Initial evaluation of ACTH levels in the jugular veins before CRH was available showed that this test was not sufficiently sensitive to diagnose Cushing disease [6, 7]. In Finding and colleagues' series of six patients with Cushing disease [7], five patients had diagnostic ACTH ratios in the petrosal sinus samples but only one had a diagnostic ratio in the jugular vein samples. At that time, the variations in drainage patterns of the inferior petrosal sinus had not been described [4], and samples were obtained from the jugular bulb. We postulated that the sensitivity of jugular vein sampling would be enhanced by use of a lower sampling site to ensure that all petrosal sinus drainage was included and by administration of CRH.

Methods

Top Methods Results Discussion Author & Article Info References

We prospectively studied 20 patients with surgically proven Cushing disease (12 women and 8 men; mean age, 40 years [range, 16 to 62 years]) and 1 male patient with proven ectopic ACTH syndrome (56 years of age). The study protocol was approved by the investigational review board of the National Institute of Child Health and Development. A diagnosis of ACTH-dependent Cushing syndrome was established by conventional laboratory studies. Gadolinium-enhanced magnetic resonance imaging of the pituitary gland was nondiagnostic in all patients. No patient had cyclic Cushing syndrome.

Patients underwent petrosal sinus sampling and jugular vein sampling before and after CRH administration in two catheterization sessions conducted at least 48 hours apart (average interval, 5 days). Petrosal sinus sampling was performed as described elsewhere [3]. For jugular vein sampling, catheters were placed in both internal jugular veins through bilateral femoral vein punctures. The catheters were positioned at the level of the angles of the mandible. This level was chosen to accommodate the inferior petrosal sinuses, which occasionally enter the jugular vein at a variable distance below the jugular bulb, and to encourage mixing of the effluent from the inferior petrosal sinuses with jugular venous flow. Samples were obtained bilaterally before and 3, 5, and 10 minutes after intravenous administration of CRH, 1 µg/kg of body weight.

For all patients, we calculated the ratio of central ACTH levels to peripheral ACTH levels in petrosal sinus samples and jugular vein samples before and after CRH was administered. A ratio of central to peripheral ACTH greater than 2 before CRH administration and greater than 3 after CRH administration in petrosal sinus samples or jugular vein samples was considered evidence of Cushing disease on the basis of cut points previously established to judge the results of petrosal sinus sampling [1]. In jugular vein samples, we disregarded ratios greater than 2 before CRH if they were not accompanied by ratios greater than 3 after CRH. Absence of diagnostic ratios was considered evidence of the ectopic ACTH syndrome.

All 20 patients with Cushing disease underwent transsphenoidal exploration by the same neurosurgeon. Nineteen patients had resection of an ACTH-secreting microadenoma, shown by positive results on immunofluorescent staining of the adenoma for ACTH and serum cortisol levels less than 2 µg/dL immediately after surgery. All 19 patients are currently in remission. One patient underwent hemihypophysectomy on the basis of results of lateralization of petrosal sinus samples when no adenoma was detected; this patient is in remission. The patient with the ectopic ACTH syndrome is in remission after removal of a neuroendocrine tumor in the left carotid sheath.

The McNemar test was used for statistical comparison of the sensitivities of petrosal sinus sampling and jugular vein sampling. P values less than 0.05 were considered significant.

Results

Top Methods Results Discussion Author & Article Info References

The Table 1 shows the results of petrosal sinus sampling and jugular vein sampling in the 20 patients with Cushing disease. Results of sampling of the inferior petrosal sinuses were positive for Cushing disease in 19 patients (sensitivity, 95% [95% CI, 75% to 99%]). One patient had a negative result on petrosal sinus sampling, a positive result on jugular sampling, and an ACTH-secreting pituitary adenoma proven surgically. Both of this patient's inferior petrosal sinuses were plexiform; this prevented normal positioning of the petrosal sinus catheters and foiled our efforts to place microcatheters (Target Therapeutics, Fremont, California) in the cavernous sinuses. Jugular vein sampling had positive results because a vein drained into the right jugular vein just above the angle of the mandible.

Jugular vein sampling had positive results in 16 of 20 patients with surgically proven Cushing disease (sensitivity, 80% [CI, 56% to 96%]). Unlike the results of petrosal sinus sampling, which were positive in all patients before and after CRH was administered, 10 of 16 (63%) jugular vein samples were positive only after CRH was administered (Table 1). In two of the four patients with negative or nondiagnostic jugular vein samples, ratios of central to peripheral ACTH were greater than 2 before CRH but less than 3 after CRH. Because our diagnostic criteria required a ratio greater than 3 after CRH administration, these two samples were considered nondiagnostic. In two patients with surgically proven Cushing disease, jugular vein samples had nondiagnostic ratios of central to peripheral ACTH before and after CRH was administered. Both of these patients had positive results on petrosal sinus sampling. Statistical analysis showed no significant difference between the sensitivity of petrosal sinus sampling and that of jugular vein sampling (P > 0.2).

Ratios of central to peripheral ACTH were higher in the petrosal sinus samples than in the jugular vein samples (Figure 1). The average ratio before administration of CRH was 17.7 (range, 2.2 to 110.5) in the petrosal sinus samples and 2.7 (range, 2.1 to 3.7) in the jugular vein samples. The average ratio after administration of CRH was 90 (range, 12.7 to 407.1) in petrosal sinus samples and 7.66 (range, 3.24 to 14.9) in the jugular vein samples. The peak ratio after administration of CRH occurred in the 3-minute petrosal sinus samples (11 of 19 samples) and were equally divided between the 3- and 5-minute jugular vein samples.

