Surgical Treatment of Major Depression

Published on 13/03/2015 by admin

Last modified 13/03/2015

Print this page

This article have been viewed 3535 times

CHAPTER 89 Surgical Treatment of Major Depression

Clement Hamani, Brian Snyder, Adrian W. Laxton, Paul E. Holtzheimer, Helen S. Mayberg, Andres M. Lozano

Depression is a prevalent and often debilitating psychiatric disorder with a 6-month prevalence of approximately 5%.¹ In the United States alone, approximately 20 million people have been diagnosed with depression at some point in their lives. Depression accounts for the greatest number of missed workdays due to illness, with an estimated economic burden of approximately $40 billion per year.²

The initial therapeutic approach for patients with depression consists of medication or psychotherapy, or both. For patients who do not respond, suitable alternatives include other medications from the same or different drug classes, augmentative pharmacologic regimens, a combination of antidepressant agents from different classes, and eventually electroconvulsive therapy (ECT). The main objectives of treatment are the remission of symptoms, restoration of daily function, and prevention of relapse and recurrence. It is estimated that only 60% to 70% of patients respond favorably to initial treatment. More important, failure to respond to any initial treatment predicts a poor response to future treatments (including ECT) and a higher likelihood of relapse.³^,⁴ Consequently, up to 40% of patients (1.5% of the general population) have chronic and refractory forms of depression. Up to 15% of patients with severe depression require hospitalization and eventually commit suicide.⁵^,⁶ Therefore, for this refractory population, surgery may be considered part of the therapeutic armamentarium.

Diagnosis

When diagnosing depression, a distinction must be made between major depressive disorder (MDD) and major depressive episode (MDE). For the diagnosis of MDE, patients must present at least five of the following symptoms for a minimum of 2 weeks: depressed mood; markedly diminished interest or pleasure or significant apathy; significant change in appetite or weight; insomnia or hypersomnia; psychomotor agitation or retardation; fatigue or loss of energy; feelings of worthlessness, excessive or inappropriate guilt, or loss of self-esteem; indecisiveness or diminished ability to think or concentrate; and recurrent thoughts of death or suicidal ideas without a specific plan, a specific plan for suicide, or an actual attempt. At least one of the five symptoms constituting a diagnosis of MDE should be a loss of interest or pleasure or a depressed mood that persists for most of the day. These symptoms must not be secondary to substance abuse or medication use, underlying medical conditions, or bereavement. They must represent a change from antecedent functioning and cause marked distress or significant impairment of social or occupational functioning. Diagnostic criteria for MDD include the occurrence of one or more MDEs with an absence of any history of manic, mixed (combined depressive and manic), or hypomanic episodes. MDEs may also occur in bipolar disorder, an illness characterized by a history of MDEs in addition to hypomanic, mixed, and manic episodes.

Although these diagnostic criteria for MDE and MDD are widely used, it is important that an in-depth phenomenologic evaluation be performed by an experienced clinician to ascertain a diagnosis of MDD. Moreover, the differential diagnosis of MDE, MDD, and bipolar disorder includes other psychiatric, neurological, and medical illnesses that likely have a different neurobiology (e.g., anxiety disorders, personality disorders, dementia, endocrine abnormalities, stroke, movement disorders). The currently accepted diagnostic criteria for these and other psychiatric disorders are codified in the fourth edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-IV) published by the American Psychiatric Association. It is worth noting that these criteria were not firmly established in their current form until 1980. Therefore, reports using diagnostic labels such as depression, mania, bipolar disorder, psychosis, and schizophrenia before that time are limited by significant diagnostic heterogeneity.

