| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Studies |
Sir George E. Clark Metabolic Unit (M.I.W., A.P.H., D.R.M., D.R.H., A.B.A.), Department of Radiology (E.M.M.), and Regional Endocrine Laboratory (B.S.), Royal Victoria Hospital, Belfast, Northern Ireland BT12 6BA
Address all correspondence and requests for reprints to: Prof. A. B. Atkinson, Sir George E. Clark Metabolic Unit, Royal Victoria Hospital, Belfast, Northern Ireland BT12 6BA.
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The most direct method of distinguishing between pituitary and ectopic ACTH secretion is bilateral inferior petrosal sinus sampling (IPSS) (9). The high diagnostic accuracy of this technique has led ourselves and others to consider it a routine investigation in the differential diagnosis of ACTH-dependent Cushing’s syndrome (10, 11, 12). However, the procedure is not available at all centers (9), and although safe in experienced hands, significant complications have been reported (13).
For this reason, other indirect, noninvasive tests continue to have an important role (9, 14). Unfortunately, these tests have been associated with limited diagnostic accuracy, especially with regard to occult ectopic ACTH syndrome (15, 16, 17). For example, in one series, a quarter of patients with ectopic ACTH secretion responded to high dose dexamethasone with greater than 50% suppression of urinary 17-hydroxysteroids (15), whereas in another report, 2 of 14 patients with an ectopic source of ACTH produced a more than 20% rise in serum cortisol after CRH stimulation (17). Reliance on such tests has led to the inappropriate referral of a significant proportion of patients with ectopic ACTH secretion for pituitary surgery (15, 16). Given these difficulties, a number of indirect tests have been reevaluated, both alone and in combination, to address the overall accuracy of the noninvasive diagnostic approach (17, 18, 19, 20, 21, 22, 23, 24, 25).
In particular, attention has focused on new stringent diagnostic criteria designed to maximize specificity and thus avoid the misclassification and mismanagement of patients with ectopic ACTH syndrome (17, 22, 23). Flack and colleagues, for example, have proposed that during the high dose dexamethasone (HDD) test (i.e. 2 mg every 6 h for 48 h), more than 90% suppression of 24-h urinary free cortisol should be required for the diagnosis of pituitary disease to achieve 100% specificity (22). In their study use of the traditional greater than 50% suppression criterion would have misdiagnosed 4 of the 10 patients with ectopic ACTH secretion. In clinical practice 0800 h serum cortisol is often used instead of urinary free cortisol on the basis of limited evidence that suppression criteria established for urinary free cortisol can be extended to serum cortisol (18, 20, 21). Although suppression criteria to obtain 100% specificity with the HDD test using serum cortisol as an end point have not been established, it seems appropriate to adopt the greater than 90% criterion proposed for urinary free cortisol by Flack et al. (22).
There are differing views on criteria for the diagnosis of pituitary-dependent disease after the CRH stimulation test (17, 18, 19, 25, 26, 27, 28). A useful meta-analysis of all relevant data published before 1989 indicated that a serum cortisol response more than 20% above basal for the diagnosis of pituitary dependency would achieve a sensitivity of 91% and a specificity of 95% (29). However, it was suggested that a greater than 50% serum cortisol response would be required to exclude confidently ectopic ACTH syndrome (29).
Previous studies have shown that when the results of the CRH test are combined with those of a separately performed HDD test, the overall diagnostic performance exceeds that obtained with either test alone (18, 19, 25). No study has assessed the usefulness of this combined approach after the stringent diagnostic criteria outlined above have been applied to both tests.
We report our experience with 53 consecutive cases of ACTH-dependent Cushing’s syndrome whose investigation included HDD testing (using 0800 h serum cortisol as the end point) and bilateral inferior petrosal sinus sampling. A number of patients also underwent CRH testing. The usefulness of the various diagnostic tests is examined along with the implications of applying the new stringent diagnostic criteria.
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
We reviewed the case records of 53 consecutive patients with ACTH-dependent Cushing’s syndrome. Our unit provides a regional endocrine service for the whole of Northern Ireland, and the present series is therefore representative of all cases in a defined geographical area. The diagnosis of hypercortisolism was suggested by increased 24-h urinary free cortisol excretion and lack of suppression of serum cortisol during a low dose dexamethasone suppression test (0.5 mg every 6 h for 48 h) (30). All patients had detectable ACTH levels and thus underwent further investigation to determine the source of ACTH secretion. The final diagnoses were as follows. Forty-four patients had pituitary-dependent Cushing’s disease established on the basis of histological confirmation or apparent cure or significant biochemical improvement after pituitary microsurgery. One patient had ectopic ACTH secretion from an oat cell carcinoma. In 8 patients the diagnosis remains unconfirmed. Four of these, for various reasons, had bilateral adrenalectomy performed in preference to pituitary surgery, 1 had medical therapy only, and 3 unconfirmed cases underwent pituitary microsurgery but had normal pituitary histology and showed no postoperative biochemical improvement. Our main analysis is based on the 44 confirmed cases of Cushing’s disease and the 1 confirmed case of ectopic ACTH syndrome.
