Gastroenterology
Volume 136, Issue 4 , Pages 1152-1154, April 2009

Reining in Recurrent Clostridium difficile Infection—Who's at Risk?

  • Christina M. Surawicz

      Affiliations

    • Corresponding Author InformationReprint requests Address requests for reprints to: Christina M. Surawicz, MD, Professor of Medicine, University of Washington, Chief, Gastroenterology, Harborview Medical Center, 325 9th Avenue, Box 359773, Seattle, WA 98104. fax: (206) 744-8698

published online 26 February 2009.

Article Outline

 

See “Prospective derivation and validation of a clinical prediction rule for recurrent Clostridium difficile infection,” BY Hu MY, Katchar K, Kyne L, et al, on page 1206.

Clostridium difficile is a gram-positive, anaerobic bacillus that can cause antibiotic-associated diarrhea, colitis, and pseudomembranous colitis (PMC). This infection is now termed C difficile infection (CDI). CDI is usually a consequence of antibiotic therapy, although sporadic cases can occur. There have been recent reports of severe CDI in “low-risk” groups, including previously healthy pregnant women and individuals without prior antibiotic use. The 3 Rs that are current challenges in CDI are: rising rates, refractory disease, and recurrent infection (RCDI).

Back to Article Outline

1. Rising Rates of CDI 

C difficile rates have increased dramatically since 2001–2002, with well-documented epidemics in hospitalized patients in the United States, Canada, and Europe. Investigators in Quebec were early to publish the increased rates of CDI, with a 5-fold increase in CDI cases from 1999 to 2004.1 CDI rates in US hospitals tripled from 1995 to 2006. A hypervirulent strain has emerged, called NAP1/BI/027, which has a polymorphism in the tcd C inhibitory gene and produces more toxins A and B in vitro. The isolates also have quinolone resistance and produce a binary toxin. Increased toxin production may account for more severe disease, and widespread quinolone use selects for this strain. It is not known if the binary toxin contributes to pathogenicity. Rising rates have prompted increased attention to infection control and antibiotic stewardship, but epidemics remain a challenge in many hospitals and nursing homes.

Back to Article Outline

2. Refractory Disease 

In addition to higher rates of CDI, the hypervirulent strain has been associated with many more severe and refractory cases, that is, cases that do not respond to standard therapy. In 2002, Muto et al2 observed a dramatic increase in severe CDI in their hospitalized patients, with severe disease requiring colectomy and an increase in fatal cases. In Quebec, where thousands of patients have been affected, Pepin et al3 estimated that excess mortality attributable to CDI could be as high as 16%. Metronidazole has been the standard therapy with response rates of 90%–95%, but with more severe disease, response rates to oral metronidazole have decreased, and more patients need additional therapy such as vancomycin and/or intravenous metronidazole or require life-saving colectomy.4 This lack of response to therapy is due to severity of disease, not to antibiotic resistance. Vancomycin is now recommended as initial therapy for patients with severe disease.

Back to Article Outline

3. Recurrent CDI 

The third R is recurrent CDI. This was initially called “relapsing PMC.” The first series of 3 cases were reported in 1979, only a year after C difficile and its toxins were identified as the cause of PMC.5 These 3 cases are illustrative—all had severe disease, 2 were older (71 and 86 years), but 1 was 44 years old, 1 relapsed at 7 days, the others at 27 and 28 days. All responded to therapy, but 1 died as a result of a second relapse. “Recurrent CDI” is now the commonly used term and refers to CDI that occurs within 1 month after successful antibiotic therapy for CDI. One factor contributing to recurrence is impaired colonization resistance, allowing for proliferation of C difficile and production of toxins and disease. Another factor may be immune response as those in a small series who developed RCDI had lower levels of immunoglobulin (Ig)A and IgG to toxin B than those who did not recur.6 Published risk factors for RCDI include older age, intercurrent antibiotics, renal disease, and prior recurrences. A recent meta-analysis identified older age, antacid medications, and continued use of non–C difficile antibiotics after diagnosis to be risk factors for CDI.7 One recurrence makes further recurrences more likely.

