Secondary Logo

Journal Logo

Basic Science Aspects

High Versus Low Volume Fluid Resuscitation Strategies in a Porcine Model (Sus scrofa) of Combined Thermal and Traumatic Brain Injury

Guenther, Timothy M.∗,†; Spruce, Marguerite W.∗,†; Bach, Lindsey M.∗,†; Caples, Connor M.∗,†; Beyer, Carl A.∗,†; Grayson, John K.; Meyers, Frederick J.; Palmieri, Tina L.∗,§; Brown, Ian E.

Author Information
doi: 10.1097/SHK.0000000000001658
  • Free
  • Editor's Choice



Thermal burns cause an estimated 50,000 hospital admissions yearly in the United States (1). Factors affecting mortality in burn injury include age, total body surface area (TBSA) burn, and presence of inhalation injury (2). Advances in resuscitation, surgical treatment, and rehabilitation have improved outcomes over the last 20 years; however, burn injuries continue to cause significant morbidity and mortality (3). Similarly, traumatic brain injury (TBI) represents another unique injury, responsible for an estimated 235,000 hospital admissions each year in the United States (4). Typically caused by rapid acceleration/deceleration forces to the head, TBI severity is dependent on the force and frequency of impact (5). Although the understanding of mechanisms of injury and pathophysiology in TBI have improved, significant knowledge gaps remain regarding the effects of management strategies on short and long-term outcomes (6).

The combined injury of burn and TBI is a historically unique pattern of poly-trauma. However, this injury pattern is increasing in the military population due to the increased use of improvised explosive devices (IED) in modern warfare. A review of 951 service members treated with burn injuries at the US Institute of Surgical Research burn center found 95 (10.0%) had a concomitant TBI (7). In this cohort, a combined burn injury and TBI was associated with increased complications including increased risk of intensive care unit stay, prolonged ventilation, higher rates of amputation, bacteremia, urinary tract infections, wound infections, and thromboembolism compared with isolated burn injury or TBI. This highlights the challenging aspects of managing this combined injury pattern and reiterates the need for a deeper understanding of the physiologic response to these two unique insults.

A major problem in the management of patients with combined burn injury/TBI is the opposing fluid management paradigms. Patients with major burns require high-volume fluid resuscitation to replace intravascular fluid depletion due to increased vascular permeability from a robust inflammatory response and increased insensible losses (8). At most centers this is accomplished by infusing a crystalloid solution, typically Lactated Ringers (LR), based on commonly used formulas such as the Parkland or the modified Brooke formula (9). In contrast, TBI patients require limitation of crystalloid infusion and avoidance of hyponatremia to minimize cerebral edema and prevent secondary brain injury (10). Given the conflicting resuscitation strategies of these two injury patterns, use of an intracranial pressure (ICP) monitor has been advocated to help guide fluid management and maintain ICP < 20 mm Hg (11). However, the optimal resuscitation strategy in these patients remains unknown. The aim of this study was to compare a more “aggressive” early fluid resuscitation strategy (using the Parkland formula) to a more “restrictive” fluid resuscitation strategy (using the modified Brooke formula) in a large animal model of combined burn injury/TBI.


Animal preparation

The Institutional Animal Care and Use Committee at David Grant Medical Center, Travis Air Force Base approved this study (FDG20180017A). All animal care and use were performed in strict compliance with the Guide for the Care and Use of Laboratory Animals in a facility accredited by the Association for Assessment and Accreditation of Laboratory Animal Care International.

After 12 h of no access to food (but access to water), castrated male Yorkshire-cross swine (4–5 months of age, 50 kg–75 kg) without evidence of disease were induced with a Tiletamine/Zolazepam intramuscular injection. Endotracheal intubation was performed and gaseous isoflurane anesthesia administered at 1.5% to 3.0% in 2 L/min oxygen with a variable FiO2 to maintain SpO2 above 90%. Positive end-expiratory pressure was kept at 5 cm H2O with tidal volumes of 6 mL/kg and a variable respiratory rate to maintain an end tidal CO2 of 40 ± 5 mm Hg. A 7 F triple lumen central line was inserted in the external jugular vein, suprapubic urinary catheter placed, and a 7 F femoral arterial line placed (Fig. 1).

Fig. 1
Fig. 1:
Schematic diagram of experimental set-up and implementation.

Burn injury model

Full-thickness contact burns were created using steel bars (1.7 kg) heated to at least 275°C with a hot plate. The temperature was confirmed with a handheld infrared thermometer (model number MS6530H, Commercial Electric Products, Cleveland, Ohio). The plates were transferred from the hot plate to the animal using adjustable tongue and groove pliers (Channellock, Meadville, Pa), and no additional pressure was applied beyond the bar's own weight. The amount of time required to induce a full-thickness burn was determined by placing the heated bars onto the animal's flank in this manner for incremental time periods. This was done for two model development swine, and representative tissue samples were then taken from each contact time region and reviewed by a pathologist to determine the thickness of the resulting burn.

