| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Studies |
,
Ramiz Çolak,
Fahrettin Keletimur,
Ahmet Selçuklu,
Kürat Ünlühizarci and
Fahri Bayram
Departments of Endocrinology (H.S.D., R.C., F.K., K.U., F.B.) and Neurosurgery (A.S.), Erciyes University, Medical School, 38039 Kayseri, Turkey
Address correspondence and requests for reprints to: Dr. Fahrettin Keletimur, Department of Endocrinology, Erciyes Üniversitesi, Tip Fakültesi, 38039 Kayseri, Turkey.
 |
Abstract |
|---|
 Top Abstract IntroductionPatients and MethodsResultsDiscussionReferences |
|---|
550 nmol/L) to the 1-µg ACTH test and
five patients showed normal cortisol responses to the 250-µg ACTH
test among the six patients with subnormal cortisol responses to the
ITT. Three patients with subnormal cortisol responses to ITT and
baseline cortisol values less than 240 nmol/L showed normal HPA axis at
the end of the first month. In the late postoperative period, at the
third month, all the patients showed normal HPA axis. In the early postoperative period of pituitary surgery, the 1-µg ACTH test results are more concordant than the 250-µg ACTH test in comparison with the ITT. Our results also indicate that HPA axis dysfunction shown by ACTH stimulation tests and the ITT in early postoperative period may be normalized 1–3 months after surgery. For this reason, we think that dynamic tests including the ITT may not be useful early after pituitary surgery.
|
Introduction |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
After pituitary adenomectomy, the most widely applied tests for evaluation of HPA axis are morning cortisol measurements, the insulin tolerance test (ITT), the short ACTH stimulation test (SAT), the metyrapone test, and CRH stimulation tests (6, 8, 10, 11, 12, 13, 14). Among those tests, the ITT has been widely accepted as the gold standard for the assessment of HPA axis (6, 15). However, it is unpleasant and is not without risk (16). The metyrapone test evaluates negative feedback of cortisol on the pituitary gland, but it does not assess the cortisol response to stress (6, 17). Additionally, this test requires 11-deoxycortisol determinations, which are not available in most laboratories, and the drug is not available in all countries. The CRH test procedure has not been well standardized, and its value in differentiating hypothalamic from pituitary causes of ACTH failure is unclear (15). It has been suggested that the SAT, which is simple, easy, and an inexpensive method, may be used to assess the HPA axis instead of the ITT (18, 19, 20, 21, 22, 23, 24). However, it has been reported that the SAT may induce false positive cortisol responses shortly after pituitary surgery (6, 10, 14, 21, 25, 26, 27).
Recently, it has been suggested that using a supraphysiological dose of exogenous ACTH (250 µg) may cause false positive responses due to maximal stimulation of the adrenal cortex, and the test should be done by using lower doses of ACTH (26, 27, 28, 29, 30, 31). Although there are some studies using low-dose ACTH in a variety of patients (26, 27, 28, 32, 33, 34), we could not find a systematic prospective study that investigated the HPA axis at early postoperative period using the 1-µg and 250-µg ACTH stimulation tests and the ITT.
Three important questions should be answered in the patients undergoing pituitary surgery: 1) is there corticotroph cell damage? 2) which of the ACTH stimulation tests, the 1 µg or the 250 µg, is reliable and can be performed to evaluate the functional capacity of the HPA axis early after surgery? 3) is the dysfunction of the HPA axis, determined in the early postoperative period, persistent or not?
In this study, our aim was to find the answers to these questions. We have carried out a prospective comparative study assessing morning serum cortisol and the serum cortisol responses to the ITT and the 1-µg and 250-µg ACTH tests within the 4th–11th days and at the third month of pituitary surgery.
|
Patients and Methods |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
Nineteen patients (11 men and 8 women; age, 24–69 yr; mean age,
51.4 ± 2.9 yr) who underwent pituitary surgery (transsphenoidal
surgery in 17 patients and transcranial surgery in 2 patients) were
included in the study (Table 1
). Local
Ethical Committee approval had been obtained before the commencement of
the study, and informed consent was obtained from each patient. All
patients had baseline pituitary hormone levels, and the ITT was
performed in all the patients in the preoperative period. Patients with
Cushing’s syndrome were excluded. Nine patients had hypothyroidism,
five patients had hypogonadism, and all were receiving appropriate
replacement therapy.
