OVERVIEW: What every practitioner needs to know

Are you sure your patient has C. difficile infection (CDI)? What should you expect to find?

  • Diarrhea is the key symptom of Clostridium difficile infection, with various degrees of abdominal cramping and pain in most patients, but accompanied by fever in a minority (28%). Occasionally, patients also have nausea and vomiting.

  • Key physical findings are a tender abdomen to palpation. There may be abdominal distension in a minority of patients. Tympany to percussion is variable. Bowel sounds can be normal, increased, or decreased with about 20% exhibiting ileus. Stool is watery and non-bloody in the majority of patients.

How did the patient develop CDI? What was the primary source from which the infection spread?

  • The source for C. difficile will be the environment of the hospital or nursing home or the hands of healthcare workers. Asymptomatic carriage of C. difficile (so-called endogenous C. difficile) occurs very infrequently, probably no more than 1-2% of community residents in Western cultures. However, patients in hospitals or nursing homes may acquire C. difficile and can carry it for weeks to months without having symptoms as a result of having a good antibody response to the toxins of C. difficile. These patients are at reduced risk of symptomatic CDI, and attempts to decolonize them should be avoided as it is almost certain to fail and can cause harm if a patient colonized with a harmless non-toxigenic strain of C. difficile is decolonized.

    Continue Reading

  • Antimicrobial use, including many chemotherapeutic agents, is by far the greatest risk for CDI. The Centers for Disease Control and Prevention (CDC) estimates that 94% of all CDI is healthcare exposure related, either through inpatient or outpatient care. Additional risks include advanced age, immunosuppression, recent gastrointestinal surgery, presence of a gastric tube, and possibly use of proton pump inhibitors (PPIs), although the latter risk is still disputed and retrospective risk analyses are about equally split on whether PPIs are a risk for CDI or for more severe CDI.

Which individuals are of greater risk of developing CDI?

  • Conditions that predispose to CDI include antimicrobial use within the past 3 months; healthcare exposure (inpatient or outpatient); advanced patient age, especially those older than 75 years of age; and patients with underlying immunosuppressive diseases.

  • The most common predisposing factor is antimicrobial exposure within the past 3 months.

Beware: there are other diseases that can mimic CDI:

  • The most common mimic of CDI is antibiotic associated diarrhea NOT caused by C. difficile. The two processes are sometimes extremely difficult to differentiate; however, CDI patients usually have more diarrheal bowel movements and are generally more ill than those without CDI. Other infectious causes of diarrhea, such as Campylobacter, Salmonella, Shigella, toxigenic E. coli, and viruses, such as norovirus, need to be ruled out if the patient is coming from the community. If the patient has been hospitalized for 3 or more days at onset of diarrhea, C. difficile is virtually the only bacterial cause of diarrhea that has been found. Norovirus, however, can be found in hospital and nursing home patients at any time.

What laboratory studies should you order and what should you expect to find?

Results consistent with the diagnosis
  • Peripheral white blood cell (WBC) is elevated in about 50% of patients with CDI. Elevation above 15,000 or 20,000/mm3 is indicative of severe CDI, depending on the severity scale used. The elevated WBC is due to increased neutrophils and a left shift is often noted. Extremely high WBC has been found in very severe complicated or fulminant CDI and may reach 50,000 to more than 100,000/mm3. These patients are usually extremely ill and may require colectomy as a life-saving procedure.

  • Serum creatinine increases of greater than or equal to 1.5 times baseline are also indicative of severe CDI and should be measured in patients as part of the initial assessment of disease severity as this will influence the choice of therapy.

  • Commonly, ascites is present with severe CDI. When paracentesis has been done on the ascites fluid, it has been found to be culture negative for C. difficile and for other colon organisms unless perforation is present.

Results that confirm the diagnosis
  • Stool examination for C. difficile or its toxins is the major diagnostic test modality.

  • A positive enzyme immunoassay (EIA) for C. difficile toxins A or B has been the most common test utilized during the past 25 years, but it is relatively insensitive, detecting only about 50-80% of patients who have diarrhea and toxigenic C. difficile in stool.

