INTRODUCTION — Clostridioides difficile is an important cause of antibiotic-associated diarrhea and one of the most common health care-associated pathogens [1]. Its clinical manifestations range from asymptomatic colonization or mild diarrhea to fulminant disease characterized by ileus, toxic megacolon, hypotension, or shock. C. difficile infection is less common in children than adults, but the incidence of C. difficile infection in children is increasing [2-5].
The clinical features and diagnosis of C. difficile infection in children will be discussed here. The microbiology, pathogenesis, epidemiology, treatment, and prevention of C. difficile infection in children and C. difficile infection in adults are discussed separately:
●(See "Clostridioides difficile infection in children: Microbiology, pathogenesis, and epidemiology".)
●(See "Clostridioides difficile infection in children: Treatment and outcome".)
●(See "Clostridioides difficile infection: Prevention and control".)
●(See "Clostridioides difficile infection in adults: Epidemiology, microbiology, and pathophysiology".)
●(See "Clostridioides difficile infection in adults: Clinical manifestations and diagnosis".)
●(See "Clostridioides difficile infection in adults: Treatment and prevention".)
CLINICAL SPECTRUM — The clinical spectrum of C. difficile infection ranges from asymptomatic colonization to fulminant colitis [6,7]. The severity of infection generally increases with age. Mortality and fulminant disease are highest in persons ≥65 years and uncommon in children [1,2,8,9]. However, the full spectrum of disease is possible in all age groups [6,10,11].
Asymptomatic colonization — Asymptomatic carriers may be a reservoir of C. difficile [12,13]. (See "Clostridioides difficile infection in children: Microbiology, pathogenesis, and epidemiology", section on 'Colonization'.)
Neonates and children <2 years of age — Asymptomatic colonization with toxigenic C. difficile is common in neonates and infants [6,14]. A meta-analysis of observational studies estimated the proportion of colonization with C. difficile increased from 15 percent (95% CI 7-25 percent) in neonates <7 days of age to 41 percent (95% CI 32-50 percent) in infants 6 to 12 months of age, the peak of asymptomatic colonization [15]. Colonization with toxigenic C. difficile increased from ≤6 percent in infants <3 months of age to 14 percent (95% CI 8-21 percent) in infants 6 to 12 months of age. (See "Clostridioides difficile infection in children: Microbiology, pathogenesis, and epidemiology", section on 'Epidemiology'.)
Multiple studies have failed to demonstrate a clear association between the detection of C. difficile toxins and diarrheal illness in children younger than two years [16-23]. The causes of this apparent resistance to illness are unknown. However, there are data to suggest that C. difficile can be pathogenic in infants and young children in select circumstances [11]. (See 'Symptomatic disease' below.)
The age at which children become universally susceptible to C. difficile infection is unknown [11]. Coinfection with additional intestinal pathogens is common in this age group but does not appear to be associated with increased illness severity [19,20,24-26]. (See "Clostridioides difficile infection in children: Microbiology, pathogenesis, and epidemiology", section on 'Pathogenesis'.)
Hospitalized children and children with comorbidities — Asymptomatic carriage has been described in hospitalized children and children with comorbidities [14,27]. In a single study of hospitalized children, similar proportions of symptomatic and asymptomatic patients had a positive C. difficile polymerase chain reaction assay [27]. Several observational studies suggest a colonization prevalence of 15 to 30 percent in children with cancer, 33 to 50 percent in children with cystic fibrosis, and 11 to 17 percent in children with inflammatory bowel disease [27-31].
Symptomatic disease
Diarrhea — Symptoms of C. difficile infection typically begin either during or several weeks after antibiotic therapy; occasionally the onset is delayed for as long as 10 weeks. Acute-onset, mild to moderate diarrhea is the most common feature of symptomatic C. difficile infection [32-34]. The diarrhea is watery and profuse; some children have >10 stools per day at the peak of the illness. Stools with grossly visible blood are uncommon, occurring in <15 percent of episodes [32-34]. Fever and lower abdominal cramping or pain are common and vomiting can occur [35]. The physical examination may demonstrate lower abdominal tenderness [36].
