FormalPara Key Summary Points

Depression has a high societal burden and complex pathophysiology, which includes microbiome-related factors.

Antidepressants may contribute to dysbiosis, which can predispose individuals to Clostridioides difficile infections.

Live biotherapeutic products represent a novel treatment option for recurrent C. difficile infections.

Introduction

Here, we postulate a possible relationship between certain antidepressants and anti-anxiety drugs, their impact on the gut microbiota and the risk of Clostridioides difficile infection (CDI). There are several components relevant to this discussion.

Depression: A Growing Societal Burden

Depression, a serious and recurrent mood disorder, is one of the most diagnosed mental disorders among adults [1, 2]. Characterized by pervasive low mood, diminished interests, and compromised cognitive function, it also presents vegetative symptoms like disrupted sleep or appetite [3]. This condition is marked by a persistent sense of sadness, emptiness, or irritability, along with notable somatic and cognitive changes, severely impairing an individual’s functionality [4]. Depression varies in types, differentiated by symptom severity, occurrence timing, and disease duration [5].

Increasingly recognized as a significant economic and social burden, depression is associated with lower quality of life and increased medical morbidity and mortality [6, 7]. Approximately 20% of the US adult population experiences at least one major depressive episode in their lifetime [8]. Globally, it affects around 322 million people, accounting for 7.5% of all years lived with disability [2]. An upward trend in depression prevalence [2] indicates its emergence as a major global health challenge [7]. Notably, depression is twice as prevalent in women compared to men [9], impacting one in every six adults [3]. The likelihood of developing depression is threefold higher in individuals with a family history of the disorder, although it can also affect those without any familial predisposition [10, 11].

The recent COVID-19 pandemic led the CDC to examine current epidemiology of anxiety and depressions. This involved the US Census Bureau Household Pulse Survey (HPS) with a significant increase in symptoms of anxiety and depressive disorders among adults aged ≥ 18 years during August 19, 2020–February 1, 2021, with the largest increases among adults aged 18–29 years. Notably the rates of depression increased in the early part of pandemic but decreased later. The average anxiety severity scores increased from 2.0 during August 19–31, 2020, to 2.3 during December 9–21, 2020 (APC = 1.5% per wave), reflecting a 13.0% increase in symptoms. Average depression severity score increased from 1.6 to 2.0, reflecting a 14.8% increase. From December 9–21, 2020, to May 26–June 7, 2021, the average anxiety severity score decreased to 1.7 (APC =  − 3.1% per wave), reflecting a 26.8% decrease; during this same period, the average depression severity score decreased to 1.4 (APC =  − 2.8% per wave), reflecting a 24.8% decrease. Nevertheless, the rate of depression and anxiety remained higher than 2019 [12].

Treatment Options for Depression

Various antidepressants and psychotherapies have proven effective for major depression in adults [13]. Treatment typically involves second-generation antidepressants like selective serotonin reuptake inhibitors (SSRIs), serotonin-norepinephrine reuptake inhibitors, serotonin modulators, and atypical antidepressants [14]. SSRIs are the standard, preferred treatment, dominating prescription trends among antidepressant classes [15,16,17]. Additional options include monoamine oxidase inhibitors, tricyclic antidepressants, norepinephrine reuptake inhibitors, and other antidepressants like trazodone, bupropion, and mirtazapine [18].

Increasing Use of Antidepressants

Antidepressants are the most frequently prescribed medication for adults aged 20–59 in the US [14], with about one in ten Americans over 12 years of age taking these medications, and over 60% have been taking antidepressants for more than 2 years [19]. The use of antidepressants has seen a nearly 400% increase across all age groups [20]. Studies highlight a significant rise in long-term prescriptions among patients who were on antidepressant monotherapy, along with a slight increase in the concurrent use of mood stabilizers or antipsychotics [21]. This trend raises concerns about the potential overuse of antidepressants, partially attributed to the condition's overdiagnosis. For example, a Spanish primary care study found a 26.5% misdiagnosis rate of depression [22]. While many patients are on long-term antidepressant therapy, research into their safety and efficacy beyond 2 years remains limited. Discontinuing these medications often leads to a higher risk of relapse or recurrence of depressive symptoms [14].

