A systematic review on gut–brain axis aberrations in bipolar disorder and methods of balancing the gut microbiota

Abstract Background Bipolar disorder (BD) is a mood disorder that affects millions worldwide. Up to half of the diagnosed patients are reported to not receive adequate treatment. This study aims to assess the relationship between the gut–brain axis and BD and to discuss and compare the efficacy of varying methods of balancing gut microbiotas in BD. Methods Using PubMed, Embase, and Google Scholar from November 2021 to February 2022, we found 5310 studies on gut microbiota and its relation to BD. Using our inclusion criteria, 5283 studies were excluded. A total of 27 full‐text articles were assessed for eligibility. Also, 12 articles that met our criteria and eligibility criteria reported on 613 BD patients. Results Most studies analyzed found an overall difference in gut microbiota composition in bipolar patients compared to healthy controls, though the alterations found were not consistent. Differences in Lactobacillus, Faecalibacterium, and Ruminococcus abundance in BD compared to controls were found to be the most consistent across a few of the studies, but their effects on the gut–brain axis conflicted. Probiotic supplementation was found to lower patient rehospitalizations and significantly improve depressive symptoms and cognitive impairments among patients with BD. Conclusions Multiple studies included in this review point toward a possible link between BD and the gut microbiota. Probiotic supplements and other gut‐balancing therapies could serve as effective adjunctive methods for the treatment of BD. Notable limitations of the studies included for analysis were small sample sizes and majority observational study designs. Furthermore, the microbiota aberrations found in patients with BD were not consistent across multiple studies. Despite these limitations, our findings demonstrate the need for further research regarding the relationship between aberrant gut microbiota profiles and BD, as well as the effectiveness of gut balancing methods as adjunctive treatments.