In a case series of 109 children (age 1 to 18 years) with C. difficile infection at a tertiary care children's hospital, the clinical features included fever (37 percent), abdominal pain (46 percent), and diarrhea with grossly visible blood (14 percent); the majority of children passed ≥6 stools per day during the peak of the illness [34].
Severe disease — There are several challenges to establishing the diagnosis of severe disease in pediatric C. difficile infection. There is no strict consensus on disease definitions, and adult scoring systems have not been validated in children.
We use the following clinical features to suggest severe C. difficile infection in children (table 1):
●Profuse diarrhea with systemic findings (fever, rigors, or severe abdominal pain, tenderness, or distention)
●Elevated white blood cell count (>15,000 cells/microL)
●Elevated age-adjusted serum creatinine level
●Pseudomembranous colitis (discussed below)
Children with features of severe disease may warrant more aggressive treatment. (See "Clostridioides difficile infection in children: Treatment and outcome".)
Pseudomembranous colitis — Pseudomembranous colitis (severe inflammation of the inner lining of the bowel) from C. difficile infection is uncommon in children. There are limited descriptions of pseudomembranous colitis in children, possibly because children with C. difficile infection often are diagnosed without endoscopy.
Intestinal perforation, ileocolic intussusception, pneumatosis intestinalis, and rectal prolapse (with pseudomembranes visible on the extruded segment) are rare manifestations [37-42]. Paralytic ileus may lead to a paradoxical absence of stool, which may delay diagnosis. Severe cases may have radiographic findings (ie, mucosal thickening, "thumbprinting" (image 1A), dilation). (See 'Radiographic features' below.)
There are reports of severe or fatal pseudomembranous colitis in neonates and infants younger than two years [43-47]. Some, but not all, cases had an identifiable risk factor such as antibiotic exposure. Other reported cases occurred in infants with gastrointestinal obstruction or prematurity but no other explanation for pseudomembranous colitis. However, given the high frequency of toxigenic C. difficile carriage in this age group, it is difficult to know definitively whether C. difficile was a causal or incidental finding.
The diagnosis of pseudomembranous colitis is confirmed by visualization of pseudomembranes on endoscopy or histologic examination of biopsied tissue. Although not usually performed in children, sigmoidoscopy or colonoscopy demonstrates characteristic adherent yellow plaques (picture 1A-B). Immediate treatment is crucial to prevent progression to more fulminant disease. (See "Clostridioides difficile infection in children: Treatment and outcome", section on 'Severe disease'.)
Fulminant disease — Fulminant C. difficile infection is characterized by ileus, toxic megacolon, hypotension, or shock (table 1). The manifestations of fulminant infection can include fever, diarrhea, severe lower quadrant or diffuse abdominal pain, abdominal distention, hypovolemia, leukocytosis, hypoalbuminemia, and lactic acidosis [48,49]. Diarrhea may be absent in patients with prolonged ileus.
Potential complications of fulminant colitis include toxic megacolon and bowel perforation [50].
●Toxic megacolon is a clinical diagnosis based upon radiographic evidence of colonic dilatation associated with severe systemic toxicity. Abdominal radiographs may demonstrate small bowel dilatation, air-fluid levels (mimicking an intestinal obstruction), and "thumbprinting" (scalloping of the bowel wall) due to submucosal edema (image 1A and image 2). (See "Toxic megacolon".)
●Bowel perforation presents with abdominal rigidity, involuntary guarding, diminished bowel sounds, rebound tenderness, and severe localized tenderness in the left or right lower quadrants. Abdominal radiographs may demonstrate free abdominal air.
Aggressive diagnostic and therapeutic interventions are warranted in the setting of fulminant C. difficile infection. Prompt surgical consultation is warranted to assess the need for surgical intervention [51]. (See "Clostridioides difficile infection in children: Treatment and outcome", section on 'Fulminant disease'.)