Safety Concerns with Antidepressant Treatments

The safety profile of SSRIs, especially in the elderly, is currently a subject of scrutiny due to significant side effects like falls, hyponatremia, and stroke [23]. A meta-analysis revealed an increased likelihood of suicide attempts in patients on SSRIs compared to those on placebo [24]. Gastrointestinal (GI) disturbances are among the most common side effects of SSRIs [25], and genetic variations in the 5-HT system can predict these side effects [26]. Although generally well tolerated, SSRIs have been associated with an increased risk of GI bleeding, potentially due to elevated gastric acid secretion and inhibition of serotonin uptake into platelets [27]. A study showed that, in patients with lymphocytic colitis (LC) treated with budesonide, levels of 5-HT were elevated compared to healthy patients, underscoring serotonin's critical role in LC's pathogenesis [28].

The Role of Dysbiosis in Depression

Depression's multifaceted origins are influenced by genetic and environmental factors [4, 29]. Ongoing genetic and non-genetic research aims to unravel the complex mechanisms underlying this disease, with many questions still unanswered [5]. Emerging evidence suggests the gut–brain–microbiome axis may play a role in the etiology of depression. The GI tract, home to an estimated 500–1000 bacterial species [30], possesses a rich genomic diversity, providing more genetic variability than the human genome [31]. The microbiota forms a symbiotic relationship with the gut mucosa, crucial for gut function and structure, nutrition and metabolism, antimicrobial protection, and immunomodulation.

A healthy microbiome can withstand microbial shifts, offering protection against harmful microbes. However, when a tipping point is reached, the microbiome becomes dysbiotic [32], leading to the depletion of Bacteriodetes and Firmicutes and subsequent disruption in microbial diversity [33, 34]. This imbalance may result in the replacement of these phyla by Proteobacteria (e.g., Escherichia coli) and Bacilli, impairing the microbiome's normal functions. The gut–brain axis, a bidirectional communication pathway, influences metabolism through short chain fatty acids (SCFAs) (e.g., butyrate), neurotransmitters [e.g., serotonin and gamma-aminobutyric acid (GABA)], hormones (e.g., cortisol), and immune system modulators (e.g., quinolinic acid). A well-functioning microbiota controls various central nervous system activities, and its perturbation might trigger neuropsychiatric effects, including depression. For instance, Naseribafrouei et al. [35] found significant correlations between gut microbiota and depression, with under-representation of Bacteroidetes in depressed patients and over-representation of certain clades within the genera Alistipes and Oscillobacter, which have been linked to inflammation and depression via inflammatory pathways [36] and neurotransmitter-like activities [37, 38].

The complex relationship between dysbiosis and depression highlights questions as to which occurs initially. Liu et al. [39] postulated the potential role of gut microbes and their metabolism and suggested further research is warranted.

Impact of Various Medications on the Microbiome Composition

The composition of the GI microbiome can be significantly influenced by a variety of medications. Dysbiosis, for instance, has been linked to the use of antibiotics (such as clindamycin, most B-lactams, and quinolones), proton pump inhibitors (PPIs)/H2 blockers, psychotropics, non-steroidal anti-inflammatory drugs, osmotic laxatives, aminosalicylic acids, antimetabolites, calcium-channel blockers, antidepressants, and antipsychotics [40,41,42,43,44,45]. In a notable study, Maier et al. [41] screened 1197 marketed drugs against 40 bacterial strains typically found in the GI tract, discovering that a significant proportion of both antibacterial (78%) and non-antibiotic drugs (27%) exhibited activity against these strains. More than half of the anti-infectives against viruses or eukaryotes exhibited anticommensal activity. This activity varied, with certain bacterial species like Roseburia intestinalis, Eubacterium rectale, and Bacteroides vulgatus being particularly sensitive, and gamma proteobacteria showing resistance.

