You are here
Community-acquired Clostridium difficile infection in children: A retrospective study
Digestive and Liver Disease, Available online 9 June 2015 In Press, Corrected Proof
Community acquired-Clostridium difficile infection (CDI) has increased also in children in the last years.
To determine the incidence of community-acquired CDI and to understand whether Clostridium difficile could be considered a symptom-triggering pathogen in infants.
A five-year retrospective analysis (January 2007–December 2011) of faecal specimens from 124 children hospitalized in the Niguarda Ca’ Granda Hospital for prolonged or muco-haemorrhagic diarrhoea was carried out. Stool samples were evaluated for common infective causes of diarrhoea and for Clostridium difficile toxins. Patients with and without CDI were compared for clinical characteristics and known risk factors for infection.
Twenty-two children with CDI were identified in 5 years. An increased incidence of community-acquired CDI was observed, ranging from 0.75 per 1000 hospitalizations in 2007 to 9.8 per 1000 hospitalizations in 2011. Antimicrobial treatment was successful in all 19 children in whom it was administered; 8/22 CDI-positive children were younger than 2 years. No statistically significant differences in clinical presentation were observed between patients with and without CDI, nor in patients with and without risk factors for CDI.
Our study shows that Clostridium difficile infection is increasing and suggests a possible pathogenic role in the first 2 years of life.
Keywords: Clostridium difficile, Diarrhoea, Infectious disease.
Clostridium difficile (C. difficile) is a spore-forming, anaerobic gram-positive bacillus, and is the most common cause of healthcare-associated diarrhoea in the United States, with significant associated morbidity, mortality and healthcare costs  . Challenging issues have recently arisen in connection with C. difficile infection (CDI) with a substantial increase in incidence, severity, resistance to standard therapy and a propensity to relapse in adulthood  . These rates have all been increasing since 2000 worldwide, especially among the recently hospitalized elderly or long-term care facility residents. Even though the majority of CDI cases are healthcare-associated in patients older than 60 years of age, the occurrence of the disease outside a healthcare facility has received a great deal of attention as a potential emerging cause of outpatient diarrheal illness  and . Moreover, a recent advisory from the Centres for Disease Control and Prevention warns of a risk of CDI in populations previously not considered at-risk  , such as young and previously healthy persons unexposed to a hospital or healthcare environment or antimicrobial therapy. Children are not usually considered a high-risk population for CDI and the role of this pathogen in infants has been debated. The colonization rate of C. difficile, especially in newborns and infants younger than 2 years varies widely (2.5–90%) , , and .
The presence of C. difficile in stools of infants is currently considered to be evidence of colonization  , and for this reason, a recent policy statement of The American Academy of Paediatrics recommended that testing of infants should be limited to those with risk factors, as Hirschsprung's disease or other severe motility disorders, or in an outbreak situation  .
This hypothesis is based on the relatively low pathogen load in the infant gut, the preferential colonization of non-toxigenic or less pathogenic C. difficile strains in infancy, the relative absence of toxin receptors or downstream signalling pathways in the immature gut mucosa, and the presence of protective factors in breast milk and in the neonatal gut flora  .
Nevertheless, cases of necrotizing enterocolitis, prolonged diarrhoea and pseudomembranous colitis associated with CDI in neonates and infants have been reported  .
One large study involving 22 US hospitals showed a steady increase from 2.6 to 4 cases per 1000 admissions in the annual incidence of CDI among paediatric inpatients over a 5-year period to 2006  . Similar results have been reported by other large multicentre cohorts , , and . Moreover, there is evidence to suggest that a large proportion of paediatric CDI are community-acquired infections and that many of these cases lack the typical risk pattern associated with exposure to antimicrobial drugs  . These changes have partly been attributed to the emergence of hyper-virulent C. difficile strains, such as the North American pulse-field gel electrophoresis type I (NAP1)-strain , , , and . However, even if CDI in children seems to be less frequent and less severe than in adults  , the hypervirulent NAP1 strain has also been identified in children  .
