Epidemiology of burn unit infections in children

Mehmet Faruk Geyik, MD,a Mustafa Aldemir, MD,b Salih Hosoglu, MD,a and Halil Ibrahim Tacyildiz, MDb Diyarbakir, Turkey
Objective: The aim of this study was to determine the epidemiology of burn unit infections, the effect of these infections on the mortality rate, and antibiotic resistance pattern of the predominant bacteria isolated from children. Patients and method: Epidemiologic data for 610 children, aged 0 to 15 years, admitted to the burn unit at Dicle University Hospital during a 5-year period were collected and analyzed. Results: In 207 patients (33.9%), 279 nosocomial infections were identified. The most common types of infections were burn wound infections (72.4%), urinary tract infections (10.8%), pneumonia (9.3%), and septicemia (7.5%). Pseudomonas aeruginosa (181 isolates) was the most common microorganism. Thirty-six patients (5.9%) died at the hospital. Sepsis was associated with mortality in 18 (50%) cases, pneumonia in 6 (17%), and varied noninfectious reasons in 12 patients (33%). P aeruginosa isolates showed high resistance to commonly used antimicrobials. Antibiotic susceptibility test results suggested that imipenem was the most effective agent for P aeruginosa and Escherichia coli strains. Conclusion: The major type of nosocomial infections in the burn unit was burn wound infections, and the majority of nosocomial infections resulted from multiple drugresistant, gram-negative bacteria. (Am J Infect Control 2003;31:342-6.)

INTRODUCTION
Pediatric burn injuries are a major environmental agent responsible for significant morbidity and mortality in developing countries. Infection is the most common cause of death after burn injury.1,2 Children are at high risk for infection as a result of the nature of the burn injury itself, the immunocompromising effects of burns, prolonged hospital stays, and intensive diagnostic and therapeutic procedures. In addition, the control and prevention of infectious diseases among children present a greater and more specialized problem because childrens immune systems are less developed than those of adults. The best management of infections is to prevent patients from becoming infected. Infections may develop easily during the hospitalization period for patients with serious burn wounds. Pseudomonas aeruginosa is the most common pathogen isolated from burn wound infection in many studies.3-5
From the Department of Infectious Diseases and Clinical Microbiology,a and the Department of General Surgery, Faculty of Medicine,b Dicle University Hospital. Reprint requests: Mehmet Faruk Geyik, MD, Infectious Diseases and Clinical Microbiology, Dicle University Hospital, 21280, Diyarbakir, Turkey. Copyright 2003 by the Association for Professionals in Infection Control and Epidemiology, Inc. 0196-6553/2003/$30.00 + 0 doi:10.1067/mic.2003.26

Although the epidemiology of burn unit infections in adults is well established in the literature, there is less information available about burn infections in children. No previous attempt has been made to evaluate surveillance of burn unit infections, relationship between mortality and infections, and the antibacterial resistance patterns of the infections in the pediatric burn cases.

MATERIALS AND METHOD

Dicle University Hospital Burn Unit is in the southeast part of Turkey. It has 3 rooms with 12 beds and a hydrotherapy room. The average admissions are about 350 patients annually. The epidemiology and microbiology of burn unit infections and the outcome of the patients treatment were studied retrospectively. The analysis was completed on the basis of a review of the charts of 610 patients admitted to the burn unit between January 1995 and June 1999. All patients with burns to >10% of total body surface area (TBSA) or burns to >2% of TBSA localized at a critical region, such as the neck, were admitted to the burn unit. The morbidity and mortality related to burn infections and the microbiologic aspects of infections were reviewed. The first wound swabs were collected immediately after admission from all patients. The samples from various sites of burn wounds (2-7 swabs per patient) were tested during the study period. A smear was taken from the wound, and it was stained with Grams stain. Then

342

Geyik et al

October 2003 343

Table 1. Patient characteristics in 610 burn patients

Patient characteristics Sex Male Female Hospital stay (mean SD) Age 0-1 y 1-2 y 2-4 y 4-7 y 7-15 y % TBSA 9 10-29 30-49 50-69 70-89 90 Cause of burn Scalding Flames Electrical Outcome Complete recovery Recovery with complication Death No of patients (%)

