Risk Factors for Nosocomial Infection and Mortality

in Burn Patients: 10 Years of Experience at a

University Hospital
Emine Alp, MD, PhD,* Atilla Coruh, Galip K. Gunay, Yalcin Yontar,
Mehmet Doganay*

To evaluate the risk factors for nosocomial infection (NI) and mortality in a university hospital, 10-year data of burn patients were assessed retrospectively. The study was conducted
at Erciyes Universitys Burn Center during 2000 and 2009. The records of 1190 patients
were obtained. Overall, 131 (11%) patients had 206 NIs with an incidence density of 14.7
infections/1000 patient days. Burn wound infection (n 109, 53%) was the most common
NI. High (%TBSA burned) and late excision were found to be the most significant risk factors for the development of NI. Pseudomonas aeruginosa was the most frequent causative
microorganism. However, the prevalence of multidrug-resistant Acinetobacter baumannii has
increased in recent years with a prevalence of 47% in 2009. The carbapenem resistance of P.
aeruginosa has decreased in recent years, whereas that of A. baumannii increased and it had a
prevalence of 94% in the last year. Conversely, the most important risk factors for mortality
were advanced age, high %TBSA and having an underlying disease. Prevention of NI is an important issue in burn units to reduce mortality rates. Early excision and wound closure are important therapeutic approaches for the prevention of burn wound infection. (J Burn Care Res
2011;XX:000 000)

Despite advances in burn care, nosocomial infections (NIs) are still serious complications in burn
patients. Seriously burned patients have an increased risk for NI because of the nature of the burn
injury itself, and NI is the most common cause of
death following burns.1 4
To survey and assess the risk factors for NIs, to
evaluate the antibiograms of causative pathogens, and
to inform clinicians of the results are important
themes for infection control and reducing the incidence of resistant pathogens. Also, the evaluation of
risk factors for mortality will help to increase the quality of care in burn centers. In this study, we evaluated
the 10-year records of burn center patients and identified the demographic characteristics, incidence of

From the *Department of Infectious Diseases and Clinical

Microbiology, Erciyes University; and Department of Plastic
and Reconstructive Surgery, Burns Unit, Faculty of Medicine,
Erciyes University, Kayseri, Turkey.
Address correspondence to Emine Alp, MD, PhD, Department of
Infectious Diseases, Faculty of Medicine, Erciyes University,
38039 Kayseri, Turkey.
Copyright 2011 by the American Burn Association.
1559-047X/2011
DOI: 10.1097/BCR.0b013e318234966c

NIs, risk factors, causative pathogens, their antibiotic

resistance, and the outcome of patients.

MATERIALS AND METHODS

Study Design
The study was conducted at Erciyes Universitys Burn
Center from 2000 to 2009. Patient data were obtained from the Burn Centers specially designed
computer system (File Maker Pro Ver 10.0) and the
surveillance reports of the Infection Control Committee. Infection Control Committee has performed
active, prospective, and patient-based surveillance in
Burn Center since 1997. Patients age, sex, type of
injury, date of accident, time of admission, location of
injury, the size of the burn wound are (TBSA%) estimated by using the Wallaces rule of nines5 method
and a more accurate assessment is performed especially in children, by using the Lund and Browder
chart,6 the abbreviated burn severity index (ABSI),7
underlying diseases (diabetes mellitus, chronic renal
failure, congestive heart failure, hypertension, and
malignancy), blood transfusion, trauma, treatment
mortality, NI, causative pathogens, antimicrobial re1

Journal of Burn Care & Research

XXX 2011

2 Alp et al

sistance, and outcome of the patients were recorded.

If the patient was exposed to smoke during flame
injury, it was defined as inhalation injury.

