Infections in Pediatric Burn Patients: An Analysis of One Hundred Eighty-One Patients

Abstract

Background:

Infectious complications are one of the most life-threatening complications and result in substantial mortality and morbidity in children who have been burned. The goal of the study is to assess the risk factors for sepsis in pediatric burn patients in a referral hospital.

Methods:

This study was performed at the Pediatric Burn Unit of Ankara Child Health and Diseases Hematology Oncology Training and Research Hospital during the period between January 2014 and June 2017. The patients were evaluated for age, sex, burn etiology, burned body surface area (BSA), the presence of inhalation injury, sepsis, positive cultures, the micro-organisms cultured samples, and septic focus.

Results:

A total of 181 patients were included in the study. The most common cause of burns was scalds in 120 patients (66.3%). Forty-one patients (22.7%) developed health-care–associated infection and sepsis. Gram-negative micro-organisms were isolated in 40 (97.6%) patients (Acinetobacter spp., Pseudomonas aeruginosa, Klebsiella pneumonia) with sepsis. Carbapenem resistance was detected in 31 (93.8%) of 40 patients. Mortality was observed in 11 patients (6.1%) in the group with sepsis. Burn surface area, burn depth, C-reactive protein (CRP) values, mortality, Garcés index, and Baux index were higher in the group with sepsis (p < 0.05). Multiple regression analysis revealed that mechanism of injury (flame), burned BSA ≥25%, C-reactive protein ≥6 mg/dL (area under the curve [AUC]: 0.76 p < 0.001 and AUC: 0.90, p < 0.001, respectively) at admission were independent parameters for development of sepsis in pediatric burn patients.

Conclusion:

Multi-drug–resistant Pseudomonas aeruginosa and Acinetobacter baumannii were important agents of blood stream infection in burned children. Burned BSA ≥25% and CRP ≥6 mg/dL were risk factors for developing sepsis in pediatric burn patients.

Treatment and care of patients who suffered extensive burns have improved greatly in recent decades. Advances in treatment of airway injury and resuscitation with hemodynamic stabilization resulted in improvement in the survival of burn patients [1]. Currently, infectious complications are one of the most life-threatening complications and cause substantial morbidity and mortality in burned patients [2,3]. Some studies found that infections lead to a mortality rate of 50%–75% in burned patients [4].

The skin is the first barrier of the human body and is the primary defense against pathogens. Loss of integrity of this organ enables micro-organisms to invade the body and causes fatal infections. Inflammatory mediators increase in burn patients when the wound remains open [5]. In children with burns and patients with burned total body surface area (TBSA) <30%, the incidence of infections increases rapidly [6].

Gram-positive bacteria including methicillin-resistant Staphylococcus aureus (MRSA) and gram-negative bacteria such as Pseudomonas aeruginosa, Acinetobacter baumannii, and Klebsiella species have been proven to be the causes of wound infections. Among increasingly reported bacteria and multi-drug–resistant bacteria, Acinetobacter baumannii and Pseudomonas aeruginosa have an important place because these bacteria quickly become multi-drug–resistant because of their intrinsic resistance mechanisms [7,8]. The goal of this study was to assess the risk factors for sepsis in children with burns in a referral medical center.

Patients and Methods

The study included the children who were followed in the intensive care burn unit (ICBU) of Ankara Child Health and Diseases Hematology Oncology Training and Research Hospital during the period between January 2014 and June 2017. This hospital is in 100-bed teaching and referral hospital in the capital of Turkey. It is also a referral center for other cities and centers in the central Anatolia region. This hospital includes an ICBU that is 10-bed unit providing care to pediatric burn patients. The burn unit has a separate surgery room inside. It has three isolated rooms and seven beds in the same room. The care of the burns includes the following steps: cleaning of the wound with normal saline and excision, escharatomy, occlusive dressing with chlorhexidine-soaked gauze, and sterile bandage. The dressings were changed every two days.

At the time of admission to our center, the wounds were swabbed. The depth of the wound and clinical parameters were evaluated. Urgent investigations including wound and blood cultures were taken at the onset of clinical and biologic signs of sepsis. The subclavian vein is the primary vessel for central vein catheterization (CVC). Wound cultures were obtained from the patients at every change of dressing. Blood cultures were taken in cases of fever. Antibiotic prophylaxis was not used in any patient as a part of routine care. Sulbactam ampicillin was started in patients with fever and no focus other than burn area. Administrative policies and care policies are evaluated periodically, and medical personnel are informed continuously. Necessary precautions such as hand washing and isolation of infected patients are mandatory steps for the burn unit.

