Lonser and colleagues 4 performed a prospective observational analysis to determine the incidence, location, and tumor characteristics of dural invasion associated with corticotropic pituitary adenomas. Eighty-seven consecutive pediatric and adult patients with Cushing disease at a single institution underwent a procedure in which the anterior dura was routinely resected and the medial cavernous sinus wall was resected when dural invasion was encountered. The uncoagulated dural specimens were processed for adrenocorticotropic hormone immunohistochemistry and were compared with highresolution MR imaging determinations. Tumor resection was accomplished by circumferential dissection of the pseudocapsule in most cases. Consistent with prior observations, 1,2,3,6,8 histological evidence of dural invasion was common, occurring in 34% of patients. Of note, 7% of patients had dural invasion of the anterior sella without apparent gross surgical invasion. Dural invasion occurred most frequently at the cavernous sinus, but also along the anterior, inferior, and posterior dura of the sella. Magnetic resonance imaging was a poor predictor of dural invasion, correctly identifying cavernous sinus invasion in only 22% of cases. Tumor size was related to invasion, yet many small tumors adjacent to the dura were invasive. In contrast to prior studies, 6 there was no apparent increase in dural invasion with age.There are several important observations in this study. First, the authors have confirmed the high frequency of dural invasion in a prospective consecutive series of patients with Cushing disease. Proximity of the tumor to dura seems to be the most important factor for invasion, and MR imaging is a poor diagnostic test to identify invasion. Even direct observation of dural invasion was inaccurate in a small number of cases. Second, dural invasion can occur at various sites within the sella, not only the cavernous sinus, as might be expected. Third, the authors' approach of extracapsular tumor dissection and resection of adjacent dura, combined with reexploration when immediate postoperative serum cortisol levels remain normal or elevated, resulted in an immediate biochemical remission rate of 98% and no early recurrences. Although this series is not controlled, the immediate remission rate is higher than that achieved in most larger studies by experienced pituitary surgeons.