(c) Bob Kentridge 1995,1996

S2 Psychopathology: Lecture 2.

Brain surgery and Psychosurgery.

How can brain-surgery alleviate psychopathologies?

In the middle ages it was widely assumed that madness (and migraines) were caused demons trapped inside a sufferer's skull. Their condition could be treated by releasing the demons. To this end patients were trepanned - that is they had holes cut into their skulls. Although we only know of the explanation dating back to the middle ages there is evidence that the practice goes back a lot further - Neolithic skulls with neat trepanning holes in them have been found. Trepanning still has its adherents today, although only in some fairly extreme 'new-age' cults, and not in the orthodox medical professions. It is a relatively safe operation if infection is avoided and causes few side-effects if, as is usually intended, no brain-tissue is damaged. It is also unlikely to have any direct therapeutic effect although there is the very slight possibility that it may have relieved headaches due to peripheral factors like muscle tension or over-pressure of fluid in the brain.

As we shall see, compared with some of the psychosurgical techniques conducted by the medical establishment in their tens of thousands trepanning looks rather benign. This must prompt us to ask what the justifications were for this expansion of brain surgery?

Justifications for brain surgery and psychosurgery.

Last week I described briefly some theories that psychological functions were localised in distinct areas of the brain. If function is anatomically localised and failings in a specific function lead to a psychopathology then it follows that the cause of a psychopathology might be anatomically localised. Surgery which destroys a region of the brain or disconnects it from the rest of the brain clearly cannot correct whatever is amiss in that region. It may, however, be the case that the signals produced by the dysfunctional region interfere with the normal operation of the rest of the brain. If this is the case then there may be a benefit to removing or disconnecting the dysfunctional region.

A good theoretical argument for surgical intervention therefore requires at least three crucial types of evidence:

Perhaps unsurprisingly such evidence is hard to come by. A weaker justification for surgery might be that some sufferers of a pathology had localised areas of their brain damaged for other reasons, perhaps accident, perhaps some other planned surgery, and that after this damage their psychopathologies were observed to improve. There is not necessarily any theoretical justification for the surgery in terms of a biological cause for the psychopathology, although one might later be developed, but there is at least evidence that the patient's suffering might be alleviated.

In the light of these requirement let us examine how various applications of psychosurgery developed.

Ailments 'treated' by brain surgery or psychosurgery.

A range of different psychopathologies have been treated at one time or another using surgery, today I will discuss: Surgical treatment of the first two classes of disorder are usually referred to as psychosurgery while the latter are regarded as brain surgery. The difference in terminology is meant to reflect the extent to which the surgery is treating an identifiable physical disorder as opposed to a psychological one.

Schizophrenia and other psychoses.

Psychosurgery was used in the treatment of schizophrenia for many years. Given our criteria for justifying psychosurgery this is strange - so little is known about the causes of schizophrenia that it is unlikely that it can be attributed to clearly to failure of one anatomically localisable function and, if, for the sake of argument, we imagined that it were due to some localised dysfunction perhaps in reward evaluation or impulsiveness, there is no evidence that complete removal of the function would correct the patients problems.

In fact, the origins of psychosurgery for schizophrenia lie in chance observation, not, however, of a brain-damaging accident alleviating a schizophrenics suffering, but rather that dogs which had had the whole of their neocortex removed were quick to become aggressive whereas dogs which had only had the temporal lobes of their neocortex removed were actually calmer than unoperated animals. These observations were made by Freiderich Golz and reported in 1892. They inspired the supervisor of an insane asylum in Switzerland, Gottlieb Burkhardt, to remove parts of the cortex of patients who were suffering from vivid hallucinations which made them very agitated. Burkhardt's patients became calmer (one became excessively calm, dying from his operation), and, in the light of the prognosis for severely disturbed schizophrenics at the time, this might be seen as some benefit - they were not, however, cured of the schizophrenia or even their hallucinations.

Although Burkhardt was widely opposed by the medical authorities of the time psychosurgery for schizophrenia was tried briefly by others but only on a very small scale until the 1940s. Again the justification arose from work on animals. In the 1920's and 30's Karl Lashley began work on the localisation of learning and memory in the rat brain. Although Lashley's results did not indicate localisation of these functions Lashley and his work encouraged a number of other studies. Heinrich Kluver in collaboration with Paul Bucy at the University of Chicago had shown that some midbrain structures in the limbic system, including the amygdala, were implicated in the control or generation of emotion in man (with evidence from stroke patients and other accidental brain-damage) and animals (with experimental surgery). In Yale Carlyle Jacobsen investigated changes in the behaviour of chimpanzees after destruction of their frontal or prefrontal cortices. These areas of the cortex are most developed in man and, as most other areas of the cortex had been found to subserve specific but basic roles in sensory and motor processes, it was speculated that 'higher' qualities of thought and reason might be localised in these areas. Jacobsen conducted well controlled experiments on his subjects. In one he investigated delayed-response learning in which an animal, having been shown a reward is prevented from seeing or obtaining it for a few moments. This is a test of memory and most animals soon settle into it. Jacobsen reported, however, that one of his animals became very angry when prevented from obtaining the reward immediately. After surgery, however, the animal became calm and could easily tackle the task. Notice that Jacobsen was not even specifically looking at emotional effect in this experiment, he was, however, reporting all significant observations he made which is good practice in animal work where one wants to obtain the maximum benefit from the fewest possible subjects.

