The Future of Cranial Neurosurgery—Adapting New Approaches

BBB: blood–brain barrier CSF: cerebrospinal fluid DTI: diffusion tensor imaging FUS: focused ultrasound GBM: glioblastoma multiforme LITT: laser interstitial thermal therapy MRFUS: MR-guided focused ultrasound MRI: magnetic resonance imaging As neurosurgical methods have advanced and effective, safe strategies for the management of a host of intracranial pathologies have evolved, the focus in recent years in many cases has shifted towards achieving these same results with less damage to normal brain tissue, and minimizing disruption of the patient's life. In this review, we aim to summarize some of the most promising technologies which promise to bring us closer to this goal. MAGNETIC RESONANCE FOCUSED ULTRASOUND As cranial neurosurgery transitions to less invasive techniques, the concept of magnetic resonance focused ultrasound (FUS) has emerged. First described by Fry and Putnam over 7 decades ago, FUS had been primarily studied to produce focal destructive lesions in deep cerebral tissue for movement disorders in animal models.1,2 Because the technology was quite bulky and required a craniectomy, the applicability of FUS was limited for cranial neurosurgery.2 Over the last 20 yr, ultrasound engineering has exponentially improved, eliminating the need for a bone defect and improving the accuracy of FUS.3 Much like laser interstitial thermal therapy (LITT), the emergence of magnetic resonance imaging (MRI) has facilitated monitoring of FUS-guided thermal ablation fields.4-6 As a result, modern MRFUS (MR-guided focused ultrasound) techniques have been developed and applied to a wide variety of neurological and nonneurological disorders including uterine fibroids and prostate cancer.7,8 Mechanism MRFUS centers acoustic waves in the desired location to generate heat in a relatively small field. The intensity of the generated energy depends on the duration and continuousness of MRFUS; ie, continuous acoustic waves over a longer duration generate enough energy to permit thermal ablation in the desired location. Once temperatures in the target location reach near 60°C, thermal coagulation and ablation ensues. If the desired goal is to prevent full ablation, the acoustic waves can be generated in pulses to reach a lower target temperature. These pulsed acoustic waves can generate a wide host of other target effects, among them the induction of microbubbles that resonate, producing internal cavitation on a microscopic level and the propagation unidirectional forces that can induce microshearing of target tissues.3 These purported tissue-sparing (nonablative) techniques are believed to facilitate other functions of MRFUS including blood–brain barrier (BBB) disruption, immunomodulation, and improved chemoradiosensitivity.3,9 Applications In the realm of brain tumors, MRFUS has had one of the most generalized applications. For deep inaccessible brain tumors, MRFUS has been employed for thermal ablation through continuous high-intensity ultrasound. Once target temperatures are achieved, the tumor cavity can be sequentially targeted with overlapping acoustic waves to treat the target lesion. FUS was initially trialed in patients with recurrent glioblastoma multiforme (GBM) with limited success. In the first study, which required a pretreatment craniectomy, 1 of the 3 treated patients had evidence of nontarget thermocoagulation resulting in significant morbidity.10 A second study improved on this technology, bypassing the need for a previous craniectomy and supplementing the ablation with MR thermal mapping similar to the method used with modern LITT.11 However, McDannold et al11 were not able to achieve high enough target temperatures for coagulation (mean target temperature was approximately 10° less than necessary). Furthermore, in this small series 1 patient died from a tumoral hemorrhage after treatment. Despite these initial failures, some successes have been reported using MRFUS for thermal ablation of malignant primary brain tumors.12,13 Beyond thermocoagulation, MRFUS is now readily applicable for a variety of uses including BBB disruption, chemoradiosensitization, and immunomodulation. Using nondestructive pulsatile waves and microbubbles, MRFUS is able to alter the tumor environment to facilitate disruption of peritumoral tight junctions. In these preclinical models, liposomal microbubbles are given intravenously; with ultrasonic waves, these bubbles resonate and transiently disrupt the endothelium in the target tissue. As a result, large intravenous therapeutics may be delivered locally with minimal systemic toxicity. This concept has been explored using multiple chemotherapeutics including doxorubicin (glioma), nitrosurea agents (glioma), bevacizumamab (glioma), and trastuzumab (breast metastasis).14-16 These studies have demonstrated variable degrees of drug penetrance and tumoral killing. Currently, clinical trials (NCT02343991) are being