Alternatives to Lithium for Bipolar Disorder: Effective Options

Several medication classes can effectively replace or supplement lithium in treating bipolar disorder, including anticonvulsant mood stabilizers like valproate and lamotrigine, and atypical antipsychotics such as quetiapine and olanzapine. The choice depends on which phase of the illness you’re treating, your side-effect tolerance, and factors like kidney function or pregnancy. While lithium remains a gold-standard treatment, its narrow therapeutic window and potential to damage the kidneys, thyroid, and parathyroid glands push many people toward alternatives that work through different mechanisms and carry different trade-offs.

Why People Look Beyond Lithium

Lithium works well for a lot of people, but it demands vigilance. It has a narrow gap between the dose that helps and the dose that becomes toxic, which means regular blood draws to keep levels in range.¹ Over time, lithium can impair kidney function, interfere with thyroid hormone production, and affect calcium regulation through the parathyroid glands. Those organ-function checks aren’t optional; they’re a permanent feature of lithium therapy. For some people, the monitoring burden alone is reason enough to explore other options. For others, real kidney damage or thyroid dysfunction forces the switch.

A nationwide population-based study found that the risk of chronic kidney disease climbed steadily with cumulative lithium exposure, roughly two-and-a-half times higher in those with the most prescriptions compared to matched controls.² When kidney function declines toward end-stage disease, clinicians face a difficult decision: stop lithium and risk destabilizing a condition that may have been stable for years, or continue lithium and accelerate renal damage.³ There are also people who simply don’t respond well to lithium, particularly those with mixed episodes where manic and depressive symptoms overlap. In those cases, the search for alternatives isn’t about side effects at all but about finding something that actually works.

Valproate as the Most Direct Substitute

Valproate (sold as divalproex or valproic acid) is the medication most often placed head-to-head against lithium in clinical trials, and it consistently performs about as well across the major phases of bipolar disorder. A large overview of systematic reviews covering dozens of randomized trials found no significant difference between valproate and lithium in treating acute mania, bipolar depression, or long-term maintenance.⁴ A Cochrane review confirmed this rough equivalence for maintenance treatment, finding that valproate beat placebo in preventing mood episodes but was statistically indistinguishable from lithium.⁵

One area where valproate may edge ahead is in mixed states. A study comparing the two drugs found that patients with a mixed presentation tended to respond better to valproate than to lithium, while those with classical, non-mixed mania did comparably on either drug.⁶ That matters because mixed episodes are common and notoriously hard to treat, and lithium’s track record in that specific subtype is weaker.

The BALANCE trial, a major randomized study of relapse prevention in bipolar I disorder, provided a useful wrinkle. Lithium monotherapy outperformed valproate monotherapy in preventing relapse, and the combination of lithium plus valproate was better still.⁷ So if you’re doing well on lithium and considering a full switch to valproate alone for maintenance, the evidence suggests you might lose a small degree of protection. The combination, however, can offer the best of both worlds for people who tolerate both drugs.

Lamotrigine for the Depressive Side

Bipolar disorder isn’t just about mania; most people with the condition spend far more of their lives in depressive episodes than in manic ones. Lamotrigine fills a gap that lithium and valproate leave partially open. It is most effective at preventing depressive episodes, which tend to dominate the clinical picture of bipolar disorder, and by keeping depression at bay it helps people sustain longer stretches of remission.⁸

A network meta-analysis of maintenance treatments ranked lamotrigine as effective for preventing any mood episode compared to placebo, but it was less potent at preventing manic or hypomanic episodes than most atypical antipsychotics or lithium itself.⁹ In practice, clinicians often pair lamotrigine with a second medication that handles the manic side. One randomized trial found that combining lamotrigine with divalproex led to fewer depressive breakthroughs than lamotrigine alone, particularly in bipolar I patients.¹⁰

Lamotrigine has a notably mild metabolic profile. It appears to have little effect on weight, in contrast to valproate or lithium, both of which can cause noticeable weight gain.¹¹ The main safety concern is a rare but serious skin reaction that can develop if the dose is increased too quickly, which is why lamotrigine is always titrated slowly over several weeks.