Jacobsen's report of his experiences with this one chimp encouraged the Portuguese neuropsychiatrist Egas Moniz to attempt to treat severe mental disorders by removing the prefrontal cortex of patients. His first surgery was performed in 1935 and, as Burkhardt had found, was generally successful in calming patients although not necessarily in reducing their psychosis. Moniz only supervised about 100 operations - one of his lobotomised patients attempted to murder him and left him paralysed with a bullet in the spine. Moniz was awarded the 1949 Nobel Prize for Medicine for his work on lobotomy. I would not want to paint Moniz as the villain of the piece. Another reason that relatively few surgeries took place under him was that he was concerned to evaluate the long-term consequences of the operation before undertaking surgery as a general treatment for patients without extreme conditions. The potential (and actual) side-effects could be devastating. It was known that damage to the frontal lobes could radically alter the personality. In 1848 a railroad worker, Phineas Gage, had been tamping blasting powder into a bored hole when it accidentally ignited sending his 1 and a quarter inch diameter, 13 pound, 3 foot 6 inch long iron tamping rod through his left cheek and out of the front part of the top of his cranium. Gage did not fall unconscious and recovered from his horrific injury. Before the accident Gage had been a diligent worker who was quietly spoken and calm. Once he had recovered from his injury he became shiftless, aggressive and incapable of holding down his previous job although offered it back by his (no doubt slightly guilty) employers.

The observed results of lobotomies included severe loss of motivation and energy, tactlessness in dealing with others, tendencies to short emotional outbursts, problems in organising actions as well as effects on their personality (not usually as extreme as Gage's). It must, however, be taken into account that the alternatives for people with severe schizophrenia were limited and terrible. The most favoured alternative 'medical' therapy was intentionally to induce diabetic comas in patients with injections of insulin. Often the patients were uncontrollable or a danger to themselves and so were left tied in straight-jackets or in padded- cells. Any treatment which calms a patient sufficiently to avoid these other practices must be judged in some sense to be justifiable.

Moniz' surgical technique was quite coarse and a number of more refined precise methods were subsequently developed, in particular in USA by Freeman and Watts. The initial developments were aimed at improving the accuracy of the surgery and involved procedures like opening holes in the front of the cranium so the frontal cortex could be lifted from underneath and localised cuts be made at the desired depth (the further back these cuts were the more radical the lobectomy (it is a lobotomy if the tissue is removed and a lobectomy if it is disconnected from the rest of the brain) and the more profound its effect) or making holes in the side of the head so that the frontal cortex could be disconnected with lateral knife cuts. These surgeries did not necessarily require general anaesthetic and the patient could often talk with the surgeon during the operation. One patient is reported to have replied when asked "what is going though your mind now" by a surgeon carrying out such a lobotomy "a knife".

Perhaps the most profound development in psychosurgery for schizophrenia, however, was the ice-pick or transorbital lobotomy. This procedure was developed on cadavers using an ice pick which was, of course, replaced with a proper surgical instrument for use on patients. The instrument is inserted into the patient's eye-socket above the eye where the skull separating the brain from the eye-socket is quite thin. A sharp tap breaks through allowing the instrument into the frontal cortex which can now be disconnected by moving the instrument from side to side. The great advantage of this technique was that it could be performed without all the rigmarole of more traditional surgery - indeed it was often performed in doctors offices rather than operating theatres. This made it quick and cheap. Following World War II psychiatric hospitals in the USA (and around the world) were overflowing with patients. The pressure of patient numbers encouraged the widespread application of treatments which were often surely inappropriate but were quick and cheap. This is the truly offensive part of the history of psychosurgery (and worth bearing in mind as a general lesson today). Large numbers of patients who were not suffering from extreme distress and who may have recovered from their problems on their own or through rest or counselling were subjected to psychosurgery which permanently effected them. By the mid 1950s these abuses of the method were increasingly recognised, in addition, in 1952 drug- treatments which calmed schizophrenics had become available. The use of psychosurgery in treating psychosis diminished greatly.

Aggressive and antisocial behaviour.