History of Psychosurgery

The history of neurosurgical intervention for psychiatric illness may be as old as the history of neurosurgery itself. The record of trephination to remove parts of the skull dates from as early as 10,000 BC.⁷ This procedure was apparently common and widespread, with more than 1500 specimens discovered from Europe, Asia, Africa, North America, Central America, South America, and Oceania.⁸^,⁹ Although the exact purposes of such trephinations are disputed, some were clearly motivated by magicotherapeutic goals, with psychiatric illnesses mistakenly thought to arise from demonic possession.¹⁰ Written records of trephining for the “relief of unexplained and unbearable pain … melancholia … or to release demons” have been dated as early as 1500 BC.¹¹

The literature of the late 19th and early 20th centuries is peppered with a few reports of neurosurgical interventions for psychiatric illness. In 1891, Gottlieb Burckhardt performed bilateral cortical resections (topectomies) in six patients institutionalized for pathologic behavior. Five survived the surgery, and although they remained psychotic, they were considered to be more placid.¹²^,¹³ Burckhardt received significant criticism after the report and abandoned the procedure. In 1910, Ludwig Puusepp reported on the severing of fibers running from the frontal to the parietal lobes in three “bipolar” patients, but he ultimately considered these to be surgical failures. Fourteen other patients who underwent frontal leucotomy were relieved of their aggressive symptoms.¹⁴ Studies by Dandy,¹⁵ Penfield,¹⁶ and Bailey¹⁷ documented how the treatment of neurosurgical diseases such as tumors and abscesses relieved mental symptoms of anxiety and depression and provided insight into the potential cerebral localization of such problems. In 1933, John Fulton and Carlyle Jacobsen operated on chimpanzees that had been trained to perform various tasks. Before the surgeries, the animals were noted to have severe “frustrational behavior” when they were not rewarded after performing their tasks poorly. After unilateral frontal lobectomies, no significant changes in behavior or temperament were noticed.¹⁸ When the procedures were done bilaterally, however, the animals displayed little agitation or concern about making poor choices on the learned tasks or the lack of reward for completion of the tasks (although general behavior was not thought to have changed). Fulton presented the data in 1935 at the International Neurologic Conference in London,¹⁹ which was attended by Egaz Moniz, a neurologist from Portugal. Inspired by Fulton’s experiments, Moniz introduced the prefrontal leucotomy (later renamed lobotomy) later that year.²⁰ The initial operations, performed with his colleague Almeida Lima, involved alcohol injections into the centrum semiovale, with the goal of disrupting frontal projections to and from the thalamus; however, because of the need to increase the number of injections and a lack of control owing to leakage of alcohol along the needle track, the authors switched to the leucotome by their ninth procedure.²¹

Also in the audience at Fulton’s presentation was American neurologist Walter Freeman. At the time, psychiatric illness was exacting a staggering toll in the United States. The estimated financial burden of psychiatric illness was $1.5 billion a year, with 450,000 patients living in asylums.²²^,²³ The results of Freeman’s early experimentation with lobotomy, performed in collaboration with neurosurgeon James Watts, were mixed. Freeman thought the Moniz procedure was inadequate, but initial attempts at a deeper lesion resulted in complications and fatalities.²⁴ Freeman continued to modify the procedure, employing x-ray and skull landmarks in an attempt to better define the appropriate target. He yearned to make the procedure safer and easier and to eliminate the need for anesthesia, the operating room, and the surgeon.²⁵ Consequently, in 1946 Freeman introduced the transorbital lobotomy, first described by Amarro Fiamberti, to the United States. He believed this procedure would realize his dream and make lobotomy more accessible to the mentally ill. During this procedure, which Freeman performed in the outpatient setting, the patient was anesthetized via electroconvulsion. An ice pick was then driven with a mallet through the orbital roof and swept in a specific fashion and direction to sever the desired tracts. Freeman performed or supervised 4000 transorbital lobotomies before the procedure fell out of favor.²⁶^,²⁷ It is estimated that between 1945 and 1955, 50,000 lobotomies were performed in the United States.¹⁹^,²³

Fueled by a combination of the variability in patient selection, the morbidity and mortality associated with injury to functioning brain tissue, the lack of objective documentation of improvement, and the sheer number of procedures, a public backlash against lobotomy developed. Fear and paranoia surrounding lobotomy and its possible use for sinister purposes or mind control permeated popular culture, as represented in works such as Tennessee Williams’s Suddenly Last Summer and Ken Kesey’s One Flew over the Cuckoo’s Nest. In 1954, chlorpromazine was approved by the Food and Drug Administration, effectively ending the lobotomy era.

