Currently, the accepted therapeutic approach to most psychiatric diseases involves a combination of psychotherapy, pharmacotherapy, and, in some instances, electroconvulsive therapy (ECT); however, some patients fail to adequately respond to all available therapeutic interventions and remain severely disabled. In these patients, surgical intervention might be considered appropriate if the overall level of functioning could be improved.

Surgery for intractable psychiatric illness involves the ablation or disconnection of brain tissue with the intent of altering abnormal affective and behavioral states caused by mental illness. It is classified as a functional neurosurgical procedure because it attempts to improve or restore function by altering underlying physiology. Unfortunately, the neurobiological basis of most psychiatric illnesses remains poorly understood, and its expression involves mental or psychic symptoms without localized physiologic abnormalities or objective physical signs. In addition, current public and professional misconceptions and biases concerning psychiatric illness often make it difficult to justify a surgical approach.

Nevertheless, surgery for psychiatric illness has been frequently applied in the past to treat a variety of psychiatric disorders, including affective disorders, obsessive-compulsive disorder (OCD), and schizophrenia. Between the years 1942 and 1954, more than 10,000 cases in England and Wales and more than 18,000 cases in the United States were performed.¹ However, despite or possibly because of such widespread use, surgery for psychiatric illness has at various times been enthusiastically endorsed and at other times flatly rejected by both the medical profession and society at large. To understand this controversy, one must appreciate the historical evolution of psychosurgery.

Psychosurgery had its modern beginning in 1936, largely through the efforts of the Portuguese neurologist Egas Moniz.² After John Fulton³ described the beneficial effects of frontal corticectomies on the behavior of two primates, Moniz felt that similar interventions might be applied to the human condition. Upon his return to Portugal, Moniz² persuaded his neurosurgical colleague, Almeida Lima, to inject alcohol into the white matter of the frontal lobes of 20 institutionalized patients. They felt that 14 of 20 showed worthwhile improvement; this initial experience received early cautious support.²

At the time, no satisfactory pharmacologic treatment options existed, and a variety of unproven somatic therapies were being used, including insulin shock therapy, Metrazol shock therapy, and ECT. Asylums for the insane were overflowing with the mentally ill, and psychiatric illness was a major public health issue; therefore, despite a lack of objective therapeutic benefits and concerns about potentially damaging side effects, psychosurgery was enthusiastically adopted by practitioners of the day. Moniz coined the term pyschosurgery for this new intervention, and his contributions were ultimately recognized in 1949 when he received the Nobel Prize in medicine "for his discovery of the therapeutic value of prefrontal leucotomy in certain psychoses."⁴

One of the most enthusiastic proponents of psychosurgery was Walter Freeman, a neuropsychiatrist. Within a few months of Moniz’s publication, Freeman⁵ performed the first prefrontal lobotomy in the US with the neurosurgical help of James Watts. The Freeman-Watts prefrontal lobotomy was performed through bilateral burr holes placed in the inferior frontal region at the level of the coronal suture. This disconnection procedure was carried out with a specially designed calibrated leucotome that was inserted blindly to the midline and swept back and forth to surgically interrupt the white matter tracts in the frontal lobes. In 1942, Freeman and Watts reported favorable results for the first 200 patients, although they did admit to a significant complication rate, including frontal lobe syndrome, seizures, apathy, decreased attention, and inappropriate behavior.⁵ They reported that the patients were "no longer distressed by their mental conflicts but also seem to have little capacity for any emotional experience."⁵ Despite these side effects, prefrontal lobotomy became widely performed throughout the US, largely because of the lack of satisfactory therapeutic alternatives and the promotional zeal of Freeman himself. Freeman became such an aggressive and enthusiastic proponent of psychosurgery that he lost the confidence and collaboration of his neurosurgeon and devised the transorbital leucotomy,6 which entailed inserting a sharp blade under the eyelids and through the thin orbital roof into the undersurface of the frontal lobes. This procedure was performed by Freeman himself on patients in the immediate postictal period after the administration of ECT. ⁶

These early surgical techniques were crude, inexact, and associated with major mortality and morbidity. Complications from these procedures were significant and included severe intracranial hemorrhage (especially of the anterior cerebral artery), postsurgical epilepsy, and inadvertent lesion enlargement.¹ Mortality rates were reported as high as 10%.⁴ Significant personality changes were seen as an organic frontal lobe syndrome manifest by decreased attention, apathy, and disinhibition. At the height of enthusiasm for psychosurgery, it was recommended for curing or ameliorating schizophrenia, depression, homosexuality, childhood behavior disorders, criminal behavior, and uncontrolled violence. It is estimated that more than 50,000 procedures were performed in the US alone between 1936 and the mid-1950s⁴; much of the controversy surrounding psychosurgery may relate to its overzealous and sometimes indiscriminate application during this period.

