1. Introduction
Clostridioides difficile, clinically known as
Clostridium difficile, is an anaerobic, gram-positive, spore-forming, toxin-producing bacillus.
C. difficile colonizes the large intestine and causes disease predominantly through its production of cytoskeletal-modifying exotoxins. These toxins mediate disease by triggering colonocyte death, causing the loss of intestinal barrier function, and provoking neutrophilic colitis [
1].
C. difficile is transmitted through the fecal–oral ingestion of spores. Although generally associated with nosocomial transmission, the incidence of
C. difficile infection in pediatric populations has increased dramatically over the past three decades in hospital, community, and outpatient settings [
2].
C. difficile infection (CDI) is thought to be the result of gastrointestinal dysbiosis that disrupts the resident microbiota, thereby creating an environment that is either favorable to colonization by
C. difficile spores or to overgrowth by indigenous
C. difficile residents [
3]. The typical insult is antibiotic use, though other factors such as advanced age, prolonged hospitalization, proton-pump inhibitor use, an immunocompromised state, and other medical comorbidities have been associated with an increased risk of
C. difficile [
2]. In the pediatric population,
C. difficile infection is associated with prior antibiotic and proton-pump inhibitor use [
4].
The conventional first-line treatment for
C. difficile in both adult and pediatric populations is antibiotic therapy, specifically metronidazole or vancomycin. Despite this, there is evidence that antibiotics can be ineffective in resolving recurrent
Clostridioides difficile infection, and instead could potentially be counteractive by perpetuating underlying dysbiosis [
5,
6].
The recurrence rate of
C. difficile in the adult population after metronidazole or vancomycin is estimated to be 20.2% and 18.4%, respectively [
7]. As a result, fecal microbiota transplantation (FMT) has emerged as a recommended treatment by the Infectious Disease Society of America (ISDA) and the Society for Healthcare Epidemiology of America (SHEA) for patients with multiple recurrences of CDI who have failed appropriate antibiotic therapy [
2]. This recommendation is supported by data that estimate that the cure rate of FMT for recurrent
C. difficile infection (rCDI) is between 89.1 and 89.7% (unweighted pooled resolution rate to weighted pooled resolution rate, respectively; 95% CI 84–93%) [
8].
Given that the rate of recurrent
C. difficile infection in the pediatric population is similar to that of adults (20–30%) [
9], FMT has also emerged in this population as a highly effective alternative to antibiotic therapy [
10]. A recent retrospective study of 372 patients from 11 months to 23 years old who underwent FMT reported that it was 81% effective with a 4.7% rate of serious adverse events during the 3-month follow-up period [
11].
Though studies in pediatric rCDI patients have shown a post-FMT increase in bacterial diversity in its recipients up to 6 months after transplant [
12,
13], the mechanism through which this healing occurs is largely unknown. In adults, short-chain fatty acids (SCFAs), a byproduct of microbial metabolism, have been implicated as a potential mediator in the post-transplant restoration of microbial diversity and richness in rCDI patients [
14,
15].
In short, SCFAs are a subset of fatty acids that are byproducts of the microbial fermentation of partially and nondigestible carbohydrates, more specifically polysaccharides, or, in the case of valeric acid, isovaleric acid, and isobutyric acid, amino acid derivatives [
16]. The butyrate production pathway involves the arrangement of acetyl CoA, which is converted to acetoacetyl CoA [
17]. The acetoacetyl CoA is then converted into butyryl CoA through two different enzymes, which makes butyrate [
17]. Propionic acid is formed from two products, either succinate or 1,2 propanediol, which are products of broken-down carbohydrates [
18]. Acetate and formate form through CO
2 reduction reactions in the Wood–Ljungdahl pathway [
19]. Isovaleric acid is formed through amino acid catabolism as an end product of oxidative phosphorylation [
20].
We chose to focus our project on the overall production of SCFAs rather than delving into individual pathways. SCFAs are able to interact with the human intestinal lumen as a fuel source, through G-protein coupling receptor (GCPR) pathways and histone deacetylation inhibition [
21]. Through these interactions, SCFAs have widespread systemic effects. Although our complete understanding of SCFAs’ role in the human body is yet to be actualized, prior studies have implicated various roles, including altering chemotaxis and phagocytosis [
21,
22]; inducing reactive oxygen species; modifying cell function and proliferation; altering gut integrity [
23]; and acting as anti-inflammatory [
21,
24], antitumorigenic, and antimicrobial molecules [
21]. In this study, we evaluated the impact of FMT on bacterial producers of SCFAs in pediatric patients with rCDI over a 12-month post-op period. Due to the introduction of diverse microbiota from FMT, we hypothesized that levels of SCFAs would increase post-op in our study population to match the profiles of healthy controls.
2. Materials and Methods
This study was approved by the Children’s Hospital of Los Angeles’ institutional review board as protocol number CCI-11-000148, and written informed consent was obtained from all participants prior to the initiation of study activities. To study the changes that occur in pediatric patients with rCDI treated with FMT, we conducted a longitudinal, retrospective, cohort study at Children’s Hospital Los Angeles and its referral sites.
We included a total of 9 subjects with rCDI who underwent fecal microbiota transplantation between January 2014 and December 2015 with ages ranging from 2 to 17 and 19 healthy controls with ages ranging from 2 to 16.
4. Discussion
This study analyzes longitudinal SCFA levels up to 12 months post-FMT in pediatric patients who have recovered from rCDI. Both the longitudinal nature of the study and the results generated shed valuable light on the mechanism and timing of FMT in the treatment of pediatric patients with rCDI.
