Transcranial focused ultrasound (FUS) ablation is an emerging incision-less treatment for neurological disorders. The factors affecting FUS treatment efficiency are not well understood. Kranion is open-source software that allows the user to simulate the planning stages of FUS treatment and to “replay” previous treatments for off-line analysis. This study aimed to investigate the relationship between skull parameters and treatment efficiency and to create a metric to estimate temperature rise during FUS. CT images from 28 patients were analyzed to validate the use of Kranion. For stereotactic targets within each patient, individual transducer element incident angles, skull density ratio, and skull thickness measurements were recorded. A penetration metric (the “beam index”) was calculated by combining the energy loss from incident angles and the skull thickness. Kranion accurately estimated the patient’s skull and treatment parameters. The authors observed significant changes in incident angles with different targets in the brain. Using the beam index as a predictor of temperature rise in a linear-mixed-effects model, they were able to predict the average temperature rise at the focal point during ablation with < 21% error (55°C ± 3.8°C) in 75% of sonications, and with < 44% (55°C ± 7.9°C) in 97% of sonications. This research suggests that the beam index can improve the prediction of temperature rise during FUS. Additional work is required to study the relationship between temperature rise and lesion shape and clinical outcomes.
Francesco Sammartino, Dylan W. Beam, John Snell and Vibhor Krishna
David Schlesinger, John Snell and Jason Sheehan
The relative performances of two plugging strategies commonly used for pituitary adenoma dose plans were evaluated in terms of factors that influence dose plan quality.
Dose plans and clinical treatment data were obtained in 108 patients treated with the Model C Gamma Knife at the University of Virginia. These data were analyzed to determine factors (including plugging strategy) influencing the quality of the dose plans in terms of beam time, conformity, dose to the optic apparatus, and plugging burden.
For both secretory and nonsecreory adenomas, beam time (psecretory < 0.001, pnonsecretory = 0.015) and plugging burden (psecretory = 0.007, pnonsecretory = 0.038) were reduced when using the customized plugging strategy. The choice of plugging strategy was found to play no significant role in conformity or dose to the optic apparatus. Other factors found to play a significant role in adenoma dose plan quality included tumor volume, prescription dose, and distance from the target to the optic pathways.
While both plugging strategies were effective at providing the required protection to the optic pathways, the authors found that the customized plugging strategy provided more efficient performance in pituitary adenoma treatments.
Stephen Monteith, John Snell, Mathew Eames, Neal F. Kassell, Edward Kelly and Ryder Gwinn
In appropriate candidates, the treatment of medication-refractory mesial temporal lobe epilepsy (MTLE) is primarily surgical. Traditional anterior temporal lobectomy yields seizure-free rates of 60%–70% and possibly higher. The field of magnetic resonance–guided focused ultrasound (MRgFUS) is an evolving field in neurosurgery. There is potential to treat MTLE with MRgFUS; however, it has appeared that the temporal lobe structures were beyond the existing treatment envelope of currently available clinical systems. The purpose of this study was to determine whether lesional temperatures can be achieved in the target tissue and to assess potential safety concerns.
Cadaveric skulls with tissue-mimicking gels were used as phantom targets. An ablative volume was then mapped out for a “virtual temporal lobectomy.” These data were then used to create a target volume on the InSightec ExAblate Neuro system. The target was the amygdala, uncus, anterior 20 mm of hippocampus, and adjacent parahippocampal gyrus. This volume was approximately 5cm3. Thermocouples were placed on critical skull base structures to monitor skull base heating.
Adequate focusing of the ultrasound energy was possible in the temporal lobe structures. Using clinically relevant ultrasound parameters (power 900 W, duration 10 sec, frequency 650 kHz), ablative temperatures were not achieved (maximum temperature 46.1°C). Increasing sonication duration to 30 sec demonstrated lesional temperatures in the mesial temporal lobe structures of interest (up to 60.5°C). Heating of the skull base of up to 24.7°C occurred with 30-sec sonications.
MRgFUS thermal ablation of the mesial temporal lobe structures relevant in temporal lobe epilepsy is feasible in a laboratory model. Longer sonications were required to achieve temperatures that would create permanent lesions in brain tissue. Heating of the skull base occurred with longer sonications. Blocking algorithms would be required to restrict ultrasound beams causing skull base heating. In the future, MRgFUS may present a minimally invasive, non-ionizing treatment of MTLE.