The clinical features of the patients during hospitalization at the ICBU were recorded. This record includes data regarding demographics (name, age, sex, etc.) and clinical features such as type, depth, and TBSA (as percentage), history of hospitalization in another center, history of results of blood and wound cultures, treatment, and mortality, etc., of the patients.

Definitions

Colonization is defined as low concentration of bacteria on the wound surface without invasive infection (pathologic diagnosis: <10⁵ bacterial per gram of tissue [American Burn Association definition]. Invasive infection (sepsis) is defined as the presence of a pathogens in a burn wound at sufficient concentrations in relation to depth, surface area, and age of the patient leading to suppurative separation of eschar or graft loss, intrusion of adjacent non-burned tissue, or causing sepsis syndrome. Pediatric sepsis is defined as two or more systemic inflammatory response syndrome (SIRS) criteria (hypothermia >36°C or hyperthermia <38.5°C, leukopenia or leukocytosis for age, bradycardia or tachycardia for age, increased respiratory rate according to age, and decreased systolic blood pressure according to age) plus confirmed or suspected invasive infection. Severe sepsis is defined as sepsis plus one of the following criteria: cardiovascular dysfunction, acute respiratory distress syndrome, or two or more organ dysfunctions. Septic shock is defined as sepsis plus cardiovascular organ dysfunction [9].

The Garcés index is an index for mortality prediction and the formula is: 40 − age of patients + the percentage of burn body surfaces for 1 (superficial), for 2 (intermediate), or for 3 (full thickness). According to this index, 0–60 points is first degree (low risk), 61–90 points is second degree (moderate risk), 91–120 points is third degree (severe risk), ≥121 is fourth degree (critical) [10]. Baux score is the total of the percentage of BSA burned (deep burn) with age of the patient [11].

Microbiologic analysis

Bottles of blood culture were incubated in the BacT/ALERT 9240 systems (bioMérieux, Marcy l'Etoile, France) for seven days [12]. When the blood culture signaled positive, it was inoculated to agar plates (chocolate, eosin methylene blue lactose sucrose, and blood) and incubated at 37°C, 5% carbon dioxide for up to 48 hours. The identification of the micro-organisms were done with VITEK-2 compact system (bioMérieux) and antibiotic susceptibility tests (including minimum inhibitory concentration levels, presence of extended spectrum β-lactamase, and carbapenem resistance) were also performed with the same system for each isolate according to the manufacturer's instructions and the Clinical and Laboratory Standards Institute's criteria [13]. Identification and antibiotic susceptibility tests of gram-positive bacteria were performed using the automated VITEK-2 system with gram-positive identification card AST-P592, supplementary Etest (bioMérieux, Durham, NC) and a disk diffusion test according to the manufacturer's instructions. Vancomycin-resistant Enterococcus spp. (VRE) and MRSA were also identified using the automated VITEK-2 system [14]. The same system was also used for the identification and antibiotic susceptibility tests of gram-negative bacteria with gram-negative identification card AST-N325, AST-N326, and AST-N327 [15]. The yeast identification was performed using API 20C AUX (bioMérieux) [16].

Statistical analysis

Descriptive analysis methods were used to evaluate clinical parameters. Mean, median, standard deviation, and interquartile range (IQR) were calculated for numeric variables. Frequency tables were used to describe categorical data. Mann-Whitney U test or independent t-test was used to compare two independent samples. Probabilities (p value) less than 0.05 were considered significant for all tests. Multiple regression analysis was also performed to evaluate risk factors of sepsis. All data were analyzed using SPSS (version 21; IBM Corp, Armonk, NY).