Both the type of medication and the number of medications impact the gut microbiome. Rogers and Aronoff [43] demonstrated that the type of medication seems to have a greater influence on the gut microbiome than the number of medications. Meanwhile, Ticinesi et al. [44], in their study on polypharmacy in the elderly, reported a significant negative correlation between the number of drugs and microbial diversity, as measured by the Chao index. The usage of PPIs, antidepressants, and antipsychotics showed a strong association with variations in bacterial abundance. To our knowledge, there have not been studies as to whether specific drugs are more of a risk factor with regard to CDI. There is a need to conduct such investigations.

Antidepressants and Their Impact on the Microbiota

Tricyclic antidepressants have been demonstrated to possess antimicrobial activity against a range of bacteria, yeasts, and protozoa [46,47,48,49,50]. SSRIs, such as sertraline, have shown excellent activity against Brucella [51] and synergistic effects with antibiotics against certain Gram-positive species [52]. Sertraline, in particular, has been noted for its broad-spectrum inhibitory effects on various bacterial species including S. aureus, Pseudomonas aeruginosa, and E. coli [53] as well as its synergy with certain antibiotics [54], and its antifungal properties [55,56,57,58].

Impact of Antipsychotics on the Microbiome

In the comprehensive screening study by Maier et al. [41], nearly all subclasses of antipsychotics exhibited anticommensal activity. Antipsychotics are divided into two broad categories: typical and atypical. Studies show typical antipsychotics, derived from in vitro research, have activity against several Gram-positive species. For example, thioridazine is effective against methicillin-susceptible S. aureus, vancomycin-resistant Enterococci, and Mycobacterium tuberculosis [59,60,61]. Fluphenazine has shown substantial activity against both Gram-positive and Gram-negative bacteria [62], while trifluoperazine has been shown to inhibit S. aureus at concentrations between 10 and 50ug/ml and strains of Shigella, Vibrio cholerae, and V. parahaemolyticus at concentrations between 10 and 100 μg/ml [63]. Prochloperazine strongly inhibits Gram-positive species like Bacillus spp. and Staphylococcus spp., with moderate effects on Gram-negative species [64]. However, it is important to consider that these are in vitro findings and may not directly translate to in vivo microbiome situations.

Atypical antipsychotic drugs (AAPs) demonstrate activity against a range of bacteria in various models, which is interesting as they target dopamine and serotonin receptors in the brain, which are absent in bacteria [41]. Phenothiazines show antibacterial activity against Mycobacterium tuberculosis and nontuberculous mycobacteria [65]. Olanzapine, interestingly, inhibited E. coli and Enterococcus faecalis in vitro at supraphysiologic concentrations but did not affect the gut microbiome in a rodent model [66]. Conversely, risperidone led to weight gain, which correlated with an altered gut microbiome in female rats [67]. Aripiprazole, a newer AAP, caused significant changes in microbiota composition, with an increase in various taxa including Clostridium, Intestinibacter, Eubacterium, and Ruminiclostridium in a rodent model [68]. Flowers et al. [69, 70] observed differential abundance of Lachnospiracea in AAP-treated patients and Alistipes in those not treated with AAPs as well as notable diversity differences in female patients treated with AAPs compared to those not treated. Another study found that risperidone led to weight gain and a significantly lower ratio of Bacteroides/Firmicutes over several months in psychiatrically ill children [71].

In cases of polypharmacy, particularly among patients with conditions like depression, the complexity increases. A large study in Japan involving metagenomic analyses of stool samples from 4198 participants revealed that depression was associated with Eubacterium, which changed upon adjustment for diazepam use. The study also found significant associations between the number of drugs taken and microbial functions, with some positive relationships with various transporters and negative associations with amino acid biosynthesis, including butyrate, aligning with a notable depletion of Clostridia species [45].

This evidence underscores the complex nature of drug–microbiome interactions, highlighting the potential broader implications of non-antibiotic drugs on the diverse consortia of microorganisms in the GI tract.

