INTRODUCTION — Clostridioides difficile is the causative organism of antibiotic-associated colitis [1]. Colonization of the intestinal tract occurs via the fecal-oral route and is facilitated by disruption of normal intestinal flora (often due to antimicrobial therapy). (See "Clostridioides difficile infection in adults: Clinical manifestations and diagnosis".)
Issues related to epidemiology and microbiology of C. difficile infection (CDI) will be reviewed here. Clinical manifestations, diagnosis, prevention, and treatment of CDI are discussed separately. (See "Clostridioides difficile infection in adults: Clinical manifestations and diagnosis" and "Clostridioides difficile infection: Prevention and control" and "Clostridioides difficile infection in adults: Treatment and prevention".) (Related Pathway(s): Clostridioides difficile infection: Treatment of adults with an initial or recurrent infection.)
EPIDEMIOLOGY
Overview — Antibiotic-associated diarrhea and colitis were well established soon after widespread use of antibiotics [2]. In 1978, C. difficile was identified as the causative pathogen in the majority of cases of antibiotic-associated colitis; many early cases were attributed to clindamycin [2,3]. Increasing use of penicillins and cephalosporins led to the implication of these antibiotic classes as precipitants as well.
From 2003 to 2006, CDI was observed to be more frequent, severe, refractory to standard therapy, and more likely to relapse than previously described [2]. These observations have been reported throughout North America and Europe and have been attributed to the emergence of a new strain designated BI, NAP1, or ribotype 027 (these designations are based on different methods for strain typing and all refer to the same strain, known as strain NAP1/BI/027). This strain appears to be more virulent than other strains [4], which may be attributable to increased toxin production compared with previous strains. Fluoroquinolone use has strongly correlated with the emergence of this strain [5], and development of fluoroquinolone resistance by outbreak strains appears to have been associated with the increasing frequency of CDI outbreaks [6]. (See 'NAP1/BI/027 strain' below.)
Since 2005, CDI due to ribotype 078 has emerged in the Netherlands; the severity is similar to CDI caused by ribotype 027. Ribotype 078 appears to affect a younger patient population, is more frequently community associated, and is genetically similar to porcine isolates [7]. Among 1366 Dutch patients hospitalized between 2006 and 2009, CDI was associated with a 2.5-fold increase in 30-day mortality (95% CI 1.4-4.3) [8].
A cross-sectional study demonstrated that ribotype may not be a significant predictor of severe CDI. Among 310 CDI cases, including 34 severe cases, the association between ribotypes 027 and 078 and severity was not statistically significant after adjustment for other covariables [9].
In 2011, an estimated 453,000 initial cases of CDI occurred in the United States; there were also an estimated 83,000 first CDI recurrences and 29,300 patients died [10]. Between 2011 and 2017, there was a decline in the incidence of CDI (154.9 to 143.6 per 100,000 persons), primarily driven by a decrease in health care associated infections [11].
Nosocomial infection — Dramatic increases in the incidence and severity of health care-associated CDI occurred after 2000, particularly in patients over age 65. However, between 2011 and 2017 the incidence of health care-associated CDI decreased from 99.6 to 73.3 per 100,000 population [11].
C. difficile carriage occurs in 8 to 10 percent of adults residing in hospitals or long-term care facilities (the carrier rate among healthy adults is about 3 percent) [12-15]. Asymptomatic C. difficile carriers are capable of shedding C. difficile spores and serve as a reservoir for environmental contamination to other hospitalized patients [16,17]. In one study including more than 250 patients without diarrhea admitted to the hospital, 79 percent were not colonized with C. difficile at the time of admission, 15 percent were colonized with a toxinogenic strain of C. difficile, and 6 percent were colonized with a nontoxinogenic strain [18]. Surprisingly, there were no differences between patients colonized with a toxinogenic strain of C. difficile and uncolonized patients with regards to recent health care contact or antimicrobial exposures. Furthermore, stool toxin concentrations in asymptomatic carriers are similar to those in patients with diarrhea and colitis [15].
