How to overcome the limitations to determine the resection margin of pituitary tumours with low-field intra-operative MRI during trans-sphenoidal surgery: usefulness of Gadolinium-soaked cotton pledgets

Abstract

Intra-operative MRI (iMRI) is used as an immediate intra-operative quality control, allowing surgeons to extend resections in situations involving residual tumour remnants. Despite these advantages, low-field iMRI has some limitations with regards to image quality and artefacts. The aim of this study is to report our experience with bone wax and Gadolinium-soaked cotton pledgets in obtaining precise tumour resection margins using low-field iMRI. The study group included 63 consecutive patients who underwent endonasal trans-sphenoidal surgery with use of intra-operative low-field iMRI (0.15 T, PoleStar N20, Medtronic Navigation, Louisville, CO, U.S.A.). The indications for intra-operative MRI use included a suprasellar or retrosellar extension (n = 23), cavernous sinus invasion (n = 21), a tumour located in the vicinity of critical anatomic structures (such as the internal carotid artery, n = 10), recurrent or revision procedures (n = 5), and pre-operative imaging revealing unusual anatomy (n = 4). Overall, among the 51 patients with intended complete tumour removal, iMRI revealed definite tumour remnants or suspicious findings in 13 patients (25.5%), leading to an extended resection and allowing completion of the resection in 10 patients. There was an increased rate of complete tumour removal from 74.5% (38 out of 51) to 94.1% (48 out of 51). The iMRI scan for complete tumour removal was more efficient in the group receiving Gadolinium-soaked cotton pledgets (85.2-100%) than in the group receiving bone wax or the conventional method (62.5-87.5%). The results of iMRI and the estimation by the surgeon concerning the extent of resection revealed a discrepancy in five patients (15.6%) in the Gadolinium-soaked cotton pledgets application group, and in 14 (45.2%) of the bone wax application group. More valuable information for determining the resection margin can be obtained with the use of contrast-soaked cottonoid packing in the tumour resection cavity during iMRI scanning. We believe that the use of this simple method reduces the false-positive results and also overcomes the disadvantages of low-field iMRI.