Clostridioides difficile Infection in Pediatric Inflammatory Bowel Disease

Abstract

Purpose of Review:

Children with inflammatory bowel disease (IBD) are at increased risk of C. difficile infection (CDI) and experience worse outcomes associated with an infection. In this article, we review recent research on the incidence, diagnosis, complications, and treatment options for CDI in children with IBD.

Recent Findings:

Children with IBD have an elevated incidence of CDI, but their CDI risk does not associate with established risk factors in adults with IBD. Existing testing methodologies are inadequate at differentiating CDI from C. difficile colonization in children with IBD. Fecal microbiota transplantation offers a durable cure for recurrent CDI.

Summary:

CDI remains a frequent occurrence in children with IBD. Careful clinical monitoring should be used to diagnose CDI and patients with co-occurring IBD and CDI require careful surveillance for worse outcomes. Future research should explore the optimal diagnosis and treatment modalities in this unique patient population.

Graphical Abstract

Introduction

Clostridioides difficile is the most common cause of antibiotic-associated diarrheal infection and accounts for a significant proportion of hospital-acquired infections [1, 2]. C. difficile is an anaerobic spore-forming bacterium that colonizes the gastrointestinal tract and opportunistically causes intestinal infections. The symptoms of C. difficile infections (CDI) include diarrhea, abdominal pain, and fever. Severe CDI can be life-threatening due to fulminant colitis and septic shock. CDI arises from the production of C. difficile toxins which disrupt intestinal epithelial integrity [1]. CDI is a major concern for children with inflammatory bowel disease (IBD) due to their perturbed intestinal microbiome and the use of immunosuppressive medications and antibiotics. Pediatric patients with IBD have an increased incidence of CDI and experience worse outcomes. High rates of C. difficile colonization in patients with IBD further complicate the diagnosis and management of these co-occurring diseases [3, 4]. CDI in pediatric IBD has unique features, differentiating it from CDI in general and in adults with IBD. Heterogeneity in diagnostic methodology and criteria complicates the comparison across studies. This review intends to summarize recent literature on CDI in the context of pediatric IBD, including its risk factors, clinical implications, diagnostic approaches, and treatment strategies.

Incidence and Associated Complications of C. difficile in IBD

In the past thirty years, the incidence of CDI in children has increased substantially [5, 6]. Currently, IBD is among the top comorbid conditions that increase the risk of CDI in children [7–9] and several studies have investigated the incidence of CDI in children with IBD compared to healthy controls. From interrogation of a large hospital discharge database which included more than 20,000 cases of pediatric CDI, children with IBD were noted to be 11 times more likely to experience CDI than children without IBD [10]. A similar study of hospital discharges in a single state in the United States displayed an incidence rate ratio of 12.7 for CDI in children with IBD relative to healthy controls [9]. Notably, adults with IBD had a lower incidence rate ratio of 4 relative to healthy adults. Unlike CDI in adults in which colonic IBD involvement is associated with an increased risk of CDI [11–13], there has been no consistent signal in pediatric IBD that colonic involvement increases the risk of CDI relative to IBD without colonic involvement [3, 7, 14–19].

CDI in pediatric IBD is a poor prognostic indicator and is often associated with complications such as disease exacerbations, hospitalizations, and the need for surgical intervention. In two large studies of hospital discharges, CDI increased the mean length of hospitalization from 6 days to 8 days in pediatric patients with IBD [15, 21]. An additional single-center study showed a similar increased length of hospitalization from 5.18 to 8.39 days in children with IBD with concurrent CDI [20]. CDI in children with IBD is also associated with an increased risk of intestinal surgery [17, 22]. In a case-control cohort study, 8% of children with IBD who experienced CDI required abdominal surgery in the following 6 months compared to 3% in children with IBD who did not have CDI (p<0.01) [20]. While CDI requiring colectomy is rare in children (0.3% of all pediatric CDIs), the vast majority (74.8%) of colectomies in a large pediatric database of children with CDI occurred in children with IBD [5]. These results parallel the surgical risk of CDI in adults with IBD [12].

Patients with IBD and CDI also often require escalation of IBD therapies. In a retrospective study of 111 pediatric IBD patients with CDI, 67% of patients required an escalation of IBD therapy following CDI, as defined as the addition of an immunomodulator or biologic agent to existing therapy [20]. Similarly, from a study of a single Canadian province, children with IBD who had CDI had an increased likelihood of requiring future use of systemic steroids or tumor necrosis factor inhibitors [7].