The clinical presentation of CDI varies widely, ranging from self-limited diarrhoea to toxic megacolon which can lead to death , , and . Colonic injury and inflammation mainly result from the production of two protein toxins (A and B) that are known to be the primary virulence factors of C. difficile, whereas toxin-negative strains are non-pathogenic  . Although the majority of pathogenic strains both produce A and B toxins, clinically relevant toxin A-negative, toxin B-positive strains have been described  . TcdC deletion-carrying Clostridium difficile strains or strains producing a binary toxin are more pathogenic.
Approximately 3% of C. difficile-colonized adults are asymptomatic  , while the culture-positive rate among hospitalized patients is about 20%  , most (85%) presenting as symptom-free faecal excretors  . In comparison, the rate of C. difficile positivity among newborns varies widely (2.5–90%) , , and  while pathogenicity in infants is still a matter of debate.
The present study aims were to determine the incidence of community-acquired (CA)-CDI in children admitted to the paediatric division of an Italian general hospital for prolonged or muco-haemorrhagic diarrhoea and describe the clinical pattern of the infection in early childhood.
2. Materials and methods
We carried out a retrospective analysis of faecal samples from 124 children (74 males, mean age 7.3 years; age range 1 month–18 years) hospitalized in the Niguarda Ca’ Granda Hospital in Milan, Italy from January 2007 to December 2011 using clinical records. During this 5-year period, the Niguarda Hospital did not increase its catchment area and the demographic parameters of the population within this remit remained substantially identical.
Patients were identified through a computerized review of records of microbiological findings. Subsequently, the clinical records of all positive patients for C. difficile toxins were retrieved and reviewed. Inclusion criteria were the presence of prolonged diarrhoea (≥3 liquid stools within 24 hours and for more than 7 days) associated with weight loss and/or fever, or muco-haemorrhagic diarrhoea, and age ranging from 1 month to 18 years. Only children with a first episode of CA-CDI were included. No hospital-associated C. difficile infection was recorded over this 5-year-period.
CA-CDI was defined as disease onset occurring in the community or within 48 hours of hospitalization in children without previous hospitalization or symptom onset >12 weeks following discharge from a hospital facility  and .
To evaluate the role of potential risk factors for CA-CDI, comorbidity data were evaluated, including pre-term birth (gestational age <37 weeks), previous gastrointestinal surgery, solid organ and stem cell transplantation, malignancies, chemotherapy, use of proton pump inhibitors, immunodeficiency, renal failure (peritoneal dialysis or haemodialysis), cystic fibrosis, inflammatory bowel disease (IBD) or Hirschsprung's disease, prior hospitalization, G- or J- tube feedings, and exposure to oral or parenteral antibiotics within the 4 previous weeks  and .
Other causes of prolonged diarrhoea were excluded by appropriate clinical evaluation and laboratory tests when indicated. Stool samples were analyzed for common infectious causes of diarrhoea (Rotavirus, Adenovirus, Salmonella, Shigella, Yersinia enterocolitica, enterohaemorragic Escherichia coli, Campylobacter sp.) and for C. difficile toxins. The laboratory diagnosis of CDI was made following a 2-step protocol, based on a laboratory screening test for GDH as described by Fenner et al.  , coupled with a confirmatory test for toxigenic C. difficile using real-time polymerase chain reaction (rPCR) for toxin B, instead of the rapid toxin A/B assay.
