From Cholera to “Fluid Creep”: A Historical Review of Fluid Resuscitation of the Burn Trauma Patient

Resuscitation of a patient with a traumatic burn presents unique and dynamic challenges, which often involve the critical care management of multiple physiological derangements. It is well known that a major burn injury can lead to burn shock and involve multiple organ systems. Burn shock has been described as the outcome of a combination of multiple factors, including hypovolemia, microcirculatory injury and dysfunction, and the release of local and systemic inflammatory mediators, which result in the body’s inability to adequately meet cellular needs.^1,2 After initial stabilization and assessment, the mainstay of early burn management is the treatment and prevention of shock with fluid resuscitation. The goal of fluid resuscitation is to replace the patient’s lost intravascular volume with crystalloid to maintain adequate tissue perfusion and organ function at the lowest physiologic cost.^3,4
There appears to be a consensus among burn care professionals that most minor burns can be orally resuscitated; however, controversy arises with the utilization of intravenous (IV) fluid therapy and the use of a number of similar resuscitation formulas (Table 1). To date, many studies have focused on the importance of fluid resuscitation in the treatment of burn patients; however, no universally accepted model for IV fluid therapy exists.^5–7 Regardless of which formula is used, it is clear that continuous individual titration of volume must be made according to the patient’s clinical response to avoid the detrimental problems associated with both over resuscitation and under resuscitation. Which formula is most appropriate and which fluid, or combination of fluids, is most advantageous continues to be debated.