Center Design and Patient Care

Erciyes University Hospital is a referral, tertiary care center in the Central Anatolian region of Turkey with 1300
beds, serving approximately 5 million people. The Burn
Center is designed separately from the other clinics with
10 single rooms and 1 double room. The four patient
rooms and nurses room have their own sink and toilet,
and there is also a sink in the dressing room.
Alcohol-based solutions are available in all patient
rooms and in the dressing room. The nurse to patient
ratio was 7:1 in every shift. Infection Control Committee provide continuous education program to healthcare personnel and also provide reminder posters in the
clinic about infection control measures to reduce infection rates. Constant surveillance of infection control
measures was not performed.
Intravenous fluids are given until a patient becomes
stable and is able to tolerate enteral feeding. Dressings
are changed daily by resident doctors and anesthesiologists after hand disinfection and taking all sterile precautions (sterile glove, gown, and mask). Early excision and
debridement is performed between 3 and 5 days postburn in full-thickness and extensive deep dermal burns
when the patients condition permits. Bactigras
(chlorhexidine-impregnated paraffin gauze dressing) is
applied on superficial partial-thickness burns, and 1%
silver sulfadiazine cream is applied on deep partial- and
full-thickness burns. Then, burn wounds are covered
with sterile pads and bandages. Split-thickness skin
grafting is done under antibiotic cover. In direct patient contact, a protective gown and disposable gloves
are used. Hands are washed with medicated soap (4%
chlorhexidine) when necessary and disinfected with
an alcohol-based solution before and after patient
contact. If needed, nonimmersion wound care is used
for wound care instead of hydrotherapy tank.
When indicated, antibiotic therapy was initiated by
the infectious disease specialist according to the previous antibiotic susceptibility pattern of the center
and reevaluated according to antibiotic susceptibility
results. No systemic prophylactic antibiotics were
used in our burn center except in the perioperative
period of debridement and auto skin grafting.

Data Collection
A designated registered nurse from the hospitals Infection Control Committee regularly assessed all patients and recorded NIs according to the Centers for
Disease and Control definitions.8 Device-associated
NI surveillance was begun after 2003. Microbial cul-

tures were processed according to current methods.9

Swab cultures were used to determine the causative organism(s) in burn wound infection and obtained when
there were clinical signs and symptoms of wound infection. Burn wound infection was diagnosed according to
local signs (pain, tenderness, swelling, purulent exudates, etc.) of infection with or without systemic signs
(fever, tachycardia, leukocytosis, etc.) of infection.2 The
bacteriological isolation and antibiotic susceptibility
tests were carried out in the microbiology laboratory of
Erciyes University. National Committee for Clinical
Laboratory Standards criteria were used for the antibiotic susceptibility tests.10

Statistical Analysis
Chi-square tests using Yates correction or the
Fishers exact test were performed to determine the
significant differences in proportions among categorical
variables. For continuous variables, a Mann-Whitney U
test was used. Univariate and bivariate analysis regarding
NI and mortality were adjusted for length of hospital
stay. Odds ratios and 95% confidence intervals were
calculated using binary logistic regression for each
model. Two-tailed P values .05 were considered to
be significant. Statistical analysis was calculated using
SPSS version 13.0 (SPSS, Chicago, IL).

RESULTS
Characteristics of Patients
During the 10-year period, 1558 patients were admitted to the Burn Center; however, only 1190
patients records were obtained. Demographic characteristics of the patients and a comparison of infected
and noninfected patients characteristics are shown in
Table 1. Seven hundred ninety-four were male
(66.8%) and 396 (33.2%) were female. The median
age of the patients was 10 years, and 684 (57%) patients were younger than 16 years. One hundred
twenty-four patients (10%) had underlying diseases.
The median %TBSA was 12.0% (range 0 99%), and
the median ABSI was 4.0 (range 114). The cause of
injury was recorded in 1171 patients. Six hundred
forty-nine patients (55.4%) had scald injury, 310
(26.5%) had flame injury, 149 (12.7%) had electrical
injury, and 63 (5.4%) had contact injury. Sixteen patients had inhalation injury attendant with flame injury. Scalding was mostly (559, 87%) seen in patients
younger than 16 years.
Of 976 patients, 774 (79.3%) were admitted on the
day of injury and 202 (20.7%) had been treated in
another hospital before admission. Unfortunately,
data about infection before admission could not be

Journal of Burn Care & Research

Volume XX, Number XX

Alp et al

Table 1. The characteristics of patients and comparison between infected and uninfected patients
Total
Patient, n (%)
Sex (male/female)
Underlying disease
Age (yr), median (2575 percentile)
Injuries
Scald
Flame
Electrical
Contact
%TBSA, median (2575 percentile) total
ABSI
Admission on the same day of injury, n (%)
Trauma, n (%)
First excision day
Length of stay
Transfusion, n (%)
Mortality, n (%)