We demonstrated, by comparison with healthy controls within a similar age range, that decreased overall fecal SCFA levels and decreased individual levels of acetic acid, butyric acid, isovaleric acid, and propionic acid are associated with recurrent
C. difficile colitis in the pediatric population. This was unsurprising, as a previous study has documented similar results in an adult population [
32].
We also witnessed a recovery in SCFA levels from baseline to month 12 in our rCDI pediatric cohort, especially isovaleric acid and propionic acid at month 1. This is correlative with disease activity, as all subjects included in this study had no evidence of
C. difficile colitis up to the study endpoint of 12 months, as well as clinical symptom resolution. Although prior literature has associated FMT success in rCDI humans with the restoration of SCFA levels [
32,
33], to our knowledge, only one has analyzed this change in humans longitudinally [
32]. The aforementioned study’s follow-up period was limited to 6 months, though, and restricted to an adult population [
32]. Therefore, this was the first study to analyze long-term (12 months) SCFA changes in
any population. Moreso, there have been no prior studies of our knowledge that have studied longitudinal changes in SCFA levels post-transplant in pediatric rCDI patients in any capacity (short-term or long-term).
The recovery of acetic acid, butyric acid, isovaleric acid, and propionic acid levels have a variety of implications for the mechanisms by which the gut heals from
C. difficile. Studies in the past have noted that the SCFAs—acetate, butyrate, propionate, and valerate—have a direct inhibitory role on the growth rate of
C. difficile in culture [
14,
34]. In hamster models, a higher level of SCFAs is protective against
C. difficile infection. This was consistent with the data that have proven that antibiotic use decreases SCFA levels, thereby suggesting that an SCFA-depleted gut can provide an environment favorable for
C. difficile growth and germination [
35]. Therefore, SCFAs are key to eradicating the pathogenic insult and preventing further disease.
Acetic acid, butyric acid, and propionic acid also have a variety of immune regulatory roles. They act as anti-inflammatory molecules through a variety of mechanisms: inhibiting the lipopolysaccharide-stimulated release of tumor necrosis factor α [
24], inhibiting the NF-kB pathway [
24], stimulating neutrophil chemotaxis [
22], and regulating interleukin levels [
36,
37,
38].
Higher acetic acid levels have been shown to decrease the permeability of the gut, thus ensuring the integrity of the mucosal epithelium [
23]. Thus, the decreased pre-treatment levels of acetic acid in our rCDI cohort, combined with the fact that
C. difficile toxins have been implicated in causing the disruption of epithelial paracellular barrier function [
39], might explain a mechanism through which
C. difficile pathogenesis occurs. Similarly, since we conclude that acetic acid levels increase in response to FMT, perhaps acetic acid is a key mediator of FMT’s anti-
C. difficile action in the pediatric population.
In addition to its role in immunity, butyrate is an important metabolite in the gut lumen, being the main source of energy for the colonocyte [
40]. It participates in a variety of roles in epithelial maintenance, including growth, differentiation, protection, and repair [
41,
42,
43,
44]. Therefore, it was unsurprising that we witnessed decreased levels in our rCDI population compared to healthy controls and an overall increase in butyric acid levels in the post-treatment era. Though month 12 butyrate levels were not significantly different by statistical definition, there was a dramatic increase in its level at month 12 compared to baseline than when compared to other time intervals (
p-value = 0.065). There was, however, a significant difference when comparing butyrate levels in pediatric patients with rCDI at baseline with healthy controls. The dichotomy between a non-significant increase in butyrate levels at post-FMT month-12 follow-up and a significant difference in butyrate levels at baseline could be due to a few factors, such as sampling size, or that there might be more follow-up time necessary to witness full butyrate level restoration.
Unlike polysaccharide-derived metabolites such as acetate, butyrate, and propionic acid, isovaleric acid is a product of amino acid fermentation and is a branched-chain fatty acid. Compared to other metabolites, prior research on isovaleric acid and its involvement in gut homeostasis is scarce. Thus, its role in the gut is vague, though there is evidence that it can lead to smooth muscle relaxation in the gut lumen through the cAMP/PKA pathway [
45]. Additionally, isovalerate has been found to potentially increase SUMOylation, leading to the inhibition of inflammation through the NF-kB pathway and preserving intestinal cell function and integrity [
46]. Though there is limited research, we would like to further explore the importance of isovaleric acid in microbiota metabolism, mucosal healing, and immune response in a future study.
The longitudinal nature of this study is significant, as SCFA profiles were only significant when examining the change from baseline to month 12. This is consistent with prior research that suggests that there are post-FMT fecal microbiota changes that occur, although this study was shorter in duration and explored an older population [
47].
We readily acknowledge the limitations of our study. First, due to the lack of patient adherence to longitudinal stool collections, the statistical power of our longitudinal data decreases over the study duration. Four of the nine children had no available stool at month six, and five of the nine had no stool at the final visit. As a result, our month-12 data only represent the SCFA profiles of four patients and thus may not be generalizable to a larger population. Additionally, the limited data may have influenced our recorded findings. Second, there is evidence that even short-term fluctuations in diet can impact the composition and metabolism of the microbiome, affecting overall SCFA composition [
48]. Though no dietary changes were reported in our population, we cannot be sure that day-to-day variance was without influence on our data, especially given our small cohort size. Third, there was no age stratification qualifier in our control group due to suggestive evidence that SCFA profiles are in stabilized concentrations after the first year of life [
30]. Additionally, we ran a supplemental analysis that excluded children above 11 years old and this yielded similar conclusions as the literature. It is evident that the healthy controls had much higher SCFA levels than the study population, suggesting age to be an insignificant confounding factor.