Once the contact time for a full-thickness burn was determined, 40% TBSA full-thickness burns were created in subsequent experimental animals. The TBSA of the animal was calculated using the known formula [TBSA (cm2) = 734 × kg0.656], and the burn size in cm2 was calculated by multiplying the TBSA by 0.4. The number of bars to use was calculated by dividing the burn size in cm2 by the surface area of one bar (117.75 cm2). To create the burn, the animal was placed in the lateral decubitus position, and the bars were placed on the animal's flank for the determined contact time to induce a full-thickness burn starting 8 cm lateral to the spine and continuing ventrally, avoiding the nipple line due to the risk of hemorrhage from large subcutaneous veins. After turning the animal, this was repeated on the contralateral flank until an adequate 40% TBSA burn was complete.

Traumatic brain injury

The TBI was then induced based on a model previously published by this lab (12). The animal was placed in the prone position, and a skin flap was created to expose the frontal skull. A 15 mm circular burr hole was created to expose but not violate the dura. A second 3 mm circular burr hole was placed away from the initial burr hole and an ICP monitor was placed (Codman ICP Express, Integra Life Sciences, Princeton, NJ) The animal's head was then secured within a custom stereotactic frame. A cortical impact device (Custom Design and Manufacturing, Richmond, Va) then deployed a 10 mm rounded polymer striking tip, in which a piezoelectric impact monitor (PCB Piezotronics, Depew, NY) connected to an oscilloscope was embedded. Impact (fixed velocity of 4.0 m/s, depth of 12 mm, and dwell time of 200 μsec) marked the beginning of 8 h of resuscitation (T0). Bone wax was used to cover the burr hole and the skin flap was closed.


Swine were then randomized to receive either an aggressive fluid resuscitation using the Parkland formula (4 cc/kg/% TBSA with half given in the first 8 h) or to a more restrictive fluid resuscitation using the modified Brooke formula (2 cc/kg/% TBSA with half given in the first 8 h). Following T0, hypertonic saline (HS) was administered to induce hypernatremia (10% HS with a target plasma sodium of 150 mEq/L–155 mEq/L), a standard intervention for TBI to reduce brain edema. (Notably, 10% HS was chosen over the commonly used 3% HS to minimize the additional volume required to maintain sodium goals because we observed significant resistance to alterations in serum sodium using lower concentrations) Norepinephrine was used to maintain mean arterial pressure (MAP) > 65 mm Hg. Resuscitation continued for 8 h during which time heart rate, blood pressure, central venous pressure, ICP, and urine output were measured. Labs were obtained at periodic intervals throughout the resuscitation. After 8 h the swine were euthanized and the brain was immediately retrieved chilled at −80°C for 30 min, sliced into 5 mm coronal slices, stained for 30 min in 1% 2,3,5-tri-phenyl tetrazolium chloride (Millipore Sigma, St. Louis, Mo), and fixed in 10% buffered formalin solution. The size of brain injury was determined by taking digital images of the front and back of each 5-mm brain slice, after which time the volume could be calculated on microscopic evaluation using image analysis software (ImageJ, NIH, Bethesda, Md) by a blinded pathologist.

A complete necropsy was then performed. Skin, kidney, liver, and brain tissue were collected and stored in 10% buffered formalin solution. After fixation, tissues were routinely processed, stained with hematoxylin and eosin, and reviewed by a pathologist blinded to study assignment. In addition to multiple sections from each brain lesion stained with hematoxylin and eosin, immunohistochemistry was performed on a selected paraffin-embedded brain section from each pig in the conventional manner to quantify the presence of β-amyloid precursor protein.

Data analysis

The primary outcome of interest was the size and degree of brain injury between the high-volume and low-volume resuscitation groups. To detect a 2-fold difference in brain lesion size between groups with similar standard deviations, an effect size of 1 was chosen. Using G∗Power 3.0.1 with an effect size of 1, alpha of 0.05, power of 0.8, and equal allocation ratios, a sample size of 14 animals per group was required. Secondary outcomes of interest included urine output, time spent at goal MAP during the resuscitation phase, time spent at goal serum sodium, laboratory assessments, and histologic assessments of the lung, liver, brain, and kidney. Histologically, each tissue was scored as: 0—no lesions seen, 1—focal lesion present, 2—multifocal lesions, 3—locally extensive lesions, or 4—generalized or diffuse lesions, and the lesions were described (13). For each pig the presence or absence of a dural tear and so-called “burst” cortical lobe (physical disruption of the neural parenchyma with laceration) were noted on gross examination. Select brain sections were further reviewed for presence or absence of subarachnoid hemorrhage and scored for severity of intraparenchymal hemorrhage, “red dead neurons” (coagulation necrosis with pyknotic nuclei, loss of nissl substance, intensely staining eosinophilic cytoplasm, and cell shrinkage), neuronal plasticity, neuropil vacuolation, and β-amyloid precursor protein (Fig. 2).

