- Email: [email protected]
Respiratory failure after pediatric scald injury
Journal of Pediatric Surgery (2011) 46, 1753–1758
www.elsevier.com/locate/jpedsurg
Respiratory failure after pediatric scald injury Dorothy V. Rocourt a , Mark Hall b , Brian D. Kenney c , Renata Fabia c , Jonathan I. Groner c , Gail E. Besner c,⁎ a
Department of Pediatric Surgery, Penn State Hershey Children's Hospital, Penn State Milton S. Hershey Medical Center, PO Box 850, Hershey, PA 17033, USA b Division of Critical Care Medicine, Nationwide Children's Hospital and The Ohio State University College of Medicine, Columbus, OH 43205, USA c Department of Pediatric Surgery, Nationwide Children's Hospital and The Ohio State University College of Medicine, Columbus, OH 43205, USA Received 29 November 2010; revised 19 April 2011; accepted 20 April 2011
Key words: Burns; Pediatric; Scald; Systemic inflammatory response syndrome
Abstract Objective: A subset of children with scald burns develops respiratory failure despite no direct injury to the lungs. We examined these patients in an effort to elucidate the etiology of the respiratory failure. Methods: The charts of pediatric patients with greater than 10% total body surface area (TBSA) scald burns were reviewed. Age, weight, burn distribution, percentage of TBSA burned, resuscitation volumes, Injury Severity Score, evidence of abuse, length of stay, days in the intensive care unit, and time and duration of intubation were recorded. Results: Two hundred thirty-two patients met our inclusion criteria. Of these, 220 patients did not require intubation, and 12 of the patients did. No patient older than 3 years or with burns less than 15% TBSA required intubation. Fluid over resuscitation was not directly associated with respiratory failure requiring mechanical ventilation. Conclusions: We report the largest published series of patients with scald burns requiring mechanical ventilation in the absence of direct airway injury. Five percent of pediatric patients required mechanical ventilation after scald injury. We believe that a combination of causes including fluid resuscitation, young patient age, small patient size, and possible activation of the systemic inflammatory immune response may be responsible for the respiratory failure. © 2011 Elsevier Inc. All rights reserved.
Scald burns are the most common type of burn injury in the pediatric population and can result in significant morbidity and mortality [1,2]. In a small subset of pediatric patients with scald burns, endotracheal intubation is required secondary to the development of respiratory failure despite the absence of inhalation injury or other direct injury to the ⁎ Corresponding author. Tel.: +1 614 722 3900; fax: +1 614 722 3903. E-mail address: [email protected] (G.E. Besner). 0022-3468/$ – see front matter © 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.jpedsurg.2011.04.018
lungs. Although experienced burn surgeons are aware of this phenomenon, reports in the literature are sparse, with only 2 previous small case series published [3,4]. The mechanism of respiratory failure in these scalded patients has yet to be defined. Fluid overresuscitation resulting in upper airway edema or decreased pulmonary compliance has been proposed as a potential etiology [4]. In addition, it is possible that activation of the innate immune response, leading to activation of the systemic inflammatory response syndrome
1754 (SIRS), may play a role, although this possibility has not been previously studied in detail. Cytokines regulate the host's response to infection, inflammation, and trauma. Burns involving 10% to 15% of the total body surface area (TBSA) result in a local inflammatory response with very little to no morbidity [5]. Major burns can lead to activation of an inflammatory cascade resulting in the development of SIRS and may also lead to multiple organ dysfunction syndrome [5]. After thermal injury, various inflammatory mediators are locally and systemically produced [5]. The morbidity associated with major burns remains high despite improved early burn care, control of infection, fluid resuscitation, and aggressive nutritional management. In this manuscript, we sought to determine the frequency with which pediatric patients with scald burns develop respiratory distress requiring intubation as well as any factors that were predictive of the development of this complication, and we report the largest published series of such cases to date.
1. Methods After hospital institutional review board approval was obtained (protocol no. 07-00370), the medical records of patients 0 to 18 years old with more than 10% TBSA scald burns admitted from January 1994 to February 2008 were reviewed. All other types of burns or combinations of burn injuries (ie, flame burns, inhalation injury, etc) were excluded from this demographic analysis. Age, weight, burn distribution, percentage of TBSA burned, resuscitation volumes, Injury Severity Score (ISS), evidence of abuse, length of hospital stay, admission to the intensive care unit (ICU), and time and duration of intubation were recorded. Fluid resuscitation was standardized to the Parkland formula, with children weighing less than 20 kg also given maintenance fluids. Demographic data were analyzed using the Wilcoxon rank sum test and Fisher's Exact test, with P b .05 considered statistically significant. Univariate logistic regression models were fit to patients who were intubated vs those who did not require endotracheal intubation in patients who were younger than 4 years old having burn areas of at least 10%. Multivariate logistic regression was then performed for patients who were intubated because there was more than 1 independent variable in the model.
2. Results A total of 398 burn patients were identified from January 1994 to February 2008 of which 232 patients met our inclusion criteria of 0 to 18 years old with more than 10% TBSA scald burns. Of the 232 patients, 90 were women (39%) and 142 were men (61%). Racial distribution included
D.V. Rocourt et al. Table 1 Comparison of patients with scald injuries requiring endotracheal intubation with patients not requiring intubation Patients who Patients who P value (SD) were intubated were not (n = 12) intubated (n = 220) Age (y) Weight (kg) % TBSA ISS LOS (d) ICU stay (d) Presence of abuse
1.4 11.1 35.8 21.4 49.8 26.9 33%
3.0 18.2 15.4 10.2 10.9 0.5 10%
.025 (0.8; 3.3) .022 (2.7; 14.3) b.001 (13.8; 5.6) b.001 (5.6; 4.5) b.001 (45; 10.2) b.001 (25.1; 2.3) .033
47 African American (21%), 151 white (66%), and 31 other (14%). The average age of the population was 1.6 years; average weight, 13 kg; average ISS, 9.0; and average TBSA, 20%. Burns were the result of hot water in 96%, hot grease in 3%, and hot wax in 1%. Of the 232 total patients, 220 patients did not require intubation. The mean age of these nonintubated patients was 3 years; weight, 18.2 kg; percentage of TBSA burned, 15.4%; ISS, 10.2; length of stay (LOS), 10.9 days; and ICU stay, 0.5 days (Table 1). Of the 232 total patients, 12 patients (5%) required endotracheal intubation (Table 1). In general, intubated patients were younger in age, smaller in size, had a larger percentage of TBSA burn, longer LOS and ICU stay, and higher ISSs than nonintubated patients. The mean age of scalded patients requiring intubation was 1.4 years (compared with 3 years in the nonintubated patients; P = .025). The mean weight of the intubated patients was 11.1 kg (compared with 18.2 kg in the nonintubated patients; P = .022). The mean TBSA of the intubated patients was 35.8% (compared with 15.4% in the nonintubated patients; P b .001). Intubated patients had a mean ISS of 21.4 (compared with 10.2 in nonintubated patients; P b .001). Mean length of hospital stay was 49.8 days (compared with 10.9 days in the nonintubated patients; P b .001), and mean ICU stay was 26.9 days (compared with 0.5 days in the nonintubated patients; P b .001). Evidence of abuse was associated with 33% of the patients who required intubation compared with 10% of nonintubated patients (P = .033). Details regarding the 12 patients who required intubation are provided in Table 2. No patient older than 3 years or who had a scald injury that covered less than 15% TBSA required intubation. The mean time to intubation and mechanical ventilation was 19 hours post injury (SD, 24.9). The mean duration of intubation was 18.8 days (SD, 16.7). These patients were resuscitated using the Parkland formula (3-4 mL Ringer lactate per kilogram per percentage of TBSA burn), with patients weighing less than 20 kg who had burns more than 15% TBSA receiving maintenance fluids 5% Dextrose Ringers Lactate (D5RL) in addition. The mean volume of fluids (resuscitation fluid plus maintenance fluid when
Respiratory failure after pediatric scald injury Table 2
1755
Demographics of patients requiring intubation
Patient no.