As early as 1941, there was much discussion among neurologists, neurosurgeons, and psychiatrists regarding the scientific, ethical, and technical aspects of psychosurgery. This debate and the frequency of harmful side effects stimulated efforts toward more rigorous trials with a goal to reduce lesion size and direct lesions to specific brain regions. With the introduction of human stereotactic techniques in 1947, methods for accurate and reproducible lesioning of cortical and subcortical structures became available. An evolution in surgical technique soon followed: Foltz and White⁷ reported their experience with anterior cingulotomy in 1962, Knight⁸ reported his experience with subcaudate tractotomy in 1964, Leksell⁹ reported his experience with anterior capsulotomy in 1972, and Kelley¹⁰ reported his experience with limbic leucotomy in 1973.

Even though the crude frontal lobotomy of earlier years had been abandoned in favor of more selective stereotactic interventions and the number of procedures performed were greatly reduced, the concern about the use of psychosurgery grew during the 1960s and 1970s. This was due to questions about efficacy, reports of permanent sequelae, and the potential for abuse—all fueled by publications in the lay press. Public debate in the US, England, and Australia prompted the formation of national commissions to investigate psychosurgery and encouraged legislation to regulate its use.

In the US, the report of the National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research11 indicated that psychosurgery was efficacious in more than half of the 400 operations performed annually between 1971 and 1973 and that no psychological deficits could be attributed to the procedures. Their conclusion was that fears of psychosurgery being used on minority and disadvantaged populations for social control were unsubstantiated. Preoperative and postoperative studies¹² by independent observers of smaller groups of patients undergoing cingulotomy demonstrated excellent results in the majority of patients, with significant improvement in full-scale verbal and performance IQ on the Wechsler IQ Scale. Detailed neuropsychological testing of patients undergoing a variety of more extensive psychosurgical procedures also failed to demonstrate any worsening except in a single category (Wisconsin Card-Sorting Test), but over half of these patients experienced a marked improvement in psychiatric symptoms.¹³ These independent evaluations of outcome and side effects appear to corroborate the prevailing published experience. More modern clinical series^14-17 report similar success rates of 25% to 60%, depending on the psychiatric diagnosis and methodology. These results argue against the initial public perception that psychosurgery is dangerous, ineffective, and experimental, but opposition remains despite its formal acceptance in position statements by organized psychiatry in many countries around the world.

The theoretical basis of surgery for psychiatric illness is based upon a thorough understanding of the pertinent anatomy, physiology, and biochemistry. While Moniz was publicizing his initial experience with prefrontal lobotomy, the concept of localized brain function was being elucidated. In 1937, Papez¹⁸ described the concept of a limbic system, proposing that the interconnections between the frontal lobes and subcortical structures form the anatomic basis for emotions and memory. The components of this circuit consist of the hypothalamus, septal nuclei, hippocampus, mamillary bodies, anterior thalamic nuclei, and cingulate gyri. This system was further expanded by McLean¹⁹ to include paralimbic structures, such as the orbital frontal cortex, the insular cortex, anterior temporal cortex, the amygdala, and the dorsomedial thalamic nuclei. These paralimbic structures form the link between the neocortex and the limbic system proper and are, therefore, strategically located to interconnect somatic and visceral stimuli with higher cortical functions and perceptions.

At present, there are several lines of evidence that implicate the limbic system in the pathophysiology of human emotion and psychiatric illness. Electrical stimulation of specific areas within the limbic system (ie, the anterior cingulum) in humans has been shown to alter both autonomic responses and anxiety.^20,21 Both cortical evoked potentials and microelectrode recordings in human cingulate gyri have demonstrated changes in response to the perception of painful stimuli.²² Stimulation of the hypothalamus in animals produces autonomic, endocrine, and complex motor effects, which suggests that the hypothalamus integrates and coordinates the behavioral expression of emotional states.²³

Contemporary neurobiological models of anxiety and affective disorders also include the fundamental role of the limbic system. Neurochemical models indicate that the affective and anxiety disorders may be mediated by monoaminergic systems, with the serotonergic system being particularly important in OCD. In a variety of other studies, dopamine, noradrenaline, serotonin, and acetylcholine have been investigated, and it appears evident that more than one system may be involved.