Of note, it remains unclear if CDI worsens the course of IBD or occurs more commonly in patients with a more severe IBD phenotype and therefore at higher risk of complications. At minimum, the diagnosis of CDI in a pediatric patient with IBD warrants careful surveillance as it may indicate a more complicated course. In sum, children with IBD experience higher rates of CDI and worse outcomes. Additional measures targeting prevention, diagnosis, and management are essential.

Risk Factors Associated with C. difficile Infections in Children with IBD

The association between IBD and higher rates of CDI is felt to be attributed to several factors. CDI has well-established risk factors in the general population including hospitalization and antibiotic exposures; these same risk factors are also more commonly experienced by patients with IBD. Furthermore, increasing evidence supports that the intestinal microbial dysbiosis associated with IBD creates a conducive environment for C. difficile colonization and infection.

While antibiotic exposure is a well-established risk factor for CDI in the general population, adults with IBD are less likely to report a recent antibiotic exposure prior to CDI [11, 23]. This difference is less pronounced in pediatric CDI, where children with IBD were no less likely to have an antibiotic exposure than non-IBD children with CDI [7, 16, 20]. Other medications, particularly acid-suppressing medications, have been associated with an increased risk of CDI in the general population, felt to be related to the loss of protective function of gastric acidity against C. difficile spores [24]. In studies of CDI in pediatric IBD, there has been no significant association between acid-suppressing medications and CDI risk [14, 16, 20]. Similarly, conflicting evidence exists regarding whether anti-inflammatory and immunomodulatory agents prescribed for IBD increase the risk of CDI. In the general population and adults with IBD, immunosuppression and steroids have been demonstrated to increase CDI risk [25]. However, in a recent meta-analysis of 14 studies on CDI in children with IBD, steroids, immunomodulators or biologics were not associated with an increased risk of CDI, again highlighting unique features of CDI in pediatric IBD [16]. Instead, only 5-aminosalicylic acid (5-ASA) was associated with increased CDI risk in children with IBD (odds ratio=1.95, 95% confidence interval 1.26–3.0) [16], although the reasons for this remain unclear.

The lack of a consistent link between immunosuppressive medications and CDI risk has led to the study of the intestinal microbiome and its associated metabolites as an additional area of focus. A diverse microbiota protects against colonization by potential pathogens, including C. difficile, termed colonization resistance [26]. Patients with IBD have a known dysbiotic gut microbiota, characterized by reduced microbial diversity and alterations in microbial composition [27–31]. This dysbiosis creates an environment that is more advantageous to the colonization and germination of C. difficile to its infectious form. When comparing the intestinal microbiome of patients with IBD with and without CDI, multiple differences have been identified. At a taxonomic level, overall microbiome diversity measures are not significantly different between samples from those with IBD+CDI versus IBD alone, but are significantly different from healthy control samples [32–34]. At a more granular level Ruminococcus gnavus and Enterococcus spp. have been identified as increased in abundance in IBD+CDI stool samples in multiple studies [32, 33]. Conversely, the anti-inflammatory taxonomic groups Faecalibacterium, Lachnospiraceae, and Roseburia are decreased in abundance in IBD+CDI samples [22]. The precise contributions of alterations in these bacterial taxonomic groups to CDI susceptibility is yet to be determined.

In a multi-omic analysis of stool from pediatric IBD patients, metabolite differences significantly discriminated CDI and non-CDI samples [34]. Several metabolites, including taurine and isocaproyltaurine, were increased in abundance in CDI versus non-CDI samples; these metabolites have been linked to intestinal inflammation and C. difficile metabolism [34]. Children with IBD have also been noted to have altered fecal levels of bile acids [34, 35], with an increased abundance of primary bile acids and decreased abundance of secondary bile acids [36, 37]. Bile acids play a key role in C. difficile pathogenesis as primary bile acids stimulate the germination of C. difficile from spores to vegetative cells. Conversely, secondary bile acids exhibit inhibitory properties towards C. difficile [38], suggesting that altered bile acid profiles in patients with IBD may be one reason for their heightened CDI risk. Indeed, increased fecal primary bile acids and decreased secondary bile acids have been noted in children with IBD at the time of CDI [34, 39]. These metabolite differences may contribute to new diagnostic modalities for CDI. In sum, the altered composition and metabolite profile of the microbiome increases the risk of C. difficile colonization and germination.