Race
Sex
Age (y)
Weight (kg)
ISS
Hours to ETT
Duration of ETT (d)
% TBSA burn
IVF first 24 h (mL/kg/% TBSA burn)
UO in first 24 h (mL/kg)
1 2 3 4 5 6 7 8 9 10 11 12 Mean values
White Black White Black White White White White White White White Black
Male Male Male Female Female Male Female Male Female Female Male Male
1.2 2.2 2.75 0.83 0.75 1.08 0.75 1.0 1.17 1.08 1.25 3.0 1.4
15 13.5 13 7.5 8.0 10 9.0 10 13 9.0 10 15.6 11.1
16 16 25 25 25 25 25 9 16 25 25 25 21
0 4 3.5 48 40 1 6 48 0 72 2 8 19
1 1 4 23 28 23 29 3 21 27 14 60 18.8
25 20 30 55 50 50 30 15 25 40 40 55 35.8
5.8 7.4 6.9 8.4 9.9 5.5 6.5 4.55 8.58 9.9 8.62 8.99 7.6
1.18 1.8 1.58 2.01 1.25 1.1 1.19 2.05 2.03 1.57 1.99 1.85 1.6
ETT indicates endotracheal tube; UO, urine output; IVF, intravenous fluid.
appropriate) administered to the intubated patients was 7.6 mL/kg per percentage of TBSA burn in the first 24 hours (SD, 1.8). The mean hourly urine output of the intubated patients was 1.6 mL/kg (SD, 0.4). Indications for intubation did not correlate with resuscitation fluid volumes. None of the patients in the intubated group had oropharyngeal burns or laryngeal edema as evidenced by direct laryngoscopy at the time of intubation. Patients intubated at referring hospitals all underwent direct laryngoscopy by the surgical critical care or pediatric ICU (PICU) staff upon arrival to our PICU. There was no sex or racial differences predictive of the need for intubation. In the entire group of 232 scalded patients, burns to the face, neck, and chest were the most common location, with a frequency of 73%. Of the 12 patients requiring intubation, 9 (75%) sustained burns predominantly to their face and chest, with the remaining 3 patients (25%) having burns localized to the upper and lower extremities and lower torso. There were no statistical differences between burn location in patients requiring or not requiring mechanical ventilation (P = 1.0). The reasons for intubation in the 12 patients were as follows. In 3 cases (25%), initial intubation was related to airway protection for transport to a level 1 burn center. These patients required mechanical ventilation for 1, 21, and 14 days, respectively. One of those patients developed acute respiratory distress syndrome (ARDS) and pneumonia. An additional 3 patients (25%) were intubated prophylactically before the development of respiratory distress. Of these 3 patients, 1 was electively intubated in the emergency department because of severe facial burns; the second was intubated in the ICU for pain control and wound care, and the third was intubated for an immediate escharotomy that was performed in the operating room and remained initially intubated for pain control during wound care. These patients required mechanical ventilation for 4, 3, and 60 days, respectively. The remaining 6 patients
(50%) were intubated for respiratory distress (stridor or labored breathing) and required mechanical ventilation for 1, 23, 28, 23, 29, and 27 days. Complications that developed in the intubated patients were as follows: 10 developed pneumonia, with 6 of the 10 developing ARDS requiring High Frequency Oscillating Ventillation (HFOV). Pneumonia was defined by either a new infiltrate on Chest radiogragh (CXR) or increased oxygen requirements with culture-positive bronchoalveolar lavage. The most common organisms cultured from bronchoalveolar lavage specimens were Candida albicans and Streptococcus viridians. One patient required tracheostomy for prolonged intubation. Other complications included line sepsis in 1 patient, abdominal compartment syndrome owing to chylous ascites in 2 patients, and respiratory arrest in 2 patients. Respiratory arrest was secondary to a mucus plug in the tracheostomy tube in 1 patient and a mucus plug in the endotracheal tube of the other patient. Both resolved promptly with suctioning with no further sequellae. There was no mortality, with all patients surviving their scald burn injuries. In our group of patients requiring intubation, 7 of the 12 patients developed ARDS as demonstrated by a PaO2/FIO2 greater than 200 and by bilateral patchy infiltrates. None of the intubated patients had a history of asthma, reactive airway disease, or were reported to have symptoms consistent with a viral upper respiratory tract infection before admission. There were 3 patients who were prophylactically intubated for transport to our facility. Despite the initial reason for intubation, 2 of these patients required mechanical ventilation for greater than 10 days. These 2 patients (40% and 25% TBSA burns, respectively) had a mean hourly urine output of 1.99 and 2.03 mL/kg, respectively, and received a volume of 8.62 and 8.58 mL/kg per percentage of TBSA, respectively. The third patient (25% TBSA burn) who did not require prolonged intubation
1756 Table 3
D.V. Rocourt et al. Predictors of endotracheal intubation
Predictor
Odds ratio
P
Age Weight % TBSA ISS Injury location Abuse, yes or no
NA NA 2.9 (1.9, 4.6) 4.6 (2.6, 8.2) NA 4.5 (1.3, 16.2)
.31 .16 b.001 b.001 .87 .034
NA indicates not applicable.
