A 10-year experience with major burns from a non-burn intensive care unit

JBUR-4256; No. of Pages 7 burns xxx (2014) xxx–xxx

Available online at www.sciencedirect.com

ScienceDirect journal homepage: www.elsevier.com/locate/burns

A 10-year experience with major burns from a non-burn intensive care unit Miguel A´ngel Ibarra Estrada *, Quetzalco´atl Cha´vez Pen˜a, Dante Ismael Garcı´a Guardado, Jose´ Arnulfo Lo´pez Pulgarı´n, Guadalupe Aguirre Avalos, Federico Corona Jime´nez Intensive Care Unit, Hospital Civil de Guadalajara ‘‘Fray Antonio Alcalde’’, Guadalajara, Jalisco, Mexico, Universidad de Guadalajara

article info

abstract

Article history:

Objective: The aim of this study was to review clinical data and outcomes of patients with

Accepted 23 December 2013

burns in a Mexican non-burn intensive care unit (ICU). Methods: We did a retrospective analysis of our single-centre database of burn patients

Keywords:

admitted to the ICU in the Hospital Civil Fray Antonio Alcalde (University Hospital). The

Major burn

sample was divided for analysis into two groups according to the outcome ‘death’ or

Prognosis

‘discharge’ from ICU.

Mortality

Results: Overall mortality was 58.2%, without a decreasing trend in mortality rates through

Non-burn ICU

the years. We identified the presence of third-degree burns (odds ratio (OR) 1.5, p = 0.003), and >49% total burned surface area (TBSA; OR 3.3, p  0.001) was associated with mortality. Mean age was higher in deceased patients (38.2 years vs. 31.3 years, p = 0.003) as was the TBSA (62.8% vs. 36.4%, p  0.001). At multivariate analysis, inhalation injury was not associated with increased mortality, but it was with more mechanical ventilation days. Early surgical debridement/cleansing was performed in most patients; however, the mean of the procedures was 1.7 per patient in both groups. Conclusion: We identified significant factors associated with mortality. These variables and prognosis from non-burn ICUs differ broadly compared with burn intensive care units (BICUs); thus, more structured, multidisciplinary and specialised treatment strategies are still needed. # 2014 Elsevier Ltd and ISBI. All rights reserved.

Major burns are injuries with necrosis at the epidermis and the dermis, resulting from thermic, chemical, electric or radiation exposure [1], with children and the elderly being the most affected [2]. Although scalding injury is the most common mechanism of injury in adults at emergency departments, direct fire is the most common mechanism in hospitalised cases, especially in men, which is associated with greater mortality [3,4]. The main risk factors associated with mortality identified so far are age, inhalation injury and total burn surface area (%TBSA) [5–8]. Age contributes significantly to mortality, as

survival in most paediatric population series is around 90– 100%. Airway injury is reported in up to 43% of all hospitalised patients with major burns, giving an 8–10-fold risk of death [9]. In addition, there is a marked correlation between %TBSA and death rising considerably from >20% TBSA [5]. In a systematic review with >186,500 patients in Europe, Brusselaers reports a mortality rate from 1.4% to 18% (maximum 34%) in major burn patients [10]; however, much of these data come from reference centres, patients with a mean %TBSA between 11% and 24%, with less strict admission criteria, and patients not necessarily critically ill. On the other

* Corresponding author at: Intensive Care Unit, Hospital Civil de Guadalajara ‘‘Fray Antonio Alcalde’’, Hospital 278, El Retiro, Specialties Building, Floor 1, Guadalajara, Jalisco 44280, Mexico. Tel.: +52 33 36728772. ´ . Ibarra Estrada). E-mail addresses: [email protected], [email protected] (M.A 0305-4179/$36.00 # 2014 Elsevier Ltd and ISBI. All rights reserved. http://dx.doi.org/10.1016/j.burns.2013.12.020 ´ , et al. A 10-year experience with major burns from a non-burn intensive care unit. Burns Please cite this article in press as: Ibarra Estrada MA (2014), http://dx.doi.org/10.1016/j.burns.2013.12.020