|
None of the patients received glucocorticoid replacement therapy after discharge and at the first, second, and third month after surgery. Serum electrolytes, blood pressure, and physical examination were monitored to ascertain HPA axis status, and basal cortisol measurement was repeated when it was clinically indicated. Additionally, the 1-µg and 250-µg ACTH stimulation tests and the ITT were performed in 3 patients characterized by subnormal cortisol responses and subnormal baseline cortisol levels at the end of the first month and in 18 patients at the end of the third month.
Study protocol
A 250 µg tetracosactrin (1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24) (Synacthen; Novartis Pharma GmbH, Nürnberg, Germany) dose was diluted in sterile saline solution to a concentration of 5 µg/mL in plastic tubes and kept at 4 C maximally for 4 months. Further dilution (1 µg) was performed by taking 0.2 mL of this solution and adding 0.8 mL saline solution and immediately was used to the patient iv (31).
ACTH stimulation tests were performed by using 1 µg and 250 µg ACTH iv as a bolus injection after an overnight fast, and blood samples were drawn at 0, 30, and 60 min (16, 28). After an overnight fast, the ITT was performed by using iv regular insulin (0.1–0.2 U/kg), and serum glucose and serum cortisol levels were measured before and after 30, 60, and 90 min. All patients had serum glucose level less than 2.2 nmol/L during the ITT. No adverse effects were seen during the tests. Peak serum cortisol levels of 550 nmol/L or greater were considered as a normal response for both the ITT and the ACTH tests.
Assays
Serum cortisol levels were determined by RIA using a commercially available kit (ICN Biomedicals, Inc., Costa Mesa, CA). This procedure has an intra-assay and interassay coefficient of variation between 5–7%, and 6–7.9%, respectively. Its sensitivity was 4.13 nmol/L. Serum glucose was determined by glucose oxidase method on a Technicon RAXT otoanalyzer (Dublin, Ireland).
Statistical analysis
Kruskal Wallis ANOVA was performed for multiple comparison. This test was followed by Mann-Whitney U test to compare the two groups. Correlations were determined by linear regression analysis. The data are expressed as mean ± SEM. P values less than 0.05 were considered statistically significant.
|
Results |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
and Figs. 1 and 2.
Both insulin-like growth factor I levels and GH responses to oral
glucose tolerance tests were normalized at the third month
postoperatively in seven of the nine patients with acromegaly. Four of
the five patients with prolactinoma had normal serum PRL levels
postoperatively. Five patients had diabetes insipidus, 10 patients had
secondary hypothyroidism, 3 patients had secondary hypogonadism, and
all received appropriate replacement therapies.
|
|
|
In the early postoperative period there was a significant correlation between the peak cortisol values of the ITT and the 1-µg ACTH test (r = 0.39, P < 0.05), but no correlation was found between the ITT and the 250-µg ACTH test (r = 0.22, P > 0.05). There was a correlation between morning serum cortisol values and peak cortisol responses to the ITT in the preoperative (r = 0.48, P < 0.01) and postoperative periods (r = 0.39, P < 0.05).
One month after surgery, the ITT and the 1- and 250-µg ACTH tests
were performed in patients (patients 1, 2, and 3) with low baseline
cortisol levels (<240 nmol/L) and subnormal cortisol responses to the
ITT. Peak cortisol responses to three tests were more than 550 nmol/L
(Table 3).
|
). The sensitivity and specificity of
the 1- and 250-µg ACTH tests compared with the ITT were 100% and
66% for the 1-µg ACTH test and 100% and 17% for the 250-µg ACTH
test, respectively. The combined discrepancy between the ITT and the
1-µg ACTH test was 10.5% (2 of 19 patients), whereas it was 26.3%
(5 of 19) between the ITT and the 250-µg ACTH stimulation test.