Fig. 2
Fig. 2:
Representative histologic pathologies of intracranial injury. Intraparenchymal hemorrhage, hematoxylin and eosin stain, ×4, scale bar equals 500 μm (A). Neuronal coagulation necrosis and cellular streaming, hematoxylin and eosin stain, ×20, scale bar equals 100 μm (B). Subarachnoid hemorrhage (arrows), hematoxylin and eosin stain, ×4, scale bar equals 500 μm (C). Neuronal coagulation necrosis, intraparenchymal hemorrhage and neuropil vacuolation (asterisks), hematoxylin and eosin stain, ×10, scale bar equals 200 μm (D).

Results are presented as either means ± standard deviation or medians (interquartile range) depending on the distribution of the underlying data. Student t tests or Wilcoxon rank-sum tests were used to assess contrasts at baseline and at the end of study using a commercially available software program (STATA version 15, Stata Corp, College Station, Tex). Hemodynamics data were plotted and evaluated with generalized estimating equations. Statistical significance was set atP = 0.01 to account for the large number of comparisons made.


Burn model development

Microscopic evaluation of dermal samples taken from incremental contact times in model development swine determined that 30 s of contact with the steel plate heated to 275°C produced a full-thickness burn (Fig. 3).

Fig. 3
Fig. 3:
Pathologic assessment of depth of thermal injury based on contact time with steel plate heated to at least 275°C. Normal porcine skin with intact dermis and adnexa (A). Representative porcine skin after 20 s partial thickness burn margin—with carbonized epidermis, coagulation necrosis of dermis, partially intact subdermal adnexa (B). Representative porcine skin after 30 s full-thickness burn margin—complete carbonization epidermis and coagulation necrosis of dermis, loss of subdermal adnexa (C). (All figures ×2, bar equals 1 mm.).

Experimental data

A total of 28 pigs underwent randomization with 14 receiving aggressive resuscitation with the Parkland formula and 14 receiving restrictive resuscitation with the modified Brooke formula (Table 1). Swine in the restrictive group on average weighed less (68.4 kg ± 7.8 vs. 76.0 kg ± 9.2, P value 0.03) and had a higher baseline heart rate (100 beats/min ± 12 vs. 89 beats/min ± 13, P value 0.04). No differences were observed between groups regarding baseline labs. During the 8 h of resuscitation, both groups received similar amounts of norepinephrine to maintain goal MAP > 65 mm Hg (35.8 mcg/kg vs. 31.9 mcg/kg, P value 0.92) and similar volumes of hypertonic saline to maintain sodium 150 to 155 (10.3 ± 1.7 vs. 11.3 ± 1.8, P value 0.12) (Table 2). However, swine in the restrictive group spent less time in the goal MAP range (51.8% vs. 58.3%, P valve < 0.01). Swine in the aggressive resuscitation group received more LR during resuscitation (76.1 mL/kg ± 20.6 vs. 42.9 mL/kg ± 11.6, P value < 0.01) and as expected gained more weight during the resuscitation (2.3 kg ± 1.3 vs. 1.1 kg ± 1.1, P value 0.01).

Table 1 - Characteristics at baseline (mean ± SD or median [IQR] depending on distribution)
Variable or analyte Restrictive (n = 14) Aggressive (n = 14) P
Weight (kg) 68.4 ± 7.8 76.0 ± 9.2 0.03
Core temperature (°C) 35.7 ± 3.8 34.5 ± 5.9 0.55
Percent TBSA (%) 40.0 ± 0.6 40.0 ± 0.6 0.98
Heart rate (beats per minute) 100 ± 12 89 ± 13 0.04
Mean arterial pressure (mm Hg) 65 [56–73] 60 [56–70] 0.43
Central venous pressure (mm Hg) 8 [5–12] 11 [9–13] 0.08
Intracranial pressure (mm Hg) 17 [13–30] 17 [4–47] 0.69
SPO2 (%) 97 [96–98] 98 [94–99] 0.84
Preinjury blood loss (mL/kg) 0.9 ± 0.4 0.8 ± 0.5 0.43
White blood cell count (×109/L) 15.0 ± 4.7 13.7 ± 3.1 0.42
Hematocrit (%) 27.0 ± 3.3 25.3 ± 3.3 0.18
Platelet Count (×109/L) 273 [247–364] 287 [268–334] 0.82
Fibrinogen (mg/dL) 403 ± 65 401 ± 62 0.95
Alkaline phosphatase (U/L) 106 [90–138] 92 [84–108] 0.09
Alanine transaminase (U/L) 39 ± 13 39 ± 11 0.90
Aspartate transaminase (U/L) 11 [10–15] 13 [10–18] 0.60
Albumin (g/dL) 2.5 [2.4–2.7] 2.6 [2.4–2.6] 0.53
Blood urea nitrogen (mg/dL) 8.4 ± 1.7 8.0 ± 2.0 0.55
Creatinine (mg/dL) 1.2 ± 0.1 1.3 ± 0.2 0.16
Total bilirubin (mg/dL) 0.3 ± 0.1 0.4 ± 0.1 0.06
pH 7.47 ± 0.05 7.44 ± 0.04 0.08
pO2/FiO2 (mm Hg/%) 470 ± 63 448 ± 109 0.52
K + (mmol/L) 3.2 [3.1–3.5] 3.4 [3.1–3.7] 0.50
Glucose (mg/dL) 94 ± 24 98 ± 19 0.68
Lactate (mmol/L) 2.8 [2.6–3.0] 3.0 [2.6–3.4] 0.34
At T30, otherwise at T0.
TBSA indicates total body surface area burned.

