| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Studies |
Departments of Medicine (G.L.B., D.A.R., S.E.), Radiology (H.G.), Pathology (K.K.), and Neurosurgery (H.S.S.), University of Toronto, Toronto, Ontario, Canada M5G-1X5
Address all correspondence and requests for reprints to: Dr. Shereen Ezzat, University of Toronto, 600 University Avenue #429, Toronto, Ontario, M5G-1X5, Canada.
| Abstract |
|---|
|
|
|---|
| Introduction |
|---|
|
|
|---|
Imaging studies are an imperfect method for identifying corticotropic microadenomas. Scanning by computerized tomography (CT) has a sensitivity of 1560% (4, 5). Contrast enhanced magnetic resonance imaging (MRI) has been reported to have a sensitivity ranging from 5491% (4, 6, 7, 8). However, false positives may occur with imaging studies; for example, MRI scanning detects pituitary microadenomas in 10% of the healthy population (9). A safe, accurate, and affordable method for localizing pituitary adenomas would be valuable for managing patients with Cushings disease.
Drainage of venous blood from each half of the pituitary gland tends to flow from the ipsilateral cavernous sinus into the inferior petrosal sinus (IPS) and then into the internal jugular vein (10, 11). Because of the proximity of the IPS to the pituitary gland, ACTH levels in the sinus are higher than in the peripheral blood. Inferior petrosal sinus sampling (IPSS), therefore, can be used to assist in the diagnosis of Cushings disease. Moreover, IPSS provides an alternative to imaging for localizing the ACTH source within the pituitary gland. However, previous studies have yielded conflicting results on the accuracy of IPSS to localize corticotropic adenomas (12, 13, 14, 15, 16, 17). The focus of this study is to compare the accuracy of IPSS with imaging to localize an ACTH-secreting source definitively established by pituitary histopathology.
| Patients and Methods |
|---|
|
|
|---|
Screening evaluations were conducted on all patients as either a 24-h urinary free cortisol (UFC) or a 1 mg overnight dexamethasone suppression test. High dose dexamethasone suppression consisted of 2 mg q6h for 48 h or as a single 8 mg overnight dexamethasone suppression test. ACTH levels were measured either by a radioimmunoassay or by immunoradiometric assay.
MRI examinations were performed using a 1.5-T superconducting magnet scanner. Sagittal and coronal T1-weighted images were acquired at 3 mm intervals before and after gadolinium-enhancement. Direct evidence for an intrasellar lesion of a low signal intensity was sought (definite lesion) or in its absence, indirect evidence such as upward bulging of the gland, lateral displacement of the stalk, or an asymmetric sellar floor (equivocal lesion). Imaging studies were independently reviewed by a single radiologist who was blinded to both the results of IPSS and the patients histopathologic findings.
Bilateral IPSS was performed after informed consent with simultaneous central and peripheral ACTH measurements obtained before and after administration of ovine CRH 100 µg iv bolus (Ferring Inc.). Catheter placement was verified before and after sampling using gentle injection of contrast. ACTH measurements were taken at 0, 2, 5, 10, and 15 min after CRH injection. In six patients, desmopressin 10 µg iv bolus was used in place of CRH (18) because of an interruption in the supply of CRH. The maximum simultaneous ratio of IPS to peripheral ACTH and the maximum intersinus gradients were then calculated. A central to peripheral gradient greater than or equal to 2.0 at baseline, or 3.0 post-CRH, was considered diagnostic of Cushings disease (12). An intersinus gradient greater than or equal to 1.4 was evidence of lateralization. An intersinus gradient less than 1.4 was indicative of either bilateral disease or a midline lesion.
Operative findings, supplemented by histological details and clinical outcomes, served as the gold standard for determining the true location of each patients pituitary lesion. Locations were categorized as either right-sided unilateral, left-sided unilateral, or midline/bilateral. Radiographic studies that failed to identify a lesion were classified as unsatisfactory. Similarly, IPSS studies that failed to cannulate both petrosal veins were classified as unsatisfactory. Patients who had both tests available were then classified as having results that were either concordant or discrepant, depending on whether the two tests did or did not demonstrate the same location of pathology.
