Physical Address
304 North Cardinal St.
Dorchester Center, MA 02124
Recent advances in the transcranial delivery of acoustic energy and magnetic resonance thermography have allowed neurosurgeons to use ultrasonography for therapeutic purposes.
Transcranial magnetic resonance–guided focused ultrasound thalamotomy is an incisionless procedure that has been approved by the U.S. Food and Drug Administration for the unilateral treatment of medication-refractory essential tremor and tremor-dominant Parkinson disease.
Focused ultrasonography has been shown to have similar clinical efficacy to other ablative and neuromodulatory techniques for the surgical treatment of unilateral tremor.
The high accuracy, safety, and efficacy of transcranial magnetic resonance–guided focused ultrasonography is leading to a resurgence in cerebral lesioning for the treatment of tremor.
Further technical advances in brain imaging and transcranial ultrasound delivery will likely lead to new frontiers in noninvasive brain mapping and less invasive treatments for neurologic and psychiatric diseases.
Neurosurgeons have rarely used ultrasonography (US) other than for diagnostic imaging. The concept of using high-intensity focused ultrasonography (HIFU) in the brain was first conceived in the early 1950s, during a time of rapid growth in stereotactic neurosurgery. William and Frank Fry, both physicists who studied US technology, described the use of HIFU to produce precise, focal lesions deep in the feline brain without violating the dura and without causing damage to the intervening tissues or blood vessels. , These seminal experiments used a four-element, 1-MHz transducer to reliably produce a US beam about 1 mm in diameter that was capable of contouring lesions with slight movements of the focus. Sonication intensities approaching 210 W/cm 2 for 4 seconds created “reversible” brain lesions in craniectomized cats and nonhuman primates. The Fry brothers collaborated with Russell Meyers, a pioneer in open surgery of the basal ganglia, to produce ultrasonic lesions in patients with Parkinson disease (PD) and other movement disorders. These treatments involved a custom stereotactic apparatus, contrast ventriculography, and a craniotomy to transmit acoustic energy to the pallidothalamic tracts. The results were favorable, with improvements in tremor and rigidity and relatively few serious side effects.
Around the same time that the Fry brothers were studying the therapeutic uses of focused US (FUS), the field of diagnostic US was developing in Lund, Sweden. Ultrasonic reflectoscopes had been designed to test material integrities in shipyards and factories. The first echocardiograms were obtained when cardiologists used these scopes to image heart motion and the mitral valve. Lars Leksell borrowed a device in 1956 and successfully identified the pineal shift in a 16-month-old child with a subdural hematoma. Although US transmission was possible in the neonatal skull, Leksell eventually became disenchanted by the limitations of transcranial US delivery through the mature, human skull because it required a cranial window. Even though Leksell acquired several patents for the use of US in the brain, he eventually pursued ionizing radiation as a less invasive method for the treatment of pain and other neurologic disorders.
Others utilized US through craniotomies for deep lesioning in the brain. Lindstrom treated 25 patients with cancer for their pain and anxiety with US leucotomy as an alternative to surgical lobotomy. , Histologic findings were ultimately available in these patients. Sixteen patients exhibited clinical improvement without microscopic change, a finding that was interpreted as a “functional lesion” selective to white matter tracts. , Heimburger reported prolonged survival in patients with glioma after sonicating through their craniotomies following initial resection. Despite these early successes, FUS lesioning necessitated craniotomy for acoustic transmission, so US therapy in the brain never gained widespread popularity among neurosurgeons.
Ultrasound can be defined as an acoustic pressure wave beyond the range of normal human hearing (>20 kHz). As with other waveform energies that follow simple harmonic motion, US is characterized by its frequency, wavelength, amplitude, power, and intensity. Safe medical imaging US at frequencies ranging from 2 to 20 MHz creates waveforms with peak intensities less than 1 W/cm , as regulated by the U.S. Food and Drug Administration (FDA). , Therapeutic HIFU for brain ablation uses lower frequencies (0.2 to 1 MHz), which better propagate through bone, and the peak intensities may exceed 1000 W/cm 2 and are therefore capable of tissue ablation or cavitation.
As US pressure waves travel through a given medium, molecules are displaced and oscillate according to the magnitude of the wave and the elasticity of the medium. Frictional energy and ultimately heat are generated by the molecules as they oscillate next to each other – a phenomenon not dissimilar to that of radiofrequency (RF) energy. With FUS, the pressure at the focus is exponentially increased, so thermal tissue ablation can be achieved in the focus without damage to surrounding tissue. ,
At high intensities, the acoustic pressure wave moving through tissue can also induce cavitation, which results when the acoustic pressure wave extracts gas that coalesces into microbubbles that oscillate in the ultrasonic field. , At lower intensities, these bubbles can expand and contract in a sustainable, periodic fashion termed stable cavitation. However, if the bubbles are subjected to higher intensities, they can potentially collapse violently (inertial cavitation) and damage tissues with extremely high temperatures, powerful jet streams, and/or free radicals. , The propensity for microbubble formation and cavitation raises concerns for HIFU ablation and has resulted in the implementation of cavitation monitoring. Theoretically, stable cavitation could be monitored and harnessed to enhance therapeutic ablations or mechanical opening of the blood-brain barrier, but inertial cavitation poses great risk to tissue because its mechanical effects are more difficult to predict and control.
