February 2011  << Back  

  Can't open the PDF? Click here for help.



The emerging use of technology for the treatment of depression and other neuropsychiatric disorders

Robert H. Howland, MD

Department of Psychiatry, Western Psychiatric Institute and Clinic, University of Pittsburgh Medical Center, Pittsburgh, PA, USA

Luann S. Shutt, MSN

Department of Psychiatry, Western Psychiatric Institute and Clinic, University of Pittsburgh Medical Center, Pittsburgh, PA, USA

Susan R. Berman, MEd

Department of Psychiatry, Western Psychiatric Institute and Clinic, University of Pittsburgh Medical Center, Pittsburgh, PA, USA

Crystal R. Spotts, MEd

Department of Psychiatry, Western Psychiatric Institute and Clinic, University of Pittsburgh Medical Center, Pittsburgh, PA, USA

Timothey Denko, MD

Department of Psychiatry, Western Psychiatric Institute and Clinic, University of Pittsburgh Medical Center, Pittsburgh, PA, USA

BACKGROUND: Our objective is to review emerging technologies intended for the treatment of depression and other neuropsychiatric disorders.

METHODS: These technologies include repetitive transcranial magnetic stimulation (rTMS), magnetic seizure therapy (MST), vagus nerve stimulation (VNS), deep brain stimulation (DBS), cortical brain stimulation (CBS), and quantitative electroencephalography (QEEG).

RESULTS: The rationale for these technologies, their mechanisms of action, and how they are used in clinical practice are described. rTMS and VNS are effective for treatment-resistant depression. DBS is effective for resistant obsessive-compulsive disorder. QEEG can help predict a patient’s response to an antidepressant. All of these technologies continue to be investigated in treatment studies.

CONCLUSIONS: As these and other emerging technologies for depression and other neuropsychiatric disorders are development and applied, psychiatrists should understand the rationale for these modalities, how they work, and how they can be used in clinical practice.

KEYWORDS: depression, treatment, technology



Neuroanatomical and brain imaging studies can identify distinct brain regions and the pathways that connect them; these pathways may underlie various psychiatric disorders. Delineating such brain circuits is relevant because these disorders are unlikely to be explained solely by focusing on simple neurotransmitter, genetic, or specific brain region abnormalities.1,2 For instance, a cortical-limbic-thalamic-striatal neural circuit that is important for understanding depression and obsessive-compulsive disorder (OCD) has been roughly outlined.3,4 Within this circuit, certain brain regions are relatively overactive whereas other regions are underactive. Effective treatments ultimately may work by modulating the function of this circuit. Drug therapies,5,6 psychotherapies,7,8 and even placebos9,10 have been associated with functional changes in this circuit, although how and where the changes occur in the circuit vary depending on the treatment modality.

During the 1940s and 1950s under the influence of Egas Moniz and Walter Freeman, many patients with severe intractable psychiatric disorders were treated by frontal lobotomy, which resulted in destruction of the white matter tracts of the frontal lobes of the brain.11,12 Although some patients improved, many others suffered irreversible personality deterioration and surgical complications. With the advent of psychotropic drugs, use of this controversial treatment declined.

Unfortunately, available therapies are not always effective. The development of modern stereotactic neurosurgical methods13 has led to a renewed interest in neurosurgical interventions. These involve selective ablation or lesioning of specific brain regions that can be effective for treatment-resistant depression (TRD)14,15 and treatment-resistant OCD (TROCD).16 Similar neurosurgical methods also have been used to implant electrodes in the brain. Stimulation by these electrodes with pacemaker-like devices can modulate brain function by stimulating or inhibiting activity of specific brain regions without causing permanent, irreversible lesions.

In this paper, we review emerging technologies intended for the treatment of depression and other neuropsychiatric disorders. These technologies include transcranial magnetic stimulation, magnetic seizure therapy, vagus nerve stimulation, cortical brain stimulation, deep brain stimulation, and quantitative electroencephalography. As these new technologies continue to emerge, clinicians need to understand the rationale for their use, how they work, and how they are used in clinical practice.

Repetitive transcranial magnetic stimulation (rTMS)

rTMS is a novel, non-invasive way to produce focal non-electrical stimulation of the brain.17 Unlike electroconvulsive therapy (ECT), rTMS is not intended to cause seizures. Instead, a high-intensity electrical current is passed through an electromagnetic coil on the scalp and rapidly turned on and off to generate repetitive magnetic pulses that are focused on particular regions of the brain, depending on the therapeutic intent. For mood disorders, rTMS usually is directed at the prefrontal cortex because neuropsychological studies, neuroimaging studies, and post-mortem investigations have implicated disturbances in the function of this brain region.4 Various stimulation parameters of rTMS have different effects on neuronal excitability.17 Low frequency stimulation inhibits neurons, whereas high frequency stimulation excites them. Typically, 5 rTMS sessions are administered weekly for 4 to 6 weeks to treat depression. The most common adverse effect is headache, but seizures have been reported with very high frequency rTMS.18 Treatment-induced hypomania or mania is possible, but uncommon.19 Adverse cognitive effects also are uncommon; in fact, rTMS may improve cognition.20 Many short-term, placebo-controlled studies using sham rTMS have used high frequency rTMS focused on the left dorsolateral prefrontal cortex (DLPFC), which is hypofunctional in depressed patients.21,22 Controlled studies of rTMS have found it effective in major depressive disorder,23,24 including TRD.25 Low frequency rTMS focused on the right prefrontal cortex also is effective,26 although this approach is not as well studied as high-frequency left-DLPFC stimulation. Some studies have found rTMS effective for treating mania.27 A double-blind controlled trial of right vs left prefrontal cortex high frequency rTMS found greater antimanic effects with right-sided stimulation, which is complementary to the known antidepressant effects of left-sided stimulation.28 The differing effects of high-frequency and low-frequency, and right vs left rTMS are consistent with the known imbalance between right and left DLPFC in mood disorders.29

Based on the results of a large multi-center study,30 rTMS has been approved by the FDA for the treatment of depressed patients who have not responded to 1 anti-depressant trial. However, its efficacy for more refractory forms of depression may be relatively less compared with ECT.25,31,32

Unfortunately, the long-term efficacy, tolerability, and safety of rTMS for mood disorders has not been well studied. In a small study of 16 medication-free patients who had responded to acute treatment with rTMS, one-half the patients had a clinically significant benefit from repeated applications of rTMS during a 4-year follow-up.33 A recent 6-month naturalistic follow-up study suggested that the therapeutic effects of rTMS are durable.34 In this study, after an acute 6-week course of rTMS treatment, patients with partial responses were tapered off rTMS and started on maintenance medication. During the 6-month follow-up, 10 of 99 patients relapsed and 38 patients had symptomatic worsening. Of the 38 patients whose symptoms worsened, 32 improved when rTMS was added as an intermittent adjunctive rescue strategy to prevent relapse.

It is somewhat surprising that studies of rTMS for anxiety disorders have not demonstrated significant benefits.35 Other studies have been conducted using rTMS targeting temporoparietal cortex for the treatment of auditory hallucinations in schizophrenia, with some evidence of efficacy.36,37 Studies of motor cortex rTMS have shown some success in stroke patients.38-40 rTMS also may be effective for the treatment of tinnitus,41,42 pain,43 and migraine headache.44 In theory, cortical stimulation using rTMS may benefit patients with obesity, but it has not been studied for this purpose.45 Other forms of nonconvulsive electrical stimulation include cranial electro-therapy stimulation,46 transcranial direct current stimulation,47 and focal electrically administered therapy,48 but these have not been well studied.

Magnetic seizure therapy (MST)

Very high frequency intensities of rTMS to induce a seizure is called MST.49,50 Because the frequency of magnetic pulses can be manipulated to enhance inhibition or excitation in targeted brain regions, MST may produce controlled seizures in selected regions of the brain, such as the prefrontal cortex, reducing seizure spread to medial temporal structures and limiting cognitive side effects. Such magnetically induced focal seizures may substitute for ECT, with less risk of cognitive impairment. Similar to ECT, MST is administered 3 times a week under general anesthesia with muscle relaxation. Earplugs are required to protect hearing from the loud vibration caused by the magnetic coil. Standard EEG electrodes used to monitor seizures may become heated in a magnetic field and must be modified to avoid scalp burns.

Clinical experience with MST for the treatment of major depressive disorder is limited. A small study of 10 patients with major depression found that MST caused less severe muscle pain, fewer headaches, fewer subjective memory complaints, and better neuropsychological function than ECT.49,50 Current studies are investigating its antidepressant effects.

Vagus nerve stimulation (VNS)

The FDA approved VNS for refractory epilepsy in 1997 and for chronic TRD in 2005. The sympathetic and parasympathetic components of the autonomic nervous system (ANS) work together to regulate function of various organs, glands, and involuntary (smooth) muscles throughout the body, including vocalization, swallowing, heart rate, respiration, gastric secretion, and intestinal motility. The vagus nerve (cranial nerve X) has been considered primarily a cholinergic (containing the neurotransmitter acetylcholine) efferent nerve that serves as the main parasympathetic component of the ANS.51 However, the vagus actually is a mixed nerve composed of 20% efferent fibers that send signals from the brain to the body and 80% afferent or sensory fibers that carry information from the body to the brain. Therefore, contrary to tradition, the most important function of the vagus nerve is transmitting and/or mediating sensory information from throughout the body to the brain.52

The right and left vagus nerves exit from the brainstem and course through the neck in the carotid sheath between the carotid artery and jugular vein, upper chest along the trachea, lower chest and diaphragm along the esophagus, and into the abdominal cavity. Along the way, branches from the vagus enervate various body structures, including the larynx, pharynx, heart, lungs, and gastrointestinal tract. In the brainstem, the sensory afferent fibers of the vagus nerve terminate in the nucleus tractus solitarius, which then sends fibers that connect directly or indirectly to different brain regions. These regions include the dorsal raphe nuclei and the locus ceruleus—major serotonin-containing neurons and norepinephrine-containing neurons, respectively—as well as the amygdala, hypothalamus, thalamus, and orbitofrontal cortex—structures that make up the limbic system.

