Clostridium difficile is considered an important pathogen in hospitalized patients aged >65 years and receiving antibiotic therapy; its role in children is still controversial.1 In several studies, high carriage rates were reported in asymptomatic neonates and young children2,3 and at present testing for C. difficile is rarely performed in children <2 years of age. Nevertheless, C. difficile can cause gastrointestinal symptoms in neonates and children,1,4 and is an important pathogen in immunocompromised young patients5 and in those with Crohn’s or Hirschsprung disease.6
Recently, attention has been paid to the emergence and spread of a highly virulent strain, BI/NAP1/027, causing increased mortality and disease severity, although the highest prevalence was observed in patients >65 years of age.7 Increased incidence has also been detected in patients previously considered at low risk, especially children and adolescents. In a report by Chernak et al8 11 of the 23 patients with community-acquired C. difficile disease were <18 years of age.
We report 3 cases of C. difficile infection in children detected within 2 months (from October to December 2006) at the University Medical Centre Ljubljana, Slovenia (summarized in Table 1, http://links.lww.com/A479). Two of the cases were severe requiring intensive or surgical treatment.7 Until now, C. difficile infections of such severity have not been seen in pediatric patients at our institution and C. difficile was not frequently sought as a cause of diarrhea in children.
A 13-year-old girl was admitted to the Department of Infectious Diseases with a 2-day history of fever, vomiting, and diarrhea resulting in syncope. She had recently completed a 5-day course of amoxicillin-clavulanic acid for skin infection. The patient developed hypovolemic shock requiring intensive care unit (ICU) admission, fluid resuscitation, and inotrope support. The laboratory was characterized by elevated C-reactive protein (330 mg/L) and leukocytosis (27 × 109/L) with increased immature cells. C. difficile infection was suspected but because the latex test for C. difficile A toxin (Becton Dickinson) was negative, the patient was given cefotaxime for presumed sepsis on admission to the ICU. Subsequently, stool tested with enzyme-linked immunosorbent assay (ELISA) for A/B toxin (Meridian) was positive. Because blood cultures remained negative, cefotaxime was substituted with oral and due to vomiting with parenteral metronidazole. Because the patient did not improve adequately, oral and later rectal vancomycin was added to the therapy. Colonoscopy demonstrated pseudomembranous colitis predominantly affecting the distal colon. C. difficile was grown in culture from the stool and from colonic aspirate. The patient gradually recovered and completed 10 days of vancomycin (including 4 days given rectally) and 14 days of metronidazole treatment. The disease course was complicated with pleural effusion requiring evacuation. The pleural fluid was classified as transudate and was probably due to hypoalbuminemia. No further investigations were performed because no fluid reaccumulated. At 2 months post-hospital discharge, the patient was considered cured and no sequelae were present.
An 8-month-old boy was hospitalized with a 24-hour history of fever, vomiting, and diarrhea. He had had surgery for Hirschsprung disease at the age of 3 months and was last hospitalized at the Department for Pediatric Surgery because of constipation problems 1 month earlier when he received penicillin orally for 10 days for perianal skin infection. His 3-year-old sister had contact with horses and developed diarrhea several days before him. The boy had nonbloody diarrhea. The course of disease was characterized by high fever and elevated C-reactive protein (214 mg/L). Stool culture was negative for Salmonella, Shigella, Campylobacter, and Escherichia coli and a latex test was negative for C. difficile A toxin. Subsequently, stool sample tested positive for the presence of C. difficile A/B toxin by ELISA and C. difficile was isolated in stool culture. Calicivirus was detected by ELISA in a stool sample obtained on day 3 of hospitalization (although it had tested negative on day 1). The course of disease was relatively mild and the patient improved with oral metronidazole therapy he received for 7 days.
