The value of preoperative brain tumor embolization is difficult to demonstrate for those who lack the benefit of having seen, in training or practice of others, what an efficient embolization is capable of. Many believe that it is ultimately unnecessary, which may be the case for many tumors. Therefore, when time comes to attack some kind of skull base monster, there is a certain lack of appreciation for how a preoperative angiogram and embolization can add game-changing value. For this to work, both surgical and embolization efforts must be coordinated towards a shared goal. The ultimate purpose is not just to control blood flow, but to minimize the overall morbidity and maximize efficiency of treatment. Thus, an overly conservative or inefficient embolization is a low risk, but ultimately unhelpful procedure. A thoughtfully aggressive embolization may be more hazardous, but pays serous dividends later. Thus, the embolization and resection should be viewed by both the patient and physicians as a package.
Much of the work on this subject was done in the 1980s and 1990s, in the pre-GDC era of more limited neurointerventional scope. Unfortunately, much of the technical sophistication achieved at that time has not been imparted to the ever-expanding group of practitioners, who have no exposure to complex preoperative tumor embolization during their training. The reason for this is lack of neurosurgical expertise in performance of complex, usually skull-base resections, at most places where training of neurointerventionalists takes place. The purpose of this section is to review some of these techniques and to keep up an interest in this subject among those who may benefit from either requesting or performing a preoperative embolization.
The majority of brain preoperative embolization targets are meningiomas, since they are both hypervascular and principally supplied by dural feeders. While the majority of convexity lesions do not require an embolization, even if highly vascular, those of the skull base or posterior fossa can be quite different. A very effective embolization, as evidenced for example by lack of contrast uptake on post-embo MRI, or diffusion restriction, can result in significant meningioma shrinkage — sometimes better than radiation — which can be very helpful in for example lesions that invade but not occlude a major venous sinus.
The goal of embolization is to reduce perioperative bleeding, both to limit blood loss and to (more importantly) allow for a more controlled resection. With exception of single supply pedicle lesions (simple tumors), the efficiency of embolization is directly related to deposition of embolic material as deeply as possible into the intrinsic tumoral vasculature. For this reason, best tumor embolizations are require small particles, which can penetrate 100 micrometer or smaller diameter vessels. Embolizations with larger particles or other methods do not penetrate as deeply into the tumor and therefore do not work as well, in most circumstances. Thus, much of tumor embolization consists of painstaking sequential catheterization of feeding arteries and deposition of particles into the tumor matrix, all the while observing for dangerous anastomoses and thinking of nontarget tissue embolization, such as cranial nerves and skin. Larger particles are less likely to cause nontarget tissue necrosis or dysfunction (cranial nerves), but are also less efficient at embolization. In my experience, cranial nerve damage with particles is relatively rare, compared with embolization of the same vessels with Onyx or even nBCA. Others may disagree with any or all of this — the purpose here is to show our methods and give reasons for why we do it this way.
Here are some cases (more on the way). Remember that preoperative tumor embolizations can be very long, and therefore the following examples are detailed. Allow sufficient time for review.
For the same reasons, it is essential to be cognizant of radiation dose when performing preoperative embolizations. The usual ALARA techniques apply — collimation, minimum patient to detector distance, etc. DSA should be performed at maximum 2 frames per second, and 1 frame per second once essential diagnostic information has been obtained. Fluoro, including live subtraction during particle deposition, can be safely done at 7.5 frames per second or, in safer territories, at 4 frames per second. Both PA and lateral cameras can often be moved during the procedure to perform embolization from different angles and therefore reduce focal skin entry doses.