Clinicians treating older patients with bipolar disorder with mood stabilizers need evidence from age-specific randomized controlled trials. The authors describe findings from a first such study of late-life mania.

The authors compared the tolerability and efficacy of lithium carbonate and divalproex in 224 inpatients and outpatients age 60 or older with bipolar I disorder who presented with a manic, hypomanic, or mixed episode. Participants were randomly assigned, under double-blind conditions, to treatment with lithium (target serum concentration, 0.80–0.99 mEq/L) or divalproex (target serum valproate concentration, 80–99 μg/mL) for 9 weeks. Participants with an inadequate response after 3 weeks received open adjunctive risperidone. The authors hypothesized that divalproex would be better tolerated and more efficacious than lithium. Tolerability was assessed based on a measure of sedation and on the proportions of participants achieving target concentrations. Efficacy was assessed with the Young Mania Rating Scale (YMRS).

Attrition rates were similar for lithium and divalproex (14% and 18% at week 3 and 51% and 44% at week 9, respectively). The groups did not differ significantly in sedation. Participants in the lithium group tended to experience more tremor. Similar proportions of participants in the lithium and divalproex groups achieved target concentrations (57% and 56%, respectively). A longitudinal mixed model of improvement (change from baseline in YMRS score) favored lithium (change in score, 3.90; 97.5% CI=1.71, 6.09). Nine-week response rates did not differ significantly between the lithium and divalproex groups (79% and 73%, respectively). The need for adjunctive risperidone was low and similar between groups (17% and 14%, respectively).

Both lithium and divalproex were adequately tolerated and efficacious; lithium was associated with a greater reduction in mania scores over 9 weeks.

临床医师使用情绪稳定剂治疗老年双相情感障碍患者需要来自特定年龄的随机对照试验的证据。作者描述了从最早的晚年躁狂症研究中得出的发现。

作者比较了224例患有躁郁症的躁郁症，躁狂症或混合发作的住院患者和门诊患者，年龄在224岁或以上的碳酸锂和divalproex的耐受性和疗效。在双盲条件下，将参与者随机分配接受锂（目标血清浓度为0.80-0.99 mEq / L）或双丙戊酸钠（目标血清丙戊酸盐浓度为80-99μg/ mL）治疗9周。3周后反应不足的参加者接受开放性辅助利培酮治疗。作者假设双丙戊酸比锂具有更好的耐受性和疗效。基于镇静措施和达到目标浓度的参与者比例评估耐受性。使用年轻躁狂症评定量表（YMRS）评估疗效。

锂和双丙戊酸钠的磨损率相似（第3周分别为14％和18％，第9周分别为51％和44％）。各组镇静作用无明显差异。锂电池组的参与者往往会感到更多的震颤。锂和双丙戊酸组中相似比例的参与者达到目标浓度（分别为57％和56％）。改善的纵向混合模型（YMRS评分偏离基线）偏爱锂（评分变化3.90； 97.5％CI = 1.71，6.09）。锂和divalproex组之间的九周应答率没有显着差异（分别为79％和73％）。组之间对辅助利培酮的需求低且相似（分别为17％和14％）。

锂和divalproex均具有足够的耐受性和有效性。锂与9周内躁狂评分的更大降低有关。