Lithium Still Best Medication for Preventing Bipolar Relapse

Daniel M. Keller, PhD

April 06, 2015

VIENNA — Naturalistic studies may reveal outcomes that randomized clinical trials miss. After hospital discharge, a 4-year, naturalistic follow-up of patients with bipolar disorder (BP) shows that even with prophylactic medication, relapse rates are high and are correlated with alcohol use, psychiatric comorbidities, life events, and the type of maintenance medication used.

"If you look into naturalistic studies over 4 or 5 years, then you can see that the relapse rate goes on, and far more patients have relapses...75% have relapses in spite of medication," compared with 50% in clinical trials, said study investigator Christian Simhandl, MD, PhD, of the Bipolar Center Wiener Neustadt, in Vienna, Austria.

Those data, from a meta-analysis of observational, naturalistic studies of 1375 BP I and BP II patients, showed a 50% relapse rate at just over 1 year.

The results were presented here at the European Psychiatric Association (EPA) 23rd Congress.

Beginning in 2000, the researchers recruited consecutively admitted BP I and BP II patients (158 BP I, 142 BP II) after the patients were discharged from a community psychiatric hospital.

In this naturalistic setting, there was no specified study drug, and treatment was at the discretion of the physician. Each patient was followed yearly for 4 years between 2000 and 2008. Follow-up included personal or telephone interviews and reviews of hospital and outpatient charts.

Within 4 years, 204 (68%) patients relapsed, at a mean of 208 ± 356.2 days. Relapses began within the first couple of months after hospital discharge and correlated with the type of index episode (P < .001) (Table). Kaplan-Meier curves showed that maintenance therapy with lithium (n = 49) prevented relapse longer than therapy with other mood stabilizers (n = 250). Half of the patients receiving lithium had a first relapse between 1300 and 1400 days vs slightly less than 400 days with the other drugs (P = .002).

Atypical antipsychotic (AAP) medications, anticonvulsants (AC), or combinations of AAP, AC, antidepressants, or lithium performed equally poorly in comparison with lithium alone in delaying relapses (P = .017).

If medication was replaced or stopped by the psychiatrist or by the patient, survival without a relapse was significantly shortened (P < .001).

Relapse rates were significantly affected by psychiatric comorbidities, alcohol dependency, and the number of life events after the index episode. The presence of alcohol use disorders conferred a hazard ratio of 2.7 for relapse (P = .005). The presence of physical comorbidities trended toward an effect but was not statistically significant.

Table. Type of Relapse Corresponding to Index Episode

Index Episode Type of Relapse (%)
  Depression Hypomania Other Syndromes
Hypomania, mania 29.7 51.4 18.9
Depression 74.8 14.6 10.6
Other syndromes 40 8.6 51.4


Dr Simhandl summarized the risk factors for relapse among BP I and II patients:

  • Depressive index episode >70%

  • Failure of lithium treatment

  • Changing/stopping medication

  • Comorbidity of alcohol use disorder in BP I and depression

  • Males with BP I

  • Two or more life events in BP I for depressive relapse

  • Subsyndromal or residual symptoms

  • Cluster C personality disorder

  • Lifetime psychotic symptoms

He concluded there is a need for better medications and treatment strategies.

"After 4 years, 75% have a relapse," he said. The strategies that Dr Simhandl recommended to reduce relapse included better adherence and self-monitoring by patients; early interventions of psychoeducation; interpersonal and social therapies; functional remediation involving mindfulness training; and cognitive behavioral therapy along with family interventions.

Session moderator Birgit Völlm, MD, PhD, professor of forensic psychiatry at the University of Nottingham, United Kingdom, said that little has changed with respect to improving BP outcomes in the past 50 years.

"We're not doing very well in developing new medications. Lithium is still the best one," she told Medscape Medical News. As with other psychiatric disorders, new medications for BP are promoted and advertised, but in the end, they do not deliver what they promise, she said.

The reasons for higher relapse rates in this study with medications other than lithium cannot be determined because Dr Simhandl did not provide detailed information about patients who relapsed, such as their level of drug abuse, Dr Völlm noted.

"But I think it's also a reflection of there not being a lot of breakthroughs in terms of developing new medications. Some of the older agents seem to be better than some of the newer ones. [It's] just frustrating," she said.

Dr Völlm agreed that in addition to medications, it would be worth trying psychological interventions, although there is not a lot of evidence of their efficacy in BP yet, and so they should be studied in that context.

Dr Simhandl and Dr Völlm report no relevant financial relationships.

European Psychiatric Association (EPA) 23rd Congress. Abstract 0181. Presented March 29, 2015.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.