg of hydrocortisone after blood sampling to prevent possible acute adrenal insufficiency. Blood samples were immediately stored on ice, centrifuged thereafter, and frozen at -28 C until later analysis. ACTH was measured as described earlier (8, 9) and by chemiluminescence immunoassays with the Nichols Advantage Specialty System (Nichols Institute Diagnostics, San Clemente, CA). As for ACTH measurement, the method for cortisol determination was changed during the years to apply up-to-date technology. To assure comparability, quality controls including linear regression analysis were performed (data not shown). Lately, cortisol has been measured by chemiluminescence immunoassays with the Nichols Advantage Specialty System (Nichols Institute Diagnostics). The reported cross-reactivity with CMP-S as reported by the manufacturer is 18% at a concentration of 400 µg/dl. CMP-S measurements have been performed with an RIA (ImmunoChem double antibody, ICN Biomedicals, Costa Mesa, CA). The cross-reactivity with cortisol is 0.23%, as reported by the manufacturer.
Statistical analysis
All data are expressed as mean ± SEM. Statistical analysis was performed on a Power Macintosh G3 with Statview. Data that are not normally distributed were logarithmically transformed before analysis by parametric testing. Paired and unpaired Student’s t tests were performed to detect differences within and between groups. Correlation coefficients were calculated for ACTH, CMP-S, and cortisol. Scheffé’s procedures were performed for correction of multiple comparisons. The present study did not include a control group, but a mean ACTH response above 150–200 ng/liter is observed in normal persons, as reported in the literature (8, 10). Conversely, the lower reference range is 75 ng/liter (11). Therefore, ROC curves were drawn for ACTH less than 200, 150, and additionally for 100 and 75 ng/liter as cut-off value and compared with CMP-S and the sum of CMP-S plus cortisol.
Results
Plasma ACTH, CMP-S, and cortisol concentrations (Fig. 1)
Mean levels for ACTH, CMP-S, and cortisol were 149 ± 9 ng/liter, 256 ± 10 nmol/liter, and 179 ± 8 nmol/liter, respectively, for all 327 tests. These values were higher in the subgroup of patients with no steroid replacement (203 ± 14 ng/liter, 307 ± 14 nmol/liter, and 219 ± 11 nmol/liter, for ACTH, CMP-S, and cortisol, respectively) compared with patients on substitution with glucocorticoids (85 ± 7 ng/liter, 193 ± 13 nmol/liter, and 130 ± 11 nmol/liter for ACTH, CMP-S, and cortisol, respectively; P < 0.001 or less for without glucocorticoids vs. with glucocorticoids). We found no dose-dependent effect of the hydrocortisone substitution dosage on ACTH and CMP-S levels. However, plasma cortisol levels in the group receiving more than 20 mg hydrocortisone were significantly lower compared with patients on lower substitution (2.5–10 mg hydrocortisone). In 13 of the 72 patients, repeated metyrapone testing confirmed the normalization of the pituitary adrenal axis, allowing patients to discontinue chronic glucocorticoid therapy.
|
)Significant correlations were found between ACTH and CMP-S (r = 0.623), ACTH and cortisol (r = 0.613), CMP-S and cortisol (r = 0.649), and between the sum of cortisol plus CMP-S vs. ACTH (r = 0.667). However, the correlation between serum ACTH and CMP-S levels was only significant for ACTH levels below 175 ng/liter (r = 0.661; P < 0.0001) in contrast to ACTH levels above 175 ng/liter (r = 0.083; P = NS). Interestingly, there was also a significant correlation between serum ACTH and cortisol that was strongest for ACTH levels below 230 ng/liter (r = 0.633; P < 0.0001) and not significant for ACTH levels above 230 ng/liter (r = 0.076; P = NS). Correlation analysis between the sum of cortisol plus CMP-S with ACTH also demonstrated a pattern with the shape of half a geometric parabola with best correlation for ACTH below 230 ng/liter (r = 0.696; P < 0.0001) and blunted effects when ACTH increased above 230 ng/liter (r = 0.137; P = NS).