Recurrence rates do not seem to be related to the treating antibiotic. The first large series of recurrent cases in 1980 documented recurrence in 11 of 79 (14%) patients with PMC treated with vancomycin.8 Since 1980, other antibiotics have also been studied and recurrence rates documented. Metronidazole has been most frequently studied, but there are also small series with teicoplanin, fusidic acid, and bacitracin (Table). In general, recurrence rates range from 10% to 20%. The most recently controlled trial comparing vancomycin and metronidazole documented similar recurrence rates (7% and 14%, respectively) with both antibiotics and no increase in those with severe disease.9

Table. Published Recurrence Rates With Various Antibiotics After Treatment of CDI
AntibioticNumber of studiesPublished recurrence rates (range)Recurrences/patients enrolledOverall recurrences
Vancomycin87%–42%41/26520%
Metronidazole414%–31%22/15519%
Teicoplanin30%–7.7%4/694.9%
Fusidic acid225%–28%13/4926.5%
Bacitracin141.6%5/1241.6%

However, there is evidence that more severe disease is associated with more recurrences. In Quebec, an increase in recurrence rates was documented during their epidemic. Rates rose from 96/622 (15%) in 1991–2002 to 104/323 (34%) in 2003–2004 and was correlated with increased severity of disease.10 Rates have subsequently decreased as the epidemic came under control.

Another risk factor seems to be inflammatory bowel disease. Two series documented a substantial increase in CDI in their inflammatory bowel disease patients,11, 12 with rates doubling and tripling in Crohn's disease and ulcerative colitis, respectively.11 Disease was more severe and immunosuppression and colitis were significant risk factors.12 A national survey of hospital discharges documented doubling in rates of C difficile among hospitalized ulcerative colitis patients between 1998 and 2004 with increased morbidity.13 The recurrence rate in 1 study was 17 of 46 (59%) patients during 2005, with 26% subsequently requiring colectomy and 22% with additional recurrences.14

Treatment is a challenge because there is no uniformly effective therapy—treatment requires antibiotics, often with pulse or tapered doses to break the cycle of infection, but these therapies are not uniformly effective.15 Probiotics, toxin binders, immunotherapy, and fecal bacteriotherapy have all been reported with variable success.16

The lack of uniformly effective therapy means that better treatments are necessary. Because overall RCDI rates are only 10%–20%, high-risk patients need to be identified for clinical trials to be feasible. The paper by Hu et al17 in this issue of Gastroenterology is an important step in that direction. The group, led by C. Kelly, reported previously reported the correlation of immune response and development of CDI and RCDI. In their prior prospective study of hospitalized patients who developed CDI, they found that risk factors for RCDI were age >65 years, severe or fulminant disease, and continued antibiotics after diagnosis for the initial CDI episode. In addition, patients with lower serum levels of IgM and IgG against toxin A had a higher risk of RCDI. In the current study, they go further to test the reliability of these criteria by developing a clinical predictor rule and testing it on a new cohort of patients with CDI.

The initial group had an unusually high recurrence rate: 22 of 44 (50%) is much higher than the expected 20%; the reasons for this are unclear, but may be related to severity of illness. Patients in the initial cohort had more serious underlying illness and more were in the intensive care unit. In addition, some were treated with both metronidazole and later vancomycin, suggesting lack of efficacy of metronidazole, an indirect measure of severity of disease. The second cohort was larger (n = 64), with recurrence in 13 (20%), which is more in line with published rates. The authors state that they did not recruit as many severely ill patients in the second study; 15 of the initial 89 died during follow-up, leaving a less severely ill study population.

Although the second cohort was less ill overall, and had a lower recurrence rate, the 3 predictive factors fared well. The high-risk group had ≥2 factors; the low-risk group had <2. A patient with all 3 predictive factors had an 87.5% probability of developing recurrence. Recurrence occurred in 7 of 19 (36.5%) in the high-risk group compared with 6 of 45 (13.3%) in the low-risk group, giving a diagnostic accuracy of 71.9%. Using a single criterion improved sensitivity but lowered specificity. However, addition of antitoxin A IgG data did not improve diagnostic accuracy indicating that simple clinical measures alone could identify many with a high risk of recurrence.