New exposure and colonization by C. difficile are more likely to lead to CDI, while patients previously colonized with C. difficile are more likely to remain asymptomatic during their hospitalization [19,20]. The magnitude of this difference was illustrated in a prospective review of 810 hospitalizations, including 618 patients with new C. difficile exposure and 192 patients with previously known C. difficile colonization [20]. Newly exposed patients developed CDI much more frequently than colonized patients (4.5 versus 1.1 percent).
These observations have been noted even when carriers and symptomatic patients are colonized with identical strains. In addition, outbreaks are frequently caused by a single strain of C. difficile even though many bacterial strains may exist in the given hospital environment [21]. As an example, two outbreaks in the same hospital were attributed to a single strain, although 31 distinct strains were isolated from 98 patients in the institution [22].
Some potential explanations for these epidemiologic observations are discussed below. Careful adherence to infection control policies is critical to successful control of C. difficile. (See 'Pathophysiology' below and "Infection prevention: Precautions for preventing transmission of infection".)
Community-associated infection — The incidence of community-associated CDI (defined as disease in patients not hospitalized during the 12 weeks prior to diagnosis) has remained unchanged between 2011 and 2017 [11].
Unusually severe CDI has been reported in populations previously thought to be at low risk, including peripartum women and healthy individuals living in the community without a history of antibiotic use, recent hospitalization, or other risk factors for C. difficile-associated disease [23-25]. As a result, consideration of C. difficile-associated infection is warranted even in the absence of a history of antibiotic exposure or recent hospitalization. Exposure to retail food products and domestic animals has been postulated as potential sources of CDI [26,27].
Recurrent infection — Recurrent CDI is defined by resolution of CDI symptoms while on appropriate therapy, followed by reappearance of symptoms within two to eight weeks after treatment has been stopped [1]. Up to 25 percent of patients experience recurrent CDI within 30 days of treatment [28]. Less commonly, recurrent CDI can occur as late as two months after discontinuation of treatment. In the United States in 2011, approximately 83,000 patients had at least one recurrence [10].
Once patients have experienced one recurrence, they are at significantly increased risk for further recurrences (multiply recurrent CDI [mrCDI]). In one retrospective cohort study including more than 45,000 patients with CDI in the United States between 2001 and 2012, the annual incidence of mrCDI increased by 189 percent (from 0.0107 to 0.0309 cases per 1000 person-years) [29]. Those who developed mrCDI were older (median age 56 versus 49 years), more likely to be female, and more likely to have used antibiotics, proton-pump inhibitors, or corticosteroids within 90 days of CDI diagnosis. Other factors associated with increased risk for mrCDI included chronic kidney disease and residence in a nursing home. The observed increase in incidence of mrCDI was independent of known risk factors for CDI, raising the possibility that the biology is different from that of CDI in general.
TRANSMISSION — Patients with C. difficile carriage are a reservoir for environmental contamination in the presence or absence of clinical infection. C difficile is highly transmissible via the fecal-oral route by ingestion of spores. The organism can be cultured readily from the hospital environment, including items and inert surfaces in patient rooms as well as the hands, clothing, and stethoscopes of health care workers [30,31].
The organism is also transmitted readily between hospital roommates [12,22]. One retrospective study noted that receipt of antibiotics by prior bed occupants was associated with increased risk for CDI in subsequent patients [32]. Measures for prevention of C. difficile transmission in hospital settings are discussed separately. (See "Clostridioides difficile infection: Prevention and control".)
Patients harboring toxinogenic C. difficile (detected via nucleic acid amplification) in their stool are capable of serving as a source for transmission of infection to others, regardless of whether toxin is detected (via enzyme immunoassay) [33].
Study of C. difficile transmission using whole genome sequencing has suggested that transmission is more likely to occur from patients with diarrhea related to active CDI than from patients with asymptomatic colonization [34].
RISK FACTORS — Antibiotic use is the most widely recognized and modifiable risk factor for CDI [35-38]. Other established risk factors include advanced age, hospitalization, and severe comorbid illness [37]. Additional risk factors include enteral feeding, gastrointestinal surgery, obesity, cancer chemotherapy, hematopoietic stem cell transplantation, inflammatory bowel disease, cirrhosis, and possibly gastric acid suppression [2,37,39-46]. The association between cancer chemotherapy and CDI may be related to the antimicrobial effect of chemotherapeutic agents and/or their immunosuppressive effects [47].