  • Stool culture for C. difficile is the most sensitive test, but turnaround is too slow for practical use, requiring at least 48 hours for culture and confirmation of a toxigenic strain.

  • Nucleic acid amplification tests (NAAT) detect the toxin genes of C. difficile in stool using either polymerase chain reaction (PCR) or loop mediated isothermal amplification (LAMP) technology. These tests are about 90-95% as sensitive as stool culture and detect only toxigenic strains of C. difficile.

  • An additional approach to diagnosis is detection of glutamate dehydrogenase (GDH) in stool using an EIA test, and combining this with a test for presence of toxin. Since GDH is found in both toxigenic and non-toxigenic C. difficile, confirming toxin or toxin gene presence is essential to rule out a non-toxigenic strain. GDH testing is less expensive than NAAT and is about 75-85% sensitive compared to stool culture.

  • For discrepant results between GDH and EIA for toxin, a NAAT test (PCR) is used to resolve the results.

  • C. difficile can also be diagnosed by visualization of pseudomembranes in the colon using flexible sigmoidoscopy or colonoscopy. This is very expensive compared to stool testing and is only 50% sensitive compared to stool testing in detection of CDI. Thus, a negative endoscopy does not rule out CDI, whereas a positive endoscopy makes the diagnosis. This low sensitivity is probably because, in early CDI, the pseudomembranes may not have formed or, if sigmoidoscopy is used, failure to visualize the entire colon.

What imaging studies will be helpful in making or excluding the diagnosis of CDI?

  • Most patients with CDI are treated without the need for any imaging studies. If imaging is needed because of ileus or suspected surgical abdomen, the abdominal/pelvic CT scan is most useful in identifying or ruling out CDI as a cause. In CDI, the abdominal CT scan typically shows thickening of the colonic mucosa, stranding, and ascites. The colon is dilated only in toxic megacolon. If present, free air indicates likely colon perforation. Abdominal CT scan $$$

  • Abdominal X-ray is less sensitive, but can demonstrate colonic dilatation and free air, indicating perforation if an upright film is also done. Abdominal X-ray $$

  • ($ = 60-125, $$ 125-500, $$$ 500-1,000, $$$$ > 1,000)

What consult service or services would be helpful for making the diagnosis and assisting with treatment?

If you decide the patient has CDI, what therapies should you initiate immediately?

Infectious Diseases, Gastroenterology, and General Surgery for severe cases would be helpful for making the diagnosis and assisting with treatment.

Stopping offending antibiotics is the first step in treating CDI. Initiation of specific treatment for CDI should be based on positive results of stool testing for toxin or the C. difficile organism or its toxin genes. If EIA testing is done in the clinical laboratory, the sensitivity is low and will miss up to 50% of cases. In these cases, treatment should be started empirically if CDI is suspected based on the clinical setting of new diarrhea in a hospitalized patient who has received antimicrobial treatment recently. About 25% of patients with CDI also resolve diarrhea with stopping of the offending antibiotic; however, in patients with severe illness, initiation of CDI-specific treatment is recommended to avoid delays in treatment that could lead to poor outcomes.

1. Anti-infective agents

If I am not sure what pathogen is causing the infection what anti-infective should I order?

Empiric treatment of CDI in the absence of a definitive diagnosis should be based on the severity of illness of the patient as follows:

  • Mild to Moderate CDI: (WBC < 15,000 or serum creatinine increase < 1.5 X baseline) Metronidazole 500 mg tid orally X 10-14 days

  • Severe CDI: (WBC > 15,000 or serum creatinine increase > 1.5 X baseline) Vancomycin 125 mg qid orally 10-14 days

  • Severe Complicated or Fulminant CDI: (hypotension, shock, megacolon, or ileus) Vancomycin 500 mg qid orally or via NG tube plus metronidazole 500 mg IV q 8 hours. If ileus is complete, consider vancomycin rectal instillation 500 mg in 100 ml normal saline every 6 hours as a retention enema.