Early History

The observations of Baron Dupuytren in the 1830s are possibly the earliest documented on the importance of fluid repletion in patients with burns. A French surgeon and pathologist, Dupuytren was best known for his description and development of surgical procedures for alleviating Dupuytren’s contracture; however, he also had extensive experience in the treatment of patients with burns. He was the first to propose a classification system for burns and the first to report a statistical study of fifty patients with burns treated at the Hotel Dieu based on their age, sex, and extent of the burn injury.^8,9 Dupuytren performed autopsies on patients with burns and correlated his findings to his knowledge of cholera patients. He further went on to present a rationale for fluid replenishment in burn victims that was similar to the treatment proposed by Sir William Brooke O’Shaughnessy in 1831 for the treatment of cholera.¹⁰ Dr. O’Shaughnessy was an Irish physician with accomplished work in pharmacology who is famous for introducing the medical use of cannabis sativa to the Western world. An outbreak of cholera in 1831 prompted O’Shaughnessy to analyze the blood of cholera patients. He reported that the resulting diarrhea “leads to dehydration, electrolytic depletion, acidosis, and nitrogen retention,” and “treatment must depend on IV replacement of the deficient salt and water.”^11–13 In 1854, Ludwig von Buhl, a German pathologist, made the correlation between the hemoconcentration that occurred in both patients with burns and cholera patients as a result of fluid loss. He went on to recommend that a saline solution be given to patients with burns either orally, subcutaneously, or intravenously.¹⁴ Certainly not all initial observations on the physiology of the burn patient were made outside the United States. As early as 1905, Haldor Sneve published research on the treatment of burns and skin grafting.¹⁵ Dr. Sneve touted the value of fluid repletion in patients with burns using a salt solution given by various routes, including enemas, to prevent shock. Unfortunately, his work was not built upon and a formal, goal-focused resuscitation formula for patients with burns would still not be realized until many years later.
In 1919, A.M. Fauntleroy, a surgeon in the US Navy, reported on 32 patients with extensive burns sustained in a shipboard coal dust explosion.¹⁶ Once the patients arrived at the hospital, morphine was given to control pain, and patients with major burns received a continuous proctoclysis of salt solution with sodium bicarbonate. After the patients were considered resuscitated they were given “water and large quantities of nutritional liquids every 2 hours by mouth including two or three eggnogs to which whiskey was added during the night.” In the fourth and fifth week, to combat what was termed “exhaustion,” a tonic of phosphorous, strychnine, and quinine was administered.¹⁷
Wartime served as a potent stimulus for advancement in the treatment of patients with burns. At the onset of World War I, physicians still did not have a clear understanding of why injured soldiers were dying of shock. The early work of Walter B. Cannon, Oswald Robertson, Arlie Bock, and Norman Keith during World War I provided great insight into the physiology of shock.^18–20 It was their work that described the significance of plasma loss, hemoconcentration as a result of plasma leakage due to damaged capillaries, and a discrepancy of vascular volume and blood volume. As a result of their work and others, blood transfusions and the liberal use of saline infusions became common in World War II.
Frank Underhill furthered understanding of burn shock beginning with his work on patients from the Rialto Theatre fire. On November 21, 1921 in New Haven, Connecticut, a standing room audience was viewing Rudolph Valentino in “The Sheik” at the Rialto Theatre when a fire broke out that killed 6 and injured 80. It was reported that prior to the movie the audience watched a stage show that used an incense burner to create atmosphere for the movie, which caused a fire backstage that quickly spread throughout the building.^21,22 Dr. Underhill measured hemoglobin, hematocrit, and chloride in the blood and blister fluid of 20 injured patients, and found the blister fluid content to be similar to plasma.²³ His early research showed that burn shock was the result of fluid shifts and set the stage for early fluid resuscitation after a burn injury. He consequently recommended replacement of the lost fluid with a salt and protein solution using hemoglobin as a guide to adjust therapy.^24,25
Nearly 21 years after Dr. Underhill’s work with the Rialto patients, the deadliest nightclub fire in United States history occurred at the Cocoanut Grove in Boston, Massachusetts (November 28, 1942). This fire claimed the lives of 492 people and injured hundreds more. It is reported that the club was filled with nearly 1000 occupants that evening, more than twice its official capacity of 460. It is speculated that a busboy lit a match to find a light bulb in the dimly lit lounge and accidentally started the fire. As the fire spread, there was little chance for the patrons to escape; the main entrance was a revolving door, which became congested and useless with panicked patrons, side doors had been locked shut to prevent people from leaving without paying their bills, and decorations covered some exit signs.^26–28 This disaster turned out to be a seminal event in the advancement of burn care and the modernization of public fire safety standards.