Bilateral inferior petrosal sinus sampling
This was attempted in all 53 cases in the present series as previously described (10), without the routine use of CRH stimulation. In 1 patient (case 44), CRH at a dose of 1 µg/kg was administered after obtaining basal samples, and further petrosal and peripheral samples were drawn after 1, 3, 5, 10, and 20 min. In another patient (case 39), bilateral inferior petrosal sinus sampling was initially performed without CRH stimulation and was then repeated using CRH as described.
Plasma ACTH (picograms per mL) was estimated by RIA as previously described (31). The coefficient of variation was 7.1% at 98 pg/mL (1 pg/mL = 0.23 pmol/L). Based on our early results from petrosal sinus sampling (10), we considered a ratio of the higher of the two petrosal sinus plasma ACTH levels to that of the peripheral venous plasma ACTH of 1.5 or greater as being consistent with pituitary-dependent hypercortisolism, but not diagnostic. In view of evidence from other series that patients with ectopic ACTH syndrome may occasionally have ratios between 1.5–2.0 (11), we used a ratio of 2.0 or more as our diagnostic cut-off. In each case the ratio obtained was interpreted alongside the actual data, with the variability of the ACTH assay in mind, to ensure that the ratio represented a meaningful difference in ACTH levels. Unless otherwise stated, all petrosal/peripheral ACTH ratios given are based on the higher of the two petrosal ACTH values.
HDD testing
Formal dexamethasone testing was carried out in 40 of the 45 patients with confirmed diagnosis and in the 8 patients in whom the diagnosis remains unconfirmed. Low dose dexamethasone was administered for 48 h (0.5 mg every 6 h), followed by high dose dexamethasone for 48 h (2 mg every 6 h). Serum cortisol was measured at 0800 h on day 1, just before the first 0.5-mg dose, and then at 0800 h daily until completion of the test (days 2–5). The percent suppression of serum cortisol was calculated as follows: % suppression = (basal value - final value)/basal value x 100. Our normal practice is to take the day 1 value as basal. In 3 of the 45 patients with confirmed diagnosis (2 with pituitary dependency and 1 with ectopic ACTH secretion) the HDD test was not directly preceded by a low dose test. An additional 2 patients (pituitary dependent) suppressed fully (serum cortisol undetectable) during preliminary low dose dexamethasone testing; therefore, the HDD test was deemed unnecessary (2). (Both of these patients had evidence of cyclical hypercortisolism.) In this analysis these 2 patients have been considered fully responsive to HDD. Serum cortisol was estimated by direct RIA as previously described (30).
CRH test
The response to ovine CRH was assessed in 23 of the 45 cases with confirmed diagnosis (including the case due to ectopic ACTH secretion) and in 4 of the cases with unconfirmed diagnosis. Ovine CRH was administered by iv bolus (1 µg/kg) at 1400 h, and serum cortisol was measured at the following times: -15, -5, 0, 5, 10, 15, 30, 60, 90, and 120 min. A maximal rise more than 50% above the basal average was taken as indicative of pituitary dependency (29).
Pituitary imaging
Forty-three of the 45 patients with confirmed diagnosis and all of the patients with unconfirmed diagnosis had fourth generation, thin cut pituitary computerized tomography (CT) performed. One patient (case 42) had magnetic resonance (MR) imaging (T1 weighted images after iv gadolinium administration) of the pituitary fossa performed instead of CT. The patient with ectopic ACTH syndrome did not have pituitary imaging.
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
. Twenty-five of the
confirmed pituitary cases had corticotroph adenomas confirmed by
histology, and 1 case had pituitary hyperplasia. In the remaining 19
pituitary cases, the diagnosis was based on apparent cure or
significant biochemical improvement after pituitary surgery. Petrosal
sinus sampling was performed in all 45 cases with confirmed diagnosis,
and bilateral catheterization was successfully achieved in 38 of 45
(84%). In 6 cases only unilateral catheterization was possible, and in
1 case, a 10-yr-old boy, only the right jugular bulb could be
catheterized (case 33). (Despite the suboptimal sampling site, a
step-up in ACTH was observed, and his results have been included in the
analysis). One patient developed a left groin hematoma after the
procedure, but otherwise there were no complications, and overall
patient tolerance was good. Several patients described an awareness of
noise after insertion of the catheter into the petrosal sinus. Of the
44 proven pituitary cases, 36 had petrosal/peripheral ACTH ratios of
2.0 or more (without CRH stimulation), giving a sensitivity of 82%.