Table 2 - Characteristics at end of study (mean ± SD or median [IQR] depending on distribution)
Variable or analyte Restrictive (n = 14) Aggressive (n = 14) P
Weight gain (kg) 1.1 ± 1.1 2.3 ± 1.3 0.01
Core temperature (°C) 38.4 [38.0–39.3] 38.7 [38.4–39.4] 0.23
Heart rate (beats per min) 127 [112–133] 113 [107–150] 0.96
Mean arterial pressure (mm Hg) 62 [54–64] 64 [62–65] 0.11
Central venous pressure (mm Hg) 8 [6–11] 10 [7–12] 0.22
Intracranial pressure (mm Hg) 22 [17–42] 23 [16–72] 0.98
SPO2 (%) 98 [97–98] 97 [96–99] 0.74
White blood cell count (×109/L) 14.7 ± 2.7 12.9 ± 3.7 0.16
Hematocrit (%) 33.4 ± 2.9 30.6 ± 3.9 0.04
Platelet count (×109/L) 314 [248–355] 309 [267–365] 0.80
Fibrinogen (mg/dL) 479 ± 89 458 ± 57 0.47
Alkaline phosphatase (U/L) 134 ± 34 114 ± 29 0.10
Alanine transaminase (U/L) 42 ± 12 41 ± 10 0.84
Aspartate transaminase (U/L) 37 [25–66] 36 [27–48] 0.85
Albumin (g/dL) 2.7 ± 0.3 2.5 ± 0.2 0.22
Blood urea nitrogen (mg/dL) 15.0 ± 2.0 12.1 ± 2.5 <0.01
Creatinine (mg/dL) 1.6 ± 0.2 1.6 ± 0.2 0.36
Total bilirubin (mg/dL) 1.4 ± 0.4 1.3 ± 0.7 0.71
pH 7.48 ± 0.04 7.47 ± 0.02 0.58
pO2/FiO ≥ 2 (mm Hg/%) 449 [431–506] 421 [387–480] 0.52
K + (mmol/L) 4.5 ± 0.6 4.4 ± 0.3 0.52
Glucose (mg/dL) 120 ± 23 114 ± 16 0.45
Lactate (mmol/L) 2.6 ± 0.7 2.7 ± 0.8 0.70
Norepinephrine (mcg/kg) 35.8 [21.9–73.7] 31.9 [25.4–70.2] 0.92
Lactated Ringer solution (mL/kg) 42.9 ± 11.6 76.1 ± 20.6 <0.01
10% saline solution (mL/kg) 10.3 ± 1.7 11.3 ± 1.8 0.12
Urine output (mL/kg/h) 5.3 ± 1.8 7.3 ± 1.8 <0.01
Time at MAP ≥ 65 mm (%) 51.8 58.3 <0.01
Time [Na+] at goal (%) 37.3 54.0 0.01
At T420, else T480.
Significantly greater than baseline, P < 0.05.
Significantly greater than baseline, P < 0.01.

Similar trends between groups were observed regarding heart rate, central venous pressure, blood pressure, and ICP throughout the 8 h of resuscitation (Fig. 4). After 8 h, urine output was higher in the aggressive resuscitation group (7.3 mL/kg/h ± 1.3 vs. 5.3 mL/kg/h ± 1, P value < 0.01). An elevated serum blood urea nitrogen (BUN) was observed in the restrictive resuscitation group (15.0 mg/dL ± 2.0 vs. 12.1 mg/dL ± 12.1 ± 2.5, P value < 0.01), although no difference was seen in serum creatinine (1.6 mg/dL ± 0.2 vs. 1.6 mg/dL ± 0.2, P value 0.36). Although higher histologic injury scores were seen in the kidney and brain of the restrictive group, neither reached statistical significance (1.5 vs. 1 for kidney, P valve 0.03 and 3 vs. 2 for brain, P valve 0.08) (Table 3) (Fig. 5). No difference was seen in volume of brain tissue injury or specific intracranial pathologies between groups on histologic evaluation (1.4 cm3 ± 0.5 vs. 1.6 cm3 ± 0.8, P value 0.51) (Table 4).

