Contamination of accidental wounds with anaerobic bacteria in children

INTRODUCTION
Today's development of means of transport, and the modernization of agricultural machinery in particular have led to frequent injuries characterized by the formation of deep pockets which are considered to be primarily infected. Wounds sustained in rural areas, or occurring in bicycle or motorcycle accidents are accompanied by a particularly high risk of contamination by various aerobic and anaerobic bacteria. The most important of anaerobic bacteria are those belonging to the Clostridium genus. Prior to the widespread use of antibiotics, Clostridia were isolated in 10-30% of peacetime wounds, and in as many as 80% of war wounds (1, 2). Clostridia are anaerobic, gram-positive, spore forming, rod-shaped bacilli found widely distributed in nature, occurring in the soil, water, plants and in the intestinal tract of animals and humans. (1) One hundred or more species have been differentiated (3); less than 20 are associated with diseases in humans (4). There are three types of clostridial infections: contamination, clostridial cellulitis and clostridial myonecrosis (5). Clostridium perfringens is most commonly responsible for the most severe manifestation, gas gangrene or clostridial myonecrosis. Other agents causing gas gangrene include C. septicum, C. novyi, C. sordelii, and C. histolyticum (1). Primary wound treatment is the basic surgical postulate aimed at preventing infection and enabling appropriate wound healing. Wound management according to Friedrich, including extensive debridment with opening of all pockets, irrigation with hydrogen peroxide if necessary, and drainage, reduces the possibility of anaerobic infection to a minimum (6).
MATERIALS AND METHODS The study included patients ranging in age from 11 months to 14 years treated in our hospital for accidental wounds in the period from January 2000 to May 2004 and with bacteriological findings of Clostridia in wound swabs. In the majority of patients the swabs were taken before surgical treatment of the wound, and some patients required periodical collection of swabs in cases when the wound had failed to heal as expected, or signs of infection appeared. The swabs received in the microbiological laboratory of our hospital were cultured in both fluid (thioglycollate broth) and solid media. Direct Gram stain was also performed.
The isolation of Clostridia was performed on blood agar medium after incubation at 37°C for 24-48 hours in anaerobic conditions (GasPak, BD). Identification of colonies was performed using the commercial test API 20 A (bioMerieux, France). Antimicrobial sensitivity of isolates was determined using the semiquantitative test, anaATB (bioMerieux, France). RESULTS Bacteria of the Clostridium genus were isolated from wound swabs in 18 * Children’s Hospital Zagreb Correspondence to: Igor Bumči, M. D., Children’s Hospital Zagreb, 10 000 Zagreb, Klaićeva 16, Croatia children. Of those, 12 children had lacerated and contused wounds, and 6 had burns. The wounds were sustained either as a result of a fall from a motorcycle or bicycle, or were inflicted by agricultural machinery. Clostridia were isolated from the wound swabs taken at the time of primary wound management in 8 children, and between the second and fifth day in 10 children.
The most frequently isolated species was Clostridium perfringens recovered in 14 children, followed by Clostridium spp. isolated in 4 children, and Clostridium butyricum and Clostridium ramosum in one child, respectively. In 6 children clostridia were the only isolate found in the wound, while 12 children with an isolated clostridium had also one or more aerobic bacteria. Antibiotic therapy was administered in 14 children. One child developed the clinical presentation of gas gangrene.