INTRODUCTION
Bipolar Disorder (BD) affects 45 million people worldwide (World Health Organization, 2022). The National Institute of Mental Health has reported that up to 50% of patients with BD fail to receive adequate treatment for their mental illness, leaving over 2 million US patients untreated (Bipolar Disorder-Fact Sheet, 2022). Globally, it has been reported by the World Health Organization (2022) that 75% of people with mental health disorders, such as BD, lack access to adequate treatment or care. It has also been found that even with adequate treatment, 37% of patients will experience a breakthrough affective episode within a year, and 60% will experience a relapse of affective symptoms in 2 years (Da Costa et al., 2016). Bipolar symptom relapse and worsening disease progression has become a growing concern, as studies have found that a significant proportion of patients are experiencing symptom relapse despite the use of adequate pharmacological agents (Fountoulakis, 2012;Geddes & Miklowitz, 2013;Perlis et al., 2006). These observed trends highlight the need for further research into possible adjunctive methods of treatment for BD.
The neurobiological mechanisms that drive BD symptoms are poorly understood, which could be a possible contributor for increasing rates of BD symptom relapse (Aizawa et al., 2019;Geddes & Miklowitz, 2013;Perlis et al., 2006). A possible environmental factor that has yet to be adequately studied in reference to BD is the contribution of intestinal microbiota aberrations. An emerging concept of a microbiota-gut-brain axis is being utilized to highlight the significant effect gut microbiota composition has on bidirectional gut-brain communication pathways (Cryan & Dinan, 2012;Painold et al., 2019).
Research is revealing that the gut microbiota is key for maintaining homeostasis, and alterations in its composition can lead to a number of disease states including those of the central nervous system (CNS) (Cryan & Dinan, 2012). In fact, the relationship between gut microbiota dysbiosis and host illness has long been linked to chronic conditions such as metabolic syndrome, irritable bowel syndrome, inflammatory bowel disease, and anorexia nervosa. (Cryan & Dinan, 2012;Mayer, 2011;Painold et al., 2019). Gut microbiota alterations have also been linked to other psychiatric disorders including BD (Coello et al., 2019;Hu et al., 2019;Painold et al., 2019). Recurring effective episodes in BD are associated with a progressive decline in cognitive and executive function, and there is growing evidence that supports a relationship between cognition and microbiota via the gut-brain axis (Cryan & Dinan, 2012;Forsythe et al., 2010;Misiak et al., 2020). Thus, the link between BD and the gut microbiome may be a useful treatment alternative, as mounting research has found significant intestinal microbiota alterations in patients with mood disorders (Aizawa et al.,  Disruptions to the neural-gut pathway have shown a clear link to alterations in the physiological stress response and behavior (Cryan & Dinan, 2012;Forsythe et al., 2010), and the high rates of cooccurring stress related to psychiatric conditions among patients with gastrointestinal (GI) disorders highlights the gut-brain axis' role in pathophysiology (Cryan & Dinan, 2012;Mayer, 2011). Research has found that stress, including social disruption, influences gut microbiota composition, and the bidirectional communication between the gut and the CNS plays a role in stress reactivity (Bailey et al., 2011;Painold et al., 2019). The body's stress response includes immune modulation (such as cytokine release), which has been linked to the development of anxiety and depression (Bailey et al., 2011). There is also evidence of a link between stress exposure and the GI barrier; increases in stress increase gut permeability (Misiak et al., 2020). The blood-brain-barrier (BBB)'s integrity depends on the intestinal microbiota, as alterations in their composition could lead to BBB impairment (Misiak et al., 2020).
The major phyla of the gut microbiome include Firmicutes, Bacteroidetes, Actinobacteria, Proteobacteria, Fusobacteria, and Verrucomicrobia, and two phyla, Firmicutes and Bacteroidetes, which represent 90% of the gut microbiota (Rinninella et al., 2019). The Firmicutes phylum is made up of over 200 genera, including Bacillus, Lactobacillus, Clostridium, Ruminicoccus, and Enterococcus, of which Clostridium constitutes 95% (Rinninella et al., 2019). Bacteroidetes major genera are Bacteroides and Prevotella (Rinninella et al., 2019). Actinobacteria is the least abundant phylum and is mainly comprised of Bifidobacterium (Rinninella et al., 2019). The gut microbiota composition is thought to vary in the same individual as well as between different individuals due to factors such as age, environmental factors, metabolic factors, and antibiotic use (Rinninella et al., 2019). A few studies have investigated the specific microbial gut composition changes that occur with stress.
An animal study found that there was a decrease in Bacteroides species (spp.) abundance as well as an increase in Clostridium abundance among stress-exposed mice as compared to controls (Bailey et al., 2011).
Chronic psychological stress has been associated with an increase in Enterobacteriaceae, Escherichia coli, and a decrease in Lactobacilli spp. (Rinninella et al., 2019).
As initially discussed, chronic stress has been found to influence gut microbiota composition through proinflammatory cytokines, such as interleukin-6 (IL-6), which was found to be correlated with higher levels of Lactobacillus, Bacilli, and Streptococcaceae . As stress induces inflammation, biological indices have confirmed inflammation as a major contributor to the pathogenesis of BD. Additionally, increased levels of cortisol in response to stress have been linked to the manic phase of BD (Misiak et al., 2020;Van Den Berg et al., 2020;Τournikioti et al., 2018), including a study that found that cortisol levels were increased for months prior to a manic relapse. This appears to coincide with research that found a significant association between cortisol and ACTH with BD. A notable finding is the relationship between HPA axis dysfunction (which was linked to developmental stressors including childhood trauma), and the clinical presentation of BD (Belvederi Murri et al., 2016).
The shifts in the microbiome composition observed in patients with BD could be of notable significance, and addressing them may contribute to symptom resolution. In this systematic review, we aim to assess how components of the gut-brain axis contribute to BD, to determine how balancing gut microbiotas affect the severity of symptoms, and to discuss and compare the efficacy of varying methods of balancing gut microbiotas in BD.

Search strategy
For this systematic review, we searched PubMed, Google Scholar, and Embase for articles published between April 2017 and January 2021.
All authors initially selected articles via manual screening of abstracts and searched the reference lists of the chosen articles for additional information that could be applied to our investigation. Articles that investigated the gut microbiome and its link to mood disorders were included.

Study selection
Studies were selected according to the following criteria: population, outcome(s) of interest or condition, study design, and context: • Population: Patients with diagnosed BD.
• Outcome(s) of interest or condition: The first outcome was determining an association between the gut-brain axis and BD. The second outcome was to document any noted changes in symptom severity of patients after correcting the gut microbiota in affected patients. And last, to evaluate existing methods of gut microbiota balancing in bipolar patients. • Study design and context: Eligible studies were randomized controlled trials, double-blind controlled trials, cohort studies, crosssectional studies, and case-control studies.