The skull is the major limiting factor for efficient delivery of US to the brain. Depending on the US frequency, the skull may reflect or absorb about 90% of the US energy. The potential for heating of the skull and adjacent soft tissues (dura and periosteum) exists if the skull overheats from absorbed US energy. Another challenge of transcranial US delivery relates to the variable thickness and density of the skull, which diffracts acoustic waves and leads to difficulty achieving a sharp acoustic focus.
Over the past two decades, two technologic advancements have addressed the high acoustic impedance and attenuation of the skull, allowing for transcranial HIFU. In the first, Clement and Hynynen and their colleagues developed a large-aperture, hemispherical transducer to maximize the area of transcranial acoustic energy delivery, thus minimizing local skull heating by distributing energy over the large surface area of the scalp and skull. Furthermore, an interface containing cold water is used to decrease the baseline temperature of the skull and keep it cooled during treatment ( Fig. 117.1 ). This also serves as a coupling medium for the US transducer with the scalp. The second advance, to account for the variability of the skull, was development of a phase-correction technology in parallel with multi-element, hemispherical transducers to correct aberrations in the acoustic beam traveling through the inhomogeneous skull. In this technology, a CT-based correction algorithm reconstructs the skull and computes the local characteristics of each location where an acoustic wave interfaces with the skull. Each element on the transducer can then be electronically adjusted and “phase shifted” to bring all of the beams to a sharp focus.
MR thermography measures temperature-dependent changes in water proton resonance frequency, thus enabling thermal monitoring during transcranial sonication. , Pre-sonication images are subtracted on a voxel-by-voxel basis from images obtained during tissue heating by US sonication. Images are acquired every 3 seconds, and tissue temperature is updated and plotted during the prescribed sonication ( Fig. 117.2 ). The images can be acquired in any of three planes (axial, coronal, and sagittal) to form a two-dimensional temperature map of change that is accurate to approximately 1°C in the plane perpendicular to the resonance frequency.
The intraoperative measurement of temperature can also be accumulated to represent the thermal dose delivered to each voxel. The Sapareto and Dewey equation was developed to estimate the thermal dose delivered to tissue on the basis of time-temperature profiles. Exposure of tissue for 240 minutes to 43°C is considered the critical dose for permanent tissue damage and cell death. Using the Sapareto and Dewey equation, one can calculate the equivalent time to achieve cell death for any temperature. This determination of thermal dose is necessary because of its nonlinear relationship with temperature, in which small increases in temperature can lead to large increases in thermal dose. This determination enables an intraoperative assessment of the ablated tissue and supports clinical decision making about additional treatment.
MRgFUS thalamotomy, like other stereotactic lesioning procedures, can be performed as a unilateral procedure for patients with severe, medication-refractory tremor. Ideal candidates tend to have asymmetric, appendicular tremors resulting in significant disabilities and functional limitations. Patients may choose this procedure because of preconceived bias against open neurosurgical procedures, implanted devices, or radiation therapies. Because this is an MR-guided procedure, patients with pacemakers, MR-incompatible implants, or claustrophobia may not be candidates for it.
Preoperative imaging studies include volumetric CT and MRI with T1- and T2-weighted sequences. Brain MRI with a high field strength and head coil is used preoperatively to obtain the highest-resolution images, which can be imported into the system for use during the procedure. CT is critical to application of the skull correction algorithm for each US element so that efficient transcranial US delivery is ensured. CT also allows for the identification of “no-pass regions” like aerated sinuses and intracranial calcifications, which can impede the passage of US. In addition, a CT scan is necessary to determine the radiodensity of cortical and cancellous bone of the skull in Hounsfield units, and the ratio of these values is used to calculate a skull density ratio (SDR). Patients with an SDR ≥ 0.45 generally undergo more efficient transmission of US to the intracranial target. However, SDR is not a perfect measure, as focal bony irregularities are often not reflected within the global average of SDR and may hinder thermal rise.
The patient is prepared on the morning of the procedure with intravenous access and compression leg stockings. A mild anxiolytic may be administered if necessary. Because hair follicles can harbor microbubbles of air, potentially leading to cavitation and scalp burns, the patient’s head is carefully shaved in its entirety. A stereotactic frame is placed as low as possible near the orbital rim and parieto-occipital junction to make maximal skull surface area available for transcranial sonication ( ). A silicone membrane is affixed to the scalp and attached to the FUS transducer, creating a coupling medium for US transduction as chilled and degassed water is circulated to cool the scalp. The transducer is manually aligned so that the natural acoustic focus approximates the stereotactic target. The patient, secured by the frame to the FUS transducer, is then positioned inside the bore of the MRI scanner to its isocenter.
