Why Did Lobotomy Win a Nobel Prize? Malaria Fever, Insulin Coma and Electroshock, 1917–1958

This is the fourth part of a series on the history of psychiatry. Part 3 followed Freud, who looked for causes in the memories his patients told him. In the same decades other physicians came at mental illness from the opposite side. Give the patient a fever. Put the patient into a coma. Set off a convulsion. Cut part of the brain. Four treatments that acted hard on the body appeared within about twenty years, and two of them won Nobel Prizes. This part sets each one beside the comparisons that were made later, to see which of them worked.

This series is a history of psychiatry. It is not advice on diagnosis or treatment. If you are struggling with your mental or physical health, please talk to a doctor or a support service. The text includes names of illnesses and expressions that are no longer in use; they are given as they appear in the sources of the time.

A disease with a known cause and no cure

The story starts with general paresis, the illness of Part 2. The organism of syphilis reaches the brain, and within a few years mind and body fail and the patient dies. In 1913 Hideyo Noguchi and a colleague found the organism in patients' brains, and the cause was settled. A cure was not.

The Viennese psychiatrist Julius Wagner-Jauregg looked back on those years in his Nobel lecture of 1927. Every textbook said that mercury, the drug then used against syphilis, was useless in paresis. Yet nobody had checked properly. A demand had been made in 1912 that a large series of untreated cases be compared with a large series of treated ones. It was, he said, nowhere put into effect.

A tally finally appeared in 1914. Among 555 patients, the share whose symptoms eased for a time was 3.9 percent in those left untreated, and 11.4 percent in those given mercury. Left alone, fewer than one patient in twenty-five improved. That figure becomes the yardstick for what follows.

1917: giving patients malaria on purpose

Wagner-Jauregg took up an old observation. In the rare cases where paresis had improved, the patient had often just been through an infectious disease with a high fever. Why not bring on such a disease deliberately? He proposed this in 1887 and named malaria as a candidate, because malaria could be stopped with quinine.

He acted on it thirty years later. In 1917 he infected nine patients with malaria. Six showed an extensive easing of their symptoms. In three of those six the recovery lasted: they went back to the work they had done before falling ill and were still doing it ten years on. In the other three the improvement did not hold. And three of the nine showed no clear benefit. Even so, this was a disease in which 3.9 percent improved untreated. The numbers were tiny, but the difference was large.

It was also dangerous. The fever passed 40 °C and often reached 41 °C. Wagner-Jauregg said in his lecture that this was frequently too much for the weakened heart of a paretic patient, and that deaths during the fever or just after it were "not by any means infrequent." He then described how changes in the dose and in the use of quinine had reduced the danger.

He made a comparison of his own as well. Patients arriving at his clinic were assigned alternately: one to malaria alone, the next to malaria followed by an anti-syphilitic drug. Symptoms cleared completely in 48.5 percent of those given the drug and in 25 percent of those who were not.

Why it worked he did not know. It was certain, he said, that the high temperature alone was not the answer: in unsuccessful cases the organisms vanished from the brain during the fever and were found there again as soon as it passed. The explanation, he concluded, was a problem for future research. The treatment went out of use once penicillin, which acts on syphilis itself, became available.

The effect was checked. The reason was not understood. And the success spread an idea: that mental illness might be cured by acting forcefully on the body.

Insulin coma: a report that nearly nine in ten improved

The next treatment was for schizophrenia. In 1933 Manfred Sakel, a physician in Vienna, reported his results with insulin, the hormone that lowers the sugar in the blood. A large dose starves the brain of sugar and the patient sinks into a coma. After about three hours the patient was brought round with sugar. This was repeated every day except Sunday, about thirty times.

Sakel reported that of more than a hundred patients who had fallen ill recently, 88 percent improved clearly and 70 percent recovered completely. The treatment spread quickly through Europe and North America. Reports of the time put the share of patients who died of it at between 0.5 and 4.5 percent.

There were doubters from the beginning. According to a 2022 study by Robert Freudenthal and Joanna Moncrieff of University College London, the leading British textbook pointed out in its 1940 edition that Sakel's figures came from outpatients, an earlier and more recoverable group than patients ill enough to be admitted. The 1944 edition said there was only one sound type of experiment: take two closely similar series of cases and treat one of them with insulin. The treatment went on regardless.