More than 300 patients were initially transported to the Boston City Hospital and an additional 114 patients were taken to the Massachusetts General Hospital; however, a majority of these patients died shortly after arrival.²⁷ Under the direction of Dr. Charles Lund, 132 patients were treated at the Boston City Hospital, whereas only 39 patients survived long enough to be treated at the Massachusetts General Hospital by a team lead by Dr. Oliver Cope.²⁹
As fate would have it, this disaster occurred at a time when both hospitals were well prepared to treat the massive influx of trauma patients. Only a week earlier, the city of Boston stockpiled emergency supplies and had conducted mock drills in preparation for a possible attack against the US coast. Additionally, the National Research Council had already started funding burn research programs conducted at the hospitals.³⁰
Patients at both institutions were initially treated with an IV fluid resuscitation of equal parts of plasma and saline. Cope based his resuscitation on using 1000-mL of fluid for each 10% total body surface area burned (TBSA) in the first 24 hours after injury.²⁹ Hematocrit, urine output, and urine nitrogen levels were followed and used as clinical indicators of resuscitation efforts. A similar method was employed at Boston City Hospital. Physicians at both hospitals noted that patients with both burns and inhalation injury were requiring more fluid during resuscitation. This raised the concern for inducing pulmonary edema. Cope decided to decrease the amount of fluid given to these patients, whereas, Lund treated them with the volumes they were requiring. In the end, no detrimental effects were reported from administering increased (needed) fluid volumes.
Numerous reports based on the clinical observations from both medical teams proved to be invaluable toward advancements in treating patients with burns.^31–34 In 1947, Cope and Moore³⁵ published the Burn Budget Formula based on their experiences with the Cocoanut Grove victims. They demonstrated that the hypovolemia produced from burn shock was due not only to fluid loss from the burn wound itself, but also from massive internal fluid shifts. Their budget formula related the patient’s fluid needs to that of a “normal sized” adult. During the first 24 hours the formula called for Lactated Ringer’s, 0.5 normal saline, colloid and glucose in water, with continuation of all of these fluids during the second 24 hours, except for the 0.5 normal saline. Kyle and Wallace later adapted Cope and Moore’s formula in 1951 in order to treat children.³⁶ In 1944, Dr. Charles Lund and Dr. Newton Browder published the renowned paper, “The estimation of areas of burns,” in which they presented the “Lund and Browder Chart” for estimation of TBSA.³⁷ This diagram is still used today with modifications made by the US Army Institute of Surgical Research.
The burn budget formula was used by many physicians; however, its use was based on the treatment of “normal-sized” adults with no adjustments for weight included. As one can now imagine this lead to the detrimental problem of fluid overload in certain patient populations. To address this issue, Dr. Everett Evans and his team developed a surface area resuscitation formula in 1952 based on a patient’s weight and TBSA burned. His work was the result of a combined research effort by the Medical College of Virginia and the US Army Medical Research and Development Command. In the first 24 hours, the formula called for normal saline to run at 1 mL/kg/% burn, colloid at 1 mL/kg/% burn, and glucose in water at 2000 mL. Additionally, the Evans groups noted that if a patient’s age or TBSA exceeded 50 they would most likely require more fluid. In utilizing their formula they stressed the importance of observing the patient’s clinical response to the fluids and not blindly allowing the formulas calculations to dictate the course.  
Further refinements in the Evans’ formula were the result of research conducted by the military. The US Army has always held a prominent place in the history of burn care. The Army had a vested interest in developing a burn care program since many of the most severely burned patients were a result of wartime injuries. The result was the creation of the US Army Institute of Surgical Research Burn Center (USAISR), the Defense Departments premier burn center located at the Brooke Army Medical Center, named for BG Roger Brooke. In 1953, Reiss and colleagues at the Army Burn Center developed the Brooke Formula, a modification of Evans’ formula. Here the emphasis was placed on increasing the amount of crystalloid and decreasing the colloid volume. They recommended that Lactated Ringer’s (LR) be run at 1.5 mL/kg/% burn, colloid at 0.5 mL/kg/% burn, and D5W at 2000 mL.³⁸