More recent neuroimaging data^24,25 reiterates the importance of the limbic and paralimbic systems in psychiatric illness. Anatomic imaging with high resolution morphometric magnetic resonance imaging (MRI) has demonstrated focal abnormalities in these striatal areas and smaller caudate nuclei in patients with OCD.^24,25 Functional neuroimaging suggests that metabolic changes in the anterior cingulate may differentiate the depressed from nondepressed state in affected individuals. Positron emission tomography (PET) studies²⁶ suggest that disease remission is associated with increased cerebral blood flow in the left dorsolateral prefrontal cortex and medial prefrontal cortex, including the anterior cingulate. Similar results have been obtained using 99m Tc. When single photon emission computed tomography was performed before and after ECT, a significant increase in tracer uptake was observed in responders to ECT when compared to nonresponders.²⁷ The increased uptake was most notable in the anterior and posterior cingulate. In a symptom provocation OCD study using PET, Rauch and colleagues²⁸ further implicated the orbitofrontal cortex, the caudate nucleus, and the anterior cingulate cortex in the pathophysiology of OCD. Other studies^29,30 support these findings and reveal a common theme correlating anterior cingulate perfusion with changes in depressive symptoms or caudate hypometabolism in OCD.

While the complex neuroanatomy, neurochemistry, and neurophysiology of psychiatric disease are not completely understood, accumulating evidence suggest that the limbic system is clearly involved and is stimulating further investigation.

Only patients with a severe, disabling, and treatment-refractory psychiatric illness should be considered for surgical intervention. The illness must prove to be refractory to systematic trials of pharmacologic, psychologic, and, when appropriate, ECT prior to considering neurosurgical intervention. It is generally agreed that surgery remains "a last resort for severely incapacitated patients who have not responded to all other state-of-the-art pharmacological and behavioral treatments."³¹ As in all medical decisions, the potential benefit from such an intervention must be balanced against the risks imposed by surgery.

The major psychiatric diagnostic groups—as defined by the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition, (DSM-IV)—that might benefit from surgical intervention include OCD and major affective disorder (ie, major depression or bipolar disorder). In many instances, patients present with mixed disorders, combining symptoms of anxiety, depression, and OCD. Due to the severity of the disease, many patients with a primary diagnosis of OCD will have concomitant treatment-refractory depression; these patients remain candidates for neurosurgery. Schizophrenia is not currently considered an indication for surgery. A history of personality disorder, substance abuse, or other Axis II symptomatology is often a relative contraindication to surgery. Only in rare instances should patients with severe violent outbursts and the potential for serious injury or self mutilation be considered for bilateral amygdalotomy, thalamotomy, hypothalamotomy, or limbic leucotomy.

Thoughtful assessment of psychosurgical candidacy requires that criteria for severity, chronicity, disability, and treatment refractoriness be operationalized to form guidelines. In this regard, chronicity would require at least 1 year of enduring symptoms without significant remission, although, practically speaking, confirmation of treatment refractoriness usually requires more than 5 years of illness prior to surgery. Severity is usually measured using validated clinical research instruments corresponding to specific indicators, such as a Yale-Brown Obsessive Compulsive Scale score of >20 for OCD or a Beck Depression Inventory score >30. Disability may be reflected, for instance, by a Global Assessment of Function score of <50.

In order to determine that a patient’s psychiatric illness is refractory to treatment despite appropriate care, the patient must be referred for surgical intervention by his or her treating psychiatrist. The referring psychiatrist must demonstrate an ongoing commitment to the patient and evaluation process and must also agree to be responsible for postoperative management. Detailed questionnaires that document the extent and severity of the illness, as well as a thorough account of the diagnostic and therapeutic history, must be provided by the psychiatrist. The specifics of pharmacologic trials should include the agents used, dose, duration, response, and the reason for discontinuation for any suboptimal trial. Adequate trials of ECT or behavioral therapy, when clinically appropriate, must also be demonstrated.

The patient and their family must also agree to completely participate in the evaluation process as well as the postoperative psychiatric treatment program. In general, only adult patients (older than 18 years of age) who are able to render informed consent and who express a genuine desire and commitment to proceed with surgery are accepted. Obviously, the surgery should only be performed to help a sick patient and never for social or political reasons.

At Massachusetts General Hospital (MGH), if the patient meets the above criteria, they undergo a more detailed presurgical evaluation by an experienced, multidisciplinary group of psychiatrists, neurosurgeons, and neurologists (the MGH Cingulotomy Assessment Committee). Thorough review of the medical record is carried out to ensure that the illness is in fact refractory to all conventional therapies. The MGH evaluation also includes an electroencephalogram, brain MRI, neuropsychological tests, and independently conducted clinical examinations by a psychiatrist, neurologist, and neurosurgeon in an out-patient setting. Electrocardiogram and appropriate blood tests are obtained to assess medical risks and to exclude organic etiologies for mental status abnormalities. Validated clinical research instruments are employed to quantify psychiatric symptom severity.

There must be unanimous agreement that the patient satisfies selection criteria, the surgery is indicated, and the requirements for informed consent are fulfilled. In a review of this topic, Stagno and colleagues³¹ provide an excellent discussion of this issue: "Freedom to choose is an especially complex issue when patients are considering surgery for OCD. The added complexity emerges because OCD patients are often desperate for relief from their disorder." Therefore, it must be determined by the committee that the patient and their family agree to participation in the pre- and postoperative evaluations and treatment processes and are able to give informed consent. A family member or close relative must also understand the evaluation process, the indications for surgery, the risks of surgery, and the alternatives to surgery. They must also agree to be available to provide emotional support for the patient during the hospitalization.

