Cavernous sinus sampling in patients with Cushing's disease
- Others, not related to the research strengths mentioned above
OBJECT Correct diagnosis and precise localization of adenomas in patients with Cushing's disease are essential for avoiding unsuccessful transsphenoidal pituitary exploration. In addition to the well-established inferior petrosal sinus sampling, preoperative cavernous sinus sampling (CSS) was introduced as a potentially improved way to predict adenoma lateralization. The authors present their results with CSS in a consecutive series of patients with Cushing's disease. METHODS During 1999-2014, transsphenoidal surgeries were consecutively performed in 510 patients with Cushing's disease. For most patients, suppression of cortisol in high-dose dexamethasone tests and stimulation of adrenocorticotropic hormone and cortisol after administration of corticotropin-releasing hormone were sufficient to prove the diagnosis of adrenocorticotropic hormone-dependent hypercortisolism. Of the 510 patients, 67 (13%) were referred to the department of neuroradiology for CSS according to the technique of Teramoto. The indications for CSS were unclear endocrine test results or negative MRI results. Data for all patients were retrospectively analyzed. RESULTS A central/peripheral gradient was found in 59 patients; lateralization to the left or right side was found in 51. For 8 patients with a central/peripheral gradient, no left/right gradient could be determined. For another 8 patients with equivocal test results, no central/peripheral gradient was found. No severe CSS-associated complications were encountered. Of the 51 patients who underwent transsphenoidal surgery, the predicted lateralization was proven correct for 42 (82%). CONCLUSIONS As MRI techniques have improved, the number of potential candidates for this invasive method has decreased in the past decade. However, because detecting minute adenomas remains problematic, CSS remains a useful diagnostic tool for patients with Cushing's disease.
- Forschungsinformationssystem des UKE