Uninfected

1190 (100)
794/396
124/1190
10 (330)

1056 (89)
705/351
104/124
8 (2.530)

649
310
149
63
12 (720)
4 (35)
774/976 (79)
27/1190 (2)
3 (08)
12 (720)
524 (44)
79 (7)

603 (93%)
245 (79%)
133 (89%)
57 (90%)
10.75 (618)
4 (35)
652/774 (84)
24/27 (89)
2 (08)
11 (717)
421 (40)
60 (6)

Infected

134 (11)
89/45
20/124
17 (432)

.940
.001
.020

46 (7%)
65 (21%)
16 (11%)
6 (10%)
26.25 (17.7532)
6 (47)
122/774 (16)
3/27 (11)
8 (511)
31 (2145)
103 (77)
19 (14)

.001

.001
.001
.001
1.000
.001
.001
.001
.001

ABSI, abbreviated burn severity index.

obtained. The median length of stay in hospital was 12

days (range: 7 45 days) and 1129 (94.8%) patients
stayed 48 hours. The median length of stay was approximately three times higher in infected patients
(31.0 days) than in uninfected patients (11.0 days).
Of the 1190 patients, 79 (6.6%) died during their
stay in the burns center, 1071 (90%) were treated,
and 40 (3.4%) patients were discharged at their own
request. The mortality rate was approximately three
times higher in infected patients (14.2%) compared
with noninfected patients (5.7%; Table 1).

Incidence of Infection
The incidence of infection in our study has changed
over the years. However, overall 131 (11%) patients had 212 NIs with an incidence density of
14.7 infections/1000 patient days. The median
length for NI development was 12.00 14.30
days. Burn wound infection (n 119, 56%) was the
most frequent NI, and nosocomial urinary tract
infection (n 37, 17.5%) and bloodstream infections (n 36, 17%) were the other frequent infections found (Table 2).

Table 2. NIs during 10 years

2000

2001

2002

NI episodes (n)
Burn wound infection
18
4
2
Urinary tract infection
1
1
4
CAUTIs/1000 catheter Not surveyed Not surveyed Not surveyed
days
Bloodstream infection
5
3
2
Central catheter infection
3
2

CLABSIs/1000 central
Not surveyed Not surveyed Not surveyed
line days
Pneumonia

VAP/1000 ventilator
Not surveyed Not surveyed Not surveyed
days
Incidence density*
16
5
4

2003 2004 2005 2006 2007 2008 2009

Total,
n (%)

10
4
8

10

2.5

12
1
4

17
12
12

18
9
9

21
5
4

119 (56)
37 (18)

3
1
8

4
1
2

5
4

4
1

36 (17)
12 (6)

3
25

3
39

1
22

8 (4)

25

16

13

23

19

17

15

* Nosocomial infection episode/1000 patient days.

NI, nosocomial infection; CAUTI, catheter-associated urinary tract infection; CLABSI, central line-associated bloodstream infection.

Journal of Burn Care & Research

XXX 2011

4 Alp et al

Risk Factors for Infection

In univariate analysis, flame injury, high %TBSA and
ABSI, admission on the same day as the burn injury,
delayed burn wound excision, and blood transfusion
were found as risk factors for the NI. Patients admitted on the same day as the burn injury had higher
%TBSA (18.30 16.08 and 9.92 10.39, P .001)
and ABSI (4.31 2.15 and 3.80 1.57, P .001).
Also, patients who had late excision had more transfusions. In the stepwise forward logistic regression
undertaken to control, the effect of confounding variables, high %TBSA, and delayed burn wound excision
were found to be the most significant risk factors for
the development of NI (Table 3).

Microorganisms and Antibiotic Resistance

Pseudomonas aeruginosa was the most frequent causative microorganism. However, the prevalence of
Acinetobacter baumannii has increased in recent
years with a prevalence of 47% in 2009. The prevalence of Staphylococcus aureus and Candida spp. was
low in our study (Figure 1). All P. aeruginosa and A.
baumannii isolates were multidrug resistant. More-

over, the carbapenem resistance of these frequently

found microorganisms has changed over the past
years. The carbapenem resistance of P. aeruginosa has
decreased in recent years, whereas that A. baumannii
has increased with a prevalence of 94% in the past year
(Figure 2).