406 (66.6) 204 (33.4) 12 9.4 20 68 230 165 127 27 382 149 43 5 4 (3.3) (11.1) (37.7) (27.1) (20.8) (4.4) (62.6) (24.4) (7.0) (0.8) (0.7)

384 (63.0) 125 (20.5) 101 (16.5) 510 (83.6) 64 (10.5) 36 (5.9)

Fig 1. Distribution of the burned patients and nosocomial infections by years. N, noninfected patients; NI, nosocomial-infected patients. No statistical significance was noted among patients with nosocomial infections and those who were not infected (P >.05).

RESULTS Epidemiology
Between January 1995 and June 1999, 610 patients were treated at the Dicle University Hospital Burn Unit. The patients ages ranged between 3 months to 15 years (mean SD, 4.8 3.9 years) (Table 1). The mean length of hospital stay was 12.0 9.4 days. After admission to the unit, 96% of the patients received antibiotics for prophylactic purposes. Four hundred forty-five cases (69.7%) had burns to more than 20% TBSA. Finally, no criteria for antibiotic prophylaxis was noted in 128 patients (21%), and antibiotic use in these cases was not justified. The mean time of antibiotic prophylaxis was 7.3 days (range, 3-15 days). Sulbactam/ampicillin (86%), clindamycin (9%), and cefazolin (6%) were the most frequently used antibiotics for prophylaxis. Ceftriaxone, ceftazidime, and imipenem were the most frequently used antibiotics for treatment purposes of burn infections. Researchers evaluated 279 nosocomial infections in 207 patients (33.9%). No statistical significance was noted among nosocomial infections vs patients without infections by years (P >.05) (Fig 1).

additional swabs were cultured every third day. The swabs were cultured for aerobic bacteria with blood agar and eosin metilen blue agar plates, and the swabs examined revealed the presence of aerobic bacteria. Antibiotic prophylaxis was initiated in patients with one of the following situations: (1) burn to >20% of TBSA; (2) admission to the unit later than 24 hours after injury; and (3) presence of contamination signs. When none of these criteria were present, antibiotic use was inappropriate. The following patients data were recorded: age, sex, hospitalization period, TBSA burned, causes of the burn, nosocomial burn wound infections, nosocomial infections in other body sites, antibiotic use after admission, antibiotic susceptibility patterns, pathogens isolated from hospital infections, outcome of the patients, and the relationship between infections and outcome. Identification of bacteria was performed according to standard bacteriologic methods. Antibiotic susceptibility tests were completed with the Kirby-Bauer disc diffusion technique.6 An infectious disease specialist diagnosed all cases of nosocomial infections with the criteria established by the Centers for Disease Control and Prevention.7 The statistical analyses were performed with SPSS (SPSS 10.0 for Windows; SPSS Inc, Chicago, IL). The chi-square test was used to compare categorical samples, and values of P <.05 were considered statistically significant.

Microbiology
In microbiologic studies of 1650 clinical samples from 170 patients, 294 bacteria were isolated during the hospitalization period. Thirty-seven cases did not have positive microbiologic results, and the diagnosis of nosocomial infection was made on the basis of only clinical findings. P aeruginosa was isolated from 181 samples (65%) and was the most frequent cause of nosocomial infections (Table 2). P aeruginosa was also the most frequent isolate from cases with septicemia.