Briefly, our protocol uses a common-antigen test, the enzyme immunoassay (EIA) or glutamase dehydrogenase (GDH), as the first step. C. difficile constitutively produces GDH in easily detectable levels, so tests based on GDH detection have an excellent sensitivity, close to 100%, even if this positivity only indicates the presence of the organism. Successively, to confirm the toxigenicity of the strain, GDH-positive specimens were analyzed by rPCR for the toxin-B gene (tcdB) as well as for tcdC mutation and binary toxin genes to confirm hypervirulence of the strain  . We chose to use rPCR for toxin B, and not the rapid toxin A/B assay as originally suggested by Fenner, because studies comparing most commercially available tests for C. difficile detection (toxin detection assays, real-time PCR for C. difficile tcdB, GDH detection assay for cytotoxin testing and cytotoxigenic culture) have found that the best and most rapid single test was PCR for the toxin B gene, as this had the highest negative predictive value , , and .
To measure CA-CDI incidence in children, we compared data from cases/1000 hospitalizations between 2007 and 2011 at our institution.
Clinical symptoms were compared among CDI-positive and CDI-negative children. Moreover, to correlate the presence of risk factors for CDI with clinical features, symptoms were compared according to the presence or not of risk factors for the infection in C. difficile-positive children.
Categorical data were analyzed using Fisher's exact test. Statistical significance was set at P < 0.05. Graph pad statistical software was used for all statistical analyses.
The study was approved by our institutional Review Board.
A total of 124 patients with prolonged or muco-haemorrhagic diarrhoea were enrolled. In 25 patients, a common cause of infection (Rotavirus in 15, Salmonella species in 8 and Campylobacter jejuni in 2) had been detected and successfully treated using rehydration and probiotics. Twelve patients had developed an exacerbation or a new diagnosis of IBD requiring appropriate therapy, and 1 had presented with a colonic polyp which had been surgically removed.
Sixty-four children had a self-limiting episode of diarrhoea without a common conventional viral or bacterial aetiology.
Twenty-two of the 124 patients (17.7%) had C. difficile toxins in their stools according to rPCR for tcdB; no CDI-positive patient had a positive stool test for tcdC mutation and/or binary toxin genes. This suggests that, in our patient population, toxigenic C. difficile-strains did not belong to hypervirulent strains.
An increase in CA-CDI incidence was observed in the 5 year period, ranging from 0.75 per 1000 hospitalizations in 2007 to 9.8 per 1000 hospitalizations in 2011 ( Fig. 1 ). No case of nosocomial C. difficile infection was recorded during this time.
No statistically significant differences in age or gender were observed: 14/22 CDI-positive children (63.6%) and 74/102 CDI-negative children (72.5%) were older than 2 years of age; 14/22 CDI-positive patients (63.6%) and 60/102 CDI-negative patients (58.8%) were male.
Regarding clinical presentation, diarrhoea, fever and abdominal pain were the most common symptoms in both CDI-positive and CDI-negative groups, followed by muco-haemorrhagic stools and vomiting. No significant differences in clinical symptoms were present between groups ( Table 1 ). Among specific risk factors for CDI, 13 CDI-positive patients (59%) showed at least one: IBD in 4 (18.1%), chemotherapy in 4 (18.1%), previous antibiotic therapy in 3 (13.6%), post-transplantation immune suppressive therapy in 1 (4.5%) and Hirschsprung's disease in 1 (4.5%). Nine patients (40.9%) had no risk factor for CDI. A comparison of symptoms in CDI-positive patients according to the presence or absence of risk factors for CDI showed no significant differences ( Table 2 ).
|Diarrhoea||22 (100%)||102 (100%)||1.0|
|Fever||12 (54.5%)||55 (53.9%)||0.9|
|Abdominal pain||7 (31.8%)||36 (35.2%)||0.8|
|Blood in stools||7 (31.8%)||35 (34.3%)||0.9|
|Vomiting||6 (27.2%)||33 (32.3%)||0.8|
|Weight loss||3 (13.6%)||15 (14.7%)||0.9|
|Mild dehydration||1 (4.5%)||5 (4.9%)||0.9|
a Some patients had more than one symptom.