Results

Patient characteristics

A total of 181 patients (77 girls, 104 boys) were included in the study between January 2014 and June 2017. At the time of hospitalization, the median age of the patients was 3 years (IQR; 1.7–6.1 years). Seventy-four percent of the cohort (n = 134) were younger than six years of age. Scalds were the most common etiology of burns and present in 120 patients (66.3%). Other causes were as follows: flame (n = 47; 26%) and electrical (n = 14; 7.7%). The mean burned BSA was 24.1 ± 14.7 of the TBSA. Among children, 53% had 1%–10% or 11%–20% BSA burned. In patient cohort, 115 patients (63.5%) had second-degree burns and 66 patients (36.5%) had third-degree burns. Compared with 80% of second-degree burns with scald, there is increased third-degree burns of electrical and flame (71.4% and 68.1%, respectively, p ≤ 0.001). Burned BSA was greater in flame burns than other types of burns (p ≤ 0.001). A total of 51 patients (28.2%) had been hospitalized in other centers for a median duration of three days (IQR: 2–11) and were then referred to our hospital. Other clinical features of the patients are provided in Table 1.

Table 1.

General Patient Characteristics

Number of patients	181
Male, n (%)/female, n (%)	104 (57.5)/77 (42.5)
Age (median, IQR, y)	3 (1.7–6.1)
0–12 mo, n	22
13–24 mo, n	36
25–72 mo, n	76
73–132 mo, n	27
133–180 mo, n	13
181–216 mo, n	7
Mechanism of injury
Scald, n (%)	120 (66.3)
Flame, n (%)	47 (26)
Electric, n (%)	14 (7.7)
Burn surface, BSA% (mean ± SD, %)	24.1 ± 14.7
1%–10%, n	31
11%–20%, n	65
21%–30%, n	42
31%–40%, n	26
41%–50%, n	3
51%–60%, n	11
61%–70%, n	2
>70%, n	1
Depth of the burn
Second-degree, n (%)	115 (63.5)
Third-degree, n (%)	66 (36.5)
Referred from other center	28.2%
Duration of hospitalization in another center, median (d, IQR)	3 (2–11)

IQR = interquartile range; BSA = body surface area; SD = standard deviation.

Additionally, for development of sepsis, receiver operating characteristic analysis showed that cutoff values for burned BSA and C-reactive protein (CRP) at admission were 25% and 6 mg/dL, respectively (AUC: 0.76 p < 0.001 and AUC: 0.90, p < 0.001). Multiple regression analysis revealed total BSA ≥25%, CRP ≥6 mg/dL and type of injury are independent risk factors for development of sepsis. Other parameters (age, history of hospitalization in another center, presence of colonization, etc.) are not statistically significant for development of sepsis (Table 2).

Table 2.

Multiple Regression Analysis of Risk Factors Affecting Sepsis Development

				95% CI
		p	OR	Lower	Upper
Step 2	Gender	0.06	3.31	0.94	11.57
	Age at admission	0.10	0.88	0.76	1.02
	Burned area of ≥25% of TBSA	<0.001	48.88	11.19	213.59
	Mechanism of injury (Ref: scald)	0.011
	Electricity	0.004	24.20	2.72	215.56
	Flame	0.054	3.40	0.98	11.49
	CRP at admission ≥6mg/dl	<0.001	10.76	3.27	35.51

CI = confidence interval; OR = odds ratio; TBSA = total body surface area; CRP = C-reactive protein.

In our burn patients, burn BSA, burn depth, CRP values, mortality, Garcés index, and Baux index were higher in the sepsis group. The rate of sepsis was substantially higher in patients with flame burns (p < 0.05; Table 3).

Table 3.

Comparison of the General Patient Characteristics with Sepsis and without Sepsis

	With sepsis	Without sepsis	p
Patient no.	41	140
Male (%)/female (%)	28 (68.3)/13 (31.7)	76 (54.3)/64 (45.7)	0.11
Age (mean ± SD, y)	5.3 ± 4.3	4.4 ± 4.0	0.91
0–12 mo, n	2	20
13–24 mo, n	6	30
25–72 mo, n	18	58
73–132 mo, n	8	19
133–180 mo, n	7	6
181–216 mo, n		7
Mechanism of injury			<0.001
Scald, n (%)	16 (39)	104 (74.3)
Flame, n (%)	20 (48.8)	27 (19.3)
Electrical, n (%)	5 (12.2)	9 (6.4)
TBSA burned (mean ± SD, %)	41.5 ± 16.1	19 ± 9.5	<0.001
1%–10%, n	1	30
11%–20%, n	2	63
21%–30%, n	9	33
31%–40%, n	14	12
41%–50%, n	2	1
51%–60%, n	10	1
61%–70%, n	2
>70%, n	1
Depth of the burn			<0.001
Second-degree, n (%)	12 (29.3)	103 (73.6)
Third-degree, n (%)	29 (70.7)	37 (26.4)
CRP at admission (mg/dL)	9.2 ± 7.4	3.2 ± 5.4	<0.001
Mortality, n (%)	11 (26.8)	0	<0.001
Garces index	76.2 ± 15.8	54.7 ± 9.9	<0.001
Baux index	46.7 ± 17.6	23.4 ± 10.7	<0.001

