Effects of lithium treatment and its discontinuation on suicidal behavior in bipolar manic-depressive disorders.

BACKGROUND Whether mood-altering treatments reduce risk of suicidal behavior remains largely unproved. METHOD We compared suicidal rates in published studies of patients treated with lithium with those who were not, and in a mood disorders clinic before, during, and after discontinuing lithium. RESULTS Published reports indicate a 7.0-fold lower rate of suicidal acts with lithium treatment of manic-depressive patients. In new findings in over 300 bipolar patients, latency from illness onset to lithium maintenance averaged 8.3 years (from 11.0 years in women with bipolar II disorder to 6.9 years in men with bipolar I disorder), but half of all suicidal acts occurred in the first 7.5 of 18.3 years at risk. Most acts (89%) occurred during depressive (73%) or dysphoric-mixed (16%) mood states and were associated with previous severe depression, prior attempts, and lower age at onset. Morbidity was reduced 2.7-fold and suicidal acts per year 6.5-fold during lithium treatment, with 8.3-fold cumulative sparing of risk by 15 years on lithium. In the first year off lithium, affective illness recurred in 67% of patients, and suicidal rates rose 20-fold but were much lower thereafter; fatalities were 14 times more frequent after discontinuation of lithium. Early morbidity was 2.5-fold lower, and suicidal risk was 2.0-fold lower after slow versus rapid discontinuation. CONCLUSION Lithium maintenance is associated with sustained reduction of suicidal acts in manic-depressive disorders. Treatment discontinuation, particularly abruptly, led to early affective morbidity and suicidal behavior. Improved diagnosis and treatment as well as earlier intervention for potentially lethal bipolar depression are urgently needed, as are studies of all mood-altering agents for effects on suicidal behavior.