We determined the accuracy of each diagnostic test by comparing it with intraoperative findings and the results of tissue histopathology. Perfect confirmation (such as when the test indicated a right-sided unilateral lesion and pathology indicated a right-sided unilateral lesion) was classified as "confirmed results". Partial disagreement about location (such as when one test indicated a right-sided unilateral lesion and the other test indicated a midline/bilateral lesion) was classified as "moderately discordant". Substantial disagreement about location (such as when one test result indicated a right-sided unilateral lesion and the other test indicated a left-sided unilateral lesion) was classified as "greatly discordant".
We calculated the frequency of unsatisfactory studies out of the total number of attempted studies. Confidence intervals were calculated using Fleiss formula (19). Rates of redundancy between the two tests were calculated using a similar method. Accuracy was calculated by examining the frequency of confirmed results and erroneous results. Additionally, separate analyses examined the accuracy of each test depending on whether results were concordant or discrepant with the alternative test. All P-values were two-tailed and unadjusted for multiple comparisons.
| Results |
|---|
|
|
|---|
Results of sellar imaging, operative findings, and final pathology are
described in Table 1
. MRI was performed
in 31 patients; the remaining 6 could not undergo MRI because of size
constraints and therefore had CT scans. A discrete lesion was
definitely visualized by MRI in 18 patients. MRI scanning was equivocal
in 6 and negative in 7 other patients. CT scans of the 6 patients for
whom MRI was unavailable were normal in 2 cases and equivocal in the
remaining 4 cases. Hence, imaging was satisfactory in 28 patients and
failed to reveal a lesion in 9 others. No adverse consequences from
imaging were observed.
|
Unilateral pathology was discovered intraoperatively in 27 patients, bilateral lesions were found in 7 patients, and a midline lesion was identified in 1 patient. In 2 other patients, no abnormal tissue was visualized during surgery. The first of these 2 patients underwent a right hemihypophysectomy, performed on the basis of lateralization demonstrated on IPSS; the surgery was curative despite the absence of a demonstrable adenoma on pathological examination. The second of these 2 patients underwent a total hypophysectomy, and pathology demonstrated the presence of diffuse corticotropic hyperplasia.
Twenty-eight patients (76%) had a pituitary corticotropic adenoma proven histologically. Seven patients (19%) had only Crookes hyaline change identified in the pituitary; one of the latter had a proven ectopic sphenoid corticotropic adenoma, and the other 6 are assumed to have had pituitary adenomas that were not identified histologically, as they achieved clinical and biochemical remission postoperatively. One patient had nodular hyperplasia (2%), and another had a diffuse form of corticotropic hyperplasia (2%). Interestingly, both cases of hyperplasia presented with a central to peripheral gradient on IPSS, while the former was associated with and the latter without an intersinus gradient. A microadenoma was assumed or proven therefore in 28 patients, of whom 25 were cured by surgery alone. A macroadenoma was present in 7 patients (19%), of whom 4 sustained a remission following transsphenoidal surgery alone. Three other patients with macroadenomas required further therapy (radiation and/or bilateral adrenalectomy).
Neither IPSS nor imaging was ideal for localizing the pituitary lesion.
IPSS accurately localized the pituitary lesion in 26 of 37 patients,
whereas imaging correctly localized the lesion in 18 of 37 patients
(diagnostic accuracy: 70% vs. 49%, P <
0.06). Both studies were not always satisfactory (Table 2
). When results were available from
both, IPSS and imaging studies were concordant for 67% of patients and
discordant for 33% of patients (Table 3
). When the two studies were concordant,
the results correctly matched the site of the lesion in most patients
(Fig. 1
). When a discrepancy existed
between the two studies, a greater proportion of IPSS results than
imaging results correctly matched the site of the lesion (63% vs.13%,
P = 0.10). Imaging was negative in 9 patients, of which
IPSS correctly localized the site of the lesion in 8 (95% confidence
interval 5099%). Unsuccessful IPS catheterization proved to be
highly inaccurate. The procedure accurately localized the lesion in
only 1 of 3 patients who underwent unilateral petrosal sampling and
contralateral jugular vein sampling. Furthermore, the absence of an
intersinus gradient in the two cases of bilateral jugular vein sampling
falsely predicted a bilateral lesion.
|
|
|
| Discussion |
|---|
|
|
|---|
A number of studies have evaluated the role of IPSS in the diagnostic approach to patients with ACTH-dependent Cushings syndrome. One study suggested that a central to peripheral gradient in ACTH of greater than 2.0 had a sensitivity of 95% and specificity of 100% (12). By using a cut-off of 3.0 following CRH administration, the sensitivity rose to 100% (12). Subsequent studies have yielded similar results (2, 16, 20, 21, 22). In our study, a central to peripheral ratio was available in 31 patients, of whom 26 (84%) had gradients of greater than 2.0. The largest gradient (111.3) was found in a patient with corticotropic hyperplasia. In four of the five patients who did not demonstrate a central to peripheral gradient, IPSS was technically unsuccessful.