Figures 3 and 4 show concordant results (normal or
subnormal peak cortisol responses in both tests) and discrepant results
(subnormal peak cortisol response to the 1- or 250-µg ACTH test, but
normal in the ITT, conversely) in the early postoperative period and
late postoperative period, respectively.
|
|
|
If we would have used 600 nmol/L as the cut-off value for peak cortisol responses, six patients would have had abnormal cortisol responses to the ITT. Four of the six patients would have had abnormal cortisol responses to the 1-µg ACTH test, and one would have abnormal cortisol responses to the 250-µg ACTH test. The combined discrepancy between the ITT and the 1-µg ACTH test would be 10.5% (2 of 19 patients), and it would have been 26.3% (5 of 19) between the ITT and the 250-µg ACTH test.
All the patients evaluated at the third month showed peak cortisol responses of 500 nmol/L (18 µg/dL), and 17 patients showed peak cortisol responses of 550 nmol/L (20 µg/dL). The second patient had a peak cortisol response of 523 nmol/L after the 1-µg ACTH test but more than 550 nmol/L after the 250-µg ACTH test and the ITT.
|
Discussion |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
Ammari et al. (36) evaluated the HPA axis by using the ITT and standard SAT in 30 patients with different hypothalamic pituitary pathology and reported that 9 of 17 patients with abnormal ITT results (<550 nmol/L) (20 µg/dL) had higher than this value in the SAT (discordant results were 53%). Similarly, Rasmuson et al. (27) determined that 4 of 16 patients with subnormal cortisol responses to the ITT had higher than 550 nmol/L (20 µg/dL) cortisol response to the 250-µg ACTH test in patients with verified or suspected hypothalamic-pituitary disorders and on long-term glucocorticoid therapy. In another study performed after 6 weeks after pituitary surgery, it has been shown that the SAT has poor diagnostic use and may be misleading if used as a screening test (35).
Recently, as partially shown in this study, 21 patients with pituitary adenomas were evaluated for functional reserve of HPA axis within 2 weeks after pituitary surgery by using the ITT or glucagon test and standard SAT, and total discrepancies were found as 24% between the tests (14).
In this study, five of six patients with subnormal cortisol responses (<550 nmol/L) (20 µg/dL) to the ITT had cortisol values higher than 550 nmol/L (20 µg/dL) in the early postoperative period after the SAT. In the early postoperative period, the total discrepancy was 26.3% (5 of 19) between the SAT and the ITT. We want to indicate that normal adrenals may still respond to a pharmacologic dose of 250 µg even when pituitary function is impaired acutely, so standard SAT is not a reliable screening test for the assessment of HPA axis just after pituitary surgery.
Some investigators suggested that those discordant results between the SAT and the ITT may be formed because of supraphysiological doses of ACTH (26, 27). The standard doses of ACTH used in the SAT is at least 1000 times higher for maximal adrenal stimulation (37) and it is unreliable until sufficient time has elapsed to allow the development of atrophy due to ACTH deficiency. Daidoh et al. (30) determined that the minimal dose of ACTH inducing peak and incremental cortisol values equivalent to those obtained with 250 µg ACTH is 0.5 µg ACTH in an iv dose-response study with normal subjects. However, 1 µg ACTH is probably the lowest physiological ACTH dose that will produce a maximal cortisol (38). Although the 1-µg test is accepted as the lowest physiologic dose in the evaluation of adrenal gland functions, this test is not standardized, and a number of problems may be emerged in nonspecialized centers without adequate training in the iv delivery of such a small quantity of the drug.
A number of different studies showed that the 1-µg ACTH test is more sensitive than the 250-µg ACTH test for assessment of the HPA axis (26, 27, 28, 32). We performed the 1-µg ACTH test in the early (on days 4–11) postoperative period and compared with the SAT and the ITT. We demonstrated that the 1-µg ACTH test is better correlated with the ITT than the 250-µg ACTH test. Only two patients with subnormal cortisol response to the ITT had normal cortisol responses to the 1-µg ACTH test. However, five of six patients had adequate cortisol response to the 250-µg ACTH test. There was no correlation between peak cortisol values of the 250-µg ACTH test and the ITT, whereas there was significant correlation between the 1-µg ACTH test and the ITT.