Extracolonic findings — Rare cases of extracolonic infection from C. difficile have been described in children, including skin and soft tissue infections, bacteremia, and osteomyelitis [52-54]. Reactive arthritis associated with C. difficile infection has also been described in adults and children [52,55-58]. (See "Reactive arthritis", section on 'Preceding infection'.)
Recurrent infection — Recurrence occurs in 12 to 30 percent of children with C. difficile infection [32,34,35,59-61]. Some children have multiple recurrences which can be refractory to conventional treatments.
Recurrent C. difficile infection is a return of symptoms with a positive assay result after a period of initial symptom resolution that occurs within eight weeks after the initial episode (typically within one to three weeks) [62-64]. The signs and symptoms of recurrence are typically similar to those of the initial episode.
Recurrent episodes may represent a relapse of the previous infecting strain or reinfection with a new strain [65]. The mechanism of recurrent disease is poorly understood. The persistence of dormant spores within the intestines, low levels of antitoxin antibody levels, and continued disturbance of intestinal flora are thought to contribute to recurrent disease [7,66,67]. Antibiotic resistance has not been shown to be a factor in recurrence.
In observational studies, recurrent C. difficile infection in children has been associated with community-associated disease, receipt of concomitant antibiotics, malignancy, tracheostomy tube dependence, recent surgery, the number of classes of antibiotic exposure, and increased inflammatory markers in stool (eg, lactoferrin, calprotectin, interleukin-8) at the time of initial infection [35,60,68,69].
Management of recurrent C. difficile infection is discussed separately. (See "Clostridioides difficile infection in children: Treatment and outcome", section on 'Management of recurrence'.)
LABORATORY FEATURES — Ancillary laboratory studies are not necessary for the diagnosis of C. difficile infection, though they can be helpful in determining disease severity (the severity scoring system used for adults relies upon laboratory findings [63], but this scoring system has not been validated for children).
Laboratory findings suggestive of severe or fulminant disease in adults include elevated white blood cell (WBC) count (>15,000 WBCs/microL) or elevated age-adjusted serum creatinine level [63,70]. Individuals with severe disease may also have lactic acidosis or serum albumin level <2.5 g/dL (25 g/L). Although leukocytosis has been reported to be an early marker of C. difficile infection in adults, preceding the onset of diarrhea by one to two days [48,49], leukocytosis does not predict C. difficile test positivity [71].
In various case series, blood in the stool (gross visible blood or stool guaiac test positive) has been found in 10 to 25 percent of children with C. difficile infection [32-34]; fecal leukocytes can also be present [33,72,73].
RADIOGRAPHIC FEATURES — Imaging studies are not necessary for the diagnosis of C. difficile infection, but they may provide the first indication of a complication related to severe disease.
Abnormal radiographic changes are uncommon in children with mild disease. For children with moderate to severe disease or complications from C. difficile infection, the following radiographic features may be seen on abdominal imaging (computed tomography, plain film, or ultrasonography) [72,74-76]:
●Thickened colon wall (image 1A-C)
●"Accordion sign" created by contrast or gas trapped between swollen haustra (characteristic but not pathognomonic)
●Wall nodularity
●Pericolonic stranding
●Ascites
●Dilated colon in patients with paralytic ileus or toxic megacolon (image 2)
●Free air (in patients with colonic perforation)
APPROACH TO DIAGNOSIS — The approach to diagnosis presented below is largely compatible with the recommendations of the American Academy of Pediatrics Committee on Infectious Diseases (2021) [77], the Infectious Diseases Society of America and Society for Healthcare Epidemiology of America (2017) [63], and the American College of Gastroenterology (2021) [78].
Clinical evaluation — The evaluation is centered on identifying clinical and laboratory findings suggestive of C. difficile infection and differentiating C. difficile infection from other causes of acute-onset diarrhea in children.
