The trusted source for
healthcare information and
Clostridium Difficile Infection in Critically Ill Patients
By Richard J. Wall, MD, MPH, Pulmonary Critical Care & Sleep Disorders Medicine, Southlake Clinic, Valley Medical Center, Renton, WA, is Associate Editor for Critical Care Alert.
Dr. Wall reports no financial relationship to this field of study.
Clostridium difficile infection (CDI) is a common and severe nosocomial infection which is being seen with alarming frequency in hospitalized patients. Although the organism was first described more than 70 years ago, clinicians didn't start seeing serious C. difficile outbreaks until the past decade. Since 2000, epidemiologists have chronicled a frightening increase in CDI severity, associated mortality, and reduced efficacy of therapies.
Intensive care unit (ICU) patients are particularly susceptible to CDI because they often have numerous risk factors. In addition, the diagnosis of CDI is often more challenging in an ICU population because the typical signs and symptoms of CDI are commonly seen in ICU patients for myriad other reasons.
Epidemiology and Risk Factors
C. difficile is a gram-positive, spore-forming anaerobic bacillus. Although common in the general environment, only 1%-4% of healthy adults carry the organism because normal colonic flora confers resistance against it. In adults, the loss of such resistance is typically due to antibiotics. Certain antibiotics are particular culprits including clindamycin, fluoroquinolones, and second- and third-generation cephalosporins. Most cases present during or shortly after antimicrobial use. Other events may impair resistance of normal flora, including bowel preparation for colonoscopy or surgery, chemotherapy, and colitis.
The incidence of CDI is rising and costs are reported to reach $1-$3 billion in the United States annually. Most cases are nosocomial and iatrogenic. Between 2000 and 2005, the number of U.S. adult hospitalizations with a diagnosis of CDI increased an average of 23% per year.1 Much of the excess disease burden has been in elderly and critically ill populations.
As incidence is rising, overall disease severity is also worsening. Between 1999 and 2004, mortality rates quadrupled. These changes are due to a new strain, which demonstrates new virulence factors and fluoroquinolone resistance. Three key factors contribute to the development of CDI: 1) impairment of normal flora with antibiotics; 2) exposure to C. difficile or its spores; and 3) compromised host health status. Other potential risk factors are age > 60 years, long length of stay, mechanical ventilation, post-pyloric tube feeding, and gastric acid suppression.2
Is There a Link with Gastric Acid Suppression?
Recent attention has turned to the possible relationship between proton pump inhibitor (PPI) use and CDI. This relationship has biologic plausibility because it is well known that gastric juices are better at killing C. difficile and neutralizing its toxin if they are more acidic. Since PPIs possess more potent acid suppression than histamine2-receptor antagonists (H2-RAs), much of the scrutiny has focused on the PPI class. It is important to recognize that this entire debate is based on observational studies and still somewhat controversial. It seems unlikely there will ever be a randomized study specifically looking at this issue, however.
In a recent retrospective cohort study of more than 101,000 patients in a tertiary medical center, Howell et al showed a dose-response effect between PPIs and CDI, wherein increasing levels of acid suppression were independently associated with an increased risk of CDI.3 Compared with no acid suppression, receipt of a daily H2-RA was associated with a 53% increase in the odds of developing CDI. The use of a PPI was associated with a 74% increase in the odds, and patients who received more than once-daily PPIs had more than a doubling of that risk.
Numerous studies have shown that PPIs are among the most frequently prescribed medications in the ICU, and the routine use of stress ulcer prophylaxis is not needed in every ICU patient. At my hospital, our current practice is shifting toward the use of more judicious stress ulcer prophylaxis and the use of PPIs only when there is a compelling need.4
The diagnosis of CDI is a clinical one based on the onset of diarrhea during or after antimicrobial use.5 The diarrhea may be associated with abdominal discomfort, fever, or leukocytosis. Up to 20% of critically ill patients may suffer from ileus and lack diarrhea. The onset of symptoms may range from 1 day to 10 weeks after antibiotics are first given, but most cases begin within 3-7 days. The diagnosis is made by demonstrating C. difficile or its toxins in a stool sample.
Pathogenic strains produce two large protein exotoxins, toxin A and toxin B. Most laboratories use enzyme immunoassays (EIAs) for detecting the toxins. These assays are quick and easy, often providing same-day results. However, EIAs are not perfect (sensitivity ≤ 90%) and an assay directed against one toxin will not recognize the other toxin. Increasingly, hospitals use EIAs that detect both toxins, but it's always best to ask if your laboratory doesn't explicitly tell you. Because the EIA is an imperfect test, trust your clinical instincts. The "gold standard" test is the cytotoxicity cell assay, but it is impractical for routine use because of cost and time delays.
Computed tomography of the abdomen is rarely used in the diagnosis of CDI. However, it may reveal patterns consistent with colitis, such as colonic wall thickening (> 4 mm), wall nodularity, pericolonic stranding, and ascites. Such characteristic findings, however, do not necessarily correlate with disease severity. Flexible sigmoidoscopy to visualize pathognomic pseudomembranes may be a useful adjunct if the diagnosis cannot be delayed, but concerns about potential perforation make this a risky endeavor and it is usually not recommended.
Treatment depends on disease severity (see Table, below).2,5 "Mild-to-moderate" cases consist of only diarrhea or cramping, unaccompanied by systemic symptoms. "Severe" cases include patients with abundant diarrhea, abdominal pain, leukocytosis, fever, or any other systemic symptoms. Although these terms are somewhat vague, most guidelines consider any patient in the ICU to have "severe" disease.
