The Water from the Spring

I

On 29 March 1948, in a psychiatric hospital on the outskirts of Melbourne, a doctor gave a fifty-one-year-old man a salt of a metal that is of no use whatever in the human body.

The patient was called Bill Brand [in the published papers he appears as W.B.], he was an ex-serviceman and he had had a manic condition for thirty years. Five of them, admitted without leaving. In his clinical record he is described as the most troublesome patient on the ward: restless, dirty, destructive, into everything.

Within five days he was calmer. Within three weeks, according to the doctor, he was another person [“as normal as anybody”, he noted]. Within three months he was discharged and went back to work.

The doctor was called John Cade, he worked alone, in a kitchen converted into a laboratory, with equipment he paid for out of his own pocket. He had just come back from three and a half years in a Japanese prison camp —where he spent the time observing other prisoners— and he was convinced that mental illnesses had a chemical cause.

His hypothesis was wrong. His treatment worked.

Both things at once, with no contradiction. That happens more than popular accounts admit: what makes a treatment useful is that the effect repeats, not that the explanation is correct. The explanation comes later and sometimes never comes.

II

It has to be told how he got there, because it is the best example in the series of a right answer reached by the wrong road.

Cade thought that mania was caused by a toxic substance the body itself made and that was got rid of through the urine. To check it, he injected urine from manic patients into guinea pigs and saw that it killed them [something almost anybody's urine does, because of the ammonia].

Then he wanted to try uric acid, which does not dissolve well, so he mixed it with a lithium salt in order to be able to inject it. The guinea pigs in that group went calm and placid.

What he had found had nothing to do with what he was looking for.

He then tried the reverse order, which is what distinguishes a researcher from a lucky man. If lithium calmed the guinea pig, the question was whether it calmed on its own or whether the rest of the mixture was needed; and, if it did it on its own, how much a human body withstood. He took it himself for days —noting pulse and discomforts— before giving it to anybody. He tried lithium alone, in guinea pigs; then on himself, for days, to see whether it was toxic; and then on ten patients.

With Brand it worked in a way that had no precedent.

III

I have taken it without knowing, like everybody, in water.

Lithium is spread through the earth's crust and appears in minuscule quantities in many mineral waters. In some springs, considerably more. In the nineteenth century whole spas were sold on that, and there was a fizzy drink, of the kind that still exists, that had lithium in the formula and had it in the name. That industry was not lying entirely and was not curing either: the substance was the right one and the quantity was not.

I bought a bottle of a water that boasts of its content: one euro eighty, a litre and a half, tasting of nothing with a mineral undertone [and a label with the analysis in minute lettering].

The sum is easy to do and it disarms you. To reach the dose somebody with bipolar disorder takes you would have to drink several thousand litres a day. Not a few litres: thousands.

That is all I can contribute with my body, and I set it down because the alternative would be inventing an experience I have not had.

What I can describe is the act of buying it. The bottle was on a shelf of spa waters, among others that boast of magnesium and of bicarbonate, with an offer sign of three for five euros. I bought it, drank it in two days and nothing happened, which was the expected result [and the only one the arithmetic allowed].

What I have seen is the part of this treatment nobody tells. The blood tests. Whoever takes it has blood checks every few months, for life, because the difference between the dose that works and the one that poisons is extremely narrow. A whole life organised around a laboratory number. The cheapest drug in the file is also the one that consumes the most appointments [blood test, adjustment, thyroid and kidney checks, year after year].

That, in a system that counts spending by prescription and not by result, makes it look expensive where it is not and cheap where it should not be. A treatment that costs pennies and demands constant monitoring is, for health accounting, more awkward than an expensive one that is taken and forgotten.

IV

Inside it is the most baffling drug in the file, because after seventy-five years we still do not properly know what it does.

What goes in is an atom, not a molecule: a small ion with a single positive charge, similar in size to sodium and potassium [the two the body uses all the time], to make nerve cells work.

That similarity is the key. Lithium enters through the same channels sodium enters by, slips into the neurons and stays inside, because the pumps that expel sodium handle it badly.

Once inside, it gets in the way in several places at once. It interferes with enzymes that take part in the recycling of internal signals, slows a chain that regulates the biological clock and the stress response, and modifies the balance between excitation and braking in circuits known to work badly in this illness.

None of that is a clean explanation. It is a list of places where it has been seen meddling.

What is clear is the narrowness. Below a certain concentration it does nothing; a little above it produces tremor, thirst, copious urine and diarrhoea; and considerably above it poisons the kidney and the nervous system severely. The whole margin fits within a factor of two.

That is why lithium is the only ordinary treatment that obliges you to measure its own quantity in the blood for ever. It is not adjusted by symptoms: it is adjusted by test.

V

What this case teaches is not that a metal cures mania.

It is that the cheapest and most effective molecule in modern psychiatry was discovered neither by an industry nor by a research programme. It was found by a psychiatric hospital doctor, alone, following a false hypothesis, with guinea pigs paid for out of his salary.

For that very reason there is almost nobody behind it to defend it. Lithium is not patented, costs pennies, and no company has any incentive to promote it over more expensive drugs; it is a carbonate, it comes out of a mine and there is nothing to license.

Of the medicines that appear in this series, this is the only one whose problem is that it makes nobody any money.

VI

The objection is supplied by the case itself.

Bill Brand was discharged, went back into the community and stopped taking it; he relapsed, was readmitted and died in 1950, with signs of lithium toxicity. The first successfully treated patient died, two years later, of the same treatment that had saved him.

Cade did not publish that with the same clarity with which he published the first part.

There is something else that obliges me to lower my voice: the reason this drug took twenty years to be used in many countries was not psychiatric scepticism. It was that in 1949, in the United States, lithium salts sold as a substitute for table salt killed several people, and the name was left marked.

The same substance, in the same year, saving in Australia and killing in America.

VII

Cade went on working in the same hospital, patented nothing, did not get rich and left Australia only to give the occasional lecture. He died in 1980.

Lithium remains, seventy-five years later, the treatment that best prevents relapses in bipolar disorder and the only one with solid evidence that it reduces suicide [which is the technical way of saying that it keeps people alive].

29 March 1948, first dose.

Five days until the change; three months until discharge.

A factor of two between the dose that works and the one that poisons.

Blood tests every few months, for life.

Bottle of mineral water with lithium: one euro eighty.

Several thousand litres a day to reach a therapeutic dose.