7.1 Introduction

It is likely that most neurosurgeons (even some who have never done awake-surgery) could briefly describe the basic steps of awake brain surgery and summarize what it is trying to accomplish. In short, you are trying to stimulate the brain with an electrode during testing (classically speech and motor) to locate eloquent brain regions and prevent yourself from injuring them, while removing a tumor or other lesion.

What is less clear to people who don’t do this kind of surgery often or at all is how to actually go about doing this. The challenge comes from how to go about finding critical areas, how to know what you are seeing is reality (and how to decide to ignore something), and what to do with the information once you have a map. Pushing an electrode into the brain is not a physically challenging maneuver on par with placing a clip on an aneurysm or removing tumor off the brainstem, but brain mapping is far more than just stimulating the brain with an electrode randomly, but instead is an intellectual challenge of figuring out what is real, and how to incorporate this into a mental map which summarizes what you are actually seeing.

This chapter (and the next) provides practical information about how to get the information and how to determine if it is real. The rest of the book is about what to do with the information.

7.2 What Are Your Goals with Brain Mapping?

Brain mapping is performed to determine if a cut into a cerebral structure (cortical or subcortical) should or should not be made before it happens. Note that the term cut has both global and local meanings that are both valid in this context, as the global cut of a glioma surgery (termed disconnection in Chapter 9) is in fact made up of a series of smaller cuts being made, ideally based on hundreds of decisions based on what should be or should not done next. Thus, performing a disconnection with brain mapping involves as large number of mini-mapping events to decide whether each individual part of the cut is ok or should be moved.

Over my career, I have seen numerous presented examples of different cases that various surgeons have “mapped,” I think it’s worth pointing out what brain mapping isn’t good for. Note it seldom makes surgery worse to test the brain; it’s just that it doesn’t add much. By understanding cases where mapping is largely useless, you have a better understanding of what it actually is useful for.

1. 
   

   Mapping a meningioma invading eloquent brain: I am not saying this is commonly done but I have seen it and it’s worth discussing why this is not helpful. First, meningiomas obviously are not infiltrating tumors, so there is no chance functional brain is running through the tumor. In addition, even if the tumor is invading the motor strip, and you stimulate the surrounding the brain and prove that it is the motor strip, knowing this fact does not mean that tell you that remaining on the tumor border will or will not be a safe method of removing the tumor or not (it may be possible to separate them without damaging the brain and it may not possible.), nor does it provide additional information of how to find the plane. In order words, it’s not to say that it’s not reasonable to leave meningioma tissue behind in the motor cortex if you believe it is invaded and is in the patient’s best interest; it’s just that mapping doesn’t help you accurately make that decision about how to manage the borders. Just because it stimulates doesn’t mean that it is not safe to carefully separate it out.

   
   
2. 
   

   Mapping a non-infiltrating tumor which presents to the cortical surface: Many series on brain mapping contain a number of metastatic tumors (or other non-infiltrating tumors) mixed in with gliomas. This is not to say that mapping is never indicated for these tumors, in many cases it is. However, it is important to recognize that these are non-infiltrating tumors, and the role of mapping is to find a safe cut into the tumor when it is deep to the surface and the best way into the tumor is not immediately obvious. Once you are inside the tumor, good microsurgical technique and staying within the lesion boundaries is usually more than enough to avoid injury to the surrounding brain. Again, it is unclear how the information from stimulation mapping makes that part of the resection safer, the boundary is the boundary and is visible most of the time.

   
   

   When a tumor like a metastasis is on the surface, not only does mapping not help you with the tumor boundary definition, but you do not need it to find a path into the tumor. Thus, any information you get does not necessarily change your strategy, but may inappropriately scare you off prior to doing what needs to get done.

   
   
3. 
   

   Mapping something you cannot sub totally resect: Here arises the thorny issue of arterial-venous malformations (AVMs) in eloquent brain. These lesions can have functioning brain in the center of the nidus, which makes them different than non-infiltrating masses like metastases. However, AVM resections ideally are en bloc removals around the margins, as cutting across the nidus is generally unwise, and there is no evidence supporting a benefit for subtotal resections of AVMs. This fact makes the surgery into an all or none event: you either take it all out, or you probably should consider not operating at all. It’s unclear in most cases how brain mapping would change this much.

Of course, there are exceptions. Mapping can help find a plane into a deep nidus, or allow removal of certain cortices which could help with resection. But in general, brain mapping is a tool for intelligent subtotal resection and AVMs aren’t very good candidates for subtotal resection.

Thus, brain mapping is most obviously helpful for gliomas, which are always infiltrative to some extent, are always subtotally resected to one degree or another, and lack clear boundaries which you need functional techniques to define. The vast majority of my brain-mapping cases are for infiltrating gliomas as mapping guides your decisions in featureless and potentially treacherous brain regions.

7.3 Which Gliomas Should be Done Awake?

Awake glioma surgery has traditionally been undertaken when the tumor is so close to the speech areas that there is significant risk to being asleep. In other words, awake surgery is something best avoided unless forced.

I am of the alternate view that almost all glioma patients should undergo surgery while awake, with the only exceptions being patients who cannot handle being awake, and patients undergoing repeat surgery for a small recurrence where it is very unlikely function will be meaningfully risked.

A rationale to keep essentially all patients awake:

1. 
   

   If you only do awake brain surgery when absolutely forced to (i.e., back of left temporal or left inferior frontal lobe), you will be bad at it and your staff and anesthesiologist will be bad at it. When someone really needs it, you will be less effective at it. When you do awake surgery all the time, everyone is better at it.

   
   
2. 
   