had a mean hourly urine output of 1.18 mL/kg and received a volume of 5.8 mL/kg per percentage of TBSA. Of the 2 patients requiring prolonged mechanical ventilation, 1 went on to develop ARDS. In contrast to these patients, the 17 patients admitted to the ICU who did not require intubation had a mean hourly urine output of 2.4 mL/kg (range, 1.753.33 mL/kg per hour) with a mean volume of resuscitation of 7.2 mL/kg per percentage of TBSA (range, 3.93-10.36 mL/ kg per percentage of TBSA). Although ascites is well described after burn resuscitation, chylous ascites is unusual [6,7]. Interestingly, 2 of the patients requiring intubation in our study developed chylous ascites within the first 3 days of burn injury. The first patient was a 9-month-old girl with a 30% TBSA burn. Her resuscitation required 6.5 mL/kg crystalloid during the first 24 hours after injury to maintain an average hourly urine output of more than 1 mL/kg. She developed Abdominal Compartment Sydrome (ACS) with bladder pressures greater than 30 mm Hg and an abdominal ultrasound confirming ascites. She developed worsening respiratory failure necessitating HFOV. She underwent decompression with a pigtail catheter. The fluid drained was milky in appearance and
Table 4
contained 88% lymphocytes. Her enteral formula was changed to Vivonex, a short-chain fatty acid formula, with resolution of her chylous ascites. She was weaned to a conventional ventilator by the third day of HFOV. The second patient was a 13-month-old girl with a 40% TBSA burn whose resuscitation required 9.9 mL/kg of crystalloid during the first 24 hours post burn to maintain an hourly urine output of more than 1 mL/kg. She developed abdominal distension and increased oxygen requirements on the ventilator. An abdominal ultrasound confirmed ascites that was drained with a pigtail catheter. The ascitic fluid was milky in appearance and contained 90% lymphocytes. Her enteral formula was changed to Vivonex, also resulting in resolution of her chylous ascites as demonstrated by repeat peritoneal fluid cell count. It is clear that volume of resuscitation contributes to the development of ACS; however, it is unclear why these 2 patients developed this complication when their hourly urine output was below the mean. In addition, both of these patients had transthoracic echocardiograms performed to assess their volume status and cardiac function, which were within normal limits with no evidence of fluid overload. Univariate logistic regression was used to determine the independent need for intubation in patients younger than 4 years with TBSA burn greater than 10% (Table 3). Both ISS and TBSA were independent predictors of patients requiring mechanical intubation. For each 5-point increase in ISS, patients had 4.6 times higher odds of requiring mechanical ventilation. For each 5% increase in TBSA, patients had a 2.9 times higher odds for intubation. Interestingly, evidence of abuse was a significant factor associated with the need for endotracheal intubation (P b .031). Patients with scald burns owing to child abuse had 4.5 times higher odds of intubation
Demographics of patients with scald burns admitted to the ICU who did not require endotracheal intubation
Patient no.
Race
Sex
Age (y)
Weight (kg)
ISS
% TBSA burn
IVF first 24 h (mL/kg/% TBSA burn)
UO in first 24 h (mL/kg)
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 Mean values
White White Other White White White Black White White Other White Black White Other Black White Black
Male Male Male Female Female Male Female Female Male Female Male Male Female Male Male Male Female
1.16 .5 4 2 1.16 6 1.33 1.25 1.08 3 3 3 1.16 1.42 1.08 1.42 2 2.1
13 8.7 20 11.4 8.6 23 9 10 9.2 16.7 15.8 17 12 12.2 11.8 11.1 10 14.0
9 16 25 9 25 9 9 25 16 25 25 16 9 16 16 16 16 14.8
15 20 32 15 30 15 15 35 23 30 30 18 15 25 20 25 15 21.2
6.46 8.11 7.4 7.98 10.5 6.3 7.22 6.6 7.72 3.93 5.9 6.11 8.67 6.25 10.36 4.95 7.91 7.2
1.98 2.23 2.56 2.03 2.01 1.75 2.21 2.13 2.19 2.43 2.39 3.25 2.61 2.25 3.07 3.33 2.01 2.4
Respiratory failure after pediatric scald injury than nonabused patients. Older age was associated with a reduced need for endotracheal intubation. In a multivariable logistic regression, only TBSA burn remained statistically significant, whereas ISS and abuse did not provide the information about mechanical ventilation requirement beyond the information contained in TBSA (P b .03). Of the 232 patients who met inclusion criteria (0-18 years old with greater than 10% TBSA scald burns), there were 17 patients admitted to the PICU who did not require endotracheal intubation (Table 4). The 17 patients admitted to the PICU who were not intubated were compared with the 12 patients admitted to the PICU requiring endotracheal intubation (Table 5). The mean age of nonintubated PICU patients was 2.1 years compared with 1.4 years in the intubated patients (P b .07). The mean weight in the nonintubated PICU patients was 14 kg compared with 11.1 kg in the intubated patients (P b .17). The mean percentage of TBSA burned in the nonintubated PICU patients was 21.2% compared with 35.8% in the intubated patients (P b .01). Nonintubated PICU patients had a mean ISS score of 14.8 compared with 21.4% in the intubated patients (P b .02). As expected, mean LOS and ICU stay were statistically greater in PICU patients requiring intubation compared with nonintubated PICU patients (P = .05 and P b .001, respectively). Interestingly, the presence of abuse was statistically greater in the intubated PICU patients compared with nonintubated ICU patients (P b .03). Importantly, there were no statistically significant differences in the mean volume of resuscitation fluids in the first 24 hours after injury in nonintubated compared with intubated PICU patients. Nonintubated patients received a mean volume of 7.2 mL/kg per percentage of TBSA burned (SD, 1.7) in the first 24 hours post injury compared with 7.6 mL/kg per
Table 5 Comparison of patients admitted to the ICU with scald injuries requiring endotracheal intubation with ICU patients not requiring intubation
Age (y) Weight (kg) % TBSA ISS Injury location, face/chest:other Abuse (%) LOS (d) ICU (d) IVF first 24 h (mL/kg/% TBSA burn) UO in first 24 h (mL/kg)
ICU patients who were intubated (n = 12)
ICU patients P value (SD) who were not intubated (n = 17)
1.4 11.1 35.8 21.4 9:3
2.1 14 21.2 14.8 14:3
.07 (0.8; 1.4) .17 (2.7; 5.3) .01 (13.8; 6.3) .02 (5.6; 6.0) .44
33 49.8 26.9 7.6
10 23.3 1.9 7.2
.03 .05 (45; 12.1) b.001 (25.1; 3.1) .54 (1.8;1.7)
1.6
2.4
.001 (0.4;0.5)
1757 percentage of TBSA burned (P = .54). Nonintubated patients had a statistically greater mean hourly urine output of 2.4 mL/kg per hour in the first 24 hours compared with 1.6 mL/kg per hour in intubated patients (P = .5). There were no complications in the patients admitted to the PICU who did not require intubation. Of the 232 patients in the study, 29 were admitted to the PICU (12 intubated and 17 nonintubated), and 203 did not require PICU admission. Most patients in the study had partial-thickness burns, with a few sustaining full-thickness burns requiring surgical intervention. Of the 203 patients who were not admitted to the PICU, 16 required operative intervention, including 6 patients who underwent burn debridement followed by dressing changes and 10 patients who required split-thickness skin grafting. Operative procedures were performed in 7 of the 12 intubated PICU patients, all of whom required split-thickness skin grafting. Operative procedures were performed in 10 of the 17 nonintubated PICU patients, including 5 patients who underwent burn debridement and 5 patients who requited split-thickness grafting.