Inclusion criteria
Human studies that examined and reported (a) a link between gut microbiotas and the physiologic causes of BD (e.g., manic, hypomanic, or depressive symptoms), (b) current documented changes to gut microbiota using probiotics, supplements, fecal implantation, or diet, (c) any functional, depressive, and/or manic symptom changes observed with gut microbiota balancing.

Exclusion criteria
Systematic review studies, editorials, case studies, commentaries, and articles irrelevant to either the development or treatment of BD (e.g., manic, hypomanic, or depressive symptoms) or the behavioral changes associated with gut microbiota.

Data collection & study assessment
All authors independently reviewed the abstracts of all the articles identified. We divided the articles into two groups of "Adopted" and "Not Adopted" based on the inclusion criteria. Then, we screened the "Adopted" articles and created a spreadsheet to include them to be used for our research work. After the final selection process, a Preferred Reporting Items For Systematic Reviews and Meta-Analysis (PRISMA) flow chart was generated following the PRISMA guidelines. Resources for this review were obtained via qualitative and quantitative analysis.

Data synthesis & analysis
Quantitative and qualitative studies based on original research that examined the gut microbiota and brain relationship in BD and nonpharmacologic methods for treating bipolar symptoms were included.
The data synthesis was conducted in a detailed summary of the included studies by table construction. The quantitative data were extracted using Microsoft Word. The data were grouped according to the objective of this study.

RESULTS
Our study included 613 BD patients (mean age: 39.25 y, women: 58%), 39 first-degree relatives, and 321 healthy controls (mean age: 36.4 y, women: 57%). Other characteristics such as BMI, medication status, lifestyle modifications, and BD I vs. BD II diagnosis were collected during this analysis and are listed in Table 1. The study selection flow chart is given in Figure 1.

TA B L E 1
Summary of the studies included in our review.

Type of study Study participant characteristics
Gut-brain axis changes in BD BD symptom changes after gut microbiota alterations Effectiveness of gut microbiota balancing methods for BD

Interventional studies
Eslami Shahrbabaki et al. F I G U R E 1 Preferred reporting items for systematic reviews and meta-analyses (PRISMA) study selection flow chart.

Gut-brain axis changes in BD
The following were observed in this analysis: (1) the composition of the gut microbiome; (2) the neurotransmitter pathway differences of gut microbiota; and (3) the gut microbiota composition and its relation to inflammation and serum lipids.

Gut microbiota composition
There are significant differences in gut microbiota composition in patients with BD as compared to healthy controls (HCs) (Coello et al., 2019;Evans et al., 2017;Hu et al., 2019;Painold et al., 2019). An interesting finding of this study was that newly diagnosed patients with BD were 2.9 times (OR 2.9, 95%CI: 1.6−5.2, Cluster, Enterobacter spp., and Clostridium Cluster IV were significantly higher in the BD group compared with HCs (p = .030, p < .001, p < .001, p < .001, and p < .001, respectively) (Lu et al., 2019). The Bacteroides/Enterobacter (B/E) ratio of the BD group was significantly lower than the HCs (p = .001).
Notably, one study found differences in gut microbiota composition between patients with BD with and without depressive symptoms .

BD symptom changes after gut microbiota alterations
We analyzed studies that investigated how balancing gut microbiotas affected the severity of bipolar symptoms, such as mania and depressive symptoms, and found conflicting results.
Eslami Shahrbabaki et al. (2020) evaluated bipolar I patients after 8 weeks of probiotic use. Blind randomization methods were used to divide patients into a placebo and probiotic groups. Patients in both groups were able to receive lithium oxide (max: 900 mg/day), sodium valproate (max: 1200 mg/day), and risperidone if needed. The probiotic group received a probiotic capsule containing 1.8 × 109 CFU/capsule Bifidobacterium bifidum, Bifidobacterium lactis, and Bifidobacterium langum, and Lactobacillus acidophilus. The YMRS mania scale and Hamilton's depression scale questionnaires were completed by a psychiatry resident at the start of the study, before the intervention (probiotic capsule), at the 4-week mark, and at the 8-week mark.
While both the probiotic and placebo groups displayed a decrease in mania and depression questionnaire scores throughout the duration of the study, it was found that patients in the probiotic group displayed significantly higher decreases in symptom severity via the Young Mania (p value = .001) and Hamilton (p-value = .001) questionnaire scores throughout the three measured points of the study. However, even though the probiotic group had a more significant improvement in questionnaire scores throughout the 8-week study as compared to the placebo group, there were no significant differences in overall mania (pvalue = .2) and depression (p-value = .5) scores between placebo and probiotics patients with type 1 BD.
In their study,