In 1953 a junior doctor, Harold Bourne, published a paper in The Lancet called "The insulin myth." The studies were small, he argued, they had no proper comparison group, and those who judged the results knew who had been treated. If patients did improve, it might be due not to the insulin but to the intensive care that came with it. Of the fourteen letters the journal received, nine took issue with him.

A trial in 1957 gave the answer. Brian Ackner and colleagues in London allocated fifty patients at random to two groups. One was put into a coma with insulin. The other was made unconscious in the same way with a sleeping drug. The results did not differ. The authors were restrained: no conclusion could be drawn about the value of the coma regime as a whole, they wrote, but insulin did not seem to be the specific therapeutic agent.

Freudenthal and Moncrieff also show that the treatment was not thrown out overnight by the trial. A new drug that appeared in the same years was simply easier to give, and in their words insulin was "quietly forgotten." That drug is the subject of the next part.

Convulsions: a treatment meant for schizophrenia that worked in depression

The third treatment was the convulsion. The Hungarian physician Ladislas Meduna examined the brains of patients with epilepsy and with schizophrenia, and thought that the supporting cells of the brain, the glia, were more abundant in one and sparser in the other. Perhaps the two diseases worked against each other. If so, bringing on epileptic fits in a patient with schizophrenia might help.

On 23 January 1934 Meduna used a drug to induce a seizure in a man of thirty-three diagnosed with catatonic schizophrenia, a form in which movement and response can almost stop. After five more seizures in three weeks, the report says, the psychosis was relieved. Meduna later described the recovery of more than half of 110 patients given drug-induced seizures.

In Rome, Ugo Cerletti heard of Meduna's method and thought of using electricity in place of the drug. He had been producing electric fits in dogs for years. But dogs sometimes died when the current crossed the heart, and he hesitated to try it on a person. So he went to the slaughterhouse in Rome and measured, on pigs, how much current caused a convulsion and how much it took to kill. The gap between the two was very wide. He decided the method would be safe in man.

The first patient was chosen in 1938: a man from northern Italy who spoke only an incomprehensible jargon and could not give his name. Seventy volts for a tenth of a second. There was no full convulsion, and the patient burst into song. Cerletti proposed a higher voltage. His assistants objected. At that moment the patient sat up and said in clear Italian: "Not a second one! It will kill me!" Cerletti gave the order to proceed, and the first electrically induced convulsion in a human followed. The scene was written down in 1960 by the American physician David Impastato, as told to him by Cerletti. Nothing in the account speaks of the patient's consent.

This treatment then took an unusual path. It began as a treatment for schizophrenia, but severe depression soon became its main use. And of the four, it alone stood up to a strict comparison. In 2003 a British review group pooled six trials, 256 patients in all, that compared real electroconvulsive therapy with a simulated one: the same procedure, anaesthetic included, but with no current passed. Depressive symptoms eased clearly more with the real treatment. The reviewers concluded that it is an effective short-term treatment for depression. Electroconvulsive therapy today is given under anaesthesia.

Nothing in the idea that two diseases cancel each other predicts an effect on depression. The treatment survived. Its original reason did not.

1935: Moniz cuts the white matter of the brain

The fourth treatment was lobotomy. The Portuguese neurologist Egas Moniz was already known around the world for inventing a way to show the blood vessels of the brain on X-ray. In 1935, at a congress in London, he heard a report by the American physiologist John Fulton and a colleague. Chimpanzees whose frontal lobes, the part of the brain behind the forehead, had been removed no longer flew into a rage when they failed at a task.

Moniz reasoned that mental illnesses marked by strong emotional tension might be relieved by cutting the links between the frontal lobes and the rest of the brain. That November, working with the surgeon Almeida Lima, he carried out the first operation. At first alcohol was injected into the frontal lobes. Later a special instrument was used to cut out several small cores, on each side, of the white matter, where the nerve fibres run in bundles. Because it cut white matter, Moniz named the operation "leucotomy." "Lobotomy" is the name it later took in the United States.

Of his first twenty patients, Moniz reported that seven were cured, seven improved and six were unchanged. There was no comparison group. His 1937 paper in an American journal ended like this: the facts speak for themselves; the recoveries have been maintained; this is "a simple operation, always safe," which may prove to be an effective surgical treatment in certain cases of mental disorder.

The operation spread. In the United States the neurologist Walter Freeman and the surgeon James Watts developed their own version, and about ten thousand operations had been done in the country by August 1949. In England and Wales more than ten thousand people were operated on between 1942 and 1954.