Fig. 4
Fig. 4:
Median heart rate (A), median central venous pressure (B), median mean arterial pressure (C), and median intracranial pressure (D) during resuscitation using restrictive and aggressive strategies following combined burn and traumatic brain injury in swine.
Table 3 - Histological injury scores (median [IQR])
Tissue Restrictive (n = 14) Aggressive (n = 14) P
Lung 0 [0–1] 0 [0–0] 0.05
Liver 1 [1–2] 1 [0–2] 0.27
Kidney 1.5 [1–3] 1 [0–1.5] 0.03
Brain 3 [2–3] 2 [0.5–3] 0.08

Fig. 5
Fig. 5:
Representative histology of organ injury. Lung showing congestion (A). Liver showing congestion (B). Brain cortex showing intraparenchymal hemorrhage and neuropil vacuolation (C). Kidney cortex showing congestion (D). (All images: hematoxylin & eosin stain, ×10, scale bar equals 200 μm.).
Table 4 - Brain injury volume and specific histological injury scores (median [IQR])
Tissue Restrictive (n = 14) Aggressive (n = 14) P
Brain injury area (cm3) 1.4 ± 0.5 1.6 ± 0.8 0.51
Dural tear 0 [0–1] 0 [0–1] 0.97
“Burst” lobe 0.5 [0–1] 1 [0–1] 0.35
Subarachnoid hemorrhage 1 [1–1] 1 [1–1] 0.12
Intraparenchymal hemorrhage 2 [2–3] 2 [1–2] 0.11
Neuronal coagulation necrosis 3 [2–3] 3 [2–3] 0.80
Neuronal plasticity 2 [2–2] 2 [2–2] 0.40
Neuropil vacuolation 2 [1–2] 2 [1–2] 0.91
ß-Amyloid precursor protein 2 [1–3] 2 [0–2] 0.17


Combined burn injury and TBI is a complex pattern of traumatic injury with a historically low incidence (7). Due to this low incidence, clinicians have been left to determine appropriate treatment approach in the absence of evidence or consensus guidelines. With trends in modern warfare and the increased use of IEDs, the incidence of combined burn injury/TBI is likely to increase, accentuating the need for development of an appropriate animal model. Swine have been used successfully to study cardiovascular disease, nutrition, pharmacology, traumatic injuries/resuscitation, and wound healing (14). The pig is useful for modeling burns due to the similarities between pig and human skin, including epidermal thickness and turnover time, dermal thickness, strong attachments to the subdermal structures, orientation of blood vessels below the dermis, and composition of the underlying fatty tissue (15). Swine burn models involving multiple different mechanisms have been developed, including immersion in hot water and contact with heated glass bottles, aluminum, or brass (16). Swine TBI models have also been developed, including the use of controlled cortical impact devices, fluid percussion, and penetrating objects (12, 17, 18). Our method of establishing combined burn injury/TBI was found to be safe, uniform, efficient, and reproducible.

In the first 24 h after a major burn, the associated inflammatory response causes significant vascular permeability (8). Early aggressive fluid resuscitation in these patients is paramount; delayed and/or inadequate resuscitation has been associated with increased mortality (19). Despite widespread evidence that early fluid resuscitation is associated with improved outcomes, the type and frequency of fluid used for resuscitation is controversial (20). Comparison studies between the Parkland and modified Brooke formula for resuscitation have been inconsistent, and no formula has shown significant advantage over the other (21). Similar to the deleterious effects of under-resuscitation, over resuscitation in burn injury (typically > 0.25 L/kg) during the first 24 h of resuscitation can lead to increased tissue swelling and complications such as abdominal compartment syndrome, extremity/truncal compartment syndrome, or cerebral edema (22). For this reason, the 2018 Advance Burn Life Support update has advocated for use of the modified Brooke formula in adults with chemical or thermal burns as the initial resuscitation strategy (23). Nevertheless, many centers have adopted “goal-directed” resuscitation strategies, whereby endpoints such as urine output are used to titrate intravenous fluid infusion rates after initiating resuscitation with either the Parkland or modified Brooke formula (24).

Even in patients with appropriate volume resuscitation, significant increases in ICP have been observed during resuscitation. A prospective study of 32 patients with an average burn of nearly 70% TBSA found elevated ICPs that peaked on the second day after burn injury to 31.4 ± 10.4 mm Hg (25). Additionally, in comparing survivors and non-survivors, patients who died had significantly higher ICPs and lower cerebral profusion pressures. Our results echoed this observation and showed that both the restrictive and aggressive resuscitation groups were associated with increasing ICP levels after initial injury and throughout 8 h of resuscitation. Although the exact mechanism of elevated ICP after burn injury is unknown, possible explanations include changes in the vascular permeability of the cerebral circulation, alterations in blood–brain barrier, and variations of cerebral perfusion autoregulation (26).

In TBI, acceleration/deceleration mechanisms cause direct injury to brain tissue that can lead to mechanical disruptions of brain tissue, alterations in cerebral blood flow, increased inflammation, and oxidative stress leading to cell death and cerebral edema (27). Acute management is aimed at maintaining cerebral perfusion and preventing secondary insults to unaffected brain tissue (28). Using our model of standardized burn and intracranial injury, no difference between the Parkland and modified Brooke formulas for resuscitation was observed regarding size of intracranial injury based on pathologic review of brain tissue. Although pathologic assessment of TBI differs from clinical assessment, these data would suggest that either resuscitation strategy did not significantly affect local secondary insults from the intracranial injury, at least in the short term within the first 8 h of combined injury (29).