Although it is difficult to distinguish diarrhea due to C. difficile infection from more common causes of diarrhea in children (ie, norovirus, rotavirus) solely based on symptoms, certain clinical and laboratory features increase the likelihood that C. difficile is a causative pathogen (table 2) [72,77,79].
The most common symptom of C. difficile infection is acute-onset watery diarrhea (defined as ≥3 loose bowel movements per day), often occurring during or shortly after a course of antibiotics. Profuse diarrhea (≥10 loose bowel movements per day) and lower abdominal pain and tenderness can occur with pseudomembranous colitis. Fever may be present. If nausea and vomiting are a significant component of the gastrointestinal symptoms, viral etiologies should be considered.
Children with severe or fulminant disease may have leukocytosis (>15,000 white blood cells/microL), elevated age-adjusted creatinine, hypoalbuminemia (<2.5 g/dL [25 g/L]), or lactic acidosis. Children with severe C. difficile infection can have a rapid reduction of stool output associated with worsening abdominal pain and distention. In such patients, abdominal imaging may be warranted to assess for the development of an ileus or toxic megacolon [48,80]. (See 'Radiographic features' above.)
The following risk factors and predisposing conditions should lower the threshold for consideration of C. difficile infection in children with acute-onset watery diarrhea (see "Clostridioides difficile infection in children: Microbiology, pathogenesis, and epidemiology", section on 'Risk factors'):
●Antibiotic exposure
●Gastric acid suppressants
●Gastrointestinal feeding device (eg, gastrostomy or jejunostomy tube)
●Predisposing conditions
●Malignancy
●Hematopoietic cell or solid organ transplantation
●Inflammatory bowel disease
●Cystic fibrosis
●Hirschsprung disease
●Immunodeficiency
●Structural or postoperative intestinal disorders
Indications for testing — Testing for C. difficile should only be performed in symptomatic children with clinically significant diarrhea (≥3 loose bowel movements per day or an acute worsening in those with chronic diarrhea) who have clinical features suggestive of C. difficile disease or predisposing conditions (table 2). Testing for C. difficile should not be performed in children who have received laxatives in the previous 48 hours [63,81].
Given the high rate of C. difficile colonization in early life, we provide the following age- and setting-dependent suggestions for testing:
●Infants <12 months of age who are not hospitalized in the neonatal intensive care unit (NICU) – Limit testing to those with Hirschsprung disease or other intestinal motility disorders.
●Infants <12 months of age hospitalized in the NICU – Testing for C. difficile in infants <12 months of age who are hospitalized in the neonatal intensive care unit (NICU) should be limited to those with [82]:
•Evidence of pseudomembranous colitis
•Clinically significant diarrhea after other causes of infectious (eg, norovirus, rotavirus, adenovirus, enterovirus) and noninfectious diarrhea (eg, opioid withdrawal, malabsorption, feeding intolerance, milk protein allergy) have been excluded (see "Approach to chronic diarrhea in neonates and young infants (<6 months)")
●Children 1 to 3 years of age – Limit testing to those in whom other causes of diarrhea have been excluded or there is a high suspicion for C. difficile infection.
●Children ≥3 years of age – Limit testing to those with clinical features compatible with C. difficile infection who have accompanying risk factors or predisposing conditions. (See 'Clinical evaluation' above.)
Indiscriminate testing of all children with diarrhea for C. difficile infection can lead to misdiagnosis and unnecessary antibiotic treatment – especially when nucleic acid amplification tests (NAATs) are used in low-risk, young children with community-onset diarrhea [83,84].
Testing for C. difficile — Laboratory testing for C. difficile infection involves detection of C. difficile toxin(s) or toxin gene(s) in a stool specimen [63,77]. The available assays cannot distinguish between colonization and infection.
Testing should only be performed on loose, watery, or semiformed stool specimens (ie, stool that takes the shape of the container) unless ileus or toxic megacolon is suspected [63,77,85]. Many laboratories do not perform routine testing of samples from children <12 months of age. For patients with suspected C. difficile infection and ileus, a perirectal swab may be necessary [86].