Modern History

In 1960, Wilson and Stirman proposed a change in the approach to burn resuscitation. They argued that patients with burns be treated solely with LR in amounts necessary to maintain an adequate urine output, and that administering blood or plasma in the first 48 hours was usually not necessary.³⁹
In 1964, Moncrief and Mason focused on the underestimation of insensible fluid loss, especially in patients with large, full-thickness burns.⁴⁰ They noted that fluid loss through the burn was considerable, and that the administration of 2000-mL of water solution free of electrolytes per 24 hours, was not adequate to replace these losses. Roe and Kinney validated this finding, and discovered that the insensible loss could be as high as 2-L per day through a dry, full-thickness burn.⁴¹
In 1965, Moyer and colleagues recommended resuscitation with crystalloid and hypertonic salt solution as a means of reducing the volume needed during resuscitation.⁴² They believed that using a hypertonic solution lead to a decreased amount of edema and exerted a positive effect on renal function by producing a higher urinary output per volume infused compared with isotonic solutions. Also of note, in the 1960s was the development of the Rule of Nines by Knaysi. The Rule of Nines provides a quick and easy method for the determination of TBSA burned and is still in widespread use today.   
The Modified Brooke formula was introduced by Dr. Basil Pruitt at the USAISR recommending LR at 2 mL/kg/% burn during the first 24 hours post burn.⁴³ In 1979, he published his findings on hemodynamic studies conducted on 10 severely burned patients treated at the Army Burn Center.⁴⁴ Fluid needs were estimated using the Brooke Formula and modifications were made according to patient response. Changes in mean blood volume, plasma volume, and cardiac output were recorded. He concluded that adult patients with extensive burns could be adequately resuscitated using LR at 2 mL/kg/% burn during the first 24 hours. In the second 24 hours, crystalloid is stopped and colloid infusion begins at a rate of 0.3 mL–0.5 mL/ kg/% burn with the addition of electrolyte free water at a volume to maintain urine output. Pruitt also challenged the use of hypertonic solutions concluding that there is “no documented need for sodium in excess of water in burn patients and such may in fact complicate post resuscitation care in burn patients who have marked evaporative water loss from the burn wound and elevated levels of rennin and angiotensin producing avid renal salt retention.”⁴⁴
The Parkland (or Baxter) formula, considered by many to be the “The Gold Standard” of fluid resuscitation, was developed by Charles Baxter at Parkland Hospital at Southwestern University Medical Center in the 1960s. This is by far the most commonly used formula in US burn centers today.⁴⁵ The formula was based on animal studies, and later trials on 11 patients with burns. In 1968, Baxter reported his results, which lead to the formation of the original Parkland formula.⁴⁶ The formula calls for LR to be administered at a rate of 4 mL/kg/% burn, with half the volume administered during the first 8 hours and the remaining volume administered over the next 16 hours, using urine output as a clinical guide. Baxter also stipulated that plasma be administered at a rate of 0.3–0.5 mL/kg/% burn during the fourth 8-hour period of the initial resuscitation, noting that crystalloid alone was not sufficient to correct the volume deficit. He would later go on to write that the use of plasma was based on animal studies, and that when applied to human studies, did not show an increase in plasma volume that was superior to LR during the first 24 hours after injury. In 1979, Baxter reported results on studies from the previous 10 years. He demonstrated that adequate fluid resuscitation can be achieved in most patients by using crystalloid administered in the range of 3.7-mL to 4.3-mL/kg/% burn.^47–49 He noted that in 400 adult patients the formula was accurate in 70% of the patients, with 12% requiring more and 18% requiring less fluid. In conclusion he stated, “lactated Ringer’s is effective as the initial and only fluid replacement for burned patients in the first 24 hours (it does restore fluid volume spaces and produces a rapid return of cardiovascular integrity); that 4 cc/kg is appropriate in about 70% of a larger population but certainly does not fit the entire population of burned patients; that plasma can be administered at any time post burn but is most effective if given between 24 and 30 hours.”⁴⁹ Further expanding on his statement that not every burn patient will require 4 cc/kg, Baxter identified specific groups of patients that would often require more fluid than predicted, including those with cutaneous burns, inhalation injuries, and electrical burns, and patients who received delayed resuscitation. Later, at the National Institute of Health consensus conference in 1981, both Baxter and Pruitt identified patients with inhalation injuries as those who would require additional fluid volumes during resuscitation, a finding that has been widely confirmed by subsequent investigators.^50–53

The Phenomenon of “Fluid Creep”

The Parkland formula was broadly accepted but has not gone unchecked throughout the years. Recent data suggest that the formula does not accurately predict fluid requirements in larger burns and that patients treated today frequently exceed the volumes predicted by the formula.^54–56 Pruitt first coined the term “fluid creep” in 2000 to describe this phenomenon of increasing resuscitation volumes and stated that clinicians should “push the pendulum back.”⁵⁷ Over resuscitation can produce significant complications such as abdominal and extremity compartment syndromes, pulmonary and cerebral edema, Acute Respiratory Distress Syndrome, and multiple organ dysfunction.^58–60 As previously noted, Baxter has acknowledged the fact that certain patient populations will require more fluid than is predicted, and stressed the importance of careful observation and monitoring of the patient’s response to necessary fluid adjustments.
In 2007, Saffle published a comprehensive review of the incidence, consequences, and possible etiologies of “fluid creep.”⁶¹ He recommended a number of potential therapeutic strategies for the treatment and prevention of overzealous fluid resuscitation including restricting early fluid resuscitation, considering the use of routine colloid, and utilizing resuscitation protocols. He also discusses the use of hypertonic saline in special populations and the possible pharmacological regulation of resuscitation. The labeling of excessive fluid volumes given during resuscitation as “fluid creep” has not gone unchallenged. Hartford wrote an editorial in response to Dr. Saffle’s article: “Fluid creep is a term that covers up for unfamiliarity with the root causes of administration of excessive volume of crystalloid and for poor and inattentive clinical management in acute burn resuscitation.”⁶² Additionally, in 2008, Blumetti et al⁶³ published a retrospective study of patients resuscitated with the Parkland formula over a 15-year period to determine accuracy based on urine output. Their review included data on 483 patients. They found that 43% received adequate resuscitation and 48% received over-resuscitation. Only 14% of adequately resuscitated and 12% of over-resuscitated patients met Parkland formula criteria. They concluded, “The actual burn resuscitation infrequently met the standard set forth by the Parkland formula” and  “patients commonly received fluid volumes higher than predicted by the Parkland formula.” They suggest that emphasis be placed on parameters used to guide resuscitation rather than calculated formula volumes.

