March 6, 2019

Probiotics for Bipolar Disorder Mania

Can they help reduce hospitalization?
Randomized controlled trial finds bipolar patients on probiotics have fewer rehospitalizations compared to those on placebo.


Dickerson F, Adamos M, Katsafanas E, et al. Adjunctive probiotic microorganisms to prevent rehospitalization in patients with acute mania: a randomized controlled trial. Bipolar Disord. 2018;20(7):614-621.


To determine if adjunctive probiotic administration reduces rehospitalization of patients with bipolar disorder 1 after a hospital stay for mania.


Randomized, double-blind, placebo-controlled 24-week intervention, with in-person visits every 4 weeks. Statistics used intention-to-treat analysis.


Lactobacillus GG and Bifidobacterium lactis strain Bb12 combined probiotic (>108 colony forming units [CFUs]) or placebo, once a day.


Sixty-six patients, aged 18 to 65, who had been admitted for mania to psychiatric inpatient or day program hospitals in Baltimore were enrolled upon discharge. Fifty-two participants completed the trial (26 in each group). All participants had been admitted with a diagnosis of bipolar 1 (single manic episode, most recent episode manic, or most recent episode mixed) or schizoaffective disorder, bipolar type (manic or mixed state; DSM-IV-TR) confirmed with the Structured Clinical Interview for Diagnosis for DSM-IV Axis I Disorders (SCID). Participants continued with regularly prescribed medications and other psychiatric treatments as usual.

Exclusion criteria included the following:

  • Substance-induced or medically induced mania
  • HIV infection or other immunodeficiency condition
  • Medical condition affecting brain or cognitive functioning and/or diagnosis of mental retardation
  • Substance abuse or dependence within 3 months prior to study start
  • History of any intravenous drug use
  • Participation in any drug trial within 30 days before study start
  • Pregnant or planning to become pregnant during the study period
  • Documented celiac disease

Study Parameters Assessed

  • Length of time until rehospitalization after discharge
  • Number of total hospitalizations in either group
  • Average number of days hospitalized in study period
  • Total days of hospitalization
  • Psychiatric indicators every 4 weeks: Brief Psychiatric Rating Scale (BPRS), the Young Mania Rating Scale (YMRS), and the Montgomery-Åsberg Depression Rating Scale (MADRS)
  • Inflammatory score using blood drawn at baseline and at study conclusion/24 weeks. Inflammatory score used these variables: immunoglobulin (Ig)G antibodies to the NR2 peptide fragment of the N-methyl-D-aspartate (NMDA) receptor, gliadin, and the Mason-Pfizer monkey virus gag protein; and IgM class antibodies to Toxoplasma gondii.

Primary Outcome Measures

The primary outcomes measured were the effects of probiotics on:

  • Inflammatory score
  • Number and length of rehospitalizations
  • Relative risk of rehospitalization
  • Psychiatric rating scales

Key Findings

  • Fewer total rehospitalization events in probiotic vs placebo group (8 vs 24; hazard ratio [HR]: 0.28, P=0.009)
  • Lower risk of first psychiatric rehospitalization in probiotic vs placebo group (HR: 0.37; 95% confidence interval [CI]: 0.15-0.91, P=0.029)
  • During the 24-week study period, the probiotic group spent a total of 182 fewer days in hospital, and average stay was shorter (mean [SD] length of stay=2.8 days [6.3] vs 8.3 days [12.4], P=0.017)
  • Psychiatric scales (MADRS, YMRS, BPRS) did not differ between groups at any follow-up visit point in the study.
  • No withdrawals from the study due to complaints about the study medication; reported adverse events were equal in both groups

The following finding trended toward significance (P=0.0596): the higher the inflammation at baseline for all patients, the higher the risk for rehospitalization.

Interestingly, the study does not quantify the change of inflammation score in the group as a whole, or compare the 2 groups. However, the authors did note that the participants who had higher inflammation scores (determined as being over the 50th percentile of a separate control group without psychiatric symptoms from another study1) were associated with the largest decrease in rehospitalization risk. The authors comment that “for individuals [who experience mania] with relatively high rates of systemic inflammation, adjunctive probiotic treatment was associated with an approximately 90% reduction in risk of rehospitalization.”

Practice Implications

This study is particularly interesting because it is investigating 2 questions at once: 1) are probiotics helpful in reducing rehospitalization for mania? and 2) if so, does the mechanism have anything to do with inflammation?

The answer to the first question was yes—in this study, adjunctive probiotic treatment for 24 weeks after hospitalization for mania did reduce the risk and length of stay of rehospitalization for mania. It’s a seemingly uncomplicated result, one that could potentially change the standard of care for bipolar disorder, schizoaffective disorder, and other conditions. Even if hospital protocol doesn’t change, outpatient practitioners can use this study’s findings to start long-term probiotic treatment for patients with a history of mania.

Patients in the probiotic group were in the hospital less, but were they any better symptomatically?

The answer to the second question—is the benefit gained by reducing inflammation?—is less clear. No data was published on final inflammation scores, or their changes, nor was there any discussion of significance between the 2 groups. Only 1 comment mentioned that individuals with higher levels of inflammation had decreased hazard ratios associated with rehospitalization—meaning, perhaps, that the protective effect of probiotics was stronger in patients with more inflammation. While encouraging and in line with studies cited in the paper’s introduction, it would be helpful to flesh out this hypothesis a bit more. Clinicians don’t typically score inflammation using specialized tests, such as the IgG and IgM markers for various pathogens used in this study. They rely more on C-reactive protein (CRP) and erythrocyte sedimentation rate (ESR), in combination with physical symptoms, to assess or approximate systemic levels of inflammation.

Note that the probiotics used in the study are fairly low-dose, at 100 million CFUs. Commonly used probiotics in naturopathic medicine are 10 to 25 billion CFU per dose, and some intensive probiotics contain over 150 billion CFU per dose.

Finally, little attention is given to the fact that psychiatric scale outcomes do not differ at any point between the probiotic and placebo group. Patients in the probiotic group were in the hospital less, but were they any better symptomatically? We hypothesize that hospitalization was the primary outcome measure of the study because it speaks the language of the medical establishment: money. Psychiatric hospitalization is grossly expensive, and often costs are even higher than what is reported by hospitals, with 5.5 days of hospitalization for bipolar disorder treatment in the United States costing $4,356 ($7,593 for 9.4 days).2 This averages out to approximately $800 per day. If the treatment of 66 people in this study saved 182 days of hospitalization (275 minus 93 days), that saves almost $145,600. That’s approximately $4,400 per patient in the 33-person probiotic group.

For those of us who see patients on an outpatient basis, the monetary numbers don’t matter quite as much as knowing that something as simple as probiotics might save patients the major life disruption caused by hospital stays. Practitioners both in and out of hospitals should consider a probiotic treatment, based on its low risk profile and large potential for benefit. Continued large-scale studies will be helpful in assessing symptomatic benefit.

Categorized Under


  1. Dickerson F, Adamos MB, Katsafanas E, et al. The association among smoking, HSV-1 exposure, and cognitive functioning in schizophrenia, bipolar disorder, and non-psychiatric controls. Schizophr Res. 2016;176:566‐571.
  2. Stensland M, Watson PR, Grazier KL. An examination of costs, charges, and payments for inpatient psychiatric treatment in community hospitals. Psychiatric Serv. 2012;63(7):666-671.