At present, surgical approaches generally consist of one of four distinct procedures: cingulotomy, subcaudate tractotomy, limbic leucotomy, and anterior capsulotomy. All are currently performed bilaterally using stereotactic techniques. In some cases, repeat procedures are performed in order to enlarge the lesions. A direct comparison of these different procedures across centers is impossible because of diagnostic inaccuracies, nonstandardized presurgical evaluation tools, center bias, and varied outcome assessment scales. However, each procedure can be discussed with respect to indications, technique, results, and complications.

Cingulotomy has been the surgical procedure of choice in North America over the last 40 years. Ballantine³² demonstrated the safety and efficacy of cingulotomy in a large number of patients, and it has been used to treat major affective disorders, chronic anxiety states, OCD, and intractable chronic pain.

Cingulotomy is performed using an MRI-compatible stereotactic frame under local anesthetic with intravenous sedation. Sagittal and oblique coronal T1-weighted images are obtained, and target coordinates are calculated for a point in the anterior cingulate gyrus 20–25 mm posterior to the tip of the frontal horns, 7 mm from midline, and 1 mm above the ventricular roof. To ensure ablation of the entire cingulate fasciculus, additional targets can be calculated at 14 mm from the midline, just superior to the roof of the lateral ventricle. Radiofrequency thermocoagulation lesions are created by inserting the electrode to the target and heating to 85ºC for 90 seconds. The electrode is then withdrawn 10 mm, and a second lesion is performed. This technique results in a lesion approximately 2 cm in height and 8–10 mm in diameter (Figure 1).¹⁴

Minor symptoms of headache, low-grade fever, and nausea are common after cingulotomy (as with all stereotactic procedures), but generally last <24–48 hours. Transient unsteady gait, dizziness, confusion, urinary retention, and isolated seizures can occur; although generally mild and self-limited, these symptoms may last up to several weeks. Permanent significant behavioral or cognitive decline has not been observed after cingulotomy. An independent analysis13 of 57 patients before and after cingulotomy found no evidence of lasting neurological or behavioral deficits after surgery. In fact, a comparison of preoperative and postoperative Weschler IQ scores demonstrated significant gains postoperatively. This improvement was greatest in patients with chronic pain and depression but negligible in those with a diagnosis of schizophrenia. Major morbidity is extremely rare, with only 4 intracranial hemorrhages and no deaths in nearly 1000 cingulotomies performed at the Massachusetts General Hospital over the past 40 years.

There is generally a delay in onset of any beneficial effect on depression and OCD following cingulotomy, with the latency being as long as 3–6 months. Ballantine³³ reported significant improvement in 62% of patients overall, 64% of patients with major depression, and 33% of patients with OCD. Modern retrospective and prospective studies using more rigid outcome criteria demonstrated that only a third of patients benefited substantially from cingulotomy. Using the identical outcome measures employed by Ballantine,³³ more recent studies¹⁴ have yielded similar success rates, underscoring the positive bias that occurs with subjective outcome rating scales. Those patients with affective disorders seem to respond better than those with OCD, and approximately 40% of patients will require more than one procedure. ¹⁴

Subcaudate tractotomy was introduced by Sir Geoffrey Knight⁸ and has been used extensively in the United Kingdom since 1963 as a treatment for major mood disorders, OCD, severe chronic anxiety states, and a variety of other psychiatric disorders. The aim of the procedure is to interrupt white matter tracts between orbital cortex and subcortical structures by placing a lesion in the region of the substantia innominata just below the head of the caudate nucleus.

The surgical procedure was initially performed using boney landmarks and ventriculography. Target coordinates were calculated as 15 mm from the midline and approximately 10–11 mm above the planum sphenoidale at the most anterior part of the sella turcica. Modern MRI-guided stereotactic techniques have now supplanted these outdated methods with direct visualization of the subcaudate region. Lesions were initially created by implanting two rows of five radioactive yttrium-90 seeds, with resulting lesion volumes in the order of 2 ml, but thermocoagulation has been employed elsewhere.⁸

Complications seen in a series¹⁵ of 208 cases were few but included transient postoperative confusion (10%), postoperative seizures (2%), and undesirable personality traits (7%). Major morbidity has included one case of coma and one case of death from inadvertent destruction of the hypothalamus when an yttrium-90 seed migrated off target.¹⁵

Early reviews⁸ of clinical outcome following subcaudate tractotomy suggested good results in 68% of patients suffering from depression and 50% of patients with OCD. Patients with schizophrenia, personality disorder, drug abuse, or alcohol abuse did poorly. A more recent review¹⁵ of the experience between 1979 and 1991 from the Geoffrey Knight National Unit for Affective Disorders in London found that only 34% of patients were well 1 year after surgery. Vegetative symptoms of depression are less likely to improve than symptoms of anxiety and depressive mood. Some patients had only temporary benefit from their initial lesion and were seen to benefit from a second lesion.¹⁵ Although there has been some criticism of this data because of short follow-up and "idiosyncratic" outcome categories, the overall results seem to parallel those of cingulotomy.