If the epidemic BI/NAP1/027 strain of C. difficile is suspected or documented, prospective randomized trials indicate that the cure rate for both vancomycin or fidaxomicin will be significantly lower compared to treatment of other C. difficile organisms. Hence, for these patients, vancomycin treatment is recommended based on the much higher cost of fidaxomicin compared to vancomycin with no evidence of a better outcome for cure or recurrence rates.

Treatment of CDI

Both the Infectious Diseases Society of America (IDSA) and the Society for Healthcare Epidemiology of America (SHEA) and the European Society of Microbiology and Infectious Disease have created guidelines for the management of this very common disease:

IDSA/SHEA Guidelines:

  • Discontinue therapy with the inciting antimicrobial agent(s) as soon as possible, as this may influence the risk of CDI recurrence.

  • When severe or complicated CDI is suspected, initiate empirical treatment as soon as the diagnosis is suspected.

  • If the stool toxin assay result is negative, the decision to initiate, stop, or continue treatment must be individualized.

  • If possible, avoid use of antiperistaltic agents, as they may obscure symptoms and precipitate toxic megacolon.

  • Metronidazole is the drug of choice for the initial episode of mild-to-moderate CDI. The dosage is 500 mg orally 3 times per day for 10–14 days.

  • Vancomycin is the drug of choice for an initial episode of severe CDI. The dosage is 125 mg orally 4 times per day for 10–14 days.

    Vancomycin administered orally (and per rectum, if ileus is present) with or without intravenously administered metronidazole is the regimen of choice for the treatment of severe, complicated CDI.

    The vancomycin dosage is 500 mg orally 4 times per day and 500 mg in approximately 100 mL normal saline per rectum every 6 hours as a retention enema, and the metronidazole dosage is 500 mg intravenously every 8 hours.

  • Consider colectomy for severely ill patients. Monitoring the serum lactate level and the peripheral blood WBC count may be helpful in prompting a decision to operate, because a serum lactate level rising to 5 mmol/L and a WBC count rising to 50,000 cells per mL have been associated with greatly increased perioperative mortality. If surgical management is necessary, perform subtotal colectomy with preservation of the rectum

  • Treatment of the first recurrence of CDI is usually with the same regimen as for the initial episode but should be stratified by disease severity (mild-to-moderate, severe, or severe complicated), as is recommended for treatment of the initial CDI episode.

    Do not use metronidazole beyond the first recurrence of CDI or for long-term therapy because of potential for cumulative neurotoxicity.

  • Treatment of the second or later recurrence of CDI with vancomycin therapy using a tapered and/or pulse regimen is the preferred next strategy.

  • No recommendations can be made regarding prevention of recurrent CDI in patients who require continued antimicrobial therapy for the underlying infection.

European Society of Microbiology and Infectious Diseases Guidelines

  • Antiperistaltic agents and opiates should be avoided.

  • In general, strive to use antibiotics covering a spectrum no broader than necessary and narrow the antibiotic spectrum of treatment after results of cultures and/or susceptibility tests become known.

  • Mild CDI (stool frequency < 4 times daily; no signs of severe colitis), clearly induced by the use of antibiotics, may be treated by stopping the inducing antibiotic. Observe patients closely for any signs of clinical deterioration and place on therapy immediately if this occurs.

  • Treatment for an initial episode and a first recurrence of CDI:

    If oral therapy is possible:

    non-severe: metronidazole 500 mg tid orally for 10 days.

    severe: vancomycin 125 mg qid orally for 10 days.

    If oral therapy is impossible:

    non-severe: metronidazole 500 mg tid intravenously for 10 days.

    severe: metronidazole 500 mg tid intravenously for 10 days + intracolonic vancomycin 500 mg in100 mL of normal saline every 4–12 h and vancomycin 500 mg qid by nasogastric tube.

    Colectomy should be performed to treat CDI in any of the following situations:

    perforation of the colon.

    systemic inflammation and deteriorating clinical condition not responding to antibiotic therapy (this includes the clinical diagnoses of toxic megacolon and severe ileus). Colectomy should preferably be performed before colitis is very severe. Serum lactate may, inter alia, serve as a marker for severity (operate before lactate exceeds 5.0 mmol/L).