More-Selective Lesioning Techniques

The side effects described in the early leucotomy-lobotomy series led many authors to seek less invasive alternatives. These so-called limited lesions and selective undercutting procedures were conducted primarily at three targets: the orbitofrontal cortex, the superior convexity of the frontal cortex, and the medial prefrontal cortex.

Orbitofrontal Cortex

As early as the 1930s, it was realized that lesions closer to the orbital and inferior aspects of the frontal lobes produced changes in emotional tone, whereas lesions involving the superolateral aspects of the frontal lobe were associated with intellectual disturbance.²⁸ In addition, stimulation of the orbitofrontal and medial prefrontal cortices was soon recognized to produce autonomic responses. These responses were thought to be predictors of the response to ablation because the ablation of corresponding regions in animals induced tranquility and loss of fear. Based on these findings, Scoville proposed the undercutting of the orbitofrontal cortex as a surgical therapy for psychiatric disorders.²⁹^–³³ The procedure involved sectioning the brain parenchyma in a plane parallel to and approximately 1 cm dorsal to the orbital surface. The anteroposterior extent of the undercutting was dictated by the distance between the rostral portion of the frontal lobe and the point of emergence of the optic nerve from the optic foramen.³⁴ The cut was extended medially until the midline was nearly reached and laterally until increased resistance indicated that the lateral cranial wall was being approached.³⁴ Patients with depression and obsessive features responded best to this procedure.²⁹^–³⁴ Of considerable importance, adverse effects were not significant when compared with those associated with leucotomy. No intellectual deterioration was noted, and there was only minor, transient blunting of personality; however, some loss of spontaneity (which did not usually culminate in apathy) and transient postoperative confusion were often reported.²⁹^–³⁴

Subsequently, Knight modified the surgery by omitting the lateral portion of the cut, confining the incision to an arc about 2.5 cm wide, passing back to the inner aspect of the frontal lobe beneath the anterior horn of the lateral ventricle.³⁵ As with other orbitofrontal undercutting procedures, patients with depression and anxiety experienced good outcomes.³⁵ Approximately 80% to 90% of patients with depression treated with selective orbitofrontal undercutting were said to respond to some extent.³⁵

Superior Convexity of the Frontal Cortex

In contrast to orbitofrontal undercutting, selective lesions of the superior convexity of the frontal cortex were performed mainly in patients with paraphrenia and severe psychotic disorders, including schizophrenia.³⁶^,³⁷ This technique was eventually abandoned owing to poor clinical results.

Medial Prefrontal Cortex

Lesions of medial prefrontal cortical structures were also attempted for the treatment of psychiatric disorders. These focused mainly on the cingulate gyrus, with the goal of disconnecting the frontal lobes from the limbic system (a more extensive rationale for the development of cingulotomy is described later). Surgery was initially performed as an open procedure, with direct visualization of the medial prefrontal cortical structures. The main goal of cingulotractotomies and subrostral cingulotomies was to lesion the portion of the cingulate gyrus ventral to the genu and rostrum of the corpus callosum.³⁸^,³⁹ Seventy percent to 80% of patients with depression were reported to respond favorably to this procedure. The rostral portion of the knee of the corpus callosum or the adjacent cingulate gyrus was also targeted³⁹^,⁴⁰; however, this so-called mesoloviotomy was not as effective as subrostral cingulotomy for alleviating the symptoms of depression.³⁹

Current Stereotactic Techniques

The development of a stereotactic frame for human use in 1947 ushered in the modern era of stereotactic neurosurgery, allowing surgeons to lesion the brain in a safer and more controlled fashion. Spiegel and colleagues ⁴¹ were the first to apply stereotactic techniques and reported promising results when lesions of the medial thalamic region were carried out to reduce emotional reactivity in psychiatric patients. Since then a variety of deep brain structures have been targeted to treat psychiatric illness. A review of the most successful of these procedures follows.