Risk factors for recurrent CDI include age >65 years, severe underlying medical disorders, need for ongoing therapy with concomitant antibiotics during treatment for CDI, serum creatinine ≥1.2 mg/dL, and lack of an antibody-mediated immune response to C. difficile toxins especially toxin B [48-54].
Risk factors for complications associated with CDI (perforation, toxic megacolon, colectomy, admission to intensive care unit, death) include older age, abnormal blood tests (white blood cell count <4000 cells/microL or ≥20,000 cells/microL, albumin <25 g/L, blood urea nitrogen >7 mmol/L, and C-reactive protein ≥150 mg/L), and abnormal vital signs (heart rate >90/minute, respiratory rate >20/minute) [55].
CDI can also occur in the absence of any risk factors [56,57].
Antibiotic use — Two major roles for antibiotics in the pathogenesis of C. difficile have been described. First, antibiotics disrupt the barrier function of the normal colonic microbiota, providing a niche for C. difficile to multiply and elaborate toxins [58]. In a small series of patients with CDI, for example, absence of Bacteroides species (a predominant species in normal colonic microbiota) was noted during CDI but returned following resolution of symptoms [59]. Second, development of C. difficile resistance to clindamycin or fluoroquinolones appears to play an important role in disease due to strains with increased virulence [5,60].
Antibiotic stewardship plays an important role in controlling CDI rates [1,61]. (See "Antimicrobial stewardship in hospital settings" and "Antimicrobial stewardship in outpatient settings".)
The antibiotics most frequently implicated in predisposition to CDI include fluoroquinolones, clindamycin, and broad-spectrum penicillins and cephalosporins (table 1) [5,60,62-65]. However, any antibiotic can predispose to colonization by C. difficile, including metronidazole and vancomycin, which are the major antibiotics used to treat CDI [66]. The use of broad-spectrum antimicrobials, use of multiple antibiotic agents, and increased duration of antibiotic therapy all contribute to the incidence of CDI [67-69].
The earliest cases of C. difficile were attributed to clindamycin, and it remains an important culprit [2,70]. A strain of C. difficile highly resistant to clindamycin was implicated in large outbreaks in the early 1990s [60]. The use of clindamycin is a specific risk factor for this strain, and colonization with the clindamycin-resistant strain increases risk for developing CDI.
Increasingly widespread use of fluoroquinolones has been correlated with C. difficile–associated diarrhea. In addition, fluoroquinolone resistance of the hypervirulent strain NAP1/BI/027 may be an important factor in nosocomial outbreaks caused by these virulent strains [5,71]. (See 'NAP1/BI/027 strain' below.)
The degree and duration of CDI risk after cessation of antibiotic treatment are not well understood. One case-control study including 337 patients with C. difficile suggested that risk was increased during antibiotic therapy and in the period of three months after cessation of therapy; the risk was highest during the first month after antibiotic use [72].
Perioperative antibiotic prophylaxis also confers risk for CDI, especially if a hospital is experiencing an outbreak of CDI.
A "herd effect" of antibiotic use has been postulated (ie, that patients not on antibiotics in regions where antibiotic use is high are at greater risk for CDI than patients not on antibiotics in regions where antibiotic use is low) [73].
Advanced age — Age seems to promote the frequency and severity of CDI. In a 2002 Quebec outbreak, the frequency of CDI among persons ≥65 years was 10-fold higher than that observed in younger adults [74]. This age-related association was also noted as early as the 1980s, prior to the epidemic emergence of hypervirulent C. difficile strains [75,76].
The reasons for this association are uncertain and may be multifactorial. Host factors, such as diminished immune response to CDI, may play a role. Alternatively, older individuals may have other comorbidities that place them at a cumulative increased risk for CDI, such as greater likelihood to require hospitalization, antibiotic use, or inherent dysbiosis.