3. Discussion Respiratory failure after cutaneous scald burn injury is relatively uncommon, with few reports in the literature to date [2,3,8,9]. Our study represents the largest series of patients with cutaneous scald burns requiring mechanical ventilation in the absence of direct airway injury to date and corroborates 2 older but smaller reports. In our study, the prevalence of patients requiring intubation after a scald burn was 0.8 or less than 1 patient a year. Despite its rarity, we were able to identify some factors that may contribute to this event. Hudson et al [3] reported 7 patients with an average age of 18 months who required mechanical ventilation after scald injuries. Three of these patients died, with 2 of the 3 patients having sustained direct thermal injury to the respiratory tract from ingestion of hot liquids. Zak et al [4] reported 7 patients (4%) of their scalded patients (mean age, 1.4 years; mean percentage of TBSA burn, 29.9%) required mechanical ventilation. In their series, the ventilated patients were younger than 2 years old with a TBSA burn of more than 20%. In our current study, 12 (5%) of scalded patients (mean age, 1.4 years; mean percentage of TBSA burn, 38.5%) required mechanical ventilation, with none of the patients having direct oropharyngeal burns. Adequate volume of resuscitation in burn patients is critical. Urine output has been the major determinant of adequacy of fluid resuscitation. In our study, we found that patients requiring endotracheal intubation had a mean hourly urine output of 1.6 mL/kg per hour with a total mean volume of resuscitation of 7.6 mL/kg per percentage of TBSA over the first 24 hours. Zak et al reported a low average hourly urine output (0.84 mL/kg per hour) within the first 24 hours after injury,
1758 arguing that fluid resuscitation was not excessive in their intubated patients. Although the mean volume of resuscitation in our group was higher than that that reported by Zak et al, the percentage of patients requiring endotracheal intubation is similar. Because fluid over resuscitation might lead to the development of respiratory failure, we compared patients admitted to the PICU who did not require mechanical ventilation to those who did. There were no differences in age, size, and, more importantly, volume of resuscitation in the first 24 hours after injury in these 2 groups. In fact, we found that patients who did not require intubation received an average volume of 7.2 mL/kg per percentage of TBSA in the first 24 hours after injury and had a mean hourly urine output of 2.4 mL/kg compared with a mean hourly urine output of 1.6 mL/kg in the intubated patients. Patients who required mechanical ventilation had larger TBSA burns and higher ISS. In both multivariate and univariate logistic regressions, TBSA burn had a higher odds for requiring endotracheal intubation. Although volume of resuscitation was not a statistically significant factor for requiring mechanical ventilation, its contribution could not be definitively excluded because of the relatively small number of patients requiring intubation. In our study, fluids in excess of the calculated resuscitation needs were administered only as needed for low urine output (b1 mL/kg per hour) or for hemodynamic instability (hypotension). To further eliminate fluid overresuscitation as a contributing factor for the need for intubation, we have developed a nurse-directed fluid resuscitation protocol in our burn unit. This protocol is designed to ensure that fluid resuscitation volumes are decreased when hourly urine outputs are high. Because the pathophysiology of respiratory failure in scalded patients remains unclear, it is possible that perturbations in the immune response may play a contributory role. Furthermore, because all of the patients who required intubation in our study had scald burns greater than 15% TBSA and because burns greater than 15% TBSA have been shown to trigger a systemic inflammatory response, it is possible that activation of SIRS may play a role [5]. Burn injury is a trigger for activation of inflammatory mediators including proinflammatory cytokines [5,10]. Serum concentrations of interleukin-6 and tumor necrosis factor-α have been found to be elevated in the early stages of burn injury [9]. KowalVern et al [11] reported a significant increase in plasma interleukin-6, tumor necrosis factor-α, and other inflammatory cytokines in burn patients with and without inhalation injury. Furthermore, Stengle et al [12] used a mouse model of 15% TBSA scald burn to demonstrate a significant elevation of pulmonary neutrophil infiltration within 6 hours of thermal injury and at 1 week postburn. In addition, activation of the SIRS typically leads to lung injury, pulmonary hypoperfusion, and respiratory failure [13]. To further elucidate the possible role of an overexuberant immune response in scald burn patients, we plan to examine parameters of the innate and acquired immune responses in these patients in the future.
D.V. Rocourt et al. In summary, the requirement for intubation, although relatively uncommon after scald injury in pediatric patients, is under appreciated. This complication should be anticipated in approximately 5% of patients, especially in patients younger than 3 years with greater than 15% TBSA scald burns. Volume of resuscitation should be tailored to maintain an average hourly urine output of 1 mL/kg with the goal of maintaining end organ perfusion; however, overresuscitation should be carefully avoided. In the current study, although fluid over resuscitation did not appear to be a major factor in the development of respiratory failure, this possibility could not be definitively excluded because of the relatively small number of patients requiring intubation. Therefore, a nurse-directed fluid resuscitation protocol has been developed in our burn unit to ensure that fluid overresuscitation is not playing a role, specifically ensuring that fluid resuscitation volumes are decreased when hourly urine outputs are high.
References [1] Drago DA. Kitchen scalds and thermal burns in children five years and younger. Pediatrics 2005;115:10-6. [2] Guzel A, Aksu B, Aylanç H, et al. Scalds in pediatric emergency department: a 5-year experience. J Burn Care Res 2009;30: 450-6. [3] Hudson DA, Jones L, Rode H. Respiratory distress secondary to scalds in children. Burns 1994;20:434-7. [4] Zak AL, Harrington DT, Barillo DJ, et al. Acute respiratory failure that complicates the resuscitation of pediatric patients with scald injuries. J Burn Care Rehabil 1999;20:391-9. [5] Nakae H, Motoyama S, Kurosawa S, et al. The effective removal of proinflammatory cytokines by continuous hemofiltration with a polymethylmethacrylate membrane following severe burn injury: reports of three cases. Surg Today 1999;29:762-5. [6] Moore-Olufemi SD, Xue H, Allen SJ, et al. Effects of primary and secondary intra-abdominal hypertension on mesenteric lymph flow: implications for the abdominal compartment syndrome. Shock 2005;23:571-5. [7] Ida J, Yamashita K, Inoue T, et al. Resuscitation fluid volume and abdominal compartment syndrome in patients with major burns. Burns 2006;32:151-4. [8] Desai MH, Nichols MM, Herndon DN. Scald injury of the respiratory tract: an unusual occurance. J Burn Care Rhabil 1987;8:210-2. [9] Bjorkl L, Svensson H. Upper airway obstruction—an unusual complication following a minor scalding injury. Burns 1999;19: 85-7. [10] Kowal-Vern A, et al. Circulating endothelial cell levels correlate with proinflammatory cytokine increase in the acute phase of thermal injury. J Burn Care Rehabil 2005;26:422-9. [11] Kowal-Vern A, Ortegel J, Bourdon P, et al. Elevated cytokine in peritoneal fluid from burned patients with intra-abdominal compartment syndrome. Burns 2006;32:563-9. [12] Stengle J, Meyers R, Pyle J, et al. Neutrophil recruitment after remote scald injury. J Burn Care Rhabil 1996;17:14-8. [13] Oda J, Yamashita K, Inoue T, et al. Acute lung injury and multiple organ dysfunction syndrome secondary to intra-abdominal hypertension and abdominal decompression in extensively burned patients. J Trauma 2007;62:1365-9.