JBUR-4256; No. of Pages 7

2

burns xxx (2014) xxx–xxx

hand, mortality has been changing around the world, with a decreasing trend at almost 50% approximately, due to advancements in initial resuscitation, antibiotics and wound care [11]. Besides a paediatric report from a Trauma Burn Unit at ‘Magdalena de las Salinas’ Hospital, Mexico [12], there are no epidemiologic data published about adult critically ill burn patients from Mexico. With the goal of finding relevant data that could help us to improve the outcomes of our patients, we did a retrospective analysis of the past 10 years’ experience in management of critically ill burn patients at our unit.

1.

Materials and methods

1.1.

Setting

Hospital Civil de Guadalajara ‘‘Fray Antonio Alcalde’’ is a thirdlevel university hospital with 964 licensed beds in which approximately 144,500 patients are attended to at the emergency room every year [13], including patients from Jalisco and neighbouring states of the country. We do not have a multidisciplinary triage system around the region for burn patients without health insurance; therefore, most patients with burns are hospitalised from an emergency department to our open, medical-surgical, 14-bed intensive care unit (ICU), as this is the only facility available for treating critically ill burn patients in our centre. The plastic surgery department is responsible for surgical and wound care of the patients.

1.2.

Patients

We did a review of our local database, including all patients with ‘major burn’ as the main diagnosis hospitalised from August 2002 to November 2012. Inclusion criteria for analysis were major burn, as defined by American Burn Association [14], namely a TBSA 25% (>20% in >40-year-old patients or >10% third-degree burns); burns involving eyes, ears, face, hands, feet or perineum; electric burns and/or co-morbid injuries like major trauma or inhalation injury. Exclusion criteria were superficial burns and <14-year-old patients. A plastic surgeon on call assessed burn extension and depth according to the Lund–Browder scheme [15]. The diagnosis of inhalation injury was based on the need for mechanical ventilation [5,11] associated with clinical data like singed nasal or facial hair, carbonaceous sputum/matter in the nose, mouth or oropharynx, as well as the accident occurring in a closed space, from a blast injury or known exposure to hot gases; a pO2/FiO2 ratio 300 at admission also supported diagnosis but was not necessary. It was not possible to discriminate between upper or lower injury because of the unavailability of bronchoscopy. The laboratory in our centre lacks the facilities to measure carboxyhaemoglobin or cyanide levels. The diagnosis of ventilator-associated pneumonia (VAP) was based on new lung infiltrates observed in chest X-rays along with purulent sputum from the endotracheal tube, fever, elevated white blood cells and/or another biochemical markers of acute inflammation at >48 h from admission; bacterial isolation was not strictly necessary. In our unit, every patient with a TBSA >15% undergoes fluid

resuscitation based on the Parkland formula [16]. Discharge from the ICU to the general ward is indicated when haemodynamic stability and/or extubation is attained, continuous cardiopulmonary monitoring is not needed, wounds are almost healing, patients are on enteral nutrition and rehabilitation has been initiated.

1.3.

Statistical analysis

The entire sample was divided into two groups ‘death’ and ‘discharge’ (from the ICU) for analysis. We made bivariate comparisons between the groups using Pearson’s chi-square or Fisher’s exact test if needed; for continuous variables, we employed Student’s t-test or Mann–Whitney U test as adequate, based on Shapiro–Wilk test for normality. To define the %TBSA associated with increased mortality, we built a receiver operating characteristic (ROC) curve and calculated L50-%TBSA (%TBSA predicting 50% mortality) with probit regression analysis. We also calculated Spearman’s correlation between nonparametric data, as well as Mantel–Haenszel’s stratified analysis and binary logistic regression between multiple variables. All p values were two-tailed and 0.05 was considered statistically significant. Analysis was made with Statistical Package for the Social Sciences (SPSS) for MAC OSX (IBM, Ver. 20.0, SPSS Inc. 2011).