|
)
ROC curves were drawn to assess the diagnostic accuracy of CMP-S levels in comparison with ACTH. When the cut-off value for ACTH was set at three different levels, namely 200 ng/liter, 150 ng/liter, or 100 ng/liter, to define a normal (safe) pituitary function, CMP-S reached a sensitivity of 61.7%, 67%, and 73%, respectively, and a specificity of 64.5%, 68%, and 66%, respectively, at a limit for CMP-S of 260 nmol/liter. In comparison, for the same cut-off values of ACTH (200, 150, and 100 ng/liter), the sum of cortisol and CMP-S reached a sensitivity of 65.7%, 71%, and 76%, respectively, and a specificity of 70.9%, 69%, and 64%, respectively (at a cut-off of CMP-S plus cortisol at 450 nmol/liter). Figure 4 shows sensitivities and specificities comparing cut-offs of CMP-S below 200 or 260 nmol/liter and CMP-S plus cortisol below 400, or below 460 in comparison to an ACTH response below 150 ng/liter. A cut-off of CMP-S at 260 nmol/liter is clearly superior to the commonly used 200 nmol/liter. Importantly, the sum of CMP-S plus cortisol yields even better sensitivity/specificity levels.
|
|
Side effects and safety
No major side effects were observed during all tests. Compliance was very good, because all patients appeared on time to deliver the blood sample. In no single case did signs of acute adrenal insufficiency appear, nor were they suspected. All patients could follow their daily work the same day of the study. There were only a few cases with nausea 1–2 h after ingestion of the metyrapone capsules, a well known side effect of the drug.
Discussion
The short metyrapone test allows the simultaneous investigation of the hypothalamic-pituitary function by direct measurement of ACTH reserve and also of circulating CMP-S levels as a test of pituitary adrenal reserve at the same time. All three centers regulating the pituitary adrenal axis can be evaluated at the same time in a single blood sample by this simple and convenient test. The classical metyrapone test, originally described by Liddle et al. (1) and still performed at some centers, is more expensive, and requires the daily collection of urine before and during metyrapone for 5 d and hospital admission because of the well known risk of acute adrenal insufficiency. The short metyrapone test was developed by the group of Jubiz et al. (12) using measurement of CMP-S. Limited data are available on this test. Because ACTH immunoassays have been cumbersome in the past and are available only in a few research centers, most results have been published using only CMP-S as a surrogate marker describing the HPA-axis (13, 14, 15, 16, 17). Only three groups besides us have reported results with direct measurement of ACTH (10, 18, 19). The present study measuring ACTH, CMP-S, and cortisol simultaneously compares the increases of both ACTH and CMP-S in a large number of tests and a large cohort of patients. Using a large number of tests, we are able to demonstrate that the commonly used cut-off for a normal CMP-S response (e.g. <200 nmol/liter) lacks sensitivity when compared with ACTH measurements.
In this study, 65% of all ACTH values were less than 150 ng/liter; an ACTH value above 150 ng/liter has been suggested as a normal response in healthy volunteers (8). In patients on chronic glucocorticoid therapy, the finding of a plasma ACTH below 100 ng/liter was indicative of pituitary-adrenal suppression. Importantly, this effect was already seen with relatively low doses of glucocorticoids (2.5–10 mg hydrocortisone). This might be explained first by the underlying disease itself, because the dosage of substitution with hydrocortisone is often adapted due to individual needs and complaints of patients. Furthermore, we cannot rule out that the degree of secondary adrenal insufficiency was equal in patients with lower and higher hydrocortisone substitution. Nevertheless, there might also have been a suppression of the pituitary-adrenal axis by the low-dose chronic glucocorticoid substitution, respectively (20).