  • Treatment for a second recurrence of CDI and later recurrences:

    If oral therapy is possible:

    vancomycin 125 mg qid orally for at least 10 days.

    Consider a taper (e.g., decreasing daily dose with 125 mg every 3 days)/pulse (e.g., a dose of 125 mg every 3 days for 3 weeks) strategy.

    If oral therapy is impossible:

    metronidazole 500 mg tid intravenously for 10–14 days plus retention enema of vancomycin 500 mg in 100 mL of normal saline every 4–12 h and/or vancomycin 500 mg qid by nasogastric tube.

  • In all the above-mentioned cases, oral vancomycin may be replaced by teicoplanin 100 mg twice daily, if available.

2. Other key therapeutic modalities
  • Hydration to replace fluids lost with diarrhea is essential and is critical to the management of patients with hypotension or shock. Avoidance of additional antibiotics is also critical, as it has been shown that continued use or addition of antibiotics increases the time to resolution of diarrhea, reduces cure rates, and increases recurrence rates in patients with CDI.

  • Use of fecal transplants to reconstitute the normal fecal microbiota in patients with multiply recurrent CDI have been highly successful with resolution of CDI recurrence in over 90% of patients, but their use remains somewhat controversial largely because of the non-standardized manner in which they have been performed. No standardized methodology for doing fecal transplants and no controlled prospective trials have been conducted, although the efficacy appears very high in observational reports. Questions of safety of stool for transplant and the appropriate testing required prior to transplant to assure safety have not been resolved. In addition, stool transplants have been administered via naso-gastric tube, colonoscope, and retention enema with no comparative efficacy data. Similarly, stool donor selection remains controversial.

  • Surgical intervention in the form of subtotal colectomy can be life-saving in severely ill patients not responding to medical management. These severely ill patients should be followed by a surgeon at the earliest indication of severe complicated or fulminant CDI, as the course is highly unpredictable as to who will require surgery. Generally, surgical mortality is reduced if patients undergo surgery before the serum lactate level reaches 5.0 mmol/L or the WBC reaches 50,000 mm3. A diverting loop ileostomy followed by colonic lavage with polyethylene glycol followed by vancomycin may be an alternative to colostomy that can reduce mortality and morbidity, but only historical controls have been compared to date.

  • Controversial therapies include use of intravenous immunoglobulin (IVIG) to treat patients with severe or recurrent CDI. There are no prospective randomized or control studies, and the results of observational studies do not provide clear evidence of benefit.

  • Also controversial is the use of probiotics as an adjunct to CDI treatment. Evidence of benefit in reducing recurrence of CDI using Saccaromyces boulardii as an adjunct to high-dose vancomycin was marginally beneficial with a barely significant p = .05, but, when combined with low or medium dose vancomycin or metronidazole, there was no benefit. A prospective randomized double-blind placebo controlled trial of S. boulardii for prevention of CDI in hospital patients taking antibiotics showed no benefit in preventing either antibiotic associated diarrhea or CDI. Other studies using probiotics for primary prevention of CDI have shown success using a probiotic drink containing lactobacillus, but the study excluded so many high risk patients and had such an extraordinarily high CDI rate in placebo patients that it was difficult to interpret. A similar trial in an Asian hospital has also raised similar questions regarding the validity of the observations.

What complications could arise as a consequence of CDI?

What should you tell the family about the patient’s prognosis?

  • Complications of CDI include ileus, toxic megacolon, colon perforation, hypotension, shock, and death. CDI rates and mortality increase with each decade of age of the patient (see Table I). Patient prognosis is clearly related to the age and underlying illnesses of the patient. Severely ill patients who are not responding to medical treatment may require surgical intervention in the form of an emergency colectomy or diverting loop ileostomy and colonic lavage with polyethylene glycol followed by vancomycin infusion into the colon.

  • Family members of patients with severe CDI older than 80 years of age should be cautioned of the significant mortality risk and potential need for surgical intervention.

Table I.
Age (Years) CDI Rate per 1000 Admissions Attributable 30-Day Mortality Rate (%)
<40 3.5 2.6
41-50 11.2 1.2
51-60 20.0 3.2
61-70 24.4 5.1
71-80 38.3 6.2
81-90 54.5 10.2
>90 74.4 14.0

How do you contract CDI and how frequent is this disease?

