Successful management of diffuse low-grade and high-grade gliomas in adults is a challenge for neuro-oncologists. Indeed, due to their highly infiltrative feature, these diseases remain incurable despite therapeutic advances. Nevertheless, the elaboration of individualized therapeutic strategies has led to an improvement of both overall survival and quality of life. In particular, the impact of surgical resection on diffuse glioma survival has been extensively demonstrated. However, this impact is significant only when the resection is total (i.e., complete removal of the T2-hyperintensity in diffuse low-grade gliomas, or complete removal of the enhancement in high-grade gliomas), or at least subtotal. Interestingly, biopsy samples within and beyond the abnormalities, defined by magnetic resonance imaging, have shown that the actual spatial extent of gliomas was underestimated by this conventional imaging modality, since glioma cells were present outside the signal abnormalities. Thus, it was suggested that the removal of a margin around the tumor visible on magnetic resonance imaging, i.e. “supratotal resection”, might improve the outcomes in diffuse gliomas. To achieve this type of supramaximal resection, while preserving the quality of life, a new concept is to switch from an image-guided surgery to a functional-guided surgery, i.e. to pursue the resection up to the eloquent neural networks using intraoperative direct electrical stimulation mapping in awake patients. The aim of this article was to review the recent data about supratotal resection, including both oncological and functional results. Favorable outcomes have recently opened the door to the principle of “preventive surgery” in incidentally discovered gliomas, and to the proposal of a medical screening.