Comparison between the aggressive and restrictive resuscitation strategies showed similar hemodynamic trends in heart rate, blood pressure, and central venous pressure in the first 8 h during resuscitation. Weight gain and total fluid administered were expectedly higher using the aggressive strategy. Although urine output was higher in the aggressive resuscitation group, urine output in both groups appeared adequate (>1 cc/kg/h) (30). Additionally, after 8 h lactate levels remained similar and the amount of norepinephrine required to maintain MAP goals was not significantly different between treatments. Taken together, both strategies appeared to provide adequate resuscitation in the first 8 h of injury, based on commonly utilized markers of end organ profusion (31).

In patients with TBI, use of hypertonic saline has shown benefit as a method of decreasing ICP (32). However in patients with concomitant burn injury/TBI, the benefit is less clear. The use of hypertonic saline in patients with isolated burns has been associated with increased risk of both acute kidney injury and death (33). The reasons for this are unclear but are thought to be related to alterations in serum sodium, anti-diuretic hormon release, and increased oxidative stress (34). There exists a scarcity of literature regarding the combined use of low volume hypertonic saline with standard burn resuscitation in the management of patients with combined burn injury/TBI. Nevertheless, maintenance of serum sodium levels >140 mEq/L to 150 mEq/L and systolic blood pressure > 90 mm Hg are both measures that have been shown to decrease secondary insults after TBI (35). Therefore, use of hypertonic saline and vasopressors should be at least considered, when combined burn injury/TBI are present. In these patients, judicious use of direct measurements of ICP can be helpful when monitoring the direct effects of resuscitation on cerebral edema (11).

Kidney injury after major burn injury is common, occurring in 30% to 60% of patients admitted for management of their burns (36, 37). Reasons for this association are uncertain but are thought to be related to decreased intravascular volume, damage from inflammatory mediators, hemoglobinuria, nephrotoxic medications used in burn management, and from sepsis from burn complications (line, wound, and pulmonary infections) (37). Our data showed that the restrictive resuscitation group was associated with higher serum BUN levels, yet serum creatinine levels were similar between groups. Although serum BUN is an imperfect marker of kidney injury and can be affected by protein intake, volume status, and renal tubular handling, this observation hints at but does not prove early renal damage occurs in the restrictive resuscitation group (38). These laboratory observations were also consistent with the observed trend in histologic evaluation of the kidneys for markers of injury seen between groups. However, these findings are somewhat confounded by the fact that swine in the restrictive group spent less time (31.2 min) in the goal MAP range, which may have led to relatively decreased renal perfusion. Taken together, the exact mechanism of renal injury is unknown and future studies looking at more specific markers of kidney injury including neutrophil gelatinase-associated lipocalin and kidney injury molecule-1 would be of benefit and would more specifically illustrate specific mechanisms of kidney injury if present. Additionally, more specific markers of lung, liver, and brain injury might highlight insults to these organ systems that were not seen on our laboratory or histology evaluation.

There are several limitations of this study. First, inflammatory markers of injury were not assessed in the model development or experimental aspect of this study. Moreover, our protocol only looked at the first 8 h of resuscitation after combined burn injury/TBI. This was chosen to increase standardization during the duration of resuscitation, as a longer duration would have required multiple teams to complete, as 2 to 3 h of preparation was needed prior to T0 of the experiment. However, comparing hemodynamic and laboratory trends over longer experimental times could have resulted in more significant differences between resuscitation strategies, as the full extent of organ injury may not be seen within 8 h of injury. Next, although no difference was observed in our primary outcome of size of brain injury between resuscitation strategies, our protocol was non-survival and a clinical assessment of the extent of TBI was not performed. This was chosen in an effort to minimize pain to the animals which would have required emergence from anesthesia for a neurologic assessment. Also, in our resuscitation algorithm swine were randomized to receive resuscitation with either the Parkland or modified Brooke formula and the rate of fluid infusion was kept constant for the duration of the experiment. At some burn centers, either the Parkland or modified Brooke formula is chosen to guide initial intravenous fluid rates, but rates are then titrated based on hourly urine output per institutional protocol (39). Our consistent rates may have led to relatively higher amounts of volume infused, based on the robust urine output that was observed. Lastly, the family-wise error rate was not controlled across the many statistical analyses that were completed but rather a P value of < 0.01 was chosen to determine significance which was based on previously utilized convention (40, 41). However, this strategy may have increased the chances of making a type 1 error.

Future studies will investigate inflammatory parameters of the model, similar endpoints beyond 8 h of resuscitation, as well as more sensitive and specific markers of early renal, hepatic, and neurologic function. Furthermore, our model of combined burn injury/TBI could also be used to evaluate the effects of more targeted therapies on ICP management in the immediate period after injury. Ultimately having a survival combined burn injury/TBI animal model would best assess the long-term effects of differing therapeutic modalities; however, logistical and ethical concerns are present when surviving animals with such severe injuries. Nevertheless, this model and these results provide some insight into the understanding and management of patients with combined burn injury/TBI.