   It is quite easy to cause a lot of havoc in parts of the cerebrum besides the speech and motor areas. If the patient is asleep, you are not monitoring any of this, as you cut into deep brain areas with complex anatomy (if you doubt the complexity of an area like the occipital lobe, glance at Chapters 5 and 6 again).

   
   
3. 
   

   There is almost no such thing as a glioma which can be done better asleep than awake. There is more to life than moving your arm and following commands. Being awake creates the possibility of protecting other functions, or taking a margin around the tumor. Being asleep just saves the patient two hours of work, and risks their higher functions to improve comfort slightly. No rational person would rather have a neurologic deficit in higher brain function or an inferior resection, than work for 1–2 hours awake.

   
   
4. 
   

   No anesthesia=no anesthesia related problems=quicker recovery.

   
   
5. 
   

   No anesthesia=no chance that anesthesia interferes with your mapping.

   
   
6. 
   

   Patients with poor preoperative function (who many people don’t map because they are hard to map) are the patients who need what little function they have the most, and the ones who have their tumor closest to bad brain areas.

In short, I map all possible patients undergoing glioma surgery because I do not see any way in which asleep surgery for gliomas improves the quality of surgery done, and I do not agree with the idea that somehow that awake surgery done by an experienced team with a good field block is somehow so bad for the patients that we should risk neurologic function to try to avoid it.

Finally, here is a brief word about asleep motor mapping. I would argue that the terms “asleep” and “motor mapping” are mutually exclusive terms; meaning that either the patients are asleep or you can map motor function. The techniques termed “asleep motor mapping” are too fraught with issues to be compared to the quality of motor system preservation that one can obtain with awake techniques. For example, the technique of motor mapping using “phase reversal,” where a strip electrode is placed on cortex and studied by an electrophysiologist who determines the location of the motor cortex by EEG phase reversal, may locate the primary motor cortex, but provides little information about where the descending corticospinal tracts are running, and it is unlikely even with the addition of motor evoked potentials that you will get as reliable an early warning as one gets with the double task performed in an awake patient. Further, direct stimulation mapping asleep fails to provide insight into the rest of the motor system, including premotor areas, the SMA, and the cumulative effect of the impact of a planned cut on inputs from the sensory system, the basal ganglia, the cerebellum, etc. As I noted in Chapter 4, a disconnected motor strip connected only to the spinal cord does not drive the anterior horn cells, and a moving but functionally useless hand is only slightly better than a plegic one. In an awake patient, all of these systems can be studied using ground truth, i.e., the patient doing a task, as opposed to a surrogate measure, or partial testing of a system. Finally, asleep motor mapping can be completely thwarted by too much anesthesia, low body temperature, as well as other physiologic problems, which is not a problem in fully awake patients.

In short, I do all patients awake because I almost always can do a more complete, and/or safer surgery awake than is possible asleep.

7.4 Set Up

As with all neurosurgical procedures, positioning is a critical detail. In brain mapping, bad positioning makes the case impossible, as it can prevent the patient from participating in testing. I also teach my residents that uncomfortable patients don’t map well. As will soon become obvious, extensive subcortical mapping requires a cooperative patient for hours sometimes, and an uncooperative patient provides only slightly more information than an asleep one.

I position all patients on their side with the lower arm off the bed slightly and placed on an arm board which is slightly below the level of the bed to prevent the lower shoulder from being kinked (Fig. 7.1). The Mayfield bed attachment hooks under the arm board to fixate the head.

One key point in glioma surgery is that most head positions used in other neurosurgical approaches are basically unwise in awake brain surgery. Extreme head rotation, flexion, or extension, not only is uncomfortable, but it makes it hard to see the testing examiners. Also one of the key things to avoid in glioma surgery is not accounting for head rotation in your assessment of proper trajectory during the disconnection phase of the mapping (see Chapter 9) which can lead to disastrous cuts made based on a flawed understanding of the anatomy. Almost all neurosurgeons (whether or not they are conscious of it) think of the cerebrum from the classic side view, and it is best to put the head as neutral and close to that as possible to eliminate or at least minimize the need to rotate the images in your head during the surgery.

This leaves us with basically four awake head positions: (Fig. 7.2)

1. 
   

   Head neutral: This is the default approach for opercular tumors or tumors in neighboring gyri.

   
   
2. 
   

   Apex tipped upward: By tipping the apex of the head upward with the head otherwise neutral (no rotation or flexion/extension), you place the SFG, MFG, top of motor/sensory strips, and superior parietal lobe into easy view. It is also helpful for accessing the insula from the superior transopecular approach when that is the appropriate angle.

   
   
3. 
   

   Apex tipped downward: Tipping the apex slightly downward in the neutral position is key for lower temporal and occipital tumors, especially when access to the hippocampus, fornix, atrium, and parahippocampal gyrus is desired. A transopercular approach to the insula from the temporal side should also be positioned this way. If the head was positioned apex up, the shoulder would block working at this inferior to superomedial angle, making life miserable.

   
   
4. 
   

   Slight contralateral rotation: This is the position of last resort as it makes mapping harder. However, when the angle of approach comes from the posteromedial parietal lobe or the upper occipital lobe, as with superior parietal lobule approaches to thalamic tumors, or approaches to splenium tumors, a neutral position causes the brain to fall across the field tying up a hand retracting the cut edge of the brain. If this position is needed, you should account for this constantly to remain oriented, and try to keep the rotation to a minimum.

   

   

We have a custom-made square-shaped bar to hold the drape off the patient’s face and we tightly attach the drape to the bar so we can use this as an attachment point for hooks. It is also important that the image guidance attachment to the Mayfield not cross the patient’s line of sight.

Finally, whenever possible, anesthesia should avoid placing lines in the contralateral arm or foot as this can severely hamper arm use and make higher level testing difficult.