From guinea pigs to manic patients: Cade’s ‘story of lithium’

I read with interest the letter from Abhishekh and Faizan (2012) on John Cade in the centenary year of his birth, in 2012, and his (re)discovery of lithium’s anti-manic effect (Cade, 1949).

Some years before, Cade (1947) had formulated the hypothesis that manic-depressive illness, by way of analogy, could be likened to thyreotoxicosis and myxoedema: mania being a state of [auto]-intoxication caused by a normal metabolite circulating in excess in the body, and melancholia the corresponding deprivative condition. He now devised a method with which to demonstrate whether such a toxic agent existed in the urine of manic patients. To this end he injected guinea pigs intraperitoneally with fresh urine from manic patients and, in way of control, from normal persons and from melancholic and schizophrenic patients. As the animals died in status epilepticus, urine from the manic patients seemingly the most lethal, he speculated whether ‘the same toxic agent is at work’. In search for such an agent, he decided first to investigate the principal nitrogenous constituents of urine: urea, uric acid, and creatinine, and ‘it was not very surprising to find that urea was the guilty substance’. Ensuing uric acid injections seemed to have a ‘slightly enhancing effect’ on urea’s toxic property, and creatinine a ‘remarkable protective action’. However, quantitative urea and creatinine assays showed that urine from manic patients ‘did not differ significantly from other urine in urea or creatinine content’. Having found though that the toxicity of some urine specimens was higher than could be expected despite the assumed neutralizing effect of creatinine, Cade now postulated ‘a third substance in urine’, which ‘neutralizes the protective action of creatinine against the toxic activity of urine’, but also ‘enhances the toxic effect of urea’.

Cade first mentioned his lithium studies in animals in the 1949 paper. As ‘the great difficulty was the insolubility of uric acid in water, so the most soluble urate was chosen – the lithium salt’.

Injecting the animals with urea saturated with lithium urate Cade observed that ‘the toxicity was far less than expected’, and ‘it looked as if the lithium ion might have been exerting a protective effect’. He now tested whether ‘lithium salts per se had any discernible effects on guinea-pigs’. ‘Large doses […] of lithium carbonate’ were injected. ‘A noteworthy result was that after a latent period […] the animals, although fully conscious, became extremely lethargic and unresponsive to stimuli […] before once again becoming normally active and timid’.

In view of these results, Cade thought it ‘worthwhile’ ‘to try lithium salts in the treatment of two distinct disorders – first mania, in view of their sedative effect, secondly epilepsy, in view of their anticonvulsant action’. As his investigations had been undertaken to demonstrate ‘some possibly excreted toxin in the urine of manic patients’, he emphasized, ‘[this] association of ideas is explicable’. From later papers by Cade (e.g. 1970), we learn that there seemed ‘no ethical contradictions to using [lithium salts] in mania’, and for the reason that his self-ingestion of lithium showed ‘no discernible ill effects’. Not least, he referred to the fact that ‘lithium salts had been in use in medical practice since the middle of the nineteenth century, albeit in a haphazard way with negligible therapeutic results’. However, he made no explicit mention of the uric acid diathesis and it previously assumed aetiological role in mood disorders.

Cade now conducted an open clinical trial with lithium, comprising ten manic patients, six with schizophrenia, and three with chronic melancholy. Within days to a couple of weeks the manic patients improved, only to relapse on discontinuation of or in non-compliance with the drug. He had no doubt that the improvement in the manic patients ‘closely paralleled treatment’, and thus he considered lithium to be a ‘specific’ anti-manic agent.

The lithium historian FN Johnson (1984), who knew him well, wrote that Cade did not regard his experiments on guinea pigs as a natural extension of the uric acid diathesis. Further, he queried whether he had any reason other than the results of his guinea pigs ‘for believing that a successful outcome in his patients was likely’ – ‘probably not – at least not in a formal, explicit way’, Johnson thought. He found it ‘hardly likely’, though, that the claims going back over a hundred years for the therapeutic effects of lithium in for instance ‘mental affections’ ‘were either totally unknown to Cade or failed to influence his thought, at least in a general way’ (e.g. works of Garrod, Hammond, da Costa, Aulde, Haig, and the Danish brothers Carl and Frits Lange).