●Testing approach – When testing for C. difficile, we suggest either a NAAT assay alone or a multistep algorithm consisting of initial enzyme immunoassay (EIA) screening for glutamate dehydrogenase (GDH) antigen and toxins A and B, followed by NAAT if EIA assays are discordant (algorithm 1) [63,87-89]; the multistep algorithm is preferred for neonates in the NICU [82], when indicated (see 'Indications for testing' above). Other approaches that include a highly sensitive and a highly specific testing modality are reasonable [63,78]. As an example, the American College of Gastroenterology suggests initial testing with the highly specific NAAT or GDH EIA, followed by the highly specific EIA for toxins A and B if initial testing is positive [78].
The high sensitivity and specificity of the NAAT assays, together with their rapid turnaround time, allow for prompt isolation and treatment of patients with C. difficile infection. This decreases the opportunity for health care-associated spread and may improve patient outcomes. The multistep algorithm is a less expensive option but takes slightly longer compared with NAAT alone.
In children with moderate to severe diarrhea with negative results on C. difficile testing, detailed evaluation for other causes of diarrhea (including underlying gastrointestinal disorders) is warranted. Repeat NAAT testing of stool should not be performed for at least seven days [63,89,90]. (See 'Differential diagnosis' below and 'Radiographic features' above.)
●Stool tests – Several diagnostic tests are available that detect C. difficile toxin(s), toxin gene(s), or toxigenic C. difficile organisms, in decreasing order of sensitivity (table 3) [63,77].
•Toxigenic culture – The reference standard method of detecting toxigenic C. difficile organisms in stool specimens is the toxigenic culture [91]. It is performed by isolating C. difficile on selective culture media with subsequent toxin testing of the isolated colonies [91]. This method is limited by slow turnaround and need for technical expertise; it is rarely employed as a routine diagnostic test.
•NAATs – NAATs, including real-time polymerase chain reaction (PCR) and loop-mediated isothermal amplification, detect the genes that encode toxin A and/or B (tcdA and tcdB). NAATs are sensitive (95 percent) and specific (>97 percent) in detecting the presence of toxin-producing C. difficile organisms; however, they do not distinguish colonization from clinical infection [63,92]. Test results are rapidly available, permitting prompt isolation and treatment of patients with C. difficile infection. However, NAATs are expensive. They also may lead to identification of incidental C. difficile carriers because of their high sensitivity.
Multiplex PCR panels can test for multiple gastrointestinal pathogens simultaneously [93]; when applied to children with acute diarrhea, C. difficile coinfection with other enteric pathogens may be detected [26,94,95]. To avoid overdiagnosis and overtreatment of C. difficile infection, multiplex panels that include C. difficile must be interpreted with caution, particularly when obtained in children at low risk for C. difficile infection, in whom detection of C. difficile likely represents colonization [26].
•EIA for C. difficile GDH – GDH is an enzyme produced by all strains of C. difficile. Since GDH antigen testing cannot distinguish between toxigenic and nontoxigenic strains, it is only useful as an initial screening test in a multistep algorithm [91,96], where positive samples undergo additional testing (ie, EIA for toxin or NAAT for toxigenic organism) and negative samples are finalized and reported. GDH antigen testing is highly sensitive and has a rapid turnaround time. Two-step algorithms combining an initial GDH assay followed by NAAT assay are sensitive, economical, and permit rapid reporting of negative results [91,97-99].
•Cell culture cytotoxicity assay – The reference standard of detecting C. difficile toxins A and/or B in stool specimens has been the cell culture cytotoxicity assay. It is performed by placing a filtered stool suspension onto cell culture. If C. difficile toxin is present, a characteristic cytopathic effect will be noted after 24 to 48 hours [91]. The cell culture cytotoxicity assay has a high sensitivity and specificity and the strongest correlation with clinical outcomes [100]. However, it is labor intensive and requires technical expertise; it is rarely employed as a routine diagnostic test.