The story of psychosurgery for schizophrenia is depressing, partly because of its abuse and partly because any justification for it in terms of attacking a physical cause of schizophrenia was so tenuous (the best example is probably the theory that schizophrenia was caused by abnormally strong reverberating cycles of neural activity which could be broken by removing part of the tissue making up the cycle - an interesting piece of theory in its day with absolutely no empirical support). In contrast, the work of Kluver, Bucy and others clearly implicated the amygdala, a structure in the midbrain, in emotional and social processes. Moreover, it makes some sense to argue that if a patient has an abnormal amygdala producing inappropriate social and emotional signals he or she might be better of with none of these signals rather than the wrong ones. Despite this justification patients with profound emotional disorders were subjected to lobotomies rather than amygdalotomies until the early 1950s. There were purely practical reasons for this - the amygdala is a relatively small structure deep in the brain - it cannot be removed or disconnected by simple 'hand-held' surgery without grave danger of damaging large amounts of the overlying cortex. The necessary technical development was the stereotaxic surgical technique for humans, in which the positions of brain structures are calculated relative to reference points on the skull revealed by X-ray and these structures are then reached by an instrument positioned by calibrated vernier adjusters. Once in position the instrument, which is quite fine to avoid damaging overlying tissue, can be used to destroy brain tissue electrically or chemically. Stereotaxic destruction of part or all of the amygdalae has been used as a treatment for uncontrollably violent patients since the early 1950's - it is often preceded by recording of brain activity with implanted electrodes. This often reveals that, in these patients, the amygdala appears to be the source of epileptic brain activity. By the late 1960s the pioneers of this work would only operate if signs of this epileptic activity could be found. The results of the surgery vary considerably. A reasonable proportion of patients (about 40% of previously profoundly disturbed patients) become emotionally so much less volatile that they can lead fairly normal lives. The most serious potential side effect is a total loss of the ability to learn new information which accompanies combined lesions of the amygdala and the nearby hippocampus in animals. The danger of this was recognised and these surgeries aim to avoid damaging both structures, usually the whole of both amygdalea is not destroyed, indeed a whole amygdala may be spared. There are, however, other effects such as emotional blunting, reduction in spontaneity, and hormonal imbalances. Again it is a case of weighing up the costs and benefits of the surgery. For uncontrollably violent patients (including those who recognise that they cannot control their own violence) this type of surgery may be the only reasonable option. There were suggestions in the 1960's that antisocial individuals, for example rioters, could be viewed as suffering from a psychopathology which should be 'treated' with this type of psychosurgery - luckily the proposed abuse never came to pass.


It has been recognised since 1817 when James Parkinson published his 'Essay on the shaking palsy' that specific structure in the midbrain are implicated in motor control and its disorders. The causes of Parkinson's disease are now quite well understood as a degeneration of dopamine producing neurons in the substantia-nigra. In the 1950s the anatomical localisation of the cause of Parkinson's disease in the basal ganglia (a set of structures including substantia-nigra) was understood, but its precise mechanism was unknown. Destruction of another basal- ganglia structure - the globus pallidus by stereotaxic surgery had resulted in some improvements in a patient with motor-disorders in 1947. Following that there were reports that surgery which had to be abandoned when the blood supply to the globus pallidus was cut off also fortuitously lead to improvements in the condition of patients with motor disorders. These isolated results lead a number of surgeons to investigate surgery of the basal ganglia more closely as a treatment for motor-disorders. It was recognised that no-one knew which dysfunctional structures might produce which motor-disorder. In response to this techniques were developed by which particular structures could be cooled so that activity in them was temporarily suppressed and the condition of the patient evaluated - the aim was to identify the structure most effecting the patients motor-problem and then destroy it. Unlike the situation of epilepsy in the amygdala, however, there was no evidence of dysfunction in any of the structures and it is unlikely that these surgeries were attacking the cause of the disorder (this is obvious in hindsight) but rather were masking it by producing addition deficits with roughly opposite effects. This approach to treating Parkinson's disease was rapidly abandoned with the introduction of L-dopa therapy which could replace the dopamine which the degenerate substantia-nigra could not produce thereby alleviating the disease. In recent years attempts have been made to permanently augment nigral dopamine production by implanting foetal nigral tissue into the brains of severe Parkinson's sufferers (sometimes control with L-dopa is very hard to achieve in these patients) - although there have been some successes the long- term outcome of this operation is very variable. At present it has been abandoned.


In discussing emotional disorders and the amygdala I mentioned that amygdalar dysfunction responsive to surgery usually appeared to be epileptic. Epilepsy tends to involve the spontaneous production of large highly organised waves of brain activity which disrupt the normal operation of the brain, often to the extent of producing a fit in a sufferer. Two different types of epilepsy can be identified, idiopathic epilepsy where the source of these wave cannot be identified or is highly variable, and focal or Jacksonian epilepsy in which the source or focus of the spontaneous activity is identifiable. Some epilepsy can be controlled by drugs, however, it has been recognised for some time that surgical removal of the focus was potentially of great benefit to profoundly epileptic patients who were not responsive to drugs. In addition, some idiopathic epilepsies have been treated by removing a portion of the lobe of the brain in which they occur. Surgical treatment for focal epilepsy continues to this day. Recordings from the electrodes in the patients brain allow the focus to identified very accurately. There is still a problem in deciding how much tissue to remove around the focus - obviously one does not want to remove normal tissue, however, incomplete removal of the focus may necessitate further surgery. It is also possible that previously normal tissue near an incompletely removed focus will itself degenerate. Recent advance in mathematical modelling of the cortex now make it possible to calculate the size of the focus from the speed and size of the epileptic waves it produces. It is fitting to note that these advance have been made in the same department in Chicago the Kluver worked in and that they were the result of work directly inspired by his own studies on the neural basis of hallucinations.