SD = standard deviation; TBSA = total body surface area; CRP = C-reactive protein.

Infection and colonization data

In total cohort, 41 patients (22.7%) developed health-care–associated infection and sepsis, 49 (27%) demonstrated colonization, and 91 (50.3%) were without infection or colonization. In the colonization group, gram-negative micro-organisms were present in 34 (69.5%) patients and gram-positive micro-organisms were present in 15 (30.5%) patients. In the colonization group, the three most common micro-organisms on wound cultures were Acinetobacter spp. (15 patients, 27.8%), Pseudomonas aeruginosa (9 patients, 16.7%) and coagulase-negative staphylococci (8 patients, 14.8%). The micro-organisms were isolated at a mean period of 11.3 ± 6 days (minimum, 1 day; maximum, 48 days) of hospitalization.

In the sepsis group, gram-negative micro-organisms were present in 40 (97.6%) patients and gram-positive micro-organism (MRSA) was present in one (2.4%) patient. In the sepsis group (n = 41), most common micro-organisms on blood cultures were Acinetobacter spp. (17 patients, 41.5%), Pseudomonas aeruginosa (13 patients, 31.7%) and Klebsiella pneumoniae (2 patients, 4.9%). None of the micro-organisms were detected in seven patients. In the sepsis group, most of the micro-organisms were carbapenem resistant (n = 31; 93.8%).

Micro-organisms were isolated at a median period of seven days (IQR: 4–14 days) of the hospitalization. In the sepsis group, 22 patients (53.7%) were referred from another center. These 22 patients had history of hospitalization for four days (IQR: 1–20 days). A comparison of the clinical features of the patients with sepsis and non-sepsis are provided in Table 3.

In the sepsis group, all patients with Acinetobacter spp. (n = 17) and all patients with Pseudomonas aeruginosa except one (n = 12) is carbapenem resistant (n = 29). In five of 29 patients (one patient with Pseudomonas aeruginosa and four with Acinetobacter spp.), sepsis was controlled with continuation of carbapenem and none of them died. One patient with Pseudomonas aeruginosa died at the third day of hospitalization without getting culture results and carbapenem continued in this patient. For the remaining patients (n = 23; 10 patients with Pseudomonas aeruginosa and 13 patients with Acinetobacter spp.) colistin, tigecycline, or meropenem infusion were used. The treatment modalities of this group were given in Table 4.

Table 4.

Additional Treatment Modalities of Patients with Carbapenem-Resistant Pseudomonas aeruginosa and Acinetobacter spp. (n = 23)

	Pseudomonas aeuruginosa	Acinetobacter spp.
	(n = 10)	(n = 13)
Colistin + meropenem infusion, n	8	8
Colistin + meropenem infusion + tigecycline, n	2	5
Mortality, n	6	3

Prognosis

In our center, median duration of hospitalization was 19 days (IQR: 10–30 days) and correlated with depth of burn (p ≤ 0.001 and r = 0.389) and BSA burned (p ≤ 0.001 and r = 0.374). Overall mortality was observed in 11 patients (6.1%) in the entire cohort and all of these patients belong to the group with sepsis (Pseudomonas aeruginosa, seven patients; Acinetobacter spp., three patients, and negative blood culture, one patient). The major reason of mortality was sepsis and multiple organ failure (Table 5).

Table 5.

Patient Outcome

Length of hospital stay (median, d)	19 (IQR:10–30)
Admission outcome
Recovery, n	170
Death, n	11
Mortality by % BSA burned	11
1%–10%, n	0
11%–20%, n	1
21%–30%, n	3
31%–40%, n	2
41%–50%, n	0
51%–60%, n	3
61%–70%, n	1
>70%, n	1
Mortality by etiology of burn
Scald, n	5
Flame, n	6
Mortality by depth of burn
Second-degree, n	3
Third-degree, n	8

IQR = interquartile range; BSA = body surface area.