The Link Between Antidepressant Use and Clostridioides difficile

Research reveals scattered evidence about the risk of developing CDI in patients treated with antidepressant medications [72, 73]. An initial study involving 14,719 patients, primarily assessing PPIs and CDI, noted this association but did not investigate specific antidepressant classes [72]. Subsequently, a study of 4407 hospitalized adults identified a significant increase in CDI risk among various antidepressant classes, including mirtazapine, nortriptyline, fluoxetine, amitriptyline, trazodone, and duloxetine [73]. A more recent large-scale study involving over 81 million patients across the US reported increased hospital-acquired CDI risk with the use of mirtazapine (OR 2.50; 95% CI 2.46–2.54), nortriptyline (OR 1.25; 95% CI 1.21–1.28) and trazodone (OR 1.31; 95% CI 1.29–1.33), contradicting earlier findings regarding fluoxetine (OR 0.94; 95% CI 0.92–0.96) [74].

Proposed Mechanisms: Antidepressants and Increased CDI Risk

The exact mechanism by which antidepressants increase CDI risk remains unclear [73,74,75]. Numerous mechanisms have been proposed [74], including serotonin level alteration [75, 76] and immune system dysregulation [77,78,79]. Antidepressants, known for amplifying serotonin (5-HT) and norepinephrine signaling [80], may contribute to GI pathologies through 5-HT signaling dysregulation [76, 81,82,83,84,85,86,87]. Experimental studies in mice suggest that increased 5-HT levels can exacerbate colonic inflammation and affect the microbiota, increasing susceptibility to colitis [76, 81]. Therefore, an increase in serotonin level in individuals using antidepressants could potentially be associated with intestinal inflammation [74,75,76].

In addition, dysregulation of the immune system has also been suggested to be a potential risk factor for CDI development [74]. One study investigated the immunomodulatory effect of SSRIs and found that paroxetine and sertraline decreased human T-cell viability through alternation of gene expression and suppression of signal transducer and activator 3 (Stat 3) and cyclooxygenase 2 (Cox 2) protein expression [78]. Another study found that some antidepressants, including reboxetine, desipramine, fluoxetine, and clomipramine, suppress the production of IFN-γ producing T-cells. These results suggest that antidepressant medications could have immunomodulatory effects by suppressing the production of cytokines [79]. Despite the various suggested hypotheses, the exact pathogenesis through which antidepressant medications increase the risk of developing CDI has not yet been well described in the literature.

Furthermore, some antidepressants may impact gut motility, exacerbating the risk of infection and colitis [88, 89]. For instance, one study found that venlafaxine tended to reduce postprandial colonic contraction [88]. The results of the study could be explained by the fact that certain classes of antidepressant medication have anticholinergic properties, including tricyclic antidepressants, paroxetine, mirtazapine, nefazodone, and monoamine oxidase inhibitors [89, 90].

The antimicrobial effects of certain antidepressants, such as sertraline and monoamine oxidase inhibitors, might also alter the gut microbiota, potentially increasing the risk of microbial alteration or resistance [54, 91,92,93]. This hypothesis is reinforced by studies demonstrating an increased risk of CDI in patients treated with antidepressants [72,73,74].

C. difficile: An Urgent Threat

C. difficile infection presents a major challenge in the US. Approximately 365,000 annual diagnoses led the Centers for Disease Control and Prevention to declare it an “urgent” threat in 2013 and again in 2019 [94]. This infection's global impact is also significant [95].

Patients with CDI commonly experience abdominal pain, diarrhea, fevers, and nausea, among other debilitating symptoms. A primary challenge in CDI management is recurrence, as standard antimicrobials primarily target the bacterial vegetative phase. The spore phase, responsible for transmission and recurrence, often persists post-treatment, leading to repeated infections. Up to 35% of patients treated for CDI experience recurrence, with subsequent recurrences affecting up to 60% of these patients [96, 97]. These recurrent episodes result in significant clinical, economic, and humanistic consequences [98].

Recurrent C. difficile infection (rCDI) is a harbinger of other things beyond the commonly considered abdominal pain and diarrheal symptomatology. With each subsequent infection, the frequency of admissions increases [99] and sepsis and colectomy rates increase. One study assessed commercially insured individuals finding that, with the initial episode, 12-month rates of sepsis and colectomy were 16.5% and 4.6%, respectively. For those with 4 or more episodes (3 recurrences), those rates were 43.3% and 10.5%, respectively [100]. A separate study of individuals 65 and over in the US showed a relatively low generalized 12-month mortality rate of 2.7%; however, with each episode, there was a remarkable increase in CDI-associated deaths, with 16.4% for those with the initial episode, but 39.0% for those with the 3rd recurrence (fourth episode) and beyond [101]. Thus, the impact on patients goes far beyond initial symptoms and presentation.