Do these criteria make sense? Yes, older age makes hospitalization and exposure to antibiotics and CDI more likely. Severity of illness has been shown to be associated with recurrences by others. In Quebec, with more severe cases, recurrence rates rose. Others have also documented poor outcomes with use of intercurrent antibiotics, which likely cause continued perturbations in the colonic flora that allows C difficile to persist. This study reinforces the importance of avoiding intercurrent antibiotics or choosing 1 with a narrow spectrum. This trio of factors is easy to use clinically and will be useful to aid future research as more effective therapies are sorely needed.

Research on RCDI should focus both on understanding pathophysiology and on treatments that prevent the first recurrence and on treat of additional recurrences. Prevention of the first recurrence would be ideal. As mentioned, recurrence rates are similar with current standard antibiotic therapies. In a small series, the new macrolide OPT-80 was tested at various doses; response rates ranged from 71% to 94%, but the recurrence rate was low, at 2 of 45 (5%), suggesting that narrow-spectrum antibiotics might have lower recurrence rates.18 Moreover, recurrence rates were only 3% with tolevamer, a nonantibiotic polymer that binds toxin, although it was not as effective as antibiotics in treating CDI in a phase III clinical trial.19 This suggests that nonantibiotic therapies should be explored, both as primary therapy or possibly as an adjunct to antibiotics. Although probiotics, including Lactobacillus GG and Saccharomyces boulardii, have been shown to prevent antibiotic-associated diarrhea, they have not been shown to prevent CDI. The probiotic S boulardii as an adjunct to antibiotic therapy helped to prevent further recurrences in those who already had a recurrence, but did not prevent recurrences in those with initial CDI. Some probiotics may be effective in preventing CDI and RCDI, but many more studies are needed. Moreover, one should be cautious because living organisms can pose a risk of infection in immunosuppressed patients.

Treating the patient who already has a recurrence is the second big challenge. The long list of published therapies attests to lack of a single effective approach. Antibiotics, different antibiotic regimens and combinations, toxin binders, probiotics, fecal bacteriotherapy, and immune approaches have all been pursued. If spores are important in pathogenesis, antisporicidal agents could be useful. Much has been written recently about protein pump inhibitor use and increased risk of gastrointestinal infection, including C difficile. This was a significant risk factor for RCDI in the recent meta-analysis7 and provides another area to be explored in preventing and treating RCDI. Finally, it may be time to apply these study results to current practice. In high-risk patients, it may make sense to use a taper and pulse antibiotic regimen to decrease the initial recurrence rate.

In summary, the epidemic strain of C difficile has refocused attention on the need for effective infection control policies to control rising rates, and better therapies to treat refractory disease and RCDI as well as strategies that prevent the first recurrence. Several new therapies are under study. This study will help to focus research on high-risk populations in the important search for better therapies for the difficult C difficile. A future issue of Gastroenterology (May 2009 supplement) will focus on gastrointestinal microbes, including CDI.