www.elsevier.com/locate/jpedsurg
Respiratory failure after pediatric scald injury Dorothy V. Rocourt a , Mark Hall b , Brian D. Kenney c , Renata Fabia c , Jonathan I. Groner c , Gail E. Besner c,⁎ a
Department of Pediatric Surgery, Penn State Hershey Children's Hospital, Penn State Milton S. Hershey Medical Center, PO Box 850, Hershey, PA 17033, USA b Division of Critical Care Medicine, Nationwide Children's Hospital and The Ohio State University College of Medicine, Columbus, OH 43205, USA c Department of Pediatric Surgery, Nationwide Children's Hospital and The Ohio State University College of Medicine, Columbus, OH 43205, USA Received 29 November 2010; revised 19 April 2011; accepted 20 April 2011
Key words: Burns; Pediatric; Scald; Systemic inflammatory response syndrome
Abstract Objective: A subset of children with scald burns develops respiratory failure despite no direct injury to the lungs. We examined these patients in an effort to elucidate the etiology of the respiratory failure. Methods: The charts of pediatric patients with greater than 10% total body surface area (TBSA) scald burns were reviewed. Age, weight, burn distribution, percentage of TBSA burned, resuscitation volumes, Injury Severity Score, evidence of abuse, length of stay, days in the intensive care unit, and time and duration of intubation were recorded. Results: Two hundred thirty-two patients met our inclusion criteria. Of these, 220 patients did not require intubation, and 12 of the patients did. No patient older than 3 years or with burns less than 15% TBSA required intubation. Fluid over resuscitation was not directly associated with respiratory failure requiring mechanical ventilation. Conclusions: We report the largest published series of patients with scald burns requiring mechanical ventilation in the absence of direct airway injury. Five percent of pediatric patients required mechanical ventilation after scald injury. We believe that a combination of causes including fluid resuscitation, young patient age, small patient size, and possible activation of the systemic inflammatory immune response may be responsible for the respiratory failure. © 2011 Elsevier Inc. All rights reserved.
Scald burns are the most common type of burn injury in the pediatric population and can result in significant morbidity and mortality [1,2]. In a small subset of pediatric patients with scald burns, endotracheal intubation is required secondary to the development of respiratory failure despite the absence of inhalation injury or other direct injury to the ⁎ Corresponding author. Tel.: +1 614 722 3900; fax: +1 614 722 3903. E-mail address: [email protected] (G.E. Besner). 0022-3468/$ – see front matter © 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.jpedsurg.2011.04.018
lungs. Although experienced burn surgeons are aware of this phenomenon, reports in the literature are sparse, with only 2 previous small case series published [3,4]. The mechanism of respiratory failure in these scalded patients has yet to be defined. Fluid overresuscitation resulting in upper airway edema or decreased pulmonary compliance has been proposed as a potential etiology [4]. In addition, it is possible that activation of the innate immune response, leading to activation of the systemic inflammatory response syndrome
1754 (SIRS), may play a role, although this possibility has not been previously studied in detail. Cytokines regulate the host's response to infection, inflammation, and trauma. Burns involving 10% to 15% of the total body surface area (TBSA) result in a local inflammatory response with very little to no morbidity [5]. Major burns can lead to activation of an inflammatory cascade resulting in the development of SIRS and may also lead to multiple organ dysfunction syndrome [5]. After thermal injury, various inflammatory mediators are locally and systemically produced [5]. The morbidity associated with major burns remains high despite improved early burn care, control of infection, fluid resuscitation, and aggressive nutritional management. In this manuscript, we sought to determine the frequency with which pediatric patients with scald burns develop respiratory distress requiring intubation as well as any factors that were predictive of the development of this complication, and we report the largest published series of such cases to date.
1. Methods After hospital institutional review board approval was obtained (protocol no. 07-00370), the medical records of patients 0 to 18 years old with more than 10% TBSA scald burns admitted from January 1994 to February 2008 were reviewed. All other types of burns or combinations of burn injuries (ie, flame burns, inhalation injury, etc) were excluded from this demographic analysis. Age, weight, burn distribution, percentage of TBSA burned, resuscitation volumes, Injury Severity Score (ISS), evidence of abuse, length of hospital stay, admission to the intensive care unit (ICU), and time and duration of intubation were recorded. Fluid resuscitation was standardized to the Parkland formula, with children weighing less than 20 kg also given maintenance fluids. Demographic data were analyzed using the Wilcoxon rank sum test and Fisher's Exact test, with P b .05 considered statistically significant. Univariate logistic regression models were fit to patients who were intubated vs those who did not require endotracheal intubation in patients who were younger than 4 years old having burn areas of at least 10%. Multivariate logistic regression was then performed for patients who were intubated because there was more than 1 independent variable in the model.
2. Results A total of 398 burn patients were identified from January 1994 to February 2008 of which 232 patients met our inclusion criteria of 0 to 18 years old with more than 10% TBSA scald burns. Of the 232 patients, 90 were women (39%) and 142 were men (61%). Racial distribution included
D.V. Rocourt et al. Table 1 Comparison of patients with scald injuries requiring endotracheal intubation with patients not requiring intubation Patients who Patients who P value (SD) were intubated were not (n = 12) intubated (n = 220) Age (y) Weight (kg) % TBSA ISS LOS (d) ICU stay (d) Presence of abuse
1.4 11.1 35.8 21.4 49.8 26.9 33%
3.0 18.2 15.4 10.2 10.9 0.5 10%
.025 (0.8; 3.3) .022 (2.7; 14.3) b.001 (13.8; 5.6) b.001 (5.6; 4.5) b.001 (45; 10.2) b.001 (25.1; 2.3) .033
47 African American (21%), 151 white (66%), and 31 other (14%). The average age of the population was 1.6 years; average weight, 13 kg; average ISS, 9.0; and average TBSA, 20%. Burns were the result of hot water in 96%, hot grease in 3%, and hot wax in 1%. Of the 232 total patients, 220 patients did not require intubation. The mean age of these nonintubated patients was 3 years; weight, 18.2 kg; percentage of TBSA burned, 15.4%; ISS, 10.2; length of stay (LOS), 10.9 days; and ICU stay, 0.5 days (Table 1). Of the 232 total patients, 12 patients (5%) required endotracheal intubation (Table 1). In general, intubated patients were younger in age, smaller in size, had a larger percentage of TBSA burn, longer LOS and ICU stay, and higher ISSs than nonintubated patients. The mean age of scalded patients requiring intubation was 1.4 years (compared with 3 years in the nonintubated patients; P = .025). The mean weight of the intubated patients was 11.1 kg (compared with 18.2 kg in the nonintubated patients; P = .022). The mean TBSA of the intubated patients was 35.8% (compared with 15.4% in the nonintubated patients; P b .001). Intubated patients had a mean ISS of 21.4 (compared with 10.2 in nonintubated patients; P b .001). Mean length of hospital stay was 49.8 days (compared with 10.9 days in the nonintubated patients; P b .001), and mean ICU stay was 26.9 days (compared with 0.5 days in the nonintubated patients; P b .001). Evidence of abuse was associated with 33% of the patients who required intubation compared with 10% of nonintubated patients (P = .033). Details regarding the 12 patients who required intubation are provided in Table 2. No patient older than 3 years or who had a scald injury that covered less than 15% TBSA required intubation. The mean time to intubation and mechanical ventilation was 19 hours post injury (SD, 24.9). The mean duration of intubation was 18.8 days (SD, 16.7). These patients were resuscitated using the Parkland formula (3-4 mL Ringer lactate per kilogram per percentage of TBSA burn), with patients weighing less than 20 kg who had burns more than 15% TBSA receiving maintenance fluids 5% Dextrose Ringers Lactate (D5RL) in addition. The mean volume of fluids (resuscitation fluid plus maintenance fluid when
Respiratory failure after pediatric scald injury Table 2
1755
Demographics of patients requiring intubation
Patient no.