CDI is a dynamic disease, which can become progressively more serious after treatment is initiated. Daily re-assessment of disease severity during treatment is important. If possible, patients should have the inciting antibiotic(s) stopped. At a minimum, changing to a more narrow-spectrum regimen should be considered. Agents with anti-peristaltic effects should be minimized, whenever possible. Of note, there is no need to retest stool samples during treatment because successfully treated patients shed C. difficile and toxins for weeks after they recover.
Metronidazole and vancomycin are the most common antibiotics used, and the organism remains susceptible to both medications in vitro. However, recent studies confirm that vancomycin should be first-line therapy in severe cases. Both agents should be given orally if patients can tolerate that route. However, metronidazole may also be given intravenously because its biliary excretion and exudation across inflamed colonic mucosa allow adequate concentrations to reach the colon.
Metronidazole is first-line therapy for mild-to-moderate disease, whereas oral vancomycin has better clinical outcomes in patients with severe disease. In a randomized trial of 150 patients, metronidazole resulted in a cure rate of only 76% in the most severe cases, compared with a 97% cure rate for vancomycin.6 There was no difference between the two agents for mild-to-moderate cases. For this reason, vancomycin is the preferred initial therapy for patients with severe illness or who are at risk for progressing to severe disease. Of note, every ICU patient with CDI in this trial was classified and treated as "severe."
In some severe cases, absent bowel motility may prevent adequate amounts of vancomycin from reaching the colon. One approach is to increase the oral vancomycin dose. Alternatively, intravenous metronidazole may be added. Another less rigorously evaluated approach is intracolonic vancomycin.
Consider surgical consultation when colitis is so severe that antibiotic efficacy is doubtful. Severe cases can progress to ileus, toxic megacolon, and an acute abdomen. Although fulminant CDI requiring colectomy is rare overall, it is occasionally a necessity in critically ill patients. Operative survival is better when surgical intervention is performed earlier in the course of antibiotic failure. Persistent leukocytosis and elevated lactate are ominous signs. Systemic symptoms increase with the severity of the colitis. For a thoughtful discussion on the role of surgery, refer to Noblett and colleagues.7
Multiple other approaches have been (or are being) attempted, but none has yet proven superior to the traditional regimens above.8 Rifaximin, nitazoxanide, and fusidic acid are equally efficacious to vancomycin or metronidazole. A randomized trial comparing oral metronidazole vs combined oral rifampin-metronidazole showed higher mortality in the combination group. Other unproven approaches include a novel macrocycle antibiotic (still under study), various probiotics (i.e., live microorganisms to restore colonic flora), intravenous immunoglobulin to boost immune response, administration of nontoxigenic C. difficile, and various agents for neutralizing toxins. Currently, the first randomized trial of fecotherapy is being conducted in the Netherlands.
Treatment Failure or Relapse
"Treatment failure" is defined as no response after 1 week, because nearly all patients should show signs of improvement within 3 days. If diarrhea does not improve, treatment should be stepped-up (see Table, below), albeit there is no robust evidence to support this advice. In patients with adynamic ileus, intravenous metronidazole should be added. Vancomycin administered via retention enema is another way to increase colonic antibiotic exposure. Surgery remains a final life-saving option.
|Table. Treatment regimens for Clostridium difficile colitis.|
|Category||Details||Treatment||Dose and Route|
|Mild-to-moderate||Non-ICU patients|| Stop (or narrow) inciting antibiotic
|500 mg PO q6h
× 10-14 days
|Severe||ICU patients|| Vancomycin||125 mg PO q6h
× 10-14 days
|Relapse #1||In most cases, assume
| Same as original treatment|
or treatment failure
|Not responding to therapy;
(e.g., hypotension, marked
| Vancomycin||500 mg q6h
× 10-14 days;
|Ileus; cannot tolerate PO
| Metronidazole||500 mg IV q6h
× 10-14 days
Despite therapy, approximately 20% of patients relapse, usually within 1-3 weeks. In the past, the return of symptoms upon completion of a treatment regimen was always thought to be from sporulation and disease recurrence. However, it is now recognized that up to 60% of "recurrences" are actually new infections with a new strain. For this reason, the antibiotic regimen that was used to treat the original episode may be used for the first recurrence. In second or subsequent relapses, however, vancomycin is always recommended.
Preventive measures are an essential part of any CDI containment program.9,10 The organism is very hardy in its vegetative state and even hardier once it sporulates. In general, diarrhea in a CDI patient should be considered a contagion.
A successful CDI preventive program requires a multidisciplinary team approach. Perhaps the most important piece is proper antibiotic stewardship. In addition, the ICU staff must promptly employ contact isolation procedures for any suspected or confirmed cases. Janitorial staff must be educated on the concept of spores and the hospital must develop an environmental decontamination program that uses chlorine-containing agents. Handwashing with soap and water is important because alcohol-containing foams do not remove spores from hands.
Infection with C. difficile is a serious problem for all hospitalized patients, particularly those in the ICU. Over the past decade, the disease has dramatically increased in both incidence and severity. Although numerous antibiotics have been tried, metronidazole and vancomycin remain the cornerstones of therapy. Preventive measures are as important as antibiotics when addressing this problem.