Limbic leucotomy was introduced by Kelly¹⁰ in 1973 and is a combination of the two previously described procedures, anterior cingulotomy and subcaudate tractotomy (Figure 2). Kelly reasoned that these two lesions might lead to a better result for the symptoms of OCD than either lesion alone by disconnecting orbital-frontal-thalamic pathways and lesioning the cingulum. Indications for this procedure have included OCD, chronic anxiety states, and major depression, along with a variety of other psychiatric diagnoses.

This procedure is carried out stereotactically, and three 6 mm lesions are placed in the posterior inferior medial quadrant of each frontal lobe, along with two lesions in each cingulate gyrus. Lesions were created using either cryoprobe or thermocoagulation. Intraoperative stimulation in the subcaudate region, and if pronounced autonomic responses were observed, this was felt to provide physiologic proof of correct lesion location.

Complications include transient confusion and urinary incontinence in the early postoperative period; persistent complaints of lethargy (12%), mild personality changes (7%), and one case of severe permanent memory loss due to inaccurate lesion placement were also noted. No patients developed seizures postoperatively, but given the 2% to 5% incidence following either cingulotomy or subcaudate tractotomy alone, one would expect this to appear in a larger series. Significant cognitive impairment was not observed, and measurements of IQ showed slight improvement postoperatively.³⁴

As with the other procedures, postoperative symptom improvement was not immediate, with a fluctuating but progressive reduction of symptoms over the first postoperative year. In patients with OCD, 89% were clinically improved; in chronic anxiety, 66% were improved; in depression, 78% were improved.³⁴ More rigorous analysis, using validated clinical rating scales and unbiased observers, has not been applied to these data, although more recent reports¹⁶ suggest benefit in smaller percentages of patients.

Although Talairach³⁵ was the first to describe anterior capsulotomy, Leksell³⁶ popularized the procedure for patients with a variety of psychiatric disorders. The goal of anterior capsulotomy is to interrupt frontothalamic connections in the anterior limb of the internal capsule, where they pass between the head of the caudate and putamen. Clinical indications for capsulotomy initially included schizophrenia, depression, chronic anxiety states, and obsessional neurosis.³⁶

Target locations are in the anterior third of the anterior limb of the internal capsule 5 mm behind the tip of the frontal horns, 20 mm lateral to midline at the level of the intercommissural plane. MRI-stereotactic techniques can now directly visualize the anterior capsule, and target coordinates are calculated based upon the individual patient’s anatomy. Lesions have generally been produced by thermocoagulation and should be at 10–12 mm in height and 4–5 mm in width (Figure 3). Occasionally, lesions have been made by radiosurgical means.

Complications of capsulotomy in a series of 115 patients include transient confusion (86%), incontinence (27%), and fatigue (32%). One patient had intracranial hemorrhage and one patient developed seizures. Weight gain is common, with an overall mean weight gain of 10%. No evidence of cognitive dysfunction was reported in more than 200 capsulotomy patients studied using a variety of psychometric tests, although poor memory and slovenly behavior was noted in a few patients.⁹

In a review of all cases of capsulotomy reported in the literature, Mindus17 found sufficient data to categorize outcome in 213 of 362 patients. Of these, 137 (64%) were deemed to have a satisfactory result, with the greatest improvement in those patients with depression and OCD.

Much of the controversy surrounding the use of surgery for intractable psychiatric illness may be attributed to the rather indiscriminate application and high complication rates of the early procedures. Modern advances in surgery and stereotactic techniques have certainly minimized side effects, but the issues of case selection and lesion location remain major considerations. One valid criticism of psychosurgery is that the theoretical basis of surgical intervention for the treatment of psychiatric illness is not well-established. Even though the scientific evidence implicates the limbic system and its interconnections in the pathophysiology of major psychiatric disorders, the neuroanatomical and neurochemical basis of emotion in health and disease remains undefined. It is possible that as the understanding of the pathophysiology of severe psychiatric illness improves, the medical community and society at large will become more accepting of these procedures.

