
An Update on the Role of Valproate in Bipolar Disorder
Key Takeaways
- Valproate is often overused for acute mania, despite guidelines recommending it as a third-line treatment option.
- Subsequent trials showed valproate's diminishing efficacy compared to lithium, carbamazepine, and antipsychotics.
Valproate's role in treating bipolar disorder is questioned as evidence suggests better alternatives exist for managing acute mania and maintenance.
BIPOLAR UPDATE
In my consulting work, I see many patients who have not done well on valproate. It often seems to be the first medication prescribed to patients with acute mania, when it should be third line, according to the latest US Department of Veterans Affairs and Department of Defense Clinical Practice Guideline for the Management of Bipolar Disorder.1 (I should disclose that I am a coauthor of that guideline.) The frequently cited Canadian Network for Mood and Anxiety Treatments guideline recommends valproate as a first-line treatment; however, they place it lower in the “hierarchy” among 8 first-line options, coming after lithium and quetiapine.2 So, they would not have you use it as the initial treatment unless you can think of a good reason to prefer it over the 2 higher-ranked options. I hope this update will have some impact on prescribers who want to use it first—or at all.
Valproate/divalproex was approved by the US Food and Drug Administration (FDA) in 1994 for use in bipolar mania after the publication of the first major study that showed efficacy. It was heavily marketed by its drug company and entered the mainstream of practice in the US. It has been used less in Europe and even less in Japan, where carbamazepine is more popular as an alternative to lithium. In the 1970s, highly respected Japanese researcher Teruo Okuma, MD, PhD, was the first to demonstrate the effect of carbamazepine in mania.
With the next 4 placebo-controlled trials for mania, the effect size vs placebo diminished dramatically. In the last 2 of these trials (1 in adults, 1 in adolescents and young adults), there were no differences from placebo.3,4 Meta-analyses of controlled studies of medications for acute mania find it to have a smaller effect size (0.16 standardized mean difference [SMD] from placebo) than lithium, carbamazepine, and several antipsychotics (SMD effect sizes range from 0.37 to 0.56).5
Maintenance trials of valproate in bipolar disorder initially showed no efficacy, and the medication did not get FDA approval as a maintenance treatment. However, there was a 2021 review that brought together 13 reports involving 9240 participants (3 studies with placebo control, 10 open-label comparisons with other mood stabilizers), which concluded that there was maintenance effectiveness for both manias and depressions.6 There are also a few small studies that suggest that valproate could be effective for acute bipolar depression, though it is not FDA approved for this use, and larger studies are needed.7 Apropos of all this, I would ask the reader: When you are prescribing valproate as a maintenance treatment, are you informing your patients that this use is off-label, and that there are other medications with solid evidence of efficacy and approval for maintenance use (like lithium and quetiapine)?
Patients with acute mixed mania (ie, meeting full criteria for mania but having 3 or more symptoms of depression concomitantly) seem to fare better on valproate. Valproate may be preferred over lithium for such patients based on a small amount of early evidence.8
Clinicians often choose valproate because of a perception that it has a milder adverse effect profile than other bipolar mania medications. However, head-to-head studies indicate that one very important adverse effect, weight gain, is greater with valproate than with lithium, olanzapine, and quetiapine.8 The risk of suicidal behaviors is twice as high on anticonvulsants such as valproate compared with controls (all anticonvulsant package inserts contain a warning about this), whereas convincing and updated evidence finds a unique reduced risk of suicidality on lithium.9 Liver dysfunction and, less commonly, pancreatitis, thrombocytopenia, and polycystic ovary syndrome are other significant adverse effects associated with valproate. It has the most teratogenicity of any medication used in bipolar disorder and has been declared unacceptable as a treatment for any woman of childbearing potential.10
In conclusion, valproate/divalproex still appears to be overused in the US. Clinicians should consider lithium, second-generation antipsychotics—especially quetiapine—and even carbamazepine (though also not for women who could become pregnant) when selecting an initial medication for patients with classic (nonmixed) bipolar mania or severe hypomania. Patients should be informed of the evidence base for efficacy and FDA approval status before being prescribed valproate for acute illness or for maintenance usage.
Dr Osser is an associate professor of psychiatry at Harvard Medical School in Boston, Massachusetts; a psychiatrist at the Veterans Affairs (VA) Boston Healthcare System, Brockton Division; and codirector of the VA National Bipolar Disorders TeleHealth Program.
References
1. Ostacher MJ, Miller CJ, Edwards-Stewart A, et al.
2. Keramatian K, Chithra NK, Yatham LN.
3. Hirschfeld RM, Bowden CL, Vigna NV, et al.
4. Wagner KD, Redden L, Kowatch RA, et al.
5. Cipriani A, Barbui C, Salanti G, et al.
6. Yee CS, Vázquez GH, Hawken ER, et al.
7. Wang D, Osser DN.
8. Mohammad O, Osser DN.
9. Wang JX, Le GH, Wong S, et al.
10. Freeman MP.
Newsletter
Receive trusted psychiatric news, expert analysis, and clinical insights — subscribe today to support your practice and your patients.