Race
Sex
Age (y)
Weight (kg)
ISS
Hours to ETT
Duration of ETT (d)
% TBSA burn
IVF first 24 h (mL/kg/% TBSA burn)
UO in first 24 h (mL/kg)
1 2 3 4 5 6 7 8 9 10 11 12 Mean values
White Black White Black White White White White White White White Black
Male Male Male Female Female Male Female Male Female Female Male Male
1.2 2.2 2.75 0.83 0.75 1.08 0.75 1.0 1.17 1.08 1.25 3.0 1.4
15 13.5 13 7.5 8.0 10 9.0 10 13 9.0 10 15.6 11.1
16 16 25 25 25 25 25 9 16 25 25 25 21
0 4 3.5 48 40 1 6 48 0 72 2 8 19
1 1 4 23 28 23 29 3 21 27 14 60 18.8
25 20 30 55 50 50 30 15 25 40 40 55 35.8
5.8 7.4 6.9 8.4 9.9 5.5 6.5 4.55 8.58 9.9 8.62 8.99 7.6
1.18 1.8 1.58 2.01 1.25 1.1 1.19 2.05 2.03 1.57 1.99 1.85 1.6
ETT indicates endotracheal tube; UO, urine output; IVF, intravenous fluid.
appropriate) administered to the intubated patients was 7.6 mL/kg per percentage of TBSA burn in the first 24 hours (SD, 1.8). The mean hourly urine output of the intubated patients was 1.6 mL/kg (SD, 0.4). Indications for intubation did not correlate with resuscitation fluid volumes. None of the patients in the intubated group had oropharyngeal burns or laryngeal edema as evidenced by direct laryngoscopy at the time of intubation. Patients intubated at referring hospitals all underwent direct laryngoscopy by the surgical critical care or pediatric ICU (PICU) staff upon arrival to our PICU. There was no sex or racial differences predictive of the need for intubation. In the entire group of 232 scalded patients, burns to the face, neck, and chest were the most common location, with a frequency of 73%. Of the 12 patients requiring intubation, 9 (75%) sustained burns predominantly to their face and chest, with the remaining 3 patients (25%) having burns localized to the upper and lower extremities and lower torso. There were no statistical differences between burn location in patients requiring or not requiring mechanical ventilation (P = 1.0). The reasons for intubation in the 12 patients were as follows. In 3 cases (25%), initial intubation was related to airway protection for transport to a level 1 burn center. These patients required mechanical ventilation for 1, 21, and 14 days, respectively. One of those patients developed acute respiratory distress syndrome (ARDS) and pneumonia. An additional 3 patients (25%) were intubated prophylactically before the development of respiratory distress. Of these 3 patients, 1 was electively intubated in the emergency department because of severe facial burns; the second was intubated in the ICU for pain control and wound care, and the third was intubated for an immediate escharotomy that was performed in the operating room and remained initially intubated for pain control during wound care. These patients required mechanical ventilation for 4, 3, and 60 days, respectively. The remaining 6 patients
(50%) were intubated for respiratory distress (stridor or labored breathing) and required mechanical ventilation for 1, 23, 28, 23, 29, and 27 days. Complications that developed in the intubated patients were as follows: 10 developed pneumonia, with 6 of the 10 developing ARDS requiring High Frequency Oscillating Ventillation (HFOV). Pneumonia was defined by either a new infiltrate on Chest radiogragh (CXR) or increased oxygen requirements with culture-positive bronchoalveolar lavage. The most common organisms cultured from bronchoalveolar lavage specimens were Candida albicans and Streptococcus viridians. One patient required tracheostomy for prolonged intubation. Other complications included line sepsis in 1 patient, abdominal compartment syndrome owing to chylous ascites in 2 patients, and respiratory arrest in 2 patients. Respiratory arrest was secondary to a mucus plug in the tracheostomy tube in 1 patient and a mucus plug in the endotracheal tube of the other patient. Both resolved promptly with suctioning with no further sequellae. There was no mortality, with all patients surviving their scald burn injuries. In our group of patients requiring intubation, 7 of the 12 patients developed ARDS as demonstrated by a PaO2/FIO2 greater than 200 and by bilateral patchy infiltrates. None of the intubated patients had a history of asthma, reactive airway disease, or were reported to have symptoms consistent with a viral upper respiratory tract infection before admission. There were 3 patients who were prophylactically intubated for transport to our facility. Despite the initial reason for intubation, 2 of these patients required mechanical ventilation for greater than 10 days. These 2 patients (40% and 25% TBSA burns, respectively) had a mean hourly urine output of 1.99 and 2.03 mL/kg, respectively, and received a volume of 8.62 and 8.58 mL/kg per percentage of TBSA, respectively. The third patient (25% TBSA burn) who did not require prolonged intubation
1756 Table 3
D.V. Rocourt et al. Predictors of endotracheal intubation
Predictor
Odds ratio
P
Age Weight % TBSA ISS Injury location Abuse, yes or no
NA NA 2.9 (1.9, 4.6) 4.6 (2.6, 8.2) NA 4.5 (1.3, 16.2)
.31 .16 b.001 b.001 .87 .034
NA indicates not applicable.