BD symptom changes after gut microbiota alterations and effectiveness of gut balancing methods
We analyzed studies that investigated how balancing gut microbiotas affected the severity of bipolar symptoms, such as mania and depressive symptoms, and found conflicting results. Other parameters, such as cognition, mood, and hospitalization, were investigated. While both the probiotic group and placebo group displayed a decrease in mania and depression questionnaire scores throughout the duration of the study, it was found that patients in the probiotic group dis-played significantly higher decreases symptom severity via the Young Mania (p value = .001) and Hamilton (p-value = .001) questionnaire scores throughout the three measured points of the study. However, even though the probiotic group had a more significant improvement in questionnaire scores throughout the 8-week study as compared to the placebo group, there were no significant differences in overall mania (pvalue = .2) and depression (p-value = .5) scores between placebo and probiotics patients with type 1 BD. Dickerson et al. (2018)

DISCUSSION
The pathogenesis of BD remains unclear. However, it has been shown that gut microbiotas play an important role in its onset and progres-  Coello et al., 2019;Cryan & Dinan, 2012;Da Costa et al., 2016;Evans et al., 2017;Forsythe et al., 2010;Fountoulakis, 2012;Hu et al., 2019;Lai et al., 2021;Lu et al., 2019;Mayer, 2011;Misiak et al., 2020;Painold et al., 2019). The gut-brain axis is a system that has numerous pathways, and any change in the molecular components of this system can result in disease for the host (Appleton, 2018). Painold et al. (2019) investigated how gut microbiota composition in BD relates to disease states and found that Lactobacillus was found to have a direct correlation with higher tryptophan levels, essentially affecting neurotransmitter levels known to play a role in BD pathogenesis. This gut-brain axis connection between the gut microbiome and brain neurotransmitters illustrates how some bacteria can be used as "psychobiotics" . In addition to cytokines, neurotransmitter precursors like tryptophan have been found to be linked to patients with BD through microbiota alterations (Appleton, 2018;Forsythe et al., 2010;Lai et al., 2021;Lu et al., 2019;Misiak et al., 2020). The relationship between the gut-brain axis and neurotransmitter availability is further highlighted by associations found between gut microbiota composition and tryptophan metabolism Valladares et al., 2013 inflammation may be a key factor in BD pathogenesis and symptomol-ogy as highlighted by previous studies (Bailey et al., 2011;Borthakur et al., 2010;Dickerson et al., 2018;Dinan & Cryan, 2017;Mayer, 2011).
Additionally, Dickerson et al. (2018), admit that their study could not have been able to capture symptom severity throughout the experimental period. As some of the participating patients had mixed BD, it is hard to tell whether the improvements noted by Dickerson et al. could have been due to the mediation of bipolar depressive symptoms.
Nonetheless, these findings support the relationship that has been found between gut microbiota, altered tryptophan metabolism, inflammation, and the HPA axis . that should be explored further in future studies.
While gut microbiota aberrations were discovered in patients with BD when compared to HCs, the species attributed to BD symptoms were not consistent across studies. Painold et al. (2019), concluded from their study that Faecalibacterium could be a distinguishing feature between patients with BD and HCs. HCs displayed higher levels of the Faecalibacterium genus and the Ruminococcaceae family compared to HCs. Their findings correlated with Evans et al. (2017), whose study associated adequate levels of Faecalibacterium with better health outcomes among patients with BD. Aside from these similar findings, however, the associated microbiota compositions varied.
Significant associations between higher levels of Actinobacteria, Coriobacteriaceae, Faecalibacterium, Enterobacter, Flavonifractor, Clostridium, and Bacteroides were found in patients with BD Lu et al., 2019;Painold et al., 2019). Studies also found differences in microbiota diversity specific to bipolar symptoms and illness severity . BD patients experiencing depressive symptoms were found to have higher levels of Enterobacteriaceae, while Clostridiaceae and Roseburia were found to be more abundant in less symptomatic patients with BD (Evans et al., 2017;Hu et al., 2019). An important factor to note about the differences in the gut diversity among patients with BD is the fact that patients were taking psychiatric medication throughout the duration of these studies. It has been shown that psychiatric medication also affects gut microbiota composition , and the lack of standardization of treatment decreased Lactobacillus levels were associated with depression and the administration of ketamine in rat species has been found to reduce depressive-like behaviors as well as suicide risk. Getachew et al. also found Lactobacillus to be increased in ketamine-treated rats. Despite the positive outcomes reported in these initial studies, further research is needed as these treatment options are understudied.