Risk Factors for Mortality

Advanced age, flame injury, high %TBSA and ABSI,
admission after 24 hours of burn injury, having an
underlying disease, and NI were found as risk factors
for mortality in univariate analysis. In the stepwise
forward logistic regression undertaken to control the
effect of confounding variables, advanced age, high
%TBSA, and having an underlying disease were found
to be the most significant risk factors for mortality
(Table 4).

DISCUSSION
To the best of our knowledge, this study has the
largest study population in a burn center in EuroAsia.
During the 10 years of our study, 1190 patients were

Table 3. Risk factors for development of NI

Univariate Analysis

Sex
Male
Female
Age
Underlying disease
No
Yes
Injury
Scald
Flame
Electrical
Contact
%TBSA
ABSI*
Admission day
24 hr
24 hr
Trauma
No
Yes
First excision day
Transfusion
No
Yes

Multiple Analysis Model

Odds Ratio

95% CI

1
1.02
1.01

0.691.49
0.991.01

.94
.163

1
1.61

0.962.69

.07

1
3.48
1.58
1.38
1.05
1.44

2.325.22
0.872.87
0.573.37
1.041.06
1.331.56

.001
.14
.48
.001
.001

1
0.11

0.040.30

.001

1
0.99
1.14

0.293.32
1.101.18

.98
.001

1
5.01

3.297.63

.001

* Because ABSI included age and TBSA, ABSI was not included in multivariate analysis.
NI, nosocomial infection; ABSI, abbreviated burn severity index; CI, confidence interval.

Odds Ratio

95% CI

1.05

1.041.06

.001

1.13

1.091.17

.001

Journal of Burn Care & Research

Volume XX, Number XX

Alp et al

Figure 1. Distribution of microorganisms during 10 years.

evaluated, and the incidence of NI was 11% and incidence density was 14.7/1000 patient days. The infection rates were lower than the other studies.1113
Despite routine infection control activities in our hospital, the infection rates did not decrease during these
period. The reasons for this are unclear. One possible
explanation is lack of compliance with infection control practices.14 Despite the bedside alcohol-based
solutions and infection control activities (education,
reminders, etc.) practiced in our hospital, hand hygiene adherence is still low (50%) (Infection Control Committee observational study, unpublished
data). The other explanation may be increasing sensitivity to infection definitions by infection control
team over time. Burn wound infection was the most
frequent (53%) NI, as shown in the literature.4 Although ventilator-associated pneumonia rates were
high in some years in our study. On the other side, no
infection was detected in other years. This is probably
due to patient characteristics and device usage rates.
In years with no infection, ventilator days were very
low, and also few patients were ventilated than the
other years (210 ventilator days vs 40 49 ventilator

Figure 2. Carbapenem resistance of Pseudomonas aeruginosa and Acinetobacter baumannii during 10 years.

days). In our study, larger BSAs were found as the

most significant risk factors for NI.
Our university hospital is a reference hospital in our
region, and severe burns are referred early to our burn
center. Patients admitted on the first day of burn
injury had higher %TBSA and ABSI, and so in univariate analysis of admittance on the first day of burn
injury was found as a risk factor. However, because of
this confounding effect, it was excluded in multivariate analysis. In addition, a confounding effect was
found between transfusion and late excision, because
patients who have delayed for burn wound excision
need more blood transfusions.15
In addition, delayed burn wound excision was
found as the most significant risk factor for NI. Early
surgical excision and temporary or permanent closure
of the burn wound are important for prevention of
infection and decreasing the stimulus of overwhelming systemic inflammatory response. Burn wounds
are sterile immediately after thermal injury; however,
wounds later become colonized with microorganisms. Immediately after injury, gram-positive bacteria
or sensitive gram-negative bacteria from endogenous
flora colonize open skin wounds. However, the use of
broad-spectrum antibiotics effective against Staphylococcus results in the emergence of resistant gramnegative pathogens, particularly P. aeruginosa.16 19
In our burn center, P. aeruginosa and A. baumannii
constituted 72% of all microorganisms isolated from
NIs, and in recent years, carbapenem-resistant A.
baumannii was the predominant pathogen (47%).
Antibiotic utilization affects the flora of the hospital
environment and the antibiotic susceptibility of microorganisms. Many reports have shown significant
relationships between antibiotic consumption and resistance.20,21 In our burn center, an increased usage
of carbapenem for NIs because of highly resistant P.