CDI, Clostridium difficile infection.
|Symptom a||Risk factors present||Risk factors absent||P-value|
|Diarrhoea||13 (100%)||9 (100%)||1.0|
|Fever||8 (61.5%)||4 (44.4%)||0.6|
|Abdominal pain||4 (30.7%)||3 (33.3%)||0.9|
|Blood in stools||3 (23.0%)||4 (44.4%)||0.3|
|Vomiting||4 (30.7%)||2 (22.2%)||0.9|
|Weight loss||2 (15.3%)||1 (11.1%)||0.9|
|Mild dehydration||0 (0%)||1 (11.1%)||0.4|
a Some patients had more than one symptom.
Treatment of CA-CDI had been started according to published guidelines  and . Briefly, a 10-day course of antimicrobial treatment (oral metronidazole or vancomycin) had been administered to CDI-positive patients (with and without risk factors) with muco-haemorrhagic or unresolved prolonged diarrhoea (from which other infective or non- infective causes of diarrhoea were excluded) or to patients with severe CDI. A single course of antimicrobial medication had been successful 18/22 (81.8%) children with CDI. No difference in efficacy between the two treatment options (metronidazole or vancomycin) was detected.
Only one child with an acute lymphoblastic leukaemia undergoing chemotherapy had developed small bowel sub-obstruction and had severe CDI requiring parenteral nutrition and vancomycin in combination with metronidazole as therapy. Three patients with CDI had a recurrence and for this reason had been successfully treated with the same medication in accordance with current treatment guidelines  .
Among patients with CDI, 8/22 children (36.3%) had been under 2 years of age and 25% had risk factors at history-taking.
Five CDI-positive children under 2 years of age (62.5%) required an antimicrobial course because of the persistence of diarrhoea, weight loss and/or fever for more than 10 days, suggesting that C. difficile could pathogenic in this age group as well.
Three patients younger than 2 years of age, with positive C. difficile toxins in their stools, did not require antimicrobial medication (2 because of spontaneous resolution of symptoms and one because a diagnosis of cow's milk-induced colitis had been suspected and successfully treated with an amino-acid formula).
The aim of our study was to investigate the incidence and the role of C. difficile in children with community-acquired prolonged diarrhoea; moreover, we investigated whether C. difficile could be considered a symptom-triggering pathogen in infants, among whom C. difficile colonization is considered to be a common finding, albeit of uncertain and controversial pathological relevance.
However, some researchers emphasize that waiting until an episode of diarrhoea reaches 14 days in duration to intensify care is ill-advised and that closer attention should be paid to patients whose acute diarrhoea is ‘prolonged’, that is, 5–7 days in duration and not yet resolved. For these reasons, we used the definition of a prolonged episode of acute diarrhoea to refer to an episode of diarrhoea lasting ≥7 days and <14 days in duration, with or without the presence of mucus or gross blood in the stools  .
In this 5-year retrospective study, we found a remarkable increase in the incidence of community-acquired CDI among children presenting with protracted diarrhoea. Within the study period, the ratio of C. difficile cases/hospitalizations increased from 0.75/1000 in 2007 to 9.8/1000 in 2011.
Well-known risk factors for CDI (especially the administration of broad-spectrum antibiotics or immunosuppressors/chemotherapy) were present in approximately 60% of patients in this series. This implies that among the remaining patients CDI had occurred in children who did not show any established risk factor for CDI. Clinical features were similar in both C. difficile-positive and C. difficile-negative children and no clinical difference between children with or without risk factors specific for CDI had been demonstrated.
The overall clinical feature of children with CA-CDI in this study appears to be less severe than in the hospital-associated disease, as shown by other studies  . Even if the negative search for tcdC mutation and binary toxin genes in our series suggests that the lack of severity could be due to the absence of hypervirulent C. difficile strains, there is not sufficient data to reach this conclusion.