Combined burn injury/TBI is a relatively rare injury pattern with increasing incidence. Therapeutic conundrums are present regarding patients with these concomitant injury patterns, especially related to the effects of fluid resuscitation on cerebral edema. Our standardized 40% TBSA burn injury and cerebral contusion creation was a safe, effective, and reproducible method to establish a large animal model for combined burn injury/TBI. Using this model, there were no differences in the size of brain injury seen on pathologic evaluation between aggressive and restrictive resuscitation groups and similar hemodynamic trends were observed with both strategies. After 8 h, a higher serum BUN was observed using the modified Brooke formula, suggesting early acute kidney injury; however, more sensitive and specific markers and prolong resuscitation time periods are needed to more fully assess this observation.


The authors acknowledge the veterinary support staff at the Clinical Investigation Facility at David Grant Medical Center for their help with this project


1. Blaisdell LL, Chace R, Hallagan LD, Clark DE. A half-century of burn epidemiology and burn care in a rural state. J Burn Care Res 33 (3):347–353, 2012.
2. Taylor SL, Lawless M, Curri T, Sen S, Greenhalgh DG, Palmieri TL. Predicting mortality from burns: the need for age-group specific models. Burns 40 (6):1106–1115, 2014.
3. Palmieri TL. Pediatric burn resuscitation. Crit Care Clin 32 (4):547–559, 2016.
4. Langlois JA, Rutland-Brown W, Thomas KE. Traumatic brain injury in the United States; emergency department visits, hospitalizations, and deaths. US Department of Health and Human Services, Centers for Disease Control Report. 2006, p. 7.
5. Dixon KJ. Pathophysiology of traumatic brain injury. Phys Med Rehabil Clin N Am 28 (2):215–225, 2017.
6. Dash HH, Chavali S. Management of traumatic brain injury patients. Korean J Anesthesiol 71 (1):12–21, 2018.
7. Graybill JC, Aden J, Rizzo JA, Clemens M, Janak J, White C, Driscoll I, Chung K. Concomittant traumatic brain injury and burns in combat trauma. J Am CollSurg 223 (4):e217, 2016.
8. Infanger M, Schmidt O, Kossmehl P, Grad S, Ertel W, Grimm D. Vascular endothelial growth factor serum level is strongly enhanced after burn injury and correlated with local and general tissue edema. Burns 30 (4):305–311, 2004.
9. Haberal M, Sakallioglu Abali AE, Karakayali H. Fluid management in major burn injuries. Indian J Plast Surg 43: (suppl): S29–S36, 2010.
10. Rajagopal R, Swaminathan G, Nair S, Joseph M. Hyponatremia in traumatic brain injury: a practical management protocol. World Neurosurg 108:529–533, 2017.
11. Cancio LC, Pruitt BA. Care of Patients with Burns and Traumatic Brain Injury. New York: Springer, 2017, 329–336.
12. Johnson MA, Williams TK, Ferencz SE, Davidson AJ, Russo RM, O’Brien WT Sr, Galante JM, Grayson JK, Neff LP. The effect of resuscitative endovascular balloon occlusion of the aorta, partial aortic occlusion and aggressive blood transfusion on traumatic brain injury in a swine multiple injuries model. J Trauma Acute Care Surg 83 (1):61–70, 2017.
13. Gibson-Corley KN, Olivier AK, Meyerholz DK. Principles for valid histopathologic scoring in research. Vet Pathol 50 (6):1007–1015, 2013.
14. Swindle MM, Smith AC. Swine in the Laboratory: Surgery, Anesthesia, Imaging, and Experimental Techniques. Boca Raton, FL: CRC press, 2015.
15. Abdullahi A, Amini-Nik S, Jeschke M. Animal models in burn research. Cell Mol Life Sci 71 (17):3241–3255, 2014.
16. Wardhana A, Lumbuun RFM, Kurniasari D. How to create burn porcine models: a systematic review. Ann Burns Fire Disasters 31 (1):65–72, 2018.
17. Cullen DK, Harris JP, Browne KD, Wolf JA, Duda JE, Meaney DF, Margulies SS, Smith DH. A Porcine model of traumatic brain injury via head rotational acceleration. Methods Mol Biol 1462:289–324, 2016.
18. Fritz HG, Walter B, Holzmayr M, Brodhun M, Patt S, Bauer R. A pig model with secondary increase of intracranial pressure after severe traumatic brain injury and temporary blood loss. J Neurotrauma 22 (7):807–821, 2005.
19. Wolf SE, Rose JK, Desai MH, Mileski JP, Barrow RE, Herndon DN. Mortality determinants in massive pediatric burns. An analysis of 103 children with > or=80% TBSA burns (> or=70% full-thickness). Ann Surg 225 (5):554–565, 1997.