One of these 36 had CRH stimulation, after which the
petrosal/peripheral ACTH ratio increased from 7.4 to 10.0 (case 44).
Forty-one of the 44 pituitary-dependent cases (93%) had a
petrosal/peripheral ACTH ratio of 1.5 or more.
|
The results of HDD testing (Table 1
) are illustrated in Fig. 1. Of the 44 proven pituitary cases, 41
met the traditional greater than 50% suppression criterion, but only
21 met the more stringent greater than 90% criterion. The patient with
ectopic ACTH syndrome failed to suppress by either criterion.
|
) are illustrated in Fig. 2. Sixteen of the 23 confirmed pituitary
cases who had the test performed produced a serum cortisol rise
exceeding 50%. The patient with ectopic ACTH secretion showed a flat
cortisol response. Pituitary CT was performed in 43 of the 44 confirmed
cases of Cushing’s disease. A pituitary microadenoma was identified in
17 patients, and a macroadenoma was found in 3 patients, 1 of whom had
evidence of suprasellar invasion (Table 1). In the remaining
pituitary-dependent case a pituitary microadenoma was identified on MR
imaging after gadolinium enhancement (case 42).
|
). If a diagnosis of pituitary
dependency was made on the basis of either a more than 90% suppression
of serum cortisol after HDD, a greater than 50% cortisol rise after
CRH, or both, then 78% sensitivity (18 of 23) was achieved. If,
however, a positive response to both tests was required to make a
confident diagnosis of pituitary disease, then only 35% (8 of 23) of
pituitary-dependent patients were correctly diagnosed. Four of the 5
pituitary-dependent patients who had a negative response to both tests
had a petrosal/peripheral ACTH ratio greater than 2.0 (cases 1, 2, 12,
and 43 with ratios of 2.9, 10.6, 2.6, and 8.9, respectively), and the
fifth (case 26) had a ratio of 1.8.
|
. During
follow-up (as shown in Table 3), no evidence of an ectopic source of
ACTH secretion emerged in any of these patients. Cases 46–48 were
thought to have pituitary-dependent Cushing’s syndrome, but pituitary
surgery was unsuccessful and pituitary histology was negative. In case
46, hypophysectomy was complicated by excessive bleeding, and bilateral
adrenalectomies were subsequently performed, the histology of which
showed bilateral hyperplasia. This patient died of an acute myocardial
infarction after 6 yr of follow-up. In case 47, only a small volume of
pituitary tissue was present in the sella turcica. She, too, had
bilateral adrenalectomies, and histological examination confirmed
bilateral hyperplasia. In case 48 the first attempt at pituitary
surgery was abandoned due to hypoxia and unusual anatomy. At a repeat
procedure 70% of the pituitary was successfully removed, but this was
complicated by postoperative meningitis. She died of bronchopneumonia 3
months after surgery. At autopsy, no residual adenoma was identified in
the pituitary gland, nor was an occult source of ACTH secretion
found.
|
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
In the present series of 53 cases, the diagnosis remains unconfirmed in 8 patients (15%). This is in keeping with a previous series in which 32 of 281 cases were undiagnosed (11%) even after biochemical testing and IPSS (11). Although such patients represent an important subgroup in clinical practice, we have not included them in our main analysis as we believe that the sensitivity of diagnostic tests is most reliably calculated in those patients with confirmed diagnosis. This is similar to the approach taken in previous series (11, 17, 18). It must be remembered that in patients with occult ectopic ACTH syndrome, there is often an interval of several years from the clinical diagnosis of hypercortisolism until the appearance of the underlying ACTH-secreting tumor (16).
In 44 patients with confirmed Cushing’s disease, bilateral IPSS without CRH stimulation yielded a petrosal/peripheral ACTH ratio of 2.0 or more in 36 (82%). After HDD, serum cortisol was suppressed by more than 90% in only 21 of 44 (48%). Of 23 proven pituitary cases who had a CRH test, 16 (70%) produced a cortisol rise of greater than 50%. Pituitary CT scanning showed evidence of a pituitary lesion in 20 of 43 (47%). This is in accordance with previous series that reported detection of pituitary adenomas on CT scan in 30–53% of patients with Cushing’s disease (33, 34). With MR imaging higher detection rates have been reported, particularly when iv contrast enhancement is used (34, 35, 36, 37). However, as pituitary microadenomas are relatively common in the general population (36, 38), it is generally agreed that pituitary imaging is of only limited diagnostic value in this situation (9, 14); it is the most useful if there is clear evidence of a pituitary macroadenoma, a situation seen in only 20% of cases of Cushing’s disease (8). Although pituitary imaging has only a small role to play in deciding whether to proceed to pituitary surgery (9), there is little doubt that MR imaging is the most accurate method of tumor localization before surgery (9, 37).
