Serum lithium levels and suicide attempts: a case-controlled comparison in lithium therapy-naive individuals

Several epidemiological studies have shown the inverse association of lithium levels in drinking water and suicide rates; however, it is necessary to perform a clinical study dealing with individual patients. We analyzed 199 patients including 31 patients with suicide attempts, 21 patients with self-harm, and 147 control patients. All were transferred to a university emergency department suffering from intoxication or injury, were aged 20 or more years, and were alive at the start of the study. The exclusion criteria consisted of suffering from schizophrenia and a past or present history of lithium therapy. These exclusions were applied because it is difficult to determine whether their suicide attempt was induced by the intent to end their life or by psychotic symptoms such as auditory hallucinations, and if the patient had received lithium therapy, the association between the small amount of lithium taken from drinking water and food and serum lithium levels cannot be detected. There was a significant difference (p = 0.043) between the three groups whereby patients with suicide attempts had significantly lower lithium levels than control patients (p = 0.012) in males but not females. Multivariate logistic regression analysis with adjustment for age and gender revealed that patients with suicide attempts had significantly lower lithium levels than control patients (p = 0.032, odds ratio 0.228, 95% CI 0.059-0.883). The limitations of the present study are the nature of observational research which cannot reveal a causal relationship and the relatively small number of subjects. The present findings suggest that higher serum lithium levels may be protective against suicide attempts in lithium therapy-naive individuals.