had a mean hourly urine output of 1.18 mL/kg and received a volume of 5.8 mL/kg per percentage of TBSA. Of the 2 patients requiring prolonged mechanical ventilation, 1 went on to develop ARDS. In contrast to these patients, the 17 patients admitted to the ICU who did not require intubation had a mean hourly urine output of 2.4 mL/kg (range, 1.753.33 mL/kg per hour) with a mean volume of resuscitation of 7.2 mL/kg per percentage of TBSA (range, 3.93-10.36 mL/ kg per percentage of TBSA). Although ascites is well described after burn resuscitation, chylous ascites is unusual [6,7]. Interestingly, 2 of the patients requiring intubation in our study developed chylous ascites within the first 3 days of burn injury. The first patient was a 9-month-old girl with a 30% TBSA burn. Her resuscitation required 6.5 mL/kg crystalloid during the first 24 hours after injury to maintain an average hourly urine output of more than 1 mL/kg. She developed Abdominal Compartment Sydrome (ACS) with bladder pressures greater than 30 mm Hg and an abdominal ultrasound confirming ascites. She developed worsening respiratory failure necessitating HFOV. She underwent decompression with a pigtail catheter. The fluid drained was milky in appearance and
Table 4
contained 88% lymphocytes. Her enteral formula was changed to Vivonex, a short-chain fatty acid formula, with resolution of her chylous ascites. She was weaned to a conventional ventilator by the third day of HFOV. The second patient was a 13-month-old girl with a 40% TBSA burn whose resuscitation required 9.9 mL/kg of crystalloid during the first 24 hours post burn to maintain an hourly urine output of more than 1 mL/kg. She developed abdominal distension and increased oxygen requirements on the ventilator. An abdominal ultrasound confirmed ascites that was drained with a pigtail catheter. The ascitic fluid was milky in appearance and contained 90% lymphocytes. Her enteral formula was changed to Vivonex, also resulting in resolution of her chylous ascites as demonstrated by repeat peritoneal fluid cell count. It is clear that volume of resuscitation contributes to the development of ACS; however, it is unclear why these 2 patients developed this complication when their hourly urine output was below the mean. In addition, both of these patients had transthoracic echocardiograms performed to assess their volume status and cardiac function, which were within normal limits with no evidence of fluid overload. Univariate logistic regression was used to determine the independent need for intubation in patients younger than 4 years with TBSA burn greater than 10% (Table 3). Both ISS and TBSA were independent predictors of patients requiring mechanical intubation. For each 5-point increase in ISS, patients had 4.6 times higher odds of requiring mechanical ventilation. For each 5% increase in TBSA, patients had a 2.9 times higher odds for intubation. Interestingly, evidence of abuse was a significant factor associated with the need for endotracheal intubation (P b .031). Patients with scald burns owing to child abuse had 4.5 times higher odds of intubation
Demographics of patients with scald burns admitted to the ICU who did not require endotracheal intubation
Patient no.
Race
Sex
Age (y)
Weight (kg)
ISS
% TBSA burn
IVF first 24 h (mL/kg/% TBSA burn)
UO in first 24 h (mL/kg)
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 Mean values
White White Other White White White Black White White Other White Black White Other Black White Black
Male Male Male Female Female Male Female Female Male Female Male Male Female Male Male Male Female
1.16 .5 4 2 1.16 6 1.33 1.25 1.08 3 3 3 1.16 1.42 1.08 1.42 2 2.1
13 8.7 20 11.4 8.6 23 9 10 9.2 16.7 15.8 17 12 12.2 11.8 11.1 10 14.0
9 16 25 9 25 9 9 25 16 25 25 16 9 16 16 16 16 14.8
15 20 32 15 30 15 15 35 23 30 30 18 15 25 20 25 15 21.2
6.46 8.11 7.4 7.98 10.5 6.3 7.22 6.6 7.72 3.93 5.9 6.11 8.67 6.25 10.36 4.95 7.91 7.2
1.98 2.23 2.56 2.03 2.01 1.75 2.21 2.13 2.19 2.43 2.39 3.25 2.61 2.25 3.07 3.33 2.01 2.4
Respiratory failure after pediatric scald injury than nonabused patients. Older age was associated with a reduced need for endotracheal intubation. In a multivariable logistic regression, only TBSA burn remained statistically significant, whereas ISS and abuse did not provide the information about mechanical ventilation requirement beyond the information contained in TBSA (P b .03). Of the 232 patients who met inclusion criteria (0-18 years old with greater than 10% TBSA scald burns), there were 17 patients admitted to the PICU who did not require endotracheal intubation (Table 4). The 17 patients admitted to the PICU who were not intubated were compared with the 12 patients admitted to the PICU requiring endotracheal intubation (Table 5). The mean age of nonintubated PICU patients was 2.1 years compared with 1.4 years in the intubated patients (P b .07). The mean weight in the nonintubated PICU patients was 14 kg compared with 11.1 kg in the intubated patients (P b .17). The mean percentage of TBSA burned in the nonintubated PICU patients was 21.2% compared with 35.8% in the intubated patients (P b .01). Nonintubated PICU patients had a mean ISS score of 14.8 compared with 21.4% in the intubated patients (P b .02). As expected, mean LOS and ICU stay were statistically greater in PICU patients requiring intubation compared with nonintubated PICU patients (P = .05 and P b .001, respectively). Interestingly, the presence of abuse was statistically greater in the intubated PICU patients compared with nonintubated ICU patients (P b .03). Importantly, there were no statistically significant differences in the mean volume of resuscitation fluids in the first 24 hours after injury in nonintubated compared with intubated PICU patients. Nonintubated patients received a mean volume of 7.2 mL/kg per percentage of TBSA burned (SD, 1.7) in the first 24 hours post injury compared with 7.6 mL/kg per
Table 5 Comparison of patients admitted to the ICU with scald injuries requiring endotracheal intubation with ICU patients not requiring intubation
Age (y) Weight (kg) % TBSA ISS Injury location, face/chest:other Abuse (%) LOS (d) ICU (d) IVF first 24 h (mL/kg/% TBSA burn) UO in first 24 h (mL/kg)
ICU patients who were intubated (n = 12)
ICU patients P value (SD) who were not intubated (n = 17)
1.4 11.1 35.8 21.4 9:3
2.1 14 21.2 14.8 14:3
.07 (0.8; 1.4) .17 (2.7; 5.3) .01 (13.8; 6.3) .02 (5.6; 6.0) .44
33 49.8 26.9 7.6
10 23.3 1.9 7.2
.03 .05 (45; 12.1) b.001 (25.1; 3.1) .54 (1.8;1.7)
1.6
2.4
.001 (0.4;0.5)
1757 percentage of TBSA burned (P = .54). Nonintubated patients had a statistically greater mean hourly urine output of 2.4 mL/kg per hour in the first 24 hours compared with 1.6 mL/kg per hour in intubated patients (P = .5). There were no complications in the patients admitted to the PICU who did not require intubation. Of the 232 patients in the study, 29 were admitted to the PICU (12 intubated and 17 nonintubated), and 203 did not require PICU admission. Most patients in the study had partial-thickness burns, with a few sustaining full-thickness burns requiring surgical intervention. Of the 203 patients who were not admitted to the PICU, 16 required operative intervention, including 6 patients who underwent burn debridement followed by dressing changes and 10 patients who required split-thickness skin grafting. Operative procedures were performed in 7 of the 12 intubated PICU patients, all of whom required split-thickness skin grafting. Operative procedures were performed in 10 of the 17 nonintubated PICU patients, including 5 patients who underwent burn debridement and 5 patients who requited split-thickness grafting.