Gastric acid suppression — Gastric acid suppression (with proton pump inhibitors [PPIs] or histamine 2 receptor antagonists) has been associated with an increased risk of CDI [23,77-88].
The risk of CDI ranges from 1.4 to 2.75 times higher among patients with PPI exposure compared with those without PPI exposure [89]. However, a causal association has not been established and the relationship between the risk of CDI and PPI dose and duration of use is uncertain [90,91].
Discontinuation of unnecessary PPIs is reasonable although there is insufficient evidence for routine discontinuation of PPIs as a measure for preventing CDI [92].
MICROBIOLOGY — C. difficile is an anaerobic gram-positive, spore-forming, toxin-producing bacillus first described in 1935 (picture 1 and picture 2) [93]. It was named "difficult clostridium" because of difficulty related to its isolation and growth on conventional media. C. difficile can exist in spore and vegetative forms. Outside the colon, it survives in spore form; spores are resistant to heat, acid, and antibiotics. Once spores are in the intestine, they convert to their fully functional vegetative, toxin-producing forms and become susceptible to killing by antimicrobial agents.
C. difficile was initially observed as a component of the intestinal microbiota in healthy newborns [93]. The pathogenic role of C. difficile was first appreciated in the 1970s when C. difficile toxin was observed in the stools of patients with antibiotic-associated pseudomembranous colitis [3,94]. This organism is now recognized as the cause of pseudomembranous colitis and other variant forms of diarrhea and colitis usually in patients exposed to antibiotics. (See "Clostridioides difficile infection in adults: Clinical manifestations and diagnosis".)
Toxins — C. difficile releases two potent exotoxins that mediate colitis and diarrhea: toxin A and toxin B. The organism is rarely invasive, and nontoxinogenic strains do not cause CDI. C. difficile toxins contain a series of contiguous repeated units at the carboxyl terminus. For toxin A, this region is important for toxin A carbohydrate receptor binding to facilitate intracellular transport as well as for antibody binding in enzyme-linked immunosorbent assays [95].
Once intracellular, toxins A and B inactivate regulatory pathways mediated by Rho family proteins that are involved in cytoskeleton structure and signal transduction via guanosine triphosphate. This disruption leads to cell retraction and rounding in tissue culture assays and as shallow ulcerations on the intestinal mucosal surface [96,97]. Both toxins also disrupt intercellular tight junctions [98,99].
In vivo, stool toxin levels correlate with disease severity [100]. Toxin A causes inflammation leading to intestinal fluid secretion and mucosal injury [101,102]. Mediators in these pathways include arachidonic acid metabolites, substance P, tumor necrosis factor, and interleukin (IL)-8, IL-6, and IL-1 [103-106]. Toxin A directly activates neutrophils, and both toxin A and B can promote neutrophil chemotaxis to localize in pseudomembranes and the underlying intestinal mucosal layer [107,108].
Toxin B is a major factor for the virulence of C. difficile and is more than 10 times more potent than toxin A on a molar basis for mediating colonic mucosal damage [97,109,110]. Thus, strains lacking toxin A can be as virulent as strains with both toxins [108,109,111-113]. Conversely, CDI-associated clinical isolates that produce toxin A but minimal or no toxin B have been identified; hence, toxin A appears to also contribute to CDI pathogenesis [114].
A minority of C. difficile clinical isolates (10 to 30 percent) are nontoxinogenic [70,115] and do not produce toxins in vivo or in vitro; these strains can colonize the gastrointestinal tract and grow normally in culture media but are not pathogenic [116].
NAP1/BI/027 strain — A "hypervirulent" strain, NAP1/BI/027, has been implicated as the responsible pathogen in selected C. difficile outbreaks since the early 2000s. The name of this strain reflects its characteristics demonstrated by different methods for strain typing: pulsed-field gel electrophoresis (NAP1), restriction endonuclease analysis (BI), and polymerase chain reaction (027). All of these designations refer to the same strain, known as strain NAP1/BI/027. Interestingly, NAP1 may not be hypervirulent in a nonepidemic setting [117].
