High-Intensity Focused Ultrasound Surgery for the Treatment of Obsessive–Compulsive Disorder

Introduction

Obsessive–compulsive disorder (OCD) is a common and disabling neuropsychiatric disorder that affects 2%–3% of the global population at some point in their life; additionally, it is known that 6%–8% of the general population has subclinical symptoms ( ). OCD is characterized by recurrent obsessions (intrusive anxious thoughts, desires, or images) and/or compulsions (repetitive ritualized behaviors used to prevent or reduce distress) ( ). In most cases patients are aware that their obsessive thoughts and compulsive behaviors are unreasonable or illogical, but they have difficulty defying them; in some cases this disorder causes marked distress, anxiety, or significant impairments in daily normal functioning and social relationships. Currently the standard therapeutic options consist of pharmacological therapies, such as selective serotonin reuptake inhibitors (SSRIs), and cognitive behavioral therapy (CBT). Although most patients with OCD will at least partially respond to standard treatment, a considerable number will continue to experience symptoms, leading to chronic functional impairments with little to no relief ( ). In such patients with disabling, treatment-resistant OCD, surgical treatments can be considered.

Several surgical procedures, including lesioning and chronic electrical stimulation of various areas along the neural circuitry associated with OCD, have been used to treat patients with severe refractory OCD. Stereotactic ablative techniques using radiofrequency, such as anterior capsulotomy and cingulotomy, have also been effective ( ). However, such procedures are invasive and the outcomes vary. Additionally, since these procedures are irreversible, concerns have been raised regarding the potential for permanent surgical complications. Compared to radiofrequency thermal ablation, stereotactic radiosurgery (e.g., gamma-knife radiosurgery) is a fully noninvasive procedure with similar clinical results, but the procedure is associated with potentially unpredictable radiation-induced adverse effects, which could be transient or permanent ( ). The biological and clinical responses to radiosurgery are also variable; but the most important limitation of radiosurgery is the considerable delay in the appearance of clinical effects, i.e., weeks to months. Thus both radiofrequency ablative procedures and stereotactic radiosurgery have irreversible features and neither can be adjusted to correct the size and/or location of the target during the procedure.

For the past two decades deep brain stimulation (DBS) has been validated as an alternative therapeutic option to ablative procedures for movement disorders, including Parkinson’s disease, essential tremor, and dystonia. The first study to use DBS in patients with treatment-refractory OCD was performed by , who positioned an electrode within the anterior limb of the internal capsule (ALIC), adapting the anatomical target used in the ablative procedure. Since this initial study the use of DBS in patients with OCD has been investigated with various targets, including the ALIC ( ), ventral caudate ( ), subthalamic nucleus ( ), inferior thalamic peduncle ( ), and nucleus accumbens ( ). DBS has certain advantages over ablative procedures, specifically its reversibility and adjustability. However, it also has disadvantages, including its inherent hardware-related complications, expense, and maintenance demands, as well as the potential for infection and hemorrhage ( ).

Recently a novel thermal ablation procedure using high-intensity focused ultrasound (HIFU), namely magnetic-resonance-guided focused ultrasound surgery (MRgFUS), has been gaining increased attention as a promising and noninvasive alternative to traditional neurosurgery owing to its unique ability to focus acoustic energy through the intact skull and on to a precise target within the brain. Unlike the neurosurgical modalities mentioned above, MRgFUS can be monitored continuously in real time with magnetic resonance imaging (MRI) and MR thermography while avoiding the potential side-effects of ionizing radiation. Several clinical phase I trials have demonstrated the feasibility and safety of using MRgFUS to treat various neurological disorders, such as chronic neuropathic pain syndrome ( ), essential tremor ( ), and tremor-dominant Parkinson’s disease ( ). In a previous study we also demonstrated that bilateral anterior capsulotomy with MRgFUS can be utilized to treat patients with medically refractory OCD without side-effects ( ).

In this chapter we describe the current understanding of the neural circuitry of OCD and review the history, general principles, and typical treatment protocols of MRgFUS. The chapter also discusses the clinical results of studies that applied MRgFUS to patients with treatment-refractory OCD to establish its clinical feasibility.