VNS generally refers to any technique that stimulates the vagus nerve. In the 1880s, clinicians noticed that manual massage and compression of the carotid artery in the neck could suppress seizures, an effect likely due to crude stimulation of the vagus.53 Subsequent animal studies found that VNS influenced brain electrical activity. Based on this work, further research found that VNS had anticonvulsant effects on experimentally-induced seizures in dogs.54 This led to clinical trials and subsequent approval of VNS for refractory epilepsy.55 Interestingly, various forms of paced breathing also can influence brain electrical activity.56,57 This effect may be mediated by vagus stimulation from the diaphragm. Therefore, cardiorespiratory stimulation of the vagus nerve that occurs during deep breathing, yoga, or aerobic exercise may explain some of their positive emotional and cognitive benefits.

Although the vagus nerve can be stimulated using a non-invasive transcutaneous method,58-60 the current clinical use of VNS involves surgical implantation of a pacemaker-like programmable pulse generator device (NCP System; Cyberonics, Inc.; Houston, TX).61 The generator is about 2 inches in diameter, about one-quarter inch thick, and weighs <1 ounce. The VNS implant surgery typically is performed under general anesthesia as an outpatient procedure. Two incisions are made: the left upper chest or left axillary border (where the generator is implanted subcutaneously) and the left neck area (where the electrode lead wire is attached to the left vagus nerve). The lead wire is passed through a subcutaneous tunnel and attached to the pulse generator. The generator is tested in the operating room and may be activated at that time, although usually it is activated about 2 weeks after the implant surgery. Surgical complications can include wound infection (<2%) and hoarseness (due to temporary or permanent left vocal cord paralysis) in about 1% of patients.

The “dose” of VNS is adjusted with a handheld computer that programs the stimulation parameters of the pulse generator; the manufacturer provides training. The settings are transmitted by a programming wand that is held on the skin directly over the implanted device. The programmable parameters and typical settings are:

  • the current charge (intensity of the electrical pulse or stimulus, measured in milliamperes [mA]), starting at 0.25 mA, average 1.0 mA, range 0.25 to 3.5 mA

  • the pulse width (the duration of each electrical pulse, measured in microseconds), average 500 microseconds, range 130 to 1000 microseconds

  • the frequency of electrical pulses (measured in hertz [Hz], average 20 Hz, range 1 to 30 Hz)

  • the on/off cycle (the stimulus on-time and off-time, measured in seconds or minutes); on-time, average 30 seconds, range 7 to 60 seconds; off-time average 5 minutes, range 12 seconds to 180 minutes.

These settings can each be adjusted to optimize efficacy and tolerability. Once the generator is activated, it will run continuously, stimulating the vagus nerve 24 hours a day. Patients can turn off VNS temporarily by holding a magnet over the device; it will start working again when the magnet is removed. VNS also can be turned on and off by the programmer. The battery life of the pulse generator ranges from 3 to 8 years, depending on the stimulus intensity, frequency, and on/off time. The pulse generator can be replaced or permanently removed in a similar surgical procedure. If replaced, it is attached to the same lead, which does not need to be replaced unless it breaks. VNS patients cannot undergo whole body MRI scans because the magnetic field may cause heating of the electrical lead, but MRI scans of the head can be done safely. Shortwave, microwave, or therapeutic ultrasound diathermy (equipment used for therapeutic tissue heating) should not be used in VNS patients, but diagnostic ultrasound is safe. Other electrical or electronic equipment such as metal detectors, microwave ovens, and cellular telephones will not affect VNS.

The adverse effects of VNS occur mostly during the “on” portion of the cycle and thus happen for very short intermittent periods. In theory, adverse effects could be related to stimulation of any body structure enervated by the vagus nerve but this is not the case for several reasons. Because approximately 80% of vagus nerve fibers are afferent, most of the electrical pulses applied to the nerve are propagated from the point of attachment toward the brain rather than the body. Also, the lead wire as it attaches to the vagus nerve is designed in such a way that the electrical pulses are directed one way, toward the brain, rather than bidirectional (up and down the nerve). Finally, the lead wire is attached to the upper cervical (neck) region of the vagus nerve. This attachment site is situated above most of the vagus branches that enervate various body structures. Therefore, VNS is not likely to affect the function of structures below this point. The most common adverse effects are hoarseness, cough, dyspnea, neck pain, dysphagia, and paresthesias in the neck. The VNS output current and/or other parameters can be adjusted to make these adverse effects tolerable in most patients over time. They disappear during VNS off-time and also when VNS is turned off temporarily with a magnet or by the programmer. Experience in epilepsy populations has shown that VNS is effective, safe, and well tolerated in patients age <18. There are no theoretical or known risks of VNS in pregnancy. VNS is safe and compatible with any psychotropic drug and ECT.

Studies of VNS for depression. The rationale for using VNS in depression is based on several converging lines of evidence.62 The vagus nerve has direct and indirect connections to limbic regions in the brain believed to be involved in mood regulation. Functional brain-imaging studies in humans confirm that VNS influences physiologic activity in these areas.55 Animal and human studies have shown that VNS stimulates dopamine, norepinephrine, serotonin, and other neurotransmitters relevant in mood disorders. In an animal model of depression, VNS has antidepressant effects similar to ECT and desipramine.63 Anticonvulsant drugs such as lamotrigine are effective antidepressants and VNS is a type of anti-convulsant therapy. Finally, VNS has positive effects on mood in epilepsy, even among patients whose seizures do not improve.

An open-label pilot study first investigated VNS in 60 patients with chronic TRD.64,65 These patients had uni-polar depression or bipolar disorder, were depressed for an average of nearly 10 years, and had not responded to an average of 16 different antidepressants (range: 3 to 44 therapies). Patients were maintained on their current medication. After implant, 2 weeks of postsurgery recovery without stimulation was followed by 10 weeks of VNS. Approximately 30% of patients were responders and 15% achieved full remission. Patients who had not responded previously to ECT were less likely to respond to VNS. Thirteen patients had not responded to >7 different drug treatments previously and none of them responded to VNS. Of the remaining patients who had not responded to <7 treatments, 39% responded to VNS. During long-term follow-up of 59 patients from this study, 44% were responders (27% remitters) at 1 year66 and 42% were responders (22% remitters) at 2 years.67 In general, patients with fewer previous unsuccessful treatments were more likely to respond or remit during long-term VNS. By 2 years, 2 patients had died (unrelated to VNS), 4 patients had dropped out, and 48 patients (81%) were still receiving VNS. This study demonstrated that the effectiveness of VNS increases with time and usually is maintained, which is contrary to the usual experience with pharmacotherapy in chronic TRD.68 Moreover, the finding that 81% of patients continued VNS suggested that it was well-accepted and that patients may have derived non-specific therapeutic benefits, even if they had not achieved a response or full remission based on depression rating scale assessments.67

A multi-center, randomized, double-blind, acute treatment study comparing active (device turned on) and sham (device turned off) VNS in 235 patients with chronic TRD also has been conducted.69 These patients had unipolar depression or bipolar disorder, and had not responded to 2 to 6 different antidepressant therapies. Patients were maintained on their current medication. After implant, 2 weeks of postsurgery recovery without stimulation was followed by 10 weeks of active or sham VNS. Response rates based on the Hamilton Rating Scale for Depression (HAMD-24) were not significantly different (15% response with active VNS vs 10% with sham VNS), but there was a significant difference based on a secondary measure, the Inventory of Depressive Symptomatology (Self-Report) (17% response active vs 7% response sham). Only 3 patients (1%) dropped out of the acute study because of adverse events.

At the end of the acute study, VNS was administered to all patients and 205 were followed in a long-term naturalistic treatment study.70 A similar cohort of 124 patients with chronic TRD who were receiving treatment as usual without VNS were recruited and followed long-term as a naturalistic comparison group.71 After 1 year, the VNS patients were significantly more likely to be improved (27% response; 16% remission) compared with the control group (13% response; 7% remission). This larger long-term study confirmed that the antidepressant response to VNS tends to increase over time and that the majority of patients maintain their response. Ninety percent of patients receiving VNS continued treatment during long-term follow-up. Only 3% dropped out because of adverse events and the rest because of lack of efficacy or other reasons. During 2 years of follow-up in this study, there was no difference in treatment outcomes for patients with bipolar vs uni-polar depression.72 Another open-label study of VNS in 11 patients with chronic TRD demonstrated a response rate of 55% and a remission rate of 27% after 1 year.73

In all of the depression studies, the relative safety, tolerability, and acceptability of VNS was good, and similar to the experiences of patients with epilepsy.65 The most common adverse effects were mild-to-moderate hoarseness, cough, dyspnea, neck pain, dysphagia, and paresthesias in the neck and often improved with time or by adjusting the VNS dose. Patients did not experience insomnia, sedation, sexual dysfunction, weight gain, and cognitive impairment often associated with antidepressants. The risk of developing hypomania or mania, which can happen with any type of antidepressant treatment, was low. Moreover, VNS did not increase the risk of suicidal thoughts, suicide attempts, or completed suicide. In fact, depression-related suicidal symptoms improved with time in most patients.

Although these studies demonstrate the effectiveness of VNS for chronic TRD, the absolute response and remission rates may seem relatively unimpressive. However, these results should be viewed in the larger context of treatment outcomes in other groups of depressed patients, especially those without TRD. For example, in the Sequenced Treatment Alternatives to Relieve Depression (STAR*D) trial74—the largest treatment study of patients with major depression ever conducted—the majority of patients had chronic or recurrent depression, but were not considered treatment-resistant and were less severely and chronically ill than patients enrolled in the VNS studies. During level 1 of STAR*D, patients were treated with citalopram for 12 weeks. Patients with an unsatisfactory response to citalopram at the end of level 1 could move to level 2 for 12 weeks. Level 2 involved either switching to bupropion, sertraline, or venlafaxine, or augmenting with a second medication by adding bupropion or buspirone to citalopram. Patients with an unsatisfactory response after level 2 could then move to level 3 for 12 weeks. Level 3 included switching to mirtazapine or nortriptyline or augmenting with a second medication by adding lithium or thyroid hormone to the level 2 medication. Patients with an unsatisfactory response after level 3 could then move to level 4 for 12 weeks. Level 4 included switching to tranylcypromine or to venlafaxine plus mirtazapine.