Critics of psychosurgery also suggest that surgery may often be performed before other alternative therapies are sufficiently tried. Currently, only patients with chronic, severe, and disabling psychiatric illness who are completely refractory to all conventional therapy are considered for surgery. This implies that well-documented systematic trials of pharmacologic, psychologic, and, when appropriate, ECT have been tried, both singly and in combination, before neurosurgical intervention

is considered. Only when a patient has failed to respond to all available and appropriate therapies is surgical intervention deemed suitable as a salvage procedure. Although the number of psychosurgical procedures performed in the world today is unknown, it is estimated that fewer than 25 patients are operated upon annually in the US and Great Britain, while only 1–2 patients/year undergo psychosurgery in Australia; therefore, claims that psychosurgery is overutilized seem greatly exaggerated.

There was a time when any patient with a severe psychiatric illness might be considered a candidate for surgical intervention, with the only major criterion being that the patient be in a "fixed state of tortured self concern."⁵ This global inclusion criterion created a very heterogeneous subject population, which made comparison of outcomes difficult. The lack of accurate psychiatric diagnosis also made it impossible to predict outcome based on clinical syndromes. To be fair to practitioners of the past, psychiatric diagnosis was less well-defined, and certain diagnoses, such as OCD, did not even exist as distinct clinical entities. It is now clear that the indications for psychosurgery are much more restrictive. There is general agreement that patients with major affective disorder, chronic anxiety states, and OCD are the best candidates for surgery. These patients must meet DSM-IV criteria for the respective diagnostic categories. Despite the fact that the majority of patients who underwent psychosurgery in the past were diagnosed with schizophrenia, schizophrenia is not currently considered an indication for surgery. Personality disorders or psychoactive substance use disorder are significant relative contraindications to surgery.

Appropriate selection of patients for surgery remains a major issue and the responsibility of the psychiatrist, guided by the informed and expert opinions of the other members of the psychosurgical team. Ethical objections about the use of psychosurgery have been addressed in all centers by insuring an informed consent from the patient and family without coercion, along with unanimous agreement among the referring and treating physicians.

Since the scientific rationale for psychosurgery is lacking, observations and conclusions regarding its use have been largely empirical and accumulated over many years from a variety of institutions. Although many psychosurgical techniques have been used in the past, four procedures—anterior cingulotomy, subcaudate tractotomy, limbic leucotomy, and anterior capsulotomy—have evolved as the safest and most effective. These procedures involve lesions of limbic or paralimbic territories or interruptions of their connections with deeper brain structures, and are performed bilaterally under modern stereotactic conditions to allow for precise identification and accurate lesioning of the target structures. Historically, all psychosurgical procedures have been directed at some component within this system; therefore, some authors prefer the term limbic system surgery to the term psychosurgery. With currently available data, it is impossible to determine whether there is one optimal surgical technique or strategy. All modern procedures seem to be safe and well-tolerated with relatively few side effects or complications.

Subjective evaluations of outcome in the past have allowed critics of psychosurgery to challenge the overall results with some validity. Given the difficulty of assessing outcome in functional, mental, and behavioral disorders, the criteria for cure or significant improvement are not clear and have never been universally agreed upon. There also exist many obstacles that prevent a direct comparison of results across centers, including diagnostic inaccuracies, nonstandardized presurgical evaluation tools, center biases, and varied outcome assessment scales. However, if a global outcome rating of symptom free or much improved is considered a satisfactory response, then in a recent review³⁷ of modern neurosurgical procedures on a series of patients with OCD, cingulotomy was effective in 56%, subcaudate tractotomy was effective in 50%, limbic leucotomy was effective in 61%, and capsulotomy was effective in 67%. In patients with major affective disorder, cingulotomy was effective in 65%, subcaudate tractotomy was effective in 68%, limbic leucotomy was effective in 78%, and capsulotomy was effective in 55%.

Based on these methods of comparison, the clinical superiority of any one procedure is not convincing, although there is a suggestion that anterior capsulotomy and limbic leucotomy may be slightly more effective in patients with OCD. Cingulotomy is most commonly used in the US, whereas in Europe, capsulotomy and limbic leucotomy are more prevalent. They all appear roughly equivalent from a therapeutic standpoint, but in terms of unwanted side effects, cingulotomy appears to be the safest of all procedures currently performed.

Another criticism of psychosurgery has been that no matter which structure in the limbic system is chosen for ablation, the clinical outcome appears similar. This lack of specificity is not necessarily evidence against its use, since many psychotropic agents and even ECT can have benefit across a wide range of mental disorders. It more likely underlines our incomplete understanding of the neurobiological basis of psychiatric disease and our failure to perform careful, controlled trials comparing different surgical techniques within the same center.

Many psychiatrists would readily accept an empirical treatment if the available clinical data supporting a favorable outcome were substantial and convincing. To date, few prospective, long-term follow-up trials have been completed, but these trials appear to support the contention that even when using the most stringent outcome criteria, cingulotomy and anterior capsulotomy are helpful in 25% to 50% of patients with intractable OCD. Using less stringent outcome criteria, improvement is seen in 50% to 70% of patients,³⁵ which is similar to earlier studies.