Subcaudate Tractotomy

In 1955, Knight reported his preliminary experience with selective undercutting of the orbitofrontal cortex for the treatment of psychiatric disorders.³⁵ In subsequent articles it was postulated that the outcome of the procedure, later named subcaudate tractotomy, derived from the disruption of fibers interconnecting the frontal lobe, substantia innominata, amygdala, and hypothalamus.⁴²^–⁴⁴ In fact, the general belief was that lesions would more specifically compromise fibers adjacent to area 13 (and to some extent area 14) and the substantia innominata (Fig. 89-1).

FIGURE 89-1 Anatomic location of the targets most commonly used during ablative procedures for the treatment of depression. As depicted, limbic leucotomy consists of a combination of cingulotomy and subcaudate tractotomy.

(From Lipsman N, Neimat JS, Lozano AM. Deep brain stimulation for treatment-refractory obsessive-compulsive disorder: the search for a valid target. Neurosurgery. 2007;61:1-11.)

A significant problem with the freehand lesioning approach is its inherent imprecision.⁴⁵ Consequently, in the 1960s Knight began to perform subcaudate tractotomy using stereotactic techniques. At first, lesions were created with radioactive yttrium.⁴⁴ Between three and five rods were implanted per hemisphere, creating lesions that measured roughly 20 × 20 × 5 mm. Based on Knight’s early reports, patients with depression responded particularly well to the procedure, compared with patients with schizophrenia and personality disorders.^43,⁴⁶^–⁵³ Overall, 34% to 50% of patients had a good outcome, with or without minimal residual symptoms. In addition, 17% to 32% of patients showed some degree of improvement but had persistent symptoms that needed treatment.^46,^47,^51,⁵² Of interest, the improvement after subcaudate tractotomy seemed to occur over weeks to months.⁴⁶ Medications were reduced and discontinued whenever possible. In the 1990s, thermocontrolled electrocoagulation replaced yttrium as the preferred means of creating lesions, with comparable results reported.⁴⁶^,⁵³

Complications of subcaudate tractotomy include confusion in the early postoperative period (usually the first day after surgery), with disorientation seen in 10% of the patients.⁴⁶^,⁴⁷ Seizures occur in 1% to 2%. The postoperative suicide rate is approximately 1%. In the early postoperative period, neuropsychological tests demonstrate decay in recognition memory tests and a marked tendency to confabulate during recall memory tasks.⁵⁴ Fortunately, these deficits appear to be transient, with no lasting effects at long-term follow-up.

Cingulotomy

The rationale for the development of cingulotomy derives from the hypothesis that the Papez circuit is of primary importance in mediating inward emotional experience and outward emotional expression.⁵⁵ Because the interruption of cingulate fibers in nonhuman primates induces a state of “tameness and placidity,” it was suggested that cingulotomy could be used to treat patients with psychiatric symptoms.⁵⁶^–⁵⁹

Cingulotomy has been extensively explored since the 1950s. The procedure has been used mainly for the treatment of obsessive-compulsive disorder (OCD), so few patients with depression have been reported in the literature.⁶⁰^–⁶⁵ In their first studies of cingulotomy at the Massachusetts General Hospital, Ballantine and colleagues⁶²^,⁶³ reported very good outcomes in patients with MDD. In a recent study from the same center, Shields and colleagues⁶⁶ reported on 33 patients with MDD who underwent one or more ablative procedures. Each of the subjects was initially treated with cingulotomy. A response to surgery was defined as a 50% reduction in the Beck Depression Inventory (BDI) score and a Clinical Global Improvement (CGI) score of less than 2. Patients who did not achieve a 35% improvement in the BDI and did not have a CGI of 2 or less were considered nonresponders and were deemed candidates for additional surgical procedures. Overall, 17 patients (52%) underwent single cingulotomies, 9 (27%) repeated cingulotomies, and 7 (21%) limbic leucotomies. Of the patients treated with one cingulotomy, 41% were considered to be responders and 35% were partial responders. Of the patients treated with multiple lesions, 25% were responders and 50% were partial responders.⁶⁶ The outcomes for multiple cingulotomies versus cingulotomy followed by limbic leucotomy were not compared.