Characteristics of this strain that contribute to the clinical and epidemiologic observations include (see 'Epidemiology' above):
●NAP1/BI/027 produces binary toxin, an additional toxin that is always absent in other C. difficile strains [70,118,119]. Binary toxin is related to the iota-toxin found in Clostridium perfringens; its role in C. difficile pathogenesis is not fully understood [120,121].
●Some studies suggest NAP1/BI/027 produces substantially larger quantities of toxins A and B than other C. difficile strains both in vitro and in vivo [122,123].
●NAP1/BI/027 is toxinotype III; most other C. difficile strains are toxinotype 0 [122,124]. Toxinotyping is based on analysis of the pathogenicity locus (PaLoc) of the C. difficile genome; this is the region that includes the genes for toxin A (tcdA), toxin B (tcdB), and neighboring regulatory genes.
●NAP1/BI/027 has a partial deletion of tcdC, a gene in the pathogenicity locus that is responsible for downregulation of toxin production [124,125]. This may contribute to the enhanced capability for production of toxins A and B.
●NAP1/BI/027 is resistant to fluoroquinolones in vitro; this was an infrequent observation in C. difficile strains prior to 2001 [5,71,74,124].
Lower clinical cure rates and increased recurrence rates have been observed among patients infected with the epidemic NAP1/BI/027 strain compared with patients infected with nonepidemic strains [126-128]. This strain has also been associated with severe disease, severe outcome (intensive care unit admission, colectomy, or death within 30 days), and death within 14 days [129,130]. However, the hypervirulent strain's genomic makeup and altered toxin-producing capabilities do not fully explain the increasing frequency and severity of disease, since historic isolates of NAP1/BI/027 possessing binary toxin genes and tcdC deletions have also been described [124]. However, a new observation is the increased resistance to fluoroquinolones; historic isolates of NAP1/BI/027 were not as resistant to fluoroquinolones as the contemporary ones. A selective advantage afforded by increased resistance to a particular antimicrobial has also been observed previously; the highly clindamycin-resistant strain was associated with epidemics in the late 1980s and early 1990s [60].
PATHOPHYSIOLOGY
Overview — C. difficile is capable of elaborating exotoxins that act upon intestinal epithelial cells and inflammatory cells, leading to tissue injury and diarrhea (figure 1).
Pathogen and host factors probably play important roles in the evolving epidemiology of CDI. C. difficile diarrhea is mediated by genes for toxin A (tcdA) and toxin B (tcdB), which inactivate members of the Rho family of guanosine triphosphatases (Rho GTPases), leading to colonocyte death, loss of intestinal barrier function, and neutrophilic colitis [131]. Toxin production and other microbiologic factors are outlined in detail above (see 'Microbiology' above). Host factor responses include toxin A and B antibody production, interleukin (IL)-8 levels, and intestinal toxin receptors [51,132-135].
Serum antitoxin antibodies are the best described host factor protecting against C. difficile pathogenesis. Asymptomatic carriers demonstrate higher serum levels of immunoglobulin (Ig)G antibodies against toxin A than patients who develop clinical CDI [51,132-134]. In a prospective study of 271 hospitalized patients, those with low or undetectable levels of antitoxin A antibodies were significantly more likely to develop diarrhea than those with detectable antibodies against toxin A (odds ratio 48, 95% CI 3.4-678) [31,132]. Mounting a serum antibody response to toxin A during an initial episode of CDI was associated with relative protection against recurrent CDI [51,136].
Colonization with nontoxinogenic strains also affords protection. This observation suggests that the initial colonizing strain may occupy a microbial niche in the large intestine that is protective against superinfection with a new toxin-producing C. difficile strain [19,137,138].
A common polymorphism in the IL-8 gene promoter is associated with increased risk for CDI and recurrence [135,139,140]. This was suggested in a report of 125 hospitalized patients (42 with C. difficile disease, 42 with C. difficile–negative diarrhea, and 41 without diarrhea) [135]. The frequency of an IL-8 allele with a single nucleotide polymorphism (genotype AA, responsible for increased IL-8 production) was significantly associated with development of C. difficile diarrhea in comparison with the two control groups (39 versus 16 and 17 percent, respectively). CDI patients also had higher median fecal IL-8 levels compared with control patients.