The level 1 response and remission rates were 47% and 28%, respectively, and the discontinuation rate due to adverse events was 26%.75 The level 2 response and remission rates for switching were 27% and 23%, respectively, with an adverse event discontinuation rate of 23%.76 The level 2 response and remission rates for augmentation were 29% and 30%, respectively, with an adverse event discontinuation rate of 17%.77 The overall level 3 response and remission rates for switching were 15% and 16%, respectively, with an adverse event discontinuation rate of 35%.78 The overall level 3 response and remission rates for augmentation were 20% and 20%, respectively, with an adverse event discontinuation rate of 16%.79 The overall level 4 response and remission rates were 17% and 10%, respectively, with an adverse event discontinuation rate of 32%.80 Hence, the efficacy and tolerability of VNS after 1 year of treatment in a group of very chronic and highly treatment-resistant patients compares favorably with pharmacotherapy outcomes after 4 levels (1 year) of vigorous STAR*D treatment in a group of less severely ill patients. Also note that VNS would be indicated for the type of patient that remain depressed after 4 treatment trials, such as those who completed treatment in STAR*D and were still depressed.

VNS for other neuropsychiatric disorders. Anticonvulsant drugs often are used to treat rapid-cycling bipolar disorder (RCBD), which is characterized by ≥4 mood episodes within 12 months. In a pilot study of 10 patients with RCBD, there were significant improvements in depressive and manic symptoms with VNS.81

The neurobiology of depressive and anxiety disorders overlap. Antidepressant and anticonvulsant therapies are effective for anxiety disorders. Studies in epilepsy patients have found that mood and anxiety symptoms often improved with VNS, and that this effect was independent of its anticonvulsant effects. In a pilot study of treatment-resistant anxiety disorders, 7 of 10 patients showed some benefit with VNS.82

In animals, VNS enhances learning and memory, perhaps mediated by its effects on the neurotransmitters acetylcholine and norepinephrine and its influence on relevant brain circuits. Human epilepsy and depression studies demonstrated that VNS did not have adverse cognitive effects, and cognitive function improved in some patients.83 In a pilot VNS-treatment study, 7 of 10 patients with Alzheimer’s disease showed clear evidence of cognitive improvement.84 After 1 year, 7 of 17 patients had improved cognition compared with baseline.85 The other 12 patients did not show significant decline in cognition compared with baseline. VNS was well tolerated in these elderly patients.

Antidepressant and anticonvulsant medications are effective for migraine and other headache disorders. In animal studies of pain, VNS has been shown to have analgesic effects.86 In 2 case series reports, describing experiences with 16 patients with chronic refractory headaches, VNS provided some benefit.87,88

The vagus nerve directly and indirectly regulates food intake and energy expenditure.89 One report in humans using VNS for epilepsy reported significant weight loss after 6 months of follow-up.90 Another report in 14 patients receiving VNS for chronic TRD also noted significant gradual weight loss during 2 years of follow-up.91 Animal studies have found that after feeding, feedback information from the stomach to the hypothalamus regulating satiety is mediated via the vagus nerve. Studies in dogs and pigs have found that VNS decreases food intake and causes weight loss.92 The methodology of VNS in these animal studies differs from the typical use of VNS (ie, left cervical VNS) in that 1 unilateral or 2 bilateral VNS pulse generators are implanted below the diaphragm; the generators are attached to 1 or both vagus nerve branches near the stomach. Unilateral and bilateral VNS has been associated with weight loss in these animals, although the weight loss is greater with bilateral VNS. A pilot study is currently in progress using subdiaphragm bilateral VNS in patients with morbid obesity.93 No findings have yet been reported.

Deep brain stimulation (DBS)

The most widely used neurosurgical form of therapeutic brain stimulation is DBS, during which electrodes are placed into deep subcortical regions of the brain.94 The particular electrode placement depends on the condition being treated. Currently, DBS that positions electrodes in various basal ganglia is FDA-approved for essential tremor95 and Parkinson’s disease.96 Forms of DBS also have been studied as investigational treatments for other movement disorders,97 epilepsy,98 chronic pain,99,100 obesity,101 and psychiatric disorders.3 Mood changes have long been reported in studies of DBS for movement disorders.102 These observations in part contributed to interest in the potential use of DBS for psychiatric disorders.3 Studies of DBS for TROCD have involved electrode placement in the region of the anterior limb of the internal capsule.103,104 Several different research groups have shown this to be effective.105-107 Recently, the FDA approved a DBS system for TROCD (Reclaim DBS Therapy; Medtronic; Minneapolis, MN). Recent studies of DBS for TRD have focused on 2 brain regions: subgenual anterior cingulate (Brodmann Area 25; Cg25) and ventral capsule/ventral striatum (VC/VS).

In depression, Cg25 is relatively overactive.108 Twenty patients with TRD were enrolled in a pilot study using Cg25 DBS.109,110 On average, they had been in a current depressive episode for 6 years and had failed at least 4 different antidepressant treatments. Seventeen patients had received ECT previously. One month after surgery, 35% of patients had responded and 10% went into remission. Six months after surgery, 60% of patients were responders and 35% were in remission. These benefits largely were maintained at 12 months. The number of serious adverse effects was small with no patient experiencing permanent deficits. Three patients had the device removed because of infection, but 1 later had the device re-implanted. One perioperative seizure occurred. Transient adverse mood changes were noted in several patients. There were no adverse cognitive effects. Preliminary brain imaging studies showed that the baseline findings of increased Cg25 metabolism and decreased DLPFC metabolism tended to normalize over time with continued DBS.

Another research group has focused on VC/VS DBS in patients with TRD. This region is close to where DBS for TROCD has been targeted. Studies of DBS for TROCD found that comorbid depression often improved. The VC/VS region is relatively underactive in depression.4 In a pilot study using VC/VS DBS for TRD, 5 patients were implanted.111 By 3 months, 3 patients were responders and 2 patients were partial responders. In a subsequent report of 15 patients from this study followed from 6 months up to 4 years, the 6-month response rate was 47% and the remission rate was 27%.112 At the last follow-up visit (mean of 24 months), the response rate was 53% and the remission rate was 33%. The patients in this study had been in a current episode of depression for at least 2 years and had failed approximately 12 different antidepressant treatments. All patients had received ECT and psychotherapy previously. The surgical procedure and stimulation with DBS was relatively well-tolerated in these patients. Transient adverse mood changes were reported, but there were no adverse cognitive effects, seizures, infections, or other serious adverse events.

Another research group has focused on the nucleus accumbens in patients with TRD. This region is relatively underactive in depression. In a pilot study using nucleus accumbens DBS for TRD, 10 patients received the implant.113 By 12 months, the response rate was 50% and the remission rate was 30%. The patients had been in a current episode of depression for an average of 11 years and had failed 8 rigorous antidepressant treatments. All patients had received ECT and psychotherapy previously. The surgical procedure and stimulation with DBS was relatively well-tolerated. Transient adverse mood changes were reported, but there were no permanent or serious adverse effects.

At least 5 different brain regions have been identified as potential targets for DBS in TRD.114 Based on the results from the 2 largest studies conducted to date, larger separate pivotal studies of Cg25 DBS for TRD110 and of VC/VS DBS for TRD112 are being initiated (ClinicalTrials.gov). These 2 clinical research efforts should provide more definitive data about the safety and efficacy of DBS for TRD.

Cortical brain stimulation (CBS)

CBS, an investigational procedure, involves placement of stimulation electrodes onto the cortical surface of the brain. The exact location depends on the condition being treated and have included placement over the temporoparietal cortex for severe tinnitus,115 over the motor cortex for stroke,116 and over the left DLPFC for TRD.117-119 In theory, cortical stimulation using CBS could be investigated for obesity, but this has not yet been studied.45 The rationale for using CBS in TRD is based in part on studies using high-frequency rTMS directed at the left DLPFC, which is underactive in depression. rTMS requires multiple weekly sessions and may require continual application over a longer time period because the acute antidepressant effects may be temporary. A pacemaker-driven surgically-implanted electrode array located over the left DLPFC might therefore provide more practical long-term benefits.

As a potential long-term adjunctive therapy for patients with TRD, 12 outpatients with severe chronic TRD were enrolled in a feasibility study to evaluate the safety and preliminary efficacy of CBS targeting the left DLPFC.119 After an observation phase of at least 8 weeks, during which the patients were taking stable doses of their current psychotropic medications, the patients were surgically implanted with an epidural cortical stimulation system (Renova DT, Northstar Neuroscience; Seattle, WA). Patients were randomized to single-blind active (device on) or sham (device off) stimulation for 8 weeks; thereafter active stimulation was provided for all patients. Psychotropic medications were not changed unless clinically indicated. Safety outcomes were assessed. Efficacy assessment included Hamilton Depression Rating Scale (HDRS), Montgomery-Åsberg Depression Rating Scale (MADRS), Global Assessment of Function (GAF), and Quality of Life Enjoyment and Satisfaction Questionnaire (QLESQ).

