This review shows that lithium and valproate have comparable anti-manic effects. However, manic patients who respond well to lithium are euphoric but manic patients who respond poorly to lithium have more than 10 past recurrences, are in a mixed state and/or have dysphoric mood and mood-incongruent psychotic features such as delusions of persecution. Valproate exhibits greater anti-manic efficacy than lithium in manic patients with a higher number of past recurrences and may be effective for patients with mixed conditions and rapid cycling. The antidepressant effects of both lithium and valproate are modest, and may necessitate concomitant lamotrigine and/or switching. Lithium may afford more prophylactic efficacy than valproate. Positive predictors for lithium prophylaxis are an episodic pattern of mania-depression-interval and a high age of illness onset, while negative predictors are a high number of previous hospitalizations, an episodic pattern of depression-mania-interval, and continuous cycling. Checking these predictors in order to determine whether lithium or valproate should be selected would appear to be useful. In addition, even very low levels of lithium in drinking water may contribute to a reduction in suicide risk in the general population. In conclusion, both lithium and valproate are useful in the treatment of bipolar disorders, and several measures exist for differentiating their clinical positioning.
2013 The Japanese Society of Clinical Neuropsychopharmacology