John W. Snell, Jason Sheehan, Matei Stroila and Ladislau Steiner
✓ The Gamma Knife has played an increasingly important role in the neurosurgical treatment of patients. Intracranial lesions are not removed by radiosurgery. Rather, the goal of treatment is to induce tumor control. During planning, the creation of dose–volume histograms requires an accurate volumetric analysis of intracranial lesions selected for radiosurgery. In addition, an accurate follow-up imaging analysis of tumor volume is essential for assessing the results of radiosurgery. Nevertheless, sources of volumetric error and their expected magnitudes must be properly understood so that the operator may correctly interpret apparent changes in tumor volume. In this paper, the authors examine the often-neglected contributions of imaging geometry (principally image slice thickness and separation) to overall volumetric error.
One of the fundamental sources of volumetric error is that resulting from the geometry of the acquisition protocol. The authors consider the image sampling geometry of tomographic modalities and its contribution to volumetric error through a simulation framework in which a synthetic digital tumor is taken as the primary model. Because the exact volume of the digital phantom can be computed, the volume estimates derived from tomographic “slicing” can be directly compared precisely and independently from other error sources. In addition to providing empirical bounds on volumetric error, this approach provides a tool for guiding the specification of imaging protocols when a specific volumetric accuracy, or volume change sensitivity, for particular structures is sought a priori.
Using computational geometry techniques, the volumetric error associated with image acquisition geometry was shown to be dependent on the number of slices through the region of interest (ROI) and the lesion volume. With a minimum of five slices through the ROI, the volume of a compact lesion could be calculated accurately with less than 10% error, which was the predetermined goal for the purposes of computing accurate dose–volume histograms and determining follow-up changes in tumor volume.
Accurate dose–volume histograms can be generated and follow-up volumetric assessments performed, assuming accurate lesion delineation, when the object is visualized on at least five axial slices. Volumetric analysis based on fewer than five slices yields unacceptably larger errors (that is, > 10%). These volumetric findings are particularly relevant for radiosurgical treatment planning and follow-up analysis. Through the application of this volumetric methodology and a greater understanding of the error associated with it, neurosurgeons can better perform radiosurgery and assess its outcome.
Stephen Monteith, Jason Sheehan, Ricky Medel, Max Wintermark, Matthew Eames, John Snell, Neal F. Kassell and W. Jeff Elias
Magnetic resonance–guided focused ultrasound surgery (MRgFUS) has the potential to create a shift in the treatment paradigm of several intracranial disorders. High-resolution MRI guidance combined with an accurate method of delivering high doses of transcranial ultrasound energy to a discrete focal point has led to the exploration of noninvasive treatments for diseases traditionally treated by invasive surgical procedures. In this review, the authors examine the current intracranial applications under investigation and explore other potential uses for MRgFUS in the intracranial space based on their initial cadaveric studies.
Stephen J. Monteith, Sagi Harnof, Ricky Medel, Britney Popp, Max Wintermark, M. Beatriz S. Lopes, Neal F. Kassell, W. Jeff Elias, John Snell, Matthew Eames, Eyal Zadicario, Krisztina Moldovan and Jason Sheehan
Intracerebral hemorrhage (ICH) is a major cause of death and disability throughout the world. Surgical techniques are limited by their invasive nature and the associated disability caused during clot removal. Preliminary data have shown promise for the feasibility of transcranial MR-guided focused ultrasound (MRgFUS) sonothrombolysis in liquefying the clotted blood in ICH and thereby facilitating minimally invasive evacuation of the clot via a twist-drill craniostomy and aspiration tube.
Methods and Results
In an in vitro model, the following optimum transcranial sonothrombolysis parameters were determined: transducer center frequency 230 kHz, power 3950 W, pulse repetition rate 1 kHz, duty cycle 10%, and sonication duration 30 seconds. Safety studies were performed in swine (n = 20). In a swine model of ICH, MRgFUS sonothrombolysis of 4 ml ICH was performed. Magnetic resonance imaging and histological examination demonstrated complete lysis of the ICH without additional brain injury, blood-brain barrier breakdown, or thermal necrosis due to sonothrombolysis. A novel cadaveric model of ICH was developed with 40-ml clots implanted into fresh cadaveric brains (n = 10). Intracerebral hemorrhages were successfully liquefied (> 95%) with transcranial MRgFUS in a highly accurate fashion, permitting minimally invasive aspiration of the lysate under MRI guidance.
The feasibility of transcranial MRgFUS sonothrombolysis was demonstrated in in vitro and cadaveric models of ICH. Initial in vivo safety data in a swine model of ICH suggest the process to be safe. Minimally invasive treatment of ICH with MRgFUS warrants evaluation in the setting of a clinical trial.