Preliminary data are available for the 8-week primary endpoint visit, for the 16-week visit, and then after an average of 52 weeks of follow-up.118 On average, patients had suffered from major depressive disorder for 27 years, the duration of their current major depressive episode was 7 years, and they failed an average of 10 different antidepressant treatments. Ten patients had previously received ECT, averaging 16 treatments. At baseline, their mean HDRS score was 35, mean MADRS score was 33, mean GAF score was 42, and mean QLESQ score was 41. At the 8-week primary endpoint, the HDRS decreased by 22% for active stimulation (n=6) vs 3% for sham stimulation (n=6); the MADRS decreased by 22% for active vs 8% for sham; the GAF increased by 23% (active) vs 12% (sham); and the QLESQ increased by 28% (active) vs 4% (sham). Results were not statistically significant. With active stimulation for all patients from week-8 to week-16, mean change scores continued to improve: 21% to 26% for HDRS, 22% to 32% for MADRS, and 25% to 46% for GAF. After an average follow-up of 52 weeks postsurgery, the HDRS had improved by 32%, the MADRS had improved by 33%, and the GAF had improved by 52% (all changes compared with pretreatment baseline). Preliminary results of brain imaging studies showed that the magnitude of decreased metabolism of the left DLPFC correlated with the magnitude of improvement with CBS.117 Therefore, greater degrees of abnormal DLPFC function predicted relatively better improvement with CBS. CBS was well tolerated. There have not been any significant surgical or stimulation-related adverse events, including adverse cognitive effects. Although not statistically significant because of the small sample size, these preliminary results demonstrate that CBS is safe and may have a positive effect that increases over time.

In contrast to stimulation of the left DLPFC, another small pilot study investigated the effects of CBS directed bilaterally over the anterior and medial prefrontal cortex in 5 patients with TRD.120 The duration of the current episode of depression in these patients ranged from 8 to 84 months, and they failed an average of 6 different anti-depressant treatments. Three patients had previously received ECT, 3 VNS, and 3 rTMS. At 7-month follow-up, the patients had an average improvement on the HDRS of 55%. Three patients achieved remission and therapy was well tolerated.

Clinical procedures with DBS and CBS

Treatment with DBS and CBS involves the surgical implantation of pacemaker-like programmable pulse generator devices similar to those used for VNS. For DBS, 2 subcutaneous generators are implanted bilaterally in the left and right upper chest under general anesthesia. Under local anesthesia, 2 bilateral craniotomies are performed near the crown of the skull. A stereo-tactic frame and MRI scanning guide placement of the electrodes deep into targeted subcortical areas in each hemisphere. The precise placement depends on the condition being treated.

For depression, intraoperative DBS testing is conducted while the patient is awake to assess the acute mood effects and electrode placement. For CBS, 2 incisions are made under general anesthesia: the left upper chest, where the generator is implanted subcutaneously, and a craniotomy over the left DLPFC, where the lead wire enters and is connected to the electrode grid sutured on the epidural surface of the DLPFC. No intraoperative testing is conducted with CBS. For both DBS and CBS, the lead wire(s) are passed through a tunnel under the skin and attached to the generator(s). Therapeutic stimulation typically starts about 1 week after implant surgery for CBS and about 4 weeks postsurgery for DBS. Seizures, bleeding, and infection are possible complications,121,122 but are more likely with DBS compared with CBS. Possible stimulation-related adverse effects of DBS include paresthesias, muscle contraction, dysarthria, and diplopia. Adverse mood effects, including hypo-mania, are usually transient and respond to stimulation adjustments. Adverse memory and cognitive effects are possible, but significant or persistent deficits are uncommon.123,124 Adverse cognitive effects were not seen in the small number of patients receiving CBS. Similar to rTMS, CBS of the left DLPFC theoretically may improve cognitive functioning.20

For DBS and CBS, stimulation parameters are adjusted using a handheld computer that programs the generator via a programming wand held on the skin over the device. The programmable parameters are the current charge (intensity of the electrical pulse or stimulus), the pulse width (the duration of each electrical pulse), the frequency of electrical pulses, and the on/ off duty cycle (the stimulus on-time and off-time). During clinical follow-up, the stimulation settings can be adjusted to optimize efficacy and tolerability. Also, different stimulation parameters can be utilized to cause excitation or inhibition of brain function in the region of interest. For example, the intent of DLPFC CBS is to increase activity of an area that is underactive in depression, whereas the intent of Cg25 DBS is to inhibit activity of an area that is overactive.

The generator operates continuously, according to the programmed parameters, 24 hours a day. Patients can turn it off temporarily by holding a magnet over the device; it will start working again by removing the magnet. Depending on the stimulation parameters, the expected battery life of CBS and DBS generators ranges from 1 to 3 years. This is considerably shorter than for VNS because they have higher energy demands than VNS stimulation. Generators can be replaced or permanently removed in a simple surgical procedure. If replaced, it is attached to the same lead, which does not need to be replaced unless it breaks. Because magnetic fields may cause heating of the electrical lead, MRI scans cannot be performed unless special shielding techniques are used. Shortwave, microwave, or therapeutic ultrasound diathermy used for therapeutic tissue heating cannot be used, but diagnostic ultrasound is safe. Metal detectors, microwave ovens, cellular telephones, and other electrical or electronic equipment will not affect the generators.

DBS and CBS are potentially viable approaches for treatment of TRD. In contrast to classic “psychosurgery,” these therapeutic brain stimulation techniques are appealing because the stimulation is non-ablative and can be modified or discontinued depending on the clinical response.

Quantitative electroencephalography (QEEG)

Electroencephalography (EEG) is an old technology used to measure brain waves and diagnose epilepsy. The EEG measures spontaneous brain electrical activity. Typically 19 to 36 electrodes are applied over the scalp, corresponding to frontal, parietal, occipital, and temporal lobes of the brain.125 Each electrode records the activity of the brain cortex underlying it. EEG recordings appear as rhythmic waveforms (“brain waves”) characterized by their frequency, measured in Hz, and their amplitude, measured as voltage. These characteristics also reflect the “power” or “energy” of the electrical activity. Slow frequency and/ or low amplitude EEG waves have less “power” than fast frequency and/or high amplitude waves. Four dominant EEG waveforms are observed: alpha (8 to 13 Hz), beta (13 to 30 Hz), delta (0.5 to 4 Hz), and theta (4 to 7 Hz) frequency bands. Alpha waves occur during relaxed wakefulness, with the eyes closed, over parietal and occipital lobes. They diminish when the eyes are open or the patient is alert. Beta waves have the fastest frequency, are seen over frontal regions and other regions during intense mental activity, and have the smallest amplitude of all EEG waves. Sedative-hypnotic drugs cause high amplitude beta waves. Delta and theta waves (slow-wave EEG) have the highest amplitude of any waves. They are normally associated with sleep, but suggest brain dysfunction when observed while awake. EEG waveforms and their power can be quantitatively analyzed precisely and reliably using computers, referred to as QEEG.

Because electrical activity represents the greatest demand on cerebral metabolism, QEEG can be used to estimate cerebral metabolism and/or perfusion (blood flow).126 From QEEG recordings, absolute power (energy in a particular EEG band) and relative power (proportion of total energy in a given band, measured as a percentage of total power) can be combined into a single measure referred to as cordance.127 Cordance can be determined at any electrode site for any frequency band (alpha, beta, delta, or theta). Cordance values range from negative to positive and these readings have been validated as a measure of cerebral perfusion and metabolism.128 Electrodes overlying cortical areas with high metabolism have higher cordance values, whereas those overlying cortex with low metabolism have lower cordance values. Areas of brain dysfunction are commonly associated with decreased metabolism/perfusion. These areas often have low QEEG power, and the low QEEG power typically is concentrated in pathologic slow-wave bands (eg, theta).

Abnormal metabolism/perfusion in the prefrontal and/or anterior cingulate cortices is a consistent finding from brain imaging studies in depression.129 These areas are closely associated parts of the limbic system, which is involved in emotional regulation. Imaging studies during treatment have reported differences in prefrontal/ cingulate activity between antidepressant responders and nonresponders. It would be logical to expect that improvement in depression would be associated with normalization of cerebral metabolism/perfusion. Curiously, many studies have found that effective treatment is associated with further reductions in cerebral metabolism/perfusion.126 This seemingly paradoxical finding likely reflects our incomplete understanding of how different interconnected neural circuits in the brain function normally, abnormally, and under the influence of various pharmacologic interventions. QEEG studies have found that theta activity—more so than other frequency bands—characterizes prefrontal/anterior cingulate function in depression.130,131 Of particular interest, theta activity changes in response to antidepressant treatments.132-134

QEEG and antidepressant drug response. Finding an effective antidepressant for a patient is a trial-and-error process. Improvement generally requires 4 to 12 weeks and patients might stop taking medication too soon because of delayed response or intolerable side effects. Less than one-half of patients achieve full remission with the first drug tried. Predicting how a patient might respond to treatment would help identify an effective treatment more quickly, avoid unnecessary medication changes, and encourage adherence.

Two randomized, double-blind, placebo-controlled studies have investigated the use of QEEG theta cordance for predicting treatment response in major depression.135 Patients first entered a 1-week, single-blind, placebo lead-in phase. Those still depressed were then randomized to fluoxetine vs placebo or to venlafaxine vs placebo for 8 weeks. QEEG was done at baseline and after 48 hours and 1 week on drug or placebo. Baseline and change from baseline QEEG values were examined in different brain regions among 4 groups of patients (drug and placebo responders and nonresponders). No regional QEEG differences were found among the groups at baseline. Drug responders uniquely showed significant decreases in prefrontal theta cordance at 48 hours and 1 week, even though clinical differences did not emerge until after 4 weeks of treatment. Patients with greater changes in cor-dance had the most complete response after 8 weeks. The cordance change was specific to the prefrontal region. By contrast, placebo responders showed a significant increase in prefrontal theta cordance starting early in treatment.136 Drug nonresponders and placebo nonresponders did not exhibit any changes in cordance. Therefore, early QEEG cordance changes (decreases within 1 week) during a placebo lead-in specifically predicted the eventual medication treatment outcome.137 Although one might expect that improvement in depression with drug therapy would be associated with increased cordance (akin to normalization of cerebral metabolism/perfusion), these results are consistent with brain imaging studies and treatment outcome discussed above.126 That effective placebo treatment is associated with changes in brain function that are distinct from those associated with effective drug treatment is a very interesting finding that has obvious clinical and research value and warrants further study.