Monitoring Resuscitation

The most commonly used measurement for monitoring fluid resuscitation in burn patients has been hourly urine output. This continues to serve clinicians well; however, this value must be used in context with frequent assessment of the patient’s vital signs, mental status, and laboratory values. Modern medicine has provided sophisticated techniques to monitor a patient’s ever changing physiologic response with the use of central venous and pulmonary artery catheters, and the measurements specific laboratory values, such as lactic acid and base excess levels. The value of this additional information, as it pertains to patients with burns, has yet been fully realized. Holm et al⁶⁴ published a prospective study in 2004 on the effects of invasive monitoring on burn shock. They randomized 50 adult patients to receive either Parkland resuscitation or intrathoracic blood volume driven (pre-load) therapy and failed to show benefit from the aggressive use of volume to increase preload and cardiac index.⁶⁴ Cancio et al⁶⁵ have investigated the contribution of base deficit and alveolar-arterial gradient in predicting mortality after burn injury. Their analysis concluded that a base deficit and increased alveolar-arterial gradient contributed independently, but modestly, to ultimate mortality after burn injury. Kamolz et al⁶⁶ conducted a prospective study to evaluate if plasma lactate is a useful parameter to estimate burn shock severity. They concluded that the initial lactate level is a useful parameter for prediction of mortality and “a better chance of survival occurs when resuscitation results in a lactate clearance to normal values within 24 hours.” Further study is needed to determine the clinical usefulness of theses parameters to guide resuscitation.

American Burn Association Guidelines

The most recent American Burn Association Practice Guidelines published in 2008 recommend: “Fluid resuscitation, regardless of solution type or estimated need, should be titrated to maintain a urine output of approximately 0.5–1.0 mL/kg/h in adults and 1.0–1.5 mL/kg/h in children; increased volume requirements can be anticipated in patients with full-thickness injuries, inhalation injury, and a delay in resuscitation; hypertonic saline should be reserved to providers experienced in this approach. Plasma sodium concentrations should be closely monitored to avoid excessive hypernatremia; Option: the addition of colloid-containing fluid following burn injury, especially after the first 12 to 24 hours postburn may decrease overall fluid requirements.”⁶⁷

Conclusion

Certainly, monumental advances have been made in burn resuscitation, which have lead to dramatically decreased mortality rates and virtually eliminated post-burn renal failure. The early work of Cope and Moore, Evans, Artz, Moyer, Baxter, Pruitt, and others have served the burn community well and continue to drive modern fluid resuscitation. The fact that debate continues as to which formula is most advantageous only serves to benefit patients with burn injuries, and stimulates healthy educational debate among clinicians. Further research is needed to investigate the phenomenon of “fluid creep.” Urine output serves as the best guide to monitor resuscitation when correlated with a patient’s overall clinical status. The importance of individualizing fluid titration to the patient’s clinical response cannot be stressed enough to avoid the detrimental problems associated with both over resuscitation and under resuscitation.  

Acknowledgements

The author thanks Maria Ford for her literature review and Matthew L. Maddox, BA, and Isaac Goldstein for manuscript review.