344 Vol. 31 No. 6

Geyik et al

Table 2. Types of microorganisms isolated from nosocomial infections in the burn unit
Wound infection UTI Pneumonia Sepsis no (%) no (%) no (%) no (%) P aeruginosa E coli S aureus Other Culture negative Total nosocomial infections 150 15 12 7 18 (74) (7) (5) (3) (9) 11 15 3 1 0 (37) (50) (10) (3) 5 (19) 0 2 (7) 0 19 (73) Total no (%)

15 (71) 181 (65) 3 (14) 33 (12) 3 (14) 20 (7) 0 8 (3) 0 37 (13) 21 (7.5) 279

202 (72.4) 30 (10.8) 26 (9.3)

Fig 2. Antibiotic resistance of major microorganisms from nosocomial infections in burned unit (BU) and hospital (H). AC, amikacin; CIP, ciprofloxacin; CTZ, ceftazidime; CX, ceftriaxone; IMP, imipenem; MEZ, mezlocillin; SAM, sulbactam/ampicillin. Thirty-six patients (5.9%) died at the hospital. Fifteen patients with septicemia had burns to >40% of TBSA. The causes of mortality were sepsis in 18 (50%) cases, pneumonia in 6 (17%), and different noninfectious reasons in 12 patients (33%). P aeruginosa isolates showed high resistance to commonly used antimicrobials. P aeruginosa was isolated in 181 patients, and 126 of the isolates were resistant to given antibiotics for prophylaxis. Antibiotic susceptibility test results suggested that imipenem was the most effective agent to P aeruginosa and E coli strains. Susceptibility patterns of P aeruginosa and E coli are shown in Fig 2. S aureus isolates were found to be resistant to methicillin in 60% samples.

UTI, Urinary tract infection.

DISCUSSION
The development of infections in burn cases is serious because of their effects on the course of the disease and patient outcomes. Many burned patients die as a result of infection during their hospital courses. Burns are a major problem that threatens child health in our region. The rate of infection in burn cases are extremely high in developing countries.2,8 This may result from the prevalence of low socioeconomic groups where poor hygienic conditions prevail; malnutrition may also play a role in the earlier establishment of the infection.8,9 Families and hospital personnel frequently touch hospitalized children. Infection is easily spread by means of this close contact. Inadequate measures to prevent cross infection by burn unit workers and visitors may also be implicated. In this study period, there was not an efficient hospital infection control program in the burn unit. Successful infection control measures depend on a burn unit built with exacting specifications, well-educated burn unit workers, surveillance of

the bacteria prevalent in the burn unit and their antimicrobial resistance patterns, and an efficient hospital infection control program.10 In children, fever is not a reliable predictor of infection. At the first suspicion of sepsis, the wound must be inspected.11 Discoloration of eschar, cellulitis of surrounding tissue, progressive separation of eschar, purulent drainage, and pain in the wound are worrisome signs. Wound infections are common problems in burn units and mostly originate from nosocomial contamination. P aeruginosa is a commonly isolated microorganism in burn wounds, to which special attention should be given.12-14 In recent years, epidemic and endemic cases resulting from multiple drugresistant P aeruginosa were identified from different centers.15 Inappropriate antibiotic use for prophylactic and treatment purposes and difficulty in isolation of the patients increase resistance to P aeruginosa strains. Long-term antibiotic use in suboptimal concentration causes occurrence of resistant microorganisms. In this study, the majority of the patients had one or more antimicrobial agents after admission for prophylaxis. Different reasons may contribute to the infection rate in this burn unit. Poor compliance to hygiene and inadequate disinfection or sterilization of mattresses, bed sheets, dressing materials, and other equipment used for the patients care, prolonged catheterization, central or peripheral lines, inefficient isolation of infected patients, overcrowding, decreased host resistance, and inappropriate antibiotic use are the most important causes for nosocomial infections in burn units.16 To control infection in the burn unit, overcrowding must be avoided, strict hand washing both before and after handling patients must be implemented, and restriction of movement within the burn unit must be established. In the study burn unit, lack of an infection control program was an important reason for failed management of some patients.