In an analysis of data from the venlafaxine vs placebo study, decreases in prefrontal cordance at the end of the initial 1-week placebo lead-in phase were significantly associated with the later development of side effects in patients taking venlafaxine but not in those taking placebo.138 Cordance changes during placebo lead-in were not related to side effects at lead-in. Changes in prefrontal brain function occurring at the end of placebo lead-in (before administration of drug) predicted the development of side effects in patients who later took venlafaxine. Prefrontal brain function changes during the placebo lead-in may therefore indicate a susceptibility to developing drug side effects.

In another study, depressed patients who had not responded to an initial antidepressant were switched to an alternative antidepressant.139 QEEG was done before starting the new treatment and after 1 week. Patients showing an early cordance decrease were more likely to respond to their new treatment. Similarly, early decreases in prefrontal cordance predicted later clinical improvement with the norepinephrine reuptake inhibitor reboxetine.140 All together, these studies demonstrate the ability of early prefrontal QEEG cordance changes to predict the response to different antidepressants, including serotonergic (eg, fluoxetine), noradrenergic (eg, reboxetine), and mixed serotonergic/noradrenergic (eg, venlafaxine) drugs.

Using QEEG requires a skilled technician to correctly apply a large number of electrodes over the entire scalp, which is impractical in routine clinical practice. Ideally, a more feasible method would require only a small number of electrodes. In addition, a clinically useful predictor of treatment outcome should be able to clearly distinguish responders from nonresponders within the first week and also predict serious adverse events.

To detect prefrontal/anterior cingulate region signals, 4 channels of EEG from the frontal regions is as accurate as a larger number of electrodes.141 Two pilot studies have investigated a simple-to-use 4-channel frontal EEG monitor to predict antidepressant response in depression.142,143 These studies used a proprietary QEEG measure (the Aspect Antidepressant Response Index [AARI]), developed by Aspect Medical Systems (Newton, MA), to predict treatment outcome. The AARI is a composite index of theta and alpha activity derived from the QEEG. The AARI ranges from 0 (low probability of response) to 100 (high probability of response). In both studies, patients were treated for 8 weeks and the QEEG was done at baseline and at weeks 1, 4, and 8. Iosifescu et al142 found that the AARI at week 1 was 73% accurate in predicting response at week 8. Poland et al143 found that the AARI at week 1 was 81% accurate in predicting response at week 8. In both studies, dose changes were allowed during treatment and the AARI predictive accuracy was best in patients with no dose change after week 1. Assessing the AARI after dose changes might enhance its predictive accuracy. The 4-channel frontal QEEG measure also has been found to predict treatment-emergent increased suicidal ideation before medication was started.144 These preliminary studies led to a larger multi-center study investigating the accuracy of the 4-channel QEEG and an updated version of the AARI in predicting treatment outcome and adverse effects in a randomized study comparing escitalopram, bupropion, and their combination.

In this multi-center study,145 the investigators examined a frontal QEEG biomarker, the Antidepressant Treatment Response (ATR) index, as a predictor of response to escitalopram and compared the ATR with other putative predictors; 375 patients with major depression had a baseline QEEG study. After 1 week of treatment with escitalopram (10 mg/d) a second QEEG was performed, and the ATR was calculated. Patients then were randomly assigned to continue escitalopram (10 mg/d) or change to alternative treatments. Seventy-three patients received escitalopram for a total of 49 days. Response and remission rates were 52.1% and 38.4%, respectively. The ATR predicted both response and remission with 74% accuracy. Neither serum drug levels nor 5HTTLPR and 5HT2a genetic polymorphisms were significant predictors. Responders had larger decreases in HAMD-17 scores at day 7, but remitters did not. Clinician prediction based upon global impression of improvement at day 7 did not predict outcome. Logistic regression showed that the ATR and early HAMD-17 changes were additive predictors of response, but the ATR was the only significant predictor of remission.

In another multi-center study,146 investigators also examined the ATR index as a predictor of differential response and remission to escitalopram, bupropion, or a combination of the 2 medications in patients with major depression; 375 patients had a baseline QEEG study preceding 1 week of treatment with escitalopram (10 mg/d), after which a second QEEG was performed and the ATR index was calculated. Patients then were randomized to continue escitalopram, switch to bupropion, or receive a combination of the 2. Clinical response was assessed using the HAMD-17 at 49 days of treatment. Accuracy of ATR in predicting response and remission was calculated. There were no significant differences between response and remission rates in the 3 treatment groups. A single ATR threshold was useful for predicting differential response to either escitalopram or bupropion monotherapy. Patients with ATR values above the threshold were >2.4 times as likely to respond to escitalopram as those with low ATR values (68% vs 28%). Patients with ATR values below the threshold who were switched to bupropion were 1.9 times as likely to respond to bupropion alone as those who remained on escitalopram (53% vs 28%). The ATR index did not predict response to combination treatment, but it may prove useful in predicting responsiveness to different individual antidepressant medications.


Many evidence-based pharmacotherapies and psychotherapies are available for depression.147 A significant minority of patients do not experience a satisfactory response to sequential trials of these therapies, including various drug-drug and drug-psychotherapy combinations.148 Can the selection of treatments for particular patients be improved and lead to better treatment outcomes? QEEG is 1 novel and promising technology approach for predicting antidepressant treatment outcome, including possible adverse effects, but needs further study for predicting treatment outcomes with psychotherapy and neurostimulation therapies.

What should be done for patients failing to respond adequately to multiple treatment trials? ECT often is considered the treatment of choice for TRD. For some patients, however, ECT may not be efficacious, well tolerated, or acceptable, or it may be contraindicated. Novel alternative approaches for TRD, which are based on advances in understanding the functional neuroanatomy of mood disorders, include various types of alternative neurostimulation therapies and neurosurgical procedures. Psychosurgery usually is considered a treatment of last resort for patients with chronic, severe, and debilitating treatment-resistant psychiatric disorders. Although ablative procedures may have a role for conditions such as TRD and TROCD,14-16 neurosurgical approaches that use implanted technologies to provide therapeutic brain stimulation are potentially more practical and acceptable because the procedure is nonablative and the stimulation can be modified or discontinued depending on the clinical response.

If ECT is the gold standard treatment for TRD, then what are the alternatives when it is ineffective, intolerable, unacceptable, or contraindicated? VNS is an FDA-approved invasive neurostimulation therapy for TRD. Other investigational invasive neurostimulation therapies include: 1) CBS directed at the left DLPFC or the medial PFC and 2) DBS directed at area Cg25 or the VC/ VS region. Potential noninvasive neurostimulation therapies include: 1) rTMS (currently FDA-approved for 1 antidepressant nonresponders); 2) MST; 3) transcutaneous VNS; 4) transcranial direct current stimulation; and 5) focal electrically administered therapy. Neurosurgical ablative therapy procedures for TRD include: 1) anterior cingulotomy; 2) anterior capsulotomy; 3) subcaudate tractotomy; and 4) limbic leucotomy.

Among the invasive neurostimulation therapies, VNS often is positioned as a treatment for ECT-intolerant or ECT-resistant depression. Other invasive therapies, such as CBS or DBS, might be considered for VNS-resistant depression. Neurosurgical ablative therapy procedures would then be considered as a treatment of last resort. Where would other noninvasive neuro-stimulation therapies (such as rTMS or MST) fall within a treatment algorithm for TRD? What is the most appropriate sequence of various neurostimulation therapies to consider for patients with TRD? Should VNS necessarily be positioned after ECT, or could it be considered a viable alternative to ECT? If sufficient data demonstrate the efficacy of CBS or DBS for TRD should they necessarily be reserved for ECT-resistant and VNS-resistant patients, or could they be considered earlier in a treatment algorithm? Are there clinical or neurobiologic features that characterize specific TRD subgroups, which might be used to identify patients who will respond preferentially to a particular neurostimulation therapy? Currently, an important clinical and research problem with neurostimulation therapies is the relative dearth of cost effectiveness data (ie, efficacy, tolerability, safety, acceptability, and cost), especially compared with each other and to ECT. Randomized controlled trials (RCTs) are ideal, but are difficult to design and carry out for these types of interventions. Designing RCTs would also depend on having sufficient effectiveness data for individual therapies to consider which therapies are most important to compare. Systematically collecting data and developing a database on treatment outcomes for these patients would be important.

Another issue to consider is whether any of these neurostimulation therapies change or influence brain function in such a way that there may be an enhanced responsiveness to the subsequent use of pharmacotherapy or psychotherapy. For example, would enhanced DLPFC function through the use of rTMS or CBS result in better cognitive functioning and enhanced psychotherapy effects? Would normalization of Cg25 activity through DBS facilitate the effects of an antidepressant? A related issue is whether there is a rationale for combining any of these neurostimulation therapies. For example, VNS has modest efficacy and a delayed onset of effect. Would CBS accelerate or increase the response to VNS? Would combining CBS and DBS, targeting different regions make sense? Would VNS or other neuro-stimulation therapies be a more effective option for long-term prophylactic treatment among ECT responders than either long-term pharmacotherapy or maintenance ECT?

DISCLOSURES: The authors have received research grant support from Cyberonics and Medtronic.