The usual target for cingulotomy is the dorsal aspect of the anterior cingulate gyrus (Brodmann’s area [BA] 24), 2 to 4 cm posterior to the anterior aspect of the frontal horn (see Fig. 89-1).^61,^62,⁶⁶^–⁶⁸ The correlation between clinical outcome and lesion location was recently investigated in patients with MDD by Steele and colleagues.⁶⁸ They suggested that the more anterior the lesion within the usually targeted anterior cingulate cortex region, the better the outcome as assessed by the Montgomery Asberg Depression Scale (MADRS) and the Hamilton Depression Rating Scale (HAMD). The optimal lesion volume was 1000 to 2000 mm³.

The most commonly reported side effects of cingulotomy are a 1% to 2% risk of seizures and a less than 5% risk of transient urinary incontinence.^60,^61,^63,^64,⁶⁹

Limbic Leucotomy

The objective of limbic leucotomy is to interrupt connections between the frontal cortex and the limbic system by disrupting white matter fibers crossing the anteromedial quadrant of the frontal lobe (underneath the caudate nucleus) and the cingulate bundle. In sum, limbic leucotomy consists of a subcaudate tractotomy combined with a cingulotomy.

Most surgeons perform two to three lesions in the anteromedial frontal lobe beneath the caudate nucleus and two to four lesions in the cingulate gyrus (BA 24).⁷⁰^–⁷⁸ In older series, 80% of patients with depression treated with limbic leucotomy were considered to benefit from surgery, half being symptom free or much improved.⁷⁰^–⁷⁵ The mean HAMD score was improved 51%.⁷⁴ The main side effects include postoperative confusion, seizures, and incontinence.

In a more recent study, six patients with MDD were treated with limbic leucotomy, either as a first procedure or after unsuccessful cingulotomy.⁷⁶ In that trial, 50% of patients were considered to be responders based on physicians’ rated assessments. In addition, 40% of patients who had both pre- and postoperative BDI scores were classified as responders. One depressed patient with a history of suicidal ideation and many serious attempts committed suicide after the procedure. Other side effects were similar to those described in earlier series, including transient somnolence and apathy (25% to 30%), postoperative seizures (19%), and bladder incontinence (mostly transient; 24%). These outcomes have been corroborated in a larger clinical series.⁶⁶

Capsulotomy

The goal of capsulotomy is to disrupt the frontothalamic fiber systems running in the anterior limb of the internal capsule (see Fig. 89-1). Some of the projections believed to be important extend from the prefrontal cortex and substantia innominata to the hypothalamus.⁷⁹ Capsulotomy has been used predominantly to treat anxiety and OCD.⁷⁹^–⁸²

Deep Brain Stimulation

Deep brain stimulation (DBS) involves the delivery of electrical current to the brain parenchyma through implanted electrodes. One of the main advantages of DBS is that most of its side effects are reversible and can be managed by adjusting stimulation parameters. This enables the modulation of disease states without irreversibly destroying neural tissue, as occurs with ablative procedures. At present, the most commonly used DBS hardware has four main components: a quadripolar electrode that is implanted into the brain parenchyma at a specified target, a plastic ring or adapter that is fixed to the cranium and holds the electrode in place, an implantable pulse generator (IPG), and extension cables that are tunneled subcutaneously from the cranial region to the chest or abdomen, connecting the DBS electrode to the IPG. The adjustable stimulation parameters are stimulus frequency, pulse width, and intensity. In addition, one can choose the electrode contact or combination of contacts to be activated.

DBS has been used extensively for the treatment of movement disorders and pain.⁸³^–⁹⁰ In addition, DBS has been investigated as a potential therapy for minimally conscious states,⁹¹ aggressiveness,⁹² cluster headaches,⁹³^,⁹⁴ obesity,⁹⁵ memory modulation,⁹⁵ and psychiatric disorders, including OCD⁹⁶^–¹⁰¹