Atypical Antipsychotics Across All Phases

The atypical (or second-generation) antipsychotics have expanded the toolkit for bipolar disorder more than any other class in the past two decades. Medications like quetiapine, olanzapine, aripiprazole, risperidone, and newer entries such as lurasidone, cariprazine, and lumateperone now carry regulatory approval for various phases of the illness. They are effective both as standalone treatments and as add-ons to mood stabilizers for acute mania, and increasingly for bipolar depression and maintenance as well.¹²

For bipolar depression specifically, a systematic review and network meta-analysis found that cariprazine, lumateperone, lurasidone, olanzapine, and quetiapine all significantly outperformed placebo in response rates.¹³ Lumateperone, one of the newest options, showed meaningful improvement in depression scores compared to placebo in both bipolar I and bipolar II patients, which is relevant because bipolar II often gets less attention in drug trials.¹⁴

For long-term maintenance, the same network meta-analysis found that olanzapine, quetiapine, aripiprazole, and several others all reduced relapse rates compared to placebo. Asenapine performed particularly well in the rankings, outperforming several established treatments for preventing episodes of any type.⁹ The practical takeaway is that if lithium isn’t working or isn’t tolerated, there are multiple antipsychotic options with solid trial evidence for keeping people stable long-term.

Metabolic Trade-Offs Worth Watching

Switching from lithium doesn’t mean you’re free of metabolic concerns. The antipsychotics in particular come with their own set of risks involving weight, blood sugar, and cholesterol. Among roughly 770 outpatients randomized to quetiapine in one study, cardiometabolic risk markers increased over 24 weeks of treatment.¹⁵ A large network meta-analysis that mapped each drug to specific metabolic outcomes found that olanzapine was the most likely to raise cholesterol and triglycerides, risperidone tended to elevate fasting blood sugar and insulin, and lurasidone was associated with higher hemoglobin A1c levels over time.¹⁶

Among mood stabilizers, valproate causes more weight gain than lithium. A randomized trial found valproate led to roughly a kilogram of weight gain in 12 weeks versus virtually none with lithium, and the fat distribution with valproate tends to settle around the abdomen, which carries higher cardiovascular risk.¹¹ Lamotrigine and carbamazepine appear to have relatively little effect on weight, making them the lightest options metabolically if they’re appropriate for your particular bipolar pattern.

Pregnancy Changes the Calculus

For people who can become pregnant, the choice of mood stabilizer takes on additional stakes. Valproate, despite its general effectiveness, is strongly associated with congenital malformations and developmental problems in offspring, and current guidance in multiple countries restricts its use in women and girls of childbearing potential.¹⁷ Lithium itself carries a risk of cardiac malformations, particularly Ebstein anomaly, though the absolute risk is small — around 1 in 2,000 live births with first-trimester exposure, roughly ten times the background rate. Carbamazepine has been linked to neural tube defects and craniofacial abnormalities and is generally not recommended as a substitute for valproate in this population.¹⁸

That leaves lamotrigine as a frequent first choice for people planning pregnancy, given its effectiveness against depressive episodes and its comparatively safer reproductive profile.⁸ Some atypical antipsychotics are also used, though reproductive safety data for many of them is less extensive. The decision is always individualized, weighing the risks of medication exposure against the well-documented risks of untreated bipolar episodes during pregnancy.

Treating Children and Adolescents

The evidence base for treating bipolar disorder in young people is thinner than in adults, and the medication landscape looks somewhat different. Second-generation antipsychotics appear to be more effective overall than mood stabilizers for pediatric mania, with five antipsychotics carrying FDA approval for this age group. Lithium remains the only mood stabilizer with FDA approval for pediatric bipolar disorder.¹⁹ Aripiprazole, olanzapine, and risperidone have all shown short-term effectiveness for both mania and mixed states in young people.²⁰

Among anticonvulsants, the pediatric picture is more complicated. Open-label trials of divalproex showed encouraging response rates in the short term, but controlled studies have been less impressive. A double-blind placebo-controlled trial found no meaningful difference between divalproex and placebo over six months of treatment in youth with bipolar disorder. Oxcarbazepine fared similarly poorly in a placebo-controlled trial of over 100 young people, and nearly a fifth of the treatment group dropped out due to side effects. Lamotrigine showed better numbers in open-label work, with response rates above 70% for manic symptoms, but these results await confirmation in rigorous placebo-controlled designs.²¹ Metabolic side effects from antipsychotics are especially concerning in younger patients, where even modest weight gain and lipid changes can set trajectories for cardiovascular risk later in life.¹⁶

Non-Drug Approaches That Have Evidence

No psychotherapy replaces mood-stabilizing medication in bipolar disorder, but several structured therapies have proven value as add-ons. Interpersonal and social rhythm therapy (IPSRT) focuses on stabilizing daily routines, especially sleep-wake cycles, under the theory that disrupted rhythms can trigger episodes. In a controlled trial, patients receiving IPSRT alongside standard medication showed significant improvements in depressive and manic symptoms, overall functioning, and even how well they responded to their mood stabilizers.²² Cognitive behavioral approaches specifically adapted for bipolar disorder, sometimes blended with elements of IPSRT and mindfulness, have also been tested for relapse prevention in randomized trials.²³