  1. Nestler EJ, Barrot M, DiLeone RJ, et al. Neurobiology of depression. Neuron. 2002;34:13–25.
  2. Ressler KJ, Mayberg HS. Targeting abnormal neural circuits in mood and anxiety disorders: from the laboratory to the clinic. Nat Neurosci. 2007;10:1116–1124.
  3. Kopell BH, Greenberg B, Rezai AR. Deep brain stimulation for psychiatric disorders. J Clin Neurophysiol. 2004;21:51–67.
  4. Surguladze S, Keedwell P, Phillips M. Neural systems underlying affective disorders. Advances in Psychiatric Treatment. 2003;9:446–455.
  5. Chen CH, Ridler K, Suckling J, et al. Brain imaging correlates of depressive symptom severity and predictors of symptom improvement after antidepressant treatment. Biol Psychiatry. 2007;62:407–414.
  6. Mayberg HS. Modulating dysfunctional limbic-cortical circuits in depression: towards development of brain-based algorithms for diagnosis and optimised treatment. Br Med Bull. 2003;65:193–207.
  7. Brody AL, Saxena S, Stoessel P, et al. Regional brain metabolic changes in patients with major depression treated with either paroxetine or interpersonal therapy: preliminary findings. Arch Gen Psychiatry. 2001;58:631–640.
  8. Goldapple K, Segal Z, Garson C, et al. Modulation of cortical-limbic pathways in major depression: treatment-specific effects of cognitive behavior therapy. Arch Gen Psychiatry. 2004;61:34–41.
  9. Benedetti F, Mayberg HS, Wager TD, et al. Neurobiological mechanisms of the placebo effect. J Neurosci. 2005;25:10390–10402.
  10. Mayberg HS, Silva JA, Brannan SK, et al. The functional neuroanatomy of the placebo effect. Am J Psychiatry. 2002;159:728–737.
  11. Pressman JD. Last resort: psychosurgery and the limits of medicine. Cambridge, United Kingdom: Cambridge University Press; 1998.
  12. Valenstein ES. Great and desperate cures: the rise and decline of psychosurgery and other radical treatments for mental illness. New York, NY: Basic Books; 1986.
  13. Binder DK, Iskandar BJ. Modern neurosurgery for psychiatric disorders. Neurosurgery. 2000;47:9–23.
  14. Eljamel MS. Ablative neurosurgery for mental disorders: is there still a role in the 21st century? A personal perspective. Neurosurg Focus. 2008;25:E4.
  15. Steele JD, Christmas D, Eljamel MS, et al. Anterior cingulotomy for major depression: Clinical outcome and relationship to lesion characteristics. Biol Psychiatry. 2008;63:670–677.
  16. Greenberg BD, Price LH, Rauch SL, et al. Neurosurgery for intractable obsessive-compulsive disorder and depression: critical issues. Neurosurg Clin N Am. 2003;14:199–212.
  17. Cowey A. The Ferrier Lecture 2004: what can transcranial magnetic stimulation tell us about how the brain works? Philos Trans R Soc Lond B Biol Sci. 2005;360:1185–1205.
  18. Loo CK, McFarquhar TF, Mitchell PB. A review of the safety of repetitive transcranial magnetic stimulation as a clinical treatment for depression. Int J Neuropsychopharmacol. 2008;11:131–147.
  19. Xia G, Gajwani P, Muzina DJ, et al. Treatment-emergent mania in unipolar and bipolar depression: focus on repetitive transcranial magnetic stimulation. Int J Neuropsychopharmacol. 2008;11:119–130.
  20. Vanderhasselt MA, De Raedt R, Baeken C, et al. The influence of rTMS over the left dorsolateral prefrontal cortex on Stroop task performance. Exp Brain Res. 2006;169:279–282.
  21. Kozel AF, George MS. Meta-analysis of left prefrontal repetitive transcranial magnetic stimulation (rTMS) to treat depression. J Psychiatr Pract. 2002;8:270–275.
  22. Rogers MA, Kasai K, Koji M, et al. Executive and prefrontal dysfunction in unipolar depression: a review of neuropsychological and imaging evidence. Neurosci Res. 2004;50:1–11.
  23. George MS, Lisanby SH, Avery D, et al. Daily left prefrontal transcranial magnetic stimulation therapy for major depressive disorder: a sham-controlled randomized trial. Arch Gen Psychiatry. 2010;67:507–516.
  24. Gross M, Nakamura L, Pascual-Leone A, et al. Has repetitive transcranial magnetic stimulation (rTMS) treatment for depression improved? A systematic review and meta-analysis comparing the recent vs the earlier rTMS studies. Acta Psychiatr Scand. 2007;116:165–173.
  25. Lam RW, Chan P, Wilkins-Ho M, et al. Repetitive transcranial magnetic stimulation for treatment-resistant depression: a systematic review and meta-analysis. Can J Psychiatry. 2008;53:621–631.
  26. Fitzgerald PB, Hoy K, McQueen S, et al. Priming stimulation enhances the effectiveness of low-frequency right prefrontal cortex transcranial magnetic stimulation in major depression. J Clin Psychopharmacol. 2008;28:52–58.
  27. Michael N, Erfurth A. Treatment of bipolar mania with right prefrontal rapid transcranial magnetic stimulation. J Affect Disord. 2004;78:253–257.
  28. Grisaru N, Chudakov B, Yaroslavsky Y, et al. Transcranial magnetic stimulation in mania: a controlled study. Am J Psychiatry. 1998;155:1608–1610.
  29. Grimm S, Beck J, Schuepbach D, Hell D, et al. Imbalance between left and right dorsolateral prefrontal cortex in major depression is linked to negative emotional judgment: an fMRI study in severe major depressive disorder. Biol Psychiatry. 2008;63:369–376.
  30. O’Reardon JP, Solvason HB, Janicak PG, et al. Efficacy and safety of transcranial magnetic stimulation in the acute treatment of major depression: a multisite randomized controlled trial. Biol Psychiatry. 2007;62:1208–1216.
  31. Eranti S, Mogg A, Pluck G, et al. A randomized controlled trial with 6-month follow-up of repetitive transcranial magnetic stimulation and electroconvulsive therapy for severe depression. Am J Psychiatry. 2007;164:73–81.
  32. McLoughlin DM, Mogg A, Eranti S, et al. The clinical effectiveness and cost of repetitive transcranial magnetic stimulation versus electroconvulsive therapy in severe depression: a multicentre pragmatic randomised controlled trial and economic analysis. Health Technol Assess. 2007;11:1–54.
  33. Demirtas-Tatlidede A, Mechanic-Hamilton D, Press DZ, et al. An open-label prospective study of repetitive transcranial magnetic stimulation (rTMS) in the long-term treatment of refractory depression: reproducibility and duration of the antidepressant effect in medication-free patients. J Clin Psychiatry. 2008;69:930–934.
  34. Janicak PG, Nahas Z, Lisanby SH, et al. Durability of clinical benefit with transcranial magnetic stimulation (TMS) in the treatment of pharmacoresistant major depression: assessment of relapse during a 6-month multisite open-label study. Brain Stimul. 2010;3:187–199.
  35. Pigot M, Loo C, Sachdev P. Repetitive transcranial magnetic stimulation as treatment for anxiety disorders. Expert Rev Neurother. 2008;8:1449–1455.
  36. Aleman A, Sommer IE, Kahn RS. Efficacy of slow repetitive transcranial magnetic stimulation in the treatment of resistant auditory hallucinations in schizophrenia: a meta-analysis. J Clin Psychiatry. 2007;68:416–421.
  37. Tranulis C, Sepehry AA, Galinowski A, et al. Should we treat auditory hallucinations with repetitive transcranial magnetic stimulation? A metaanalysis. Can J Psychiatry. 2008;53:577–586.
  38. Málly J, Dinya E. Recovery of motor disability and spasticity in post-stroke after repetitive transcranial magnetic stimulation (rTMS). Brain Res Bull. 2008;76:388–395.
  39. Mansur CG, Fregni F, Boggio PS, et al. A sham stimulation-controlled trial of rTMS of the unaffected hemisphere in stroke patients. Neurology. 2005;64:1802–1804.
  40. Nowak DA, Grefkes C, Dafotakis M, et al. Effects of low-frequency repetitive transcranial magnetic stimulation of the contralesional primary motor cortex on movement kinematics and neural activity in subcortical stroke. Arch Neurol. 2008;65:741–747.
  41. Plewnia C, Reimold M, Najib A, et al. Moderate therapeutic efficacy of positron emission tomography-navigated repetitive transcranial magnetic stimulation for chronic tinnitus: a randomised controlled pilot study. J Neurol Neurosurg Psychiatry. 2007;78:152–156.
  42. Pridmore S, Kleinjung T, Langguth B, et al. Transcranial magnetic stimulation: potential treatment for tinnitus? Psychiatry Clin Neurosci. 2006;60:133–138.
  43. Leo RJ, Latif T. Repetitive transcranial magnetic stimulation (rTMS) in experimentally induced and chronic neuropathic pain: a review. J Pain. 2007;8: 453–459.
  44. Brighina F, Piazza A, Vitello G, et al. rTMS of the prefrontal cortex in the treatment of chronic migraine: a pilot study. J Neurol Sci. 2004;227:67–71.
  45. Alonso-Alonso M, Pascual-Leone A. The right brain hypothesis for obesity JAMA. 2007;297:1819–1822.
  46. Bystritsky A, Kerwin L, Feusner J. A pilot study of cranial electrotherapy stimulation for generalized anxiety disorder. J Clin Psychiatry. 2008;69:412–417.
  47. Boggio PS, Rigonatti SP, Ribeiro RB, et al. A randomized, double-blind clinical trial on the efficacy of cortical direct current stimulation for the treatment of major depression. Int J Neuropsychopharmacol. 2008;11:249–254.
  48. George MS, Nahas Z, Borckardt JJ, et al. Brain stimulation for the treatment of psychiatric disorders. Curr Opin Psychiatry. 2007;20:250–254.