While controversy exists regarding the exact choice of surgical procedure to be employed, there is unanimous agreement that the presurgical evaluation must be performed by committed multidisciplinary teams with expertise and experience in the surgical treatment of psychiatric illness. Diagnosis based on a formal classification scheme is essential. Although it is impossible to mandate across all centers, prospective trials employing standardized clinical instruments with long-term follow-up are needed. Comparisons of preoperative and postoperative functional status (in addition to target psychiatric symptoms) remain an important parameter in characterizing outcome. The operation is not curative and should only be considered as one aspect in the overall management of these patients. All centers with experience emphasize the importance of postoperative rehabilitation and the need for ongoing psychiatric follow-up. It appears that many patients are greatly improved after surgery and the complications or side effects are few.

Despite the advent of new, effective psychopharmacologic agents, it is generally felt by centers employing this form of surgical intervention that psychosurgery remains an important therapeutic option for disabling psychiatric disease and is probably underutilized; however, caution regarding its use must be urged to ensure that the overzealous and indiscriminate application of this form of therapy never recurs. Hopefully, as our understanding of the neurobiological basis of psychiatric illness improves, the rationale and theoretical basis for surgical intervention will become apparent. Until then, only carefully controlled, prospective long-term follow-up studies by independent observers can improve our empirical assessment of psychosurgery.

Surgical intervention can be helpful in certain patients with severe, disabling, and treatment refractory psychiatric disease, including major affective disorders, OCD, and chronic anxiety states. Surgical interventions should only be carried out by an expert multidisciplinary team with experience in these disorders. Surgery should be considered as one part of an entire treatment plan and must be followed by an appropriate psychiatric rehabilitation program. Many patients are greatly improved after surgery, and the complications or side effects are few. Surgical intervention remains an important therapeutic option for disabling psychiatric disease and is probably underutilized.

1. Tooth JC, Newton MP. Leucotomy in England and Wales 1942-1954. Reports on public health and medical subjects. No 104. London, United Kingdoms: Her Majesty's Stationary Office; 1961.

2. Moniz E. Prefrontal leucotomy in the treatment of mental disorders. Am J Psychiatry. 1937;93:1379-1385.

3. Fulton JF, Jacobsen CF. Fonctions des lobes frontaux; etude comparee chez l’homme et les singes chimpanzes. Proceedings of the International Neurological Congress. London; 1935:552.

4. Valenstein ES. Great and Desperate Cures. The Rise and Fall of Psychosurgery and Other Radical Treatments for Mental Illness. New York, NY: Basic Books; 1986.

5. Freeman W, Watts JW. Psychosurgical in the Treatment of Mental Disorders and Intractable Pain. Springfield, [MASS?]: Charles C Thomas; 1950.

6. Freeman W. Transorbital leucotomy. Lancet. 1948;2:371-373.

7. Foltz EL, White LE Jr. Pain relief by frontal cingulotomy. J Neurosurg. 1962;19:89-94.

8. Knight GC. The orbital cortex as an objective in the surgical treatment of mental illness. The development of the stereotactic approach. Br J Surgery. 1964;51:114-124.

9. Herner T. Treatment of mental disorders with frontal stereotactic thermal lesions. A follow-up study of 116 cases. Acta Psychiatrica Neurolog Scand. 1961;158:36.

10. Kelly D, Richardson A, Mitchell-Heggs N. Technique and assessment of limbic leucotomy, In: Laitinen LV, Livingston KE, eds. Surgical Approaches in Psychiatry. Baltimore, Md: University Park Press; 1973:165-173.

11. National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research. Report and Recommendations: Psychosurgery. Department of Health and Human Services. Pub No (OS) 77-002. Washington DC: US Government Printing Office; 1979.

12. Vasko T, Kullberg G. Results of psychological testing of cognitive functions in patients undergoing stereotactic psychiatric surgery. In: Hitchcock ER, ed. Modern Concepts in Psychiatric Surgery. Amsterdam: Elsevier Biomedical Press; 1979:303.

13. Corkin S, Twitchell TE, Sullivan EV. Safety and efficacy of cingulotomy for pain and psychiatric disorders. In: Hitchcock ER, Ballantine HT, Myerson BA, eds. Modern Concepts in Psychiatric Surgery. Amsterdam: Elsevier; 1979:253-272.

14. Spangler, WJ, Cosgrove, GR, et al. Magnetic resonance image-guided stereotactic cingulotomy for intractable psychiatric disease. Neurosurgery. 1996;38:1071.