There is now substantial evidence that bilateral IPSS is the most reliable method of distinguishing between pituitary and nonpituitary ACTH excess (11, 14). Indeed, when combined with CRH stimulation, a diagnostic accuracy of 100% has been reported in a series of 220 patients, including 17 with ectopic ACTH syndrome (11). Although in the past we have not routinely used CRH stimulation, we now advocate this approach, in keeping with most other centers (11, 14). Obviously, the present analysis reflects our previous experience using IPSS without CRH stimulation and to some extent underestimates the current usefulness of the procedure when CRH is routinely used.
Even with CRH stimulation, the results of IPSS can be misleading in certain unusual situations. Two cases of proven ectopic ACTH syndrome have been reported in whom false positive results occurred after CRH stimulation (39). Periodic hypercortisolism was documented in each, and it was suggested that the patients may not have been hypercortisolemic at the time of sampling, emphasizing that petrosal sinus sampling is only a useful discriminatory test in the presence of hypercortisolism (40).
The need for simultaneous bilateral sampling has previously been emphasized (10) and is supported by the data from the present analysis. Of the 38 patients with proven pituitary disease who had bilateral sampling successfully performed, 27 had a lower petrosal/peripheral ACTH ratio of less than 2.0. It is recognized, however, that even in the most experienced hands, bilateral catheterization is not always possible (11). Results from unilateral sampling are only of diagnostic value if an elevated petrosal/peripheral ratio is found. Similarly, the need for sampling higher than that in the jugular bulb has long been recognized, as blood from the jugular bulb is admixed with blood from different areas of the brain (41). In a young boy from our series (case 33) sampling was only possible from the right jugular bulb due to the patient’s age and sinus development. In that case the ratio of jugular bulb to peripheral ACTH was 2.0. If anything, one might have expected a higher ratio had the inferior petrosal sinuses been sampled, and his results were therefore interpreted as indicative of pituitary disease.
For many years the HDD suppression test has retained a central role in the differential diagnosis of Cushing’s syndrome (9, 14, 42), and traditionally greater than 50% suppression of urinary 17-hydroxysteroid after HDD for 48 h has been taken as evidence of pituitary disease (2). This suppression criterion has subsequently been extended to urinary free cortisol (43, 44) and serum cortisol (18, 20, 21). Although this provides a sensitive test for the diagnosis of pituitary dependency (93% sensitivity in the present series), it is clear from previous reports that patients with ectopic ACTH syndrome may achieve greater than 50% suppression of serum cortisol after HDD (20, 21). In the absence of specific criteria for serum cortisol suppression after HDD to achieve 100% specificity, we have used the greater than 90% suppression criterion developed for urinary free cortisol (22). Given limited evidence that the suppressibility of serum cortisol in response to HDD may be slightly less than that of urinary free cortisol (21), it is likely that the new 90% criterion established for urinary free cortisol can be extended to serum cortisol without loss of test specificity.
Although various centers, including our own, routinely use IPSS in the differential diagnosis of ACTH-dependent Cushing’s syndrome (10, 11, 12), others advocate a more selective use of IPSS, based on the results of biochemical and radiological investigations (37). For example if a patient with ACTH-dependent Cushing’s syndrome has a clearly defined adenoma on CT or MR scanning, suppresses on HDD testing by 50% or more, and has no evidence of a neoplasm on chest x-ray, many physicians may consider referral for pituitary surgery without IPSS. Had we used such an approach, based on pituitary imaging and HDD testing, 19 of the 44 proven pituitary cases (43%) would have been diagnosed without the need for IPSS. However, any reduction in the number of cases requiring IPPS must be balanced against the small, but significant, risk of failing to diagnose ectopic ACTH syndrome and inappropriate referral of such patients for pituitary surgery. Given the diagnostic limitations of both pituitary imaging and traditional HDD testing already discussed, we do not recommend that patients be referred for pituitary surgery on the basis of these tests alone.