3. Discussion Respiratory failure after cutaneous scald burn injury is relatively uncommon, with few reports in the literature to date [2,3,8,9]. Our study represents the largest series of patients with cutaneous scald burns requiring mechanical ventilation in the absence of direct airway injury to date and corroborates 2 older but smaller reports. In our study, the prevalence of patients requiring intubation after a scald burn was 0.8 or less than 1 patient a year. Despite its rarity, we were able to identify some factors that may contribute to this event. Hudson et al [3] reported 7 patients with an average age of 18 months who required mechanical ventilation after scald injuries. Three of these patients died, with 2 of the 3 patients having sustained direct thermal injury to the respiratory tract from ingestion of hot liquids. Zak et al [4] reported 7 patients (4%) of their scalded patients (mean age, 1.4 years; mean percentage of TBSA burn, 29.9%) required mechanical ventilation. In their series, the ventilated patients were younger than 2 years old with a TBSA burn of more than 20%. In our current study, 12 (5%) of scalded patients (mean age, 1.4 years; mean percentage of TBSA burn, 38.5%) required mechanical ventilation, with none of the patients having direct oropharyngeal burns. Adequate volume of resuscitation in burn patients is critical. Urine output has been the major determinant of adequacy of fluid resuscitation. In our study, we found that patients requiring endotracheal intubation had a mean hourly urine output of 1.6 mL/kg per hour with a total mean volume of resuscitation of 7.6 mL/kg per percentage of TBSA over the first 24 hours. Zak et al reported a low average hourly urine output (0.84 mL/kg per hour) within the first 24 hours after injury,
1758 arguing that fluid resuscitation was not excessive in their intubated patients. Although the mean volume of resuscitation in our group was higher than that that reported by Zak et al, the percentage of patients requiring endotracheal intubation is similar. Because fluid over resuscitation might lead to the development of respiratory failure, we compared patients admitted to the PICU who did not require mechanical ventilation to those who did. There were no differences in age, size, and, more importantly, volume of resuscitation in the first 24 hours after injury in these 2 groups. In fact, we found that patients who did not require intubation received an average volume of 7.2 mL/kg per percentage of TBSA in the first 24 hours after injury and had a mean hourly urine output of 2.4 mL/kg compared with a mean hourly urine output of 1.6 mL/kg in the intubated patients. Patients who required mechanical ventilation had larger TBSA burns and higher ISS. In both multivariate and univariate logistic regressions, TBSA burn had a higher odds for requiring endotracheal intubation. Although volume of resuscitation was not a statistically significant factor for requiring mechanical ventilation, its contribution could not be definitively excluded because of the relatively small number of patients requiring intubation. In our study, fluids in excess of the calculated resuscitation needs were administered only as needed for low urine output (b1 mL/kg per hour) or for hemodynamic instability (hypotension). To further eliminate fluid overresuscitation as a contributing factor for the need for intubation, we have developed a nurse-directed fluid resuscitation protocol in our burn unit. This protocol is designed to ensure that fluid resuscitation volumes are decreased when hourly urine outputs are high. Because the pathophysiology of respiratory failure in scalded patients remains unclear, it is possible that perturbations in the immune response may play a contributory role. Furthermore, because all of the patients who required intubation in our study had scald burns greater than 15% TBSA and because burns greater than 15% TBSA have been shown to trigger a systemic inflammatory response, it is possible that activation of SIRS may play a role [5]. Burn injury is a trigger for activation of inflammatory mediators including proinflammatory cytokines [5,10]. Serum concentrations of interleukin-6 and tumor necrosis factor-α have been found to be elevated in the early stages of burn injury [9]. KowalVern et al [11] reported a significant increase in plasma interleukin-6, tumor necrosis factor-α, and other inflammatory cytokines in burn patients with and without inhalation injury. Furthermore, Stengle et al [12] used a mouse model of 15% TBSA scald burn to demonstrate a significant elevation of pulmonary neutrophil infiltration within 6 hours of thermal injury and at 1 week postburn. In addition, activation of the SIRS typically leads to lung injury, pulmonary hypoperfusion, and respiratory failure [13]. To further elucidate the possible role of an overexuberant immune response in scald burn patients, we plan to examine parameters of the innate and acquired immune responses in these patients in the future.
D.V. Rocourt et al. In summary, the requirement for intubation, although relatively uncommon after scald injury in pediatric patients, is under appreciated. This complication should be anticipated in approximately 5% of patients, especially in patients younger than 3 years with greater than 15% TBSA scald burns. Volume of resuscitation should be tailored to maintain an average hourly urine output of 1 mL/kg with the goal of maintaining end organ perfusion; however, overresuscitation should be carefully avoided. In the current study, although fluid over resuscitation did not appear to be a major factor in the development of respiratory failure, this possibility could not be definitively excluded because of the relatively small number of patients requiring intubation. Therefore, a nurse-directed fluid resuscitation protocol has been developed in our burn unit to ensure that fluid overresuscitation is not playing a role, specifically ensuring that fluid resuscitation volumes are decreased when hourly urine outputs are high.
References [1] Drago DA. Kitchen scalds and thermal burns in children five years and younger. Pediatrics 2005;115:10-6. [2] Guzel A, Aksu B, Aylanç H, et al. Scalds in pediatric emergency department: a 5-year experience. J Burn Care Res 2009;30: 450-6. [3] Hudson DA, Jones L, Rode H. Respiratory distress secondary to scalds in children. Burns 1994;20:434-7. [4] Zak AL, Harrington DT, Barillo DJ, et al. Acute respiratory failure that complicates the resuscitation of pediatric patients with scald injuries. J Burn Care Rehabil 1999;20:391-9. [5] Nakae H, Motoyama S, Kurosawa S, et al. The effective removal of proinflammatory cytokines by continuous hemofiltration with a polymethylmethacrylate membrane following severe burn injury: reports of three cases. Surg Today 1999;29:762-5. [6] Moore-Olufemi SD, Xue H, Allen SJ, et al. Effects of primary and secondary intra-abdominal hypertension on mesenteric lymph flow: implications for the abdominal compartment syndrome. Shock 2005;23:571-5. [7] Ida J, Yamashita K, Inoue T, et al. Resuscitation fluid volume and abdominal compartment syndrome in patients with major burns. Burns 2006;32:151-4. [8] Desai MH, Nichols MM, Herndon DN. Scald injury of the respiratory tract: an unusual occurance. J Burn Care Rhabil 1987;8:210-2. [9] Bjorkl L, Svensson H. Upper airway obstruction—an unusual complication following a minor scalding injury. Burns 1999;19: 85-7. [10] Kowal-Vern A, et al. Circulating endothelial cell levels correlate with proinflammatory cytokine increase in the acute phase of thermal injury. J Burn Care Rehabil 2005;26:422-9. [11] Kowal-Vern A, Ortegel J, Bourdon P, et al. Elevated cytokine in peritoneal fluid from burned patients with intra-abdominal compartment syndrome. Burns 2006;32:563-9. [12] Stengle J, Meyers R, Pyle J, et al. Neutrophil recruitment after remote scald injury. J Burn Care Rhabil 1996;17:14-8. [13] Oda J, Yamashita K, Inoue T, et al. Acute lung injury and multiple organ dysfunction syndrome secondary to intra-abdominal hypertension and abdominal decompression in extensively burned patients. J Trauma 2007;62:1365-9.