Presently disease outcomes and severity of CDI in children are still poorly understood compared to adults, and the more aggressive clinical course seen in some cases of paediatric CDI can possibly be explained by underlying co-morbidities. Furthermore, the epidemiology of paediatric CDI remains largely uncharted and there are currently no established predictors of CDI-related disease severity in paediatric patients. In particular, it has been recently demonstrated that a severe clinical outcome is uncommon in CDI in children in contrast to adults, regardless of the C. difficile strain encountered  . Clinical practice guidelines for CDI diagnosis recommend that only stools from patients with diarrhoea should be tested for C. difficile and that testing once eradication has been achieved is unnecessary  .
As demonstrated in our study and in accordance with the literature, the detection of C. difficile toxin in stools is not evidence of its causative role in children with diarrhoea, particularly in infants. For this reason faecal examination for C. difficile in infants should be limited to selected cases  .
However, our data suggest that the heaviest burden of paediatric CDI is community-acquired and that CA-CDI is, on the whole, increasing, thereby supporting the research hypothesis that prolonged and/or muco-haemorrhagic diarrhoea is independent of the presence of risk factors in all children. Exposure to antibiotics has been confirmed to be the most important modifiable risk factor for CDI in children and any effort that focus on antibiotic stewardship in paediatric outpatient health care settings are needed  .
The limitations of our study include the use of retrospective data, the single-centre analysis, and the possibility of heightened referral and disease awareness biases. Another limitation of our study is the decision to confirm the laboratory diagnosis of CDI using the single toxin B test as a confirmatory test for toxigenic C. difficile, without measuring toxin A. It has been suggested that toxin B assay failed to detect up to 34.9% of CDI  not only because of different C. difficile strains, but also due to physiological fluctuations of C. difficile toxin production during the course of disease. For these reasons, the Committee on Infectious Disease of American Academy of Paediatrics suggests to check for both toxins  .
However, we chose to use a 2-step protocol, with a confirmatory test for toxigenic C. difficile using the most sensitive rPCR instead of EIA for toxin B to partially overcome this limitation.
Further prospective paediatric studies are needed to explore the association between the presence of genetically identified C. difficile strains, symptomatic infection and co-morbidities or other predictors of disease severity specific to the paediatric age.
Similarly to recently published data, over 30% of C. difficile-positive children were infants and most had required antimicrobial treatment to clear their infection  suggesting that C. difficile-positive specimens in these cases likely represent true infection; therefore prospective and age-matched case-control studies are needed to improve our understanding of the pathogenic role of the microorganism in the first 2 years of life.
Conflict of interest
The contributions of Giuseppe Scibilia MD, Anna Marcuzzi BS and Gabriel R Bougyue MSc are gratefully acknowledged.
-  E.R. Dubberke, D.N. Gerding, D. Classen, et al. Strategies to prevent Clostridium difficile infection in acute care hospital. Infection Control and Hospital Epidemiology. 2008;29:S81-S89
-  J.G. Bartlett. Clostridium difficile: progress and challenge. Annals of the New York Academy of Sciences. 2010;1213:62-69
-  M.D. Cairns, R.A. Stabler, B. Wren. The continually evolving Clostridium difficile species. Future Microbiology. 2012;7:945-957