conducted to evaluate the safety of MRFUS to transiently break down the BBB for delivery of intracranial doxorubicin. Similar methods are being used to evaluate the efficacy of MRFUS for chemoradiosensitization. Nonablative pulses of MRFUS can provide mild hyperthermia (≈40°C) in the target tissues. Previously, it has been demonstrated that mild hyperthermia can work synergistically to improve the efficacy of radiation therapy. In 1991, FUS was utilized to induce mild hyperthermia in 15 glioma patients prior to daily radiation dosing. Due to lack of adequate thermometry, target temperatures were variable and success was not optimal.17 However, the mechanism of delivering hyperthermia to deep-seated cerebral lesions has recently improved with safer, less invasive transcranial and thermal monitoring techniques. Additionally, MRFUS has a chemosensitizing ability that utilizes shearing and cavitation forces of the ultrasonic waves to induce reactive oxygen species. This type of therapy, termed sonodynamic therapy, utilizes chemotherapeutics to augment the cytotoxicity of ultrasonic therapy.3 Drugs such as doxorubicin, 5-ALA, porphyrins, and curcumin all can be converted into higher energy states after exposure to FUS which can thereby induce free radical formation and eventually the apoptotic cascade.3,18-20 Furthermore, since many of these agents have a higher affinity for hypermetabolic tissues, their tumor specificity may be higher, permitting a higher drug concentration into the FUS target tissue. Functional Neurosurgery Although the field of MRFUS burgeoned in neurooncology, similar concepts using MRFUS can be applied to functional neurosurgery. The first report of MRFUS for functional neurosurgery was intended for the treatment of refractory pain disorders. In these patients with neuropathic pain, MRFUS targeted the contralateral thalamus with nearly 57% of patients attaining some degree of pain relief.21 Similar approaches are now being investigated in preclinical and clinical models for trigeminal neuralgia and psychiatric disorders. MRFUS for Parkinson's disease has also been reported recently in small patient series with moderate benefit. Magara et al22 reported a 60% improvement in tremors after multiple sonications for larger lesional volume of the pallidothalamic tracts. Other small retrospective series also described similar improvements in tremor-dominant Parkinson's disease after lesioning the ventral intermediate nucleus (50%) and globus pallidus (55%), although some of the improvements were only temporary.23,24 Recently, indications for MRFUS have expanded to potentially include mesial temporal lobe epilepsy, neuromodulation, and even Alzheimer's disease. To date, the majority of these functional studies were performed in Vivo, investigating the possible role of MRFUS for a wide breadth of neurological diseases.25,26 Benefits/Limitations MRFUS offers several benefits over traditional neurosurgical therapies. The incision-less noninvasive technique provides patients an added benefit over open surgery and circumvents simple complications such as wound infections and incisional bleeding. The thermal fields generated by MRFUS also avoid the dangers of radiation-based therapies such as wound breakdown, radiation necrosis, or radiation-induced neoplasms. MRFUS can also offer multiple treatments in 1 session with an option for repeated short interval treatments, which cannot be performed with conventional radiotherapy, LITT, or radiosurgery. The combination of nontoxic repeated treatments and real-time thermal monitoring provides a unique safety profile for patients with complex neurological illnesses. Although there is an emerging utility of MRFUS for noninvasive treatment of neurological disease, there are certainly some limitations. As with stereotactic radiosurgery and interstitial thermal therapy, MRFUS may be limited by the size of the lesion. For larger lesions (>1-2 cm.), several treatment sessions may be necessary over a longer period of time to achieve the desired ablation. Some larger lesions may not be amenable to MRFUS due to perilesional/peritumoral edema after treatment, and therefore may require surgical intervention. Additionally, there is an undetermined risk of hemorrhage associated with MRFUS; as mentioned previously, several reports of lesional hemorrhages have been reported after MRFUS ablations. Lastly, despite technical improvements, there is still a small risk of nontarget treatment effects if the MRFUS thermal gradients are not precise. Nevertheless, as our specialty becomes more facile with the use of MRFUS and as the technology becomes more commonplace, its role will likely increase in coming years. Time will tell in which areas of neurosurgery MRFUS can carve a lasting niche for itself, but it is poised to become an invaluable tool in our arsenal. LASER INTERSTITIAL THERMAL THERAPY Background Amongst the armamentarium of neurosurgical treatment modalities holding