Neural Circuitry of Obsessive–Compulsive Disorder

Despite the remarkable developments in neuroimaging techniques and the abundant experience gained in performing functional neurosurgeries, the exact pathophysiology of OCD currently remains unknown. Several studies that examined OCD using positron emission tomography demonstrated increased regional cerebral blood flow and metabolic activity within the orbitofrontal cortex, anterior cingulate cortex (ACC), and caudate nucleus ( ) which could be magnified by provoking OCD symptoms ( ). Interestingly, decreased activity in the dorsolateral prefrontal cortex was also noted in individuals with OCD ( ). Executive functions, such as planning, are often impaired in OCD patients, and functional MRI during planning tasks revealed decreased frontal–striatal responsiveness, mainly in the dorsolateral prefrontal cortex and caudate nucleus ( ). Frontal–striatal control of the limbic structures which mediate the inputs to the amygdala may also be decreased in patients with OCD and may be responsible for certain clinical manifestations, including the fear of contamination ( ). These findings indicate that abnormalities in the frontal–striatal neural circuitry play important roles in the pathogenesis of OCD.

Based on these findings, the general organization of the neural circuits involved in OCD is as follows. These circuits project from several separate but functionally connected specific territories in the frontal cortex to corresponding targets within the striatum, and then connect to the basal ganglia and thalamus via direct and indirect pathways, finally sending recurrent projections back to the original frontal territory where each loop began ( ). A multicircuit hypothesis of OCD mentioned that the primary pathogenic mechanism is dysregulation of the basal ganglia and limbic–striatal circuitry, which modulate neuronal activity in and between the orbitofrontal cortex and ACC ( ). In more simplified terms, the neural circuitry involved in OCD consists of three components ( ): the cortico–thalamic pathway, a positive feedback loop from the orbital and prefrontal cortex to the thalamus via the ALIC; the cortico–striato–thalamo–cortical loop, an inhibitory pathway connecting the orbitofrontal cortex, caudate nucleus, globus pallidus, and thalamus that receives serotonergic projections from the midbrain into the striatum; and portions of the limbic system, including the hippocampus, mammillary bodies, and fornix, linking to the thalamus and ACC. Because excessive stimulation of the cortico–thalamic pathway or decreased inhibition of the cortico–striato–thalamo–cortical loop might lead to OCD symptoms, the ALIC, ACC, and nucleus accumbens are often used as therapeutic targets in patients with OCD ( ).

History of High-Intensity Focused Ultrasound

The first investigation to use HIFU for noninvasive ablation was performed by Lynn et al. in . They produced focal lesions deep inside the brain and spinal cord, which generated well-demarcated tissue changes through the intact skull without damaging the surrounding nontargeted tissue. However, the delivered ultrasound energy invariably caused thermal injuries to the skin and underlying tissues, including the scalp, muscles, and even the meninges, resulting from absorption of the ultrasound while passing through the skull bone ( ). Moreover, the irregular thickness and inhomogeneous density of the intact skull caused ultrasonic beam aberrations that disturbed focusing. In the early 1950s William and Francis Fry demonstrated that HIFU could be used after a craniectomy to target deep-seated areas of the basal ganglia in primate models ( ); thus in the following years investigations of HIFU used craniectomies to avoid skull heating, ultrasound attenuation, ultrasonic beam distortion, and reflection ( ). Over time, substantial important technical developments were achieved, and as a result several studies investigated using HIFU as a potential alternative modality to contemporary neurosurgical procedures and radiosurgery ( ). However, as distinct from the need to create an acoustic window by removing parts of the skull, another factor limiting the utility of HIFU on the brain was the lack of adequate imaging modalities that could accurately visualize the target and detect real-time temperature changes during thermal ablation.

To overcome this problem, Fry et al. attempted to sonicate through the intact skull without performing a craniectomy in the 1970s and 1980s. In doing so, the authors demonstrated that while it was possible to focus the ultrasound energy in a manner that would create focal brain lesions through the intact skull, the resultant foci were severely distorted and shifted ( ). A solution to this problem only became clinically feasible in the 1990s with the development of the multielements phased array ultrasound transducers; these transducers permitted the focusing of ultrasound beams by correcting for phase distortions based on computed tomography (CT) scans, which were used to measure the skull thickness and density ( ).

The other major revolution that permitted HIFU to be used clinically was the improvement in MRI techniques. MRI, which was developed in the 1980s, has a much better sensitivity than other imaging modalities in terms of target localization and is capable of measuring temperature changes with considerable accuracy. Because of its temperature sensitivity, the focal volume of a thermal lesion can be localized and controlled well before irreversible tissue injury is induced. Thus with the development of ultrasound phased arrays to correct for cranial distortions, MRI to provide intraoperative guidance, and MR thermography to monitor the thermal deposition during treatment, modern HIFU technology is able to overcome its previous limitations ( ).

Principles of Transcranial Magnetic-Resonance-Guided Focused Ultrasound Surgery