Discussion

Burn injuries are community health problems throughout the world including developing countries. The data regarding the burn injuries were based mainly on adult studies. Our study aimed to investigate the clinical features of pediatric burn patients in a referral hospital.

Other pediatric series reported that most of the children were younger than five years of age [1,7,17 –20]. The majority of patients in our study was also younger than six years of age. Age stratification of other pediatric series reveal different age groups. In the study by Belba et al. [17], infantile age group (<2 years of age) constituted approximately 9% of the pediatric age group. The study by Elsous et al. [20], infantile age group constituted 36.3% of the cohort. Similarly, in our series the infantile age group constitutes 32% of the patients.

Another important factor that affects the burn size is the type of the burn. Flame injuries are usually deeper and cause more damage to the surrounding tissue [21]. In our study, burn depth is higher in flame and electrical burns than scalds, and burned BSA is higher in flame burns compared with other types of burns.

After resuscitation and fluid management in the acute period, infections cause approximately 75% of mortality in burn patients [22]. Sepsis that develops in the burn patient is unlike the sepsis that develops in other situations. Burn patients lose their skin, which is the primary barrier against micro-organisms. The risk of sepsis is increased when their wound healing is delayed. Once the burned area reaches 15% of total BSA, cytokine storm begins in the organism [5]. Rogers et al. [23] showed that full-thickness burns (TBSA >30%) and burn patients with flame and inhalation injury had higher risk of infectious complications. Belba et al. [17] found infectious complications are higher in burns with >40% and colonization is noticed in patients with burns of 0%–20% BSA. In the study by Dermijian et al. [21], burn surface >40% is the independent risk factor for mortality. Additionally, burned surface area and depth of the lesion is also important for the development of infectious complications. Mechanism of the injury also affects the severity of the burn lesion. Severity (extent and depth) of the burns is higher in flame and electrical burns [25]. Among causes other than scalding, flame injuries are deeper and lead to more destruction of the surrounding tissues. This may facilitate colonization of micro- organisms. In our study, patients in the group with sepsis have higher mean burned total BSA. The group with sepsis also has more patients with third-degree burns.

The percentage of the patients with flame and electrical burns are higher in the group with sepsis than in non-sepsis group. In addition, multiple regression analysis showed that burned area >25% of TBSA, mechanism of injury, and CRP >6 mg/dL at admission were risk factors affecting development of sepsis.

Despite considerable improvements in initial management and fluid resuscitation, infections are the major complication in burned children [19]. Conditions that lead to the development of health-related infections in burn patients are the immune-suppressing effect of burns, long-term hospitalization, multiple diagnosis, and treatment applications that are challenging for the burn team [17]. Belba et al. [17] investigated health-care–associated infections in 181 burn patients. In their study, 12% and 44% of the patients developed health-care–associated infections and colonization, respectively. Other studies demonstrated that the health-care–associated infection prevalence rate was 20.6% [27,28]. Similar to these results, in our study, the prevalence of sepsis was 22.6%.

Studies have reported differences in distribution of causative agents. Rosanova et al. [29] studied 110 burn patients and showed that multi-resistant Pseudomonas aeruginosa and Acinetobacter baumannii were the most common microorganisms. Ronat et al. [30] stated that Pseudomonas aeruginosa was the major micro-organism that lead to wound infection and bacteremia. In other studies, gram-positive agents were reported as the most common micro-organisms. Gastmeier et al. [31] reported that the most common pathogen was Staphylococcus aureus, which is isolated in 20.7% of the patients. Devrim et al. [7] studied 206 patients and showed that gram-positive organisms constitute 66.4% of micro-organisms. These differences may be caused by isolation of micro-organisms in different time intervals. Gram-positive micro-organisms (Streptococcus pyogenes and Staphylococcus aureus) colonize burn wounds in one to two days [32,33]. Therefore, gram-positive bacteria are more probable to develop sepsis in early periods of follow-up. Lee et al. [34] showed that infections in gram-positive bacteria developed in initial periods of burn, which are replaced by gram-negative micro-organisms in later periods.