significant promise for the future is LITT. LITT involves the use of a stereotactically placed catheter to deliver focused thermal energy, thereby achieving local tissue destruction. The genesis of thermal therapy in neurosurgery can be traced back to early studies in the 1970s which suggested that not only could thermal energy be used to produce cellular necrosis, but also that malignant cells may be particularly sensitive to thermal damage.27-29 These studies gave way to the early use of lasers in the laboratory setting, and shortly thereafter, in 1990, to the first use of LITT for an intracranial neoplasms.30-32 Widespread adoption of this technology, however, was limited due to an inability to effectively monitor local temperature changes, and therefore control the extent of ablation. With the advent of effective temperature monitoring technology using MR thermography, which harnesses the temperature dependence of the proton resonance frequency, LITT’s usage has significantly increased in recent years.33-36 Indeed, while LITT had early acceptance in the world of neurooncology for the treatment of tumors, it has increasing utility for a variety of conditions including radiation necrosis, epilepsy, spinal metastases, and preoperative devascularization.33,37-40 Mechanism Central to the understanding of LITT is to establish that a laser is simply a form of nonionizing radiation delivered in a focused beam.35 The efficacy of laser therapy in ablating neural tissue derives from the ability of the laser to achieve rapid photothermal heating via the transfer of photon energy.35,41 By rapidly heating the tissue beyond a threshold level, the laser achieves damage to cellular DNA, ultimately leading to cellular necrosis and gliosis.34,42-46 It is critical, however, that this thermal energy be delivered in a predictable and reliable pattern without causing excessive heat, which could lead to undesirable sequelae. Therefore, stereotactic guidance, temperature feedback, and prevention of overheating are critical components of modern LITT.33,35,46 Temperature feedback is achieved via MRI thermometry which can be overlaid on preoperative MRI planning images, thereby allowing the surgeon to achieve accurate ablation of a desired target volume via real-time imaging (Figure 1).33,35,47 The development of catheter cooling systems has largely addressed the problem of overheating, thereby allowing for larger treatment volumes by preventing tissue overheating and subsequent carbonization and vaporization.46FIGURE 1: A, MR thermography showing progressive heating of target tissues during LITT. B, Postablation MRI showing ablation volume.Risks and Benefits There are a number of distinct benefits of LITT over other, more invasive treatment options. Perhaps most notable is the fact that LITT obviates the need for a large craniotomy. This in turn allows for not only shorter hospital stays, but also for short interval chemotherapy and/or radiation treatment.27,33,48 At our institution, we frequently start adjuvant therapy within a week of LITT. Given its minimally invasive nature, LITT also serves as an attractive alternative for patients who would otherwise be too unhealthy to withstand open cranial surgery.36 Additionally, the small diameter of the laser catheter and focused nature of the laser allows for ablation of otherwise surgically inaccessible lesions.49 LITT also serves as a salvage treatment options for patients who have failed existing therapeutic avenues. One such application is for recurrent dural-based lesions that have failed multiple rounds of surgery and radiation with no other viable treatment options.50 As is the case with any invasive procedure, LITT carries its own set of risks and potential complications. The most commonly described complications include hemorrhage, incorrectly positioned catheters, and unintended thermal injury to eloquent brain tissue.51 A recent paper reviewing the literature noted 4 cases of hemorrhage with 2 secondary to device positioning, 1 secondary to device removal, and 1 resulting from incomplete dural puncture.51 While these complications are of course concerning, the authors of the paper offer several viable means of avoiding such complications in the future, including the use of CTA fusion to MRI to avoid critical vascular structures and needle puncture of the dura to prevent epidural hematoma formation.51 Likewise, unintended thermal damage to surrounding tissue can be mitigated by carefully selecting targets which are in close proximity to cerebrospinal fluid (CSF) spaces which can serve as “heat sinks.”51 Finally, complications arising from inaccurate catheter placement can be minimized with a strong working knowledge of the neuronavigation system being used and its limitations. Applications The use of LITT in treating neurooncologic conditions is becoming increasingly prevalent. The most common applications include gliomas (both surgically inaccessible and recurrent),49 