During the imaging phase, a series of localizer MR images are acquired so that higher-quality T1- or T2-weighted images can be prescribed orthogonally to the line connecting the anterior commissure (AC) and posterior commissure (PC). These “reference” images are fused to the preoperatively obtained CT scan and highest-resolution MR images.
After image acquisition, stereotactic planning can be performed with neuronavigation software either by indirect targeting from AC-PC measurements or by direct targeting to an imaged structure. The ventral intermediate (VIM) nucleus of the thalamus, defined in the Schaltenbrand and Wahren neurosurgical thalamic nomenclature, is indirectly targeted in the treatment of essential tremor (ET) and tremor-dominant PD (TDPD). Our typical target for lesioning is planned at 25% of the AC-PC distance anterior to the PC and 14 mm lateral to midline, with adjustments made laterally in the case of ventriculomegaly and in the dorsal dimension for cases of less efficient transcranial delivery (typically in patients with low SDR skull scores). The VIM nucleus receives input from the cerebellum via the dentatorubrothalamic tract (DRT) and can also be targeted using diffusion tensor imaging (DTI) connectivity.
Final planning preparations are made before treatment is begun. Fiducial markers are designated to aid the operator with visual detection of intraprocedural movement, although the current system will automatically recognize a movement of the patient in relation to the transducer. In this case, the prescribed sonication will be halted. Lastly, a final adjustment of the transducer is made so that the natural focus is positioned to precisely match the stereotactic target.
The treatment phase relies heavily on MR thermography using the proton resonance frequency shift method to monitor the size, shape, and location of heating in multiple planes as well as on continuous clinical evaluation for safety and symptom response. The treatment phase involves three stages of sonication: (1) focus alignment, (2) target verification, and (3) therapeutic ablation.
In the alignment phase, low-energy sonications (i.e., 150 watts × 10 seconds = 1500 joules) are prescribed to elevate the temperature at the target to a peak voxel temperature ranging from 40°C to 45°C. With short exposure times, this temperature range theoretically does not reach the threshold at which tissue damage would occur. The thermal spot location is assessed in each orthogonal plane with two-dimensional MR thermography. If necessary, the system can electronically adjust the thermal spot to precisely match the planned target location.
During the verification stage, the acoustic energy dose is slowly escalated through an increase in power, duration, or both to elevate target temperatures moderately. Each brief sonication is then assessed with MR thermography and clinical evaluation. We have observed that neurological effects begin to manifest at temperatures in the “low 50s,” most notably transient or partial tremor suppression or paresthesia during VIM thalamotomy. Thermal neuromodulation in the range of 50°C to 55°C can be utilized to validate the stereotactic target, but it is important to recognize that even this early heating could result in a thermal dose that is sufficient to damage tissue. If the peak temperature deviates more than 1 mm from the planned target or symptom response helps refine the targeting, the acoustic focus can be adjusted by means of electronic steering by reprogramming of the acoustic elements. This thermal neuromodulation provides a means to personalize and optimize the treatment from clinical feedback.
Once the heating pattern has been verified to the proper location and clinical target, the acoustic energy is increased to temperatures necessary for therapeutic ablation, typically peak voxel temperatures in the range of 55°C to 60°C (see Fig. 117.2 ). The typical duration of FUS heating, 10 to 20 seconds, results in a thalamic lesion analogous to that produced by RF lesioning at 70°C for 60 seconds. We are cautious about excessive heating because microscopic hemorrhages have been observed with temperatures above 60°C in animal models. The lesion can be assessed with T2-weighted MRI while the patient is still coupled to the US transducer, although higher-quality images are typically obtained with the water drained. Early imaging typically shows a small area (2 to 3 mm) of T2 hyperintensity at the site of ablation and can be used to plan additional ablation, which is performed until satisfactory tremor suppression is observed. In our practice, a tandem ablation is made 2 mm dorsal to the original target to enlarge the size of the lesion within the VIM nucleus.
During transcranial MRgFUS thalamotomy for tremor, the patient is awake to allow for clinical monitoring throughout the procedure. Between sonications, the physician is able to interact with the patient to monitor his or her condition and to perform clinical assessments for somatosensory symptoms and motor performance as well as for tremor response (see ) Resting and postural tremors are easily observed, and intention tremor can be assessed with finger-to-nose testing and/or by having the patient perform simple spiral and line drawings.
Become a Clinical Tree membership for Full access and enjoy Unlimited articles
If you are a member. Log in here