Multi-drug–resistant gram-negative bacteria-related infections are an serious issue in hospitals. Goverman et al. [35] reported 14 children over an 11-year period. In their study, response rate was 78.6% and mortality was 14.3% [35]. In our study, colistin was used either alone or in addition meropenem infusion and tigecycline. Its success is more predominant in Acinetobacter spp.

Risk factors for burn-related infections have been investigated. Patients with previous meropenem treatment, gram-negative colonization on admission, escharotomy, and superficial partial thickness burn are at increased risk for developing multi-drug–resistant gram-negative infection [36]. In another adult study, TBSA ≥10% and abbreviated burn severity index (ABSI) ≥3 were associated with a substantially higher incidence of blood stream infections caused by Pseudomonas sp., Enterococcus sp., and Candida sp. [37]. In our study total BSA ≥25% was found to be a risk factor for developing sepsis in pediatric burn patients.

We acknowledge the limitations of our study. First, our study was based on retrospective medical records. Although our center is a referral burn unit, the results were generated using data from a single hospital. Additionally, our study included patients followed between 2014 and 2017. Therefore, the comparison of burn infections with before 2014 could not be evaluated. Validation of risk factors for burn infection in a multi-center study is needed.

In conclusion, gram-negative bacteria such as multi-drug–resistant Pseudomonas aeruginosa and Acinetobacter baumannii were important agents of blood stream infections in patients with burns. Burned BSA ≥25% and CRP ≥6 mg/dL were risk factors for developing sepsis in pediatric burn patients.

Footnotes

Funding Information

No funding was received.

Author Disclosure Statement

No competing financial interests exist.

References

Santucci

, Gobara

, Santos

, et al. Infections in a burn intensive care unit. J Hosp Infect, 2003; 53:6–13.

Appelgren

, Björnhagen

, Bragderyd

, et al. A prospective study of infections in burn patients. Burns, 2002; 28:39–46.

Benmeir

, Sagi

, Greber

, et al. An analysis of mortality in patients with burns covering 40 per cent BSA or more: A retrospective review covering 24 years (1964–88). Burns, 1991; 17:402–405.

Luterman

, Dacso

, Curreri

. Infections in burn patients. Am J Med, 1986; 81:45–52

Greenhalgh

DG.

Sepsis in the burn patient: A different problem than sepsis in the general population. Burns Trauma, 2017; 5:23.

Ramirez-Blanco

, Ramirez-Rivero

, Diaz-Martinez

, et al. Infection in burn patients in a referral center in Colombia. Burns, 2017; 43:642–653.

Devrim İ, Kara

, Düzgöl

et al. Burn-associated bloodstream infections in pediatric burn patients: Time distribution of etiologic agents. Burns, 2017; 43:144–148.

Potron

, Poirel

, Nordmann

. Emerging broad-spectrum resistance in Pseudomonas aeruginosa and Acinetobacter baumannii: Mechanisms and epidemiology. Int J Antimicrob Agents, 2015; 45:568–585.

Goldstein

, Giroir

, Randolph A; International Consensus Conference on Pediatric

Sepsis

. International pediatric sepsis consensus conference: Definitions for sepsis and organ dysfunction in pediatrics. Pediatr Crit Care Med, 2005; 6:2–8.

10.

Garcés

, Tapia

, Hoecher

, et al. Clasificación y pronósticode los quemados. Asistencia Públ, 1971; 1:5–9.

11.

Dahal

, Ghimire

, Maharjan

, et al. Baux's and abbreviated burn severity score for the prediction of mortality in patients with acute burn injury. JCMS Nepal, 2015; 11:24–27.

12.

Wilson

, Weinstein

, Reller

. Automated blood culture systems. Clin Lab Med, 1994; 14:149–169.

13.

Clinical and Laboratory Standards Institute. M100-S25 Performance Standards for Antimicrobial Susceptibility Testing; Twentieth Informational Supplement. Wayne, PA: Clinical and Laboratory Standards Institute; 2015.

14.

Bobenchik

, Hindler

, Giltner

, et al. Performance of Vitek 2 for antimicrobial susceptibility testing of Staphylococcus spp. and Enterococcus spp. J Clin Microbiol, 2014; 52:392–397.