Journal of Burn Care & Research

XXX 2011

6 Alp et al

Table 4. Risk factors for mortality (adjusted for length of hospital stay)
Univariate Analysis

Sex
Male
Female
Age
Injuries
Scald
Flame
Electrical
Contact
%TBSA
ABSI*
Admission day
24 hr
24 hr
Underlying disease
No
Yes
Trauma
No
Yes
First excision day
Transfusion
No
Yes
Infection
No
Yes

Multiple Analysis

Odds Ratio

95% CI

Odds Ratio

95% CI

1
1.34
1.02

0.842.14
1.011.03

.226
.001

1.02

1.011.03

.006

1
4.18
1.18
0.84
1.07
1.88

2.506.99
0.354.04
0.322.23
1.061.09
1.672.12

.001
.787
.727
.001
.001

1.07

1.061.08

.001

1
2.11

0.944.73

.071

1
2.48

1.394.4

.002

1
2.29

1.124.69

.023

1
2.45
0.92

0.837.28
0.870.98

.107
.007

1
0.91

0.561.47

0.689

1
3.03

1.555.94

0.001

* Because ABSI included age and TBSA, ABSI was not included in multivariate analysis.
ABSI, abbreviated burn severity index; CI, confidence interval.

aeruginosa probably induced carbepenem-resistant

A. baumannii, whereas extensive use of colistin
thereafter may have increased the sensitivity of P.
aeruginosa to carbapenems.22,23
The survival rates of burn patients have improved
in recent years because of advances in modern medical care, advances in fluid resuscitation, nutritional
support, pulmonary care, burn wound care, and infection control practice. However, infection complications are still the leading cause of death in burn
patients. In patients with more than 40 TBSA%, 75%
of all deaths were related to burn wound infection or
other infection complications.2 Despite the fact that
NI was not the most important risk factor for mortality in our study, infected patients had approximately three times higher mortality rate than uninfected patients. Moreover, advanced age, underlying
diseases, and higher %TBSA were the most significant
risk factors for mortality in the study. In univariate
analysis, patients admitted on the first day of the burn

injury had higher %TBSA and ABSI; however, the risk

of mortality was double in patients admitted more
than 24 hours postburn. Adoption of transfer criteria
similar to those of the American Burn Association
guidelines may encourage earlier transfer and decrease mortality in these patients.24
In conclusion, extensive burns and late burn
wound excision were significant risk factors for
NIs. Also, NI is an important risk factor for mortality. Moreover, advanced age, comorbidities, burn
size, and late transfer of the patients to the burn center were the other significant risk factors for mortality.
To reduce the burn wound infections hereafter, we
increased the number of healthcare providers including the nurses, and so we have made a better quality
of patient care and trying to reduce the crosscontamination between the burn patients. Our burn
centers infrastructure has been also renewed including air conditioning and Hepa filtration systems at the
beginning of 2011. Experimentally, our studies with

Journal of Burn Care & Research

Volume XX, Number XX

stem cells are still in progress, which may be a valuable

tool in the treatment of severe burns in the near future. Conversely, The Ministry of Health has a system
for the early and appropriate transfer of burn patients
by ambulances in recent years.