Tordjman et al. (26) compared the cortisol responses of 1, 5, and 250 µg doses of ACTH in 10 patients with pituitary disease and abnormal results of the ITT or metyrapone test, 9 patients with pituitary disease and normal results of the ITT or metyrapone test, and 7 normal subjects. They suggested that the 1-µg ACTH stimulation test can replace standard the 250-µg ACTH stimulation test for the evaluation of milder forms of secondary adrenal insufficiency. Similarly, Rasmuson et al. (27) reported that none of 16 patients with subnormal cortisol responses (<500 nmol/L) (18 µg/dL) to the ITT showed adequate response to the 1-µg ACTH test. In these two studies, the cut-off value for adequate cortisol responses for the ITT was lower than ours. In our study, two of seven patients with inadequate responses to the ITT showed adequate response to the 1-µg ACTH test. Abdu et al. (39) showed that the sensitivity of the 1-µg ACTH test was 100%, whereas the 250-µg ACTH test produced 3% falsely reassuring results when the pass criterion was more than 600 nmol/L (22 µg/dL).
Mayenknecht et al. (40) evaluated the 1-µg ACTH test and the 250-µg ACTH test in 44 patients with pituitary disease and 35 control subjects. They found a high correlation between 30- and 60-min cortisol levels in the 250-µg ACTH test and between cortisol responses to the 1-µg ACTH test and 30-min cortisol responses of the 250-µg ACTH test. In our study, we found that peak cortisol value was reached at 30 min in 37% and at 60 min in 63% of the patients in the 250-µg ACTH test. In the 1-µg ACTH test, 84% of patients had the peak cortisol value at 30 min and 16% at 60 min. For this reason, we have preferred to take peak cortisol values at 30 or 60 min instead of 30-min cortisol values as a criterion for interpretation of the 1- and 250-µg ACTH tests.
It has been suggested that the morning cortisol levels may be a predictor whether further evaluation of the HPA axis by dynamic testing is necessary or not (11, 13, 23, 27, 41, 42). Watts and Tindall (11) suggested that in 2–3 days after pituitary surgery the morning serum cortisol value of more than 250 nmol/L (9.0 µg/dL) indicates a normal reserve of the HPA axis and no further test or treatment is necessary, and the patients with basal cortisol values of less than 240 nmol/L (8.7 µg/dL) should be discharged receiving glucocorticoid therapy, and then at the first month after surgery they should be reevaluated for corticotrophin deficiency. In the other studies, similar suggestions have been reported (13, 41). In this study, basal cortisol levels correlated with peak cortisol responses to the ITT in the pre- and postoperative periods. The morning serum cortisol levels were found less than 250 nmol/L (9.0 µg/dL) in four of six patients with abnormal ITT results. Two of these six patients had higher cortisol values than 250 nmol/L (9.0 µg/dL), such as 349 nmol/L (12.6 µg/dL) and 366 nmol/L (13.3 µg/dL). Although the morning cortisol level is often indeterminant, values over 500–550 nmol/L (18–20 µg/dL) would certainly indicate normal adrenal function and such patients could avoid further testing. We suggest that, unless the basal cortisol value was higher than 500–550 nmol/L, physiologic dose glucocorticoid should be given until the establishment of normal HPA axis.