15.

Quesada

, Giménez

, Molinos

, et al. Performance of VITEK-2 Compact and overnight MicroScan panels for direct identification and susceptibility testing of Gram-negative bacilli from positive FAN BacT/ALERT blood culture bottles. Clin Microbiol Infect, 2010; 16:137–140.

16.

Valenza

, Strasen

, Schäfer

, et al. Evaluation of new colorimetric vitek 2 yeast identification card by use of different source media. J Clin Microbiol, 2008; 46:3784–3787.

17.

Belba

, Petrela

, Belba

. Epidemiology of infections in a burn unit, Albania. Burns, 2013; 39:1456–1467.

18.

Santucci

, Gobara

, Santos

, et al. Infections in a burn intensive care unit: Experience of seven years. J Hosp Infect, 2003; 53:6–13.

19.

Hassen

, Khalifa

, Daiki

. Epidemiological and bacteriological profiles in children with burns. Burns, 2014; 40:1040–1045.

20.

Elsous

, Salah

, Ouda

. Childhood burns: An analysis of 124 admissions in the Gaza Strip. Ann Burns Fire Disasters, 2015; 28:253–258.

21.

Alp

, Coruh

, Gunay

, et al. Risk factors for nosocomial infection and mortality in burn patients: 10 years of experience at a university hospital. J Burn Care Res, 2012; 33:379–385.

22.

Ansermino

, Hemsley

. Intensive care management and control of infection. BMJ, 2004; 329:220–223.

23.

Rodgers

, Mortensen

, Fisher

, et al. Predictors of infectious complications after burn injuries in children. Pediatr Infect Dis J, 2000; 19:990–995.

24.

Dermijian

Adjusting a prognostic score for burned children with logistic regression. J Burn Care Rehabil, 1997; 18:313–316.

25.

Tekin

, Yolbaş

, Selçuk

, et al. An evaluation of pediatric burn patients over a 15-year period. Ulus Travma Acil Cerrahi Dergisi, 2012; 18:514–518.

26.

Nagesha

, Shenoy

, Chandrashekar

. Study of burn sepsis with special reference to Pseudomonas aeruginosa. J Indian Med Assoc, 1996; 94:230–233.

27.

Nosocomial Infections National Surveillance Service (NINSS). Surveillance of hospital acquired bacteraemia in English hospitals 1997–2002. London: Public Health Laboratory Service; 2002.

28.

Louis

, Bihari

, Suter

. The prevalence of nosocomial infections in intrensive care units in Europe. European Prevalence of infection in Intensive care (EPIC) study. JAMA, 1995; 274:639–644.

29.

Rosanova

, Stamboulian

, Lede

Risk factors for mortality in burn children.Braz J Infect Dis 2014;18:144–9

30.

Ronat

, Kakol

, Khoury

, et al. Highly drug-resistant pathogens implicated in burn-associated bacteremia in an Iraqi burn care unit. PLoS One, 2014; 9:e101017.

31.

Gastmeier

, Weigt

, Sohr

, et al. Comparison of hospital-acquired infection rates in paediatric burn patients. J Hosp Infect, 2002; 52:161–165.

32.

Erol

, Altoparlak

, Akcay

, et al. Changes of microbial flora and wound colonization in burned patients. Burns, 2004; 30:357–361.

33.

Barret

, Herndon

. Effects of burn wound excision on bacterial colonization and invasion. Plast Reconstr Surg, 2003; 111:744–750

34.

Lee

, Jang

, Choi

, et al. Blood stream infections in patients in the burn intensive care unit. Infect Chemother, 2013; 45:194–201.

35.

Goverman

, Weber

, Keaney

, Sheridan

. Intravenous colistin for the treatment of multi-drug resistant, gram-negative infection in the pediatric burn population. J Burn Care Res, 2007; 28:421–426.

36.

Vickers

, Dulhunty

, Ballard

, et al. Risk factors for multidrug-resistant Gram-negative infection in burn patients. ANZ J Surg, 2018; 88:480–485.

37.

Fochtmann-Frana

, Freystätter

, Vorstandlechner

, et al. Incidence of risk factors for bloodstream infections in patients with major burns receiving intensive care: A retrospective single-center cohort study. Burns, 2018; 44:784–792.