In contrast to the findings for the ACTH response to metyrapone, only 37% of CMP-S responses were abnormal to metyrapone testing when a safe response was defined as CMP-S above 200 nmol/liter (10, 14, 17, 21). There was also a significant decrease of CMP-S during chronic substitution with glucocorticoids. However, this decrease was only 159%, compared with the group without glucocorticoid substitution. This was significantly less than the measured reduction of ACTH during chronic glucocorticoid substitution of 238%. Thus, our findings suggest that pituitary ACTH might be a more sensitive marker for glucocorticoid suppression, because partial ACTH deficiency may be detected earlier using stimulated ACTH values.
Importantly, when circulating ACTH reaches 150–175 ng/liter, the dose-response curve for CMP-S flattens. Our data show a strong correlation between ACTH and CMP-S for ACTH of less than 175 ng/liter. In contrast, no correlation was seen for circulating ACTH levels above 175 ng/liter. These results are in concordance with a recent paper analyzing 115 metyrapone tests (10).
A uniform definition of a normal ACTH response to metyrapone does not exist. Although some suggested an ACTH peak above 200 ng/liter (8), others suggested above 150 ng/liter (10) or only above 75 ng/liter (11). There is more accordance with regard to a normal response of CMP-S, although this arbitrary limit has not been validated by another test. A normal peak value is commonly defined by a CMP-S level above 200 nmol/liter (10, 14, 17, 21). However, it is important to point out that an exact definition of a normal ACTH or CMP-S is impossible based on the fact that the patterns of adrenal insufficiency are highly variable, ranging from mild, often overseen, and completely reversible symptoms to death. Therefore, a true "gold standard" for the measurement of adrenal or pituitary insufficiency does not exist. It is widely thought that the insulin-hypoglycemia test may be the gold standard test on the basis that a stressful situation is produced, which may result in adrenal insufficiency. Therefore, most of the definitions, ours included, present only presumed safety values and not true normal values. Our results suggest a safe ACTH response in the range between 150 and 175 ng/liter just before CMP-S flattens. Interestingly, we also found strong positive correlation between ACTH and cortisol as well as CMP-S. It has been well documented by Jubiz et al. (22) that the inhibition of 11-ß hydroxylase is declining in the early morning hours and that the cortisol increases in parallel with the ACTH and the declining effects of metyrapone. The positive correlation of cortisol and CMP-S or ACTH could be partially explained by the cross-reactivity of the cortisol measurement with CMP-S. However, in clinical practice cortisol is usually not corrected for cross-reactivity with CMP-S using this assay. Therefore, the positive correlation between ACTH and cortisol levels can be used to strengthen the test criteria. This is illustrated by the low sensitivity of only 47% (specificity, 82%) for CMP-S levels below 200 nmol/liter for a given cut-off value for the ACTH response of more than 150 ng/liter. Sensitivity of CMP-S compared with ACTH can be increased to 67% (specificity, 68%) when a normal CMP-S response is defined at above 260 nmol/liter. Importantly, the sum of CMP-S plus cortisol (with a cut-off value of 450 nmol/liter for the sum of both parameters) increases the sensitivity and specificity to 71% and 69%, compared with ACTH. Thus, our data suggest that a cut-off value for CMP-S of 260 nmol/liter (rather than 200 nmol/liter) should be used to define a safe response. Alternatively, we suggest using the sum of cortisol plus CMP-S and defining a safe response at 450 nmol/liter (corresponding to an ACTH > 150 ng/liter). This increases both specificity and sensitivity compared with CMP-S alone, and the obtained results are less discordant to ACTH measurements.
Recently, it has been suggested that the overnight metyrapone test identifies more patients with possible ACTH deficiency than the insulin-hypoglycemia test (23). But this study was performed in a small number of patients, and only CMP-S was measured during metyrapone testing, which lacks specificity and sensitivity in comparison to ACTH. Measurement of CMP-S may be valuable under certain circumstances to separate between primary adrenal insufficiency when baseline ACTH levels are high and following metyrapone administration, CMP-S concentrations are low, and ACTH levels do not change (19). The value of the low-dose ACTH test is unclear. It has been suggested that the low-dose ACTH test may be better than the high-dose ACTH test (6) and equal to metyrapone testing (5). However, Mayenknecht et al. (24) compared low-dose and high-dose ACTH tests to metyrapone and insulin-hypoglycemia tests and found that these tests identified patients with moderately to severely pathological insulin and metyrapone tests, but not those with slightly pathological reference tests.