Zhiyuan Xu, Carissa Carlson, John Snell, Matt Eames, Arik Hananel, M. Beatriz Lopes, Prashant Raghavan, Cheng-Chia Lee, Chun-Po Yen, David Schlesinger, Neal F. Kassell, Jean-Francois Aubry and Jason Sheehan
In biological tissues, it is known that the creation of gas bubbles (cavitation) during ultrasound exposure is more likely to occur at lower rather than higher frequencies. Upon collapsing, such bubbles can induce hemorrhage. Thus, acoustic inertial cavitation secondary to a 220-kHz MRI-guided focused ultrasound (MRgFUS) surgery is a serious safety issue, and animal studies are mandatory for laying the groundwork for the use of low-frequency systems in future clinical trials. The authors investigate here the in vivo potential thresholds of MRgFUS-induced inertial cavitation and MRgFUS-induced thermal coagulation using MRI, acoustic spectroscopy, and histology.
Ten female piglets that had undergone a craniectomy were sonicated using a 220-kHz transcranial MRgFUS system over an acoustic energy range of 5600–14,000 J. For each piglet, a long-duration sonication (40-second duration) was performed on the right thalamus, and a short sonication (20-second duration) was performed on the left thalamus. An acoustic power range of 140–300 W was used for long-duration sonications and 300–700 W for short-duration sonications. Signals collected by 2 passive cavitation detectors were stored in memory during each sonication, and any subsequent cavitation activity was integrated within the bandwidth of the detectors. Real-time 2D MR thermometry was performed during the sonications. T1-weighted, T2-weighted, gradient-recalled echo, and diffusion-weighted imaging MRI was performed after treatment to assess the lesions. The piglets were killed immediately after the last series of posttreatment MR images were obtained. Their brains were harvested, and histological examinations were then performed to further evaluate the lesions.
Two types of lesions were induced: thermal ablation lesions, as evidenced by an acute ischemic infarction on MRI and histology, and hemorrhagic lesions, associated with inertial cavitation. Passive cavitation signals exhibited 3 main patterns identified as follows: no cavitation, stable cavitation, and inertial cavitation. Low-power and longer sonications induced only thermal lesions, with a peak temperature threshold for lesioning of 53°C. Hemorrhagic lesions occurred only with high-power and shorter sonications. The sizes of the hemorrhages measured on macroscopic histological examinations correlated with the intensity of the cavitation activity (R2 = 0.74). The acoustic cavitation activity detected by the passive cavitation detectors exhibited a threshold of 0.09 V·Hz for the occurrence of hemorrhages.
This work demonstrates that 220-kHz ultrasound is capable of inducing a thermal lesion in the brain of living swines without hemorrhage. Although the same acoustic energy can induce either a hemorrhage or a thermal lesion, it seems that low-power, long-duration sonication is less likely to cause hemorrhage and may be safer. Although further study is needed to decrease the likelihood of ischemic infarction associated with the 220-kHz ultrasound, the threshold established in this work may allow for the detection and prevention of deleterious cavitations.
Stephen J. Monteith, Ricky Medel, Neal F. Kassell, Max Wintermark, Matthew Eames, John Snell, Eyal Zadicario, Javier Grinfeld, Jason P. Sheehan and W. Jeff Elias
Transcranial MR-guided focused ultrasound surgery (MRgFUS) is evolving as a treatment modality in neurosurgery. Until now, the trigeminal nerve was believed to be beyond the treatment envelope of existing high-frequency transcranial MRgFUS systems. In this study, the authors explore the feasibility of targeting the trigeminal nerve in a cadaveric model with temperature assessments using computer simulations and an in vitro skull phantom model fitted with thermocouples.
Six trigeminal nerves from 4 unpreserved cadavers were targeted in the first experiment. Preprocedural CT scanning of the head was performed to allow for a skull correction algorithm. Three-Tesla, volumetric, FIESTA MRI sequences were performed to delineate the trigeminal nerve and any vascular structures of the cisternal segment. The cadaver was positioned in a focused ultrasound transducer (650-kHz system, ExAblate Neuro, InSightec) so that the focus of the transducer was centered at the proximal trigeminal nerve, allowing for targeting of the root entry zone (REZ) and the cisternal segment. Real-time, 2D thermometry was performed during the 10- to 30-second sonication procedures. Post hoc MR thermometry was performed on a computer workstation at the conclusion of the procedure to analyze temperature effects at neuroanatomical areas of interest. Finally, the region of the trigeminal nerve was targeted in a gel phantom encased within a human cranium, and temperature changes in regions of interest in the skull base were measured using thermocouples.