Bright light therapy, borrowed from the treatment of seasonal depression, has shown promise for bipolar depression as well. A systematic review and meta-analysis of randomized controlled trials concluded that it is an effective adjunctive treatment for reducing depressive symptoms in people with bipolar disorder.²⁴ Timing matters: midday bright light exposure is generally preferred over early morning use, because morning light can occasionally trigger mania in susceptible individuals. A related and more experimental approach involves blocking blue light in the evening using amber-tinted lenses, creating a kind of “virtual darkness” that may help regulate circadian rhythms and reduce sleep onset delays. The underlying idea is that blue light suppresses melatonin production, and blocking it after sundown could mimic the darkness that helps stabilize mood. Preliminary case reports suggest some patients with bipolar disorder fall asleep faster with this approach, though large trials are lacking.²⁵

Nutritional Supplements as Add-Ons

Two dietary supplements have enough trial data to warrant mention, though neither is a replacement for prescription medication. Omega-3 fatty acids, the kind found in fish oil, showed a moderate benefit for bipolar depression in a meta-analysis of five pooled datasets.²⁶ The effect was specific to depressive symptoms; the same analysis found no benefit for mania. N-acetylcysteine (NAC), an antioxidant available over the counter, also showed a moderate effect size favoring it over placebo for bipolar depression across six clinical trials, though the confidence interval was wide enough to suggest considerable uncertainty about the true benefit.²⁷ Both are used as adjuncts to standard medication, not as standalone treatments. They’re inexpensive, generally well-tolerated, and unlikely to cause harm, which makes them reasonable to discuss with a prescriber even though their effects are modest.

Emerging Drug Treatments

Esketamine, the nasal spray already approved for treatment-resistant unipolar depression, has been explored in bipolar depression as well. A naturalistic multicenter study compared esketamine in bipolar treatment-resistant depression against unipolar treatment-resistant depression and found significant reductions in depressive symptoms in both groups, with no significant difference in response or remission rates between them. No treatment-emergent manic switches were observed, which had been a key safety concern.²⁸ These are early data, and esketamine is not yet approved for bipolar depression, but the absence of manic switching is encouraging.

Pharmacogenomic testing, which analyzes genes that influence how you metabolize psychiatric medications, has been a hope for years. In theory, knowing which enzyme variants you carry could guide the choice of mood stabilizer or antipsychotic. In practice, the research supports the idea that genetics plays a role in treatment response, but robust, clinically actionable predictors remain elusive.²⁹ Some clinicians order pharmacogenomic panels, and they can sometimes help avoid drugs you’re likely to metabolize poorly, but they can’t yet reliably tell you which medication will work best for bipolar disorder specifically.

How to Safely Transition Off Lithium

If you and your clinician decide to move away from lithium, how you stop matters as much as what you switch to. Abrupt discontinuation carries a substantially higher risk of relapse compared to a gradual taper, and the evidence is consistent enough that this should be treated as a rule, not a suggestion.³⁰ One review recommended reducing the dose over at least four weeks, and ideally up to three months, with continued monitoring of symptoms and mental state for two years after the medication has been fully stopped.³¹

There’s also an older concern that some people who stop lithium and then relapse may not respond as well when they restart it. While the evidence on this so-called discontinuation-induced refractoriness is debated, the possibility adds another reason to taper slowly and to overlap the new medication with lithium during the transition rather than stopping one and starting the other cold. A cross-taper, where the replacement medication is brought up to a therapeutic dose before lithium is fully withdrawn, is the standard approach in most clinical settings.

Electroconvulsive Therapy for Severe or Resistant Episodes

For people who have not responded to multiple medication trials, electroconvulsive therapy (ECT) remains an option that tends to be underutilized relative to its effectiveness. A case series of patients with treatment-resistant bipolar disorder who were referred for ECT found that all who consented and received treatment responded, despite having failed previous medication regimens and despite several presenting with psychotic features.³² ECT is generally considered for acute, severe episodes rather than as a maintenance strategy, though some patients do receive periodic maintenance treatments. Its reputation still suffers from outdated perceptions, but modern ECT is administered under anesthesia with muscle relaxants and has been refined considerably in terms of both safety and the precision of electrical stimulation.