In ordinary clinical practice, glucocorticoid therapy is given against the acute adrenal insufficiency during surgery and the postoperative period (43). It remains customary to use glucocorticoid therapy during the course of the operation and in the immediate postoperative phase. In recent studies, it has been demonstrated that glucocorticoid administration is not necessary in patients undergoing pituitary surgery pre- or perioperatively (8, 9, 13, 44). In addition, Arafah et al. (9) demonstrated that pituitary function may be normalized in one or two days postoperatively. Hout et al. (8) showed that routine glucocorticoid therapy is not indicated in patients with normal HPA axis preoperatively. The same investigators suggested an alternative approach that glucocorticoid therapy may be given for 1–2 days. We gave prednisolone before anesthesia induction and stopped within 36 h of surgery in the patients, including a patient with abnormal ITT results preoperatively. It is difficult to decide whether glucocorticoid replacement therapy is, or is not, necessary for a patient after pituitary surgery. Although a glucocorticoid replacement can prevent adrenal insufficiency in patients with permanent corticotroph cell damage, the overreplacement may cause a suppression of ACTH secretion and frustrate postoperative evaluation of HPA integrity (6). In the presence of suspicious secondary adrenal failure, witholding the steroids may be harmful to the patients and the present study clearly demonstrates that physiologic glucocorticoid replacement should be given until the time as the HPA axis can be appropriately tested. In this way, one would avoid adrenal insufficiency while also avoiding overreplacement and subsequent suppression of the HPA axis. The ACTH deficiency demonstrated in the early postoperative period may not be permanent (7, 9, 11). We clearly demonstrated that impairment of the HPA axis in the immediate postoperative period was improved at the first or third month after pituitary surgery, and either of these tests in the late postoperative period (i.e. at 2–3 months) might suffice in testing the HPA axis. As shown in the study, it is an open question as to which of these is the best methodology to determine adrenal functions in the late postoperative period.
In conclusion, because the patients, particularly those with macroadenomas, may have secondary adrenal insufficiency, a great attention should be given to the adrenal functions in the postoperative period. Repeated morning serum cortisol levels may be an indicator of postoperative HPA function, but it has limited usefulness in the establishment of probable impaired HPA axis. The 1-µg ACTH stimulation test is more sensitive than the 250-µg ACTH test in comparison with the ITT, for the evaluation of the HPA axis, in the immediate postoperative period after pituitary surgery. However, neither the ITT nor ACTH tests (1 or 250 µg) correctly estimates HPA axis status late postoperatively. Our data indicate that the use of dynamic testing may be inappropriate in the early postoperative period, particularly the use of the 250-µg test, which may have a low sensitivity to identify secondary adrenal insufficiency with respect to the ITT. Our results also indicate that HPA axis dysfunction may be normalized at 1–3 months after pituitary surgery and dynamic testing might be better delayed until the late postoperative period, but the "safe" option is to keep patients on physiologic glucocorticoid replacement for 1–3 months postsurgery and then assess the HPA axis.
Received February 11, 2000.
Revised June 22, 2000.
Accepted June 29, 2000.
|
References |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  M. Christ-Crain, S. Jutla, I. Widmer, O. Couppis, C. Konig, H. Pargger, J. Puder, R. Edwards, B. Muller, and A. B. Grossman Measurement of Serum Free Cortisol Shows Discordant Responsivity to Stress and Dynamic Evaluation J. Clin. Endocrinol. Metab., May 1, 2007; 92(5): 1729 - 1735. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Agha, J. W. Tomlinson, P. M. Clark, G. Holder, and P. M. Stewart The Long-Term Predictive Accuracy of the Short Synacthen (Corticotropin) Stimulation Test for Assessment of the Hypothalamic-Pituitary-Adrenal Axis J. Clin. Endocrinol. Metab., January 1, 2006; 91(1): 43 - 47. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Salvatori Adrenal Insufficiency JAMA, November 16, 2005; 294(19): 2481 - 2488. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. H. Courtney, A. S. McAllister, P. M. Bell, D. R. McCance, H. Leslie, B. Sheridan, and A. B. Atkinson Low- and Standard-Dose Corticotropin and Insulin Hypoglycemia Testing in the Assessment of Hypothalamic-Pituitary-Adrenal Function after Pituitary Surgery J. Clin. Endocrinol. Metab., April 1, 2004; 89(4): 1712 - 1717. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. K. Gleeson, B. R. Walker, J. R. Seckl, and P. L. Padfield Ten Years on: Safety of Short Synacthen Tests in Assessing Adrenocorticotropin Deficiency in Clinical Practice J. Clin. Endocrinol. Metab., May 1, 2003; 88(5): 2106 - 2111. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. R. Williamson, M. Albert, and M. Charneux Corticosteroids for Patients With Septic Shock JAMA, January 1, 2003; 289(1): 42 - 42. [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 |