CDI can occur with taking any antibiotic; however, certain antibiotics have been more frequently associated with CDI, namely cephalosporins, clindamycin, and fluoroquinolones. In addition, certain penicillins, such as ampicilin, amoxicillin, ampicillin-sulbactam, and amoxicillin-clavulanate, are also considered high risk.

In the absence of antibiotic exposure, the risk of CDI is virtually non-existent because of the protection afforded by the “colonization resistance” of the normal gastrointestinal microbiota. In addition, healthcare exposures account for well over 90% of CDI cases, presumably because of the increased risk of ingesting spores of the organism in healthcare environments. Spores are ingested in the healthcare environment either through direct contact with contaminated surfaces by the patient or by contact with the hands of healthcare workers who fail to practice good hand hygiene and glove wearing.

Patients who take antibiotics and ingest C. difficile spores do not necessarily become symptomatic with diarrhea. Most patients have protective antibodies directed at the toxins of C. difficile and remain asymptomatic but colonized with C. difficile. These patients are at reduced risk of symptomatic C. difficile infection but pose a potential risk to transmit C. difficile spores to other patients, but the extent of that risk has not been determined.

The majority of “shedding” of C. difficile spores into the hospital environment occurs from symptomatic patients with diarrhea, particularly those who are incontinent of stool. About 40% of strains of C. difficile in hospitals are non-toxigenic and incapable of producing disease symptoms. Patients who acquire these strains are colonized and protected from acquiring toxigenic C. difficile.

Piglets, horses, dogs, cats, and calves have all been documented to have CDI. There is evidence of the same organism that infects animals, particularly piglets, also infecting humans. Food contamination by C. difficile spores has been documented in beef, pork, poultry, leafy vegetables, and root vegetables, but no direct connection of CDI to exposure in food has ever been documented.

What pathogens are responsible for this disease?

How does C. difficile cause disease?

C. difficile is by far the most common infectious cause of diarrhea related to antibiotic ingestion and is spread through ingestion of spores of the organism from the environment (usually healthcare environment) or the hands of healthcare workers who fail to exercise good hand hygiene or to wear gloves.

The C. difficile spores are able to vegetate in the gastrointestinal (GI) tract following ingestion and, in the presence of a disrupted microbiota (caused by antibiotic use), are able to colonize the colon where they produce two toxins, toxin A and toxin B, that are responsible for the diarrhea and colitis that ensues. Toxin A is an enterotoxin that also causes cell cytotoxicity, and toxin B is a cell cytotoxin. Both toxins cause epithelial cell death through glucosylation of Rho GTPases leading to cell cytoskeleton reorganization, cell rounding, and apoptosis mediated by caspases. The inflammatory response is largely one of neutrophil infiltration of the mucosa and submucosa with extrusion of cellular debris and proteinaceous material onto the mucosal surface to form the pseudomembrane, which is seen at endoscopy as pseudomembranous colitis (PMC) in about 50% of all patients diagnosed with CDI (see Figure 1).

Figure 1.

Current Pathogenesis Model for CDI

Some C. difficile also possess a third toxin, called binary toxin, which also acts by glucosylating epithelial cells. One strain of epidemic C. difficile, known as North American pulse field type 1 (NAP1) or restriction endonuclease analysis (REA) group BI, or PCR ribotype 027, contains the three toxins (A, B, and binary) and is currently the most common strain found among isolates in North America; however, it is not clear whether binary toxin is contributing to the higher morbidity and mortality seen in patients infected with this strain.

Other rare causes of diarrhea associated with antibiotic use:

Rarely, antimicrobial associated diarrhea has been attributed to strains of Clostridium perfringens that produce enterotoxin. The detection of C. perfringens diarrhea is difficult because of the high frequency of the organism in stool but the low frequency of the presence of enteroxin in C. perfringens. Stool enzyme-linked immunosorbent assay (ELISA) for detection of C. perfringens enterotoxin has identified C. perfringens diarrhea transmission in hospital and estimated its frequency at about one-fourth that of CDI.