2.

Results

The total sample included 146 patients, 28 (19.2%) were women and 118 (80.8%) were men; the mean age was 35 years. Only second-degree (n = 86, 58.9%) and third-degree burns (n = 60, 41.1%) were registered. Mean TBSA was 51.8%. The most common cause was direct fire (n = 120), as seen in Fig. 1. With respect to other acute co-morbidities, 74 patients (50.7%) suffered inhalation injury, three (2.1%) had a diagnosis of mild traumatic brain injury, three (2.1%) had medullary trauma, seven (4.8%) had long bone fracture and laparotomy for traumatic acute abdomen was performed in three (2.1%) patients.

Direct fire Electric burn Deflagration 2,7% 15,1%

82,2%

Fig. 1 – Causes of burns, expressed in percentages.

´ , et al. A 10-year experience with major burns from a non-burn intensive care unit. Burns Please cite this article in press as: Ibarra Estrada MA (2014), http://dx.doi.org/10.1016/j.burns.2013.12.020

JBUR-4256; No. of Pages 7

3

burns xxx (2014) xxx–xxx

Of the patients, 49 (33.6%) developed VAP. Other infections observed were urinary tract infection in 16 patients (11%) and five (3.4%) patients had soft-tissue infection. Tracheostomy was performed in 13 patients (8.9%). Regarding surgical management carried out by non-burn plastic surgeons, 95% of the patients underwent surgical debridement/cleansing, performed before their admission in the ICU, with a mean of 1.7 procedures per patient. Seven patients died before the surgical procedure for extreme haemodynamic instability in the first 6 h of admission. In 17 patients (11.6%), fasciotomy was performed and escharotomy was performed in four patients (2.7%). Wound coverage was done with silver sulphadiazine-impregnated gauzes in all patients. There was just one procedure of autologous skin graft. Mean length of stay (LOS) in the ICU was 10 days (range, 1– 67 days).

2.1.

Mortality and associated factors

Mortality was 58.2% (n = 85), and as seen in Table 1, patients who died had more mean TBSA of 68.2% compared with 36.4%

among the survivors. The mean age was also higher in patients who died, that is, 38 and 31 years old, respectively. Half of all the deaths occurred in the first 7 days. Spearman’s correlation between %TBSA deciles and death was statistically significant, with calculated rho 0.9 and rho2 0.79 ( p = 0.003; Fig. 2); we found a sudden increase in death from 50% TBSA, so we built an ROC curve with %TBSA obtaining a cut-off at 49% with an area under the curve of 0.83 (Fig. 3); for this variable, OR for death was 3.3 ( p  0.001) with 81% and 79% for sensitivity and specificity, respectively. L50-%TBSA was calculated at 43% by the probit regression method ( p = 0.03). We did a multivariate analysis with binary logistic regression using the variable >49% TBSA (Table 2). After adjustment for gender, age, third-degree burns, electric burns, inhalation injury and VAP, we calculated an OR of 3.7 ( p  0.001), similar to the calculated risk at bivariate analysis. In addition, we found that a major age (>38 years) is associated with an increased risk of death, after adjustment for the same covariables (OR 2.18; p = 0.002). We did not find an association between inhalation injury or VAP and mortality ( p = 0.99 for both).

