Failure of fear extinction is a core feature of posttraumatic stress disorder (PTSD).1 Recently, it was confirmed that the amygdala and the orbitofrontal cortex are crucial for both fear acquisition and fear extinction.2 The amygdala was found to have neurons active only during fear acquisition, and other neurons active only during fear extinction.3 In essence, the balance of activity between these 2 neuronal populations determines whether if an incoming stimulus is feared or not feared. This balance is under the influence of several cognitive domains, including memory, reward, and executive function.
In PTSD, the equilibrium is shifted heavily toward fear acquisition. The majority of patients spontaneously regain the capacity for fear extinction over time4 or with the help of treatment.5,6 Nonetheless, some patients with severe PTSD seem unable to recover the ability of fear extinction and remain refractory to both standard and novel psychotherapeutic or psychopharmacologic treatments.7 For these patients, direct modulation of the neural activity in the amygdala may permit fear extinction. This article describes the rationale for using deep brain stimulation (DBS) and initial results from the first-ever clinical trial.
Deep Brain Stimulation
Deep brain stimulation involves inserting electrodes in precise cerebral targets and then connecting the leads to a pulse generator (similar to a pacemaker) inserted in a subclavicular pocket. The generator controls the electrical signal (amplitude, pulse width, pulse frequency) delivered to the brain target and can be adjusted with use of a noninvasive programmer. In 1997, the FDA approved DBS for patients with Parkinson disease or essential tremor. Since then, its efficacy in these movement disorders has been confirmed in several studies.8,9
The mechanism by which the small electrical pulses of DBS influence activity is not clear. Clinically, DBS functionally inhibits the activity of local neurons.10 One theory describes “frequency jamming,” a concept similar to cardiac overdrive pacing in which the resultant high-frequency neuronal signal is meaningless and discounted by the rest of the brain.11
Over the years, DBS has demonstrated a strong safety profile.12 The risks of electrode insertion are mitigated with targeting based on high-quality magnetic resonance imaging (MRI) and computed tomography (Figure). Unlike a destructive lesion, DBS is reversible, and the implanted system can be removed in its entirety. Histologic analyses have shown only a small amount of scarring around the electrode tip.13 In movement disorder treatment, clinical experience has shown that stimulation-related adverse effects (AEs) are reversible with readjustment of stimulation parameters by external programmer.14
Novel Applications of DBS
The advantageous safety profile of DBS has permitted its evaluation in the treatment of other conditions thought to have malfunctioning networks at their core—such as intractable epilepsy (in resective surgery noncandidates).15,16 Although several trials have shown promising results of using DBS for treatment-resistant depression,17 the results of pivotal sham-controlled trials have been mixed.18,19 Obsessive-compulsive disorder, on the other hand, received the FDA humanitarian device exemption designation on the basis of positive long-term results.20 In treatment-resistant depression and obsessive-compulsive disorder, functional neuroimaging has identified DBS targets.21,22 Functional MRI or positron emission tomography (PET) images can be compared at resting state, at symptomatic state, and after treatment response. Nodes hyperactive during a symptomatic state and less active after successful treatment can be targeted with high-frequency DBS to directly reduce the hyperactivity and indirectly modulate or normalize the overall function of the circuit.23
Given the functional MRI and O15 (oxygen-15) PET evidence of amygdala hyperactivity in patients with PTSD having core symptoms,24-26 the authors hypothesized that high-frequency DBS targeting of the amygdala would improve PTSD-associated hyperarousal and reexperiencing symptoms in treatment-refractory patients. Indirect data supporting this hypothesis include a correlation between amygdala hyperactivity of increased intensity and symptom severity measured with the Clinician-Administered PTSD Scale (CAPS),27 and a correlation between reduced pretreatment amygdala hyperactivity and successful cognitive-behavioral treatment.28,29
Using a rodent model in which a novel object serves as a cue reminder of foot shocks (traumatic event), the authors tested the hypothesis that amygdala DBS would reduce PTSD-like symptoms.30 When untreated rats were presented with the object in their cage a week after the initial exposure, they immediately buried the object under bedding to avoid being reminded of the shocks. In contrast, rats treated with DBS did not bury the object. In most cases, in fact, they played with it.
The authors also replicated their results but with the addition of rats treated with paroxetine.31 Using the same rodent model, they found DBS superior to paroxetine in treating PTSD-like symptoms. This study had a crossover design: DBS and sham DBS. Briefly, 20 rats received an electrode in the amygdala and were exposed to inescapable shocks in the presence of the cue object. The rats were then randomly assigned to a DBS group (10 rats) or a sham-DBS group (10 rats). After 1 week, behavioral testing showed fear extinction in the DBS group and no improvement in the sham-DBS group. Then the groups were switched: The rats originally treated with DBS received no treatment, and the rats that were originally sham-treated underwent DBS. One week later, behavioral testing showed acquisition of fear extinction in all the rats. These results suggested DBS can be effective even when delayed after establishment of fear persistence and PTSD symptoms. These results also showed that DBS effects persist even after therapy discontinuation.
Similarly, other investigators have reported that the role of the amygdala is not limited to fear acquisition; it extends to fear expression. A lesion in the amygdala can prevent fear expression even if the disruption is performed subsequent to fear-conditioning training.32 This finding is important for humans, as DBS would be initiated during the chronic phase of the disorder, after failure of less invasive treatment options, such as pharmacotherapy and psychotherapy.