•EIA for toxin A and/or B – Stool EIA for toxin A and/or B is widely used for the diagnosis of C. difficile infection. These tests are inexpensive and provide results within hours. Their major limitation is poor sensitivity (70 percent, 95% CI 66-74) [63,92]. Repeat testing after an initial negative EIA test has little clinical utility [101]. In addition, colonized children can have a positive toxin test, so EIA may not fully differentiate between colonization and symptomatic C. difficile infection [31,102].
Endoscopy — Endoscopy is rarely needed to diagnose C. difficile infection. However, it may be warranted when an alternative diagnosis is suspected and visualization and/or biopsy of the intestinal mucosa is required. In such cases, sigmoidoscopy or colonoscopy may demonstrate pseudomembranous colitis and can provide fresh luminal content for C. difficile testing.
Pseudomembranous colitis is strongly supported by the observation of 2 to 5 mm yellowish plaques (picture 1A-B). Findings of typical pseudomembranous colitis usually are absent in flares of IBD associated with C. difficile infection.
DIAGNOSIS — The diagnosis of C. difficile infection is based on a combination of clinical and laboratory findings (table 2): clinically significant diarrhea (≥3 loose bowel movements per day) or ileus with a positive C. difficile test. The finding of pseudomembranous colitis (by endoscopy or histology) is suggestive of C. difficile infection and should prompt testing [63,77,79].
The diagnosis of C. difficile infection in children can be challenging. Validated clinical criteria that reliably distinguish between C. difficile colonization (where a positive stool assay is an incidental finding) and infection do not exist, nor is there a stool test or testing algorithm that can differentiate between colonization and infection. However, identifying clinical features suggestive of C. difficile makes it more likely that C. difficile infection is present in a child with a positive stool test. (See 'Symptomatic disease' above.)
DIFFERENTIAL DIAGNOSIS — The differential diagnosis of C. difficile infection in children depends upon the clinical setting [103,104]:
●Nonbloody diarrhea
•Viral enteritis – Mild C. difficile infection may be indistinguishable from diarrhea due to enteric viruses such as norovirus, rotavirus, and many others. In infants and young children, mild, self-limited diarrhea is most likely to be viral. Testing for viral pathogens may be helpful in establishing a diagnosis of viral diarrhea, but generally is not necessary. (See "Diagnostic approach to diarrhea in children in resource-abundant settings" and "Clinical manifestations and diagnosis of rotavirus infection", section on 'Clinical manifestations'.)
•Parasitic infection – Intestinal parasites, such as Giardia intestinalis, Cryptosporidium, and others, may cause relatively prolonged watery diarrhea. Parasites generally are diagnosed with stool microscopy or antigen-based assays. (See "Cryptosporidiosis: Epidemiology, clinical manifestations, and diagnosis", section on 'Microscopy' and "Giardiasis: Epidemiology, clinical manifestations, and diagnosis", section on 'Stool microscopy'.)
●Bloody diarrhea
•Bacterial infection – Salmonella, Shigella, Yersinia, Campylobacter, and toxigenic or invasive Escherichia coli may present with fever, inflammatory and/or bloody diarrhea, or laboratory signs of inflammation (eg, leukocytosis). Testing for bacterial etiologies may be warranted for children with community-acquired diarrhea. However, bacteria other than C. difficile rarely cause hospital-onset diarrhea [105]. (See "Shigella infection: Epidemiology, clinical manifestations, and diagnosis", section on 'Clinical manifestations' and "Shiga toxin-producing Escherichia coli: Clinical manifestations, diagnosis, and treatment", section on 'Clinical features'.)