Diagnostic Dilemma: C. difficile or IBD Flare

Due to the high rate of CDI in patients with IBD, current algorithms suggest testing for C. difficile in the setting of an IBD flare, noting that symptoms associated with CDI and a flare of IBD are indistinguishable [40, 41]. However, the ability to appropriately differentiate CDI from a flare in patients with IBD is highly confounded by high rates of C. difficile colonization in children with IBD. Colonization is loosely defined as the presence of toxigenic C. difficile in the intestinal microbiome without symptoms associated with C. difficile [42]. C. difficile colonization rates in children with IBD have been reported to be between 10–25% [4, 9, 43].

Currently, most clinical laboratories test for C. difficile using nucleic acid amplification tests (NAAT), toxin enzyme immunoassays (EIA), and/or glutamate dehydrogenase (GDH) [44], all with variable sensitivity and specificity. NAAT are polymerase chain reaction (PCR)-based tests that detect the presence of DNA segments encoding the C. difficile toxins and are often preferred related to high sensitivity [44, 45]. However, their use has led to higher rates of C. difficile detection and concerns regarding overdiagnosis [46, 47]. Toxin EIAs detect C. difficile toxin A and/or toxin B but have overall low sensitivity [48]. GDH is a metabolic enzyme present in the majority of C. difficile strains independent of whether the C. difficile strain encodes pathogenic toxins; that is, GDH positivity alone cannot be used to diagnose the presence of toxigenic C. difficile [44].

Current testing recommendations for CDI are to either: A) use a two-step testing approach with a combination of the aforementioned tests or B) use a highly sensitive test (e.g. NAAT) while meeting careful clinical criteria. The overall goal of these recommendations was to increase the sensitivity and specificity of the testing approach and minimize the risk of false-positive tests as a result of C. difficile colonization [41, 45, 47]. With a two-step algorithm, the first step of testing is the use of a highly sensitive test, either GDH EIA or NAAT (screening test), which should then be followed by a toxin EIA (toxin test). In the case of discordant results between GDH and toxin EIA, NAAT may be used to arbitrate the results [45].

Notably, existing CDI testing guidelines were developed for use in the general non-IBD population, and none were originally evaluated in children [40, 45]. More recent data have suggested some limitations in the use of testing algorithms in these patient populations. In a cohort of children positive for C. difficile by NAAT, including patients with IBD, there was no difference in toxin test positivity when comparing symptomatic versus asymptomatic colonized patients [43]. These results suggest that a toxin test may not adequately differentiate colonization from disease in children, including those with IBD. In addition, studies evaluating outcomes of adult patients with IBD comparing those with screen-positive/toxin-negative results to screen-positive/toxin-positive results have demonstrated variable results. Some studies have demonstrated no differences in clinically relevant outcomes [49–51], while others have demonstrated an increased likelihood of escalation of IBD therapies and increased response to antibiotics in those who are toxin positive [52].

Ongoing research aims to identify more definitive biomarkers of CDI, particularly in patients with IBD. Promising candidates may include antibody responses to C. difficile and metabolite signatures characteristic of CDI [34]. In the interim, the American College of Gastroenterology recommends empiric treatment for CDI even in the absence of a toxin EIA if clinical suspicion of CDI is high [40].

Antibiotic Treatment for C. difficile in IBD

Following diagnosis, rapid initiation of antibiotic treatment is essential to prevent further CDI-related complications. Pediatric IBD-specific treatment guidelines do not exist for CDI; thus, current recommendations as discussed below are based on adult data [53].

Oral vancomycin is currently recommended for an initial episode of non-severe CDI in patients with IBD [54]. Alternatively, fidaxomicin may also be considered for the treatment of non-severe CDI. For severe, fulminant CDI, rectal vancomycin may be considered as an alternative to oral vancomycin. Likewise, intravenous metronidazole can be considered for fulminant CDI. For recurrent CDI episodes, current recommendations support pulse-tapered vancomycin or fidaxomicin. Pulse-tapered vancomycin consists of 125mg oral vancomycin daily for 14 days, then twice daily for 7 days, once daily for 7 days, and finally once every 2–3 days for 2–8 weeks [53]. Fidaxomicin for recurrent CDI consists of a twice daily administration for a 10-day course. Following completion of antibiotic course for CDI, testing for C. difficile eradication (i.e. test of cure) is not recommended [45].