Table 1 – Characteristics and bivariate analysis for groups ‘‘death’’ and ‘‘discharge’’. Death n = 85 (%)

Discharge n = 61 (%)

OR

p valuea

IC 95%

Male

19 67.9%

9 32.1%

1.19

0.28

0.88–1.61

Age (years)

38.2

31.3

Variable

%TBSA Surgical cleans (mean of precedures)

62.84%

36.47%

0.003b

2.3–11.5

b

19.9–32.7

<0.001 b

0.45–0.40

1.76

1.79

>49% TBSAd

67 84.3%

13 15.7%

3.3

<0.001

2.15–5.12

3rd Degree burn

44 73.3%

16 26.7%

1.5

0.003

1.15–1.95

Electric burn

7 31.8%

15 68.2%

0.7

0.006

0.27–0.94

Inhalation injury

36 48.6%

38 51.4%

0.7

0.01

0.53–0.94

Ventilator associated pneumonia

23 57.5%

17 42.5%

1.003

0.98

0.72–1.38

Soft-tissue infection

1 20%

4 80%

<0.91

0.43

3c

0.52–17.4

Urinary tract infection

7 43.8%

9 56.2%

1.38

0.19c

0.78–2.46

Surgical cleans

83 59.7%

56 40.3%

1.39

0.44c

0.58–3.31

Fasciotomy

12 70.6%

5 29.4%

0.81

0.29c

0.57–1.14

1

1.18c

0.65–3.18

3.42

0.14c

0.54–21.61

Escharotomy

Acute kidney injury a b c d

2 50% 7 87.5%

2 50% 1 12.5%

Chi-square test. Student’s t-test. Fisher’s exact test was required. Variable obtained after construction of ROC curve with %TBSA and death.

´ , et al. A 10-year experience with major burns from a non-burn intensive care unit. Burns Please cite this article in press as: Ibarra Estrada MA (2014), http://dx.doi.org/10.1016/j.burns.2013.12.020

JBUR-4256; No. of Pages 7

4

burns xxx (2014) xxx–xxx

Table 2 – Multivariate analysis of risk factors for death.

100

Variable

90 80

Mortality

70 60 50 r2= 0.79; p= 0.003

40

Male sex Age (>38 years old) 3rd Degree burn Electric burn Inhalation injury Ventilator associated pneumonia >49% TBSAb

Adjusted ORa 0.54 2.18 39.3 2.06 20.57 18.54 3.72

p 0.54 0.002 0.99 0.16 0.99 0.99 <0.001

a

Hosmer and Lemeshow test for goodness of fit, p = 0.97 with 8 degrees of freedom. b Total burn surface area.

30 20 10 0 <30%

30-39% 40-49% 50-59% 60-69% 70-79% 80-89%

>90%

marked difference in patients with >49% of TBSA, indeed at 30 days, only three of them survived whereas 76% of patients with less %TBSA had been discharged alive from the ICU.

% TBSA

2.3. Fig. 2 – Spearman’s correlation between %TBSA deciles and death.

Another factor associated with death at the bivariate analysis was third-degree burn with a calculated OR at 1.5 ( p = 0.003). Electric burns were associated with a lower risk of death (OR 0.7; p = 0.006; Table 1). All 22 cases of electrical injuries were caused by high-voltage (>1000 V) high-tension power lines. Although mortality had an apparent decreasing trend in past years, this change was not significant at Kruskall–Wallis test ( p = 0.44, Fig. 4).

2.2.

Survival analysis

1,0

Sensitivity

0,8

0,6

0,4

0,2

0,0 0,0

0,2

Inhalation injury had a weak but statistically significant correlation with ventilator days (rho = 0.22, p = 0.006), being associated with 3 more days of mechanical ventilation in patients with such an injury (6 days vs. 3 days, respectively ( p = 0.004, IC 95% 1.1–5.8)). On the other hand, because inhalation injury seemed to be a protective factor for death as observed in the bivariate analysis (Table 1) and in Spearman’s correlation (rho 0.19 ( p = 0.017)), we did a Mantel– Haenszel stratified analysis taking %TBSA, third-degree burns and VAP as covariates, giving a statistically non-significant difference ( p = 0.67, p = 0.09 and p = 0.6 for covariates, respectively).

3.

Increased risk of death also was reflected in survival analysis for %TBSA deciles, in a Kaplan–Meier curve (Fig. 5); we see the

0,4

0,6

0,8

1,0

1 - Specificity

Fig. 3 – ROC curve to finding %TBSA cutoff predicting death. Area under the curve 0.83.