Some studies have correlated IPSS data with surgical findings (2, 3). However, many studies included fewer than a dozen patients. Furthermore, few reports have compared IPSS against histopathology. One review suggested that an intersinus gradient of greater than 1.4 could accurately predict the location of a microadenoma in 7080% of patients (2). In our study, IPSS accurately located the site of the lesion in 70% of patients. The significance of a negative intersinus gradient has achieved little attention in previous studies; in our study, an intersinus gradient less than 1.4 correlated with a bilateral tumor in three of six patients and diffuse corticotropic hyperplasia in a fourth patient.
There are several major reasons why a large intersinus ACTH gradient may not be found on IPSS despite a known underlying pituitary lesion. First, corticotrophs are primarily located in the median wedge of the adenohypophysis (23), hence, many adenomas arise in this central location and may not be associated with an intersinus gradient. Second, some patients with Cushings disease, like two in our series, have primary diffuse or nodular corticotropic hyperplasia without an adenoma. The absence of an ectopic or eutopic source of stimulation by CRH leaves the etiology of such hyperplasia unknown. As noted in our patient with diffuse hyperplasia, however, this can be an additional cause of an absent intersinus gradient on IPSS. Finally, some procedures will be technically impossible because of the marked anatomical variation in the number and size of anastomotic channels between the IPS and vertebral, basilar, and epidural venous plexuses (10).
There are also reasons why a large intersinus ACTH gradient may falsely localize the tumor. For example, Mamelak et al. (24) demonstrated asymmetrical venous drainage in as many as 39% of patients undergoing IPSS. In their study, IPSS correctly lateralized the tumor in 86% of cases with symmetrical drainage, but only 44% of patients with asymmetrical patterns. The authors suggested that routine venography is necessary to correctly interpret results of IPSS (24). Inaccurate results have also been found after previous pituitary surgery. Previous transsphenoidal surgery, however, did not alter the accuracy of IPSS in two of our patients. Earlier treatment with medications that lower serum cortisol levels may eliminate the suppression of nontumorous corticotrophs and attenuate the intersinus gradient. Ectopic location of the lesion may be an additional factor. In our series, a central to peripheral gradient was documented in a patient with an ectopic corticotropic adenoma situated within the sphenoid sinus (25).
Pathology failed to confirm the diagnosis of a corticotropic adenoma in six patients. In patients with cortisol excess, corticotrophs develop accumulation of bundles of intermediate filaments representing low molecular weight cytokeratins. This morphologic alteration, known as Crookes hyalinization, is characteristic of nontumorous corticotrophs and is the pathologic hallmark of Cushings syndrome. In our six patients, only Crookes hyalinization could be identified, indicating the presence of an adenoma that may have been too small to be detected in routine histology or lost during surgical resection but provided pathologic confirmation of the diagnosis of Cushings syndrome.
Our study has several limitations, of which three merit emphasis. First, only a small number of patients with Cushings disease underwent imaging, IPSS, and definitive surgery. This is a large sample size compared with most other analyses of IPSS, yet is not sufficient for sophisticated statistical analyses. Second, IPSS is typically an arduous procedure and in many cases is conducted only when imaging is inconclusive. This introduces selection bias, which might cause our results to under-estimate the relative accuracy of imaging and over-estimate the relative accuracy of IPSS. Third, IPSS was not performed on all patients as a screening procedure. As a consequence, our study can not provide a precise estimate of the specificity, positive predictive value, or other characteristics of test accuracy.
A great deal of controversy surrounds the question of when to perform IPSS. To justify the procedure, there must be unequivocal evidence of cortisol excess. Indeed, normal subjects or patients who have pseudo-Cushings may show an intersinus ACTH or AVP concentration gradient (17, 26). Morbidity from the procedure is low but not negligible. For example, two patients with permanent brainstem injury following IPSS have been reported in the literature (27). The test is expensive and is associated with a total fee of $1125 (Canadian dollars) at our facility including technical and professional components. Our study suggests that a key indication for IPSS is when a patients high-dose dexamethasone suppression testing is diagnostic of Cushings disease, yet imaging fails to detect an adenoma. More importantly, this procedure should be restricted to centers with a skilled and dedicated team that has expertise in invasive neuro-endocrinology.
