Epidemic Clostridium difficile
http://www.100md.com
《新英格兰医药杂志》
To the Editor: The two articles by McDonald et al.1 and Loo et al.2 (Dec. 8 issue) document the spread of NAP1 Clostridium difficile isolates of toxinotype III in North America. This emerging strain is thought to be more virulent than earlier strains, perhaps related to its increased capacity to produce toxins.3 However, NAP1 is not solely responsible for the increasing prevalence and severity of C. difficile–associated disease. At our hospital, the rate of C. difficile–associated disease has greatly increased, and we have documented substantial treatment failure and mortality.4 Only 18 percent of isolates have been of the epidemic NAP1, toxinotype III strain; most were NAP2, toxinotype 0. Whereas McDonald et al. found that NAP1 caused at least 50 percent of C. difficile–associated disease at five hospitals, non-NAP1 strains predominated at three others.
Thus, the overall burden of C. difficile–associated disease may be attributable, in part, not only to the spread of new, potentially more virulent NAP1 isolates but also to other factors, such as the presence in hospitals of older, debilitated adults with a greater severity or range of underlying diseases than in the past and the increased use of wide-spectrum antibiotics to which C. difficile has developed resistance.
Daniel M. Musher, M.D.
Nancy Logan, M.A.
Michael E. DeBakey Veterans Affairs Medical Center
Houston, TX 77030
daniel.musher@med.va.gov
Vaibhav Mehendiratta, M.B., B.S.
Baylor College of Medicine
Houston, TX 77030
References
McDonald LC, Killgore GE, Thompson A, et al. An epidemic, toxin gene-variant strain of Clostridium difficile. N Engl J Med 2005;353:2433-2441.
Loo VG, Poirier L, Miller MA, et al. A predominantly clonal multi-institutional outbreak of Clostridium difficile-associated diarrhea with high morbidity and mortality. N Engl J Med 2005;353:2442-2449.
Warny M, Pepin J, Fang A, et al. Toxin production by an emerging strain of Clostridium difficile associated with outbreaks of severe disease in North America and Europe. Lancet 2005;366:1079-1084.
Musher DM, Aslam S, Logan N, et al. Relatively poor outcome after treatment of Clostridium difficile colitis with metronidazole. Clin Infect Dis 2005;40:1586-1590.
To the Editor: McDonald et al. and Loo et al. conclude that a more virulent strain of C. difficile was responsible for an outbreak of severe diarrhea in Quebec, Canada, in 2003 and in eight health care institutions in six U.S. states between 2000 and 2003. Available data suggest that the incidence of C. difficile–associated disease is increasing.1,2 We estimated the quarterly incidence of the disease in 100 to 119 hospitals from 1999 through the second quarter of 2005 by querying the code for the disease (00845, as listed in the International Classification of Diseases, 9th Revision, Clinical Modification) in the University HealthSystem Consortium database (www.uhc.edu), which contains discharge summaries and Uniform Billing Code of 1992 data. Median rates of C. difficile–associated disease for the four quarters of 1999 were 3.9 to 4.5 cases per 1000 admissions (maximum rate, 10 to 22 cases). During the first two quarters of 2005, those rates had increased significantly, to 7.9 to 8.4 cases per 1000 admissions (maximum rate, 46 to 58 cases) (Figure 1). Overall rates per 1000 admissions were significantly greater in the mid-Atlantic region (7.2), the Midwest (7.9), and New England (6.7) than they were in the Southeast (4.5), the West (4.9), and the Central Southwest (3.9). C. difficile–associated disease was significantly more common in smaller hospitals than in larger facilities. These data suggest that rates of the disease have doubled during the past six years in this sample of university teaching hospitals in the United States.
Figure 1. Incidence of Cases of C. difficile–Associated Disease.
The solid line connects the mean numbers of cases per 1000 population (P<0.001). The box plots show the median values and interquartile ranges, and the I bars represent the 10th and 90th percentiles.
Ronald E. Polk, Pharm.D.
Virginia Commonwealth University
Richmond, VA 23298
rpolk@hsc.vcu.edu
Micheal Oinonen, Pharm.D., M.P.H.
University HealthSystem Consortium
Oak Brook, IL 60523
Amy Pakyz, Pharm.D.