Why did it go so far? In a 1998 article for the Nobel Foundation's website, the Swedish psychiatrist Bengt Jansson listed the circumstances. There was no other treatment for patients who had been in hospital for years. In 1943, American mental institutions admitted 100,000 new patients and discharged 67,000. The hospitals were overflowing.

In 1949 Moniz received the Nobel Prize in Physiology or Medicine. The presentation speech said that among patients with depression or obsessions "the great majority" of those operated on had recovered and become capable of working. For schizophrenia the prospects were less favourable, but quite a few patients could be released; and where they could not, nursing was much simplified, because after the operation the patient could be kept in a "quiet" ward. The speech also acknowledged that personality changed after the operation. For people already complete invalids through illness, it said, this might be of small importance, while in other cases the decision to operate had to be weighed very strictly.

Families saw something else. According to Jansson, the Swedish psychiatrist Gösta Rylander reported in 1948 what one mother said: "She is my daughter but yet a different person. She is with me in body but her soul is in some way lost."

The change in personality was not something learned only later. According to Jansson it had been noticed by the end of the 1930s. He quotes an American report of 1949 on schizophrenic patients after the operation. They were no longer distressed by their mental conflicts, it said, but they also seemed to have little capacity left for any emotional experience, pleasant or otherwise. Nurses and doctors described them over and over as "dull, apathetic, listless, without drive or initiative," childlike and dependent.

There was physical harm as well. Figures presented at a conference in 1953 put the death rate from the operation at between 0.8 and 2.5 percent. About 10 percent of those operated on were known to have problems with epilepsy afterwards.

The changes, in other words, were known. They were judged a small price for people who were severely ill, and the operations went on.

Compared with patients who were not operated on

Studies that compared lobotomy with no operation came only after the operation had spread.

In 1958 A. A. Robin, a physician at Runwell Hospital in England, took 198 patients who had been operated on. To each he matched, from the hospital records, an unoperated patient of the same sex and similar age at admission. The operation had not improved the chances of leaving hospital. It had not reduced readmissions among those who left. In Canada in 1964, K. G. McKenzie and G. Kaczanowski compared 183 patients over five years with a closely similar group who had not had the operation, and found no significant difference in the rate of discharge. Leucotomy, they concluded, does not produce remissions significantly beyond what would be expected without it.

Neither was a randomised trial. Both chose similar patients from records after the fact, so the evidence is weaker than in the insulin trial. Even so, as soon as a comparison group was set beside it, the effect could no longer be seen.

A survey of England and Wales found that of 9,284 patients operated on, 41 percent were rated "recovered or greatly improved." Read alone, that looks like a treatment that works. But without knowing what the figure would have been with no operation, 41 percent has no fixed meaning. It is the problem Eysenck pointed to in Part 3. In the same survey, 4 percent of the patients had died by the time of follow-up.

What had been done, and what had not

First, what had been done. General paresis, a disease whose cause was known, had its first effective treatment. Severe depression had a treatment that is still in use. And it had become thinkable that mental illness could be treated through the body. For doctors in hospitals with nothing else to offer, that was a great change.

What had not been done was telling the treatments apart. All four entered the world in the same way: try the method on a small number of patients, report the share who improved, set no comparison beside it. From those reports alone, the two that would later be confirmed cannot be told from the two for which no difference was found. The exceptions were general paresis, where a figure for untreated patients existed from the start, and electroconvulsive therapy, which was compared with a simulated treatment long afterwards.

The price of not being able to tell them apart differed from one treatment to the next. With insulin coma, some patients died of the treatment itself. With lobotomy, the benefit was never confirmed and the harm, a changed personality, remained. Brain tissue that has been cut does not grow back. A drug that does not work can be stopped. This operation could not be undone.

As for the reasons, none turned out as expected. Fever therapy was never explained. Convulsive therapy worked in a different illness from the one its theory was about. If patients improved on insulin, it was apparently not because of the insulin. Lobotomy had a coherent explanation, and it was the effect that could not be confirmed. The gap seen in Part 3, between having a satisfying explanation and having a tested one, appears here in a form that touched people's bodies and lives.

According to Jansson, what pushed lobotomy out of the treatment of schizophrenia was a single drug that appeared in 1952. The next part takes up that drug and the other medicines for mental illness that were found almost by chance.