The trigeminal nerves were clearly identified in all cadavers for accurate targeting. Sequential sonications of 25–1500 W for 10–30 seconds were successfully performed along the length of the trigeminal nerve starting at the REZ. Real-time MR thermometry confirmed the temperature increase as a narrow focus of heating by a mean of 10°C. Postprocedural thermometry calculations and thermocouple experiments in a phantom skull were performed and confirmed minimal heating of adjacent structures including the skull base, cranial nerves, and cerebral vessels. For targeting, inclusion of no-pass regions through the petrous bone decreased collateral heating in the internal acoustic canal from 16.7°C without blocking to 5.7°C with blocking. Temperature at the REZ target decreased by 3.7°C with blocking. Similarly, for midcisternal targeting, collateral heating at the internal acoustic canal was improved from a 16.3°C increase to a 4.9°C increase. Blocking decreased the target temperature increase by 4.4°C for the same power settings.
This study demonstrates focal heating of up to 18°C in a cadaveric trigeminal nerve at the REZ and along the cisternal segment with transcranial MRgFUS. Significant heating of the skull base and surrounding neural structures did not occur with implementation of no-pass regions. However, in vivo studies are necessary to confirm the safety and efficacy of this potentially new, noninvasive treatment.
John W. Thompson, Omar Elwardany, David J. McCarthy, Dallas L. Sheinberg, Carlos M. Alvarez, Ahmed Nada, Brian M. Snelling, Stephanie H. Chen, Samir Sur and Robert M. Starke
Cerebral aneurysm rupture is a devastating event resulting in subarachnoid hemorrhage and is associated with significant morbidity and death. Up to 50% of individuals do not survive aneurysm rupture, with the majority of survivors suffering some degree of neurological deficit. Therefore, prior to aneurysm rupture, a large number of diagnosed patients are treated either microsurgically via clipping or endovascularly to prevent aneurysm filling. With the advancement of endovascular surgical techniques and devices, endovascular treatment of cerebral aneurysms is becoming the first-line therapy at many hospitals. Despite this fact, a large number of endovascularly treated patients will have aneurysm recanalization and progression and will require retreatment. The lack of approved pharmacological interventions for cerebral aneurysms and the need for retreatment have led to a growing interest in understanding the molecular, cellular, and physiological determinants of cerebral aneurysm pathogenesis, maturation, and rupture. To this end, the use of animal cerebral aneurysm models has contributed significantly to our current understanding of cerebral aneurysm biology and to the development of and training in endovascular devices. This review summarizes the small and large animal models of cerebral aneurysm that are being used to explore the pathophysiology of cerebral aneurysms, as well as the development of novel endovascular devices for aneurysm treatment.
Benjamin Davidson, Karim Mithani, Yuexi Huang, Ryan M. Jones, Maged Goubran, Ying Meng, John Snell, Kullervo Hynynen, Clement Hamani and Nir Lipsman
Magnetic resonance imaging–guided focused ultrasound (MRgFUS) is an emerging treatment modality that enables incisionless ablative neurosurgical procedures. Bilateral MRgFUS capsulotomy has recently been demonstrated to be safe and effective in treating obsessive-compulsive disorder (OCD) and major depressive disorder (MDD). Preliminary evidence has suggested that bilateral MRgFUS capsulotomy can present increased difficulties in reaching lesional temperatures as compared to unilateral thalamotomy. The authors of this article aimed to study the parameters associated with successful MRgFUS capsulotomy lesioning and to present longitudinal radiographic findings following MRgFUS capsulotomy.
Using data from 22 attempted MRgFUS capsulotomy treatments, the authors investigated the relationship between various sonication parameters and the maximal temperature achieved at the intracranial target. Lesion volume and morphology were analyzed longitudinally using structural and diffusion tensor imaging. A retreatment procedure was attempted in one patient, and their postoperative imaging is presented.
Skull density ratio (SDR), skull thickness, and angle of incidence were significantly correlated with the maximal temperature achieved. MRgFUS capsulotomy lesions appeared similar to those following MRgFUS thalamotomy, with three concentric zones observed on MRI. Lesion volumes regressed substantially over time following MRgFUS. Fractional anisotropy analysis revealed a disruption in white matter integrity, followed by a gradual return to near-baseline levels concurrent with lesion regression. In the patient who underwent retreatment, successful bilateral lesioning was achieved, and there were no adverse clinical or radiographic events.
With the current iteration of MRgFUS technology, skull-related parameters such as SDR, skull thickness, and angle of incidence should be considered when selecting patients suitable for MRgFUS capsulotomy. Lesions appear to follow morphological patterns similar to what is seen following MRgFUS thalamotomy. Retreatment appears to be safe, although additional cases will be necessary to further evaluate the associated safety profile.