20. Navickis RJ, Greenhalgh DG, Wilkes MM. Albumin in burn shock resuscitation: a meta-analysis of controlled clinical studies. J Burn Care Res 37 (3):e268–e278, 2016.
21. Chung KK, Wolf SE, Cancio LC, Alvarado R, Jones JA, McCorcle J, King BT, Barillo DJ, Renz EM, Blackbourne LH. Resuscitation of severely burned military casualties: fluid begets more fluid. J Trauma 67 (2):231–237, 2009.
22. Ivy ME, Atweh NA, Palmer J, Possenti PP, Pineau M, D’Aiuto M. Intra-abdominal hypertension and abdominal compartment syndrome in burn patients. J Trauma 49 (3):387–391, 2000.
23. Association AB. Advanced Burn Life Support Course: Provider Manual 2018 Update. Chicago, IL: American Burn Association, 2018.
24. Guilabert P, Usúa G, Martín N, Abarca L, Barret JP, Colomina MJ. Fluid resuscitation management in patients with burns: update. Br J Anaesth 117 (3):284–296, 2016.
25. Gueugniaud P, Jauffray M, Bertin-Maghit M, Bouchard C, Durand F, Fonrouge J, Petit P. Cerebral oedema after extensive thermal injury: prognostic significance of early intracranial and cerebral perfusion pressures. Ann Burns Fire Disast 10:72–77, 1997.
26. Shin C, Kinsky MP, Thomas JA, Traber DL, Kramer GC. Effect of cutaneous burn injury and resuscitation on the cerebral circulation in an ovine model. Burns 24 (1):39–45, 1998.
27. Vella MA, Crandall ML, Patel MB. Acute management of traumatic brain injury. Surg Clin North Am 97 (5):1015–1030, 2017.
28. Sheridan RL, Schaefer PW, Whalen M, Fagan S, Stoddard FJ Jr, Schneider JC, McConkey B, Cancio LC. Case records of the Massachusetts General Hospital. Case 36-2012. Recovery of a 16-year-old girl from trauma and burns after a car accident. N Engl J Med 367 (21):2027–2037, 2012.
29. Ma X, Aravind A, Pfister BJ, Chandra N, Haorah J. Animal models of traumatic brain injury and assessment of injury severity. Mol Neurobiol 56 (8):5332–5345, 2019.
30. Chiao HY, Chou CY, Tzeng YS, Wang CH, Chen SG, Dai NT. Goal-directed fluid resuscitation protocol based on arterial waveform analysis of major burn patients in a mass burn casualty. Ann Plast Surg 80: (2s suppl 1): S21–S25, 2018.
31. Pan J, Peng M, Liao C, Hu X, Wang A, Li X. Relative efficacy and safety of early lactate clearance-guided therapy resuscitation in patients with sepsis: a meta-analysis. Medicine (Baltimore) 98 (8):e14453, 2019.
32. Fatima N, Ayyad A, Shuaib A, Saqqur M. Hypertonic solutions in traumatic brain injury: a systematic review and meta-analysis. Asian J Neurosurg 14 (2):382–391, 2019.
33. Huang PP, Stucky FS, Dimick AR, Treat RC, Bessey PQ, Rue LW. Hypertonic sodium resuscitation is associated with renal failure and death. Ann Surg 221 (5):543–554, 1995.
34. Belba M. Our experience in the treatment of burn shock by hypertonic lactated saline solution. Ann Burns Fire Disasters 18 (2):61–67, 2005.
35. Jain V, Choudhary J, Pandit R. Blood pressure target in acute brain injury. Indian J Crit Care Med 23: (suppl 2): S136–S139, 2019.
36. Steinvall I, Bak Z, Sjoberg F. Acute kidney injury is common, parallels organ dysfunction or failure, and carries appreciable mortality in patients with major burns: a prospective exploratory cohort study. Crit Care 12 (5):R124, 2008.
37. Clark AT, Li X, Kulangara R, Adams-Huet B, Huen SC, Madni TD, Imran JB, Phelan HA, Arnoldo BD, Moe OW, et al. Acute kidney injury after burn: a cohort study from the parkland burn intensive care unit. J Burn Care Res 40 (1):72–78, 2019.
38. Vaidya VS, Ferguson MA, Bonventre JV. Biomarkers of acute kidney injury. Annu Rev Pharmacol Toxicol 48:463–493, 2008.
39. Paratz JD, Stockton K, Paratz ED, Blot S, Muller M, Lipman J, Boots RJ. Burn resuscitation—hourly urine output versus alternative endpoints: a systematic review. Shock 42 (4):295–306, 2014.
40. Noble WS. How does multiple testing correction work? Nat RBiotechnol 27 (12):1135–1137, 2009.
41. Zhu W. p<0.05,<0.01,<0.001,<0.0001,<0.00001,<0.000001, or< 0.0000001…. J Sport Health Sci 5 (1):77–79, 2016.

Combined burn injury and traumatic brain injury; modified Brooke formula; Parkland formula; resuscitation; thermal injury; traumatic brain injury

Copyright © 2020 by the Shock Society