In the late 1800s and early 1900s the Langes were the first systematically to use and recommend lithium carbonate in the treatment and prophylaxis of mood disorders, which they believed were caused by the uric acid diathesis (Schioldann, 2001, 2009, 2011). However, when in the early 1900s this die-hard diathesis was proven fallacious, lithium’s thymoleptic effect fell into oblivion until its rediscovery by Cade.

The cardinal issue raised by Johnson has now been re-examined by the present author (Schioldann, 2009). My study showed that Cade’s hypothesis of manic-depressive illness was not dissimilar to what a number of previous investigators had discussed, as for instance summarized by Kraepelin in his work on manic-depressive illness (English edition, 1921) with which Cade would have been acquainted. Kraepelin dismissed Carl Lange’s views outright. Cade could have been aware of this. However, unless he had had access to Lange’s work via its German edition in 1896, it would not have been available to him, being ‘locked in the Danish language’. This fact, according to Callahan and Berrios (2005), ‘caused an incorrect history of the “discovery” of lithium treatment that historians are finding hard to resolve’.

Moreover, Cade would have been acquainted with a number of the old authors on the presumed connection between gouty conditions, nutrition impurities, the presence of some poison in the blood, and mood disorders and their treatment with alkalis, e.g. lithium salts. In his 1949 paper he did mention Garrod’s important work (1859), but not his use of lithium in gouty mania.

Samuel Gershon (Schioldann, 2009) for one had called Cade’s animal experiments and their interpretation into question. In his opinion, ‘the introduction of lithium […] would seem to have been quite serendipitous, as we do not have any significant basis for its reinvestigation’. Mogens Schou (Schioldann, 2009) for his part found Cade’s work ‘indeed strange’ – ‘the hypothesis which started his work was crude. His experimental design was not particularly clear. And his interpretation of the animal data may have been wrong’. Schou’s own attempts to replicate them indicated that the lethargy reported by Cade was caused by toxic over-dosage rather than a specific action of lithium. And why, Schou asked, ‘would a compound counteracting the effect of intraperitoneal urea be of psychiatric interest’. Notwithstandingly, added he: ‘this is the marvel of the thing – a spark jumped in John Cade’s questing mind, and he performed that therapeutic trial which eventually changed life for manic-depressive patients all over the world’.

My work concluded that Cade’s experimental animal studies cannot be considered to be documentation of scientific fact. It is not unlikely, therefore, that Cade did have knowledge, not revealed by him, that made his hypothesis and outcome of his clinical trial not so unlikely. Consistently with this opinion, rather than based on erroneous, irreproducible observations in guinea pigs, but with beforehand knowledge, Cade decided to undertake his clinical pilot study.

Further, I disputed that Cade’s choice of lithium in his animal experiments rested solely upon its greater solubility of its urate salt. Rather, I found direct evidence that a possible therapeutic efficacy of lithium in mania was not unsuspected by Cade in his comment, on 6 March 1948, in the case of WB, his first patient to receive lithium: ‘Chronic mania. This extremely high blood uric acid result [17.5 mg/%] is suspect’; 13 April 1948: ‘[WB] has been on large dosages of lithium citrate for a fortnight’. If not a determinant for Cade, this could have firmed his resolve to undertake his clinical trial.

As there was a lapse of 23 days – from 6 March to 29 March 1948 – before Cade instituted WB’s lithium treatment, it can be speculated, but not proven, that it was during this time that he self-administered lithium to establish what dosage to give to the patients, his being guided by various pharmacopoeias and other sources. Consistently with this view, he made an informed, not a fortuitous, choice.

What Cade ingeniously discovered was that lithium’s psychotropic effect (Cade, 1970) was attached to the lithium ion (Schioldann, 2009). Thus, paradoxically, he severed any association between lithium’s thymoleptic effect and uric acid, and uric diathesis, a nosological entity, which did not exist. Consistently, in a reply letter to a relative of his lithium patient, No. 9, RT, he wrote: ‘Please let me reassure you on several points. [RT’s] mental condition is not due to “poison in the blood” so that no treatment directed to neutralize such a poison would be of the slightest use’ (Schioldann, 2009).

Cade’s rediscovery, scientifically verified by Schou in 1954 (Schioldann, 2006), revolutionized the treatment of manic-depressive illness and gave birth to modern psychopharmacology, 3–4 years before the advent of chlorpromazine, and changed the face of psychiatry.

See Letter by Abhishekh and Faizan, 2012, 46(1): 68–69.