Back to Article Outline

References 

  1. Pepin J, Valiquette L, Alary ME, et al. Clostridium difficile-associated diarrhea in a region of Quebec from 1991 to 2003; a changing pattern of disease severity. Can Med Assoc. 2004;171:466–472
  2. Muto CA, Pokrymwka M, Shutt K, et al. A large outbreak of Clostridium difficile-associated disease with an unexpected proportion of deaths and colectomies at a teaching hospital following increased fluoroquinolone use. Infect Control Hosp Epidemiol. 2005;26:273–280
  3. Pepin J, Valiquette L, Cossette B. Mortality attributable to nosocomial Clostridium difficile-associated disease during an epidemic caused by a hypervirulent strain in Quebec. CMAJ. 2005;173:1037–1041
  4. Pepin J, Valiquette L, Gagnon S, et al. Outcomes of Clostridium difficile-associated disease treated with metronidazole or vancomycin before and after the emergence of NAP1/027. Am J Gastroenterol. 2007;102:2781–2788
  5. George WL, Volpicelli NA, Stiner DB, et al. Relapse of pseudomembranous colitis after vancomycin therapy. N Engl J Med. 1979;301:414–415
  6. Kyne L, Warny M, Qamar A, et al. Association between antibody response to toxin A and protection against recurrent Clostridium difficile diarrhoea. Lancet. 2001;357:189–193
  7. Garey KW, Sethi S, Yadav Y, et al. Meta-analysis to assess risk factors for recurrent Clostridium difficile infection. J Hosp Infection. 2008;70:298–304
  8. Bartlett JG, Tedesco FJ, Shuss S, et al. Symptomatic relapse after oral vancomycin therapy of antibiotic-associated pseudomembranous colitis. Gastroenterology. 1980;78:431–434
  9. Zar FA, Bakkanagari SR, Moorthi LST, et al. A comparison of vancomycin and metronidazole for the treatment of Clostridium difficile-associated diarrhea, stratified by disease severity. Clin Infect Dis. 2007;45:302–307
  10. Pepin J, Alary M-E, Valiquette L, Raiche E, et al. Increasing risk of relapse after treatment of Clostridium difficile colitis in Quebec, Canada. Clin Infect Dis. 2006;40:1591–1597
  11. Rodemann JF, Dubberke ER, Reske KA, et al. Incidence of Clostridium difficile infection in inflammatory bowel disease. Clin Gastroenterol and Hepatol. 2007;5:339–344
  12. Issa M, Vijayapal A, Graham MB, et al. Impact of Clostridium difficile on inflammatory bowel disease. Clin Gastroenterol Hepatol. 2007;5:345–351
  13. Nguyen GC, Kaplan GG, Harris ML, et al. A national survey of the prevalence and impact of Clostridium difficile infection among hospitalized inflammatory bowel disease patients. Am J Gastroenterol. 2008;103:1443–1450
  14. Issa M, Anathakrishnan AN, Binion DG. Clostridium difficile and inflammatory bowel disease. Inflamm Bowel Dis. 2008;14:1432–1442
  15. Surawicz CM, McFarland LV, Greenberg RN, et al. The search for a better treatment for recurrent Clostridium difficile disease: use of high-dose vancomycin combined with Saccharomyces boulardii. Clin Infect Dis. 2000;31:1012–1017
  16. Stepan C, Surawicz CM. Treatment strategies for recurrent and refractory Clostridium difficile-associated diarrhea. Gastroenterol Hepatol. 2007;1:295–305
  17. Hu MY, Katchar K, Kyne L, et al. Prospective derivation and validation of a clinical prediction rule for recurrent Clostridium difficile infection. Gastroenterology. 2009;136:1206–1214
  18. Louie T, Miller M, Donskey C, et al. Clinical outcomes, safety, and pharmacokinetics of OPT-80 in a Phase 2 Trial with Patients with Clostridium difficile infection. Antimicrob Agents Chemother. 2009;53:223–228
  19. Louie TJ, Peppe J, Watt CK, et al. Tolevamer, a novel nonantibiotic polymer, compared with vancomycin in the treatment of mild to moderately severe Clostridium difficile-associated diarrhea. Clin Infect Dis. 2006;43:411–420

 Conflicts of interest The author is on the speaker's bureau for Viropharma and Biocodex Inc.

PII: S0016-5085(09)00203-0

doi:10.1053/j.gastro.2009.02.023

Refers to article:

  • PodcastEditorial Accompanies Article Prospective Derivation and Validation of a Clinical Prediction Rule for Recurrent Clostridium difficile Infection , 15 December 2008

    Mary Y. Hu, Kianoosh Katchar, Lorraine Kyne, Seema Maroo, Sanjeev Tummala, Valley Dreisbach, Hua Xu, Daniel A. Leffler, Ciarán P. Kelly
    Gastroenterology April 2009 (Vol. 136, Issue 4, Pages 1206-1214)

Gastroenterology
Volume 136, Issue 4 , Pages 1152-1154, April 2009