  49. Lisanby SH, Luber B, Schlaepfer TE, et al. Safety and feasibility of magnetic seizure therapy (MST) in major depression: randomized within-subject comparison with electroconvulsive therapy. Neuropsychopharmacology. 2003;28:1852–1865.
  50. Spellman T, McClintock SM, Terrace H, et al. Differential effects of high-dose magnetic seizure therapy and electroconvulsive shock on cognitive function. Biol Psychiatry. 2008;63:1163–1170.
  51. George MS, Sackeim HA, Rush AJ, et al. Vagus nerve stimulation: a new tool for brain research and therapy. Biol Psychiatry. 2000;47:287–295.
  52. Zagon A. Does the vagus nerve mediate the sixth sense? Trends Neurosci. 2001;24:671–673.
  53. Lanska DJ. J.L. Corning and vagal nerve stimulation for seizures in the 1880s. Neurology. 2002;58:452–459.
  54. Zabara J. Inhibition of experimental seizures in canines by repetitive vagal stimulation. Epilepsia. 1992;33:1005–1012.
  55. Vonck K, Van Laere K, Dedeurwaerdere S, et al. The mechanism of action of vagus nerve stimulation for refractory epilepsy: the current status. J Clin Neurophysiol. 2001;18:394–401.
  56. Stancak A, Kuna M, Srinivasan , et al. Kapalabhati— yogic cleansing exercise. II. EEG topography analysis. Homeost Health Dis. 1991;3:182–189.
  57. Stancák A, Pfeffer D, Hrudová L, et al. Electroencephalographic correlates of paced breathing. Neuroreport. 1993;4:723–726.
  58. Dietrich S, Smith J, Scherzinger C, et al. A novel transcutaneous vagus nerve stimulation leads to brainstem and cerebral activations measured by functional MRI. Biomed Tech (Berl). 2008;53:104–111.
  59. Kraus T, Hosl K, Kiess O, et al. BOLD fMRI deactivation of limbic and temporal brain structures and mood enhancing effect by transcutaneous vagus nerve stimulation. J Neural Transm. 2007;114:1485–1493.
  60. Ventureyra EC. Transcutaneous vagus nerve stimulation for partial onset seizure therapy. A new concept. Childs Nerv Syst. 2000;16:101–102.
  61. Terry RS, Tarver WB, Zabara J. The implantable neurocybernetic prosthesis system. Pacing Clinical Electrophysiol. 1991;14:86–93.
  62. George MS, Rush AJ, Sackeim HA, et al. Vagus nerve stimulation (VNS): utility in neuropsychiatric disorders. Int J Neuropsychopharmacol. 2003;6:73–83.
  63. Krahl SE, Senanayake SS, Pekary AE, et al. Vagus nerve stimulation (VNS) is effective in a rat model of antidepressant action. J Psychiatr Res. 2004;38:237–240.
  64. Rush AJ, George MS, Sackeim HA, et al. Vagus nerve stimulation (VNS) for treatment-resistant depressions: a multicenter study. Biol Psychiatry. 2000;47:276–286.
  65. Sackeim HA, Rush AJ, George MS, et al. Vagus nerve stimulation (VNS) for treatment-resistant depression: efficacy, side effects, and predictors of outcome. Neuropsychopharmacology. 2001;25:713–728.
  66. Marangell LB, Rush AJ, George MS, et al. Vagus nerve stimulation (VNS) for major depressive episodes: one year outcomes. Biol Psychiatry. 2002;51:280–287.
  67. Nahas Z, Marangell LB, Husain MM, et al. Two-year outcome of vagus nerve stimulation (VNS) for treatment of major depressive episodes. J Clin Psychiatry. 2005;66:1097–1104.
  68. Dunner DL, Rush AJ, Russell JM, et al. Prospective long-term multicenter study of the naturalistic outcomes of patients with treatment-resistant depression. J Clin Psychiatry. 2006;67:688–695.
  69. Rush AJ, Marangell LB, Sackeim HA, et al. Vagus nerve stimulation for treatment-resistant depression: a randomized controlled acute phase trial. Biol Psychiatry. 2005;58:347–354.
  70. Rush AJ, Sackeim HA, Marangell LB, et al. Effects of 12 months of vagus nerve stimulation in treatment-resistant depression: a naturalistic study. Biol Psychiatry. 2005;58:355–363.
  71. George MS, Rush AJ, Marangell LB, et al. A one-year comparison of vagus nerve stimulation with treatment as usual for treatment-resistant depression. Biol Psychiatry. 2005;58:364–373.
  72. Nierenberg AA, Alpert JE, Gardner-Schuster EE, et al. Vagus nerve stimulation: 2-year outcomes for bipolar versus unipolar treatment-resistant depression. Biol Psychiatry. 2008;64:455–460.
  73. Corcoran CD, Thomas P, Phillips J, et al. Vagus nerve stimulation in chronic treatment-resistant depression. Br J Psychiatry. 2006;189:282–283.
  74. Rush AJ, Fava M, Wisniewski SR, et al, and the STAR*D Study Team. Sequenced treatment alternatives to relieve depression (STAR*D): rationale and design. Control Clin Trials. 2004;25:119–142.
  75. Trivedi MH, Rush AJ, Wisniewski SR, et al, and the STAR*D Study Team. Evaluation of outcomes with citalopram for depression using measurement-based care in STAR*D: implications for clinical practice. Am J Psychiatry. 2006;163:28–40.
  76. Rush AJ, Trivedi MH, Wisniewski SR, et al, and the STAR*D Study Team. Bupropion-SR, sertraline, or venlafaxine-XR after failure of SSRIs for depression. N Engl J Med. 2006;354:1231–1242.
  77. Trivedi MH, Fava M, Wisniewski SR, et al, and the STAR*D Study Team. Medication augmentation after the failure of SSRIs for depression. N Engl J Med. 2006;354:1243–1252.
  78. Fava M, Rush AJ, Wisniewski SR, et al, and the STAR*D Study Team. A comparison of mirtazapine and nortriptyline following two consecutive failed medication treatments for depressed outpatients: a STAR*D report. Am J Psychiatry. 2006;163:1161–1172.
  79. Nierenberg AA, Fava M, Trivedi MH, et al, and the STAR*D Study Team. A comparison of lithium and T(3) augmentation following two failed medication treatments for depression: a STAR*D report. Am J Psychiatry. 2006;163:1519–1530.
  80. McGrath PJ, Stewart JW, Fava M, et al, and the STAR*D Study Team. Tranylcypromine versus venlafaxine plus mirtazapine following three failed antidepressant medication trials for depression: a STAR*D report. Am J Psychiatry. 2006;163:1531–1541.
  81. Marangell LB, Suppes T, Zboyan HA, et al. A 1-year pilot study of vagus nerve stimulation in treatment-resistant rapid-cycling bipolar disorder. J Clin Psychiatry. 2008;69:183–189.
  82. George MS, Ward HE Jr, Ninan PT, et al. A pilot study of vagus nerve stimulation (VNS) for treatment-resistant anxiety disorders. Brain Stimul. 2008;1:112–121.
  83. Sackeim HA, Keilp JG, Rush AJ, et al. The effects of vagus nerve stimulation on cognitive performance in patients with treatment-resistant depression. Neuropsychiatry Neuropsychol Behav Neurol. 2001;14:53–62.
  84. Sjögren MJ, Hellström PT, Jonsson MA, et al. Cognition-enhancing effect of vagus nerve stimulation in patients with Alzheimer’s disease: a pilot study. J Clin Psychiatry. 2002;63:972–980.
  85. Merrill CA, Jonsson MA, Minthon L, et al. Vagus nerve stimulation in patients with Alzheimer’s disease: additional follow-up results of a pilot study through 1 year. J Clin Psychiatry. 2006;67:1171–1178.
  86. Multon S, Schoenen J. Pain control by vagus nerve stimulation: from animal to man…and back. Acta Neurol Belg. 2005;105:62–67.
  87. Lenaerts ME, Oommen KJ, Couch JR, et al. Can vagus nerve stimulation help migraine? Cephalalgia. 2008;28:392–395.
  88. Mauskop A. Vagus nerve stimulation relieves chronic refractory migraine and cluster headaches. Cephalalgia. 2005;25:82–86.
  89. Berthoud HR. The vagus nerve, food intake, and obesity. Regul Pept. 2008;149:15–25.
  90. Burneo JG, Faught E, Knowlton R, et al. Weight loss associated with vagus nerve stimulation. Neurology. 2002;59:463–464.
  91. Pardo JV, Sheikh SA, Kuskowski MA, et al. Weight loss during chronic cervical vagus nerve stimulation in depressed patients with obesity: an observation. Int J Obes (Lond). 2007;31:1756–1759.
  92. Sobocki J, Królczyk G, Herman RM, et al. Influence of vagal nerve stimulation on food intake and body weight: results of experimental studies. J Physiol Pharmacol. 2005;56(suppl 6):27–33.
  93. Roslin M, Kurian M. The use of electrical stimulation of the vagus nerve to treat morbid obesity. Epilepsy Behav. 2001;2(suppl):S11–S16.
  94. Pereira EA, Green AL, Nandi D, et al. Deep brain stimulation: indications and evidence. Expert Rev Med Devices. 2007;4:591–603.
  95. Pilitsis JG, Metman LV, Toleikis JR, et al. Factors involved in long-term efficacy of deep brain stimulation of the thalamus for essential tremor. J Neurosurg. 2008;109:640–646.
  96. Moro E, Lang AE. Criteria for deep-brain stimulation in Parkinson’s disease: review and analysis. Expert Rev Neurother. 2006;6:1695–1705.
  97. Johnson MD, Miocinovic S, McIntyre CC, et al. Mechanisms and targets of deep brain stimulation in movement disorders. Neurotherapeutics. 2008;5:294–308.
  98. Halpern CH, Samadani U, Litt B, et al. Deep brain stimulation for epilepsy. Neurotherapeutics. 2008;5: 59–67.
  99. Duncan GH, Bushnell MC, Marchand S. Deep brain stimulation: a review of basic research and clinical studies. Pain. 1991;45:49–59.
  100. Levy RM, Lamb S, Adams JE. Treatment of chronic pain by deep brain stimulation: long term follow-up and review of the literature. Neurosurgery. 1987;21:885–893.
  101. Halpern CH, Wolf JA, Bale TL, et al. Deep brain stimulation in the treatment of obesity. J Neurosurg. 2008;109:625–634.
