One of the most important potential limitations of presurgical mapping using blood oxygen level dependent functional magnetic resonance imaging (BOLD fMRI) is the phenomenon of neurovascular uncoupling (NVU).¹ The phenomenon of NVU refers to the breakdown of the normal neurovascular coupling cascade that ultimately results in generation of the BOLD response to a neural stimulus.² NVU can lead to erroneous interpretation of clinical fMRI examinations and can be an important confound in research studies relating to structural brain lesions such as brain neoplasms, arteriovenous malformations and other vascular malformations, cortical dysplasias and other epileptogenic lesions.^3,4 The effects of brain tumor-related NVU on task-based BOLD fMRI have been previously published.⁵ Several articles have recently been published suggesting that there may be potential advantages of using resting state BOLD fMRI (rsfMRI) as a pre-operative mapping tool.⁶ The purpose of this study is to demonstrate that the problem of brain tumor-related NVU is a significant issue with respect to resting state BOLD fMRI similar to task-based BOLD fMRI, in which signal detectability can be compromised by breakdown of normal neurovascular coupling.We evaluated seven de novo brain tumor patients who underwent resting state fMRI as part of comprehensive clinical fMRI exams at 3T. Scanning was performed on a 3.0 T Siemens Trio MRI with a 12-channel head matrix coil. Imaging protocol included 3D T1 MPRAGE (TR=2300 ms, TI= 900 ms, TE= 3.5 ms, 9° FA, 24-cm FOV, 256x 256x176 matrix, slice thickness 1 mm) as well as 2D T2 FLAIR (TR=9000 ms, TI=2500 ms, TE=116 ms, 141° FA, 17.2cm x 23cm FOV, 240x320x53 matrix, slice thickness 3 mm with 3 mm gap between slices) for structural imaging. For rsfMRI, 180 volumes were acquired using a standard 2D GE-EPI T2* weighted BOLD sequence (TR=2000 ms, TE=30 ms, 90° FA, 24-cm FOV, 64x64x33 matrix, 4 mm slice thickness with 1 mm gap between slices, interleaved acquisition). Each patient was instructed to remain still with eyes closed without falling asleep during the scanning period of 6 minutes. Each patient demonstrated evidence of potential NVU based on results of the clinical task-based fMRI scans. Such potential NVU was determined by demonstration of abnormally decreased or absent task based activation in expected eloquent cortex corresponding to regional decreased breath hold (BH) cerebrovascular reactivity (CVR) in the absence of corresponding neurological deficits as described in several previous publications.⁵ For each of the seven patients who demonstrated evidence of NVU on task-based motor fMRI, we performed both an independent component analysis (ICA) and an atlas-based parcellation-based seed correlation analysis (SCA) of the resting state fMRI data. For each patient, ipsilesional (IL) and contralesional (CL) regions of interest (ROIs) comprising primary motor and somatosensory cortices (see Figure 1) were used to evaluate BOLD signal changes on Z score maps derived from both ICA and SCA analysis for evidence of NVU. A subsequent two-tailed t-test was performed to determine whether statistically significant differences between the two sides were present that were consistent with NVU. In seven patients, overall decreased BOLD signal (based on suprathreshold voxels in ICA and SCA-derived Z-score maps) was noted in IL compared to CL ROIs (p<0.01), consistent with NVU. All of the results are presented in Table 1. An example of analysis for an individual patient is provided in Figure 2.Our study demonstrates evidence of potential neurovascular uncoupling on both SCA-derived Z-score maps for sensorimotor seed region and ICA-derived sensorimotor component maps in all seven patients with perirolandic primary glial neoplasms.We have demonstrated that NVU can result in unexpectedly decreased or absent activation BOLD signal changes on rsfMRI comparable to previously published findings on standard motor task-based fMRI.