LIMITATIONS
While our initial search yielded many results, only 12 were original experimental studies involving BD patients. Our cumulative sample size was 613 bipolar (BD) patients and 312 HCs. This is an ideal cumulative size, however, outside of one study (Evans et al., 2017), which had a sample size of 115 bipolar patients, and 64 control subjects, the other studies had relatively small sample sizes even before losing subjects to follow-up. This left room for sampling error and left all but one of the articles included in this study with reduced design power. Additionally, most of these studies were of cohort, case-control, or cross-sectional design, preventing causal conclusions being drawn. Due to the different study designs, some potential biases could be present. Last, there were inconsistencies in the amount of potential confounding data collected per study, which could also make room for additional biases. While a few findings were replicated between studies, overall, they were very inconsistent. Despite these limitations, our study has demonstrated the need for further research regarding gut balancing methods and their impact on the treatment of BD.

CONCLUSIONS
In this systematic review, we aimed to assess how components of the gut-brain axis contribute to BD, to determine how balancing gut microbiotas affect the severity of symptoms, and to discuss and compare the efficacy of varying methods of balancing gut microbiotas in BD. There are significant associations between BD pathogenesis and alterations of gut microbiome diversity via immunomodulation, the HPA axis, and neurotransmitter alterations. Specific microbiome species could serve as therapeutic targets as adjunctive therapies for patients with BD to manage depressive and manic symptoms. Additionally, other gut-balancing methods, such as ketamine, charcoal, and vitamin D supplementation, have also shown promise as adjunctive agents. Despite these promising findings, current studies investigating the gut-brain axis relationship among patients with BD are limited by small sample sizes and a lack of adequate replication of results. However, the conclusions drawn from this review highlight that there may be specific bacteria common to patients with BD that could serve as screening tools and therapeutic targets, of which future studies could investigate.
We recommend that future studies place emphasis on the following: 1. Identify common microbiome alterations that can be linked to BD.
The studies included in this review reported microbiome alterations among patients with BD, however, those alterations varied.
There have yet to be additional studies that have replicated the findings of the articles mentioned.
2. Additional investigations into gut-balancing methods aside from probiotics.
3. Further investigation of how gut microbiome alterations serve as epigenetic factors of the pathogenesis of BD. One of the included studies found a link between gut microbiota alterations and the epigenetic impact on the gene ARNTL, which is thought to play a role in BD pathogenesis . Additional studies that replicate this finding or find other possible epigenetic impacts would be impactful.

ACKNOWLEDGMENTS
We are very grateful to Gibson Anugwom and Funso Oladunjoye for their expert guidance and support throughout our analysis.

DISCLAIMER
The products used for this research are commonly and predominantly used products in our area of research and country. There is absolutely no conflict of interest between the authors and producers of the products because we do not intend to use these products as an avenue for any litigation but for the advancement of knowledge. Also, the research was not funded by the producing company rather it was funded by the personal efforts of the authors.

CONSENT
As per international standards or university standards, Participants' written consent has been collected and preserved by the authors.

CONFLICT OF INTEREST STATEMENT
The authors have declared that no competing interests exist.

DATA AVAILABILITY STATEMENT
Data sharing not applicable to this article as no datasets were generated or analyzed during the current study.

PEER REVIEW
The peer review history for this article is available at https://publons. com/publon/10.1002/brb3.3037