Inhalation injury

Discussion

The main goal of this work was to characterise a specific group of patients, namely critically ill burn patients not treated at specialised centres, where there is no availability of all new options in management and/or monitorisation for them, as we know that 90% of all burn-related deaths worldwide occur in developing countries like ours [17]. Despite the mean ICU LOS in patients with TBSA >40% being around 35 days (from 8 to 74 days) [11], it was lower in our series because 90% of all deaths (which accounted for most patients of the total cohort) were early, occurring in <2 weeks; in fact, 97.6% of all deaths occurred at 35 days. Besides, it is worth to remark that our mixed ICU is in continuous need for availability of beds for new and more severely ill patients; therefore, it is not uncommon to discharge to the general ward to be hastened just after haemodynamic stability and/or extubation is attained. Of all surviving patients, 98% were discharged to the general ward at 35 days, no readmission to the ICU for new complications were needed; however, we did not register how many discharges were hastened nor hospital LOS. Compared with most series, the difference in mortality was striking, in that in many series it is around 30%, whereas in our unit it was almost twice. Taking into account %TBSA as one of the main predictors of death, this result was not unexpected, because in our patients the mean %TBSA was even more than

´ , et al. A 10-year experience with major burns from a non-burn intensive care unit. Burns Please cite this article in press as: Ibarra Estrada MA (2014), http://dx.doi.org/10.1016/j.burns.2013.12.020

JBUR-4256; No. of Pages 7

5

burns xxx (2014) xxx–xxx

100 p= 0.44

Mortality (%)

90 80 70 60 50 40 30

2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012

Year Fig. 4 – Mortality through the years expressed in percentages. p value for trend obtained by Kruskall–Wallis test.

patients with 60% TBSA in other developing countries [19] unlike our unit in which mortality of patients with that burn extension was not so high (78%). In fact, this matches with the results of a Belgian validation cohort with patients admitted to burn centres, where patients with 60% TBSA had an observed mortality of 75% [5]. Despite this, an alarming fact is that mortality has not changed since 10 years, unlike the rest of the world [10]. We also confirmed that more burn extension means increased risk for mortality, and it is worth noting that besides %TBSA and age, there were no other co-morbidities independently associated with death at the multivariate analysis; VAP only was statistically significant at survival analysis, but the difference between the groups was almost null (41% vs. 42%). Albeit not always a direct association

twice compared with most reports, where it does not exceed 25–30%. Our calculated L50-%TBSA was also very different, because this has been reported from 60% to 77% [18] and resulting much lower in this series (43%), which reflects the severity of disease in our patients. Although the probit method is accepted as the best one to make periodic evaluations and comparisons between similar characteristics groups, there are no related data with adult patients to compare [12]. Because of this, we used the ROC curve method to determine the cut-off %TBSA that would predict death, obtaining good sensitivity and specificity. Given that our patients are a subgroup with a greater severity of burns, and therefore greater expected mortality, it is important to note that mortality reaches up to 100% in

1,0 Log Rank p= <0.001

0,9 0,8 40-49%

0,7

Survival

<30%

0,6 30-39%

0,5 0,4 0,3 0,2

50-59%

0,1 80-89%

0,0

70-79%

60-69%

>90%

0

10

20

30

40

50

60

70

Time (days) Fig. 5 – Kaplan–Meier curve, survival analysis by %TBSA deciles. ´ , et al. A 10-year experience with major burns from a non-burn intensive care unit. Burns Please cite this article in press as: Ibarra Estrada MA (2014), http://dx.doi.org/10.1016/j.burns.2013.12.020