Recurrent disease — The mechanism of CDI recurrence following initial infection is not fully understood. It may be due to the presence of persistent spores from the initial infection (relapse) or reinfection by exogenous C. difficile.
Impairment of the host immune response to C. difficile toxins may also be an important mechanism for recurrence. Asymptomatic carriers of C. difficile tend to have high serum antibody levels against C. difficile toxins [132], while patients with recurrent CDI tend to have lower antitoxin antibody levels than patients with a single brief episode of diarrhea [51,133,136,141].
Antibiotic resistance does not appear to be a factor in recurrence. However, treatment with metronidazole or vancomycin for an initial episode of CDI may alter the colonic microenvironment (with regard to microbiota or other factors), potentially increasing susceptibility to reinfection and recurrence. (See "Clostridioides difficile infection in adults: Treatment and prevention".)
C. difficile in neonates — The absence of intestinal receptors for C. difficile toxins may be protective against CDI. Although this observation has been made in an animal neonate model, it is speculated that it may also play a role in protection against C. difficile in human newborns [142]. These individuals have a relatively high asymptomatic C. difficile carriage rate (up to 100 percent) in the setting of elevated C. difficile toxin stool titers [142-144]. Serum and stool Ig levels subsequently develop by two years of age in about 60 percent of healthy children and are detectable through adulthood [145,146].
INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5th to 6th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10th to 12th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.
Here are the patient education articles that are relevant to this topic. We encourage you to print or email these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient info" and the keyword(s) of interest.)
●Beyond the Basics topic (see "Patient education: Antibiotic-associated diarrhea caused by Clostridioides difficile (Beyond the Basics)")
SUMMARY
●Epidemiology – Clostridioides difficile is a significant nosocomial pathogen and can cause both health-care associated and community-associated infection. Dramatic increases in the severity and refractoriness of C. difficile infection (CDI) in multiple outbreaks around the world have been attributed, in part, to the NAP1/BI/027 strain. (See 'Epidemiology' above and 'NAP1/BI/027 strain' above.)
●Transmission – C difficile is highly transmissible via the fecal-oral route by ingestion of spores. Transmission is more likely to occur from patients with diarrhea than from patients with asymptomatic colonization. (See 'Transmission' above.)
●Risk factors – Although CDI can occur in the absence of any risk factors, the most common risk factors for CDI include antibiotic use, advanced age, hospitalization, and severe comorbid illness.
•Antibiotics associated with CDI – The antibiotics most frequently implicated in predisposition to CDI include fluoroquinolones, clindamycin, penicillins, and cephalosporins. (See 'Antibiotic use' above.)
•Recurrent CDI – Risk factors for recurrent CDI include age >65 years, severe underlying medical disorders, need for ongoing therapy with concomitant antibiotics during treatment for CDI, serum creatinine ≥1.2 mg/dL, and lack of an antibody-mediated immune response to C. difficile toxins, especially toxin B. (See 'Risk factors' above.)
•Complications – Risk factors for complications associated with CDI (perforation, toxic megacolon, colectomy, admission to intensive care unit, death) include older age, abnormal blood tests (eg, leukopenia or leukocytosis, hypoalbuminemia, acute kidney injury), and abnormal vital signs. (See 'Risk factors' above.)
●Pathophysiology
•Colonization of C. difficile occurs via the fecal-oral route usually in the context of disrupted intestinal microbiota due to antimicrobial therapy. (See 'Pathophysiology' above.)
•C. difficile produces two toxins that bind to receptors on intestinal epithelial cells, leading to inflammation and diarrhea. Individuals with low or undetectable levels of antibody against C. difficile toxins are more likely to develop diarrhea than those with detectable anti-toxin antibody levels. (See 'Toxins' above and 'Pathophysiology' above.)
79 : Iatrogenic gastric acid suppression and the risk of nosocomial Clostridium difficile infection.
87 : Proton-Pump Inhibitor Use and the Risk of Community-Associated Clostridium difficile Infection.