If IPSS is to be avoided, then whatever diagnostic strategy is used in its place should be highly specific so that the small number of patients with ectopic ACTH secretion are accurately diagnosed. One approach might be to employ a combination of biochemical tests using stringent diagnostic criteria to achieve 100% specificity. In the present analysis 23 patients with proven pituitary disease underwent both HDD and CRH testing. Using stringent response criteria for the diagnosis of pituitary dependency, the sensitivity of the 2 tests combined was 78% (18 of 23) if a positive response to 1 or other of the tests (or both) is taken as evidence of pituitary disease. If these criteria were 100% specific, then a positive response to either test would completely exclude the possibility of ectopic ACTH secretion and obviate the need for IPSS. Unfortunately, the evidence to date does not permit this degree of confidence. Although the data used to develop the stringent response criteria were based on large numbers of patients, the actual numbers with ectopic ACTH secretion were still relatively small (22, 29). Since publication of the greater than 90% suppression criterion for urinary free cortisol (22), at least 1 patient with an ACTH-secreting bronchial carcinoid tumor has been reported in whom urinary free cortisol was suppressed by more than 90% after HDD (14). With regard to the CRH test, a separate patient with ectopic ACTH secretion has been noted to produce a greater than 50% cortisol rise in response to CRH (17). It seems only a matter of time before more such cases are reported. Based on the evidence available, we believe that reliance on 1 indirect test (even with the stringent response criteria described) does not afford the degree of confidence required to refer a patient with ACTH-dependent Cushing’s syndrome for pituitary surgery.
A more cautious approach might be to insist that stringent criteria for both the HDD test and the CRH test are fulfilled before a confident diagnosis of pituitary dependency is made. In the present series this would have identified only 35% (8 of 23) of proven pituitary cases. The chance of a patient with ectopic ACTH secretion surpassing both of the stringent response criteria would appear small, and it could be argued that further testing in such cases would be unnecessary. However, bilateral IPSS would be required to establish the diagnosis in the remaining 65%. It has been suggested previously that when the HDD and CRH tests are combined, only cases with discordant results require further evaluation to exclude ectopic ACTH secretion (19). Clearly, when stringent response criteria are applied, such recommendations are no longer appropriate, as in the present study a significant number of cases with proven pituitary disease had concordant negative results (5 of 23).
In conclusion, bilateral inferior petrosal sinus sampling without CRH stimulation was diagnostic of pituitary dependency in 82% of confirmed cases of Cushing’s disease. The procedure was safe and well tolerated. If the procedure is not to be performed in all cases, then it is essential that stringent response criteria are applied to whatever alternative tests are used, and preferably a positive response to more than one indirect test should be sought before referring a patient for surgery. This means that a significant proportion of patients will still require petrosal sinus sampling.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Received May 14, 1999.
Revised December 29, 1999.
Accepted January 11, 2000.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  N. Hayashi, M. Kurimoto, M. Kubo, N. Kuwayama, K. Kurosaki, S. Nagai, and S. Endo The Impact of Cavernous Sinus Drainage Pattern on the Results of Venous Sampling in Patients With Suspected Cushing Syndrome AJNR Am. J. Neuroradiol., January 1, 2008; 29(1): 69 - 72. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F Castinetti, I Morange, H Dufour, P Jaquet, B Conte-Devolx, N Girard, and T Brue Desmopressin test during petrosal sinus sampling: a valuable tool to discriminate pituitary or ectopic ACTH-dependent Cushing's syndrome Eur. J. Endocrinol., September 1, 2007; 157(3): 271 - 277. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 I. S. Kaskarelis, E. G. Tsatalou, S. V. Benakis, K. Malagari, I. Komninos, D. Vasiliadou, S. Tsagarakis, and N. Thalassinos Bilateral inferior petrosal sinuses sampling in the routine investigation of Cushing's syndrome: a comparison with MRI. Am. J. Roentgenol., August 1, 2006; 187(2): 562 - 570. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. Swearingen, L. Katznelson, K. Miller, S. Grinspoon, A. Waltman, D. J. Dorer, A. Klibanski, and B. M. K. Biller Diagnostic Errors after Inferior Petrosal Sinus Sampling J. Clin. Endocrinol. Metab., August 1, 2004; 89(8): 3752 - 3763. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
I. Ilias, R. Chang, K. Pacak, E. H. Oldfield, R. Wesley, J. Doppman, and L. K. Nieman Jugular Venous Sampling: An Alternative to Petrosal Sinus Sampling for the Diagnostic Evaluation of Adrenocorticotropic Hormone-Dependent Cushing's Syndrome J. Clin. Endocrinol. Metab., August 1, 2004; 89(8): 3795 - 3800. [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 |