The Broad Burden of CDI

The recurrent nature of CDI imposes a substantial burden on the healthcare system. It is estimated that individuals with a single episode of CDI visit the emergency room 1.5 times on average, while those with multiple recurrences visit up to 4.6 times [102]. The cost of managing these patients escalates with each recurrence, increasing from an estimated US$71,980 for those with initial infection to $207,733 in those with three or more recurrences [102].

The largest burden of rCDI, goes well beyond the physical manifestations of the infection. As a clinician, we attempt to treat the patient, not just the specific disease. Acknowledging the overarching impact of rCDI on patients is essential. For the patient, the immediate physical impacts of rCDI along with the need to repeatedly seek further medical attention, potential complications such as sepsis and colectomy, and the financial costs of care have a massive impact on patient’s quality of life and psychological state that can be beyond comprehension for clinicians. This impact is all encompassing. One study used a survey format to assess those with active rCDI or a distant history (> 3 months after most recent treatment) considering 5 broad categories including physical (e.g., abdominal pains and diarrhea), mental (e.g., stress, anxiety, post-traumatic stress disorder, depression), adaptation (e.g., changes to activities of daily living dietary habits, socialization skills and also clothing requirements), relationships (e.g., comfortable seeing those friends and relatives they had previously), and productivity (e.g., ability to work). Regardless of timing of the most recent rCDI (currently active or greater than 3 months in the past), more than 50% of participants expressed a significant impact on all categories except productivity, where those with active rCDI were impacted greater than 50%, but approximately 30% for those with a distant history [103]. From this study, we can see the profound impact rCDI has on patients and their daily life, even with a history of being successfully treated at least 3 months prior. Given the known associations of CDI and the microbiota, there are theories that perhaps these circumferential impacts on patients might be a microbiota-driven effect, but further research is needed.

The impact of rCDI extends beyond physical symptoms, significantly affecting patients' quality of life and psychological well-being. In particular, patients with rCDI appear to have remarkably high rates of depression. Using the National Inpatient Sample (2018 and 2019), the largest publicly available all-payer inpatient care database in the United States, 21.5% of the 84,055 hospitalizations with a diagnosis of rCDI had a co-diagnosis of depression compared to 14.2% in those without CDI. CDI patients had 56% greater chance of having a diagnosis of depression compared with the general population. In fact, patients with CDI had significantly higher odds of having depression than those with congestive heart failure (OR 1.38, 95% CI 1.32–1.44; p < 0.001), diabetes (OR 1.32, 95% CI 1.25–1.38; p < 0.001), end-stage kidney disease on dialysis (OR 1.49, 95% CI 1.42–1.55; p < 0.001), and metastatic cancer (OR 1.89, 95% CI 1.81–1.98; p < 0.001) [104].

The CDIFF32 questionnaire, a tool assessing the impact of rCDI on physical, mental, and social well-being domains, underscores this broad impact [105]. This assessment tool, introduced in 2016, has been instrumental in understanding and quantifying the effects of rCDI on patients' lives [106].

The CDIFF 32 tool comprises 32 questions in 3 separate domains, which are cumulated to yield an overall score.

There appears to be an association between the health-related quality of life tool, especially the mental domain and the response to C. difficile therapy involving restorative microbiota.

Live Biotherapeutic Products in rCDI Treatment

Live biotherapeutic products (LBPs), defined as non-vaccine biologicals composed of live microorganisms, have been approved for rCDI treatment. In late 2022 and early 2023, the FDA approved two such products: fecal microbiota, live-JSLM (Rebyota™; RBL) and fecal microbiota spores, live-BPRK (Vowst™; VOS). These products aim to supplement microorganism deficiencies in the colon, targeting the spore phase of C. difficile to reduce recurrence.

