Psychiatric Classics #1: Lithium for Mania
Classic Paper: Cade JF. Lithium salts in the treatment of psychotic excitement. Med J Aust 1949; 2:349-352
some wells in the British Isles were known for their salubrious effects on mental illness; this may have been due to their lithium content
Prior to 1949, treatments for mania were limited. That year, John Cade published a paper showing the usefulness of lithium in treating patients with mania ("psychotic excitement").
Interestingly enough, the finding was apparently a surprise to Cade. He was studying guinea pigs in order to see whether uric acid added to the convulsive toxicity of urea, but he needed to find a way to make uric acid soluble in water to be able to inject it into the guinea pigs. (Confusingly enough, urea and uric acid have almost nothing to do with one another chemically.)
For this, he used the lithium salt of urate, and was surprised to find that it was protective against the urea-induced convulsions. He then injected lithium carbonate alone into guinea pigs, and noted that after a couple of hours, they became lethargic and unresponsive to stimuli.
Skipping straight from this effect in guinea pigs (not even a disease model!! -- this would never be allowed today) to humans, Cade then reports on 10 cases of patients with mania who were successfully treated with lithium, including longitudinal cases of chronic mania where the mania subsided during lithium treatment and recrudesced when lithium was discontinued.
Other interesting aspects of this paper:
Cade notes that historically, water from certain wells was associated with improvements in mental illness, and speculates that "it is very likely that their supposed efficacy was a real efficacy and directly proportional to the lithium content of the waters."
Cade notes that lithium treatment "would be much preferred" to what is usually now considered the cruel treatment of prefrontal leucotomy, even though this (1949) was the year that the Nobel prize was awarded for it, and its use continued into the mid-1950s.
All of the cases reported on were men between ages 40 and 65 years old, indicating a total lack of evidence for generalization of the effect across more diverse patient populations.
Recent meta-analysis (2013) has shown that antipsychotics are more effective than lithium in the treatment of acute mania (e.g., the standardized mean difference in manic symptoms for haloperidol is -0.56, while for lithium it is -0.37), but lithium is still often used in combination with antipsychotics in the treatment of mania.
Overall, this short paper is among the best I've read in terms of scientific puzzle solving, although you could argue that Cade got lucky.
References
Cade JF. Lithium salts in the treatment of psychotic excitement. 1949. Bull World Health Organ. 2000;78(4):518-20.
Cipriani A, Barbui C, Salanti G, et al. Comparative efficacy and acceptability of antimanic drugs in acute mania: a multiple-treatments meta-analysis. Lancet. 2011;378(9799):1306-15.
Doig MT, Heyl MG, Martin DF. Lithium and mental health. J Chem Educ. 1973;50(5):343-5.