Virginia Commonwealth University
Richmond, VA 23298
References
Archibald LK, Banerjee SN, Jarvis WR. Secular trends in hospital-acquired Clostridium difficile disease in the United States, 1987-2001. J Infect Dis 2004;189:1585-1589.
McDonald LC, Owings M, Jernigan DB. Increasing rates of Clostridium difficile infection among patients discharged from US short-stay hospitals, 1999–2003. Emerg Infect Dis (in press). (Accessed February 23, 2006, at http://www.cdc.gov/ncidod/EID/vol12no03/05-1064.htm.)
To the Editor: Loo et al. concluded that resistance to fluoroquinolones may have selected for a hypervirulent strain of C. difficile. However, some aspects of antimicrobial practice were not considered. Patients with C. difficile–associated diarrhea received antibiotics more frequently than did controls and received 46 percent more antibiotics (mean number of antibiotics received, 1.9 vs. 1.3; P<0.001 for both comparisons). The study found that 20.7 percent of patients in the case–control study but only 2.8 percent of the whole cohort were not exposed to antibiotics within six weeks before the diagnosis of C. difficile–associated diarrhea; these rates are incompatible. Both antibiotic polypharmacy and the duration of treatment (which were not examined) affect the incidence of C. difficile–associated diarrhea and may clearly confound risk analyses for antimicrobial agents.1,2 Hospital formularies and policies limit antimicrobial use and thus also affect the risk of C. difficile–associated diarrhea.3 Without information on policies and practices (e.g., defined daily doses and attack rates), it is not possible to assess fully the contribution of specific antibiotics to this outbreak.
A substantial spread of the outbreak strain almost certainly reflected suboptimal infection-control practices. Indeed, a reduction in the incidence of C. difficile–associated diarrhea occurred after the implementation of "major infection-control measures." Appropriately controlled studies are required to determine the risk of C. difficile–associated diarrhea that is linked to the use of specific antibiotics.4
Mark H. Wilcox, M.D.
Jane Freeman, Ph.D.
Leeds Teaching Hospitals
Leeds LS1 3EX, United Kingdom
mark.wilcox@leedsth.nhs.uk
Dr. Wilcox reports having received consulting fees, financial support to attend meetings, and research funding from AstraZeneca, Bayer, Genzyme, Pfizer, and Wyeth.
References
Wistrom J, Norrby SR, Myhre EB, et al. Frequency of antibiotic-associated diarrhoea in 2462 antibiotic-treated hospitalized patients: a prospective study. J Antimicrob Chemother 2001;47:43-50.
Bignardi GE. Risk factors for Clostridium difficile infection. J Hosp Infect 1998;40:1-15.
Wilcox MH, Freeman J, Fawley W, et al. Long-term surveillance of cefotaxime and piperacillin-tazobactam prescribing and incidence of Clostridium difficile diarrhoea. J Antimicrob Chemother 2004;54:168-172.
Thomas C, Stevenson M, Riley TV. Antibiotics and hospital-acquired Clostridium difficile-associated diarrhoea: a systematic review. J Antimicrob Chemother 2003;51:1339-1350.
To the Editor: McDonald et al. and Loo et al. report increasing morbidity that has been linked to C. difficile–associated diarrhea. This apparent increase in complications caused by a spore-forming organism may partially be a result of the increasing use of alcohol-based hand hygiene. At our institution, an 875-bed hospital, 13 confirmed cases of bloodstream infection associated with bacillus species occurred in 2004. The hands of 60 ward nurses were randomly cultured, and bacillus species grew in 80 percent of the cultures, whereas few staphylococci or gram-negative organisms were identified.
The guidelines of the Centers for Disease Control and Prevention (CDC)1 regarding the use of alcohol-based hand hygiene should be revised to encourage hand washing with soap and water when organisms that are resistant to alcohol-based cleaners, such as C. difficile, are identified in the local environment.
Kentaro Iwata, M.D.
Asako Doi, M.D.
Naoko Furuya, R.N.
Kameda Medical Center
Kamogawa 296-8602, Japan
kentaro-iwata@kameda.jp
References
Boyce JM, Pittet D. Guideline for hand hygiene in health-care setting: recommendations of the Healthcare Infection Control Practices Advisory Committee and HICPAC/SHEA/APIC/IDSA Hand Hygiene Task Force. MMWR Recomm Rep 2002;51:1-45.
Drs. McDonald and Gerding reply: Dr. Musher and colleagues note their experience with a recent increase in cases of treatment failure and mortality linked to C. difficile and the association of the bacterium with predominance of a NAP2 strain over the NAP1 epidemic strain. We have found that the molecular pattern of NAP2 as characterized by pulsed-field gel electrophoresis correlates with the pattern of the J strain ( J7 and J9) as found on restriction-enzyme analysis.1 Although the J strain has been an important cause of outbreaks in U.S. hospitals, it has not been associated with increased disease severity. Thus, it would be important to determine whether it was the NAP1 strain or the NAP2 strain that was associated with more severe cases in the experience of Musher and colleagues. Nonetheless, we agree that recent increases in incidence and disease severity may be due to factors other than the emergence of NAP1.
Dr. Polk and colleagues found that the rates of hospitalization for C. difficile infection doubled between 1999 and 2005 in a group of approximately 100 university hospitals. This appears to be a slower increase than the estimated doubling of rates in all U.S. nonfederal hospitals between 2000 and 2003.2 However, considered together, these findings suggest rates of C. difficile–associated disease are continuing to increase and that the effect in smaller hospitals may be equal to or greater than that in larger hospitals. Higher rates of incidence in certain geographic regions may reflect the prevalence of more virulent strains, such as NAP1, in those regions. These results again emphasize the need for all health care providers to prescribe antimicrobial agents judiciously and to comply with infection-control measures.
One of the most important of these measures consists of hand hygiene; increasing the use of alcohol-based waterless hand sanitizers has been an important method of increasing hand-hygiene compliance among health care providers. The outbreak described by Iwata and colleagues is interesting in that it was caused by a bacillus species, which, like C. difficile, is capable of forming spores that resist the action of alcohol. However, we are not aware of increased outbreaks of infection caused by bacillus species, nor are we aware of data implicating the use of alcohol-based waterless hand sanitizers as the cause of C. difficile outbreaks. In contrast, data suggest that the use of alcohol-based waterless hand sanitizers is probably not important in recent increases in cases of C. difficile.3 Nonetheless, existing guidance indicates that in addition to wearing gloves,4 it is prudent for health care personnel to wash their hands with soap and water, rather than with alcohol-based waterless hand sanitizers, when caring for a patient with C. difficile–associated disease in an outbreak setting.
L. Clifford McDonald, M.D.
Centers for Disease Control and Prevention
Atlanta, GA 30333
cmcdonald1@cdc.gov
Dale N. Gerding, M.D.
Hines Veterans Affairs Hospital
Hines, IL 60141
References
Johnson S, Samore MH, Farrow KA, et al. Epidemics of diarrhea caused by a clindamycin-resistant strain of Clostridium difficile in four hospitals. N Engl J Med 1999;341:1645-1651.
McDonald LC, Owings M, Jernigan DB. Increasing rates of Clostridium difficile infection among patients discharged from U.S. short-stay hospitals, 1999–2003. Emerg Infect Dis (in press). (Accessed February 23, 2006, at http://www.cdc.gov/ncidod/EID/vol12no03/05-1064.htm.)
Ligi C, Kohan C, Dumigan DG, Boyce JM. Lack of association between the incidence of Clostridium difficile-associated disease (CDAD) and the use of alcohol-based hand rub. In: Proceedings of the 15th Annual Scientific Meeting of the Society for Healthcare Epidemiology of America, Los Angeles, April 9–12, 2005.
Johnson S, Gerding DN, Olson MM, et al. Prospective, controlled study of vinyl glove use to interrupt Clostridium difficile nosocomial transmission. Am J Med 1990;88:137-140.
Dr. Loo and colleagues reply: We agree with Drs. Wilcox and Freeman that other aspects of antimicrobial practice are needed to evaluate the specific risk associated with individual antibiotics. We evaluated the use of multiple antibiotics but examined neither the duration of antibiotic use nor the institutional use of antibiotics. These potential confounders should be incorporated into future studies. We have reported an erratum in Table 3 and in the corresponding text (which will appear in a future issue of the Journal) regarding the number of patients who were exposed to antibiotics in the case–control study. For the characteristic "any exposure to antibiotics," Table 3 should have read that 223 of case patients received antibiotics (94.1 percent), as compared with 159 of controls (67.1 percent).
There were multiple factors that contributed to the magnitude of this outbreak.1 However, despite the implementation of important infection-control interventions, average rates of nosocomial C. difficile–associated diarrhea in 10 of our 12 study hospitals were twice the baseline rate of 5 to 6 cases per 1000 admissions. We have two historical C. difficile isolates from 1987 that were of the same predominant pulsovar; however, they were susceptible to the newer fluoroquinolones. It is likely that the increased use of fluoroquinolones and the acquisition of fluoroquinolone resistance were critical events that contributed to the proliferation of this strain.
Dr. Musher and colleagues point out that strains other than NAP1 can cause increased severity of disease. They did not correlate severe disease with specific strains and did not report attributable mortality to C. difficile. In our study, we found three non-NAP1 strains that were associated with severe disease. We agree that severe morbidity and mortality related to C. difficile–associated diarrhea is correlated with the age of the patient, as was demonstrated in our study. However, among our 12 study hospitals, there were no differences in the age distribution of patients admitted during the outbreak period as compared with that of patients admitted in the two years preceding the outbreak.
Dr. Polk and colleagues note that rates of C. difficile–associated diarrhea are increasing in other areas of the United States. It would be interesting to analyze C. difficile isolates from these hospitals to determine whether there is a common predominant strain.
We agree with Dr. Iwata and colleagues that the CDC guidelines should be revised to emphasize hand washing with soap and water when health care workers are exposed to spore-forming bacteria, such as C. difficile, particularly in outbreak situations. These spores are known to be resistant to killing by alcohol.2 Further studies are required to fully delineate the role of alcohol-based hand-hygiene agents in C. difficile outbreaks.
Vivian G. Loo, M.D.
McGill University Health Centre
Montreal, QC H3G 1A4, Canada
vivian.loo@muhc.mcgill.ca
Michael D. Libman, M.D.
St. Mary's Hospital
Montreal, QC H3T 1M5, Canada
André Dascal, M.D.
Sir Mortimer B. Davis Jewish General Hospital
Montreal, QC H3T 1E2, Canada
Since publication of the article, Dr. Loo reports having received lecture fees from Sanofi-Aventis.
References
Loo VG, Libman MD, Miller MA, et al. Clostridium difficile: a formidable foe. CMAJ 2004;171:47-48.
Wullt M, Odenholt I, Walder M. Activity of three disinfectants and acidified nitrite against Clostridium difficile spores. Infect Control Hosp Epidemiol 2003;24:765-768.
Thus, the overall burden of C. difficile–associated disease may be attributable, in part, not only to the spread of new, potentially more virulent NAP1 isolates but also to other factors, such as the presence in hospitals of older, debilitated adults with a greater severity or range of underlying diseases than in the past and the increased use of wide-spectrum antibiotics to which C. difficile has developed resistance.
Daniel M. Musher, M.D.
Nancy Logan, M.A.
Michael E. DeBakey Veterans Affairs Medical Center
Houston, TX 77030
daniel.musher@med.va.gov
Vaibhav Mehendiratta, M.B., B.S.
Baylor College of Medicine
Houston, TX 77030
References
McDonald LC, Killgore GE, Thompson A, et al. An epidemic, toxin gene-variant strain of Clostridium difficile. N Engl J Med 2005;353:2433-2441.
Loo VG, Poirier L, Miller MA, et al. A predominantly clonal multi-institutional outbreak of Clostridium difficile-associated diarrhea with high morbidity and mortality. N Engl J Med 2005;353:2442-2449.
Warny M, Pepin J, Fang A, et al. Toxin production by an emerging strain of Clostridium difficile associated with outbreaks of severe disease in North America and Europe. Lancet 2005;366:1079-1084.
Musher DM, Aslam S, Logan N, et al. Relatively poor outcome after treatment of Clostridium difficile colitis with metronidazole. Clin Infect Dis 2005;40:1586-1590.
To the Editor: McDonald et al. and Loo et al. conclude that a more virulent strain of C. difficile was responsible for an outbreak of severe diarrhea in Quebec, Canada, in 2003 and in eight health care institutions in six U.S. states between 2000 and 2003. Available data suggest that the incidence of C. difficile–associated disease is increasing.1,2 We estimated the quarterly incidence of the disease in 100 to 119 hospitals from 1999 through the second quarter of 2005 by querying the code for the disease (00845, as listed in the International Classification of Diseases, 9th Revision, Clinical Modification) in the University HealthSystem Consortium database (www.uhc.edu), which contains discharge summaries and Uniform Billing Code of 1992 data. Median rates of C. difficile–associated disease for the four quarters of 1999 were 3.9 to 4.5 cases per 1000 admissions (maximum rate, 10 to 22 cases). During the first two quarters of 2005, those rates had increased significantly, to 7.9 to 8.4 cases per 1000 admissions (maximum rate, 46 to 58 cases) (Figure 1). Overall rates per 1000 admissions were significantly greater in the mid-Atlantic region (7.2), the Midwest (7.9), and New England (6.7) than they were in the Southeast (4.5), the West (4.9), and the Central Southwest (3.9). C. difficile–associated disease was significantly more common in smaller hospitals than in larger facilities. These data suggest that rates of the disease have doubled during the past six years in this sample of university teaching hospitals in the United States.
Figure 1. Incidence of Cases of C. difficile–Associated Disease.
The solid line connects the mean numbers of cases per 1000 population (P<0.001). The box plots show the median values and interquartile ranges, and the I bars represent the 10th and 90th percentiles.
Ronald E. Polk, Pharm.D.
Virginia Commonwealth University
Richmond, VA 23298
rpolk@hsc.vcu.edu
Micheal Oinonen, Pharm.D., M.P.H.
University HealthSystem Consortium
Oak Brook, IL 60523
Amy Pakyz, Pharm.D.
Virginia Commonwealth University
Richmond, VA 23298
References
Archibald LK, Banerjee SN, Jarvis WR. Secular trends in hospital-acquired Clostridium difficile disease in the United States, 1987-2001. J Infect Dis 2004;189:1585-1589.
McDonald LC, Owings M, Jernigan DB. Increasing rates of Clostridium difficile infection among patients discharged from US short-stay hospitals, 1999–2003. Emerg Infect Dis (in press). (Accessed February 23, 2006, at http://www.cdc.gov/ncidod/EID/vol12no03/05-1064.htm.)
To the Editor: Loo et al. concluded that resistance to fluoroquinolones may have selected for a hypervirulent strain of C. difficile. However, some aspects of antimicrobial practice were not considered. Patients with C. difficile–associated diarrhea received antibiotics more frequently than did controls and received 46 percent more antibiotics (mean number of antibiotics received, 1.9 vs. 1.3; P<0.001 for both comparisons). The study found that 20.7 percent of patients in the case–control study but only 2.8 percent of the whole cohort were not exposed to antibiotics within six weeks before the diagnosis of C. difficile–associated diarrhea; these rates are incompatible. Both antibiotic polypharmacy and the duration of treatment (which were not examined) affect the incidence of C. difficile–associated diarrhea and may clearly confound risk analyses for antimicrobial agents.1,2 Hospital formularies and policies limit antimicrobial use and thus also affect the risk of C. difficile–associated diarrhea.3 Without information on policies and practices (e.g., defined daily doses and attack rates), it is not possible to assess fully the contribution of specific antibiotics to this outbreak.
A substantial spread of the outbreak strain almost certainly reflected suboptimal infection-control practices. Indeed, a reduction in the incidence of C. difficile–associated diarrhea occurred after the implementation of "major infection-control measures." Appropriately controlled studies are required to determine the risk of C. difficile–associated diarrhea that is linked to the use of specific antibiotics.4
Mark H. Wilcox, M.D.
Jane Freeman, Ph.D.
Leeds Teaching Hospitals
Leeds LS1 3EX, United Kingdom
mark.wilcox@leedsth.nhs.uk
Dr. Wilcox reports having received consulting fees, financial support to attend meetings, and research funding from AstraZeneca, Bayer, Genzyme, Pfizer, and Wyeth.
References
Wistrom J, Norrby SR, Myhre EB, et al. Frequency of antibiotic-associated diarrhoea in 2462 antibiotic-treated hospitalized patients: a prospective study. J Antimicrob Chemother 2001;47:43-50.
Bignardi GE. Risk factors for Clostridium difficile infection. J Hosp Infect 1998;40:1-15.
Wilcox MH, Freeman J, Fawley W, et al. Long-term surveillance of cefotaxime and piperacillin-tazobactam prescribing and incidence of Clostridium difficile diarrhoea. J Antimicrob Chemother 2004;54:168-172.
Thomas C, Stevenson M, Riley TV. Antibiotics and hospital-acquired Clostridium difficile-associated diarrhoea: a systematic review. J Antimicrob Chemother 2003;51:1339-1350.
To the Editor: McDonald et al. and Loo et al. report increasing morbidity that has been linked to C. difficile–associated diarrhea. This apparent increase in complications caused by a spore-forming organism may partially be a result of the increasing use of alcohol-based hand hygiene. At our institution, an 875-bed hospital, 13 confirmed cases of bloodstream infection associated with bacillus species occurred in 2004. The hands of 60 ward nurses were randomly cultured, and bacillus species grew in 80 percent of the cultures, whereas few staphylococci or gram-negative organisms were identified.
The guidelines of the Centers for Disease Control and Prevention (CDC)1 regarding the use of alcohol-based hand hygiene should be revised to encourage hand washing with soap and water when organisms that are resistant to alcohol-based cleaners, such as C. difficile, are identified in the local environment.
Kentaro Iwata, M.D.
Asako Doi, M.D.
Naoko Furuya, R.N.
Kameda Medical Center
Kamogawa 296-8602, Japan
kentaro-iwata@kameda.jp
References
Boyce JM, Pittet D. Guideline for hand hygiene in health-care setting: recommendations of the Healthcare Infection Control Practices Advisory Committee and HICPAC/SHEA/APIC/IDSA Hand Hygiene Task Force. MMWR Recomm Rep 2002;51:1-45.
Drs. McDonald and Gerding reply: Dr. Musher and colleagues note their experience with a recent increase in cases of treatment failure and mortality linked to C. difficile and the association of the bacterium with predominance of a NAP2 strain over the NAP1 epidemic strain. We have found that the molecular pattern of NAP2 as characterized by pulsed-field gel electrophoresis correlates with the pattern of the J strain ( J7 and J9) as found on restriction-enzyme analysis.1 Although the J strain has been an important cause of outbreaks in U.S. hospitals, it has not been associated with increased disease severity. Thus, it would be important to determine whether it was the NAP1 strain or the NAP2 strain that was associated with more severe cases in the experience of Musher and colleagues. Nonetheless, we agree that recent increases in incidence and disease severity may be due to factors other than the emergence of NAP1.
Dr. Polk and colleagues found that the rates of hospitalization for C. difficile infection doubled between 1999 and 2005 in a group of approximately 100 university hospitals. This appears to be a slower increase than the estimated doubling of rates in all U.S. nonfederal hospitals between 2000 and 2003.2 However, considered together, these findings suggest rates of C. difficile–associated disease are continuing to increase and that the effect in smaller hospitals may be equal to or greater than that in larger hospitals. Higher rates of incidence in certain geographic regions may reflect the prevalence of more virulent strains, such as NAP1, in those regions. These results again emphasize the need for all health care providers to prescribe antimicrobial agents judiciously and to comply with infection-control measures.
One of the most important of these measures consists of hand hygiene; increasing the use of alcohol-based waterless hand sanitizers has been an important method of increasing hand-hygiene compliance among health care providers. The outbreak described by Iwata and colleagues is interesting in that it was caused by a bacillus species, which, like C. difficile, is capable of forming spores that resist the action of alcohol. However, we are not aware of increased outbreaks of infection caused by bacillus species, nor are we aware of data implicating the use of alcohol-based waterless hand sanitizers as the cause of C. difficile outbreaks. In contrast, data suggest that the use of alcohol-based waterless hand sanitizers is probably not important in recent increases in cases of C. difficile.3 Nonetheless, existing guidance indicates that in addition to wearing gloves,4 it is prudent for health care personnel to wash their hands with soap and water, rather than with alcohol-based waterless hand sanitizers, when caring for a patient with C. difficile–associated disease in an outbreak setting.
L. Clifford McDonald, M.D.
Centers for Disease Control and Prevention
Atlanta, GA 30333
cmcdonald1@cdc.gov
Dale N. Gerding, M.D.