JBUR-4256; No. of Pages 7

6

burns xxx (2014) xxx–xxx

between those co-morbidities and death is found [17]; this suggests that aside from wound management, the rest of the care of patients (early fluid resuscitation, mechanical ventilation, sepsis, antibiotic use, etc.) is optimal enough to avoid worsening in prognosis as we would think. An important weakness of this work is we did not register evolution in time from initial injury and resuscitation level previously received to hospitalisation in ICU, a likely heterogeneous variable that could have biased the calculated risk for death and should have been included in the multivariate analysis, as we know early deaths are due to inappropriate resuscitation [10]. In our patients, half of the deaths occurred in the first week. Besides the retrospective nature of this study, one of the most relevant weaknesses is that diagnosis of inhalation injury was made at critical care/emergency physician’s discretion and without bronchoscopy. Bronchoscopy could have allowed us to determine if airway injury was high or low, each one with a different prognosis from baseline. Currently in our centre, there are >30 different physicians with different levels of training who could be responsible for making the initial diagnosis of inhalation injury in every patient from triage at the emergency room to admission to the ICU; this heterogeneity and non-standarisation in diagnosis suggests an inadequate reasoning of criteria, giving a seeming ‘randomised’ diagnosis, explaining why in this study inhalation injury was only associated with more mechanical ventilation days, not death. Another weakness in this study was the definition of VAP, according to the new CDC’s surveillance paradigm for ventilator-associated events [21], which is based on new respiratory deterioration and isolation of pathogenic organism by cultures, we surely obtained a falsely elevated number of VAP cases; therefore, we must interpret with caution its lack of association with mortality, as happened with inhalation injury. The lack of more independent factors associated with mortality in this cohort also reflects the huge potential remaining to improve our patients’ prognosis. For example, improving wound care or increasing performance of bronchoscopy, which is not only a diagnosis adjunct but also associated with a significant reduction in mechanical ventilator days, hospital and ICU LOS, as well as hospital costs [20].

4.

Conclusion

In conclusion, epidemiological data and outcomes of critically ill burn patients managed in non-specialised non-burn ICUs are not comparable with those reported in the current literature; therefore, we must take into account different cut-offs to estimate risks and prognosis. Degree and extent of burn, as well as age were the only variables associated with increased mortality, which is not as high compared with other developing countries; this reflects that the rest of the support given to patients is appropriate and there is much more potential to increase survival, namely performance of bronchoscopy, using skin grafts and/or dressings and more frequent surgical cleansing/debridement, among others.

Conflict of interest statement The authors declare that there is no conflict of interest that could inappropriately influence this work.

Ethical approval As this was a retrospective study, and at non-patient data level, ethics committee approval or informed consent was not needed.

Acknowledgements We thank Hilario Coronado Magan˜a, MD, Department Chief of Intensive Care Unit, for his support in permission and reviewing of this work. In addition, we acknowledge Paulina Go´mez Figueroa MD, for her technical help in the transcription of data.

references

[1] Hettiaratchy S, Papini R. Initial management of a major burn: I – overview. BMJ 2004;328(June (7455)):1555–7. [2] Lloyd EC, Rodgers BC, Michener M, Williams MS. Outpatient burns: prevention and care. American Family Physician 2012;85(January (1)):25–32. [3] Koller J, Orsag M, Ondriasova E, Graffinger I, Bukovcan P. Analysis of 1119 burn injuries treated at the Bratislava Burn Department during a five-year period. Acta Chirurgiae Plasticae 1994;36(3):67–70. [4] Cutillas M, Sesay M, Perro G, Bourdarias B, Castede JC, Sanchez R. Epidemiology of elderly patients’ burns in the South West of France. Burns: Journal of the International Society for Burn Injuries 1998;24(March (2)):134–8. [5] Belgian Outcome in Burn Injury Study G. Development and validation of a model for prediction of mortality in patients with acute burn injury. British Journal of Surgery 2009;96(January (1)):111–7. [6] Germann G, Barthold U, Lefering R, Raff T, Hartmann B. The impact of risk factors and pre-existing conditions on the mortality of burn patients and the precision of predictive admission-scoring systems. Burns: Journal of the International Society for Burn Injuries 1997;23(May (3)): 195–203. [7] Galeiras R, Lorente JA, Pertega S, Vallejo A, Tomicic V, de la Cal MA, et al. A model for predicting mortality among critically ill burn victims. Burns: Journal of the International Society for Burn Injuries 2009;35(March (2)):201–9. [8] Coste J, Wasserman D, Venot A. Predicting mortality in adult burned patients: methodological aspects of the construction and validation of a composite ratio scale. Journal of Clinical Epidemiology 1996;49(October (10)): 1125–31. [9] Brusselaers N, Monstrey S, Blot S. The FLAMES score accurately predicts mortality risk in burn patients (Gomez M et al., 2008). Journal of Trauma 2009;67(August (2)):415. [10] Brusselaers N, Monstrey S, Vogelaers D, Hoste E, Blot S. Severe burn injury in Europe: a systematic review of the incidence, etiology, morbidity, and mortality. Critical Care 2010;14(5):R188.

