Condition: Bipolare Störung

Across randomized maintenance trials, lithium was more effective than placebo at preventing relapse to any mood episode, with a random-effects relative risk of about 0.65 (95% CI 0.50 to 0.84). The protective effect was clear for manic relapse and weaker and less certain for depressive relapse. Lithium remains a first-line maintenance treatment in the CANMAT and ISBD guidelines. Measured in: Adults with bipolar disorder in five randomized placebo-controlled maintenance trials, 770 participants. Several older maintenance trials used designs that recruited people already stable on lithium and then withdrew it in the placebo arm, which can exaggerate the apparent benefit; the effect on depressive episodes is the weaker part of the evidence.

In a network meta-analysis of acute mania, the drugs with the highest probability of being most effective, measured by change on the Young Mania Rating Scale versus placebo, were haloperidol (standardized mean difference -0.56), risperidone (-0.50) and olanzapine (-0.43). Overall, antipsychotics were significantly more effective than mood stabilizers such as lithium and valproate for calming an acute manic episode, though tolerability differed between drugs. Measured in: 16,073 adults with acute mania across 68 randomized controlled trials. Efficacy for the acute episode is not the same as long-term suitability: several of the most effective antimanic drugs carry sedation, movement side effects or metabolic effects that matter over the years someone takes them, and lithium's edge is in maintenance, not acute mania.

In a network meta-analysis of the atypical antipsychotics approved for bipolar depression, all of them beat placebo on response (a 50% or greater fall on the Montgomery-Asberg Depression Rating Scale). The number needed to treat for response was lowest for lurasidone (5), then quetiapine (6), olanzapine (10) and cariprazine (12). Lurasidone and cariprazine caused less weight gain than olanzapine and quetiapine, so the choice trades efficacy against side effects. Measured in: adults with bipolar depression across 18 randomized controlled trials of atypical antipsychotic monotherapy (lurasidone, quetiapine, olanzapine, cariprazine, aripiprazole and ziprasidone). These drugs differ sharply in side effects (weight gain, sedation, metabolic effects), so the most effective on paper is not automatically the right choice for a given person, and the trials were acute, not long-term.

In the World Health Organization World Mental Health Survey Initiative, the aggregate lifetime prevalence was 0.6% for bipolar I disorder, 0.4% for bipolar II, 1.4% for subthreshold bipolar, and 2.4% for the bipolar spectrum overall. Three quarters of people with a bipolar spectrum disorder also met criteria for another psychiatric disorder, most often an anxiety disorder (63%). Rates were higher in high-income than low-income countries, but severity and comorbidity were similar across settings, and treatment need was often unmet. Measured in: 61,392 community adults across 11 countries in the Americas, Europe and Asia, assessed with the WHO Composite International Diagnostic Interview version 3.0. What could explain it instead: Prevalence estimates depend on the diagnostic threshold and interview method used, and structured lay-administered interviews can over-count or under-count bipolar spectrum cases relative to clinician diagnosis, which is one reason rates vary widely between countries.. Cross-country estimates ranged widely (from about 0.1% to 3.3%), reflecting real differences and differences in case ascertainment, and a lay-administered interview is not the same as a clinical diagnosis.

In randomized placebo-controlled trials, second-generation antidepressants added on top of a mood stabilizer or an antipsychotic produced a small reduction in acute bipolar depression symptoms but did not increase the proportion of people who reached clinical response or remission. Prolonged use was associated with a higher risk of switching into mania or hypomania, so the authors concluded antidepressants have at best a modest short-term role as an add-on, never as sole treatment. Measured in: Adults with acute bipolar depression across randomized double-blind placebo-controlled trials of adjunctive second-generation antidepressants. The trials tested newer antidepressants added to existing treatment, not older classes or monotherapy, and were short-term, so they do not describe long-term maintenance or the higher-risk drugs.