Geyik et al

October 2003 345

The pattern of bacterial resistance is important for epidemiologic surveillance and treatment options in patients with burns. The antimicrobial treatment closely depends on the bacterial resistance pattern. The antimicrobial resistance patterns seen in this study gives serious cause for concern because the predominant bacterial isolates were highly resistant to the commonly available antimicrobial agents in many burn units. In fact some new -lactamase variations were defined from P aeruginosa strains from burn units in Turkey.17 At the study burn unit, antibiotic use was highly excessive during the study period, and commonly multidrug-resistant bacteria were isolated. Although it is known that the spectrum of antibacterial resistance varies from unit to unit and time to time,18 the distribution of bacteria and antibacterial resistance of these strains found in this study should serve as a guide for the next empirical treatment. A marked increase in the number of nosocomial infections resulting from methicillin-resistant S aureus has been reported in many countries.19 In Cremers study,5 the most frequent microorganisms from burn wound infections were found to be Pseudomonas spp, (49%) Staphylococci, (18%), E coli (18%) and Streptococcus faecalis (10%). A study20 from Poland reported that the most common microorganisms were P aeruginosa (31%) and S aureus (29%), followed by Proteus mirabilis, Burkholderia cepecia, and Klebsiella pneumoniae in burn unit infections. In this study, P aeruginosa was the most predominant microorganism in patients with burns. During the 5-year period of the study, several antipseudomonal drugs, such as ceftazidime and imipenem, along with aminoglycosides were used depending on culture and susceptibility results. In the study burn unit, long-term prophylactic antibiotics were used in inpatients, and this may have contributed to the increasing resistance to ceftazidime and other antibiotics. Clinicians may feel an obligation to use more antibiotics for prophylaxis to prevent infection because of poor hygiene conditions in the burn unit. Septicemia is the most important reason leading to mortality in burn cases after the first 3 days. In patients with burns to >40% of TBSA, the risk of septicemia and deadly complication are highly possible. Previous blood cultures are the best tools for presumptive diagnosis of the septic agent and early antibiotic therapy in serious burn cases.15 Despite the current use of broadspectrum antibiotics and aggressive diagnostic and therapeutic techniques, P aeruginosa bacteremia occurs in 10% of patients with burns with a mortality rate of 80%.21,22 Sepsis was noted in 21 of 207 (10.1%) patients with nosocomial infections. P aeruginosa was noted as a causative agent in 15 cases (71%) of the sepsis. In this series, the rate of septicemia was similar to

previous series.13,15,23 The high incidence of P aeruginosa septicemia in the study unit might have been a result of poor isolation facilities, inappropriate antibiotic use, and inefficient infection control practice. It is often the wound or the lung that serves as a focus for sepsis. Therefore, researchers concentrated on both areas. The main source of septicemia in these patients appeared to be the burn wound itself. P aeruginosa was resistant to multiple tested drugs. The incidence of P aeruginosa resistant to ceftazidime (90%) and imipenem (25%) was high. In addition, the incidence of E coli resistant to cephalosporins was high. P aeruginosa, found commonly in the hospital environment, may lead to contamination and nosocomial infections.2 Overcrowding is one of the most important causes for cross-contamination infections in burn units.24 In this study, the nosocomial infection rates were similar throughout of the study years. In conclusion, because of the limited number of burn units, the density of patients per unit was high in southeast Turkey. Antimicrobial prophylaxis was not given on the basis of rational indications in the burn unit. An effective infection control program should be established at all burn units. The prevention of nosocomial burn wound infections may be minimized with fastidious wound care. It is important that the staff from burn units and infection control units work together to control infections. Team effort can help avoid infections and solve problems. The education of the burn unit staff and establishment of a follow-up program are important for preventing nosocomial infections.
References
1. Lari AR, Panjeshahin MR, Talei AR, Rossignol AM, Alaghehbandan R. Epidemiology of childhood burn injuries in Fars province, Iran. J Burn Care Rehabil 2002;23(1):39-45. 2. Lari AR,Alaghehbandan R, Nikui R. Epidemiological study of 3341 burns patients during three years in Tehran, Iran. Burns 2000;26:49-53. 3. Morrison AJ, Wenzel RP. Epidemiology of infections due to Pseudomonas aeruginosa. Rev Infect Dis 1984;6:627-42. 4. Tredget EE, Shankowsky HA, Joffe AM, et al. Epidemiology of infections with Pseudomonas in burn patiens: the role of hydrotherapy. Clin Infect Dis 1992;15:941-9. 5. Cremer R, Ainaud P, Le-Bever H, Fabre M, Carsin H. Nosocomial infections in a burn unit. Results of a prospective study over a year. Ann Fr Anesth Reanim 1996;15:599-607. 6. Koneman EW, Allen SD, Janda WM, Schreckenberger PC, Winn WC. Color atlas and textbook diagnostic microbiology, 4th ed. Philadelphia: JB Lippincott; 1992. 7. Garner JS, Jarvis WR, Emori TG, Horan,TC, Hughes JM. CDC definition for nosocomial infections, 1988. Am J Infect Control 1988;16: 128-40. 8. Mabogunje OA, Khwaja MS, Lawrie JH. Childhood burns in Zaria, Nigeria. Burns Incl Therm Inj 1987;13(4):298-304. 9. Esposito S. Immune system and surgical site infection. J Chemother 2001;13(1):12-6. 10. Ozumba UC, Jiburum BC. Bacteriology of burn wounds in Enugu, Nigeria. Burns 2000;26:178-80. 11. Barillo DJ, Mc Manus AT. Infection in burn patiens. In: Armstrong D,