In summary, we conclude from our study that metyrapone is a potent stimulator of ACTH secretion (mean rise > 200 ng/liter), comparable with the insulin-hypoglycemia test, and even stronger than the corticotropin-releasing factor test. The good correlation between circulating CMP-S and ACTH below ACTH levels of 175 ng/liter suggests that the adrenal cortex is maximally stimulated when ACTH rises above 150–175 ng/liter. Measurement of ACTH and CMP-S allows testing the pituitary and adrenocortical reserve in a single test and provides a methodological control. It is concluded that the ACTH response is the more sensitive parameter to evaluate the HPA axis, compared with CMP-S. During metyrapone testing, cortisol increases dose-dependently, demonstrating good correlation with ACTH levels below 230 ng/liter. We suggest using, if available, direct measurements of ACTH greater than 150 ng/liter as cut-off or the sum of cortisol plus CMP-S at a limit of 450 nmol/liter to define a safe pituitary response.
Acknowledgments
We appreciate the kind support of the endocrinology nurses Silvia Alsmer and Maya Kunz.
Footnotes
B. M. was supported by the "Sondesprogram zur Fordesung des akademischen Nachwuchses der Universitat Basel."
Abbreviations: CMP-S, Compound-S (11-deoxycortisol); HPA, hypothalamus-pituitary-adrenal; NS, not significant; ROC, receiver/response operating characteristic.
Received December 10, 2001.
Accepted September 7, 2002.
References
This article has been cited by other articles:
 |
|
 |
|
  R. F. A. de Lind van Wijngaarden, B. J. Otten, D. A. M. Festen, K. F. M. Joosten, F. H. de Jong, F. C. G. J. Sweep, and A. C. S. Hokken-Koelega High Prevalence of Central Adrenal Insufficiency in Patients with Prader-Willi Syndrome J. Clin. Endocrinol. Metab., May 1, 2008; 93(5): 1649 - 1654. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Z. Thomas and G. L Fraser An Update on the Diagnosis of Adrenal Insufficiency and the Use of Corticotherapy in Critical Illness Ann. Pharmacother., September 1, 2007; 41(9): 1456 - 1465. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Aneja and J. A Carcillo What is the rationale for hydrocortisone treatment in children with infection-related adrenal insufficiency and septic shock? Arch. Dis. Child., February 1, 2007; 92(2): 165 - 169. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Salvatori Adrenal Insufficiency JAMA, November 16, 2005; 294(19): 2481 - 2488. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. F. Dallman Adrenocortical function, feedback, and alphabet soup Am J Physiol Endocrinol Metab, September 1, 2005; 289(3): E361 - E362. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
I. E. Widmer, J. J. Puder, C. Konig, H. Pargger, H. R. Zerkowski, J. Girard, and B. Muller Cortisol Response in Relation to the Severity of Stress and Illness J. Clin. Endocrinol. Metab., August 1, 2005; 90(8): 4579 - 4586. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Y. Liu, S. M. Pincus, D. M. Keenan, F. Roelfsema, and J. D. Veldhuis Joint synchrony of reciprocal hormonal signaling in human paradigms of both ACTH excess and cortisol depletion Am J Physiol Endocrinol Metab, July 1, 2005; 289(1): E160 - E165. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. S. Nylen and B. Muller Endocrine Changes in Critical Illness J Intensive Care Med, March 1, 2004; 19(2): 67 - 82. [Abstract] [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 |
, n = 174), only 2.5–10 mg (
, n = 24), 11–20 mg (
, n = 71), and > 20 mg (
, n = 48). *, P < 0.005 or less vs. 
). Cut-off values for CMP-S and CMP-S plus cortisol (bold) are presented in each of the four panels.