A 10-year-old girl with Down syndrome was admitted to the pediatric ICU after complete surgical correction of Fallot tetralogy. In the early postoperative period she developed fever and received teicoplanin and ceftazidime. Despite antibiotic treatment, the patient’s condition deteriorated and, subsequently, she was given imipenem and fluconazole. The patient required intensive therapy including peritoneal dialysis for renal failure. Eight days after surgery, she developed a distended, painful abdomen followed by watery and later bloody diarrhea, peritonitis, and paralytic ileus resulting in multiorgan failure and shock. After insertion of peritoneal drains, 500–800 mL of odorless, brownish fluid was collected daily. The stool tested positive for C. difficile A/B toxin by ELISA and for rotavirus by ELISA and electron microscopy. C. difficile toxin was also positive in peritoneal fluid. Despite oral vancomycin, the patient’s condition further deteriorated and oral metronidazole and rectal (because of paralytic ileus) vancomycin were added to therapy. Sigmoidoscopy demonstrated pseudomembranous colitis. Because intestinal perforation was suspected, laparotomy was performed 52 days after heart surgery, demonstrating necrotic transverse colon with several abdominal abscesses and numerous perforations of the terminal ileum. Peritoneal lavage with terminal ileostomy placement and colectomy was performed. The patient received 21 days of vancomycin and 33 days of metronidazole treatment and was discharged from the hospital 93 days after heart surgery in a stable clinical condition.
An unusual cluster of 3 C. difficile infections in pediatric patients is described. All 3 cases were positive for C. difficile toxin A, but strains were isolated only from cases 1 and 2 in the diagnostic laboratory where they were frozen and subsequently made available for molecular characterization. Both strains were grouped in the toxinotype 09 and correspondingly had no genes for binary toxin when tested by PCR method.10 Both strains were ribotyped as described by Stubbs et al.11 Epidemiologic association of case 3 with a hospitalized case (case 2) could not be ruled out because no isolate was available for case 3. Still, in case 1 the strain was community-acquired and was in ribotype profile different from the strain from case 2.
Two patients (case 2 and 3) had underlying conditions previously reported to be associated with C. difficile disease (Hirschprung disease and Down syndrome) and both had previously been hospitalized. The source of infection in case 1 was less obvious. There was indirect contact with a hospital environment because the patient’s mother worked in the hospital as a nurse. She has not been tested. Also, the girl had regular contact with a horse. The animal tested negative for C. difficile, but testing was performed more than 2 months after the girl’s disease. The horse was asymptomatic at the time of sampling, but occasionally received antibiotics because of various problems due to advanced age. Indirect contact with a horse was also present in case 2 via the patient’s sister who developed a mild diarrheal episode (with no stool examination) before her brother. The second horse also tested negative for C. difficile. The animals could therefore not be proven as sources of infection in our cases but horses have been increasingly recognized as important hosts for C. difficile, as well as pigs and calves.12
Apart from the above described cases, at the Department of Infectious Diseases C. difficile toxin was positive in 6 additional symptomatic children with milder disease in the period from November 2006 to May 2007. Four of these patients also tested positive for enteric viruses (2 for rotavirus and 2 for calicivirus), one had simultaneous pneumococcal bacteremia, and one had adenoviral upper respiratory tract infection. All patients had previously been treated with antibiotics. When comparing these patients with the 3 severe cases, significantly lower C-reactive protein (CRP) values (mean value 37 versus 259 mg/L) and white blood cell (WBC) counts (14.4 versus 24.3 × 109/L) were seen in mild cases and in all severe cases a left shift was observed.
Reports from North America suggest that the gradual emergence of C. difficile in populations previously considered at low risk seems to correlate with the increased incidence and severity of C. difficile infections reported in recent years.8,13 Although there is no national C. difficile surveillance system in Slovenia, the situation could be different from that in the United States and in some European countries in that epidemic type BI/NAP1/027 has so far not been detected and none of the hospitals had reported increased infection rates (Rupnik, unpublished data). In other countries, fulminant community-acquired cases requiring hospitalization were associated with variant toxinotypes (other than toxinotype 0) and producing binary toxin, but not necessarily of the type BI/NAP1/027.8,14,15 In contrast, strains isolated in this study were nonvariant toxinotypes 0, indicating again that such strains can cause severe disease.