ACKNOWLEDGMENTS
We thank the Infection Control Committee nurses
(Bilge Kiran, Dilek Altun, and Safiye Ersoy) for their great
assistance during data collection and thank Ahmet Ozturk
for statistical analysis. We also thank the Editing Office of
Erciyes University for suggestions about English usage.
REFERENCES
1. Murray CK. Burns. In: Mandell GL, Bennett JE, Dolin R,
eds. Principles and practice of infectious diseases.
Philadelphia: Churchill Livingstone; 2010. p. 39059.
2. Church D, Elsayed S, Reid O, Winston B, Lindsay R. Burn
wound infections. Clin Microb Rev 2006;19:403 4.
3. Macedo JLS, Santos JB. Nosocomial infections in a Brazilian
Burn Center. Burns 2006;32:477 81.
4. Taneja N, Emmanuel R, Cahri PS, Sharma M. A prospective
study of hospital-acquired infections in burn patients at a tertiary
care referral centre in North India. Burns 2004;30:6659.
5. American College of Surgeons. Advanced trauma life support. 6th ed. Chicago: American College of Surgeons; 1997.
6. Lund CC, Browder NC. The estimation of areas of burns.
Surg Gynecol Obstet 1944;79:352 8.
7. Andel D, Kamolz LP, Niedermayr M, Hoerauf K, Schramm
W, Andel H. Which of the Abbreviated Burn Severity Index
variables are having impact on the hospital length of stay?
J Burn Care Res 2007;28:163 6.
8. Horan T, Andrus M, Dudeck MA. CDC/NHSN surveillance
definition of health-care associated infection and criteria for
specific types of infections in the acute care setting. Am J
Infect Control 2008;36:309 32.
9. Murray PR, ed. Manual of clinical microbiology. 9th ed.
Washington, DC: ASM Press; 2007.
10. National Committee for Clinical Laboratory StandardsPerformance Standards for Antimicrobial Susceptibility Testing. National Committee for Clinical Standards, Wayne
(M100-S9, 1999).

Alp et al

11. Oncul O, Ulkur E, Acar A, et al. Prospective analysis of nosocomial infections in a Burn Care Unit, Turkey Indian. J Med
Res 2009;130:758 64.
12. Santucci SG, Gobara S, Santos CR, Fontana C, Levin AS.
Infections in a burn intensive care unit: experience of seven
years. J Hosp Infect 2003;53:6 13.
13. Mayhall CG. The epidemiology of burn wound infections:
then and now. Clin Infect Dis 2003;37:54350.
14. Altun D, Alp E, Baysal S, Kran B, Aygen B, Bir universite
hastanesinde eriskin ve pediatri yogun bakm unitesinde
calsan saglk personelinde el hijyenine uyum oranlar ve
dogru eldiven kullanm, Hastane Infeksiyonlar Kongresi,
Antalya, 2008.
15. Mosier MJ, Gibran NS. Surgical excision of the burn wound.
Clin Plast Surg 2009;36:61725.
16. Sharma BR. Infection in patients with severe burns: causes
and prevention thereof. Infect Dis Clin N Am 2007;21:
7459.
17. Rezaei E, Safari H, Naderinasab M, Aliakbarian H. Common
pathogens in burn wound and changes in their drug sensitivity. Burns 2011;37:8057.
18. Rajput A, Saxena R, Singh KP, et al. Prevalence and antibiotic
resistance pattern of metallo-beta-lactamase-producing Pseudomonas aeruginosa from burn patients experience of an
Indian tertiary care hospital. J Burn Care Res 2010;31:
264 8.
19. Altoparlak U, Erol S, Akcay MN, Celebi F, Kadanali A. The
time-related changes of antimicrobial resistance patterns and
predominant bacterial profiles of burn wounds and body flora
of burned patients. Burns 2004;30:660 4.
20. Towner KJ. Acinetobacter: an old friend, but a new enemy.
J Hosp Infect 2009;73:355 63.
21. Wibbenmeyer L, Danks R, Faucher L, et al. Prospective analysis of nosocomial infection rates, antibiotic use, and patterns
of resistance in a burn population. J Burn Care Res 2006;27:
152 60.
22. Jung JY, Park MS, Kim SE, et al. Risk factors for multi-drug
resistant Acinetobacter baumannii bacteremia in patients
with colonization in the intensive care unit. BMC Infect Dis
2010;10:228.
23. Lee SO, Kim NJ, Choi SH, et al. Risk factors for acquisition
of imipenem-resistant Acinetobacter baumannii: a casecontrol study. Antimicrob Agents Chemother 2004;48:
224 8.
24. Hospital and prehospital resources for optimal care of patients with burn injury: guidelines for development and operation of burn centers. American Burn Association. J Burn
Care Rehabil 1990;11:98 104.