Cavernous and inferior petrosal sinus sampling and dynamic magnetic resonance imaging in the preoperative evaluation of Cushing's disease.

ABSTRACT

The surgical management of Cushing's disease is often complicated by difficulties detecting corticotropic adenomas. Various diagnostic modalities are used when conventional magnetic resonance imaging (MRI) is negative or inconclusive. We sought to analyze our use of two such modalities in the surgical management of Cushing's disease (1) cavernous/inferior petrosal sinus sampling (central venous sampling, CVS) for adrenocorticotropic hormone and (2) dynamic MRI (dMRI). We conducted a single-center, retrospective review of all patients with Cushing's disease treated by a single neurosurgeon with endonasal transsphenoidal surgery. Accuracy of adenoma localization with CVS and dMRI was analyzed. Ninety-one consecutive patients were included. Pathology confirmed an adenoma in 66. Preoperative dMRI and CVS were performed in 40 and 37 patients, respectively, with 20 undergoing both studies. Surgical pathology was positive for adenoma in 31 dMRI patients, 25 CVS patients, and 13 who underwent both. Among patients with pathology confirming an adenoma, dMRI identified a lesion in 96.8% and correctly lateralized the lesion in 89.7%, while CVS correctly lateralized in 52.2-65.2% (depending on location of sampling). Among patients with both studies, dMRI and CVS correctly lateralized in 76.9 and 61.5-69.2%, respectively. Accuracy of CVS improved if only patients with symmetric venous drainage were considered. In this mixed population of Cushing's disease patients, dMRI was more accurate than CVS at localizing adenomas, supporting the use of advance MRI techniques in the work-up of Cushing's disease. CVS, however, remains an important tool in the workup of Cushing's syndrome.