Received November 14, 1997.
Accepted April 7, 1998.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
S. Jehle, J. E. Walsh, P. U. Freda, and K. D. Post Selective Use of Bilateral Inferior Petrosal Sinus Sampling in Patients with Adrenocorticotropin-Dependent Cushing's Syndrome Prior to Transsphenoidal Surgery J. Clin. Endocrinol. Metab., December 1, 2008; 93(12): 4624 - 4632. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. Tsagarakis, D. Vassiliadi, I. S. Kaskarelis, J. Komninos, E. Souvatzoglou, and N. Thalassinos The Application of the Combined Corticotropin-Releasing Hormone plus Desmopressin Stimulation during Petrosal Sinus Sampling Is Both Sensitive and Specific in Differentiating Patients with Cushing's Disease from Patients with the Occult Ectopic Adrenocorticotropin Syndrome J. Clin. Endocrinol. Metab., June 1, 2007; 92(6): 2080 - 2086. [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] |
||||
![]() |
D. Batista, M. Gennari, J. Riar, R. Chang, M. F. Keil, E. H. Oldfield, and C. A. Stratakis An Assessment of Petrosal Sinus Sampling for Localization of Pituitary Microadenomas in Children with Cushing Disease J. Clin. Endocrinol. Metab., January 1, 2006; 91(1): 221 - 224. [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] |
||||
![]() |
H. Raff and J. W. Findling A Physiologic Approach to Diagnosis of the Cushing Syndrome Ann Intern Med, June 17, 2003; 138(12): 980 - 991. [Full Text] [PDF] |
||||
![]() |
N. Patronas, N. Bulakbasi, C. A. Stratakis, A. Lafferty, E. H. Oldfield, J. Doppman, and L. K. Nieman Spoiled Gradient Recalled Acquisition in the Steady State Technique Is Superior to Conventional Postcontrast Spin Echo Technique for Magnetic Resonance Imaging Detection of Adrenocorticotropin-Secreting Pituitary Tumors J. Clin. Endocrinol. Metab., April 1, 2003; 88(4): 1565 - 1569. [Abstract] [Full Text] [PDF] |
||||
![]() |
V. Lefournier, M. Martinie, A. Vasdev, P. Bessou, J.-G. Passagia, F. Labat-Moleur, N. Sturm, J.-L. Bosson, I. Bachelot, and O. Chabre Accuracy of Bilateral Inferior Petrosal or Cavernous Sinuses Sampling in Predicting the Lateralization of Cushing's Disease Pituitary Microadenoma: Influence of Catheter Position and Anatomy of Venous Drainage J. Clin. Endocrinol. Metab., January 1, 2003; 88(1): 196 - 203. [Abstract] [Full Text] [PDF] |
||||
![]() |
E J Evanson Imaging the pituitary gland Imaging, April 1, 2002; 14(2): 93 - 102. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. Lienhardt, A. B. Grossman, J. E. Dacie, J. Evanson, A. Huebner, F. Afshar, P. N. Plowman, G. M. Besser, and M. O. Savage Relative Contributions of Inferior Petrosal Sinus Sampling and Pituitary Imaging in the Investigation of Children and Adolescents with ACTH-Dependent Cushing's Syndrome J. Clin. Endocrinol. Metab., December 1, 2001; 86(12): 5711 - 5714. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Boscaro, L. Barzon, and N. Sonino The Diagnosis of Cushing's Syndrome: Atypical Presentations and Laboratory Shortcomings Arch Intern Med, November 13, 2000; 160(20): 3045 - 3053. [Abstract] [Full Text] [PDF] |
||||
![]() |
K. E. Graham, M. H. Samuels, G. M. Nesbit, D. M. Cook, O. R. ONeill, S. L. Barnwell, and D. L. Loriaux Cavernous Sinus Sampling Is Highly Accurate in Distinguishing Cushing's Disease from the Ectopic Adrenocorticotropin Syndrome and in Predicting Intrapituitary Tumor Location J. Clin. Endocrinol. Metab., May 1, 1999; 84(5): 1602 - 1610. [Abstract] [Full Text] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |