- Email: [email protected]
Analysis of Clostridium difficile-associated diarrhea among patients hospitalized in tertiary care academic hospital
Diagnostic Microbiology and Infectious Disease 52 (2005) 153 – 155 www.elsevier.com/locate/diagmicrobio
Notes
Analysis of Clostridium difficile-associated diarrhea among patients hospitalized in tertiary care academic hospital Gayane Martirosiana,c,T, Adam Szcze˛snya,b, Stuart H. Cohenb, Joseph Silva Jr.b a Department of Medical Microbiology, Medical University of Silesia, 40-572 Katowice, Poland Division of Infectious and Immunologic Diseases, Department of Internal Medicine, University of California-Davis Medical Center, Sacramento, CA 95817, USA c Department of Histology and Embryology, Center of Biostructure, Research Warsaw Medical University, 02-0041 Warsaw, Poland Received 21 July 2004; accepted 31 December 2004
b
Abstract The frequency of Clostridium difficile strains in stool samples of patients with diarrhea hospitalized in the hematology/oncology, surgery, orthopedics, transplantology ward, and emergency room of Davis Medical Center was analyzed. A total of 786 stool samples collected from patients with diarrhea and 180 samples taken from the hospital environment were cultured for C. difficile by routine methods. There were 119 strains of C. difficile isolated: 97 (12.3%) strains from patients’ stools (no enteropathogen other than C. difficile was detected in these stool samples) and 22 (12.2%) strains from the hospital environment. It was confirmed that hospital environment plays an important role in transmission of C. difficile by AP-PCR and PCR ribotyping. Among 97 C. difficile strains isolated from patient’ stools 25 were nontoxigenic (A /B ), 67 were toxigenic (A+/B+), and 5 strains were toxin B-positive/toxin A-negative. Analysis of concomitant symptoms among hospitalized patients with diarrhea demonstrated significantly longer duration of diarrhea caused by nontoxigenic strains than in cases of diarrhea caused by toxigenic strains. On the other hand, among patients infected by toxigenic strains, significantly higher leukocytosis and longer duration of fever were observed. The resistance of isolated C. difficile strains to erythromycin and clindamycin indicated the possibility of transmission in the hospital strains with macrolide-lincosamide-streptogramin B resistance type. D 2005 Elsevier Inc. All rights reserved. Keywords: Clostridium difficile; Diarrhea; Toxins
Clostridium difficile-associated diarrhea (CDAD) has become a major clinical problem with the increased use of different antibiotics (clindamycin, penicillins, cephalosporins, and others) (Berild et al., 2003) and some chemotherapeutics (Anand and Glatt, 1993). Incidences of C. difficile-associated diseases were described even after treatment of patients with vancomycin (Szcze˛sny et al., 2002) or metronidazole drugs of choice for treatment of C. difficileassociated diseases. CDAD and colitis occur primarily in hospitalized patients in intensive care units, surgical wards, hematology/oncology units, particularly among those older than 60 years (Cohen et al., 1997). Although the most important risk factor is the use of antibiotics, other risk factors T Corresponding author. Department of Medical Microbiology, Medical University of Silesia, 40-572 Katowice, Poland. Tel.: +48-32-2526075; fax: +48-32-252-6075. E-mail addresses: [email protected], [email protected] (G. Martirosian). 0732-8893/$ – see front matter D 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.diagmicrobio.2004.12.015
such as long-term hospitalization, surgery, immunosuppression, chemotherapy, and radiotherapy also may be predisposing factors to C. difficile-associated diseases (Barbut and Petit, 2001). Toxins A and B are considered the major virulence factors in C. difficile. Genes for toxins A and B (tcdA and tcdB) are located on a large 19.6-kb chromosomal fragment called the pathogenicity locus, which is a prerequisite for virulence (Poxton et al., 2001). Toxigenic strains of C. difficile produce both toxins, whereas nontoxigenic strains lack both toxins and genes. Recently, mutant strains with only one toxin were described as a causative agents of hospitalacquired diarrhea (Cohen et al., 1998; Rupnik, 2001). In this communication, we analyzed the occurrence of CDAD among patients hospitalized in the hematology/ oncology (294), surgery (318), orthopedics (58), transplantology wards (60), and emergency room (56) of Davis Medical Center and the frequency of C. difficile strains in the environment of these hospital units (75, 62, 14, 15, and 14, respectively).
154
G. Martirosian et al. / Diagnostic Microbiology and Infectious Disease 52 (2005) 153 – 155
A total of 786 stool samples collected from 786 patients with diarrhea and 180 samples taken from the hospital environment were cultured for C. difficile on Columbiacycloserine-cefoxitin-amphotericin B agar between October 2001 and August 2002. Plates were incubated anaerobically at 37 8C for 48–72 h. Two to seven suspected colonies were isolated from each plate and identified by routine laboratory methods (Martirosian et al., 2000). Each strain was tested for susceptibility to erythromycin and clindamycin by E-test method (AB Biodisk, Sweden) (Ackermann et al., 2003). For detection toxins (A and B) of C. difficile strains Tox A/B ELISA test (TechLab, Blacksburg, VA), TCD toxin A test (Becton Dickinson, Franklin Lakes, NJ) and PCR for toxin A and B genes (primer pairs YT28–YT29 and YT17–YT19) were performed as previously described (Kuhl et al., 1993; Martirosian et al., 2003). AP-PCR and PCR ribotyping were performed to compare C. difficile strains isolated from stool samples and from hospital environment (Martirosian et al., 1995). Student t-test was used for statistical analysis of clinical data and Gel-Compare Program (Quantity One, Gel-doc, Word 97) was used for analysis of PCR results. There were 119 strains of C. difficile isolated: 97 (12.3%) strains from patients’ stools (no enteropathogen other than C. difficile was detected in these stools) and 22 (12.2%) strains from the hospital environment. In all studied hospital units, strains of C. difficile were isolated from environmental samples, except transplantology ward (no isolates were recovered). By using AP-PCR and PCR-ribotyping, it was confirmed that hospital environment plays an important role in transmission of C. difficile; similarity was confirmed in 59% of environmental and fecal strains (data not shown). Among 97 C. difficile strains isolated from patient’ stools, 25 were nontoxigenic (A /B ), 67 were toxigenic (A+/B+), and 5 strains were toxin B-positive/toxin A-negative. Analysis of concomitant symptoms among hospitalized patients with diarrhea demonstrated significantly longer duration of diarrhea caused by nontoxigenic strains than in cases of diarrhea caused by toxigenic strains. On the other hand, among patients infected with toxigenic strains, significantly higher leukocytosis and longer duration of fever was observed. Of 36 C. difficile-positive hematology/ oncology patients, 23 underwent previous surgery. Majority (22/36) of CDAD cases in hematology/oncology ward occurred after using of cytostatics (13 cases) and penicillins with h-lactamase inhibitors (9 cases). In addition, lymphoma/leukemia is usually diagnosed more frequently in young adults ( b40 years old) than other cancers, which are more common in old adults ( N60 years old) ( P b 0.05). Therefore, hematology/oncology patients, independent of their ages, are those at major risk for infection with C. difficile. Antibiotic susceptibility testing to erythromycin and clindamycin demonstrated a high degree of resistance (MIC N 256 Ag/mL) to both antibiotics in 9 of 13 nontoxigenic C. difficile strains isolated in the hematology/ oncology ward.
This study has some limitations. In stool samples of studied patients, no enteropathogens other than C. difficile were detected. However, stool samples were not tested for noroviruses. It is well known that noroviruses represent the most important cause of nonbacterial gastroenteritis worldwide. In industrialized countries, noroviruses may be responsible for up to 80% of all outbreaks of gastroenteritis. Outbreaks may affect all age groups and generally occur in crowded communities, including hospitals and nursing homes. The incubation time of Norovirus gastroenteritis is 24– 48 h, and symptoms include nausea, vomiting, diarrhea, abdominal pain, low-grade fever, headache, and myalgia. The duration time of Norovirus infection is generally 1–3 days. Our studied group of patients represents mainly mild diarrhea, with duration time longer than 4 days (depending on patient’ status and type of isolated C. difficile strain) and fever (from subfebrile to high). All C. difficile-positive patients were previously treated with antibiotics; they did not present upper gastrointestinal symptoms (nausea and vomiting) specific for Norovirus gastroenteritis. The absence of vomiting and other clinical findings among this group of patients are probably a good indication against Norovirus. However, to exclude Norovirus infection among hospitalized patients, Norovirus testing of stool samples is recommended in clinically motivated cases. Further studies with a larger patient populations are required to explain why nontoxigenic strains can cause diarrhea and which virulence factors other than toxins A and B play a role in pathogenesis of C. difficile-associated diseases.
References Ackermann G, Degner A, Cohen S, Silva Jr J, Rodloff AC (2003) Prevalence and association of macrolide-lincosamide-streptogramin B (MLSB) resistance with resistance of moxifloxacin in C. difficile. J Antimicrob Chemother 51:599 – 603. Anand A, Glatt AE (1993) C. difficile infection associated with antineoplastic chemotherapy: a review. Clin Infect Dis 17:109 – 113. Barbut F, Petit JC (2001) Epidemiology of C. difficile-associated infections. Clin Microbiol Infect 7:405 – 411. Berild D, Smaabrekke L, Halvorsen DS, Lelek M, Stahlsberg EM, Ringertz SH (2003) C. difficile infections related to antibiotic use and infection control facilities in two university hospitals. J Hosp Infect 54:202 – 206. Cohen SH, Tang YJ, Muenzer J, Gumerlock PH, Silva Jr J (1997) Isolation of various genotypes of C. difficile from patients and the environment in an oncology ward. Clin Infect Dis 24:889 – 893. Cohen SH, Tang Y, Hansen B, Silva J (1998) Isolation of a toxin Bdeficient mutant strain of C. difficile in a case of recurrent C. difficileassociated diarrhea. Clin Infect Dis 26:410 – 412. Kuhl SL, Tang YJ, Nawarro L, Gumerlock PH, Silva Jr J (1993) Diagnosis and monitoring of C. difficile infections with the polymerase chain reaction. Clin Infect Dis 16:234 – 238. Martirosian G, Kuipers S, van Belkum A, Verbrough H, MeiselMikolajczyk F (1995) PCR ribotyping and arbitrarily primed PCR for typing of C. difficile from Polish maternity hospital. J Clin Microbiol 33:2016 – 2021. Martirosian G, van Belkum A, Pituch H, Obuch-Woszczatyn´ski P, MeiselMikolajczyk F (2000) Are rapid immunoassays for in vivo detection of
G. Martirosian et al. / Diagnostic Microbiology and Infectious Disease 52 (2005) 153–155 toxin A sufficient for diagnostic purposes of antibiotic-associated diarrhea. Anaerobe 6:15 – 19. Martirosian G, Popielska J, Marczyn´ska M (2003) Occurrence of Clostridium difficile in fecal samples of HIV-infected children in Poland. Anaerobe 9:295 – 297. Poxton IR, McCoubrey J, Blair G (2001) The pathogenicity of C. difficile. Clin Microbiol Infect 7:421 – 427.
155
Rupnik M (2001) How to detect C. difficile variant strains in a routine laboratory. Clin Microbiol Infect 7:417 – 420. Szcze˛sny A, Kan´ski A, Martirosian G (2002) Incidence of pseudomembranous colitis after vancomycin-treated MRSA infection. Clin Microbiol Infect 8:58 – 59.
Notes
Analysis of Clostridium difficile-associated diarrhea among patients hospitalized in tertiary care academic hospital Gayane Martirosiana,c,T, Adam Szcze˛snya,b, Stuart H. Cohenb, Joseph Silva Jr.b a Department of Medical Microbiology, Medical University of Silesia, 40-572 Katowice, Poland Division of Infectious and Immunologic Diseases, Department of Internal Medicine, University of California-Davis Medical Center, Sacramento, CA 95817, USA c Department of Histology and Embryology, Center of Biostructure, Research Warsaw Medical University, 02-0041 Warsaw, Poland Received 21 July 2004; accepted 31 December 2004
b
Abstract The frequency of Clostridium difficile strains in stool samples of patients with diarrhea hospitalized in the hematology/oncology, surgery, orthopedics, transplantology ward, and emergency room of Davis Medical Center was analyzed. A total of 786 stool samples collected from patients with diarrhea and 180 samples taken from the hospital environment were cultured for C. difficile by routine methods. There were 119 strains of C. difficile isolated: 97 (12.3%) strains from patients’ stools (no enteropathogen other than C. difficile was detected in these stool samples) and 22 (12.2%) strains from the hospital environment. It was confirmed that hospital environment plays an important role in transmission of C. difficile by AP-PCR and PCR ribotyping. Among 97 C. difficile strains isolated from patient’ stools 25 were nontoxigenic (A /B ), 67 were toxigenic (A+/B+), and 5 strains were toxin B-positive/toxin A-negative. Analysis of concomitant symptoms among hospitalized patients with diarrhea demonstrated significantly longer duration of diarrhea caused by nontoxigenic strains than in cases of diarrhea caused by toxigenic strains. On the other hand, among patients infected by toxigenic strains, significantly higher leukocytosis and longer duration of fever were observed. The resistance of isolated C. difficile strains to erythromycin and clindamycin indicated the possibility of transmission in the hospital strains with macrolide-lincosamide-streptogramin B resistance type. D 2005 Elsevier Inc. All rights reserved. Keywords: Clostridium difficile; Diarrhea; Toxins
Clostridium difficile-associated diarrhea (CDAD) has become a major clinical problem with the increased use of different antibiotics (clindamycin, penicillins, cephalosporins, and others) (Berild et al., 2003) and some chemotherapeutics (Anand and Glatt, 1993). Incidences of C. difficile-associated diseases were described even after treatment of patients with vancomycin (Szcze˛sny et al., 2002) or metronidazole drugs of choice for treatment of C. difficileassociated diseases. CDAD and colitis occur primarily in hospitalized patients in intensive care units, surgical wards, hematology/oncology units, particularly among those older than 60 years (Cohen et al., 1997). Although the most important risk factor is the use of antibiotics, other risk factors T Corresponding author. Department of Medical Microbiology, Medical University of Silesia, 40-572 Katowice, Poland. Tel.: +48-32-2526075; fax: +48-32-252-6075. E-mail addresses: [email protected], [email protected] (G. Martirosian). 0732-8893/$ – see front matter D 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.diagmicrobio.2004.12.015
such as long-term hospitalization, surgery, immunosuppression, chemotherapy, and radiotherapy also may be predisposing factors to C. difficile-associated diseases (Barbut and Petit, 2001). Toxins A and B are considered the major virulence factors in C. difficile. Genes for toxins A and B (tcdA and tcdB) are located on a large 19.6-kb chromosomal fragment called the pathogenicity locus, which is a prerequisite for virulence (Poxton et al., 2001). Toxigenic strains of C. difficile produce both toxins, whereas nontoxigenic strains lack both toxins and genes. Recently, mutant strains with only one toxin were described as a causative agents of hospitalacquired diarrhea (Cohen et al., 1998; Rupnik, 2001). In this communication, we analyzed the occurrence of CDAD among patients hospitalized in the hematology/ oncology (294), surgery (318), orthopedics (58), transplantology wards (60), and emergency room (56) of Davis Medical Center and the frequency of C. difficile strains in the environment of these hospital units (75, 62, 14, 15, and 14, respectively).
154
G. Martirosian et al. / Diagnostic Microbiology and Infectious Disease 52 (2005) 153 – 155
A total of 786 stool samples collected from 786 patients with diarrhea and 180 samples taken from the hospital environment were cultured for C. difficile on Columbiacycloserine-cefoxitin-amphotericin B agar between October 2001 and August 2002. Plates were incubated anaerobically at 37 8C for 48–72 h. Two to seven suspected colonies were isolated from each plate and identified by routine laboratory methods (Martirosian et al., 2000). Each strain was tested for susceptibility to erythromycin and clindamycin by E-test method (AB Biodisk, Sweden) (Ackermann et al., 2003). For detection toxins (A and B) of C. difficile strains Tox A/B ELISA test (TechLab, Blacksburg, VA), TCD toxin A test (Becton Dickinson, Franklin Lakes, NJ) and PCR for toxin A and B genes (primer pairs YT28–YT29 and YT17–YT19) were performed as previously described (Kuhl et al., 1993; Martirosian et al., 2003). AP-PCR and PCR ribotyping were performed to compare C. difficile strains isolated from stool samples and from hospital environment (Martirosian et al., 1995). Student t-test was used for statistical analysis of clinical data and Gel-Compare Program (Quantity One, Gel-doc, Word 97) was used for analysis of PCR results. There were 119 strains of C. difficile isolated: 97 (12.3%) strains from patients’ stools (no enteropathogen other than C. difficile was detected in these stools) and 22 (12.2%) strains from the hospital environment. In all studied hospital units, strains of C. difficile were isolated from environmental samples, except transplantology ward (no isolates were recovered). By using AP-PCR and PCR-ribotyping, it was confirmed that hospital environment plays an important role in transmission of C. difficile; similarity was confirmed in 59% of environmental and fecal strains (data not shown). Among 97 C. difficile strains isolated from patient’ stools, 25 were nontoxigenic (A /B ), 67 were toxigenic (A+/B+), and 5 strains were toxin B-positive/toxin A-negative. Analysis of concomitant symptoms among hospitalized patients with diarrhea demonstrated significantly longer duration of diarrhea caused by nontoxigenic strains than in cases of diarrhea caused by toxigenic strains. On the other hand, among patients infected with toxigenic strains, significantly higher leukocytosis and longer duration of fever was observed. Of 36 C. difficile-positive hematology/ oncology patients, 23 underwent previous surgery. Majority (22/36) of CDAD cases in hematology/oncology ward occurred after using of cytostatics (13 cases) and penicillins with h-lactamase inhibitors (9 cases). In addition, lymphoma/leukemia is usually diagnosed more frequently in young adults ( b40 years old) than other cancers, which are more common in old adults ( N60 years old) ( P b 0.05). Therefore, hematology/oncology patients, independent of their ages, are those at major risk for infection with C. difficile. Antibiotic susceptibility testing to erythromycin and clindamycin demonstrated a high degree of resistance (MIC N 256 Ag/mL) to both antibiotics in 9 of 13 nontoxigenic C. difficile strains isolated in the hematology/ oncology ward.
This study has some limitations. In stool samples of studied patients, no enteropathogens other than C. difficile were detected. However, stool samples were not tested for noroviruses. It is well known that noroviruses represent the most important cause of nonbacterial gastroenteritis worldwide. In industrialized countries, noroviruses may be responsible for up to 80% of all outbreaks of gastroenteritis. Outbreaks may affect all age groups and generally occur in crowded communities, including hospitals and nursing homes. The incubation time of Norovirus gastroenteritis is 24– 48 h, and symptoms include nausea, vomiting, diarrhea, abdominal pain, low-grade fever, headache, and myalgia. The duration time of Norovirus infection is generally 1–3 days. Our studied group of patients represents mainly mild diarrhea, with duration time longer than 4 days (depending on patient’ status and type of isolated C. difficile strain) and fever (from subfebrile to high). All C. difficile-positive patients were previously treated with antibiotics; they did not present upper gastrointestinal symptoms (nausea and vomiting) specific for Norovirus gastroenteritis. The absence of vomiting and other clinical findings among this group of patients are probably a good indication against Norovirus. However, to exclude Norovirus infection among hospitalized patients, Norovirus testing of stool samples is recommended in clinically motivated cases. Further studies with a larger patient populations are required to explain why nontoxigenic strains can cause diarrhea and which virulence factors other than toxins A and B play a role in pathogenesis of C. difficile-associated diseases.
References Ackermann G, Degner A, Cohen S, Silva Jr J, Rodloff AC (2003) Prevalence and association of macrolide-lincosamide-streptogramin B (MLSB) resistance with resistance of moxifloxacin in C. difficile. J Antimicrob Chemother 51:599 – 603. Anand A, Glatt AE (1993) C. difficile infection associated with antineoplastic chemotherapy: a review. Clin Infect Dis 17:109 – 113. Barbut F, Petit JC (2001) Epidemiology of C. difficile-associated infections. Clin Microbiol Infect 7:405 – 411. Berild D, Smaabrekke L, Halvorsen DS, Lelek M, Stahlsberg EM, Ringertz SH (2003) C. difficile infections related to antibiotic use and infection control facilities in two university hospitals. J Hosp Infect 54:202 – 206. Cohen SH, Tang YJ, Muenzer J, Gumerlock PH, Silva Jr J (1997) Isolation of various genotypes of C. difficile from patients and the environment in an oncology ward. Clin Infect Dis 24:889 – 893. Cohen SH, Tang Y, Hansen B, Silva J (1998) Isolation of a toxin Bdeficient mutant strain of C. difficile in a case of recurrent C. difficileassociated diarrhea. Clin Infect Dis 26:410 – 412. Kuhl SL, Tang YJ, Nawarro L, Gumerlock PH, Silva Jr J (1993) Diagnosis and monitoring of C. difficile infections with the polymerase chain reaction. Clin Infect Dis 16:234 – 238. Martirosian G, Kuipers S, van Belkum A, Verbrough H, MeiselMikolajczyk F (1995) PCR ribotyping and arbitrarily primed PCR for typing of C. difficile from Polish maternity hospital. J Clin Microbiol 33:2016 – 2021. Martirosian G, van Belkum A, Pituch H, Obuch-Woszczatyn´ski P, MeiselMikolajczyk F (2000) Are rapid immunoassays for in vivo detection of
G. Martirosian et al. / Diagnostic Microbiology and Infectious Disease 52 (2005) 153–155 toxin A sufficient for diagnostic purposes of antibiotic-associated diarrhea. Anaerobe 6:15 – 19. Martirosian G, Popielska J, Marczyn´ska M (2003) Occurrence of Clostridium difficile in fecal samples of HIV-infected children in Poland. Anaerobe 9:295 – 297. Poxton IR, McCoubrey J, Blair G (2001) The pathogenicity of C. difficile. Clin Microbiol Infect 7:421 – 427.
155
Rupnik M (2001) How to detect C. difficile variant strains in a routine laboratory. Clin Microbiol Infect 7:417 – 420. Szcze˛sny A, Kan´ski A, Martirosian G (2002) Incidence of pseudomembranous colitis after vancomycin-treated MRSA infection. Clin Microbiol Infect 8:58 – 59.