•Noninfectious inflammatory or bloody diarrhea – Noninfectious causes of bloody diarrhea include:
-Inflammatory bowel disease (see "Clinical presentation and diagnosis of inflammatory bowel disease in children")
-Neonatal necrotizing enterocolitis, which has not been associated with C. difficile, even though heavy colonization is common in this age group
-Gastrointestinal mucositis in patients receiving chemotherapy or radiation therapy
●Pseudomembranous colitis – Cases of pseudomembranous colitis, in the absence of C. difficile infection, have been attributed to rotavirus, adenovirus, E. coli O157, Salmonella, Shigella, and antibiotic-induced Staphylococcus aureus enterocolitis [106-110]. (See "Shigella infection: Epidemiology, clinical manifestations, and diagnosis", section on 'Clinical manifestations' and "Clinical manifestations and diagnosis of rotavirus infection", section on 'Clinical manifestations' and "Shiga toxin-producing Escherichia coli: Clinical manifestations, diagnosis, and treatment", section on 'Clinical features'.)
●Fulminant colitis – Fulminant colitis must be differentiated from other causes of acute intra-abdominal disease (eg, appendicitis, intussusception, inflammatory bowel disease) [103]. (See "Causes of acute abdominal pain in children and adolescents" and "Emergency evaluation of the child with acute abdominal pain".)
SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Clostridioides difficile infection".)
SUMMARY AND RECOMMENDATIONS
●Clinical and laboratory features – The clinical spectrum of Clostridioides difficile infection ranges from asymptomatic colonization to fulminant colitis. The full spectrum of disease is possible in children. (See 'Clinical spectrum' above.)
Symptoms of C. difficile infection typically begin either during or within a few weeks after antibiotic therapy. Mild to moderate diarrhea is the most common presentation of symptomatic C. difficile infection. The diarrhea is usually watery and may be accompanied by fever, lower abdominal cramping, and other systemic signs. The physical examination is typically normal or may demonstrate lower abdominal tenderness. (See 'Diarrhea' above.)
Laboratory findings that suggest severe or fulminant C. difficile infection include elevated white blood cell count (>15,000 cells/microL) or elevated age-adjusted serum creatinine level. (See 'Laboratory features' above.)
Certain patient characteristics and clinical features are more indicative that gastrointestinal symptoms in a given patient are due to C. difficile infection (table 2). In the absence of these factors, C. difficile is less likely than other causes of diarrhea. (See 'Approach to diagnosis' above.)
●Indications for testing – Testing for C. difficile should only be performed in symptomatic children with clinically significant diarrhea (≥3 loose bowel movements per day) who have clinical features suggestive of C. difficile infection or predisposing conditions (table 2). (See 'Indications for testing' above.)
●Approach to testing – A number of stool tests are available for C. difficile infection (table 3).
We suggest either a nucleic acid amplification test (NAAT) assay or a multistep algorithm consisting of a glutamate dehydrogenase (GDH) antigen assay combined with a toxin assay, followed by NAAT testing if the results are discordant (algorithm 1). Testing should only be performed on loose, watery, or semiformed stool specimens and is rarely indicated in children <12 months of age. (See 'Testing for C. difficile' above.)
●Diagnosis – The diagnosis of C. difficile infection in children is based on a combination of clinical and laboratory findings: clinically significant diarrhea (acute change in stooling habits and ≥3 loose bowel movements per day) or ileus with a positive C. difficile diagnostic test. Findings of pseudomembranous colitis (by endoscopy or histology) suggest C. difficile infection and should prompt further testing. (See 'Diagnosis' above.)
●Differential diagnosis – The possibility that C. difficile is an incidental finding, particularly in infants and young children, must be considered. Even if such children have positive tests for toxigenic C. difficile, alternative causes of diarrhea (viral, parasitic, and other bacteria) often are more likely and treatment for C. difficile may not be indicated. (See 'Differential diagnosis' above.)
ACKNOWLEDGMENT — The UpToDate editorial staff acknowledges Michael Cooperstock, MD, MPH, who contributed to an earlier version of this topic review.
Do you want to add Medilib to your home screen?