RBL, consisting of a broad consortium of microorganisms including Bacteroidetes and Firmicutes, is administered via a single rectal installation post-standard care antimicrobial and a brief washout period. The PUNCH-CD3 trial, a multicenter, double-blind, randomized, placebo-controlled trial, evaluated RBL in patients experiencing their first recurrence (second episode) or subsequent episodes of rCDI. Participants received at least 10 days of standard care antimicrobial followed by RBL. At 8 weeks, Bayesian analysis, leveraging data from the PUNCH-CD2 phase 2 study, revealed an efficacy rate of 70.6% for RBL compared to 57.5% for the placebo group (posterior probability of superiority = 0.991) [107]. To underscore RBL's effectiveness in reducing rCDI, a notable enhancement in patients' quality of life was reported, gauged by the CDIFF32 questionnaire [108]. Within the PUNCH-CD3 trial, a cohort of patients completed CDIFF32 at weeks 1, 4, and 8. In both arms of the study, there were significant improvements in total, physical, mental, and social categories. In the mental domain, RBL-treated patients showed a significantly greater improvement than placebo. When adjusting for confounding factors, RBL showed greater improvement than placebo in the total, physical, and mental domains. Most interestingly, within the non-responders, the cohort that received placebo had no change in their CDIFF32; however, those who received RBL showed numerical trends in improvement in total, physical, mental, and social domains [108].

This finding is fascinating since this implies that, perhaps, the microbiota restoration using RBL, which includes a wide array of microorganisms such as Bacteroidetes, may adjust the microbiome diversity/composition and address not only C. difficile but other elements that impact mental and social elements for patients.

VOS, focusing on a defined consortium of microorganisms, has also shown promising results in reducing rCDI and improving quality of life outcomes [109, 110]. In the pivotal phase 3 ECOSPOR III trial, CDIFF32 scores were collected at baseline and at weeks 1 and 8. The VOS patient cohort had greater overall improvement at both time frames compared with placebo and the impact on the total and physical scores at weeks 1 and 8 were significantly greater for VOS compared with placebo [110]. These results show that VOS, despite having a narrow consortium, can positively impact overall and physical elements of quality of life. The ECOSPOR III trial had infrequent recurrences in the VOS arm, so analysis and comparison of those who recurred after receiving VOS were limited by small comparators. Therefore, trends that were seen with RBL could not be correlated with this product among those who did recur. In theory, with a narrow consortium, the potential secondary benefits of VOS on the microbiota that might inadvertently improve quality of life, should theoretically be less than a product like RBL which has a broad consortium, but there are no data to support those claims at the current time.

These trials highlight the potential of LBPs in not only preventing CDI recurrences but also improving overall patient well being. As our understanding of the microbiota's impact on mental health and disorders like depression evolves, these findings may guide future therapeutic approaches.

Discussion

The current trend of increasing depression and anxiety in tandem with changes in the epidemiology of C. difficile prompted study of the role of previously unrecognized drugs which may contribute to gastrointestinal dysbiosis. The disturbance of the normal gut flora has been shown to adversely affect the gut–brain axis by virtue of changing the levels of critical compounds such as GABA.

We hypothesize that it maybe the underlying depressive state which contributes to patients being at risk of CDI, as there seem to be some commonalities in terms of dysbiosis compared with those with depression. Most interestingly, those who receive LBPs seem to respond strongly in terms of the mental impacts. This combination of concepts leads to our theory of the connection. In future research, it would be incredibly helpful to differentiate those with long- versus short-term depression and see how they fared following LBP administration.

In summary, the complexities of two seemingly disparate conditions, depression and rCDI, may be linked by dysbiosis of the GI tract. Some drugs used to manage depression and other psychiatric conditions have antimicrobial effects which can lead to disturbance of the gut microbiome and may pre-dispose patients to CDI. This shift in bacterial flora creates a disturbance in the metabolites such as GABA which drive the gut–brain axis. In turn, this brain dysfunction may exacerbate depression and perhaps related conditions. This cycle can be further driven by more antidepressants, increased dose, or new agents, leading to further dysbiosis of the GI tract. This is illustrated in Fig. 1.

Fig. 1
figure 1

Cycle of depression, dybiosis and C. difficile

Full size image

A multi-disciplinary approach to managing these recurrent conditions is needed to minimize the dysbiosis, and perhaps LBPs may play a role.