´ , et al. A 10-year experience with major burns from a non-burn intensive care unit. Burns Please cite this article in press as: Ibarra Estrada MA (2014), http://dx.doi.org/10.1016/j.burns.2013.12.020

JBUR-4256; No. of Pages 7 burns xxx (2014) xxx–xxx

[11] Brusselaers N, Hoste EA, Monstrey S, Colpaert KE, De Waele JJ, Vandewoude KH, et al. Outcome and changes over time in survival following severe burns from 1985 to 2004. Intensive Care Medicine 2005;31(December (12)):1648–53. [12] Cuenca-Pardo J, de Jesus Alvarez-Diaz C, Compres-Pichardo TA. Related factors in burn children. Epidemiological study of the burn unit at the ‘‘Magdalena de las Salinas’’ Traumatology Hospital. Journal of Burn Care & Research: Official Publication of the American Burn Association 2008;29(May–June (3)):468–74. [13] Un dı´a tı´pico en el hospital. Hospital Civil de Guadalajara; 2013, http://www.hcg.udg.mx [accessed October 17]. [14] American Burn Association. Burn Center Referral Criteria; 2013, www.ameriburn.org [accessed June 26]. [15] Lund CCBN. The estimation of areas of burns. Surgery Gynecology and Obstetrics 1944;79:352. [16] Gibran NS. Committee on Organization, Delivery of Burn Care ABA. Practice Guidelines for burn care, 2006. Journal of Burn Care & Research: official publication of the American Burn Association 2006;27(July–August (4)):437–8. [17] Forjuoh SN. Burns in low- and middle-income countries: a review of available literature on descriptive epidemiology,

[18]

[19]

[20]

[21]

risk factors, treatment, and prevention. Burns: Journal of the International Society for Burn Injuries 2006;32(August (5)):529–37. Saffle JR, Davis B, Williams P. Recent outcomes in the treatment of burn injury in the United States: a report from the American Burn Association Patient Registry. Journal of Burn Care & Rehabilitation 1995;16(May–June (3 Pt 1)): 219–32. discussion 88–9. Kumar V, Mohanty MK, Kanth S. Fatal burns in Manipal area: a 10 year study. Journal of Forensic and Legal Medicine 2007;14(January (1)):3–6. Carr JA, Phillips BD, Bowling WM. The utility of bronchoscopy after inhalation injury complicated by pneumonia in burn patients: results from the National Burn Repository. Journal of Burn Care & Research: official publication of the American Burn Association 2009;30(November–December (6)):967–74. Centers for Disease Control and Prevention. Surveillance for Ventilator Associated Events; 2013, http://www.cdc.gov/ nhsn/acute-care-hospital/vap/index.html [accessed November 23rd].

´ , et al. A 10-year experience with major burns from a non-burn intensive care unit. Burns Please cite this article in press as: Ibarra Estrada MA (2014), http://dx.doi.org/10.1016/j.burns.2013.12.020

7