Hines Veterans Affairs Hospital
Hines, IL 60141
References
Johnson S, Samore MH, Farrow KA, et al. Epidemics of diarrhea caused by a clindamycin-resistant strain of Clostridium difficile in four hospitals. N Engl J Med 1999;341:1645-1651.
McDonald LC, Owings M, Jernigan DB. Increasing rates of Clostridium difficile infection among patients discharged from U.S. short-stay hospitals, 1999–2003. Emerg Infect Dis (in press). (Accessed February 23, 2006, at http://www.cdc.gov/ncidod/EID/vol12no03/05-1064.htm.)
Ligi C, Kohan C, Dumigan DG, Boyce JM. Lack of association between the incidence of Clostridium difficile-associated disease (CDAD) and the use of alcohol-based hand rub. In: Proceedings of the 15th Annual Scientific Meeting of the Society for Healthcare Epidemiology of America, Los Angeles, April 9–12, 2005.
Johnson S, Gerding DN, Olson MM, et al. Prospective, controlled study of vinyl glove use to interrupt Clostridium difficile nosocomial transmission. Am J Med 1990;88:137-140.
Dr. Loo and colleagues reply: We agree with Drs. Wilcox and Freeman that other aspects of antimicrobial practice are needed to evaluate the specific risk associated with individual antibiotics. We evaluated the use of multiple antibiotics but examined neither the duration of antibiotic use nor the institutional use of antibiotics. These potential confounders should be incorporated into future studies. We have reported an erratum in Table 3 and in the corresponding text (which will appear in a future issue of the Journal) regarding the number of patients who were exposed to antibiotics in the case–control study. For the characteristic "any exposure to antibiotics," Table 3 should have read that 223 of case patients received antibiotics (94.1 percent), as compared with 159 of controls (67.1 percent).
There were multiple factors that contributed to the magnitude of this outbreak.1 However, despite the implementation of important infection-control interventions, average rates of nosocomial C. difficile–associated diarrhea in 10 of our 12 study hospitals were twice the baseline rate of 5 to 6 cases per 1000 admissions. We have two historical C. difficile isolates from 1987 that were of the same predominant pulsovar; however, they were susceptible to the newer fluoroquinolones. It is likely that the increased use of fluoroquinolones and the acquisition of fluoroquinolone resistance were critical events that contributed to the proliferation of this strain.
Dr. Musher and colleagues point out that strains other than NAP1 can cause increased severity of disease. They did not correlate severe disease with specific strains and did not report attributable mortality to C. difficile. In our study, we found three non-NAP1 strains that were associated with severe disease. We agree that severe morbidity and mortality related to C. difficile–associated diarrhea is correlated with the age of the patient, as was demonstrated in our study. However, among our 12 study hospitals, there were no differences in the age distribution of patients admitted during the outbreak period as compared with that of patients admitted in the two years preceding the outbreak.
Dr. Polk and colleagues note that rates of C. difficile–associated diarrhea are increasing in other areas of the United States. It would be interesting to analyze C. difficile isolates from these hospitals to determine whether there is a common predominant strain.
We agree with Dr. Iwata and colleagues that the CDC guidelines should be revised to emphasize hand washing with soap and water when health care workers are exposed to spore-forming bacteria, such as C. difficile, particularly in outbreak situations. These spores are known to be resistant to killing by alcohol.2 Further studies are required to fully delineate the role of alcohol-based hand-hygiene agents in C. difficile outbreaks.
Vivian G. Loo, M.D.
McGill University Health Centre
Montreal, QC H3G 1A4, Canada
vivian.loo@muhc.mcgill.ca
Michael D. Libman, M.D.
St. Mary's Hospital
Montreal, QC H3T 1M5, Canada
André Dascal, M.D.
Sir Mortimer B. Davis Jewish General Hospital
Montreal, QC H3T 1E2, Canada
Since publication of the article, Dr. Loo reports having received lecture fees from Sanofi-Aventis.
References
Loo VG, Libman MD, Miller MA, et al. Clostridium difficile: a formidable foe. CMAJ 2004;171:47-48.
Wullt M, Odenholt I, Walder M. Activity of three disinfectants and acidified nitrite against Clostridium difficile spores. Infect Control Hosp Epidemiol 2003;24:765-768.