Lithium in bipolar disorder: can drug concentrations predict therapeutic effect?

The evidence is reviewed for effective serum lithium concentrations for the acute and prophylactic treatment of mania and depression in patients with bipolar disorder. The efficacy of lithium in the treatment of acute manic episodes has been recognised for several decades, primarily using concentrations in the range of 0.8 to 2 mmol/L. The number of patients responding increases as the serum lithium concentration increases, although individual patients may respond at lower concentrations (<0.8 mmol/L). Lithium doses and serum concentrations similar to those used to treat acute mania have been studied in bipolar depression, with no evaluation of a relationship between concentration and clinical response. Several prospective controlled trials have evaluated this relationship in the prophylactic treatment of bipolar disorder. Maintaining higher serum lithium concentrations (0.8 to 1 mmol/L) improves the likelihood of good effect in prophylactic treatment, although individual patients may do well on lower concentrations. Despite the paucity of evidence to specifically support the efficacy of lithium at lower serum lithium concentrations in the elderly, lower target ranges (0.5 to 0.8 mmol/L) are commonly recommended due to an increased sensitivity to adverse effects, particularly neurotoxicity. The serum lithium concentrations recommended in adults have been applied to children; however, this has not been studied. Overall, the evidence suggests a relationship between serum lithium concentration and therapeutic effect, although the exact nature of this relationship is not clear. For example, it is not known why some people respond to lower concentrations and others do not. There are many factors that influence studies trying to elucidate this relationship. Many of these factors are related to the interpretation of the serum lithium concentration. In summary, patients have an increased chance of responding to lithium if 12-hour serum lithium concentrations at steady state are above 0.8 mmol/L. Many patients will respond to lower concentrations (0.4 to 0.7 mmol/L), but we are unable to identify these patients a priori. The relationship between serum lithium concentrations and adverse effects is also very important in determining appropriate target lithium concentrations. The current best advice is to individualise the target serum lithium concentrations based on efficacy and tolerability and to optimise the interpretation of these concentrations by ensuring within-patient consistency with respect to dosage schedule, lithium preparation and the timing of blood sampling.