DISCUSSION
Anaerobic bacteria are widely distributed in nature. They can also be found in humans as part of normal skin flora, as well as in nasopharyngeal, oropharyngeal, oral, intestinal and urogenital tract mucosa. The majority of anaerobic infections are endogenous, while anaerobic infections such as food intoxication, gastroenteritis, enteritis, and wound contamination are exogenous. As clinical recognition of anaerobic infection is not always simple, and thus neither is the decision on appropriate therapy, the clinical material requires bacteriological analysis. As a rule, material from all parts of the body which are not expected to be contaminated by the normal flora, is assessed for anaerobic bacteria as well as for facultative anaerobes and aerobic microorganisms. This includes, among others, wound material. Fine-needle or -syringe aspiration specimens or tissue samples are optimal material for the isolation of anaerobic bacteria (7). In our study, all specimens from which anaerobes were isolated were collected from swabs. Having the microbiological laboratory within the premises of our hospital, we were able to analyse the swabs immediately after collection. Bacteria of the Clostridium genus were recovered from the swabs taken during primary wound treatment in 8 children, of whom 4 (66,6%) had burns and 4 (33,3%; 4/12) had lacerated and contused wounds. This may be explained by the common practice in our hospital which implies mandatory taking of swabs from burn wounds during the primary and each subsequent treatment, while in other accidental wounds the necessity of swab collection is a matter of the clinician's assessment. This is why 8 children had their swabs with isolates positive for Clostridium taken between the second and fifth day when wound infection was observed. In 2 children with burn wounds Clostridia were not isolated from the swab at the time of primary treatment, but in the first child on the second and in the second child on the third day post injury, which is considered secondary contamination. In our patients Clostridium perfringens was most frequently recovered, in as many as 12 children. Clostridium perfringens is mentioned in literature as a clostridial species most usually isolated from the soft tissues (1). Although Clostridium ramosum is cited in literature as about as common (1), in our study it was recovered in only one child. Clostridium ramosum is a species otherwise most usually isolated from faecal samples in children, and is rarely associated with a serious clinical presentation (8). However, V a n d e r V o r m e t a l . describe a case of gas gangrene in an immuno- compromised child caused by Clostridium ramosum (9).
Some other types of clostridia are more frequently recovered from other clinical material or in particular groups of patients, such as diabetics and patients suffering from gastrointestinal tract malignancies and leukemia (10, 11), as well as children with cyclic neutropenia (12, 13). Thus Clostridium septicum is the most common causative agent of spontaneous non-traumatic gas gangrene (10, 11). There was no C. septicum isolated from the swabs of our patients. Mi - n u t t i e t a l . describe a pediatric patient with non-traumatic gas gangrene caused by Clostridium perfringens, which is a rare occurrence, not only because gas gangrene is uncommon in the pediatric population but also because non-traumatic gas gangrene is infrequently associated with this particular type of clostridia (14). Clostridium perfringens is most significantly responsible for gas gangrene associated with trauma (1, 5). In the case of our patient with the clinical presentation of gas gangrene Clostridium perfringens also proved to be the cause. Gas gangrene can in no way be considered a disease of the past, because despite early wound treatment, new antimicrobial drugs, and all other modern methods of treatment, histolytic clostridia still remain a cause of death and loss of extremities (15). Our patient with a developed clinical picture of gas gangrene, also required amputation of the lower extremity. Some other clinical manifestations caused by Clostridium perfringens such as meningitis (16) and sepsis (17) are described in literature. Clostridium as the only isolate was found in one third of our patients (6/18), while in 12 mixed flora was isolated. In a study on a larger number of samples (14), 84% of soft-tissue infections harboring clostridia also had other bacteria (18). Brook in his investigation found polymicrobial infection in 85% of children, and only 15% Clostridium sp in clear culture (8). Antimicrobial therapy was administered in 77.8% of our patients (14/18).
Penicillin G is the antibiotic treatment of choice for infections caused by clostridia, in spite of some studies indicating that resistance of clostridia to penicillin is on the increase (1). Our patients received either penicillin or clindamycin. In literature, treatment with prophylactic antibiotics is recommended for contaminated wounds (1). In our opinion, in cases of wounds where primary contamination is suspected, as well as in wounds with healing per secundam, and where there are no obvious clinical signs of purulent infeciton, in addition to adequate surgical management, antibiotic therapy should also be administered. Though serious manifestations caused by Clostridium sp. in children are rare in accidental wounds, we consider taking swabs for isolating anaerobic bacteria indispensable during primary wound treatment.

REFERENCES
1. Bennett Lorber. Gas gangrene and other Clostridium- associated diseases. In: Mandell GL, Bennett JE, Dolin R. Mandell, Douglas, and Bennett's principles and practice of infectious diseases. 5th ed. Philadelphia: Churchill Livingstone, 2000: 2549-2561.
2. Bradić I. Kirurške infekcije klostridijima. In: Bradić I. Kirurgija, 1st ed. Zagreb: Medicinska naklada, 1995: 91-92.
3. Holt JG, Krieg NR, Sneath PHA, Staley JT. Bergey's Manual of Determinative Bacteriology. 9th ed. Baltimore: Williams and Wilkins, 1994: 560.
4. Hatheway CL. Toxigenic clostridia. Clin Microbiol Rev 1990;3:66-98.
5. United States Deparment of Defens. Clostridial Infections. Emergency War Surgery NATO Hand- book: Part II. Response of the Body to Wounding: Chapter XI: Infection). Available from: http:www. vnh.org/EWSurg/ch11/11ClostridialInfections.html
6. Hančević J, Turčić J, Antoljak T. Rana. In: Bradić I. Kirurgija, 1st ed. Zagreb: Medicinska naklada, 1995:97-108.
7. Allen SD, Emery CL, Lyerly DM. Clostridium. In: Murray PR. Manual of clinical microbiology.8th ed. Washington, D.C: ASM Press, 1993;835-56. 8. Brook I. Clostridial infection in children. J Med Microbiol 1995;42:78-82.
9. Van der Vorm ER, von Rosenstiel IA, Spanjaard L, Sankert J. Gas gangrene in an immunocompromised girl due to a Clostridium ramosum infection. Clin Infect Dis 1999;28:923-4.
10. Hengster P, Pernthaler H. Gas gangrene: necropsy is imperative. Lancet 1996;347:553.
11. Chin RL, Martinez R, Garmel G. Gas gangrene from subcutaneous insulin administration. Am J Emerg Med 1993;11:622-625.
12. Bar-Joseph G, Halberthal M, Sweed Y, Bialik V, Shoshani O, Etzioni A. Clostridium septicum infection in children with cyclic neutropenia. J Pediatr 1997;131:317-9.
13. Bratton SL, Krane EJ, Park JR, Burchette S. Clostridium septicum infections in children. Pediatr Infect Dis J 1992;11:569-75. 14. Minutti CZ, Immergluck LC, Schmidt ML. Spontaneous gas gangrene due to Clostridium perfringens. Clin Infect Dis 1999;28:159-60. 15. Hart GB, Lamb RC, Strauss MB . Gas gangrene I. A collective review. J Trauma 1983;23:991-5.
16. Penrose-Stevens A, Ibrahim A, Redfern RM. Localized pneumocephalus caused by Clostridium perfringens meningitis. Br J Neurosurg 1999;13: 85-6.
17. Moustoukas NM, Nichols RL, Voros D. Clostridial sepsis: unusual clinical presentations. South Med J 1985;78:440-4.
18. Sherwood LG, Thadepalli H. Isolation of Clostridium in human infections: evaluation of 114 cases. J Infect Dis 1975;131:S81-5.