-  L. Benson, X. Song, J. Campos, et al. Changing epidemiology of Clostridium difficile-associated disease in children. Infection Control and Hospital Epidemiology. 2007;28:1233-1235
-  Centers for Disease Control Prevention (CDC). Severe Clostridium difficile-associated disease in populations previously at low risk-four states. Morbidity and Mortality Weekly Report. 2005;54:1201-1205
-  R.P. Bolton, S.K. Tait, P.R. Dear, et al. Asymptomatic neonatal colonization by Clostridium difficile. Archives of Disease in Childhood. 1984;59:466-472
-  S. Jangi, J.T. Lamont. Asympomatic colonization by Clostridium difficile in infants: implications for disease in later life. Journal of Pediatric Gastroenterology and Nutrition. 2010;51:2-7
-  M. Delmée, G. Verellen, V. Avesani, et al. Clostridium difficile in neonates; serogrouping and epidemiology. European Journal of Pediatrics. 1988;147:36-40
-  K. Bryant, L.C. McDonald. Clostridium difficile infections in children. Pediatric Infectious Disease Journal. 2009;28:1-2
-  G.E. Schutze, R.E. Willoughby, Committee on Infectious Disease, et al. Clostridium difficile infection in infants and children. Pediatrics. 2013;131:196-200
-  C. Pothoulakis, J.T. Lamont. Microbes and microbial toxins: paradigms for microbial–mucosal interaction. II: The integrated response of the intestine to Clostridium difficile toxins. American Journal of Physiology. Gastrointestinal and Liver Physiology. 2001;280:G178-G183
-  M. Pokorn, A. Radsel, M. Cizman, et al. Severe Clostridium difficile associated disease in children. Pediatric Infectious Disease Journal. 2008;27:944-946
-  J. Kim, S.A. Smathers, P. Prasad, et al. Epidemiological features of Clostridium difficile-associated disease among inpatients at children's hospital in the Unites States, 2001–2006. Pediatrics. 2008;122:1266-1270
-  C.M. Nylund, A. Goudie, J.M. Garza, et al. Clostridium difficile infection in hospitalized children in the United States. Archives of Pediatrics and Adolescent Medicine. 2011;165:451-457
-  M.D. Zilberberg, G.S. Tillotson, C. McDonald. Clostridium difficile infections among hospitalized children United States, 1997–2006. Emerging Infectious Diseases. 2010;16:604-609
-  S.J. Sammons, P. Toltzis. Recent trends in the epidemiology and treatment of C. difficile infection in children. Current Opinion in Pediatrics. 2013;25:116-121
-  L.C. McDonald, G.E. Killgore, A. Thompson, et al. An epidemic, toxin gene-variant strain of Clostridium difficile. New England Journal of Medicine. 2005;353:2433-2441
-  V.G. Loo, L. Poirier, M.A. Miller, et al. A predominantly clonal multi institutional outbreak of Clostridium difficile-associated diarrhea with high morbidity and mortality. New England Journal of Medicine. 2005;353:2442-2449
-  M. Warny, J. Pepin, A. Fang, 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
-  M. Cohen. Clostridium difficile infections: emerging epidemiology and new treatment. Journal of Pediatric Gastroenterology and Nutrition. 2009;48:S63-S65
-  V. Morinville, J. McDonalds. Clostridium difficile-associated diarrhea in 200 Canadian children. Canadian Journal of Gastroenterology. 2005;19:497-501
-  P. Toltzis, J. Kim, M. Dul, et al. Presence of the epidemic North American Pulsed Field type I Clostridium difficile strain in hospitalized children. Journal of Pediatrics. 2009;154:607-608
-  B. Elliott, B.J. Chang, C.L. Golledge, et al. Clostridium difficile-associated diarrhea. Internal Medicine Journal. 2007;37:561-568
-  C.P. Kelly, C. Pothoulakis, J.T. LaMont. Clostridium difficile colitis. New England Journal of Medicine. 1994;330:257-262
-  J. Guarner, J. Bhatnagar, A.L. Shane, et al. Correlation of the detection of Clostridium difficile toxins in stools and presence of the clostridia in tissue of children. Human Pathology. 2010;41:1586-1592
