The Shed Where Lithium Was Found

Melbourne, 1948, in a disused kitchen

John Cade was thirty-six when he came home from the war. He had spent three and a half years as a prisoner in Changi after the fall of Singapore, working as a camp doctor, watching several thousand men closely and under conditions no ethics committee would ever approve. What he took from it was a conviction: that some of the mental disturbance he saw was physical in origin, and that the psychiatry of his day, which had almost nothing to offer people with mania, was looking in the wrong place.

In 1948 he was Senior Medical Officer at the Repatriation Mental Hospital at Bundoora, on the outskirts of Melbourne. He had no research budget, no assistants and no laboratory. He worked in a disused kitchen on the hospital grounds, in his own time, keeping his experimental animals nearby and his reagents in a small refrigerator.

A toxin in the urine, and a great many guinea pigs

Cade’s theory was that mania was caused by an excess of some substance the body produced, and that depression was a deficiency of the same thing. If such a substance existed, it should be excreted. So he collected urine from patients with mania, from patients with depression and schizophrenia, and from healthy people, and injected it into the abdomens of guinea pigs.

Urine from patients with mania killed the guinea pigs at lower volumes than urine from anyone else. Encouraged, Cade set about finding the culprit. Urea turned out to be the lethal component, but urea concentration did not differ between the groups, which was awkward. He wondered whether uric acid modified urea’s toxicity, and here he hit a mundane practical problem. Uric acid is barely soluble in water.

To get it into solution he needed a soluble urate salt. The most soluble one available to him was lithium urate. He used it because it dissolved.

Animals that were calm, not asleep

The guinea pigs given lithium urate did not die as expected. Cade then tested lithium carbonate on its own, and observed something he had not gone looking for. The animals became placid. Turned on their backs, they lay still and looked back at him rather than righting themselves in a panic. They were not unconscious. They were, in the word he used, unresponsive.

A modern reader should be sceptical here, and Cade’s later reviewers have been. Lithium in those doses probably made the animals generally unwell rather than specifically calm, and his reasoning about urea and uric acid was wrong from beginning to end. This is one of the clearer cases in medicine of the right conclusion reached through the wrong argument.

He took lithium citrate and lithium carbonate himself first, for some weeks, to check that the doses were tolerable. Then he gave it to patients.

The paper that almost nobody read

Cade treated ten patients with mania, six with schizophrenia and three with melancholia. Lithium did nothing useful for the last two groups. In mania the effect was unmistakable.

His first patient was a man in his fifties who had been in a state of chronic manic excitement for years and was regarded as a permanent resident of the hospital. Within three weeks he was settled. Within a few months he had left, gone home and returned to his old job. Cade published the results in the Medical Journal of Australia in September 1949, under the title Lithium Salts in the Treatment of Psychotic Excitement. It was five pages long, single-authored, and described one of the most consequential findings in the history of psychiatry.

Why the discovery sat on a shelf

The timing could hardly have been worse. In the same year, 1949, lithium chloride was being sold in the United States as a salt substitute for patients on low-sodium diets, mostly people with heart disease. Several were poisoned, and some died. Lithium acquired a reputation as a dangerous industrial chemical almost exactly as Cade was reporting that it was a treatment.

It did not help that the finding came from a single doctor, in a small hospital, in a journal read almost entirely within Australia, describing an uncontrolled series of ten patients, using a naturally occurring salt that nobody could patent and therefore nobody had a commercial reason to promote.

What rescued it was careful work by Mogens Schou and colleagues in Denmark, who ran placebo-controlled trials through the 1950s and later established that lithium prevented episodes rather than merely treating them. Blood level monitoring made safe dosing possible. The United States did not approve lithium for mania until 1970, twenty-one years after Cade’s paper.

Still here, seventy-five years on

Lithium remains a first-line long-term treatment for bipolar disorder in essentially every major guideline. Newer medicines have arrived, some of them easier to use, and none has clearly displaced it for preventing relapse. It needs blood tests, attention to kidney and thyroid function, care in pregnancy and caution with dehydration and with several common drugs. Prescribed and monitored properly, it is one of the more reliably effective treatments in medicine. It is also worth separating the monitoring a medicine needs from the question of whether it is addictive, which the guide to dependence on psychiatric medication takes apart properly.

Its reputation for reducing suicide risk is more contested than is usually admitted. Meta-analyses of older randomised trials and a considerable body of observational data point that way, but a large randomised trial in United States veterans, reported in 2022, did not find the benefit and was stopped early. It is a genuinely open question, and it is worth knowing that it is open.

As for the trivia that everyone reaches for: 7 Up did contain lithium citrate when it was launched in 1929, and the lithium was gone by around 1950. Lithia spring waters were sold for decades on vague promises about nerves and gout. The popular version of that story is tidier than the record, but the outline is true. A salt that had been drifting around the margins of medicine for a century turned out to work, and it was found by a man in a converted kitchen who was chasing the wrong idea.