When caring for patients with gastrointestinal symptoms, especially when high WBC counts and/or elevated CRP are present, clinicians should consider C. difficile infection in the differential diagnosis. This is particularly important in countries that have not yet experienced epidemics with the new hypervirulent type 027 and where the physicians caring for patients at low risk for C. difficile infection are not aware of this problem. In some cases, as exemplified by this report, because of low sensitivity the rapid microbiologic tests give false negative results, which is one of the well known problems in C. difficile diagnostics. Therefore, not only rapid C. difficile specific tests but also tests with high sensitivity and close cooperation between clinicians and microbiologists are needed to ensure early recognition and treatment of this severe disease.
1. Wilson ME. Clostridium difficile
and childhood diarrhea: cause, consequence, or confounder. Clin Infect Dis
2. Matsuki S, Ozaki E, Shozu M, et al. Colonization by Clostridium difficile
of neonates in a hospital, and infants and children in three day-care facilities of Kanazawa, Japan. Int Microbiol
3. Kato H, Kato N, Watanabe K, et al. Application of typing by pulsed-field gel electrophoresis to the study of Clostridium difficile
in a neonatal intensive care unit. J Clin Microbiol
4. Enad D, Meislich D, Brodsky NL, Hurt H. Is Clostridium difficile
a pathogen in the newborn intensive care unit? A prospective evaluation. J Perinatol
5. Wolfhagen MJ, Meijer K, Fluit AC, et al. Clinical significance of Clostridium difficile
and its toxins in faeces of immunocompromised children. Gut
6. Morinville V, McDonald J. Clostridium difficile
-associated diarrhea in 200 Canadian children. Can J Gastroenterol
7. Kuijper EJ, Coignard B, Tull P; the ESCMID Study Group for Clostridium difficile
(ESGCD); EU Member States and the European Centre for Disease Prevention and Control (ECDC). Emergence of Clostridium difficile
-associated disease in North America and Europe. Clin Microbiol Infect
8. Chernak E, Johnson CC, Weltman A, et al. Severe Clostridium difficile
-associated disease in populations previously at low risk–four states. MMWR Morb Mortal Wkly Rep
9. Rupnik M, Avesani V, Janc M, von Eichel-Streiber C, Delmee M. A novel toxinotyping scheme and correlation of toxinotypes with serogroups of Clostridium difficile
isolates. J Clin Microbiol
10. Stubbs S, Rupnik M, Gibert M, Brazier J, Duerden B, Popoff M. Production of actin-specific ADP-ribosyltransferase (binary toxin) by strains of Clostridium difficile
. FEMS Microbiol Lett
11. Stubbs SL, Brazier JS, O’Neill GL, Duerden BI. PCR targeted to the 16S–23S rRNA gene intergenic spacer region of Clostridium difficile
and construction of a library consisting of 116 different PCR ribotypes. J Clin Microbiol
12. Rupnik M. Is Clostridium difficile
-associated infection a potentially zoonotic and foodborne disease? Clin Microbiol Infect
13. McDonald LC, Owings M, Jernigan DB. Increasing rates of Clostridium difficile
infection among patients discharged from US short-stay hospitals, 1996–2003. Emerg Infect Dis
14. Barbut F, Decré D, Lalande V, et al. Clinical features of Clostridium difficile
-associated diarrhoea due to binary toxin (actin-specific ADP-ribosyltransferase)-producing strains. J Med Microbiol
15. Terhes G, Urban E, Soki J, Hamid KA, Nagy E. Community-acquired Clostridium difficile
diarrhea caused by binary toxin, toxin A, and toxin B gene-positive isolates in Hungary. J Clin Microbiol