View larger version (17K): [in this window] [in a new window]   Figure 1. Ratios of central adrenocorticotropic hormone (ACTH) to peripheral ACTH in inferior petrosal sinus (IPS) samples and internal jugular vein (IJV) samples before and after administration of corticotropin-releasing hormone (CRH). Dashed lines indicate the lowest ratio considered diagnostic for Cushing disease. Open circles represent patients with Cushing disease; closed circles represent the single patient with the ectopic ACTH syndrome.

In the patient with the ectopic ACTH syndrome, ratios of central to peripheral ACTH were less than 2 before and less than 3 after administration of CRH in the petrosal sinus samples and jugular vein samples (Figure 1).

Discussion

Top Methods Results Discussion Author & Article Info References

In our small series, jugular vein sampling was not as accurate as petrosal sinus sampling for distinguishing pituitary ACTH production from ectopic ACTH production. However, jugular vein sampling is a much simpler procedure that can be performed easily and safely in institutions that lack experience in the more technically demanding and hazardous procedure of inferior petrosal sinus sampling. Of 20 patients with surgically proven Cushing disease, jugular vein samples established the correct diagnosis in 16 (sensitivity, 80% [CI, 75% to 99%]) and produced false-negative results in 4. The patient with proven ectopic ACTH syndrome had negative results on jugular vein sampling and petrosal sinus sampling.

During the study, we performed petrosal sinus sampling and jugular vein sampling on three additional patients presumed to have ectopic ACTH syndrome. All three patients lacked diagnostic ratios of central to peripheral ACTH on petrosal sinus sampling and jugular vein sampling, did not have suppression with high-dose dexamethasone (8 mg twice daily), and did not respond to CRH. Each patient had a normal magnetic resonance imaging scan of the pituitary gland. A source of ectopic ACTH production was not detected by computed tomography or magnetic resonance imaging of the chest and abdomen or by somatostatin-receptor scintigraphy. The anatomic source of the ectopic ACTH syndrome has not been located in these three patients, and the diagnosis is considered clinically established but not anatomically proven. Thus, we excluded these three patients from the analysis.

One patient with an ACTH-secreting microadenoma of the pituitary gland proven surgically lacked diagnostic ACTH ratios on petrosal sinus sampling before and after CRH was administered. This occurrence seems to be uncommon: It was encountered only once in more than 500 patients with proven Cushing disease [8]. In both cases, the inferior petrosal sinus on the side of the adenoma was plexiform. Although the inferior petrosal sinus could be accessed from the jugular vein, it drained preferentially into another vessel. In our overall experience [1, 2], petrosal sinus sampling has a 99% sensitivity as a differential diagnostic test.

When venous sampling for Cushing syndrome was originally described, the disappointing results of jugular vein sampling compared with those of petrosal sinus sampling were probably caused by sampling of the jugular bulb, which often lies above the level of inflow from the inferior petrosal sinuses. In our series, the jugular vein catheter was positioned at the angle of the mandible (that is, below the level of inflow from the petrosal sinuses). In addition, CRH was not available at the time of the early studies. In our series of jugular vein sampling, administration of CRH was essential for positive results in 10 of 16 samples.

A major advantage of sampling the internal jugular vein is that it is simpler and safer than sampling the inferior petrosal sinuses. Although our small series did not show a statistically significant difference between the two methods, larger series suggest a higher sensitivity for petrosal sinus sampling (99%) [1, 2]. On the basis of clinical experience, we recommend that patients whose jugular vein samples do not have diagnostic ratios of central to peripheral ACTH undergo petrosal sinus sampling. We also recommend that petrosal sinus sampling be performed in patients who have diagnostic ratios on jugular vein sampling only before administration of CRH. We encountered two such patients; their petrosal sinus samples showed a brisk response to CRH. Failure to detect a response to CRH in the jugular vein samples was probably due to streaming of the petrosal sinus effluent in the jugular veins. Positioning the sampling catheters along the medial wall of the jugular veins may improve the sensitivity of jugular vein sampling. In addition, performing a Valsalva maneuver during sampling may facilitate mixing in the jugular vein and thereby improve sensitivity. Both modifications are the subject of ongoing investigations.

Other investigators [9, 10] have recommended sampling the cavernous sinuses with Tracker microcatheters (Target Therapeutics) as an alternative to petrosal sinus sampling, especially when CRH is not available. We did not find this technique helpful in a series of 15 patients [11], 3 of whom had elevated ACTH levels only after CRH was administered. We find it hard to justify this more invasive and expensive procedure. The attraction of jugular vein sampling is its simplicity and safety. Now that CRH (Ferring Laboratory, Suffern, New York) is approved for diagnostic use by the U.S. Food and Drug Administration, jugular vein sampling done after administration of CRH may become one of the least expensive single tests with which to distinguish pituitary ACTH production from ectopic ACTH production. We envision that this test would be used to confirm the diagnosis of Cushing disease in patients with a positive response to dexamethasone suppression [12], CRH administration [13], or metyrapone stimulation [14].

From Warren Grant Magnuson Clinical Center, National Institute of Neurological Disorders and Stroke, and National Institute of Child Health and Human Development, National Institutes of Health, Bethesda, Maryland.

Dr. Oldfield: Surgical Neurology Branch, National Institute of Neurological Disorders and Stroke, National Institutes of Health, Bethesda, MD 20892.

Dr. Nieman: Developmental Endocrinology Branch, National Institute of Child Health and Human Development, National Institutes of Health, Bethesda, MD 20892.