  102. Saleh C, Okun MS. Clinical review of deep brain stimulation and its effects on limbic basal ganglia circuitry. Front Biosci. 2008;13:5708–5731.
  103. Lipsman N, Neimat JS, Lozano AM. Deep brain stimulation for treatment-refractory obsessive-compulsive disorder: the search for a valid target. Neurosurgery. 2007;61:1–13.
  104. Van Laere K, Nuttin B, Gabriels L, et al. Metabolic imaging of anterior capsular stimulation in refractory obsessive-compulsive disorder: a key role for the subgenual anterior cingulate and ventral striatum. J Nucl Med. 2006;47:740–747.
  105. Abelson JL, Curtis GC, Sagher O, et al. Deep brain stimulation for refractory obsessive-compulsive disorder. Biol Psychiatry. 2005;57:510–516.
  106. Greenberg BD, Malone DA, Friehs GM, et al. Three-year outcomes in deep brain stimulation for highly resistant obsessive-compulsive disorder. Neuropsychopharmacology. 2006;31:2384–2393.
  107. Nuttin BJ, Gabriëls LA, Cosyns PR, et al. Long-term electrical capsular stimulation in patients with obsessive-compulsive disorder. Neurosurgery. 2003;52:1263–1274.
  108. Agid Y, Buzsáki G, Diamond DM, et al. How can drug discovery for psychiatric disorders be improved? Nat Rev Drug Discov. 2007;6:189–201.
  109. Mayberg HS, Lozano AM, Voon V, et al. Deep brain stimulation for treatment-resistant depression. Neuron. 2005;45:651–660.
  110. Lozano AM, Mayberg HS, Giacobbe P, et al. Subcallousal cingulate gyrus deep brain stimulation for treatment-resistant depression. Biol Psychiatry. 2008; 64:461–467.
  111. Greenberg B, Friehs G, Carpenter L, et al. Deep brain stimulation: clinical findings in intractable depression and OCD (Abstract). Neuropsychopharmacology. 2004;29:S32.
  112. Malone DA, Dougherty DD, Rezai AR, et al. Deep brain stimulation of the ventral capsule/ventral striatum for treatment-resistant depression. Biol Psychiatry. 2009;65:267–275.
  113. Bewernick BH, Hurlemann R, Matusch A, et al. Nucleus accumbens deep brain stimulation decreases ratings of depression and anxiety in treatment-resistant depression. Biol Psychiatry. 2010;67:110–116.
  114. Hauptman JS, DeSalles AA, Espinoza R, et al. Potential surgical targets for deep brain stimulation in treatment-resistant depression. Neurosurg Focus. 2008;25:E3.
  115. Friedland DR, Gaggl W, Runge-Samuelson C, et al. Feasibility of auditory cortical stimulation for the treatment of tinnitus. Otol Neurotol. 2007;28:1005–1012.
  116. Harvey RL, Nudo RJ. Cortical brain stimulation: a potential therapeutic agent for upper limb motor recovery following stroke. Top Stroke Rehabil. 2007;14:54–67.
  117. Dougherty DD, Thase ME, Howland RH, et al. Feasibility study of an implantable cortical stimulation system for patients with major depressive disorder. Biol Psychiatry. 2008;63:294S–295S.
  118. Howland RH, Thase ME, Evans K, et al. Feasibility study of an implantable cortical stimulation system for patients with major depressive disorder. Paper presented at: American Psychiatric Association Annual Meeting; May 3-8, 2008; Washington, DC.
  119. Kopell BH, Kondziolka D, Dougherty DD, et al. Feasibility study of the safety and effectiveness of an implantable cortical stimulation system for subjects with major depression (Abstract 863). Neurosurgery. 2007;61:215.
  120. Nahas Z, Anderson BS, Borckardt J, et al. Bilateral epidural prefrontal cortical stimulation for treatment-resistant depression. Biol Psychiatry. 2010;67:101–109.
  121. Grill WM. Safety considerations for deep brain stimulation: review and analysis. Expert Rev Med Devices. 2005;2:409–420.
  122. Hamani C, Lozano AM. Hardware-related complications of deep brain stimulation: a review of the published literature. Stereotact Funct Neurosurg. 2006;84:248–251.
  123. Fields JA, Tröster AI. Cognitive outcomes after deep brain stimulation for Parkinson’s disease: a review of initial studies and recommendations for future research. Brain Cogn. 2000;42:268–293.
  124. Woods SP, Fields JA, Tröster AI. Neuropsychological sequelae of subthalamic nucleus deep brain stimulation in Parkinson’s disease: a critical review. Neuropsychol Rev. 2002;12:111–126.
  125. Daly DD, Pedley TA. eds. Current practice of clinical electroencephalography, 2nd ed. New York, NY: Raven Press; 1990.
  126. Leuchter AF, Cook IA, Uijtdehaage SH, et al. Brain structure and function, and the outcomes of treatment for depression. J Clin Psychiatry. 1997;58(suppl 16):22–31.
  127. Leuchter AF, Cook IA, Lufkin RB, et al. Cordance: a new method for assessment of cerebral perfusion and metabolism using quantitative electroencephalography. Neuroimage. 1994;1:208–219.
  128. Leuchter AF, Uijtdehaage SH, Cook IA, et al. Relationship between brain electrical activity and cortical perfusion in normal subjects. Psychiatry Res. 1999;90:125–140.
  129. Drevets WC, Gadde KM, Krishnan KRR. Neuroim-aging studies of mood disorders. In: Charney DS, Nestler EJ, eds. Neurobiology of mental illness, 2nd ed. New York, NY: Oxford University Press; 2004:461-490.
  130. Bruder GE, Stewart JW, Tenke CE, et al. Electroencephalographic and perceptual asymmetry differences between responders and nonresponders to an SSRI antidepressant. Biol Psychiatry. 2001;49:416–425.
  131. Pizzagalli D, Pascual-Marqui RD, Nitschke JB, et al. Anterior cingulate activity as a predictor of degree of treatment response in major depression: evidence from brain electrical tomography analysis. Am J Psychiatry. 2001;158:405–415.
  132. Knott VJ, Telner JI, Lapierre YD, et al. Quantitative EEG in the prediction of antidepressant response to imipramine. J Affect Disord. 1996;39:175–184.
  133. Knott V, Mahoney C, Kennedy S, et al. EEG correlates of acute and chronic paroxetine treatment in depression. J Affect Disord. 2002;69:241–249.
  134. Ulrich G, Haug HJ, Fähndrich E. Acute versus chronic EEG effects of maprotiline- and in clomipramine-treated depressive inpatients and the prediction of therapeutic outcome. J Affect Disord. 1994;32:213–217.
  135. Cook IA, Leuchter AF, Morgan M, et al. Early changes in prefrontal activity characterize clinical responders to antidepressants. Neuropsychopharmacology. 2002;27:120–131.
  136. Leuchter AF, Cook IA, Witte EA, et al. Changes in brain function of depressed subjects during treatment with placebo. Am J Psychiatry. 2002;159:122–129.
  137. Hunter AM, Leuchter AF, Morgan ML, et al. Changes in brain function (quantitative EEG cordance) during placebo lead-in and treatment outcomes in clinical trials for major depression. Am J Psychiatry. 2006;163:1426–1432.
  138. Hunter AM, Leuchter AF, Morgan ML, et al. Neurophysiologic correlates of side effects in normal subjects randomized to venlafaxine or placebo. Neuropsychopharmacology. 2005;30:792–799.
  139. Cook IA, Leuchter AF, Morgan ML, et al. Changes in prefrontal activity characterize clinical response in SSRI nonresponders: a pilot study. J Psychiatr Res. 2005;39:461–466.
  140. Cook IA, Leuchter AF, Greenwald SD, et al. QEEG cordance predicts treatment response to reboxetine in major depression. Paper presented at: American Psychiatric Association Annual Meeting; May 17-22, 2003; San Francisco, CA.
  141. Leuchter AF, Kofol T, Greenwald SD, et al. Source localization of anterior cingulate activity: comparison of 4 vs. 24 channel EEG techniques. Paper presented at: American Psychiatric Association Annual Meeting; May 21-26, 2005; Atlanta, GA.
  142. Iosifescu DV, Greenwald S, Smith CP, et al. Frontal EEG at 1 week predicts clinical response to SSRI treatment in major depressive disorder. Paper presented at: American Psychiatric Association Annual Meeting; May 20-25, 2006; Toronto, Canada.
  143. Poland RE, Greenwald S, Smith CP, et al. Frontal EEG at 1 week predicts response to treatment with citalopram in MDD. Paper presented at: American Psychiatric Association Annual Meeting; May 20-25, 2006; Toronto, Canada.
  144. Iosifescu DV, Greenwald S, Devlin P, et al. Pretreatment frontal EEG and changes in suicidal ideation during SSRI treatment in major depressive disorder. Acta Psychiatr Scand. 2008;117:271–276.
  145. Leuchter AF, Cook IA, Marangell LB, et al. Comparative effectiveness of biomarkers and clinical indicators for predicting outcomes of SSRI treatment in major depressive disorder: results of the BRITE-MD study. Psychiatry Res. 2009;169:124–131.
  146. Leuchter AF, Cook IA, Gilmer WS, et al. Effectiveness of a quantitative electroencephalographic bio-marker for predicting differential response or remission with escitalopram and bupropion in major depressive disorder. Psychiatry Res. 2009;169:132–138.
  147. Howland RH. Therapeutic armamentarium for treating depression. Postgrad Med. 2010;122:66–93
  148. Rush AJ, Trivedi MH, Wisniewski SR, et al. Acute and longer-term outcomes in depressed outpatients requiring one or several treatment steps: a STAR*D report. Am J Psychiatry. 2006;63:1905–1917.

CORRESPONDENCE: Robert H. Howland, MD, Associate Professor of Psychiatry, University of Pittsburgh School of Medicine, Western Psychiatric Institute and Clinic, 3811 O’Hara Street, Pittsburgh, PA 15213, USA, E-MAIL: HowlandRH@upmc.edu