An additional rare cause of antimicrobial associated diarrhea is Klebsiella oxytoca, which causes a hemorrhagic antibiotic associated diarrhea distinctly different from the non-hemorrhagic diarrhea of C. difficile. Selective media for culture of K. oxytoca and for detection of hemorrhagic E. coli should be employed if bloody diarrhea is present.

Staphylococcus aureus enterocolitis was long considered the major cause of antimicrobial associated diarrhea in the 1950s and 1960s; however, since the advent of CDI in the 1970s, the documentation of S. aureus enterocolitis has been rare, estimated in one prospective study to occur about 1/60 as often as a cause of antibiotic associated diarrhea as C. difficile.

What other clinical manifestations may help me to diagnose and manage CDI?

Occasionally, patients with CDI present with unexplained leukocytosis and little evidence of gastrointestinal symptoms. Patients with severe disease may also present with an ileus rather than diarrhea.

The history almost always includes diarrhea and abdominal discomfort at some point in the evolution of CDI.

Physical exam may reveal absence of bowel sounds and distension with tenderness when ileus is present.

What other additional laboratory findings may be ordered?

Fecal lactoferrin may be present in CDI; its presence does not make the diagnosis of CDI, and CDI is often diagnosed in the absence of lactoferrin in stool.

How can CDI be prevented?

There are no known effective prophylactic methods to prevent CDI, other than avoidance of antimicrobials. Those that have been tried include oral metronidazole (no data and probably ineffective because of 100% absorption in the absence of diarrhea) and oral vancomycin, which will probably prevent CDI if the organism is ingested while it is being taken, but, when stopped, the patient will remain at risk of CDI because of the microbiota disruption as a result of taking vancomycin. Probiotics have also been used as primary prevention of CDI, but results remain inconclusive, including the use of a dairy probiotic drink, lactobacillus preparations, and Saccharomyces boulardii.

Two vaccines are currently undergoing clinical trials as are intravenous monoclonal antibodies against toxins A and B and oral spores of non-toxigenic C. difficile for prevention of CDI; however, definitive results of these studies are not expected for several years.

WHAT’S THE EVIDENCE for specific management and treatment recommendations?

Cohen, S, Gerding, DN, Johnson, S. “Clinical practice guidelines for infection in adults: 2010 update by the Society for Healthcare Epidemiology of America (SHEA) and the Infectious Disease Society of America (IDSA)”. Infect Control Hosp Epidemiol. vol. 31. 2010. pp. 431-55. (This is the most recent from SHEA/IDSA, and its treatment recommendations are provided as part of this chapter. The guideline is currently undergoing update because of the approval of fidaxomicin for treatment of CDI in 2011 and the increased adoption of NAAT testing for CDI in clinical laboratories, which has increased diagnostic sensitivity markedly compared to EIA tests. New guidelines are not expected to be available until 2013.)

Bauer, MP, Kuijper, EJ, van Dissel, JT. “European Society of Clinical Microbiology and Infectious Diseases (ESCMID): treatment guidance document for infection (CDI)”. Clin Microbiol Infect. vol. 15. 2009. pp. 1067-79. (This is widely used in the European Union and and the United Kingdom and antedates the SHEA/IDSA guideline. Recommendations for treatment from this guideline are provided as part of this chapter. Approval of fidaxomicin for treatment in Europe did not occur until 2012 and no EU treatment recommendations have been made to date as to its use, nor has a guideline update schedule been announced.)

Petrella, LR, Sambol, SP, Cheknis, A. “Decreased cure and increased recurrence rate for infection caused by the epidemic BI strain”. Clin Infect Dis. vol. 55. 2012. pp. 351-7. (This paper analyzes the treatment outcome of patients infected with the BI/NAP1/027 strain of C. difficile in two large randomized controlled studies and demonstrates that treatment cure is reduced and recurrence rate of CDI is increased when patients are infected with the BI/NAP1/027 epidemic strain of C. difficile compared to other C. difficile strains. This reduced cure and increased recurrence was statistically significant for both vancomycin and fidaxomicin treatments by multivariate analysis.)