-  D.J. Riddle, E.R. Dubberke. Clostridium difficile infection in the intensive care unit. Infectious Disease Clinics of North America. 2009;23:727-743
-  M.H. Samore, P.C. DeGirolami, A. Tlucko, et al. Clostridium difficile colonization and diarrhea at a tertiary care hospital. Clinical Infectious Diseases. 1994;18:181-187
-  R. Turco, M. Martinelli, E. Miele, et al. Proton pump inhibitors as a risk factor for paediatric Clostridium difficile infection. Alimentary Pharmacology and Therapeutics. 2010;31:754-759
-  T.J. Sandora, M. Fung, K. Flaherty, et al. Epidemiology and risk factors for Clostridium difficile infection in children. Pediatric Infectious Disease Journal. 2011;30:580-584
-  L. Fenner, A.F. Widmer, G. Goy, et al. Rapid and reliable diagnostic algorithm for detection of Clostridium difficile. Journal of Clinical Microbiology. 2008;46:328-330
-  S. Johnson, S.A. Kent, K.J. O’Leary, et al. Fatal pseudomembranous colitis associated with a variant Clostridium difficile strain not detected by toxin A immunoassay. Annals of Internal Medicine. 2001;135:434-438
-  K. Eastwood, P. Else, A. Charlett, et al. Comparison of nine commercially available Clostridium difficile toxin detection assays, a real-time PCR assay for C. difficile tcdB, and a glutamate dehydrogenase detection assay to cytotoxin testing and cytotoxigenic culture methods. Journal of Clinical Microbiology. 2009;47:3211-3217
-  P.D. Stamper, R. Alcabasa, D. Aird, et al. Comparison of a commercial real-time PCR assay for tcdB detection to a cell culture cytotoxicity assay and toxigenic culture for direct detection of toxin-producing Clostridium difficile in clinical samples. Journal of Clinical Microbiology. 2009;47:373-378
-  S.H. Cohen, D.N. Gerding, S. Johnson, et al. Clinical practice guidelines for Clostridium difficile infection in adults: 2010 update by the Society of Healthcare Epidemiology of America (SHEA) and the Infectious Disease Society of America (IDSA). Infection Control and Hospital Epidemiology. 2010;31:431-455
-  M.P. Bauer, E.J. Kuijper, J.T. van Dissel. European Society of clinical Microbiology and Infectious Disease (ESCMID): treatment guidance document for Clostridium difficile infection (CDI). Clinical Microbiology and Infection. 2009;15:1067-1069
-  WHO. Persistent diarrhoea in children in developing countries: memorandum from a WHO meeting. Bulletin of the World Health Organization. 1988;66:709-717
-  S.R. Moore. Update on prolonged and persistent diarrhea in children. Current Opinion in Gastroenterology. 2011;27:19-23
-  K.L. Schwartz, I. Darwish, S.E. Richardson, et al. Severe clinical outcome is uncommon in Clostridium difficile infection in children: a retrospective cohort study. BMC Pediatrics. 2014;14:28
-  C.M. Surawicz, L.J. Brandt, D.G. Binion, et al. Guidelines for diagnosis, treatment, and prevention of Clostridium difficile infection. American Journal of Gastroenterology. 2013;108:478-498
-  J.O. Shim. Clostridium difficile in children: to treat or not to treat?. Pediatric Gastroenterology, Hepatology & Nutrition. 2014;17:80-84
-  J.M. Wendt, J.A. Cohen, Y. Mu, et al. Clostridium difficile infection among children across diverse US geographic locations. Pediatrics. 2014;133:651-658
-  J.E. Markowitz, K.A. Brown, P. Mamula, et al. Failure of single-toxin assays to detect Clostridium difficile infection in pediatric inflammatory bowel disease. American Journal of Gastroenterology. 2001;96:2688-2690
a Pediatrics Division Niguarda Ca’ Granda Hospital, Milan, Italy
b Microbiology Service, Niguarda Ca’ Granda Hospital, Milan, Italy
⁎ Corresponding author at: Pediatrics Division, Niguarda Ca’ Granda Hospital, P.le Ospedale Maggiore, 3, 20162 Milan, Italy. Tel.: +39 0264442433; fax: +39 0264443915.
© 2015 Editrice Gastroenterologica Italiana S.r.l., Published by Elsevier B.V.