346 Vol. 31 No. 6

Cohen J, editors. Infectious diseases. Spain: Harcourt Publishers; 1999. p. 3-8. Leblebiciog lu H. Common pathogens in nosocomial infections. J Klimik 1993;6:106-10. McManus AT, Mason AD, McManus WF, Pruitt BA. Twenty-five year review of Pseudomonas aeruginosa bacteremia in a burn center. Eur J Clin Microbiol 1985;4:219-23. Martin MA. Nosocomial infections in intensive care units: an overview of their epidemiology, outcome, and prevention. New Horiz 1993;1:162-71. Gang RK, Bang RL, Sanyal SC, Mokaddas E, Lari AR. Pseudomonas aeruginosa septicaemia in burns. Burns 1999;25:611-6. Pandit DV, Gore MA, Saileshwar N, Dedhar LP. Laboratory data from the surveillance of a burns ward for the detection of hospital infection. Burns 1993;19:52-5. Hall LMC, Livermore DM, Gu D, Akova M, Akaln HE. OXA-11, an expended-spectrum variant of OXA-10 (PSE-2) -lactamase from Pseudomonas aeruginosa.Antimicrob Agent Chemother 1993;37:1637-44. Song W, Lee KM, Kang HJ, Shin DH, Kim DK. Microbiologic aspects of

Geyik et al
predominant bacteria isolated from the burn patients in Korea. Burns 2001;27:136-9. Broadhead JM, Parra DS, Skelton PA. Emerging multiresistant organism in the ICU: epidemiology, risk factors, surveillance, and prevention. Crit Care Nurs Q 2001;24(2):20-9. Ziolkowski G, Tomanek E, Sakiel S, Dudzinska G, Ekiel A, RogalaZawada D. Bacteriologic examinations of patients with burns. Med Dosw Mikrobiol 1993;45:233-6. Kolmos HJ, Thuesen B, Nielsen SV, Lohmann M, Kristoffersen K, Rosdahl VT. Outbreak of infection in a burn unit due to Pseudomonas aeruginosa originating from contaminated tubing used for irrigation of patients. J Hosp Infect 1993;24:11-21. Richard P, Le Floch R, Catherine C, Pannier M, Espaze E, Richet H. Pseudomonas aeruginosa outbreak in a burn unit: role of antimicrobials in the emergence of multiple resistant strains. J Infect Dis 1994;170:377-83. Lessava MI, Hadjiiski OG. Staphylococcal infections in the Sofia Burn Centre, Bulgaria. Burns 1996;22:279-82. Gupta M, Gupta OK,Yaduvansh RK, Upadhyahy J. Burn epidemiology: the Pink City scene. Burns 1993;19:47-51.

12. 13.

19.

20.

14.

21.

15. 16.

22.

17.

23. 24.

18.