15. Hodgkiss, AD, Malizia, AL, Bartlett, JR, et al. Outcome after the psychosurgical operation of stereotactic subcaudate tractotomy, 1979-1991. J Neuropsychiatry. 1995;7:230.

16. Poyton AM, Kartsounis LD, Bridges PK. A prospective clinical study of stereotactic subcaudate tractotomy. Psychological Med. 1995;25:763-770.

17. Mindus P, Rasmussen SA, Lindquist C. Neurosurgical treatment for refractory obsessive-compulsive disorder: implications for understanding frontal lobe functions. 1994;6(4):467-477.

18. Papez JW. A proposed mechanism of emotion. Arch Neurol Psychiatry. 1937;38:725-743.

19. McLean PD. Some psychiatric implications of physiologic studies on the frontotemporal portion of limbic system. Electroenceph Clin Neurophysiol. 1952;4:407-418.

20. Laitinen LV. Emotional responses to subcortical electrical stimulation in psychiatric patients. Clin Neurol Neurosurg. 1979;81:148-157.

21. Talairach J, Bancaud J, Geier S. The cingulate gyrus and human behaviour. Electroencephalogr Clin Neurophysiol. 1973;34:45-52.

22. Hutchison WD, Davis KD, Lozano AM, et al. Pain-related neurons in the human cingulate cortex. Nat Neurosci. 1999;2(5):403-405.

23 Hess WR. Diencephalon: Autonomic and Extrapyramidal Functions. New York, NY: Grune & Stratton; 1954.

24. Weilburg JB, Mesulam MM, Weintraub S, et al. Focal striatal abnormalities in a patient with obsessive compulsive disorder. Arch Neurol. 1989;46:233-236.

25. Luxenberg JS, Swedo SE, Flament MF, et al. Neuroanatomical abnormalities in obsessive compulsive disorder detected with a quantitative x-ray computed t tomography. Am J Psychiatry. 1988;145:1089-1093.

26. Mayberg HS, Brannon HK, Mahurin RK, et al. Cingulate function in depression: a potential predictor of treatment response. Neuroreport. 1997;8:1057-1061.

27. Bonne O, Krausz Y, Bocher M, et al. Increased cerebral blood flow in depressed patients responding to electroconvulsive therapy. J Nucl Med. 1996;37:1075-1080.

28. Rauch SL, Jenike MA, Alpert NM, et al. Regional cerebral blood flow measured during symptom provocation in obsessive-compulsive disorder using 15-O-labelled CO2 and positron emission tomography. Arch Gen Psychiatry. 1994;51:62-70.

29. Baxter LR, Schwartz JM, Phelps ME, et al. Reduction of prefrontal glucose metabolism common to three types of depression. Arch Gen Psychiatry. 1989;46:243-250.

30. Baxter LR Jr, Schwartz JM, Bergman KS, et al. Caudate glucose metabolic rate changes with both drug and behavior therapy for obsessive compulsive disorder. Arch Gen Psychiatry. 1992;49:681-689.

31. Stagno SJ, Smith ML, Hassenbusch SJ. Reconsidering "psychosurgery" issues of informed consent and physician responsibility. J Clin Ethics. 1994;5:217-223.

32. Ballantine HT, Giriunas IE. Treatment of intractable psychiatric illness and chronic pain by stereotactic cingulotomy. In: Schmidek HH and Sweet WH, eds. Operative Neurosurgical Techniques. New York, NY: Grune & Stratton; 1982:1069-1075.

33. Ballantine HT, Bouckoms AJ, Thomas EK, et al. Treatment of psychiatric illness by stereotactic cingulotomy. Biol Psychiatry. 1987;22:807-819.

34. Mitchell-Heggs N, Kelly D, Richardson A, McLeigh J. Further exploration of limbic leucotomy. In: Hitchcock ER, Ballantine HT Jr., Meyerson BA, eds. Modern Concepts in Psychosurgery. Amsterdam: Elsevier/North-Holland Biomedical Press; 1979.

35. Talairach J, Hecaen H, David M. Lobotomie prefrontale limitee par electrocoagulation des fibres thalamo-frontalis leur emergence du bras anterior de la capsule interne. Proceedings of the 4th Congress Neurologique Internationale. Paris: Masson; 1949:141.

36. Leksell L. A stereotaxic apparatus for intracerebral surgery. Acta Chir Scand. 1949;99:229-233.

37. Cosgrove GR, Rauch SL. Psychosurgery. In: Gildenberg PH, ed. Functional Neurosurgery. Philadelphia, Pa: W. B. Saunders Co; 1995.

• Dr. Cosgrove is associate professor of surgery at Harvard Medical School and attending neurosurgeon at Massachusetts General Hospital in Boston, MA.
• Disclosure:The author does not have an affiliation with or financial interest in a commercial organization that might pose a conflict of interest.

Take the Quiz		Back to CME Topics