14 Surgery Around the Command and Control Axis: The Default Mode, Control, and Frontal Aslant Systems
Abstract
The initiation axis is a novel concept in neurosurgical anatomy which is based on the idea that some part of the brain needs to give the “start” signal for goal-directed action to occur. This chapter describes this anatomy and discusses methods for avoiding injury to the initiation axis during brain surgery.
14.1 Introduction
For a long time, the frontal lobe has been a synonym for “safe” in neurosurgery. In most surgical management schemes, the “eloquent areas” are principally the motor strip, the language areas, and perhaps the supplementary motor area (SMA). The term “eloquent” is a principal term in the neurosurgical community that refers to areas of the brain which when transgressed cause a visible observable deficit, usually one that can easily be described by a nonexpert, for example, the inability to talk or move one’s arm. In this paradigm, our only mandate is to avoid destroying these areas, and we theoretically have carte blanche to do whatever with the rest of the largest lobe in the human brain.
Of late, this paradigm is being challenged by a growing voice, as it is unlikely that the lobe evolved to absorb cerebrospinal fluid, and all parts of the brain are doing something. Certainly, we know this from neuropsychology, as frontal lobe syndromes are reproducible, roughly localizable to gross areas of the frontal lobe, and debilitating, and the functional imaging community has been arguing this for some time. In this paradigm, we should respect all parts of the brain as if they were the motor cortex.
I will argue that uncritically applying either strategy is ridiculous and deprives patients of an ideal approach. While neither deficit is ideal, reducing an adult patient to a child is hardly better than paralyzing the hand, and in many ways, it leaves a less functional patient: some people with motor deficits can hold jobs as compared to patients with severely impaired judgment. At the same time, conventional wisdom is not entirely baseless, and the frontal lobe clearly has more redundancy, bilaterality, and plasticity than the primary motor cortex in many people, and approaching malignant cancers in the frontal lobe overly cautiously to “save function” frequently ends in saving neither the function nor the patient. A key problem in balancing our goals is very limited knowledge we have about how the frontal lobe works and how to do rational surgery there.
It is beyond any question that the medial frontal lobe can punish an unwary neurosurgeon, often in ways which are hard to nail down using 5-point scales, but which damage the patient’s quality of life in profound ways. Again, a patient who can talk, but rarely speaks or initiates any activities is almost certainly less functional than the one who struggles to find words. A patient with a paralyzed arm is easier to rehabilitate than one who merely stares blankly into space.
This chapter addresses the command and control axis of the frontal lobe which summarizes the parts of the lobe that initiate behaviors and transmit intentions to areas that can act on them. This axis comprises several interconnected large-scale functional networks that have been found to play key roles in these processes and, which I am increasingly convinced, underlie many of the clinical syndromes we see with surgery or around in the medial frontal lobe; these syndromes can often be overlooked if you limit your assessments to “moving all four extremities,” but families do notice these as such syndromes have often more profound implications. Knowing the exact anatomy of these areas can and should alter your surgical plan, as it is probable that many syndromes which we have long accepted as inevitable consequences of certain operations can be avoided in many people with better knowledge of anatomy, especially connectomics.
14.2 The Command and Control Axis Defined
We take it as a given that there is a system of places in the brain that makes the decisions, such as one is thirsty, and translates them into a series of actions starting from reaching for a glass of water and putting it to the mouth to drink it. While hypothalamic, motor planning, visual, somatosensory, attention, and executive systems all are part of this series of events, it is self-evident that some part of the brain needs to say “Go,” or in other words, needs to tell to stop one task and start another. Given that a large percentage of problems with initiation and motivation occur, in clinical neurology, due to medial frontal damage, it also seems self-evident that at least some of the key machinery for initiation and motivation is located in the medial frontal lobe. Given that much of the experience with akinetic and abulic problems have come from stroke, trauma, and surgical manipulation, such as with the anterior interhemispheric approaches which are for imprecise lesions, we have long had a poor understanding of this problem and how to avoid it.
This is not entirely understood by anyone at present, but a large body of evidence supports the idea that a coordinated interplay of three large-scale cerebral networks—the default mode network (DMN), the central executive or control network (CTRL), and the salience network—is strongly linked with the process of transitioning from internal to external mental states, and several related behaviors such as goal-directed behavior versus mind wandering. The proximity of these networks to the SMA and specific clinical observations related to the nature of syndromes from these areas raise the possibility that this is an axis needed to get actions going, and that the consequences for disrupting this axis are some form of failure to start an action or the lack of motivation to do so. In some ways, this is an axis similar to the visual pathway where lesions in different places lead to different variants on the same basic issue, though obviously initiation is a subtler problem which is harder to definitively identify than a field cut. The subsequent text provides some summary about what is known about these networks, our work defining the anatomy of these networks as precisely as possible which will provide some ideas of how they exactly interconnect with each other, and eventually insights into how to make good decisions regarding appropriate surgery for gliomas in these parts of the brain. We will start with the key large-scale functional networks and then segue into a discussion about the frontal aslant tract (FAT) which contains the SMA and salience network interconnections.
14.3 The Large-Scale Functional Networks
The DMN was first noted by Marcus Raichle and colleagues in 2001, when they noted areas in the anteromedial frontal lobe, the posterior cingulate cortex (PCC), and the lateral parietal lobe that were activated only in the task-negative state. Since then, thousands of reports confirm these areas to activate on correlated time courses (thus forming a functional network), and to anticorrelate with several other networks involved in performing externally directed goals, notably the CTRL network. The DMN is probably the most consistently identifiable network in the brain that is involved in numerous complex cognitive abilities, such as speech, theory of mind, and memory, among others.
Fig. 14‑1 provides a simple schematic of DMN–CTRL interactions. The salience network is a third network which appears to be the key in this transition. Failure to alternate these networks has been found in minimally conscious and vegetative patients, and it has been shown to be impaired in schizophrenics with severe negative symptoms. Thus, it seems reasonable to hypothesize that disruption of this system might impair patients’ ability to organize their thoughts, create a plan, and transition toward executing the plan, though obviously this is a complex process.
14.3.1 Network Anatomy
The Human Connectome Project (HCP) recently published their scheme for parcellating the human neocortex based on functional connectivity and physical characteristics. Fig. 14‑2 provides an example of the scheme which we have heavily utilized to describe brain connectivity in ways which can be compared, reproduced, and utilized by surgeons. Next we describe the anatomy of the large-scale functional networks in HCP parcellation format, based on coordinate based meta-analysis, combined with diffusion spectrum tractography to provide the best possible anatomic model for these networks, given existing technology. There is extreme interindividual variability in the human cortex, and gliomas can cause functional reorganization, so these models are merely the starting point in the discussion, but a key starting point compared to a previous lack of knowledge of these areas.