along with focally recurrent radio-resistant metastases52 and radiation necrosis46,53,54 (Figure 2). While large-scale studies demonstrating the relative effectiveness of this treatment have yet to be performed, preliminary analysis suggests similar complication rates to those seen in open surgery for recurrent gliomas.33 In addition to its use in treating malignant lesions, there has been data to suggest LITT as an effective treatment modality for recurrent dural-based lesions.50FIGURE 2: Pre- and postablation contrast-enhanced MRI in a patient with a deep-seated metastasis which failed stereotactic radiosurgery.In parallel, LITT is emerging as a viable treatment option for mesial temporal lobe epilepsy. One notable study included 13 patients with mesial temporal lobe epilepsy treated with LITT and reported 77% of patients achieving meaningful seizure reduction and 54% free of any disabling seizures.55 There was only 1 complication reported in this study, which was a visual field deficit not directly related to laser ablation itself, but rather resulting from an error in stereotactic placement of an alignment rod.55 Use of LITT has also been reported in the treatment of epileptogenic activity arising from focal cortical dysplasia,56 tuberous sclerosis,56 cavernous malformations,57 and periventricular nodular heterotropia.58,59 LITT has also found relevance in the sphere of spine oncology as an alternative to separation surgery.33,60 Early studies have shown promising reduction in the volume of epidural tumors, though the technology at this stage is still quite cumbersome and time consuming.61 TUBULAR RETRACTORS Tubular retractor systems have been popularized in recent years to reduce the amount of subcortical white matter damage classically associated with the approach to any deep-seated lesion. The use of such a system was first documented in 1987, when Kelly et al62 used a metal tube attached to a Leksell frame as a retractor. Most modern systems allow for progressive dilation, thus limiting the transection of white matter tracts. Furthermore, by applying an equal amount of pressure radially on the brain surrounding the tube, these retractor systems are thought to minimize the amount of brain injury compared to traditional flat retractor blades. Though the diameter of the retractor tubes, which is usually between 12 and 28 mm in the commercially available systems, is larger than that of most endoscopes, they allow for bimanual surgical technique and the use of the operating microscope. Two systems are currently commercially available in the United States, BrainPath (Nico Corporation, Indianapolis, Indiana; Figure 3), and ViewSite Brain Access System (Vycor Medical Inc, Boca Raton, Florida). Both consist of an inner obturator that helps bluntly separate white matter tracts, and a clear outer cannula that can be anchored to a fixed retractor system. At our institution, a trajectory is planned preoperatively with the aim of minimizing transgression of eloquent brain using anatomic landmarks and if necessary diffusion tensor imaging (DTI). With the help of neuronavigation, an approximately 3 cm craniotomy is then turned centered on the entry point of the trajectory. Whenever possible, a sulcus is dissected to allow for transsulcal of the In cases the this the tube is placed through a small is then using technique (Figure If the retractor can be to bring of larger into the field of of the is limited by the tube, we have recently been an at the of the to the cavity and that no tumor in the the tube is in a along the (Figure of the BrainPath system (Nico one of the commercially available during of a using a Pre- and MRI of a with a deep was performed using a and the patient was in neurological on series the safety and efficacy of retractor systems in the management of a variety of and pathologies have been described the use of a 12 mm retractor for the of This method many of the complications of the approach without limiting the surgeon the way et a series of 20 patients using the BrainPath (Nico retractor for and lesions of The retractor was through a transsulcal approach in all and was employed was achieved in and 3 patients which by the of the To the amount of tissue damage by et the of 20 patients who of deep-seated lesions with the help of Though there was only a minimal increase in there was a significant increase in areas of diffusion that a degree of injury still studies will on the amount of tissue damage by and on may help this that promise to provide the results of surgical techniques with patient risk for neurological and shorter hospital are on the from techniques such as MRFUS to methods at the damage of open as seen with these though in are still in their and in many cases the is still on their method of use and we are that with their these technologies will our to deliver neurosurgical with patient and in the is a for The other authors have no or in any of the or described in this