Pooling individual patient data from five randomized trials, more people responded to lamotrigine than placebo in bipolar depression on both the Hamilton (relative risk 1.27, 95% CI 1.09 to 1.47) and Montgomery-Asberg (1.22, 1.06 to 1.41) depression scales, with a larger effect in people who were more severely depressed at baseline. In maintenance, lamotrigine reduces the risk of a new depressive episode. Its weakness is the manic pole: it does little for acute mania and must be started slowly because of a rare serious rash. Measured in: 1,072 adults with bipolar depression across five randomized placebo-controlled trials (individual patient data). The acute antidepressant effect is modest and clearest in more severe depression; lamotrigine does not treat mania, and it must be titrated slowly to reduce the risk of a serious rash (Stevens-Johnson syndrome).

In a systematic review of randomized controlled trials, structured group psychoeducation added to medication reduced the risk of relapse in bipolar disorder, with the effect clearest for manic and mixed relapse and for people earlier in the illness. Individually delivered psychoeducation showed less consistent benefit. Psychoeducation teaches people to recognize early warning signs, keep regular routines, adhere to treatment, and act on a relapse plan. Measured in: Adults with bipolar disorder across randomized controlled trials of psychoeducation versus usual care or active comparators. The benefit was strongest for group formats and for people not in an acute episode and earlier in the course of illness; it is an add-on to medication, not a replacement, and trials varied in content and length.

In a randomized trial, people who received interpersonal and social rhythm therapy (IPSRT) in the acute phase went longer without a new mood episode over two years than those given intensive clinical management, even though the two groups reached remission at similar rates (70% versus 72%). IPSRT works by steadying daily rhythms, sleep, meals and social routines, on top of medication, which is one of the more direct levers in bipolar disorder because circadian disruption can precipitate episodes. Measured in: 175 adults with bipolar I disorder randomized to IPSRT or intensive clinical management. The two therapies reached remission at the same rate; IPSRT's advantage was in staying well afterward, and it is an adjunct to medication, not a standalone treatment.

Pooled analyses found that adjunctive omega-3 fatty acids produced a modest improvement in bipolar depressive symptoms (a significant but small effect size around 0.3), with the benefit tied to EPA-predominant formulations added on top of standard treatment. There was no benefit for manic symptoms. The depression finding rests on a small number of trials and shows signs of publication bias, so it is best read as a low-risk adjunct, not a treatment in its own right. Measured in: Adults with bipolar disorder across pooled randomized trials of adjunctive omega-3 fatty acids. Few trials, heterogeneity between them, and evidence of publication bias make the effect uncertain; omega-3 does not treat mania and does not replace a mood stabilizer.

In a randomized trial, adding family-focused therapy (psychoeducation, communication training and problem-solving with the person and their relatives) to medication reduced mood relapses and rehospitalizations over two years compared with an individually focused treatment. The effect is thought to run partly through lowering high-conflict, high-criticism family environments that predict relapse, and partly through better recognition of early warning signs. Measured in: 53 recently hospitalized adults with bipolar disorder randomized to family-focused therapy (n=28) or individually focused treatment (n=25), all on mood-stabilizing medication. The trial was modest in size (53 patients), the two treatments may have differed in total contact time, and the approach depends on having involved family members, so it does not fit everyone's circumstances.

Multiple lines of evidence indicate that disrupted sleep can both predict and precipitate manic episodes: a change in the sleep pattern often precedes a change in mood, and deliberately curtailing sleep can tip some people with bipolar disorder into mania or hypomania. Because of this, stabilizing sleep and daily rhythm is treated as a core priority in bipolar disorder, both as a way to reduce relapse and as an early-warning signal to watch. Measured in: Reviews of experimental, observational and clinical studies of sleep and circadian function in bipolar disorder. This is review-level evidence assembling many smaller studies, not a single large trial of sleep regularization for relapse prevention, and not everyone who loses sleep switches, so it describes a strong and consistent risk relationship, not a fixed cause in every case.