Importantly, immunosuppressive medications should not be discontinued during the treatment of CDI. Due to the significant overlap between CDI and symptoms of an IBD flare, and the diagnostic difficulties previously discussed, guidelines recommend a careful and ongoing evaluation for signs of active colitis. If ongoing diarrhea without improvement on day 3 or 4 of C. difficile-targeted antibiotics, escalation of immunosuppression should be considered [41].

Fecal Microbiota Transplantation for Recurrent C. difficile in IBD

Just as children with concurrent IBD and CDI are at risk of IBD-related complications, they are also at risk of CDI-associated adverse events, including recurrent infection. Recurrent CDI (rCDI) is defined as a second CDI episode within 8 weeks of a prior CDI. From several studies in children with IBD, a range of recurrence rates have been reported from 25% to 34% [7, 20, 55].

Current guidelines recommend the consideration of fecal microbiota transplantation (FMT) for treatment of rCDI in patients with IBD [41, 45, 53]. FMT is the installation of processed stool from a healthy donor in an attempt to replace a dysbiotic intestinal microbiome and has been widely used to treat rCDI in the general population [56].

A multicenter retrospective cohort study showed that FMT was as efficacious in treating recurrent CDI in children with IBD as non-IBD children (76% vs. 81%, p=0.17) [57]. Among children with IBD, FMT was more likely to succeed using fresh donor stool, in patients without active diarrhea prior to FMT, and with a shorter duration between CDI diagnosis and the FMT procedure. Hospitalization related to IBD occurred in 13% of children in the 3-months following FMT, but it could not be determined if these hospitalizations were attributable to FMT, the patients’ IBD, or a combination of factors [57]. Some adult studies have suggested improvement in IBD-disease activity following FMT for the treatment of CDI [58]. In a study of 50 adults with IBD, 73% of Crohn’s disease patients and 62% of ulcerative colitis patients had clinical improvement in their IBD following FMT for rCDI [58]. Conversely, only 38% of pediatric patients had improvement in their IBD status in the 3 months after FMT in a large multicenter study, although results were limited by the retrospective study design [57]. As FMT protocols and microbial therapeutics become increasingly standardized, they warrant continued study in children with CDI and IBD.

CONCLUSIONS

Children with IBD are at increased risk of primary and rCDI, likely related to underlying gastrointestinal inflammation, altered intestinal microbiota, and frequent exposure to healthcare settings. Compared to adults with IBD, children with IBD who experience CDI are less likely to have baseline colonic disease. Additionally, steroids, immunomodulators, and biologics are not established risk factors for CDI in children with IBD. Children with IBD with CDI experience longer hospitalizations, increased risk of intestinal surgery, and increased likelihood of requiring escalation of IBD therapies although it remains unclear if C. difficile worsens IBD outcomes or, alternatively, occurs more frequently in patients with severe or refractory IBD.

Symptom overlap between IBD flares and CDI creates a significant diagnostic dilemma. Moreover, existing CDI testing methodologies are limited in their to differentiate CDI from C. difficile colonization, both of which are common in children with IBD. Current recommendations support the testing for C. difficile when symptoms of CDI and/or an IBD flare are present. Due to diagnostic challenges, when antibiotics are used for the treatment of CDI in patients with IBD, a careful assessment for resolution of symptoms is critical, and escalation of immunosuppression in the setting of an inadequate antibiotic response should be considered. While antibiotics remain the mainstay of treatment for an initial episode of CDI in children with IBD, high rates of rCDI require the consideration of alternative treatment modalities. FMT has emerged as an effective treatment for rCDI in children with IBD.

Research on CDI in pediatric IBD has been limited by retrospective, single-center studies, and variability in those studies’ CDI testing methodology. Current diagnosis and treatment guidelines are extrapolated from adult studies. The unique features of CDI in pediatric IBD compared to adult IBD necessitate targeted prospective study of CDI in pediatric IBD.